cgiar science forum 2009 - IPC Global Platform

http://crop.scijournals.org/content/vol50/Supplement_1

ii CROP SCIENCE, VOL. 50, MARCH–APRIL 2010CROP SCIENCE, VOL. 50, MARCH–APRIL 2010 WWW.CROPS.ORG

Science Forum 2009 was held on 16-17 June, in Wageningen, the Netherlands. It brought together more than 300 scientists from 55 countries, from advanced research centers, national agricultural research systems, donors, and NGOs to explore recent scientifi c advances and to identify where there exists real potential to capture these advances to further international development. It also explored new modalities for research collaboration between the CGIAR and its partners. A selection of papers has been collected to illustrate the subjects and outcomes of the conference. Special thanks are extended to guest editor Brent Godshalk and to the CGIAR Science Forum 2009 Organizing Task Force: Rudy Rabbinge (Chair), Jeff Sayer, Gebisa Ejeta, Ren Wang, Mark Holderness, Bram Huijsman, Fred Cholick, Rodomiro Ortiz, Manuel Lantin, Haruko Okusu, and Christine Deane.

Forward iii

Introduction v

Resilient Natural Resource Systems

Should Enhanced Resilience Be an Objective of Natural Resource Management Research for Developing Countries?

Brian Walker, Jeff Sayer, Neil L. Andrew,

and Bruce Campbell

S-10

Boundary Work and the Complexity of Natural Resources Management

Peter P. Mollinga

S-1

The Future of Food: Developing More Nutritious

Diets and Safer Food

The Future of Food: Scenarios for 2050

Bernard Hubert, Mark Rosegrant, Martinus

A. J. S. van Boekel, and Rodomiro Ortiz

S-33

Biofortifi cation—A Sustainable Agricultural Strategy for Reducing Micronutrient Malnutrition in the Global South

Howarth E. Bouis and Ross M. Welch

S-20

Relearning Old Lessons for the Future of Food—By Bread Alone No Longer: Diversifying Diets with Fruit and Vegetables

John D. H. Keatinge, Farid Waliyar,

Ramni H. Jamnadas, Ahmed Moustafa,

Maria Andrade, Pay Drechsel, Jacqueline

d’A. Hughes, Palchamy Kadirvel, and

Kartini Luther

S-51

CGIAR SCIENCE FORUM 2009SCIENCE FOR DEVELOPMENT: MOBILIZING GLOBAL LINKAGES

A SUPPLEMENT TO CROP SCIENCE, VOLUME 50, NO. 2MARCH–APRIL 2010

Transforming Agricultural Science, Research,

and Technology Generation

Information and Communication Technologies—Opportunities to Mobilize Agricultural Science for Development

Peter Ballantyne, Ajit Maru, and Enrica

M. Porcari

S-63

Beyond the Yield Curves: Exerting the Power of

Genetics, Genomics, and Synthetic Biology

Mobilizing Science to Break Yield Barriers

Ronald L. Phillips

S-99

Climate Risk Management for Adaptation to Climate Variability and Change

Walter E. Baethgen

S-70

Rapid Determination of Gene Function by Virus-induced Gene Silencing in Wheat and Barley

Cahid Cakir, Megan E. Gillespie, and

Steven R. Scofi eld

S-77

Breeding and Cereal Yield Progress

R. A. (Tony) Fischer and Gregory

O. Edmeades

S-85

Eco-effi ciencies in Agro-ecosystems

Eco-effi cient Agriculture: Concepts, Challenges, and Opportunities

Brian A. Keating, Peter S. Carberry, Prem

S. Bindraban, Senthold Asseng, Holger

Meinke, and John Dixon

S-109

Enhancing Eco-effi ciency in Agro-ecosystems through Soil Carbon Sequestration

R. Lal

S-120

More than Eco-effi ciency is Required to Improve Food Security

S. E. Park, S. M. Howden, S. J. Crimp, D.

S. Gaydon, S. J. Attwood, and P. N. Kokic

S-132

Agriculture Beyond Food: Science for

a Biobased Economy

6J. W. A. Langeveld, J. Dixon, and J.

F. Jaworski

S-142

Biorefi neries– A Path to Sustainability?

Rajni Hatti-Kaul

S-152

TABLE OF CONTENTS

crop science, vol. 50, march–april 2010 iii

forward

There is widespread acknowledgement that a major chal-lenge to global development is to double food and fiber production within the next three decades. More food with better characteristics will be needed to meet the demands of an increasing better fed world population. This should be done with less water, nutrients, pesticides, and other external inputs. The productivity rise per ha is needed for environmental reasons and to spare land for nature and biodiversity. This increase in food production will have to be achieved during a period when pressures on land for biofuels, industrial raw materials, and urbanization are also increasing dramatically.

The Consultative Group on International Agricul-tural Research (CGIAR) has been a catalyst in remarkable increases in agricultural productivity in the develop-ing world for almost half a century. It strives to produce international public goods through agricultural research for sustainable food and fiber production. The CGIAR’s impacts are felt through better nutrition, improved public health, poverty reduction, and raised standards of living in developing countries.

The CGIAR is an alliance whose 64 Members sup-port 15 international research centers, working in col-laboration with hundreds of government and civil society organizations, as well as private enterprises throughout the world. CGIAR Members include 21 developing and 26 industrialized countries, four co-sponsors, and 13 other international organizations.

Through strategic investments in agricultural and natural resources management research, the CGIAR leverages donor funding to strengthen food security. CGIAR expenditures amounted to more than $500 mil-lion in 2008—the single biggest investment made to mobilize science to improve food security and contribute to the cornerstone of sustainable development worldwide. Today, more than 8,000 CGIAR scientists and staff are active in over 100 countries.

Throughout its history the CGIAR has worked to mobilize the best in science to contribute to international development. Working in partnership with national agricultural research systems, with universities and pub-lic research institutes in both developed and developing countries, the CGIAR has contributed to sustainable and poverty-reducing development through the research and research-related activities of its centers. Major pro-ductivity gains have been achieved and the CGIAR has also contributed to improved policies and institutional arrangements, which provide a solid foundation for better management of agriculture and natural resources.

Notwithstanding the progress that has been made, the scope and depth of challenges to achieving sustain-able agricultural systems is increasing. To meet these challenges the CGIAR has evolved over time, shifting its focus and structure to adapt to the needs of the present, while at the same time investing in research to meet the demands of the future. The CGIAR has grown from a small group of research centers focused on raising agricul-tural productivity, largely through crop breeding activi-ties, to a strategic alliance of research centers hosting a range of partnerships and collaborative engagements, and addressing a broader and more diverse set of research-for-development activities. CGIAR centers now address the sustainable development needs of forest, aquatic, range-land, and irrigation systems.

Seven distinct phases of development are evident in its history:

Published in Crop Sci 50:iii–iv (2010). doi: 10.2135/cropsci2010.06.0001for © Crop Science Society of America | 677 S. Segoe Rd., Madison, WI 53711

All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the mate-rial contained herein has been obtained by the publisher.

Crop SCienCe: CGiar SCienCe forum 2009

iv www.crops.org crop science, vol. 50, march–april 2010

1. Plant breeding activities oriented to high-yielding varieties of rice, wheat, and maize, the major food staple crops (1960–1965);

2. Plant breeding activities underpinned by agronomic technologies tailored to the needs of the high-yield-ing varieties. Special emphasis was given to crop protection, irrigation, soil fertility, plant nutrition (1965–1975);

3. Farming systems research to fine-tune the agronomic and technological activities to the specific needs of the various socio-economic contexts of farming sys-tems, and more socio-economic research to address distorted policies and weak institutions (1975–1980);

4. Broadening the scope of research by including biodi-versity, natural resource management, natural eco-systems, and agro-forestry (1980–1990);

5. The introduction of Eco-regional Programs aimed at achieving a higher degree of integration across broader application domains (1990–1998);

6. Growing recognition that global programs should be addressed by broader coalitions amongst the centers, advanced research institutes and their national part-ners, and thus the introduction of Challenge Pro-grams (1998–2007);

7. The imperative for broader mobilization of sci-ence and the need to fuel a more inclusive research and innovation system for agriculture and natural resources. The focus has been on stimulating greater public and private sector investment, and extending partnerships with a broader range of research suppli-ers and users and with civil society (present).

The CGIAR’s agenda has broadened considerably as it has shifted emphasis from producing new and improved crop varieties, to developing approaches, articulating problems, and devising with its partners common agendas and roadmaps to solutions. Similarly, the CGIAR’s activi-ties as a moderator, facilitator, stimulator, and a bridge to broader stakeholder groups have increasingly been in demand.

The role of the Science Council of the CGIAR is to oversee the quality and relevance of its research to address the CGIAR mission. In pursuing this mandate, the Sci-ence Council works to enhance opportunities for the CGIAR to engage with and mobilize other relevant pro-

viders of science, in line with the organization’s agricul-tural and natural resource development objectives. 2010 will be an exciting year of change in the CGIAR—the collection of individual research centers with their own agendas and sometimes competing funding needs is now to be replaced by a unified research system which will exploit synergies amongst the centers and their external partners. This will lead to more inclusive thinking and the ability to mobilize optimal consortia of research and development providers with improved links to the ulti-mate beneficiaries of research—the farmers themselves. The transaction costs and overheads of this system will decrease. There will be a division of labor between the centers and their consortium and the fund council, which is responsible for the allocation of resources to the work of achieving CGIAR impacts.

In the newly restructured CGIAR, the Science Council is succeeded by an Independent Science and Partnership Council. The new ISPC remains dedicated to mobilizing the best science and fostering partnerships that generate conditions conducive to translating science and scientific breakthroughs into innovations that support sustainable and adequate food production. In pursuit of this aim, the Science Council convened Science Forum 2009, which brought together more than 300 participants from 55 countries, to examine scientific advances that offer significant opportunities for agriculture and natu-ral resources development. The Forum was organized in cooperation with the CGIAR Secretariat at the World Bank, the Alliance of the CGIAR Centers, the Global Forum on Agricultural Research, and Wageningen Uni-versity and Research Centre. The current issue presents a selection of papers from the Science Forum in the hope and expectation that it will raise the profile of the science-for-development mission of the CGIAR, expand the net-work of research organizations mobilized to address the CGIAR mission, and raise even greater awareness of the urgency of intensifying research investments to meet the development challenges of the coming decades.

Prof. Rudy RabbingeISPC Chair

CGIAR

crop science, vol. 50, march–april 2010 v

symposia

Science for Development: Mobilizing Global Partnerships

Christine Deane,* Gebisa Ejeta, Rudy Rabbinge, and Jeff Sayer

in the developing world. Tactical partnerships are vital to conduct and to inform the research. Entering into part-nership arrangements that present a genuine value-added proposition has been central to the CGIAR’s successes in its impact-oriented research.

Partnerships that connect development-focused agri-cultural research with the extensive research capacities of the developed and emerging economies can increase the returns on donor countries’ broader investment in research. The CGIAR has been leveraging donor funds through its strategic investment choices for almost half a century. As a consortium of 15 agricultural research Cen-ters, largely situated in developing countries, and support-ing several major collaborative research programs, it is uniquely positioned as a partner in research.

Planning and conducting global public goods research requires highly effective partnerships between the appro-priate providers of ideas, new knowledge, technology, social science understanding, and policy processes. Ensur-ing that pathways to disseminate useful advances are available will equally require cooperation between the CGIAR and other parts of the research community that help to translate and adapt research results to ensure on-the-ground impact.

In its current phase of growth, the CGIAR is com-mitted to building stronger, better, and more inclusive partnerships to strengthen impact. The Science Council (which now, in recognition of this imperative, has become

Investment in agricultural research has consistently delivered high rates of return in terms of increasing

productivity, improving nutrition, and reducing poverty.1 This is the mandated aim of the Consultative Group on International Agricultural Research (CGIAR). A route therefore exists through agricultural research to boost food resources, improve livelihoods, enhance quality of life, and underpin the stability of societies in the devel-oping world. Supporting quality and relevant agricultural research must be a priority for all development coopera-tion as well as stability and security agendas.

The challenges to food security in developing coun-tries are great, but so are the opportunities for developing solutions. The number and scope of possible avenues for research are extensive. They range from scientific prob-lems, where the focus is on bio-molecular and biophysical research, to regional- and global-level challenges, where meshing technological innovations with social, institu-tional, and economic research dominates. Demands exceed the funds available. The importance of selecting strategic research directions that will have the greatest impact is clear. The CGIAR invests in research that increases the productivity of agricultural and natural resource systems

Published in Crop Sci. 50:v–viii (2010). doi: 10.2135/cropsci2010.12.0001symp © Crop Science Society of America 677 S. Segoe Rd., Madison, WI 53711 USA

All rights reserved. No part of this periodical may be reproduced or trans-mitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for print-ing and for reprinting the material contained herein has been obtained by the publisher.

C. Deane, CGIAR Independent Science & Partnership Coun-cil, Food and Agriculture Organization of the United Nations, Viale delle Terme di Caracalla, 00153 Rome, Italy; G. Ejeta, Dep. of Agronomy, Purdue Univ., 915 West State St., West Lafayette, IN 47906, USA; R. Rabbinge, Wageningen Univ., Costerweg 50, 6701BH Wageningen, The Netherlands; J. Sayer, International Union for Conservation of Nature, 28 rue Mauverney, CH-1196, Gland, Switzerland. Received 24 Feb. 2010. *Corresponding author ([email protected]).

1Investment in agricultural research has delivered an average rate of return of 43% in 700 development projects evaluated in developing countries. For the poorest people, GDP growth originating in agriculture is about four times more effective in reducing poverty than GDP growth originating outside the sector. World Development Report 2008, Agriculture for Development, The World Bank.

vi www.crops.org crop science, vol. 50, march–april 2010

the Independent Science and Partnership Council) sup-ports the development of better ways of organizing how agricultural and natural resources management research is conducted. As a contribution to this, the Science Coun-cil convened the CGIAR Science Forum 2009, which brought together more than 300 participants from 55 countries. The meeting focused on identifying innovative science, and the arrangements that can help to mobilize it more effectively, for development goals. It explored recent scientific advances in six areas that have the potential to improve agriculture, and it highlighted opportunities for new linkages between distinct research communities. The context for these discussions was set by opening presenta-tions that examined the challenges and opportunities to improve agriculture and natural resource management in developing countries. They also put forward ways to mobilize research resources and agricultural know-how more effectively. Six parallel workshops, each focusing on rapidly-moving areas of research, formed the basis of the two-day agenda:

• Resilient natural resource systems;• The future of food: developing more nutritious diets

and safer food;• Information and communication technologies (ICTs):

transforming agricultural science, research, and technology generation;

• Beyond the yield curve: exerting the power of genet-ics, genomics, and synthetic biology;

• Eco-efficiencies in agro-ecosystems; and• Agriculture beyond food: science for a bio-based

economy.Each workshop addressed the following questions:

What is the frontier science in this area? Where is the potential for impact on sustainable food security and international development goals? What are the research needs, and what other changes are needed, if this potential is to be realized? Which applications are the most feasible for achieving impact through take-up into national agri-cultural research systems within a short time-frame (3–5 years), and which have longer time horizons?

The current issue of Crop Science offers a selection of the papers presented and debated during these workshops.

Participants agreed that the major challenge ahead is to double primary food production in a sustainable way within the next three decades, and that this increase must be achieved largely through improvements in agricultural productivity, as limits on the availability of land and the scarcity of other inputs will persist. In addressing this chal-lenge, the Forum concluded that there is a pressing need to introduce new paradigms for agricultural productivity and agricultural research.

One example of this is the concept of resilience. The impacts of climate change, market volatility, and other global, regional, and local shocks (e.g. induced by political

volatility or environmental changes) have raised the sug-gestion that in striving to enhance food security, resilience should be considered alongside productivity. There had been some perception that the concepts of ‘resilience’ and ‘productivity’ are incompatible, but although there may be necessary trade-offs between the two, these concepts should be considered to be complementary. Attaining sustainable increases in production requires that the processes involved do not have secondary effects that feedback negatively on the functioning of the resource base, and therefore on flows of other valued goods and services. Adopting a resilience approach aims to ensure that gains in productivity lead to sustained increases in human well-being. In this current issue of Crop Science, Walker et al. (2010) explore the con-cept of resilience, and examine when and how enhanced resilience might become an objective of research.

Broadening the foundation of the global food supply is similarly a crucial factor in striving to improve food secu-rity and enhance nutrition. Currently we rely on just 15 plant species and eight animal species for more than 90% of all human food (Convention on Biological Diversity, 2008). The intensive research focus on wheat, rice, and maize often discounts many valuable opportunities in underuti-lized crops. The Forum explored untapped opportunities that exist here, such as developing better cultivation meth-ods for underutilized crops, and the use of biofortification to boost micronutrient content. Bouis and Welch (2010), Hubert et al. (2010), and Keatinge et al. (2010) discuss the challenges and opportunities in this field.

Developments in ICTs are transforming the ways in which knowledge, information, and data are generated, used, and shared in many different fields. These develop-ments are widely recognized to have valuable applications in efforts to improve agricultural productivity, promote sustainable practices, and contribute to the efficient opera-tion of markets. Although the potential is great, harness-ing this potential for the benefit of the developing world requires greater strategic investment and adaptation to the specific needs of developing countries. Ballantyne et al. (2010) explore these issues further.

Powerful new molecular tools developed in recent decades offer ever greater opportunities to develop improved crops. The process of identifying gene function is particularly intractable in wheat and barley, due to the size and complexity of their genomes, and inherent diffi-culties in achieving genetic transformation in these plants. Cakir et al. (2010) present a system of virus-induced gene silencing as a useful new tool that overcomes many of these obstacles and promises to greatly facilitate the assess-ment of gene function. It was noted during the meeting that achieving increases in yields requires a combination of improved genotypes, optimal agronomic management, and the timely availability of appropriate inputs. Research aimed at improving the genetic makeup of crops, therefore,

crop science, vol. 50, march–april 2010 www.crops.org vii

to yield bio-based products and production systems that compliment rather than compete with food production. Langeveld et al. (2010) and Hatti-Kaul (2010) discuss the challenges, the opportunities, and the complexities of these issues.

Mollinga (2010) emphasized that more attention should be focused on the boundaries between differ-ent research disciplines, and those between research and policy, when addressing sustainable agricultural produc-tion and natural resource management in the developing world. Indeed, a consistent theme that arose during the meeting was the need for more cross-disciplinary coop-eration in research programs that seek to address these kinds of complex challenges. Incentive mechanisms that are well-crafted and adapted to facilitate these kinds of collaborations were considered essential. The need for people with ‘T-shaped’ skill sets (i.e. being specialists in one discipline and knowledgeable across complimen-tary disciplines at the same time) was also highlighted. Another common theme that emerged in discussion was the importance of considering without prejudice all pos-sible research tools and technologies that can potentially contribute to solutions.

As a cooperative effort of the CGIAR centers, the Science Council, and advanced research institutes, the CGIAR Science Forum 2009 represented a positive step towards broadening the engagement of the international scientific and development communities in research that holds significant potential to contribute to food security. The Forum identified areas of rapid scientific innovation where opportunities exist, and explored new modali-ties for research collaboration and partnership. It debated which areas of research hold the greatest promise for inter-national development, where the most pressing research needs are, and what kinds of collaborative work and part-nerships should be supported to realize this potential. The meeting also provided a forum in which to engage the broader research community in the science-for-develop-ment mission.

The findings and concluding directions arising from the Forum, presented in part in the papers in this current issue of Crop Science, can serve to guide the current debate on refining and refocusing the CGIAR’s research agenda. They also offer elements for consideration in broader dis-cussions on designing science-for-development agendas in other contexts. Some of the key conclusions have immedi-ate relevance to the CGIAR, and could be adopted as part of the new research agenda (CGIAR Mega Programs). As the CGIAR transitions to a new research agenda, there is, for example, an opportunity to factor in the concept of resilience across the Mega Programs. Likewise, crop improvement programs relying on advances in molecular genetics should be more closely developed with programs aimed at developing a better understanding of relevant

should advance hand-in-hand with better knowledge of crop physiology and agronomy, and a greater capac-ity to understand the complex phenotypic responses of crops in the field. Discussions on this topic concluded that research programs on genetic improvement should take this into considerations in their design, and avoid viewing molecular genetic improvement in disciplinary isolation. Fischer and Edmeades (2010) and Phillips (2010) discuss this in more depth. Climate change featured in discussions throughout the Forum, and Baethgen (2010) emphasizes the need to consider risk management in policymaking for adaptation to climate variability.

The need to increase efficiency in the use of resources in agricultural production was also raised in discussion throughout the Forum. Production ecological principles are based on the notion that a production factor is most efficiently used when other required factors are at their optimum. Inputs should therefore be balanced to crops’ needs in time and space, considering location-specific ecological conditions, in order to yield the highest returns on those inputs. Spreading these principles is urgently needed, as they can help to improve agriculture in devel-oping countries. The concept of eco-efficiency is multi-dimensional, encompassing both ecological and economic dimensions of sustainable agriculture. Although social and institutional aspects of sustainability are often not explic-itly captured in eco-efficiency measures, they present sig-nificant obstacles as well as opportunities when trying to transition to more eco-efficient agriculture, and as such, they should be taken into account. Risk remains a critical factor influencing the uptake of more eco-efficient mea-sures. In striving to achieve greater eco-efficiencies and increase sustainable food production in the developing world, those risks most relevant in the context of develop-ing country agriculture must be taken into consideration. Keating et al. (2010), Lal (2010), and Park et al. (2010) discuss different aspects of these issues.

The advent of bio-based products (e.g. flavors, biomol-ecules, bioplastics etc.) and the development of a bio-based economy have often been hailed as a means of expanding sustainable production, but there are critical resource con-cerns, which are particularly acute in the developing world (e.g. the conflicting demands for resources used for food and for biofuel production). As bio-based production is being increasingly promoted and adopted, rational advice to guide the development of initiatives and incentives is essential. Debate on this topic at the Forum explored these concerns, and questioned whether research should focus on improving the efficiencies of first-generation biofuels, for example, or whether greater effort should be directed towards first assessing the social, economic, and environ-mental impacts of their production in developing coun-tries. An alternative focus would be to direct research to a far greater extent towards initiatives that are more likely

viii www.crops.org crop science, vol. 50, march–april 2010

crop physiology and agronomy. Further, by including social as well as economic and environmental criteria in the development of eco-efficiency tools and technologies, research portfolios would take on a more holistic approach to the issues constraining development.

Ultimately, the CGIAR Science Forum 2009 called for a new paradigm in agricultural research for devel-opment, where partners in research, donors, and com-munities in the developing world together realize better strategies to reach the goal of doubling food production within the next 30 years.

ReferencesBaethgen, W.E. 2010. Climate risk management for adaptation of

climate variability and change. Crop Sci. 50(2):S-70–S76.Ballantyne, P., A. Maru, and E.M. Porcari. 2010. Informa-

tion and communication technologies– Opportunities to mobilize agricultural science for development. Crop Sci. 50(2):S-63–S-69.

Bouis, H.E., and R.M. Welch. 2010. Biofortification– A sustain-able agricultural strategy for reducing micronutrient malnu-trition in the Global South. Crop Sci. 50(2):S-20–S-32.

Cakir, C., M.E. Gillespie, and S.R. Scofield. 2010. Rapid deter-mination of gene function by virus-induced gene silencing in wheat and barley. Crop Sci. 50(2):S-77–S-84.

Convention on Biological Diversity. 2008. What’s the Prob-lem? [Online]. Available at http://www.cbd.int/agro/whatstheproblem.shtml (verified 24 Feb. 2010).

Fischer, R.A., and G.O. Edmeades. 2010. Breeding and cereal

yield progress. Crop Sci. 50(2):S-85–S-98.Hatti-Kaul, R. 2010. Biorefineries– A path to sustainability? Crop

Sci. 50(2):S-152–S-156.Hubert, B., M. Rosegrant, M.A.J.S. van Boekel, and R. Ortiz.

2010. The future of food: Scenarios for 2050. Crop Sci. 50(2):S-33–S-50.

Keating, B.A., P.S. Carberry, P.S. Bindraban, S. Asseng, H. Meinke, and J. Dixon. 2010. Eco-efficient agriculture: Concepts, chal-lenges, and opportunities. Crop Sci. 50(2):S-109–S-119.

Keatinge, J.D.H., F. Waliyar, R.H. Jamnadas, A. Moustafa, M. Andrade, P. Drechsel, J.d’A. Hughes, P. Kadirvel, and K. Luther. 2010. Re-learning old lessons for the future of food– By bread alone no longer: Diversifying diets with fruits and vegetables. Crop Sci. 50(2):S-51–S-62.

Lal, R. 2010. Enhancing eco-efficiency in agro-ecosystems through soil carbon sequestration. Crop Sci. 50(2):S-120–S-131.

Langeveld, J.W.A., J. Dixon, and J.F. Jaworski. 2010. Development perspectives of the biobased economy: A Review. Crop Sci. 50(2):S-142–S-151.

Mollinga, P.P. 2010. Boundary work and the complexity of natural resources management. Crop Sci. 50(2):S-1–S-9.

Park, S.E., S.M. Howden, S.J. Crimp, D.S. Gaydon, S.J. Attwood, and P.N. Kokic. 2010. More than eco-efficiency is required to improve food security. Crop Sci. 50(2):S-132–S-141.

Phillips, R.L. 2010. Mobilizing science to break yield barriers. Crop Sci. 50(2):S-99–S-108.

Walker, B., J. Sayer, N.L. Andrew, and B. Campbell. 2010. Should enhanced resilience be an objective of natural resource man-agement research for developing countries science? Crop Sci. 50(2):S-10–S-19.

S-10 www.crops.org cropscience,vol.50,march–april2010

symposia

The Green Revolution was a product of conventional agri-cultural research applied to developing country situations. It

focused on improving productivity and emphasized agricultural intensification, often using high inputs. The Green Revolution has been criticized on several fronts; for instance, that it failed to benefit farmers in low-potential areas and that it exposed its ben-eficiaries to risks associated with narrowly specialized agricultural systems. Farmers whose livelihoods were based on a broad range of products were thought to be less vulnerable than those depen-dent on a single crop and cropping system (Conway, 1997).

Recent initiatives to address poverty alleviation through agricultural research have responded to these criticisms by giving more attention to sustainability and, more recently, to resilience (von Braun et al., 2009). This has been motivated by the realiza-tion that poor farmers in developing countries are vulnerable to the external shocks that are likely to be caused by increased cli-matic variability, global economic volatility, civil disturbances, and disruption of supplies of agricultural inputs (FAO, 2008). There have been a number of recent publications on food secu-rity that have discussed the merits of agricultural systems that are more diverse and less dependent on external inputs (notably those derived from scarce fossil fuels). There has been a wave of interest

Should Enhanced Resilience Be an Objective of Natural Resource

Management Research for Developing Countries?

Brian Walker,* Jeff Sayer, Neil L. Andrew, and Bruce Campbell

aBsTRaCTProductivityenhancementhastraditionallybeenthemainfocusofagriculturalresearchtoallevi-atepovertyandenhance foodsecurityofpoorfarmers in the developing world. Recently, theharmfulimpactofclimatechange,economicvol-atility,andotherexternalshocksonpoorfarmershas led to concern that resilience should fea-turealongsideproductivityasamajorobjectiveofresearch.Theapplicabilityofrecentworkonresilient social–ecological systems to theprob-lemsofpoorfarmersisreviewed,andproposalsaremadeforissuesthatneedtobeaddressedindeterminingwhenandhowenhancedresiliencemightbecomeanobjectiveofresearch.

B. Walker, CSIRO Sustainable Ecosystems, Box 284, Canberra ACT 2601, Australia; J. Sayer, International Union for Conservation of Nature, 28 rue Mauverney, CH-1196 Gland, Switzerland; N.L. Andrew, The WorldFish Center, P.O. Box 500, Penang, Malaysia; B. Campbell, CGIAR Challenge Program on Climate Change, Agriculture and Food Security (CCAFS), Dep. of Agriculture and Ecology, Univ. of Copenhagen, Rolighedsvej 21, 1958 Frederiksberg C, Denmark. WorldFish Contribution no. 1936. Received 2 Oct. 2009. *Corresponding author ([email protected]).

Published in Crop Sci. 50:S-10–S-19 (2010). doi: 10.2135/cropsci2009.10.0565 Published online 27 Jan. 2010. © Crop Science Society of America | 677 S. Segoe Rd., Madison, WI 53711 USA

All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

cropscience,vol.50,march–april2010 www.crops.org S-11

in shifting emphasis away from productivity enhancement and toward sustainability and resilience (McNeely and Scherr, 2003; IAASTD, 2009a, 2009b; UNEP, 2009).

Much of the interest in sustainability and resilience in agriculture comes from the industrialized world and is manifest in significant movements supporting organic agriculture and local self-sufficiency. There is a rich set of literature addressing these issues (McNeely and Scherr, 2003; Millennium Ecosystem Assessment, 2005; Pollan, 2006; Ronald and Adamchak, 2008). Recently, the logic of more diverse, locally sustainable agriculture has been applied to the developing world and the issue of increas-ing the resilience of poor developing world farmers has emerged as a significant concern. However, there is little empirical evidence to demonstrate how resilience may be enhanced. Green revolution technologies implicitly address resilience to climate variability, pest and disease outbreaks, and economic shocks through investments in improved water management, use of pesticides, and improved markets, capital accumulation, etc. There is an assumption that resilience might be better enhanced through promotion of extensive, low-input, highly biodi-verse agricultural systems (UNEP, 2009), but the empiri-cal evidence to support this hypothesis appears to be largely lacking. The objective of this paper is to explore the issues of when and how it might be appropriate to redirect investments toward enhanced resilience.

WHaT is “REsiLiENCE”?Many definitions of resilience exist (Brand and Jax, 2007). For the purposes of this paper, we will use the definitions adopted by Walker et al. (2004), “the capacity to absorb dis-turbance and reorganize while undergoing change so as to still retain essentially the same function, structure, identity, and feedbacks,” and Cumming et al. (2005, p. 976), “the ability of the system to maintain its identity in the face of internal change and external shocks and disturbances.”

These definitions imply that resilience is a desirable attribute. However, agricultural research for development is often addressing the needs of the extreme poor who are struggling to escape from agricultural systems that are highly resistant to change. In the context of poor devel-oping country farmers, the need is clearly to change but to do so in ways that do not increase exposure to risks. The challenge is therefore to progress to more produc-tive systems while at the same time retaining or increasing resilience to external shocks.

Walker et al. (2004) discuss a number of attributes of natural resource systems that influence resilience. The ones that are of most significance for agriculture are the following.

Thresholds and Tipping pointsAn essential feature of resilience is the existence of lim-its, or thresholds, beyond which significant change will

occur. If such change is of zero probability, then there is no fundamental problem for resource management. This is because such a system is always reversible within tech-nology and resource constraints (as in Fig. 1a). If a mis-take is made, or the managers change their minds, there is no fundamental obstacle in moving to another state of the system. In systems with nonlinear dynamics, however, the likelihood of alternate system regimes (alternate stable states) is high. A shift (intended or unintended) from one to the other can be irreversible or very hard to reverse.

Conventional natural resource management has tended to assume that ecosystems, agro-ecosystems, and social–ecological systems are predictable, controllable, and follow smooth trajectories (i.e., they don’t exhibit dis-continuous changes). Management has focused on aver-age conditions and on particular time and space scales. It has mostly ignored extreme events, and it has assumed that getting the system into some particular state and then keeping it there will maximize the flow of benefits.

However, many social–ecological systems exhibit threshold-type changes. If these thresholds are exceeded, changes in feedbacks will cause them to shift toward a dif-ferent state. Examples occur in agricultural, forestry, and fisheries systems, which do not recover after being changed by human or natural disturbances beyond some critical level. They may “break down” and remain in different, low-production states, even after human use is withdrawn.

Resilience is a feature of social and ecological systems and governance is clearly an important determinant of resilience. The resilience of governance systems is deter-mined largely by the attributes of networks, trust, human capital, leadership, etc. (Walker et al., 2004). A particular feature of threshold changes and recovery—hysteresis—is illustrated in Fig. 1.

The likelihood of alternate stable states is what makes the concept of resilience so important. The bigger the dif-ference between the levels of the two states, and the bigger the hysteresis effect (i.e., the more the controlling variable needs to be reversed before the state of the system “flips” back), the greater is the significance of that particular aspect of resilience.

specified and General ResilienceResilience is often seen as specific to an external driver of change; for instance, of a particular fish stock to fishing intensity, or of crop production to a drought (Carpen-ter et al., 2001). However, increased resilience to specific disturbances may cause the system to lose resilience to others. The “highly optimized tolerance” theory (Doyle and Carlson, 2000) shows how systems that become very robust to frequent kinds of disturbance necessarily become fragile in relation to infrequent kinds.

An important question is whether it is only the resilience of agricultural production (for example) that is of concern,


system defined by new state variables. It means intro-ducing new components and new ways of making a liv-ing, and often changes in the scales at which the system functions. This is the general problem that agricultural research for development is confronting.

Many production systems do not meet the needs of local communities. And some existing agricultural systems will not be viable under changed climate conditions; simple incremental adaptation will not suffice. These systems will need to be transformed into new kinds of agro-ecosystems. Such a transformational change may require that totally new germplasm, crops, farming systems, institutions, and policies are all put into place in a short space of time.

Resilience and transformability are both necessary attributes of systems. Building resilience to cope with external change is the appropriate action in some cir-cumstances. In others cases, incremental adaptations to a changing environment may amount to “digging the hole deeper.” The question facing policymakers will increas-ingly become: “Which parts of our (locality, region, or country), or which components or sectors, need enhanced resilience (to ensure their present, preferred states can continue), and which parts need to be transformed?” This is a fundamental societal choice and the legitimacy of the decision-making process is critical.

Changing to persistResilience requires that a system can change and should not be equated with resisting change. Keeping a system in some particular state may reduce its resilience. Allow-ing a system to change is necessary for maintaining the

or the resilience of broader attributes of livelihoods. Some specialized agricultural technologies or production systems may be less resilient to external challenges than the diverse production systems they replace. For instance, encouraging millions of small farmers in Africa to adopt hybrid seeds only available from a few distant producers and requiring high fertilizer inputs may greatly increase their incomes, but render them highly vulnerable to any external disrup-tions to the supply of agricultural inputs.

“General resilience” does not consider any particular kind of shock, or any particular aspect of the system that might be affected, and is, therefore, used both normatively and positively, implying that the general capacity to deal with shocks is deemed to be a good thing. The capacity of people and institutions to learn and adapt, and to self-organize and reorganize is critical to building resilience (Folke et al., 2003; Walker et al., 2004; Berkes and Seixas, 2005; Kooiman et al., 2005; Folke, 2006; Mahon et al., 2008). This capacity to respond to surprises is especially important in enabling man-agers to adapt (McLain and Lee, 1996). Building the capacity to adapt is therefore a key element of enhancing resilience. The concept of generalized resilience implies that the attri-butes that enable a system to cope with one kind of shock (e.g., a tsunami) are similar to those needed to respond to a different kind of shock (the global financial crisis).

Enhancing Resilience vs. TransformationWhen a society is trapped in an undesirable system regime and recovery to its former state, or movement to some new configuration of the system, is not possible, the only option is to transform into a different kind of system: a

Figure 1. The four possible responses of the equilibrium state of a system (here denoted by the state of a capital stock) to changes in an underlying, controlling variable. The equilibrium state is the amount of the capital stock that the system will eventually reach if the controlling variable is held constant at any particular level. (a) and (b) represent systems with no alternate equilibrium states; the response is smoothly changing in (a) and a step change in (b). The lateral arrows in (c) and (d) represent the direction of change. (c) and (d) involve hysteretic responses, where the return path is different from the development path, resulting in two possible stable (equilibrium) states for the same level of the controlling variable. (From Walker et al., 2009a).


resilience of the system’s current configuration. Change is also needed to shift the system to an alternate regime if that is desired; for example, in the crop–livestock systems in western Niger, from a low and declining state of soil fertility and crop production to a higher, self-maintaining state (Fernandez et al., 2002). Change is also needed to transform systems to different configurations when that is necessary (e.g., from a low-production livestock system to a new way of making a living).

Resilient systems are learning systems. Ecosystems adapt through exposure to shocks; for example, by the reorganization of species assemblages following a distur-bance. Social systems learn in multiple ways. Policy and management actions need to enable and foster learning. Learning requires providing safe spaces for experimenta-tion, and rewarding novelty and experiments, rather than having them prevented and penalized. This same need is explicitly recognized in the “safe arenas” concept in the field of transition theory and practice (Kemp et al., 2007).

Estimating or measuring ResilienceHow does one know if the resilience of the system is increasing or decreasing? For a well-defined threshold, such as water table depth and salinity, it may be possible to measure whether or not the state of the system (water table depth) is getting closer to the threshold and, therefore, whether resilience is declining. For others, such as a shift from a clean, high-diversity river system to one dominated by algal blooms and with low biodiversity, the position of the threshold may not be known, and managers will need to monitor changes in the attributes that likely determine the threshold, such as flow rates, inflow levels of pollutants,

abundance and diversity of fish and zooplankton species, and use these as indicators of changes in resilience.

If a threshold is known to exist, then it is important to learn about it. This is a difficult area for both science and management, but two approaches are worth considering. The first is development of a typology of thresholds with respect to the systems they are likely to occur in. A start has been made on developing a database for a very general framework (Walker and Meyers, 2004), but it calls for a wide research effort. A second, very pragmatic, approach is that in use by the Kruger National Park in South Africa. It involves identifying “thresholds of potential concern” (TPCs) based on available information and knowledge of related systems. The list is regularly revised and the top few TPCs are used to guide both research and manage-ment (Biggs and Rogers, 2003).

CoNsEQUENCEs oF a REsiLiENCE pERspECTiVE FoR NaTURaL REsoURCE maNaGEmENT

A number of recurring principles that are important in understanding resilience can be identified from com-parisons of resilience among social–ecological systems (Walker et al., 2006):

1. Allow systems to vary and to probe the boundaries of resilience.

Attempts to resist change reduce resilience. A common objective of policies aimed at optimizing some particular product or outcome is to identify an “optimal” state of the system, and then to somehow try to keep it in that particular state. Keeping a system in one particular state leads to changes that make the system less resilient. For instance, preventing fire in an attempt to keep a forest in its present state leads eventually to the loss of species that are fire tolerant. They are outcompeted by species that do not have to channel resources into thick bark, resistant cell structures, dormant stem buds that enable them to resist or recover from fire. The longer fire is prevented, the more flammable material accumulates and the more vulnerable the forest becomes to fire. To keep a forest resilient to fire, it is necessary to periodically allow the forest to burn. To keep a community, an organization, or a society resilient, it has to be exposed to subcritical levels of the kinds of disturbances to which it needs to be resilient.

2. Multiple scales and cross-scale effects.It is not possible to understand or manage a social–

ecological system by focusing only on the scale of primary interest. All systems are structured and function at multiple interconnected scales, and cross-scale effects strongly deter-mine the overall trajectory of the system as a whole—the concept of “panarchy” (see Holling et al., 2002). Resource managers tend to operate at a single scale—for instance, the farm or the forest—but building resilience at one scale can reduce resilience at other scales. In developing policy or

Figure 2. Ten interacting thresholds in the Goulburn-Broken Catchment (GB region) in South East Australia, at three scales and in three domains. The kind and magnitude of a shock will determine which threshold is most likely to be crossed, and the subsequent cascading effect through the system. Crossing one particular threshold may either increase or decrease the likelihood of another being crossed. (From Walker et al., 2009b).


management proposals, one needs to consider the broader context and the effects of changes at finer and greater scales. The so-called “Dutch disease” is a well-documented exam-ple of how macroeconomic changes driven by development based on oil and gas exploitation can have harmful impacts on other sectors of the economy (Wunder, 2003, 2008; Wunder and Sunderlin, 2004). There are numerous exam-ples of mining, infrastructure, and agro-industrial develop-ments having positive impacts on the economy at national scales, but having harmful impacts on the livelihoods of certain sectors of the population.

3. Multiple thresholds across scales and domains.In addition to cross-scale effects there are cross-domain

effects—interactions between the ecological, social, and economic domains. They are made especially difficult by the fact that the three domains function at different scales in both time and space. Threshold effects can occur at each scale and in each domain. As an example, Fig. 2 depicts 10 known or strongly suspected thresholds in the Goulburn-Broken Catchment, in South East Australia, at three spa-tial scales and in the three domains. The kind of shock the region experiences will determine which of the thresholds might be crossed. Crossing a particular threshold may then initiate a cascading effect in crossing other thresholds; and it may also lessen the likelihood of crossing certain others.

4. Controlling variables.Comparisons of resilience in several regions/sys-

tems suggest that, at any one scale, there are only three to five critical variables that determine the dynamics of the system (Walker et al., 2006). Identifying these criti-cal variables is fundamental to management. The Kruger National Park approach of iteratively identifying a priority list of “thresholds of potential concern” is an interesting application (Biggs and Rogers, 2003).

5. Pursuing narrowly defined efficiency reduces resilience.Efficiency is taken to be “good” in virtually all policy

developments. Where it really does only eliminate waste or redundancy, this is justified. But in many cases what is apparently redundant is actually “response diversity,” in resilience terms (Elmqvist et al., 2003). A farming system with many annual and perennial crops is more resilient to external fluctuations in weather, markets, input supplies, etc., than is a single, high-production commodity crop sys-tem. The pursuit of economic efficiency needs to be accom-panied by analysis of unintended resilience consequences.

oppoRTUNiTiEs aND CHaLLENGEsWe consider that the following seven challenges and opportunities will need to be addressed in understanding resilience of developing country agro-ecosystems.

1. Defining the system and providing context.A critical first step in any resilience assessment is to

define the system of concern. We need to be clear about the resilience “of what” and have an understanding of resilience

“to what.” What are the variables of concern? What is the normal disturbance regime of the system? What shocks, pressures, or internal changes is the system subject to? Even if the main focus is on the natural system, the social aspects of management responses strongly influence the dynam-ics of the linked social–ecological system. How we define the identity of a social–ecological system is important from both a technical and political point of view. Defining the identity of the system addresses not only the “of what,” “to what,” but, as well, the “for whom” questions (Carpenter et al., 2001; Lebel et al., 2006; Nadasdy, 2007). It requires perceiving and understanding the system as a linked system with strong interactions between the social and ecological domains, often across scales.

The Resilience Alliance Workbooks (http://www.resalliance.org/index.php?id=3874&sr=1&type=pop [ver-ified 26 Dec. 2009]) provide a list of questions to guide such assessments. For many applications a simpler list might be developed, but some analysis is necessary to clarify the controlling variables of the system, to prioritize issues both within and external to the system, and to identify a con-stituency and set of rights and institutions that “fit” the sys-tem (Young, 2002; Andrew et al., 2007; McClanahan et al., 2008; Evans and Andrew, 2009). There is a long history of developing and testing ecological and social methods for developing country contexts that would constitute the tools for the differential diagnosis of agro-ecosystems. Such methods are critical in the developing world because, in most instances, long-term resource-rich analyses of systems are neither possible nor desirable. Integration and adapta-tion of rapid participatory methods for resilience analysis is an area that needs further research effort.

Difficulties in defining “the system” can often be resolved by explicitly defining the spatial and time scales over which resilience is of concern. Fast variables at one scale are often slower variables at, and hence controlled by, the scale above. A closely related idea is the notion of lay-ered interventions. It calls for identifying the set of impor-tant barriers to advancing human wellbeing, and how and in what order to deal with them. It is not good enough to deal with only some of them. A single remaining barrier can prevent progress. Reducing or removing these barri-ers is equivalent to addressing the limiting factors to gen-eral resilience, and also transformability. It is necessary to encompass the whole system of problems to identify the key leverage points for change. Partial solutions do not work.

2. Thresholds and the importance of integrated natural resource management.

Thresholds in the behavior of complex systems are difficult to recognize and are most often “seen” after they have been crossed. In the institutionally weak, data-sparse world in which researchers operate in developing country contexts, this is the norm. Resilience management that seeks to keep a production system away from thresholds


needs to know something about where those thresholds might be (see Estimating or Measuring Resilience, above). The “rule of hand” [see Point 4, Controlling variables, above] suggests that there are three to five critical, con-trolling (often slowly changing) variables that determine the dynamics of the system at a given scale. Special atten-tion should be paid to thresholds in these controlling vari-ables that lead to changes in system behavior. Trying to identify controlling feedbacks is a useful way to approach the problem.

A threshold may be crossed as the result of an external shock (a tsunami or a civil war) or the cumulative effects of internal stresses (chopping down too many trees or catch-ing too many fish). Although many systems will at some point exhibit threshold changes in dynamics, it would be wrong to focus research and management attention only on identifying thresholds. Production systems can be made less vulnerable to the threat of external shocks without knowing when they will occur (building general resil-ience). Management within this domain has been well articulated as integrated natural resource management (Sayer and Campbell, 2001). As outlined earlier, integrated natural resource management shares many of the principles and concepts with resilience thinking and most of the field methods and analytical tools will be the same.

3. Values.As emphasized earlier, resilience per se is neither good

nor bad; it is a property of the current configuration of a system. It is critical not to conflate understanding of the resilience of some system configuration (value-free) with judgments about its desirability (value-based). The chal-lenge is to strengthen the capacity of society to manage resil-ience; to enhance it where appropriate or to erode it and help transform systems that are in undesirable states. The overall goal has to be to preserve the flow of economic, social, and environmental benefits to society as a whole.

Value judgments will always be needed and those judgments should be made by legitimate decision makers (Lebel et al., 2006; Nadasdy, 2007). Depending on one’s priorities and values, the current state of a system may or may not be desirable. Many undesirable social–ecologi-cal system configurations are highly resilient; for instance, forestry operations by military regimes and illegal fishing in the seas of developing countries.

“Resilience of whom?” (Lebel et al., 2006) is an ethical question and, except in the most egregious cases, legitimate but opposing perspectives may be held. In his critique of resilience thinking, Nadasdy (2007, p. 216) makes a further point: “the more one has invested (ecologically, socially, or economically) in existing social–ecological relations and institutions, the more one is likely to view resilience as ‘good’. Those who are marginalized or excluded are less likely to view a collapse of existing social/institutional structures as an unmitigated disaster. Indeed, they may even

embrace the kind of radical socio-ecological change brought about by a system shift. The valorisation of resilience, then, represents a decision—at least implicitly—to endorse the socio-ecological status quo.” In essence, Nadasdy makes the case for resilience as a value-free proposition for analysis. We note, however, that the poorest and weakest are likely to fare worse in the transition.

4. Reconciling “sustainable development” and “resilience”.Sustainable development is a societal goal that the

world has adopted. Some definitions emphasize stability and control—of the environment, society, or the econ-omy—using institutions of governance. Since a resilience perspective is counter to this (it assumes that responses of ecosystems to human use have limited predictability and control), it might be seen as opposed to conventional sus-tainable development. In some ways it is, where sustainable development invokes goals of equilibrium and optimiza-tion as embodied in metrics such as maximum sustainable yield. However, as posited by Lebel et al. (2006), resilience should rather be regarded as a necessary system property for sustainability in the face of change and uncertainty, furthering our endeavor to achieve sustainable develop-ment rather than challenging it. They assert that strength-ening the capacity of societies to manage resilience is critical to effectively pursuing sustainable development.

It is useful to distinguish between “resilience,” the system property, “resilience based-management,” and “resilience-based development”—that is, designing a development strategy that leads to the maintenance or enhancement of resilience. We deal with this last aspect of resilience in the following point.

5. Development pathways and path dependence.Some development pathways will likely lead to greater

resilience than others. For instance, one might argue that the Millennium Ecosystem Assessment “Adapting Mosaic” sce-nario would be more resilient than the “Global Orchestra-tion” scenario (Millennium Ecosystem Assessment, 2005), and that, if so, there should be more attention to research that would favor this scenario. Similarly the International Assessment of Agricultural Knowledge, Science and Tech-nology for Development (IAASTD) provides analysis that supports agricultural development pathways that are locally adapted and less reliant on outside inputs of technology or agro-chemicals (the latter derived from declining fos-sil fuels) (IAASTD, 2009a, 2009b). A resilience approach suggests that there should be more research focused on these multiple precision agricultural models rather than on conventional specialized models centered on a very small number of crops and valuing economies of scale, standard-ization, and specialization.

6. Transformation.“Transformational change” is much needed to meet

the food security challenges of the developing world. In reality, transformation is a tricky ethical arena (Olsson et


al., 2005, 2008; Kristjanson et al., 2009). The questions facing managers (broadly defined) with respect to the transformation of production systems include:

· Which parts of the (locality, region, country), or which components or sectors, need enhanced resil-ience to ensure that their present states can continue, and which parts need to be transformed?

· If a production system is to be transformed, who decides what the changes will be? Trying to determine the “best” new system can lead into the same kind of trap the social–ecological system is currently in, and it may be better to allow for a learning approach within a range of acceptable new trajectories; that is, decide on where NOT to go, avoid those pathways, and allow self-orga-nization within the range of acceptable futures.

· Transformation will favor some people over others—Who will lose and who will be the winners? What process will ensure the legitimacy of the decisions that lead to such redistributions of wealth and influence? Do scientists have any legitimate role in this process?

· The transformation process may be chaotic and unpre-dictable, throwing up new actors seeking advantage, creating new, visionary leaders that catalyze societal change in good ways, causing unexpected ecologi-cal phase shifts, etc. What responsibility do agents of change have for transformations that make things worse in the fight against poverty?

· Can managers capitalize on windows of opportunity to create the sense of urgency needed to overcome resistance to change (Olsson et al., 2005, 2008)?

· Transformations of production systems will prompt dis-cussions about the tradeoffs between different sorts of landscapes. The currency used in such discussions will likely be in terms of ecosystem services (including bio-diversity conservation). Very quickly this will lead to issues of ecosystem valuation and the problems of non-monetary valuation of, for example, rivers and forests.

7. Governance—partnerships, networks, and forums.Exclusive, centralized forms of management have failed

to deliver sustainable and equitable use of natural resources in the developing world (Berkes, 2003; Charles, 2001; Var-jopuro et al., 2008). Inclusion of a diverse, but appropriate set of stakeholders will include better problem definition and ownership, a more diverse knowledge base for decision mak-ing, greater legitimacy, and, therefore, better compliance and commitment to agreed courses of action, and conflict resolu-tion (Jentoft, 2000; Bryan, 2004). Recent work on so-called boundary processes—individuals and organizations—has been identified in the success (or lack of success) in many developing country situations. Their success in enabling cross-scale and cross-institution communication and coop-eration depends in all cases on their being identified by all players as accountable and trusted (Guston, 2001; Carr and Wilkinson, 2005; McNie, 2007; Kristjanson et al., 2009).

A resilience approach clearly anticipates or leads many of these trends, and so the types and modalities of research required to support resource management are also chang-ing. Creation of new knowledge will remain a corner-stone activity, but increasingly the role of research may be to understand the processes and necessary conditions for transformational change. Research may more explic-itly seek to build general resilience. Concrete examples of these new modalities may include:

· Supporting the creation of national and regional forums to take leadership of management change and to set the research agenda. In fisheries, for example, the African Union’s New Partnership for Africa’s Development (NEPAD) has articulated the AU-NEPAD Action Plan for the Development of Afri-can Fisheries and Aquaculture, which has become a key component of NEPAD’s Comprehensive Afri-can Agriculture Program.

· Facilitate learning networks that encourage local insti-tutions to become learning organizations to build resilience. For example, in the Greater Mekong region, the Wetlands Alliance (www.wetlandsal-liance.org [verified 26 Dec. 2009]), a network of >30 organizations, works with “dialog partners” to address institutional aspects of poverty through capacity building for wetlands management.

· The current interest in landscape and ecosystem approaches to fisheries, forest, and agro-ecological system problems are implicitly driven by consider-ations of resilience. They seek a more balanced and sustainable approach to productivity enhancement and address the flows of multiple benefits.

An overall conclusion in regard to governance in developing world social–ecological systems is the need for decentralization and devolution of power. Centralized control leads to frequent inappropriate actions through application of one-size-fits-all policies, and because it involves long feedback times it does not match the speed at which decisions need to be made. The model of polycen-tricity and distributive governance (e.g., Marshall, 2009) is more in line with developing country needs.

CoNCLUsioNsThere are few examples of resilience thinking being for-mally incorporated into the natural resource manage-ment programs of developing countries. A comparison of resilience in some 15 social–ecological systems around the world led to the identification of 10 guidelines that might be applied in the agricultural and natural resources management programs of these countries (Anderies et al., 2006). Several of these have profound implications for the way in which scientists approach agricultural and natural resources research. They also have profound implications


for the management of natural resources in the developing world, and we advocate their wider adoption. They are:

· Neither ecosystems nor social systems can be managed in isolation. Their strong interactions and multiple feedbacks must be taken into account.

· Managers must intervene at multiple scales, under-stand how the focal scale interacts with other scales, what is happening in the levels above and below, and what effects cross-scale processes are likely to exert.

· Slow variables need to be understood. Identifying the key controlling variables with threshold effects that determine alternate system regimes is important. There are typically no more than a few such key variables that are important at any one scale.

· Manage for diversity. Simplifying production, ecologi-cal, or social systems for increased efficiency carries with it a reduction in response diversity, so that the sys-tem becomes more vulnerable to stresses and shocks.

· Accept that maintaining resilience incurs costs. There may be a tradeoff between short-term benefits from high efficiency under narrowly constrained circum-stances and the long-term performance of a more resil-ient regime with reduced costs of crisis management.

· Make strategic interventions. Focus on identifying the key points for intervention in the social–ecologi-cal system that can avoid undesirable pathways and alternate regimes. Successful intervention requires investment in adaptive capacity.

· Understand underlying mental models. Successful outcomes depend on expanding and connecting the mental models that exist across the stakeholder groups so as to increase their mutual understanding and thereby the social system’s capacity to act.

· Embrace adaptive governance. Introduce flexible, dynamic institutional and governance structures so that key intervention points can be addressed at the appropriate scales and times.

· Recognize windows for transformation. If a system has already moved onto an undesirable trajectory that is unacceptable and efforts to move off it are failing, there comes a point at which adaptation is no longer ecologically, socially, or economically feasible. When transformation is the only option, the sooner it is rec-ognized and acted on, the lower the transition costs and the higher the likelihood of success.

· Recognize that vulnerability cannot be eliminated. Strategies that enhance robustness to particular types of shocks necessarily give rise to new vulnerabilities in other domains.

Our overall conclusion is that the primary goal in inter-national agricultural research is to shift people out of their highly resilient condition of poverty into a more productive condition as defined by a broad set of livelihood attributes without making them vulnerable to external shocks such

as those caused by climate variability, economic volatility, pandemics, etc. In some circumstances, resilience may hin-der escape from a poverty trap to a more desirable state. Where resilience is an obstacle to change, then transforma-tion has to be actively sought. This means moving to a dif-ferent kind of system, defined by different variables, with a different way of making a living. Enhancing transformabil-ity is a major need in the developing world. The literature on resilience could enhance the ability of science to bring solutions to the needs of the rural poor without exposing them to some of the risks that may result from overly simple solutions focusing solely on yield increases and efficiency. We therefore advocate the broader adoption of systems approaches to agricultural research. Such approaches must be based on a thorough understanding of the context within which farming, fishing, and forestry take place, a process of continuous experimentation and learning that involves producers working alongside scientists and the integra-tion of many knowledge systems. It requires that scientists take into account the broad set of attributes of the system that ultimately determine the livelihoods of the rural poor (Sayer and Campbell, 2004).

aCKNoWLEDGmENTsThis paper is based on a background paper prepared for a work-shop at the Science Forum 2009 of the Consultative Group on International Agricultural Research (CGIAR). We are grateful to the panelists and speakers in the workshop for their insights: Patrick Caron, William Clarke, Sir Gordon Conway, Den-nis Garitty, Line Gordon, Hannah Jaenicke, Peter Mollinga, Richard Munang, Meine van Noordwijk, Thomas Rosswall, Marten Scheffer, and Mark Smith. We also thank the numerous participants who engaged in the lively discussion.

REFERENCEsAnderies, J.M., B.H. Walker, and A.P. Kinzig. 2006. Fifteen wed-

dings and a funeral: Case studies and resilience-based man-agement. Ecol. Soc. 11(1): Article 21.

Andrew, N., C. Béné, S.J. Hall, E.H. Allison, S. Heck, and B.D. Ratner. 2007. Diagnosis and management of small-scale fish-eries in developing countries. Fish Fish. 8:227–240.

Berkes, F. 2003. Alternatives to conventional management: Les-sons from small-scale fisheries. Environments 31:1–14.

Berkes, F., and C.S. Seixas. 2005. Building resilience in lagoon social–ecological systems: A local-level perspective. Ecosys-tems 8:967–974.

Biggs, H.C., and K.H. Rogers. 2003. An adaptive system to link science, monitoring, and management in practice. p. 59–80. In J.H. Du Toit et al. (ed.) The Kruger experience. Island Press, Washington, DC.

Brand, F.S., and K. Jax. 2007. Focusing the meaning(s) of resil-ience: Resilience as a descriptive concept and a boundary object. Ecol. Soc. 12(1): Article 23.

Bryan, T.A. 2004. Tragedy averted: The promise of collaboration. Soc. Nat. Resour. 17:881–896.

Carpenter, S.J., B. Walker, J.M. Anderies, and N. Abel. 2001. From metaphor to measurement: Resilience of what to what? Ecosystems 4:765–781.

Carr, A.J.L., and R.L. Wilkinson. 2005. Beyond participation:


Boundary organizations as a new space for farmers and scien-tists to interact. Soc. Nat. Resour. 18:255–265.

Chambers, R. 1997. Whose reality counts? Putting the first last. Intermediate Technol., London.

Charles, A.T. 2001. Sustainable fishery systems. Blackwell Sci., London.Conway, G. 1997. The doubly green revolution. Cornell Univ.

Press, Ithaca, NY.Cumming, G.S., G. Barnes, S. Perz, M. Schmink, K.E. Sieving,

J. Southworth, M. Binford, R.D. Holt, C. Stickler, and T. Van Holt. 2005. An exploratory framework for the empirical measurement of resilience. Ecosystems 8:975–987.

Doyle, J., and J. Carlson. 2000. Power laws, highly optimized tolerance, and generalized source coding. Phys. Rev. Lett. 84:5656–5659.

Elmqvist, T., C. Folke, M. Nystrom, G. Peterson, J. Bengston, B. Walker, and J. Norberg. 2003. Response diversity, ecosystem change and resilience. Front. Ecol. Environ. 1:488–494.

Evans, L., and N. Andrew. 2009. Diagnosis and the management constituency of small-scale fisheries. The WorldFish Ctr. Working Pap. 1941. Available at http://www.worldfishcenter.org/resource_centre/WF_2416.pdf (verified 23 Dec. 2009). The WorldFish Ctr., Penang, Malaysia.

FAO. 2008. The state of food and agriculture 2008. Available at http://www.fao.org/docrep/011/i0100e/i0100e00.htm (verified 23 Dec. 2009). Food and Agric. Organ. of the United Nations, Rome.

Fernandez, R.F., E.R.M. Archer, A.J. Ash, H. Dowlatabadi, P.H.Y. Hiernaux, J.F. Reynolds, C.H. Vogel, B.H. Walker, and T.W. Iegand. 2002. Degradation and recovery in socio-ecological systems: A view from the household/farm level. p. 297–323. In J.F. Reynolds and D.M. Stafford Smith (ed.) Global desertifica-tion. Do humans cause deserts? Dahlem Univ Press., Berlin.

Folke, C. 2006. Resilience: The emergence of a perspective for social–ecological systems analyses. Global Environ. Change 16:253–267.

Folke, C., J. Colding, and F. Berkes. 2003. Synthesis: Building resilience and adaptive capacity in social–ecological systems. p. 352–387. In F. Berkes et al. (ed.) Navigating social–ecologi-cal systems: Building resilience for complexity and change. Cambridge Univ. Press, Cambridge, UK.

Guston, D.H. 2001. Boundary organizations in environmental policy and science: An introduction. Sci. Technol. Hum. Val-ues 26:399–408.

Holling, C.S., L.H. Gunderson, and G.D. Peterson. 2002. Sus-tainability and panarchies. p. 63–102. In L.H. Gunderson and C.S. Holling (ed.) Panarchy: Understanding transformations in human and natural systems. Island Press, Washington, DC.

IAASTD. 2009a. A synthesis of the Global and Sub-global IAASTD reports. Available at http://www.agassessment.org/index.cfm?Page=IAASTD%20Reports&ItemID=2713 (veri-fied 23 Dec. 2009). Island Press, Washington, DC.

IAASTD. 2009b. Towards multifunctional agriculture for social, environmental and economic sustainability. Policy issue brief. Available at http://www.agassessment.org/index.cfm?Page=IAASTD%20Reports&ItemID=2713 (verified 23 Dec. 2009). Island Press, Washington, DC.

Jentoft, S. 2000. Legitimacy and disappointment in fisheries man-agement. Mar. Pollut. 24:141–148.

Kemp, R., D. Loorbach, and J. Rotmans. 2007. Transition manage-ment as a model for managing processes of co-evolution towards sustainable development. Int. J. Sust. Devel. World Ecol. 14:1–15.

Kooiman, J., M. Bavinck, S. Jentoft, and R. Pullin (ed.) 2005. Fish for life: Interactive governance for fisheries. Amsterdam Univ. Press, Amsterdam.

Kristjanson, P., R.S. Reid, N. Dickson, W.C. Clark, D. Rom-ney, R. Puskur, S. MacMillan, and D. Grace. 2009. Linking

international agricultural research knowledge with action for sustainable development. Proc. Natl. Acad. Sci. USA 106:5047–5052.

Lebel, L., J.M. Anderies, B. Campbell, C. Folke, S. Hatfield-Dodds, T.P. Hughes, and J. Wilson. 2006. Governance and the capacity to manage resilience in regional social–ecological systems. Ecol. Soc. 11: Article 19.

Mahon, R., P. McConney, and R.N. Roy. 2008. Governing fish-eries as complex adaptive systems. Mar. Pollut. 32:104–112.

Marshall, G.R. 2009. Polycentricity, reciprocity, and farmer adop-tion of conservation practices under community-based gover-nance. Ecol. Econ. 68:1507–1520.

McClanahan, T.R., J.C. Castilla, A.T. White, and O. Defeo. 2008. Healing small-scale fisheries by facilitating complex socio-ecological systems. Rev. Fish Biol. Fish. 19:33–47.

McLain, R.J., and R.G. Lee. 1996. Adaptive management: Prom-ises and pitfalls. Environ. Manage. 20:437–448.

McNeely, J.A., and S.J. Scherr. 2003. Ecoagriculture: Strategies for feeding the world and conserving wild biodiversity. Island Press, Washington, DC.

McNie, E.C. 2007. Reconciling the supply of scientific infor-mation with user demands: An analysis of the problem and review of the literature. Environ. Sci. Pollut. 10:17–38.

Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being. Vol. 2: Scenarios. Available at http://www.mil-lenniumassessment.org/en/Scenarios.aspx (verified 4 Jan. 2009). Island Press, Washington, DC.

Nadasdy, P. 2007. Adaptive co-management and the gospel of resil-ience. p. 208–226. In D. Armitage et al. (ed.) Adaptive co-man-agement: Collaboration, learning and multi-level governance. Univ. of British Columbia Press, Vancouver, Canada.

Olsson, P., S. Carpenter, L. Gunderson, P. Ryan, L. Lebel, C. Folke, and C.S. Holling. 2005. Shooting the rapids: Navigating transi-tions to adaptive ecosystem governance. Ecol. Soc. 11: Article 18.

Olsson, P., C. Folke, and T.P. Hughes. 2008. Navigating the tran-sition to ecosystem-based management of the Great Barrier Reef, Australia. Proc. Natl. Acad. Sci. USA 105:9489–9494.

Pollan, M. 2006. The omnivore’s dilemma. Penguin Books, New York.

Ronald, P.C., and R.W. Adamchak. 2008. Tomorrow’s table: Organic farming, genetics and the future of food. Oxford Univ. Press, New York.

Sayer, J.A., and B. Campbell. 2001. Research to integrate produc-tivity enhancement, environmental protection and human development. Conserv. Ecol. 5: Article 32.

Sayer, J.A., and B. Campbell. 2004. The science of sustainable development; local livelihoods and the global environment. Cambridge Univ. Press, Cambridge, UK.

UNEP. 2009. The environmental food crisis—The environment’s role in averting future food crises. A UNEP rapid response assessment. Available at http://new.unep.org/pdf/FoodCrisis_lores.pdf (veri-fied 23 Dec. 2009). United Nations Environ. Progr., Nairobi.

Varjopuro, R., T. Gray, J. Hatchard, F. Rauschmayer, and H. Wit-tmer. 2008. Introduction: Interaction between environment and fisheries and the role of stakeholder participation. Mar. Pollut. 32:158–168.

von Braun, J., D. Byerlee, C. Chartres, T. Lumpkin, N. Olembo, and J. Waage. 2009. Towards a strategy and results framework for the CGIAR: Draft report by the Strategy Team 21 Octo-ber 2009. Available at http://www.cgiar.org/exco/exco17/exco17_srf&mps.pdf (verified 4 Jan. 2009).

Walker, B., L. Gunderson, A. Kinzig, C. Folke, S. Carpenter, and L. Schultz. 2006. A handful of heuristics and some propo-sitions for understanding resilience in social–ecological sys-tems. Ecol. Soc. 11(1): Article 13.

Walker, B., and J.A. Meyers. 2004. Thresholds in ecological and


social–ecological systems: A developing database. Ecol. Soc. 9(2): Article 3.

Walker, B., L. Pearson, M. Harris, K.-G. Maler, C.-Z. Li, R. Biggs, and T. Baynes. 2009a. Incorporating resilience in the assess-ment of inclusive wealth: An example from South East Austra-lia. Environ. Resour. Econ. doi: 10.1007/s10640-009-9311-7.

Walker, B.H., N. Abel, J.M. Anderies, and P. Ryan. 2009b. Resil-ience, adaptability, and transformability in the Goulburn-Broken Catchment, Australia. Ecol. Soc. 14(1): Article 12.

Walker, B.H., C.S. Holling, S.C. Carpenter, and A.P. Kinzig. 2004. Resilience, adaptability and transformability. Ecol. Soc. 9(2): Article 5.

Wunder, S. 2003. Oil wealth and the fate of the forest: A compara-tive study of eight tropical countries. Routledge, Taylor & Francis Group, New York.

Wunder, S. 2008. Oil wealth and the fate of the forest: Linking resource booms, macroeconomics, and the environment. Routledge, London.

Wunder, S., and W.D. Sunderlin. 2004. Oil, macroeconomics, and forests: Assessing the linkages. World Bank Res. Observ. 19:231–257.

Young, R.O. 2002. The institutional dimensions of environmen-tal change: Fit, interplay, and scale. MIT Press, London.


SympoSia

Globally, malnutrition, including both overt nutrient defi-ciencies as well as diet-related chronic diseases (e.g., heart

disease, cancer, stroke, and diabetes), is responsible for more deaths than any other cause, accounting for >20 million mortalities annually (Kennedy et al., 2003; WHO and FAO, 2003). Malnu-trition also contributes to increased morbidity, disability, stunted mental and physical growth, and reduced national socioeconomic development (WHO and FAO, 2003). Micronutrient malnutri-tion alone afflicts more than two billion people, mostly among resource-poor families in developing countries, with Fe, I, Zn, and vitamin A deficiencies most prevalent (Kennedy et al., 2003). More than five million childhood deaths occur from micronutri-ent malnutrition every year (Anonymous, 2007). Leading global economists have identified investing in strategies to reduce mal-nutrition as the most cost-effective investments governments can make (Anonymous, 2008).

What causes malnutrition? Dysfunctional food systems that cannot supply all the nutrients and health-promoting factors

Biofortification—A Sustainable Agricultural Strategy for Reducing Micronutrient

Malnutrition in the Global South

Howarth E. Bouis and Ross M. Welch*

aBSTRaCTMinerals and vitamins in food staples eatenwidely by the poor may be increased eitherthroughconventionalplantbreedingor throughuseoftransgenictechniques,aprocessknownasbiofortification.HarvestPlusseekstodevelopand distribute cultivars of food staples (rice[Oryza sativa L.], wheat [Triticum aestivum L.],maize [Zea mays L.], cassava [Manihot escu-lentaCrantz],pearlmillet [Pennisetum america-numLeeke],beans[Phaseolus vulgarisL.],sweetpotato[Ipomoea batatasL.])thatarehighinFe,Zn,andprovitaminAthroughaninterdisciplinaryglobalallianceofscientificinstitutionsandimple-mentingagencies indevelopinganddevelopedcountries.Biofortifiedcropsoffer a rural-basedintervention that, by design, initially reachesthesemoreremotepopulations,whichcompriseamajorityoftheundernourishedinmanycoun-tries,and thenpenetrates tourbanpopulationsas production surpluses are marketed. Thus,biofortification complements fortification andsupplementationprograms,whichworkbest incentralizedurbanareasandthenreachintoruralareaswithgoodinfrastructure.Initialinvestmentsinagriculturalresearchatacentrallocationcangeneratehigh recurrentbenefitsat lowcostasadapted biofortified cultivars become widelyavailableincountriesacrosstimeatlowrecurrentcosts.Overall,threethingsmusthappenforbio-fortificationtobesuccessful.First,thebreedingmustbesuccessful—highnutrientdensitymustbecombinedwithhighyieldsandhighprofitabil-ity.Second,efficacymustbedemonstrated—themicronutrientstatusofhumansubjectsmustbeshowntoimprovewhenconsumingthebioforti-fied cultivars as normally eaten. Third, the bio-fortifiedcropsmustbeadoptedbyfarmersandconsumedbythosesufferingfrommicronutrientmalnutritioninsignificantnumbers.

H.E. Bouis, HarvestPlus, c/o International Food Policy Research Insti-tute, 2033 K St. NW, Washington, DC 20006; R.M. Welch, USDA-ARS, Robert W. Holley Center for Agriculture and Health, Tower Rd., Ithaca, NY 14853-2901. Received 23 Sept. 2009. *Corresponding author ([email protected]).

Abbreviations: CGIAR, Consultative Group on International Agricultural Research; CIAT, International Center for Tropical Agriculture; IBM, intermated B73 ´ Mol7; IFPRI, International Food Policy Research Institute; LPS, lipopolysaccharides; NILs, near-iso-genic lines; PAC, Program Advisory Committee; QTL, quantitative trait loci; RI, recombinant inbred.




required for human life in sustainable ways are responsible. However, food systems that feed the disadvantaged are very complex (Sobal et al., 1998). Therefore, dysfunctions in numerous interacting factors can result in inadequate sup-plies of nutrients reaching the most vulnerable populations (World Bank, 2007). Importantly, because food systems are dependent on agricultural products as their source of most nutrients, agricultural systems must be contributing to this worldwide quandary in public health (Welch, 2001).

Unfortunately, agricultural systems have never been explicitly designed to promote human health and, instead, mostly focus on increased profitability for farmers and agricultural industries. Agriculture met the challenge of feeding the world’s poor during the “Green Revolution,” focusing primarily on three staple crops—rice (Oryza sativa L.), wheat (Triticum aestivum L.), and maize (Zea mays L.). These crops provided enough energy to prevent widespread famines in many developing nations. An unforeseen conse-quence of that agricultural revolution was the rapid rise in micronutrient malnutrition in many nations that adopted the cropping systems that prevented large-scale starvation (Welch and Graham, 1999). Agriculture must now formu-late new policies that not only provide enough calories to meet the energy needs of the poor but also deliver all the essential nutrients needed for adequate nutritional health.

Sustainable solutions to malnutrition will only be found by closely linking agriculture to nutrition and health and by formulating agriculture, nutrition, and health poli-cies to reflect this need (Graham et al., 2007; Hawkes and Ruel, 2006; Rouse and Davis, 2004; World Bank, 2007). It is shortsighted if the world once again focuses only on delivering the energy needs of resource-poor people dur-ing the current food crisis (Casey and Lugar, 2008; Zaro-costas, 2009) without also giving those affected the crops and other agricultural products needed for adequate nutri-tion required for healthy and productive lives.

Humans require at least 44 known nutrients in adequate amounts and consistently to live healthy and productive lives (Table 1). Many agricultural tools (e.g., diversification, crop selection, fertilizers, cropping systems, soil amend-ments, small livestock production, aquaculture, etc.) could be used to increase the nutrient output of farming systems (Graham et al., 2007). Biofortification (developing food crops that fortify themselves) is the first agricultural tool now being employed to address micronutrient malnutrition worldwide. Conventional breeding has been the primary focus of programs to enhance staple food crops with suf-ficient levels of Fe, Zn, and provitamin A carotenoids to meet the needs of at-risk populations in the Global South (Hotz et al., 2007; White and Broadley, 2009).

The biofortification strategy is a feasible means of reaching rural families that only have limited access to markets and healthcare facilities needed to provide for-tified foods and nutritional supplements because it is

targeted at this population. Once implemented, bioforti-fication will lower the number of micronutrient-deficient people requiring interventions dependent on supplemen-tation and fortification programs (see Fig. 1). Thus, bio-fortification complements other interventions and is a means to provide micronutrients to the most vulnerable people in a comparatively inexpensive and cost-effective way, using an agricultural intervention that is sustainable (Bouis, 1999; Nestel et al., 2006; Pfeiffer and McClafferty, 2007; Qaim et al., 2007).

HarvestPlus is the CGIAR’s Biofortification Chal-lenge program. It is directed at using plant breeding as an intervention strategy to address micronutrient malnutri-tion by producing staple food crops with enhanced levels of bioavailable essential minerals and vitamins that will have measurable impact on improving the micronutrient status of target populations, primarily resource-poor peo-ple in the developing world. Impressive progress has been made at meeting the goals of the HarvestPlus program set forth at its inception in 2003, but much remains to be done (Bouis et al., 2009).

Three primary issues have been identified that are required to make biofortification successful: (i) a biofor-tified crop must be high yielding and profitable to the farmer, (ii) the biofortified crop must be shown to be effi-cacious and effective at reducing micronutrient malnutri-tion in humans, and (iii) the biofortified crop must be acceptable to both farmers and consumers in target regions where people are afflicted with micronutrient malnutri-tion. The HarvestPlus program has addressed all of these issues (Hotz et al., 2007). This program has been able to assemble a multi-CGIAR Centers team along with col-laborators from numerous universities, nongovernmen-tal organizations, in-country agencies, and international institutions comprising plant scientists, plant breeders, food scientists, nutritionists, economists, and commu-nication and behavioral specialists to tackle these issues. The program model developed by HarvestPlus has been successful in developing transdisciplinary team-research programs among CGIAR Centers and across diverse disciplines (see Web site at http://www.harvestplus.org/ [verified 22 Dec. 2009]).

ConvenTional BReeding To BiofoRTify STaple food CRopS

The task of plant breeders attempting to biofortify staple food crops is to increase the micronutrient level in the edible product of a staple food crop to have measurable impact on improving the nutritional health of individu-als at high risk of developing micronutrient malnutrition. For this to be accomplished, plant breeders must work closely with food scientists and nutritionists to develop target micronutrient levels for their breeding programs. Considerations must include not only micronutrient


trials will be formally submitted to the Varietal Release Committees for further testing and, once approved, will be officially released within the target country. This pro-cess may take up to 8 yr to complete. Once implemented, baseline nutritional studies will be compared to postdis-semination impact and effectiveness studies in both control and intervention locations to establish if biofortified crops can improve the micronutrient status of people in target populations. To facilitate seed dissemination, market chain analysis, production capacity for seed increases, consumer acceptance studies, and development of a favorable policy environment for the production of biofortified crops will also be required for successful and sustainable implementa-tion of the biofortification strategy.

USing feRTilizeRS To enhanCe miCRonUTRienT elemenTS in STaple food CRopS

Both macronutrient fertilizers containing N, P, K, and S, and certain micronutrient fertilizers (e.g., Zn, Ni, I, Co, Mo, and Se) can have significant effects on the

concentrations in the edible portions of crops, but also the amount of the nutrient that can be absorbed by the consumer, after processing and cooking, when eaten in a traditional diet for the target population. This can be a difficult task. Numerous genes may be involved in con-trolling the amount of a mineral element that is absorbed by roots, translocated to shoots, remobilized from vegeta-tive tissues, and deposited in edible portions of seeds and grains in forms that are utilizable in the person eating the crop (Welch, 1986, 1995). Further, environmental factors and cultural practices (e.g., edaphic, climatic, agronomic, etc.) can interact with plant-gene expression to influence the amount of a micronutrient accumulated in a seed or storage organ. Additionally, various dietary factors can interact to determine how much of a micronutrient can be absorbed and utilized by people eating the biofortified staple plant food (i.e., the bioavailable amount) (Hotz et al., 2007; Ortiz-Monasterio et al., 2007; Welch, 2001).

The HarvestPlus program has set needed levels for Fe, Zn, and provitamin A carotenoids in target crops after addressing these issues. Table 2 list these target levels and assumptions used to set levels for target populations in the developing world (Bouis et al., 2009). These target levels are very conservative estimates and are estimates and will be changed if deemed necessary as new data and informa-tion merits adjustment. Figure 2 and Table 3 summarize the progress being made in the HarvestPlus program to develop biofortified crops. Once high-yielding bioforti-fied crop cultivars are developed that meet target nutrient levels, they will be disseminated widely.

HarvestPlus will disseminate the biofortified seeds through established partnerships with country agencies for delivering biofortified seeds to farmers and, ultimately, to the consumer. The HarvestPlus program will do this in several stages. First, national agricultural research and extension programs will multiply the seeds and test the biofortified lines at multiple locations in trials throughout the target country to determine environmental ´ genetic interactions on expression of the high-micronutrient traits in the biofortified crops. Selected promising lines from these

Table 1. The known essential nutrients for human life†.

Air, water, and energy Protein (amino acids) Lipids–Fat (fatty acids) Macrominerals Essential trace elements VitaminsOxygenWater

Carbohydrates

HistidineIsoleucineLeucineLysine

MethioninePhenylalanine

ThreonineTryptophan

Valine

Linoleic acidLinolenic acid

NaK

CaMgSPCl

FeZnCuMn

IF

SeMo

Co (in B12)B

A (retinol)D (calciferol)

E (a-tocopherol)K (phylloquinone)C (ascorbic acid)

B1 (thiamin)B2 (riboflavin)

B3 (niacin)B5 (pantothenic acid)

B6 (pyroxidine)B7 (biotin)

B9 (folic acid, folacin)B12 (cobalamin)

†Numerous other beneficial substances in foods are also known to contribute to good health.

Figure 1. Frequency distribution of Fe adequacy in a population. Biofortification improves status for those less deficient and maintains status for all at low cost. Iron adequacy for a population is indicated as 12.0 mg dL−1 on the plot. Biofortification will shift the population into a more Fe-sufficient range.


accumulation of nutrients in edible plant products (Alla-way, 1986; Grunes and Allaway, 1985). Other micronu-trient fertilizers have very little effect on the amount of the micronutrient accumulated in edible seeds and grains when they are applied to soils or when used as foliar sprays (Welch, 1986). This is especially true for those micronu-trient elements with limited phloem sap mobility such as Fe (Welch, 1999). Some examples of the effects of fertilizer practices on the micronutrient concentrations in edible plant parts are given below. For more detailed informa-tion concerning the effects of fertilization practices on micronutrient accumulation in plant foods, refer to R.M. Welch’s “Importance of seed mineral nutrient reserves in crop growth and development” (Welch, 2001).

For certain essential micronutrient elements (e.g., Zn, Ni, I, and Se), increasing soil-available supply to food crops can result in significant increases in their concentrations in edible plant products (Graham et al., 2007; Welch, 1995).

For example, increasing the supply of Zn to pea (Pisum sati-vum L.) plants at levels in excess of that required for maxi-mum yield has been shown to increase the concentration of bioavailable Zn in seeds (Peck et al., 1980; Welch et al., 1974). Furthermore, increasing the supply of Zn and Se to wheat improved the amount of bioavailable Zn and Se in wheat grain (Cakmak, 2008; Haug et al., 2008; House and Welch, 1989). Increasing Zn levels via Zn fertilization has also been shown for navy beans (Phaseolus vulgaris L.), as well as other crops (Moraghan, 1980; Welch, 1986). For Fe, providing more to plants than required for maximum yield does little to further increase the Fe in edible seeds and grains. Interestingly, the micronutrient I, supplied in irrigation water, can greatly increase the levels of I in edible portions of food crops, alleviating the debilitating disease cretinism, as well as other I-deficiency disorders in populations dependent on irrigated food crops grown on low-I soils (Cao et al., 1994; Ren et al., 2008). In Finland,

Table 2. Information and assumptions used to set target levels for micronutrient content of biofortified staple food crops.

Amount eaten or nutrient Criteria

Rice (polished)

Wheat (whole)

Pearl millet (whole)

Beans (whole)

Maize (whole)

Cassava (fresh wt.)

Sweet potato (fresh wt.)

Per capita consumption Adult women (g/d) 400 400 300 200 400 400 200Children 4–6 yr (g/d) 200 200 150 100 200 200 100

Fe % of EAR† to achieve: 30

EAR, nonpregnant, nonlactating women (µg/day)

1460

EAR, children 4–6 yr (µg/d) 500

Micronutrient retention after processing (%)

90 90 90 85 90 90 90

Bioavailability (%) 10 5 5 5 5 10 10

Baseline micronutrient content (µg/g) 2 30 47 50 30 4 6

Additional content required (µg/g) 11 22 30 44 22 11 22

Final target content (µg/g) 13 52 77 94 52 15 28

Final target content as dry wt. (µg/g) 15 59 88 107 60 45 85

Zn % of EAR to achieve: 40

EAR, nonpregnant, nonlactating women (µg/d)

1860

EAR, children 4–6 yr of age (µg/d) 830

Micronutrient retention after processing (%)

90 90 90 90 90 90 90

Bioavailability (%) 25 25 25 25 25 25 25




Final target content as dry wt. (µg/g) 28 38 66 56 38 34 70

Provitamin A % of EAR to achieve: 50

EAR, nonpregnant, nonlactating women (µg/d)

500

EAR, children 4–6 yr of age (µg/d) 275

Micronutrient retention after processing 50 50 50 50 50 50 50

Bioavailability ratio (µg:RE‡) 12:1 12:1 12:1 12:1 12:1 12:1 12:1




Final target content as dry wt. (µg/g) 17 17 23 34 17 48 91† EAR, estimated average requirement.‡ RE, retinyl esters.


Se added to fertilizers and applied to soils increased the Se status of the entire Finnish population (Mäkelä et al., 1993). Thus, fertilizers can be used as an effective agricul-tural tool to improve the nutritional health of people in the developing world. Graham et al. (2007) discuss such food system strategies in detail.

The BioavailaBiliTy iSSUeIncreasing the concentrations of micronutrients in sta-

ple food crops is only the first step in making these foods richer sources of these nutrients for humans. As stated pre-viously, this is because not all of the micronutrients in plant foods are bioavailable to humans who eat these foods. Plant foods can contain substances (i.e., antinutrients) that inter-fere with the absorption or utilization of these nutrients in humans (Welch and Graham, 1999). In general, staple food

Figure 2. Micronutrient content of staple crops, across varieties from HarvestPlus screening activities. PVAC = provitamin A carotenoids.

Table 3. Breeding progress as of 2007–2008 (iron, zinc, provitamin A expressed as percent of breeding target in lines at indi-cated stage of breeding).

Crop

Screening Crop improvement G ´ E† testing Launch

Screening gene/trait identification

validation

Early development

parent building

Intermediate product

developmentFinal product development

Performance G ´ E testing in target countries

Release prelaunch seed multiplication

Sweet potato NARS‡ Uganda Program Introduction NARS Uganda

Breeding Provitamin A 100% target 100% 100% 100% 100%

Fast-track Uganda, Mozambique 100% 100%

Maize

Breeding Provitamin A 100% target 60% 50% NA§

Cassava

Breeding Provitamin A 100% target >75% >75% 50% ³30% Fast-track Democratic Republic of Congo NA

Bean

Breeding Fe 100% target 60% 40–50% 40–50%

Fast-track Rwanda 40–50%

Rice, polished

Breeding Zn 100% target 100% 75–100% 75–100% ³30%Wheat

Breeding Zn 100% target 100% ³30% ³30%Pearl millet

Breeding Fe 100% target 100% 75–100% 50–75%†G ´ E, genotype ´ environment interaction.‡NARS, National Agricultural Research Systems.§NA, not applicable.


crop seeds and grains contain very low bioavailable levels of Fe and Zn (i.e., about 5% of the total Fe and about 25% of the total Zn present in the seed is thought to be bio-available). Increasing the bioavailable amounts of Fe from 5 to 20% would be equivalent to increasing the total Fe by fourfold. Using conventional breeding, it should be geneti-cally much easier to greatly improve the bioavailability of Fe and Zn compared with increasing their total content by this magnitude. Antinutrients that depress Fe and Zn bioavailability (such as phytate and certain polyphenolics) or promoter substances (such as ferritin) have fewer genes involved in their biosynthesis and metabolism compared with the uptake, transport, and deposition of Fe and Zn in edible seeds and grains (e.g., >4000 genes have been shown to be up-regulated or down-regulated in controlling Fe homeostasis in higher plants). The fewer the genes needed to breed for makes the job of breeding for the trait easier.

Determining the bioavailability of micronutrients in plant foods to humans is pervaded with numerous com-plexities. A myriad of factors interact to ultimately deter-mine the bioavailability of a particular micronutrient to an individual eating a mixed diet within a given environment (Fairweather-Tait and Hurrell, 1996; Graham et al., 2001; House, 1999; Van Campen and Glahn, 1999). Because of this complexity, the data obtained using various bioavail-ability model systems are always equivocal and dependent on the experimental design used to obtain the data. Only data obtained on reducing the prevalence of micronutrient deficiencies among those afflicted using feeding trials in test populations under free-living conditions can delineate the actual effectiveness of using micronutrient-enriched culti-vars of plant foods as an intervention tool. However, it is impractical to test in this way the bioavailability of selected micronutrients in numerous genotypes of staple plant foods that can be generated in plant breeding programs (Graham and Welch, 1996; Graham et al., 2001). Thus, model bio-availability systems must be used for crop screening pur-poses but ultimately tested in target populations.

idenTifying moleCUlaR maRkeRS in CeReal CRopS To enhanCe BioavailaBle iRon CRopS

An integrated genetic, physiological, and biochemi-cal strategy can be used to identify molecular markers for improving Fe bioavailability in cereal crops. The inter-mated B73 ´ Mo17 (IBM) recombinant inbred (RI) maize population can be employed to identify these markers (Lee et al., 2002). The RI populations are maintained map-ping populations, developed for plant breeders. The maize IBM population is a valuable resource for the analysis of quantitative traits and is the maize breeders’ community standard for genetic mapping, as it has a large number of members (302), extensive recombination, and an exten-sive number of molecular genetic markers (Falque et al.,

2005; Sharopova et al., 2002). B73, a parent from the IBM mapping population, was also used in the Maize Genome Sequencing Project and this facilitates molecular genetic analyses. Scientists at the USDA-ARS Robert W. Holley Center for Agriculture and Health at Cornell University in Ithaca, NY, collected a data set using RIs to find genetic links to improving Fe bioavailability from mature maize kernels using an in vitro Caco-2 cell model. These data were then analyzed using single-marker analysis to identify quantitative trait loci (QTL) that regulate this trait.

The Caco-2 cell line bioassay identified genetic loci in this breeding population associated with increased Fe bio-availability. The identified loci were on six chromosomes and explained 54% of the variance observed in RIs from a single year–location. Three of the largest Fe bioavailabil-ity QTL were successfully isolated in near-isogenic lines (NILs). The NILs are lines that are >90% genetically iden-tical to each other; yet this population contained significant differences in the levels of kernel-Fe bioavailability. The NILs were grown 3 yr after the initial RI population used in the first Caco-2 cell screening experiment. These find-ings confirm the identification of the QTL from the first screening. This is the first genetic analysis for seed-Fe bio-availability and an excellent example for Fe biofortification in a staple food crop. The magnitude of improvement in Fe bioavailability observed in the NILs was comparable to that reported for the highest transgenic events (Drakakaki et al., 2005). This preliminary study was a proof-of-concept study showing the power of using genetic tools to deter-mine which factors in plant foods impact bioavailable Fe from staple food crops. This breeding strategy shows great promise as a tool for plant breeders in the future. However, animal models–human trials should be conducted to sub-stantiate these Caco-2 cell model findings before attempt-ing to breed biofortified maize crops using the identified markers. Preliminary data comparing Caco-2 cell model data with data from a poultry model Fe bioavailability study using high- and low-bioavailable maize kernel recombinant inbred lines is very encouraging (see Fig. 3).

inhiBiToR and pRomoTeR SUBSTanCeS

Plant foods (especially staple seeds and grains) con-tain various antinutrients (Table 4) in differing amounts, depending on both genetic and environmental factors that can reduce the bioavailability of dietary nonheme Fe, Zn, and other micronutrients to humans (Welch, 2001; Welch and House, 1984). Dietary substances that pro-mote/enhance the bioavailability of micronutrients in the presence of antinutrients are also known whose levels are controlled by genes but also influenced by environmen-tal factors (Table 5). Current plant molecular, biological, and genetic modifications, combined with plant breeding approaches, now make it possible to reduce or eliminate


antinutrients from staple plant foods, or to significantly increase the levels of promoter substances in these foods (Becker and Frei, 2004; Forssard et al., 2000; Genc et al., 2005; King, 2002; Theil et al., 1997; Welch, 2002; White and Broadley, 2009). Given these options (i.e., to decrease antinutrients or to increase promoters in staple plant foods), which is the wisest path to pursue?

Plant breeders could breed for genotypes that contain lower concentrations of antinutrients or molecular biolo-gists could alter plant genes in ways that reduce or even eliminate antinutrients from plant food meals. However, doing so is not without risk and should be done with cau-tion because many antinutrients are major plant metabo-lites that may play important roles in plant metabolism, in plant abiotic stress resistance, and in plant resistance to crop pests or pathogens (Graham et al., 2001). Additionally,

some of the antinutrients, such as phytate and polyphe-nols, may play important beneficial roles in human diets by acting as anticarcinogens or by promoting health in other ways such as in decreasing the risk of heart disease or diabetes (Anonymous, 1996; Saied and Shamsuddin, 1998; Shamsuddin, 1999; Zhou and Erdman, 1995). Thus, plant breeders and molecular biologists should be aware of the possible negative consequences of changing antinutri-ents in major plant foods before they attempt to alter food crops in this fashion (Graham and Welch, 1996).

Some promoter compounds are normal plant metabo-lites. Only a few genes control their levels in plants and only small changes in their concentration may have signif-icant effects on the bioavailability of micronutrients. Thus, breeding for increased levels of these promoters should be relatively easy compared with breeding for higher levels

Figure 3. Bioavailable Fe in maize kernels from two recombinant inbred lines (RILs) of maize determined using either the in vitro Caco-2 cell model (cell ferritin level was used as a proxy for Fe bioavailability) or using a poultry model and blood hemoglobin as a measure of Fe bioavailability (R.P. Glahn and E. Tako, unpublished data, 2009). Total dietary Fe and kernel Fe levels were about equal for both high- and low-Fe-bioavailability maize RILs used in the poultry model.

Table 4. Examples of antinutrients in plant foods that reduce the bioavailability of essential trace elements and examples of major dietary sources (modified from Graham et al., 2001).

AntinutrientsEssential micronutrient

metal inhibited Major dietary sourcesPhytic acid or phytin Fe, Zn, Cu, Ni Whole legume seeds and cereal grainsCertain fibers (e.g., cellulose, hemicellulose, lignin, cutin, suberin)

Fe, Zn, Cu Whole cereal grain products (e.g., wheat, rice, maize, oat, barley [Hordeum vulgare L.], rye [Secale cereale L.])

Certain tannins and other polyphenolics Fe Tea [Camellia sinensis (L.) Kuntze], coffee (Coffea arabica L.), beans, sorghum [Sorghum bicolor (L.) Moench]

Hemagglutinins (e.g., lectins) Fe Most legumes and wheat

Goitrogens I Brassicas and Alliums

Heavy metals (e.g., Cd, Hg, Pb) Fe, Zn Contaminated leafy vegetables and roots


of Fe and Zn, which involves numerous genes and their interactions with the environment. Therefore, it is highly recommended that plant breeders and molecular biologists closely scrutinize the strategy of increasing promoter sub-stances in food crops when attempting to improve food crops as sources of micronutrients for people (Graham et al., 2007; Welch and Graham, 1999, 2004).

pReBioTiCS aS pRomoTeRS of miCRonUTRienTS

Which known plant food promoter substances should be targeted for increasing in staple plant foods through bio-fortification to improve Fe and Zn bioavailability? Unfor-tunately, there is a dearth of knowledge concerning Fe and Zn promoters in staple plant foods. The well-known Fe promoter and antioxidant ascorbate could be increased in staples, although it is not stable because it can be oxidized to dehydroascorbate during storage, food preparation, and cooking, losing its promoter properties (Combs, 2008). Thus, ascorbate may not be a good target promoter for plant breeding. The amino acid cysteine is also known to promote Fe and Zn bioavailability. Breeding for higher lev-els of cysteine-rich peptides and proteins could be achieved (Lucca et al., 2001; White and Broadley, 2009). However, cysteine also is prone to oxidation to the disulfide cystine during processing and cooking, potentially losing its pro-motion properties by oxidation of its metal-binding sulf-hydryl functional group. The Fe stored as phytoferritin (a 450,000-Da protein) in seeds is a bioavailable source of Fe in staple food crops. It protects up to 4500 ferric-Fe atoms stored in its Fe cage from binding to antinutrients such as phytate (Lonnerdal, 2009). Breeding for enhanced levels of phytoferritin in staple food crops appears to be a viable strategy if genetic engineering approaches are used (Lucca et al., 2006; White and Broadley, 2009), although the genetic diversity in seed-phytoferritin accumulation in the genomes of the major staple food crop seeds is not known. If enough genetic diversity existed for this trait in these genomes, then conventional breeding could be used to increase phytoferritin in these crops.

One very promising area related to improving the bio-availability of Fe and other micronutrients in staple food

crops is the role of nondigestible carbohydrates as enhanc-ers of micronutrient bioavailability. Within the past decade, numerous studies have reported promoter effects of various nondigestible carbohydrates on Ca, Mg, Fe, Cu, and Zn absorption in animal models and in humans, even when consumed in diets containing high amounts of antinutrients from staple food crops. Much of this research has focused on fructans, the fructo-oligosaccharides including inulin. The mode of action of fructans is the result of their promoting the growth of beneficial microbiota primarily within the caecum and colon, which has systemic effects on improving micronutrient absorption and utilization. These carbohy-drates are classed as prebiotics—substances that significantly promote the growth of beneficial bacteria (i.e., probiotics) in the distal small intestine and the large intestine. Increases in probiotic bacteria in the intestine have been shown to have beneficial systemic effects on a number of metabolic pathways in the human body. Research into human gut microbiota and their effects on human nutrition and health is in its infancy. Yet, it is clear that the effect of our intes-tinal microbiota on our ability to utilize food, nutrients, and phytochemicals is immense (Dethlefsen et al., 2007; FAO and WHO, 2006; Manning and Gibson, 2004). With respect to Fe nutriture, probiotics may play a critical role in Fe absorption from the diet and this is discussed below.

The hUman “SUpeRoRganiSm”— The Body, iTS miCRoBeS, and TheiR Role in iRon BioavailaBiliTy

The human intestine contains more bacteria than the eukaryotic cells of the body (i.e., at least 10 trillion micro-bial cells compared with about one trillion body cells). The metabolic activity of these organisms is equal to that of the body’s vital organs and can account for 60% of the dry weight of feces (Steer et al., 2000). Studies have shown that host–microbe interactions are essential to normal mamma-lian physiology, including metabolic activity and immune homeostasis (Dethlefsen et al., 2007). Their activity pro-vides energy from undigested food substrates, trains the immune system, prevents growth of pathogens, transforms certain nutrients and beneficial phytochemicals into utiliz-able substrates, synthesizes certain vitamins, defends against

Table 5. Examples of substances in foods reported to promote Fe and Zn bioavailability and examples of major dietary sources (modified from Graham et al., 2001).

Substance Trace element Major dietary sourcesCertain organic acids (e.g., ascorbic acid, fumarate, malate, citrate) Fe and/or Zn Fresh fruits and vegetables

Hemoglobin Fe Animal meats

Certain amino acids (e.g., methionine, cysteine, histidine) Fe and/or Zn Animal meats

Long-chain fatty acids (e.g., palmitate) Zn Human breast milk

Se I Seafoods, tropical nuts

b-carotene Fe Green and orange vegetablesInulin and other nondigestible carbohydrates (prebiotics) Fe, Zn Chicory (Cichorium intybus L.), garlic (Allium sativum L.),

onion (Allium cepa L.), wheat, Jerusalem artichoke (Helian-thus tuberosus L.)


certain diseases, stimulates cell growth, prevents some aller-gies, improves mineral absorption, produces anti-inflam-matory effects, and improves gut health in general.

Low-grade inflammation (i.e., systemic inflamma-tion) can occur because of changes in the bacteria popula-tions colonizing the intestine from certain dietary habits. For example, high fat intake has been reported to increase the proportion of gram-negative to gram-positive bac-teria in the intestine (Cani et al., 2008). Gram-negative bacteria contain the endotoxin lipopolysaccharides (LPS) in their cell walls; gram-positive bacteria contain no LPS. Endotoxemia, resulting from intestinal epithelium exposure to cell-wall LPS from gram-negative bacteria, causes a cellular immune signaling cascade that results in the inflammatory response (Bensinger and Tontonoz, 2008; Schiffrin and Blum, 2002). Inflammation can lead to up-regulation of the genes encoding the biosynthesis of the Fe-regulation peptide hormone hepcidin. Injection of humans with LPS dramatically increased serum IL-6 and urinary hepcidin within 6 h and reduced serum Fe concentrations by 57% within 22 h (Kemna et al., 2005). Hepcidin is primarily produced in the liver. It is trans-located to intestinal enterocytes where it suppresses the induction of Fe deficiency response genes in the apical and basal membranes of mucosal cells, lowering their abil-ity to absorb and utilize Fe from the diet and to transfer Fe across their basolateral membrane into the blood. This can lead to the anemia of inflammation even when diets contain adequate levels of bioavailable Fe, as a host defense mechanism to inhibit the growth of infectious bacteria.

Changes in the bacterial profile of the gut to a higher gram-positive (e.g., Firmicutes bacteria) to a gram-neg-ative bacteria (e.g., Proteobacteria) ratio has been shown to result in reduced inflammation and lower LPS levels in the intestine and an improvement in mucosal barrier function (Cani et al., 2008; Wang et al., 2006). Further-more, prebiotics, such as fructans, stimulate the growth of beneficial gram-positive (probiotic) bacteria at the expense of gram-negative bacterial growth (Bouhnik et al., 2007; Salminen et al., 1998). Beneficial gram-positive bacteria, such as bifidobacteria, do not degrade intestinal mucous glycoproteins, which promote a healthier micro-villus environment by reducing intestinal permeability to gram-negative bacteria. This results in less LPS enter-ing the blood (Cani et al., 2007; Griffiths et al., 2004; Teitelbaum and Walker, 2002). Therefore, changes in the ratio of gram-positive to gram-negative bacteria in the intestine and their link to inflammation may provide an Occam’s razor explanation for the effects of prebiotics on up-regulating the genes for Fe absorption by enterocytes in the intestine.

Changing the gut microbiota populations to more gram-positive bacteria may also have enhancing effects on Zn absorption, but little experimental evidence exists.

Providing prebiotics may overcome the negative effects of antinutrients on Fe and Zn bioavailability because many bacteria in the gut can degrade antinutrients, such as phy-tate and polyphenols, releasing their bound metals (such as Fe and Zn) and allowing them to be absorbed by entero-cytes lining the intestine. Probiotics’ systemic effects on inducing the genes controlling the absorption of Fe and other metals from the intestine may enhance the bio-availability of these micronutrients. Of equal and possibly more importance is the role of prebiotics on improving gut health and the intestine’s ability to absorb and uti-lize numerous nutrients, regulate the immune system, and protect against invasion by pathogenic organisms. Thus, increasing the levels of prebiotics in staple food crops is an extremely important strategy to enhance the nutrition and health of malnourished people everywhere, especially resource-poor families with poor gut health living in less sanitary environments.

developmenT impaCTAs briefly summarized above, reducing micronutrient

malnutrition improves cognitive ability, reduces morbid-ity and mortality, and improves work productivity.

In an analysis of commercial fortification, Horton and Ross (2003) estimate that the present value of each annual case of Fe deficiency averted in South Asia is approxi-mately US$20.2. Consider the value of 1 billion cases of Fe deficiency averted in 16–25 yr after a biofortification research and development project was initiated (100 mil-lion cases averted per year in South Asia). The nominal value of US$20 billion (1 billion cases ´ a value of US$20 per case) must be discounted because of the lags involved between the times that investments are made in bioforti-fication and when benefits are realized. At a 3% discount rate the present value would be approximately US$10 bil-lion, and at a 12% discount rate the present value would be approximately US$2 billion. This benefit is far higher than the cost of breeding, testing, and disseminating high-Fe and high-Zn cultivars of rice and wheat for South Asia (more than US$100 million in nominal costs).

aChievaBle goalS foR The ShoRT- and long-TeRm

HarvestPlus’s experience in the dissemination of bio-fortified crops is limited to orange sweet potato (Ipomoea batatas L.), which is very high in provitamin A. A pub-lished pilot study in Mozambique showed that (i) behavior can be changed among farmers by switching from pro-duction of white to orange cultivars, and change in con-sumption to orange cultivars by their families; and that (ii) vitamin A deficiency can be improved (Low et al., 2007). As a result, vitamin A deficiency among preschool children in treatment villages declined from 60 to 38%, while vitamin A deficiency remained constant in control


villages. HarvestPlus is now concentrating on identifying activities and messages that will effect this same behavior change at the lowest cost possible.

The dissemination strategy for nutrients that are invisible (Fe and Zn) will piggyback on superior agro-nomic characteristics of the newly introduced cultivars. For example, high-Fe beans that are drought- and heat-tolerant are undergoing national release trials in Africa.

In developing detailed plans for delivery of bioforti-fied crops and to achieve realistic goals for delivery during 2014 to 2019, HarvestPlus management realized that the number of crops being developed under HarvestPlus II would need to be reduced. Given progress to date, Har-vestPlus can now anticipate release dates for the bioforti-fied products (Table 6).

new modaliTieS foR ReSeaRCh CollaBoRaTionS—faCToRS affeCTing ReSeaRCh CollaBoRaTion

Interdisciplinary exchange/communication is crucial for the success of HarvestPlus. Such interactions become increasingly productive as experience is gained, that is, over time and at a series of meetings. HarvestPlus has an advantage from experience that was gained by a subset of the collaborating institutions in precursor projects, but many new non-CGIAR collaborators have participated since 2003. To motivate true collaboration, it is impor-tant that the collaborating institutions share a common set of shared goals/objectives, which must be jointly dis-cussed and agreed on. Understanding across disciplines is hindered by technical language, which is either not com-monly understood or has different connotations to dif-ferent disciplines. These barriers must be surmounted. This all takes time and the give and take of interacting on repeated occasions.

The optimal situation in terms of team-building is one in which the partner institutions are all known at the start of the planning process. Competitive bidding can hinder this process of team-building in three ways. First, if one does not know that their proposal will be selected, either he/she will be more reluctant to fully buy into the planning process, or may not have been invited to partici-pate in the planning process at all. Second, a winning bid-der has agreed to undertake a specific activity. Challenge Programs must be flexible as ongoing research and exter-nal circumstances dictate changes in overall plans. Unless fully integrated into a culture of teamwork, the winning bidder may be reticent to alter the terms of reference of the winning bid, which may have taken quite a substantial amount of work to prepare. Third, it is usually expected that competitive bids will be decided only on the basis of technical competence, perhaps also with a value placed on capacity building. However, ability/willingness to

collaborate across disciplinary boundaries is essential and difficult to assess in evaluating formal proposals.

goveRnanCe ThRoUgh diSTRiBUTed deCiSion-making poweR inCReaSeS TRanSaCTion CoSTS

Building consensus among collaborating institutions is vital to the success of HarvestPlus. The Program Direc-tor reports to a Project Advisory Committee which has ultimate decision-making power over workplans and bud-gets, as well as the Directors General of CIAT and IFPRI. Such a structure inherently forces consensus-building.

Nevertheless, consensus-building requires consider-able transactions costs. The Program Management Team must have flexibility to make operational decisions, subject to Program Advisory Committee (PAC) oversight every 6 mo, within the strategic boundaries set by the PAC. The PAC members do not represent stakeholder institutions (except for minority representation of CIAT and IFPRI), but do represent a broad spectrum of scientific disciplines, career work experiences, and nations around the world. This governance system has worked well.

ConSideR oUTReaCh To The pUBliC aT inCepTion

We took the decision to change the name of the Bio-fortification Challenge Program to HarvestPlus as a way to reach out more effectively to the public. We felt that this was important in terms of (i) sustaining donor sup-port for a long-term program, and (ii) meeting one of the goals of the Challenge Programs to raise the public pro-file of the CGIAR Centers. Not everyone agreed with the decision; several scientists were reticent to use such an “imprecise” title. However, the decision-making process was highly participatory, the decision approved by a large majority, accepted, and behind us. Time has proven that this was a good decision.

SUmmaRyThe biofortification strategy seeks to take advantage of

the consistent daily consumption of large amounts of food staples by all family members, including women and chil-dren who are most at risk for micronutrient malnutrition. As a consequence of the predominance of food staples in the diets of the poor, this strategy implicitly targets low-income households. After a one-time investment in devel-oping seeds that fortify themselves, recurrent costs are low and germplasm may be shared internationally. It is this mul-tiplier aspect of plant breeding across time and distance that makes it so cost-effective. Once in place, production and consumption of nutritionally improved cultivars is highly sustainable, even if government attention and international funding for micronutrient issues fade. Biofortification


provides a feasible means of reaching malnourished popula-tions in relatively remote rural areas, delivering naturally fortified foods to people with limited access to commer-cially marketed fortified foods, which are more readily available in urban areas. Biofortification and commercial fortification, therefore, are highly complementary.

Ultimately, good nutrition depends on adequate intakes of a range of nutrients and other compounds, in combinations and levels that are not yet completely under-stood. Thus, the best and final solution to eliminating undernutrition as a public health problem in developing countries is to provide increased consumption of a range of nonstaple foods. However, this will require several decades to be realized, informed government policies, and a relatively large investment in agricultural research and other public and on-farm infrastructure.

In conceptualizing solutions for a range of nutritional deficiencies, interdisciplinary communication between plant scientists and human nutrition scientists holds great potential. Human nutritionists need to be informed, for example, about the extent to which the vitamin and min-eral density of specific foods, as well as compounds that promote and inhibit their bioavailability, can be modified through plant breeding. Plant breeders need to be aware of both the major influence that agricultural research may have had on nutrient utilization in the past (e.g., the bio-availability of micronutrients in modern cultivars vs. bio-availability in traditional cultivars), and the potential of plant breeding for future improvements in nutrition.

RefeRenCeSAllaway, W.H. 1986. Soil–plant–animal and human intrrelation-

ships in trace element nutrition. p. 465–488. In W. Mertz (ed.) Trace elements in human and animal nutrition. Academic Press, Orlando, FL.

Anonymous. 1996. Viewpoint: The optimal breeding strategy is to increase the density of promoter compounds and micronu-trient minerals in seeds; caution should be used in reducing antinutrients in staple food crops. Micronutr. Agric. 1:20–22.

Anonymous. 2007. Global childhood malnutrition. Lancet 367:1459.

Anonymous. 2008. The Copenhagen Consensus 2008—Results. p. 1–6. Available at http://www.copenhagenconsensus.com/default.aspx?ID=953 (verified 21 Dec. 2009). The Copenha-gen Consensus Cent., Frederiksberg, Denmark.

Becker, K., and M. Frei. 2004. Improving the nutrient availability

in rice—Biotechnology or biodiversity? Agric. Rural Devel. 11:64–65.

Bensinger, S.J., and P. Tontonoz. 2008. Integration of metabolism and inflammation by lipid-activated nuclear receptors. Nature 454:470–477.

Bouhnik, Y., L. Raskine, K. Champion, C. Andrieux, S. Penven, H. Jacobs, and G. Simoneau. 2007. Prolonged administration of low-dose inulin stimulates the growth of bifidobacteria in humans. Nutr. Res. 27:187–193.

Bouis, H.E. 1999. Economics of enhanced micronutrient density in food staples. Field Crops Res. 60:165–173.

Bouis, H.E., C. Hotz, B. McClafferty, J.V. Meenakshi, and W. Pfeiffer. 2009. Biofortification: A new tool to reduce micro-nutrient malnutrition. In Int. Congr. of Nutrition, 19th, Bangkok, Thailand. 4–9 Oct. 2009.

Cakmak, I. 2008. Enrichment of cereal grains with zinc: Agro-nomic or genetic biofortification? Plant Soil 302:1–17.

Cani, P., A. Neyrinck, F. Fava, C. Knauf, R. Burcelin, K. Tuohy, G. Gibson, and N. Delzenne. 2007. Selective increases of bifi-dobacteria in gut microflora improve high-fat-diet-induced diabetes in mice through a mechanism associated with endo-toxaemia. Diabetologia 50:2374–2383.

Cani, P.D., R. Bibiloni, C. Knauf, A. Waget, A.M. Meyrinck, N.M. Delzenne, and R. Burcelin. 2008. Changes in gut microbiota control metabolic endotoxemia-induced inflam-mation in high fat diet–induced obesity and diabetes in mice. Diabetes 57:1470–1481.

Cao, X.Y., X.M. Jiang, A. Kareem, Z.H. Dou, M.R. Rakeman, M.L. Zhang, T. Ma, K. O’Donnell, N. DeLong, and G.R. DeLong. 1994. Iodination of irrigation water as a method of supplying iodine to a severely iodine-deficient population in Xinjiang, China. Lancet 344:107–110.

Casey, R.P., and R.G. Lugar. 2008. A call for a strategic U.S. approach to the global food crisis. p. 1–9. CSIS Press, Wash-ington, DC.

Combs, G.F., Jr. 2008. The vitamins. Third ed. Elsevier Academic Press, San Diego, CA.

Dethlefsen, L., M. McFall-Ngai, and D.A. Relman. 2007. An ecological and evolutionary perspective on human–microbe mutualism and disease. Nature 449:811–818.

Drakakaki, G., S. Marcel, R. Glahn, E. Lund, S. Pariagh, R. Fischer, P. Christou, and E. Stoger. 2005. Endosperm-specific co-expression of recombinant soybean ferritin and Aspergillus phytase in maize results in significant increases in the levels of bioavailable iron. Plant Mol. Biol. 59:869–880.

Fairweather-Tait, S.J., and R.F. Hurrell. 1996. Bioavailability of minerals and trace elements. Nutr. Res. Rev. 9:295–324.

Falque, M., L. Decousset, D. Dervins, A.M. Jacob, J. Joets, J.P. Martinant, X. Raffoux, N. Ribiere, C. Ridel, D. Samson, A. Charcosset, and A. Murigneux. 2005. Linkage mapping of

Table 6. Schedule of product release for biofortified products.

Crop Nutrient Countries of first release Agronomic trait Release year†

Sweet potato Provitamin A Uganda, Mozambique High yielding, virus resistance, drought tolerance 2007

Bean Fe, Zn Rwanda, Democratic Republic of Congo Virus resistance, heat and drought tolerance 2010

Pearl millet Fe, Zn India Mildew resistance, drought tolerance 2011

Cassava Provitamin A Nigeria, Democratic Republic of Congo High yielding, virus resistance 2011–2012

Maize Provitamin A Zambia High yielding, disease resistance, drought tolerance 2011–2012

Rice Zn, Fe Bangladesh, India Disease and pest resistance, submergence tolerance 2012–2013

Wheat Zn, Fe India, Pakistan Disease resistance, lodging 2012–2013†Approved for release by national governments after 2–3 yr of testing.


1454 new maize candidate gene loci. Genetics 170:1957–1966.FAO and WHO. 2006. Probiotics in food. Health and nutritional

properties and guidelines for evaluation. World Health Organ./Food and Agric. Organ. of the United Nations, Rome.

Forssard, E., M. Bucher, F. Mächler, A. Mozafar, and R. Hurrell. 2000. Review. Potential for increasing the content and bio-availability of Fe, Zn and Ca in plants for human nutrition. J. Sci. Food Agric. 80:861–879.

Genc, Y., J.M. Humphries, G.H. Lyons, and R.D. Graham. 2005. Exploiting genotypic variation in plant nutrient accumulation to alleviate micronutrient deficiency in populations. J. Trace Elem. Med. Biol. 18:319–324.

Graham, R.D., and R.M. Welch. 1996. Breeding for staple-food crops with high micronutrient density. Int. Food Policy Res. Inst., Washington, DC.

Graham, R.D., R.M. Welch, and H.E. Bouis. 2001. Addressing micronutrient malnutrition through enhancing the nutri-tional quality of staple foods: Principles, perspectives and knowledge gaps. Adv. Agron. 70:77–142.

Graham, R.D., R.M. Welch, D.A. Saunders, I. Monasterio, H.E. Bouis, M. Bonierbale, S. de Hann, G. Burgos, G. Thiele, R. Liria, C.A. Meisner, S.E. Beebe, M.J. Potts, M. Kadiajn, P.R. Hobbs, R.K. Gupta, and S. Twomlow. 2007. Nutritious sub-sistence food systems. Adv. Agron. 92:1–74.

Griffiths, E.A., L.C. Duffy, F.L. Schanbacher, H. Qiao, D. Dryja, A. Leavens, J. Rossman, G. Rich, D. Dirienzo, and P.L. Ogra. 2004. In vivo effects of bifidobacteria and lactoferrin on gut endotoxin concentration and mucosal immunity in Balb/c mice. Dig. Dis. Sci. 49:579–589.

Grunes, D.L., and W.H. Allaway. 1985. Nutritional quality of plants in relation to fertilizer use. p. 589–619. In O.P. Engel-stad (ed.) Fertilizer technology and use. SSSA, Madison, WI.

Haug, A., R.D. Graham, O.A. Christopherson, and G.H. Lyons. 2008. How to use the world’s scarce selenium resources efficiently to increase the selenium concentration in food. Microb. Ecol. Health Dis. 19:209–228.

Hawkes, C., and M.T. Ruel. 2006. Understanding the links between agriculture and health. p. 1–32. In C. Hawkes and M.T. Ruel (ed.) Int. Food Policy Res. Inst., Washington, DC.

Horton, S., and J. Ross. 2003. The economics of iron deficiency. Food Policy 28:51–75.

Hotz, C., B. McClafferty, C. Hawkes, M. Ruel, and S. Babu. 2007. From harvest to health: Challenges for developing biofortified staple foods and determining their impact on micronutrient status. Food Nutr. Bull. 28:S271–S279.

House, W.A. 1999. Trace element bioavailability as exemplified by iron and zinc. Field Crops Res. 60:115–141.

House, W.A., and R.M. Welch. 1989. Bioavailability of and inter-actions between zinc and selenium in rats fed wheat grain intrinsically labeled with 65Zn and 75Se. J. Nutr. 119:916–921.

Kemna, E., H. Tjalsma, C. Laarakkers, E. Nemeth, H. Willems, and D. Swinkels. 2005. Novel urine hepcidin assay by mass spectrometry. Blood 106:3268–3270.

Kennedy, G., G. Nantel, and P. Shetty. 2003. The scourge of “hid-den hunger”: Global dimensions of micronnutrient deficien-cies. Food Nutr. Agric. 32:8–16.

King, J.C. 2002. Biotechnology: A solution for improving nutrient bioavailability. Int. J. Vit. Nutr. Res. 72:7–12.

Lee, M., N. Sharopova, W.D. Beavis, D. Grant, M. Katt, D. Blair, and A. Hallauer. 2002. Expanding the genetic map of maize with the intermated B73 ´ Mo17 (IBM) population. Plant Mol. Biol. 48:453–461.

Lonnerdal, B. 2009. Soybean ferritin: Implications for iron status of vegetarians. Am. J. Clin. Nutr.

Low, J.W., M. Arimond, N. Osman, B. Cunguara, F. Zano, and D. Tschirley. 2007. A food-based approach introducing orange-fleshed sweet potatoes increased Vitamin A intake and serum retinol concentrations in young children in rural Mozam-bique. J. Nutr. 137:1320–1327.

Lucca, P., R. Hurrell, and I. Potrykus. 2001. Genetic engineering approaches to improve the bioavailability and the level of iron in rice grains. Theor. Appl. Genet. 102:392–397.

Lucca, P., S. Poletti, and C. Sautter. 2006. Genetic engineering approaches to enrich rice with iron and vitamin A. Physiol. Plant. 126:291–303.

Mäkelä, A.-L., V. Näntö, P. Mäkelä, and W. Wang. 1993. The effect of nationwide selenium enrichment of fertilizers on selenium status of healthy Finnish medical students living in south western Finland. Biol. Trace Elem. Res. 36:151–157.

Manning, T.S., and G.R. Gibson. 2004. Prebiotics. Best Pract. Res. Clin. Gastroenterol. 18:287–298.

Moraghan, J.T. 1980. Effects of soil temperature on response of flax to phosphorus and zinc fertilizers. Soil Sci. 129:290–296.

Nestel, P., H.E. Bouis, J.V. Meenakshi, and W. Pfeiffer. 2006. Bio-fortification of staple food crops. J. Nutr. 136:1064–1067.

Ortiz-Monasterio, J.I., N. Palacios-Rojas, E. Meng, K. Pixley, R. Trethowan, and R.J. Pena. 2007. Enhancing the mineral and vitamin content of wheat and maize through plant breeding. J. Cereal Sci. 46:293–307.

Peck, N., D.L. Grunes, R.M. Welch, and G.E. MacDonald. 1980. Nutritional quality of vegetable crops as affected by phospho-rus and zinc fertilizers. Agron. J. 72:528–534.

Pfeiffer, W.H., and B. McClafferty. 2007. HarvestPlus: Breeding crops for better nutrition. Crop Sci. 47:S88–S105.

Qaim, M., A.J. Stein, and J.V. Meenakshi. 2007. Economics of biofortification. Agric. Econ. 37:119–133.

Ren, Q., F. Fan, Z. Zhang, X. Zheng, and G.R. DeLong. 2008. An environmental approach to correcting iodine deficiency: Supplementing iodine in soil by iodination of irrigation water in remote areas. J. Trace Elem. Med. Biol. 22:1–8.

Rouse, T.I., and D.P. Davis. 2004. Exploring a vision: Integrating knowledge for food and health. A workshop summary. Board on Agric. and Nat. Resources, Division on Earth and Life Stud., Natl. Res. Counc. of the Natl. Acad. of Sci. Natl. Acad. Press, Washington, DC.

Saied, H.T., and A.M. Shamsuddin. 1998. Up-regulation of the tumor suppressor gene p53 and WAF1 gene expression by IP6 in HT-29 human colon carcinoma cell line. Anticancer Res. 18:1479–1484.

Salminen, S., C. Bouley, M.-C. Boutron-Ruault, J.H. Cum-mings, A. Franck, G.R. Gibson, E. Isolauri, M.-C. Moreau, M.B. Roberfroid, and I. Rowland. 1998. Functional food sci-ence and gastrointestinal physiology and function. Br. J. Nutr. 80:S147–S171.

Schiffrin, E.J., and S. Blum. 2002. Interactions between the microbiota and the intestinal mucosa. Eur. J. Clin. Nutr. 56:S60–S64.

Shamsuddin, A.M. 1999. Metabolism and cellular functions of IP6: A review. Anticancer Res. 19:3733–3736.

Sharopova, N., M.D. McMullen, L. Schultz, S. Schroeder, H. San-chez-Villeda, J. Gradiner, K. Edwards, T. Ruff, J.C. Register, C. Brouwer, R. Thompson, R. Velasco, E. Chin, M. Lee, W. Woodman-Clikeman, M.J. Long, E. Liscum, K. Cone, G. Davis, and E.H. Coe, Jr. 2002. Development and mapping of


SSR markers for maize. Plant Mol. Biol. 48:463–481.Sobal, J., L.K. Khan, and C. Bisogni. 1998. A conceptual model

of the food and nutrition system. Soc. Sci. Med. 47:853–863.Steer, T., H. Carpenter, K. Tuohy, and G.R. Gibson. 2000. Per-

spectives on the role of the human gut microbiota and its modulation by pro- and prebiotics. Nutr. Res. Rev. 13:229–254.

Teitelbaum, J.E., and W.A. Walker. 2002. Nutritional impact of pre- and probiotics as protective gastrointestinal organisms. Annu. Rev. Nutr. 22:107–138.

Theil, E.C., J.W. Burton, and J.L. Beard. 1997. A sustainable solu-tion for dietary iron deficiency through plant biotechnology and breeding to increase seed ferritin control. Eur. J. Bio-chem. 51:S28–S31.

Van Campen, D.R., and R.P. Glahn. 1999. Micronutrient bio-availability techniques: Accuracy, problems and limitations. Field Crops Res. 60:93–113.

Wang, Z.M., G.M. Xiao, Y.M. Yao, S.M. Guo, K.M. Lu, and Z.M. Sheng. 2006. The role of bifidobacteria in gut barrier function after thermal injury in rats. J. Trauma-Injury Infect. Crit. Care 61:650–657.

Welch, R.M. 1986. Effects of nutrient deficiencies on seed produc-tion and quality. Adv. Plant Nutr. 2:205–247.

Welch, R.M. 1995. Micronutrient nutrition of plants. Crit. Rev. Plant Sci. 14:49–82.

Welch, R.M. 1999. Importance of seed mineral nutrient reserves in crop growth and development. p. 205–226. In Z. Rengel (ed.) Mineral nutrition of crops. Fundamental mechanisms and implications. Food Products Press, New York.

Welch, R.M. 2001. Micronutrients, agriculture and nutrition; linkages for improved health and well being. p. 247–289. In

K. Singh et al. (ed.) Perspectives on the micronutrient nutri-tion of crops. Sci. Publ., Jodhpur, India.

Welch, R.M. 2002. Breeding strategies for biofortified staple plant foods to reduce micronutrient malnutrition globally. J. Nutr. 132:495S–499S.

Welch, R.M., and R.D. Graham. 1999. A new paradigm for world agriculture: Meeting human needs—Productive, sustainable, nutritious. Field Crops Res. 60:1–10.

Welch, R.M., and R.D. Graham. 2004. Breeding for micronutri-ents in staple food crops from a human nutrition perspective. J. Exp. Bot. 55:353–364.

Welch, R.M., and W.A. House. 1984. Factors affecting the bio-availability of mineral nutrients in plant foods. p. 37–54. In R.M. Welch and W.H. Gabelman (ed.) Crops as sources of nutrients for humans. ASA, Madison, WI.

Welch, R.M., W.A. House, and W.H. Allaway. 1974. Availability of zinc from pea seeds to rats. J. Nutr. 104:733–740.

White, P.J., and M.R. Broadley. 2009. Biofortification of crops with seven mineral elements often lacking in human diets—Iron, zinc, copper, calcium, magnesium, selenium and iodine. New Phytol. 182:49–84.

WHO and FAO. 2003. Joint WHO/FAO expert consultation on diet, nutrition and the prevention of chronic diseases. World Health Organ., Geneva, Switzerland.

World Bank. 2007. From agriculture to nutrition: Pathways, syn-ergies and outcomes. Rep. 40196-GLB. Int. Bank for Recon-struction and Dev./The World Bank, Washington, DC.

Zarocostas, J. 2009. Malnutrition and hunger rise as economic cri-sis hits world’s poorest people. BMJ 338:B1651.

Zhou, J.R., and J.W. Erdman, Jr. 1995. Phytic acid in health and disease. Crit. Rev. Food Sci. Nutr. 35:495–508.

crop science, vol. 50, march–april 2010 S-33

SympoSia

Providing humankind with enough food has been a challenge throughout the ages. This topic remains of importance, but

food production has changed considerably in the last 50 to 100 yr. Food security and quality improved tremendously in the indus-trialized world, but increasing obesity suggests humankind’s dif-ficulty in handling overweight. The impact of food production on the environment has also become problematic. However, on a global scale, food security and quality are not yet realized, and even though the situation has changed in the past century, we are still faced with tremendous challenges that require new food options to provide nutritious and healthy diets to overcome mal-nutrition. The following options may assist in this endeavor:

The Future of Food: Scenarios for 2050

Bernard Hubert, Mark Rosegrant, Martinus A. J. S. van Boekel, and Rodomiro Ortiz*

abStractThis background article addresses key chal-lenges of adequately feeding a population of9 billion by 2050, while preserving the agro-ecosystemsfromwhichotherservicesarealsoexpected. One of the scenario-buildings usesthe Agrimonde platform, which considers thefollowing steps: choosing the scenarios andtheirunderlyingbuildingprinciples,developingquantitative scenarios, and building completescenariosbycombiningquantitative scenarioswith qualitative hypotheses. These scenariosconsiderhowfoodissueslinktoproduction,forexample, thepercentageofanimal vs. vegetalcalorieintakeinthefulldiet.ThefirstsectionofthisarticlediscussesAgrimonde GOandAgri-monde 1 scenarios,which indicate thatglobaleconomic growth and ecological intensifica-tionremainasmainchallengesfor feedingtheearth’sgrowingpopulationtowardthemid-21stcentury.Thesecondsectionprovides theout-comesof theanalysisofalternative futuresforagriculturalsupplyanddemandandfoodsecu-rity to 2050, based on research done for theInternational Assessment of Agricultural Sci-enceandTechnologyforDevelopment.Thelastsectionofthisarticleprovidesasummaryanal-ysisof foodsystemsand functions,aswellastheroleoffoodtechnologythataddresssomeoftheglobalchallengesaffectingthesupplyofmorenutritiousandhealthydiets. Italsohigh-lightsthefoodproductionbynovelmeans(e.g.,alternativesforanimalproductsbasedonplantmaterials)andincreasingthepresenceofpoten-tially health-promoting compounds in food toimprovehumanandanimalhealth.Finally, thisarticle proposes priority areas that should beincludedinfurtheragri-foodresearch.

B. Hubert, French Initiative for International Agricultural Research, (FI4AR), Agropolis International, Ave. Agropolis, F-34394 Montpel-lier Cedex 5, France; M. Rosegrant, International Food Policy Research Institute, 2033 K St, NW, Washington, DC 20006-1002; M.A.S.J. van Boekel, Dep. Agrotechnology and Food Sciences, Wageningen Univ., PO Box 8129, 6700 EV Wageningen, The Netherlands; R. Ortiz, Cen-tro Internacional de Mejoramiento de Maíz y Trigo (CIMMYT). R. Ortiz, current address: Martin Napanga 253, Apt. 101, Miraflores, Lima 18, Peru. B. Hubert, M. Rosegrant, and M.A.S.J. van Boekel contrib-uted equally to this manuscript. Received 21 Sept. 2009. *Correspond-ing author ([email protected]).

Abbreviations: AKST, Agricultural Knowledge Science and Technology; FAO, Food and Agriculture Organization of the United Nations; GDP, gross domestic product; IAASTD, International Assessment of Agricultural Science and Technology for Development; IMPACT, International Model for Policy Analysis of Agricultural Commodities and Trade; MEA, Millennium Ecosystem Assessment; OECD, Organisation for Economic Co-operation and Development.



S-34 www.crops.org crop science, vol. 50, march–april 2010

1. Increasing the content of micronutrients in the edi-ble parts of crops through plant breeding (i.e., bio-fortification).

2. Producing protein-rich products by novel means, based on plant materials, as alternatives for ani-mal products.

3. Eliminating potentially toxic compounds in sta-ple foods.

4. Reducing constitutive or microbial toxins in selected staples that impact food quality, safety, and human health.

5. Evaluating biofortification strategies in the con-text of other approaches, such as diversification of diet, to improve the diets of nutritionally disadvan-taged people.

World food prices rose dramatically between 2000 and 2008 before beginning to decline later in 2008. A major cause of soaring food prices was the rapid growth in demand for biofuels, which has diverted land from food production. Other factors, many of them long term, have also contributed to the current food supply-and-demand situation. Rapid economic growth and urbanization, par-ticularly in Asia, have driven rapid demand for meat and for maize (Zea mays L.) and soybeans [Glycine max (L.) Merr.] for livestock feed. Improved economic growth in Africa has increased the demand for staples such as rice (Oryza sativa L.) and wheat (Triticum aestivum L.). Mean-while, agricultural productivity growth, especially in developing countries, continues to drop and the decline of global food stocks in the last 5 yr has led to very tense cereal markets, worldwide. Growing water scarcity and climate change are also increasingly affecting food pro-duction and prices. Poor and food-insecure households are among the hardest hit by rising food prices and, sub-sequently, by the global economic recession. Although households that are net sellers of food are benefiting, most poor households are net buyers of food.

This article gives some scenarios of the future of food toward 2050. Two scenarios are built on Agrimonde fore-sight models, which address challenges for feeding the world (Agrimonde, 2009). The third scenario ensues from analyzing alternative futures for agricultural supply and demand, and food security. This article ends summariz-ing food systems and functions, and how food technol-ogy addresses some of these global challenges affecting the supply of more nutritious and healthy diets.

agriculture and Food in the World oF 2050: ScenarioS and challengeS For a SuStainable developmentAgrimonde was established as a collective instrument—led by the French Initiative for International Agricultural

Research on behalf of the Institut National de la Recherche Agronomique and the Centre de Coopération Internatio-nale en Recherche Agronomique pour le Développement (CIRAD)—for analyzing global food and agricultural issues under the scenario of feeding 9 billion people by 2050, while preserving agro-ecosystems from which other services and products are expected (including cli-mate change, carbon storage, biodiversity, bio-energy, or bio-materials) (Chaumet et al., 2009). The variables con-sidered for the analysis are multifarious, including geopo-litical, social, cultural, sanitary, economical, agronomical, ecological, or technological, to cite just a few (Agrimonde, 2009). The global scale at which such issues are raised does not preclude reflections at the regional level, which are necessary to account for the diversity of the world’s food and agriculture, and their interactions, especially through trade that contains other key variables.

The classical scenario method is based on a first step of exhaustive recording of all kinds of variables likely to impact on the future of the system studied, within the timeline chosen for a future’s study (De Jouvenel, 2000). The classical method of scenario-building would not be suitable considering the number and the diversity of vari-ables, as well as the importance of articulating the regional and global contexts. This exercise would have been both unwieldy and largely indecipherable by combining the hypotheses on all the key variables for the future of a given agro-ecosystem investigated at both regional and global levels. The method was, therefore, adapted by building a tool based essentially on the complementarity of quan-titative and qualitative analyses. The quantitative mod-ule Agribiom was developed by B. Dorin and T. Le Cotty (CIRAD, Montpellier, France) by formulating quantita-tive hypotheses at the regional level, on a limited num-ber of variables, thereby reducing the complexity, while affording an entry point for in-depth qualitative reflec-tion on all the dimensions of the agro-ecosystem. This scenario-building considers the following main steps: (i) choosing scenarios and their underlying building prin-ciples, (ii) developing quantitative scenarios, and (iii) building complete scenarios by combining quantitative scenarios with qualitative hypotheses.

choice and principles of the ScenariosFor the 2006–2008 phase, the Agrimonde project chose the Millennium Ecosystem Assessment (MEA) scenarios, in particular Global Orchestration, to analyze it from the angle of food and agricultural systems, and to construct a single new scenario that departed from those of the MEA scenarios (Agrimonde, 2008). The MEA scenarios, which are references in international debates, were origi-nally built to study the future of ecosystems. Hence, they are not necessarily the most relevant for considering the future of food and agricultural systems. It is nevertheless

crop science, vol. 50, march–april 2010 www.crops.org S-35

friendly agricultural practices. These two scenarios are constructed differently; while Agrimonde GO is essentially a trend scenario starting from the current situation, Agri-monde 1 is built on the basis of sustainability objectives that are supposed to be met by 2050, and explores the trajecto-ries that would enable them to be attained.

Two underlying principles constitute the Agrimonde 1 and Agrimonde GO scenarios:

1. Assessing the capacity for each large region of the world to satisfy its food needs in 2050, thereby implying that interregional trade would be consid-ered only after evaluating the extent to which agri-cultural production in each region covered local needs.

2. Identifying the effects of future population trends independently of the large international migratory flows, so that the implications of expected popula-tion explosions could be examined fully with regard to each region’s capacity to feed its own population.

In its present form, Agrimonde 1 as a tool limits the construction of scenarios for the world’s food and agricul-ture in 2050 in several ways. First, there are no precise and complete quantitative estimations for the consequences of climate change on the world’s agriculture. Consequently climatic phenomena (greater variability, alterations in rainfall, rising temperatures, or melting of certain areas) have not been taken into account. Nonetheless, the panel of experts, inspired by the scenarios from the Intergov-ernmental Panel on Climate Change, modulated their hypotheses in relation to the surface areas under crops and to the possible yields in 2050 in the different regions. Sec-ond, even if the notion of pressure on natural resources is dominant in the analysis in various respects (e.g., defor-estation resulting from the extension of farmlands, water shortages induced by climatic and demographic changes, or deterioration of the quality of the soil and water caused by farming practices), the quantitative module does not integrate indicators of the consumption of natural resources, such as quantities of water or energy consumed.

Finally, Agrimonde 1 is based on the hypothesis that agricultural development is a driving force of global eco-nomic development and poverty alleviation (World Bank, 2008). This tool nevertheless enables us to verify whether the supposed regional increases in agricultural production effectively contributes to sufficient economic develop-ment, especially to avoid mass migration.

Food consumption in 2050In the Agrimonde scenarios, as in the MEA scenarios, “food availability” serves as an approximation of food consump-tion. It is calculated as the balance between the calorie equivalent of quantities of available foodstuffs (produc-tion + imports − exports ± stock variations) to feed the human population in a region (i.e., excluding animal

interesting to compare the two types of approaches: one regarding ecosystems and the other regarding the human activities that have the strongest impact on ecosystems.

As a baseline comparison, Agrimonde chose to reconstruct the MEA scenario Global Orchestration, which is a trend scenario on food consumption, but with different underlying societal priorities. Global Orchestration is the MEA scenario with the largest reduction of poverty and malnutrition. It is based on both the liberalization of trade and on major tech-nical advances in terms of agricultural yields. The priority given to economic development in this scenario, neverthe-less, results in an exclusively reactive management of ecosys-tems and environmental problems. This scenario was called Agrimonde GO because it was reconstructed on the basis of the quantification method adopted in Agrimonde, and because the population hypotheses chosen for this scenario are not precisely those used in the MEA.

The MEA scenarios are exploratory because they explore the consequences of changing trends by starting with the present situation. Some experts, including those involved in the MEA, indicated the need for a desirable scenario on the future of ecosystems. As a result, a new scenario (Agrimonde 1) was developed. The hypothesis of Agrimonde 1 uses as reference points a combination of the MEA scenario and the one proposed by Griffon (2006), who describes agriculture considering all characteristics of sustainability and the potential and conditions of a “doubly green revolution” (Conway, 1997). This type of agriculture would be characterized by agricultural pro-duction technologies that both preserve ecosystems and allow for development through agriculture in countries lacking capital, where the implementation of production systems requiring intensive use of equipment, pesticides, and fertilizers is limited. The same “population pressure” hypotheses are used for comparing the Agrimonde 1 sce-nario to the Agrimonde GO scenario.

Agrimonde 1 can be regarded as a normative forecast-ing scenario because it aims to explore the meaning and conditions of existence of a scenario on the development of a sustainable food and agricultural system. The idea was to better understand the meaning of such development, with the dilemmas and the main challenges that this type of scenario entail, and through the changes and disconti-nuities that it implies.

The World in 2050, as described in Agrimonde 1, is based above all on sustainable food conditions, allowing for the reduction of inequalities in food and health through a drastic reduction of both undernourishment and excessive food intake. The World in 2050 will need to implement a set of actions to intensify productive systems and to increase production in most regions. These actions will meet the fol-lowing objectives: satisfying the growing demand, allow-ing for the development of income from agriculture in rural areas of the Global South, and developing environmentally


feed, non–food uses, seeds, and postharvest losses), and the number of inhabitants of that region. It reflects the quantity of calories available to consumers, at home and through other channels, and includes calories that will be lost between the purchase of the products and their ingestion. It should not be confused with the quantity of calories actually ingested, which is difficult to estimate. In terms of ingestion, the net energy needs of a human being are around 2000 to 3000 kcal daily, depending on sex, height, weight, and intensity of physical activity.

Food consumption trends are very different between Agrimonde GO and Agrimonde 1 (Fig. 1). Agrimonde GO uses the hypotheses from the MEA Global Orchestration scenario in which economic growth largely explains con-sumption levels. Total availabilities at regional and world levels are given in the MEA report, but they were not split by product. Precise extrapolations were made in the Agri-monde report to quantify the food consumption hypoth-eses of the Agrimonde GO scenario. Agrimonde GO can qualify as a trend scenario in terms of the evolution of the total food calorie consumption, where economic growth boosts consumption in all regions to reach a mean global availability of 3590 kcal per capita daily and substantially reducing undernourishment.

The Agrimonde 1 scenario is clearly distinguished from the Agrimonde GO trend scenario. The income–food con-sumption nexus is not the main determinant because of great concerns for health, equity, and the environment. The hypothesis of food availability that the Agrimonde expert panel selected for 2050 is 3000 kcal per capita daily in all regions, notwithstanding certain regional particu-larities visible in the breakdown in terms of animal calorie sources (monogastric, ruminants, and halieutic). This set of hypotheses is in sharp contrast with the trends observed between 1961 and the beginning of the 21st century. It cor-responds to a slow growth of food availability per capita in most regions up to 2050, except in sub-Saharan Africa, where the per capita food availability will increase by 20% in 50 yr, and the Organisation for Economic Co-opera-tion and Development (OECD)–1990 region, where it will decrease by one-fourth (Fig. 1). The 3000 kcal are broken into 2500 kcal of plant products and 500 kcal of animal products. Within animal products, the proportion due to monogastrics is increasing in all regions, whereas the pro-portion due to ruminants is declining despite high levels in OECD-1990 countries, the former Soviet Union and Latin America, and an increase in sub-Saharan Africa. Calories of aquatic origin increased their share in varying propor-tions, which are linked to regional productive possibilities. Although the oceans are a considerable source of food pro-duction, fishing will face structural limits related to several factors (overfishing, artificialization of the littoral, pollu-tion, accelerated erosion of the biodiversity). It is assumed that marine aquaculture can increase at a faster pace than

it has over the past 40 yr, but at a different pace depending on the region. In Agrimonde 1, the pace of the development of marine aquaculture is high in Asia, OECD-1990, and Latin America, and moderate in the other regions. Relative stability in per capita availability of calories from freshwater fish is expected, as the existing (and increasing) tension over freshwater availability prevents any increase in freshwater fishing. Trends in relation to population increases in each region were thereafter calculated.

The set of hypotheses on food consumption assumes that people’s diets will depart from current tendencies as they take into account the objectives of sustainable devel-opment, which will ensue from the mounting pressure on resources and public health problems associated with human diets. It is a very strong set of hypotheses, as it implies that consumers, producers, and public policymak-ers will take into account the global and local impacts of modes of food production and consumption on health and the environment. This set of hypotheses corresponds to four challenges:

1. The wide gap between the observed availability and the nec-essary availability for food security. The actual mean daily availability in 2000 was close to 4000 kcal per capita daily in the OECD-1990 zone and just under 4500 kcal per capita daily in the United States, whereas the Food and Agriculture Organization (FAO) of the United Nations deems satisfactory a mean daily per capita availability of 3000 kcal to guarantee that each individual has sufficient healthy food (FAO, 2002). These gaps can be explained by the distri-bution of diets within the population, by the fact that in rich countries the 3000-kcal threshold may be simply exceeded, and by a great proportion of loss between the available food and actual consumption, linked to consumption habits.

2. The importance of equity in a sustainable development scenario. Instead of using the assumption suggested by Collomb (1999) that each region attains at least 3000-kcal per capita daily threshold, with some countries exceeding that level, Agrimonde chose to test a stronger hypothesis that there will be a con-vergence of average availabilities of food worldwide.

3. The food/health nexus. A daily per capita availability of 3000 kcal may have positive consequences in terms of public health by (i) maintaining the proportion of undernourished people at a relatively low level, thus reducing the risks of malnutrition in develop-ing countries; and (ii) limiting overconsumption, a source of nontransmissible food-related diseases such as obesity. Public actions aimed at changing food-related behaviors are a response to the current increase in obesity.

4. The relationship between diet and the pressure on natu-ral resources. The aim of adequately feeding 9 billion


people in 2050 implies that, irrespective of the pro-duction methods, there will be considerable pressure on natural resources that will increase along with the growing proportion of animal products in people’s diets. The production of animal calories requires a substantial volume of plant calories, water, and energy. In addition, breeding ruminants generates greenhouse gases directly or indirectly (e.g., through animal fodder, processing, and transport). This last component is increasing with the intensification of production. Caution is nevertheless required, con-sidering the environmental impact of animal pro-duction. One can also consider that there is an advantage in producing animals that optimize the use of plant resources (e.g., grazing on pastures, which humans cannot digest). Systems have, however, been intensified over the past 40 yr, which has resulted in shrinking pastures and concentrates, especially for grains. Producing ruminants still has the advantage of using land that is often unfit for crops (e.g., high altitudes, slopes, or semiarid areas), and of storing carbon on such lands. Furthermore, ruminants also have various uses because they represent a form of capital for their owner, provide organic fertilizer, are often used as draft animals, and are sources of food and regular income for populations often among the poorest in the world.

The World in 2050 in the agrimonde ScenariosThe analysis of scenarios, in terms of coherence and action levels, and their comparison, enabled the identification of certain qualitative hypotheses in the Agrimonde 1 scenario. On this basis, the factors that had not yet been considered in the analysis, but that were likely to have a decisive impact

on the world’s food and agriculture during the period leading up to 2050, were sought. These factors have been grouped into seven main themes: (i) the global context, (ii) international regulations, (iii) the dynamics of agricultural production, (iv) the dynamics of biomass consumption, (v) the actors’ strategies, (vi) knowledge and technologies in the field of food and agriculture, and (vii) sustainable development. A complete scenario was built by developing hypotheses on these different dimensions, with a concern for the overall coherence and plausibility of the scenarios. A possible account of the Agrimonde 1 scenario is proposed here, as well as that of Agrimonde GO, which corresponds to the MEA experts’ forecasts (Carpenter et al., 2005). This article will focus on Points vi and vii.

agrimonde go: Feeding the World by Making Global Economic Growth a PriorityThe global availability of calories for consumption as food, per day and per capita, will increase by 818 calo-ries between 2000 and 2050. The steepest increases will be in Asia, sub-Saharan Africa, and Latin America, and the number of children suffering from malnutrition in developing countries will decrease by a factor of 2.5 dur-ing the first half of the century. This trend, stimulated by the rapid economic growth and intense urbanization, will be accompanied by a richer protein content of diets as people consume more meat and fish. It will result in the growth of obesity in many regions (Asia, Africa), where new nutrition policies need to be implemented.

Technological development will allow for more intensive farming, as well as for an extensive use of fer-tilizers and genetically modified crops. The vast major-ity of farms, both small and large, will become highly mechanized and industrial. Local know-how will often

Figure 1. Mean regional food availability (daily kcal per capita) trends in Agrimonde Global Orchestration (AGO) and Agrimonde 1 (AG1) scenarios. The data used for this figure (1961–2003) ensued by reprocessing data from the Food and Agriculture Organization (FAO) of the United Nations. FSU: former Soviet Union States, LAM: Latin America, OECD: Organisation for Economic Co-operation and Development (or so-called developed world), MENA: Middle East and North Africa, SSA: sub-Saharan Africa.


be replaced by standardized industrial methods and the variety of agricultural species will be reduced. Multina-tional firms are a predominant feature of this scenario, as they will increase their share of plant and animal produc-tion, primarily through the development of new genetic strains. Nevertheless, it needs to increase the cultivated area by 18%, with yields close to 33,000 kcal ha−1 daily.

agrimonde 1: Feeding the World by Preserving EcosystemsIn 2050, diets in the various regions of the world would converge regarding calorie intake. On average about 3000 kcal per capita daily would be available worldwide. Cul-tural particularities would nevertheless maintain some diversity in the distribution of the various food sources. Increasing income would not lead to a convergence of diets toward western diets. Even though in certain regions, espe-cially in sub-Saharan Africa, food consumption trends are initially based on economic development, they also stem from behavioral changes in most regions. For instance, in a region like OECD-1990, the mean calorie consumption has declined from 4000 to 3000 kcal per capita daily. This steep downward trend is the result of less wastage by users or in catering systems, and more effective nutrition poli-cies. The maintenance of diversity of diets also helps to solve problems of deficiencies in micronutrients, primar-ily through the consumption of fruit and vegetables. The fast growth of the proportion of raw products compared with processed products, recorded at the beginning of the century, has leveled off. This is a symptom of the diver-sification of food systems. It also stems from regulations that have placed strong constraints on agri-food companies’ information and communication on nutrition in the rich countries, encouraging them to limit the degree of product processing, while continuing to sell products that are inno-vative, in terms of practicality and variety.

From 2000 to 2050, the agri-industrial model, initially clearly dominant, merges increasingly with the local food and agricultural systems based on short circuits and on the diversity of small and medium-sized farms and processing enterprises, especially in the developing world. The tendency toward standardization, internationalization, and concentra-tion around a limited number of multinational firms declines. This change is facilitated by national and regional strategies to ensure food security, and by the considerable impact of corporate social responsibility (CSR) on large firms’ strat-egies. The agri-food sector is strongly affected by CSR as consumers in the rich countries prove to be more and more concerned about food issues, due to the spread of the sustain-able food concept and following the “hunger riots.” They pressure agri-food firms, often via nongovernmental and consumer organizations, to take on their particular role in economic development and the reduction of malnutrition, as well as in the struggle against obesity. According to this

scenario, the increase of cropping area needed is almost 39% more than the current state, with yields varying from 20,000 to 30,000 kcal ha−1 daily. In this case, there is a huge need for new models of agricultural activities facing new ways of combining the ecological and productive functions of agro-ecosystems in the same area corresponding to a model that can be qualified as “integrationist.” It is based on the com-bination of different types of productive systems in a given territory, adapted to the local ecosystems in such a way as to maintain it in the form of a mosaic of ecosystems pro-ducing a diversity of services (e.g., purifying and regulating water resources, soil conservation, maintenance of landscape structures and biodiversity, or carbon fixation). This model involves different types of farming (such as livestock, forestry, or crops) in the same territory, on the same farm or on dif-ferent farms, overlapping to differing degrees (see the mode of ecological intensification in the Agrimonde 1 scenario for the North Africa–Middle East, sub-Saharan Africa, Latin America, and Asia regions).

ecological intensification, performance criteria, and (ir)reversibility of choicesToday the concept of ecological intensification essentially refers to tailoring technical options, rather than prescrib-ing a set of processes that can be applied uniformly every-where. These so-called technical options encompass social, economic, spatial, and political options that are not inci-dental and have probably not been sufficiently explored. However, enough is known about the options that have accompanied the process of rationalization (also known as “modernization”) of North American and European agriculture, so that they enable us to clarify the conditions required for a particular option.

In the Agrimonde 1 scenario, the agricultural perfor-mance criteria are no longer limited to tech-economic indicators. They encompass a range of indicators at the ter-ritorial level that pertain to the efficiency of agricultural practices regarding water quality, biodiversity, and soil quality conservation, as much as on commercialized pro-duction. In this scheme, the different types of productive systems described above are no longer exclusive, but they are complementary to each other by allowing for efficient management of the diversity of the ecosystems involved. The Agrimonde 1 scenario is a fine illustration of such com-plementarity. For instance, in Latin America forests are devoted no longer to clearing for land use or protection, but to intermediate forms corresponding to various agro-forestry models. In Asia humid areas are not all drained, but rather they are valued as a source of grazing land in dry seasons or for combined agricultural and aquaculture projects. In North Africa–Middle East and in sub-Saharan Africa, rangelands with low forage productivity become key elements in grazing routes that use a diversity of envi-ronments and biological corridors, enabling the fauna and


flora to circulate. The same applies to hedges, small woods, and orchards, habitats for many crop auxiliaries and coarse substances that preserve the soil and low-lying vegeta-tion from the effects of wind and rain. In the Agrimonde 1 scenario, farms with a low level of efficiency, in terms of exclusively tech-economic criteria, play an important role in this respect in 2050. They make the multifunctionality of agriculture fully meaningful; that is, not only a farming activity that provides goods and services apart from agri-cultural goods, whether for food or not, but also one of the activities practiced in a territory by some of the households living there. In this sense it is both the territory and the households that are multifunctional, as agriculture as such represents only one of these functions.

The Agrimonde 1 scenario integrates a change of view-point on the multifunctionality of agriculture, assessed as essential by both the recommendations of the 2008 Interna-tional Assessment of Agricultural Knowledge, Science and Technology for Development (IAASTD, 2008) and by the World Develop-ment Report 2008 (World Bank, 2008) on agricultural issues. One of the first tasks to make it meaningful would consist of producing performance criteria to evaluate the accomplish-ment of the different functions, not merely to remunerate them, but to frame them politically and to administer them. It would become evident that in such a scheme the differ-ent types of agriculture complement one another rather than having to fit into a single model (e.g., from commercial spe-cialized to family multipurpose farming).

In both scenarios, the question remains as to the real capacity for emerging new technology options, which are affected also by other factors including social, economic, and local development issues. It could prove difficult to break away from past choices that are embedded in current techni-cal solutions (e.g., mechanization, fertilizer and pesticide use, or genetic engineering) as well as in cognitive systems (such as knowledge and know-how, representations of nature, pol-lution, or landscapes) and in the values of the main actors involved. Are we not trapped in a technical rationalization? It is a sort of lock-in that other sectors have also experienced—except we cannot do without agriculture!

Food Supply and demand For Food SecurityThis section examines the new realities of the global food system, presenting the outcomes of the analysis of alterna-tive futures for agricultural supply and demand and food security to 2050, based on research done for the IAASTD (Rosegrant et al., 2009).

drivers that influence the Future of Food Supply and demand and Food Security

Drivers that influence the future of food supply and demand and food security include any natural or human-induced factors that directly or indirectly influence the

future of agriculture. Indirect drivers include demo-graphic, economic, sociopolitical, scientific and techno-logical, cultural and religious, and biogeophysical change. Important direct drivers include changes in food con-sumption patterns, natural resource management, land use, climate, energy, and labor. The key quantitative driv-ers in this scenario assessment are summarized below.

baseline Quantitative modeling assumptionsThe baseline case forecasts a world developing out to 2050 as it does today, without deliberate interventions requir-ing new or intensified policies. The key assumptions of the reference case include:

1. Population. The baseline (as well as alternative policy experiments) uses the United Nations medium vari-ant projections (United Nations, 2005), with the global population increasing from slightly more than 6.1 billion in 2000 to >8.2 billion in 2050. Popula-tion growth drives changes in food demand.

2. Overall economic growth. Economic growth assump-tions are based on the TechnoGarden scenario of the MEA (Carpenter et al., 2005). Incomes are expressed as MER-based values. The TechnoGarden scenario assumptions are near the midrange growth scenarios in the literature for the world as a whole and most regions. In some regions, such as sub-Saharan Africa, the scenario is relatively optimistic.

3. Agricultural productivity. Agricultural productivity values are based on the MEA (TechnoGarden sce-nario) and the recent FAO interim report projections to 2030/2050 (FAO, 2006). The MEA assumptions have been adjusted from the TechnoGarden sce-nario assumptions to conform to FAO projections of total production and per capita consumption in meats and cereals, and to our own expert assessment. The main recent technological change develop-ments, with continued slowing of growth overall, have been taken into account. Growth in numbers and slaughtered carcass weight of livestock has been similarly adjusted.

4. Nonagricultural productivity. In the reference case, in general, productivity growth is projected to be lower in nonagricultural than in agricultural sectors. The nonagricultural gross domestic product (GDP) growth rates are based on the MEA TechnoGarden scenario, but with adjustments to align with World Bank medium-term projections. While the relatively higher productivity in agriculture largely reflects the declining trends in agricultural terms of trade, this is not translated into higher output growth in agricul-tural sectors relative to nonagricultural sectors. This broadly confirms Engel’s Law that the budget share of food falls with increasing income.


5. Disparities in growth rates among developing coun-tries are projected to remain high, while more developed regions will see more stable growth out to 2050. Developed regions will see relatively low and stable to declining growth rates between 1 and 4% yr−1. Latin America is also expected to experience stable growth rates, though slightly higher than for developed regions—between 3.5 and 4.5% yr−1. The GDP growth in East and Southeast Asia is expected to be stable, with relatively high rates of 4 to 7% yr−1. In particular, China’s economy is projected to slow from the 10% growth in recent years to a more stable rate of 5.6% yr−1. On the other hand, growth in South Asia—following strong reforms and initiatives focusing on macroeconomic stabili-zation and market reforms—is expected to lead to improved income growth in that subregion of 6.5% yr−1. The Middle East and North Africa is expected to see GDP growth rates averaging 4% yr−1. Growth in sub-Saharan Africa has been low in the recent past, but there is room for recovery, which is pro-jected to lead to modest to strong growth just under 4% yr−1. Growth in Central and Western Africa is expected within the 5 to 6% range. Growth in East and Southern Africa is expected at <4% out to 2025, followed by more rapid growth of 6 to 9% by 2050.

6. Trade policies. Today’s trade conditions are presumed to continue out to 2050. No trade liberalization or reduction in sectoral protection is assumed for the reference scenario.

7. Climate change. Climate change is both driving dif-ferent outcomes of key variables of the baseline (such as crop productivity and water availability) and is an outcome of the agricultural projections of the refer-ence run, due to land-use changes and agricultural emissions, mainly from the livestock sector. Medium energy outcomes are assumed in the baseline. The B2 scenario was used for the analysis. From the available B2 scenario, the ensemble mean of the results of the HadCM3 model for the B2 scenario was used. The pattern scaling method applied was that of the Cli-mate Research Unit, University of East Anglia. The “SRES B2 HadCM3” climate scenario is a transient scenario depicting gradually evolving global climate from 2000 through 2100.

8. Biofuels. The baseline, based on actual national bio-fuel plans, assumes continued expansion in produc-tion of biofuels through 2025, although the rate of expansion declines after 2010 for the early rapid growth countries such as the United States and Bra-zil. Under this scenario, significant increases in bio-fuel feedstock demand occur in many countries for commodities such as maize, wheat, cassava (Mani-hot esculenta Cranz), sugar, and oil seeds. By 2020,

the United States is projected to put 130 million t of maize into biofuel production; European coun-tries will use 10.7 million t of wheat and 14.5 mil-lion t of oil seeds; and Brazil will use 9 million t of sugar equivalent. We hold the volume of biofuel feedstock demand constant starting in 2025 to repre-sent relaxed demand for food-based feedstock crops created by the rise of new technologies that convert nonfood grasses and forest products.

models used in the StudyTwo types of models were used for the study: partial agricultural equilibrium models and computable general equilibrium (CGE) models. Both types were used for analyses at the national (India and China) and regional or global levels. The partial equilibrium agricultural sector model—International Model for Policy Analysis of Agri-cultural Commodities and Trade, or IMPACT (Rosegrant et al., 2002)—provided insights into long-term changes in food demand and supply at a regional level, taking into account changes in trade patterns using macroeconomic assumptions as an exogenous input.

The IMPACT model was developed at the beginning of the 1990s, on the realization that there was a lack of long-term vision and consensus among policymakers and researchers about the actions that are necessary to feed the world in the future, reduce poverty, and protect the natural resource base. This model has been used in several important research publications, which examine the link-age between the production of key food commodities and food demand and security at the national level. The most comprehensive set of results for IMPACT are published in the book Global Food Projections to 2020 (Rosegrant et al., 2001). These projections are presented with details on the demand system and other underlying data used in the projections work, and cover both global and regionally focused projections. This IMPACT model was further expanded through inclusion of a water simulation model, as water was perceived as one of the major constraints to future food production and human well-being.

The Global Trade and Environmental Model (GTEM)—A CGE model, developed by the Australian Bureau of Agricultural and Resources Economics (Aham-mad and Mi, 2005), was used to validate the GDP and pop-ulation input data to achieve cross-sectoral consistency and to implement trade analysis. The GTEM is a multiregion, multisector, dynamic, general equilibrium model of the global economy, which addresses policy issues with global dimensions and issues where the interactions between sec-tors and between economies are significant. This includes international climate change policy, international trade and investment liberalization, and trends in global energy mar-kets. In addition, the IAASTD analyses used the integrated assessment model IMAGE 2.4 (Eickhout et al., 2006) for


climate change impacts and land use, and the livestock spatial location–allocation model SLAM (Thornton et al., 2002, 2006) for a more detailed livestock assessment.

baseline results

Food Supply and Demand

The baseline was a 3-yr average centered on 2000 for all input parameters and assumptions for driving forces. Fol-lowing this baseline, global cereal production increases 0.9% yr−1 for the 2000–2050 period. The year 2000 reflects a 3-yr moving average for 1999 to 2001, and 2050 reflects a 3-yr moving average of 2048 to 2050 unless noted oth-erwise. Growth of food demand for cereals slows during the 2000–2025 period and again from 2025 to 2050, from 1.4 to 0.4% yr−1. The demand for meat products (beef, sheep, goat, pork, and poultry) grows more rapidly but also slows somewhat after 2025, from 1.8 to 1% annually.

Changes in cereal and meat consumption per capita vary significantly among regions (Fig. 2 and 3). Over the projections period, per capita demand for cereals as food is expected to decline by 27 kg in East Asia and the Pacific and by 11 kg in Latin America and the Caribbean. On the other hand, demand is projected to increase by 21 kg in sub-Saharan Africa. Per capita meat demand is projected to more than double in South Asia and sub-Saharan Africa, almost double in East Asia and Pacific, and increase by 50% in the Middle East and North Africa. In the developed countries, only a minor 4% increase is projected, given that demand is already very high.

Total cereal demand is projected to grow by 1048 mil-lion metric t, or by 56%: 45% of this increase is expected for maize; 26% for wheat; 8% for rice; and the remainder for millet [Pennisetum glaucum (L.) R. Br.], sorghum [Sor-ghum bicolor (L.) Moench.], and other coarse grains. Rapid growth in meat and milk demand in most of the devel-oping world will put strong demand pressure on maize and other coarse grains as feed. Globally, cereal demand as feed increases by 430 million t for the 2000–2050 period; that is, a 41% of total cereal demand increase. Slightly more than 60% of total demand for maize will be used as animal feed, and a further 16% for biofuels.

How will expanding food demand be met? For meat in developing countries, increases in the number of ani-mals slaughtered have accounted for 80 to 90% of pro-duction growth during the past decade. Although there will be significant improvement in animal yields, growth in numbers will continue to be the main source of pro-duction growth. In developed countries, the contribution of yield to production growth has been greater than the contribution of numbers growth for beef and pig meat, while for poultry, numbers growth has accounted for about two-thirds of production growth. In the future, carcass weight growth will contribute an increasing share

of livestock production growth in developed countries, as numbers expansion is expected to slow.

For the crops sector, water scarcity is expected to increasingly constrain production, with little additional water available for agriculture due to slow supply increases and rapid shifts of water away from agriculture in key water-scarce agricultural regions in China, India, plus the Middle East and North Africa. Climate change will increase heat and drought stress in many of the current breadbaskets in China, India, and the United States, and even more so in the already stressed areas of sub-Saharan Africa. Once plants are weakened from abiotic stresses, biotic stresses tend to set in, and the incidence of pest and diseases tends to increase.

With declining availability of water and land that can be profitably cultivated, area expansion is not expected to contribute significantly to future production growth. In the baseline, cereal harvested area expands from 660 million ha in 2000 to 694 million ha in 2020 before con-tracting to 632 million ha by 2050. The projected slow growth in crop area places the burden to meet future cereal demand on crop yield growth.

Although yield growth will vary considerably by com-modity and country, in the aggregate and in most coun-tries it will continue to slow. The global yield growth rate for all cereals is expected to decline from 1.96% yr−1 in 1980–2000 to 1.01% in 2000–2050. In developed coun-tries, annual average cereal yield growth is estimated at 0.96% yr−1 during 2000 to 2050, 0.9% in East Asia and the Pacific, and 1.07% in South Asia. Slightly higher yield growth is expected in the Middle East and North Africa, Latin America and the Caribbean, and sub-Saharan Africa; that is, at 1.16, 1.25, and 1.59% yr−1, respectively. As can be seen in Fig. 4, area expansion is significant for projected food production growth only in sub-Saharan Africa (23%), Latin America and the Caribbean (9%), and the Middle East and North Africa (7%).

Food Trade, Prices, and SecurityIn the last few years, real prices of food have increased dramatically as a result of changes in biofuel/climate policies, rising energy prices, declining food stocks, and market speculation. Projections reported here show that higher food price trends are likely to stay as a result of increased pressures on land and water resources, adverse impacts from climate variability and change, and rapidly rising incomes in most of Asia. Given underinvestment in agriculture over the past few decades, and projected slow growth in investment in the baseline and poor govern-ment policies in response to rising food prices in many countries, it is unlikely that the supply response will be strong enough in the short to medium term.

Maize, soybean, rice, and wheat prices are projected to increase by 60 to 97% in the baseline (Fig. 5) and prices for


beef, pork, and poultry by 31 to 39%. Impacts of higher food prices on net food purchasers will be substantial, depress-ing food demand in the longer term, increasing childhood malnutrition rates, and reversing progress made in several low-income countries on nutrition and food security.

World food trade is expected to continue to increase, with cereals trade projected to increase from 257 million t in 2000 to 584 million t by 2050, and trade in meat prod-ucts rising from 16 million t to 64 million t. Expanding trade will be driven by the increasing import demand from the developing world, particularly sub-Saharan Africa, East Asia and the Pacific, and South Asia, where net cereal imports will grow by >200% (Fig. 6). Sub-Saharan Africa

Figure 2. Per capita availability of cereals as food in 2000 and change for 2000 to 2050 by region. EAP: East Asia and the Pacific, LAC = Latin America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA = sub-Saharan Africa.

Figure 3. Per capita availability of meats in 2000 and change for 2000 to 2050 by region. EAP: East Asia and the Pacific, LAC: Latin America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.

Figure 4. Sources of cereal production growth (2000–2050) by region. EAP: East Asia and the Pacific, LAC: Latin America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.

Figure 5. International food prices ($US t−1) of selected grains in 2000, and projected for 2025 and 2050.


will face the largest increase in food import bills despite the significant area and yield growth expected during the next 50 yr in the baseline. By 2050, the Middle East and North Africa is expected to account for 33% of net cereal imports, sub-Saharan Africa for 25%, and China for 19%.

With most developing countries unable to increase food production rapidly enough to meet growing demand, the major exporting countries—mostly in high-income countries and in Eastern Europe and Central Asia—will play an increasingly critical role in meeting global food consumption needs. The United States and Europe are a critical safety valve in providing relatively affordable food to developing countries. However, given the strong demand for food crops as feedstock for biofuels in the short to medium term, net cereal exports in these countries are projected to decline over the next decade before rebound-ing after food-crop use for biofuel feedstock is expected to decline. For example, net maize exports from the United States are expected to decline from 40 million t in 2000 to 17 million t in 2015 before rebounding and increasing to 62 million t by 2025. Net wheat exports are projected to grow to 48 million t in Russia, 41 million t in the United States, and to around 20 million t in Australia, Canada, Central Europe, and Kazakhstan. Net meat exports are expected to double in developed countries and to sharply increase in Latin America. Brazil’s net meat exports are expected to increase 10-fold over the 50-yr time horizon.

The substantial increase in food prices will slow growth in calorie consumption due to both direct price impacts and reductions in real incomes for poor consum-ers who spend a large share of their income on food. As a result, there will be little improvement in food security for the poor in many regions. In sub-Saharan Africa, daily cal-orie availability is expected to stagnate up to 2025 before slowly increasing to 2762 kilocalories by 2050, compared with 3000 or more calories available, on average, in most

other regions. Only South Asia (excluding India) fares worse, with only 2654 kilocalories available on average by 2050. Several regions are projected to experience declin-ing calorie availability between 2000 and 2025 (Fig. 7).

In the reference run, malnutrition among children up to 60 mo will continue to decline slowly in most regions, but remains high by 2050, with progress far below that envisioned in the Millennium Development Goals (Fig. 8). Childhood malnutrition is projected to decline from 149 million children in 2000 to 130 million children by 2025 and 99 million children by 2050. The decline will be greatest in Latin America at 51%, followed by Central/West Asia and North Africa, and East Asia and the Pacific at 46 and 44%, respectively. Progress is slowest in sub-Saharan Africa. By 2050, an 11% increase is expected—to 33 million children in the region—despite significant income growth and rapid area and yield gains, as well as substantial progress in supporting services that influence well-being outcomes, such as female secondary education and access to clean drinking water.

alternative inveStmentS in agricultural KnoWledge, Science and technology (aKSt)Three alternative AKST scenarios out to 2050 were analyzed to examine their implications for food sup-ply, demand, trade, and security. The first two scenarios examine the outcome of different levels of investments in crop yield and livestock numbers growth (AKST high and AKST low). A third scenario analyzes the implica-tions of even more aggressive growth in agricultural R&D together with advances in complementary sectors (AKST high plus). These include investments in irriga-tion infrastructure represented by accelerated growth in irrigated area and efficiency of irrigation water use, by accelerated growth in access to drinking water, and

Figure 6. Net trade in cereals in 2000 and projected for 2025 and 2050. EAP: East Asia and the Pacific, LAC: Latin America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.


greater investments in secondary education for females, an important indicator for human well-being (Table 1).

The AKST high variant that presumes increased investment in AKST, results in higher food production growth, which in turn reduces food prices and makes food more affordable to the poor when compared with the reference world. Under AKST high, cereal produc-tion increases by 7% and by an even stronger 17% under the AKST high plus variant. Under AKST high, rice prices decline by 46%, wheat prices by 57%, and maize prices by 65%, compared with the 2050 baseline value. On the other hand, if investments decrease faster than in the recent past, prices would rapidly increase, by 96% for rice, 174% for wheat, and 250% for maize compared with the 2050 baseline value (Fig. 9).

Despite these strong changes in AKST behavior, yield growth will continue to contribute most to future cereal production growth under both the AKST low and AKST high variants. Under AKST low, however, the contribution of area growth to overall production growth is projected to increase compared to the baseline, from 23 to 35% for sub-Saharan Africa, and from 11 to 29% in Latin America and the Caribbean. For developing countries as a whole, area change would contribute 13% to overall production growth, up from a negative 4% (contraction of area) under the baseline.

This growth, coupled with rapid expansion of the livestock population under AKST high, requires expansion of grazing areas in sub-Saharan Africa and elsewhere, which could lead to further forest conversion into agricultural use.

What are the implications of more aggressive produc-tion growth on food security and trade? Under AKST high, developing countries cannot meet the rapid increases in food demand through domestic production alone. As a result, net cereal imports from developed countries would increase by 70% compared with the reference run. Net cereal imports are projected to increase from 72 to 125 million t in sub-Saharan Africa, and from 93 to 100 mil-lion t in the Middle East and North Africa, but drop by almost half in China. Under AKST low, on the other hand, high food prices lead to depressed global food mar-kets and reduced global trade in agricultural commodities.

Sharp increases in international food prices as a result of the AKST low variant depress demand for food and reduce availability of calories. Average daily kilocalorie availability per capita declines by 850 calories in sub-Saharan Africa, pushing the region below the generally accepted minimum level of 2000 calories and thus also below the levels of the base year 2000. On the other hand, under the AKST high and AKST high plus scenarios,

Figure 7. Calorie availability in 2000 and projected for 2025 and 2050. EAP: East Asia and the Pacific, LAC: Latin America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.

Figure 8. Number of malnourished children in 2000 and projected for 2025 and 2050, baseline. EAP: East Asia and the Pacific, LAC: Latin America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.


calorie availability increases in all regions compared with 2000 and baseline levels.

Calorie availability—together with changes in comple-mentary service sectors, including female secondary educa-tion, female-to-male life expectancy at birth, and access to clean drinking water—can help explain changes in child-hood malnutrition levels (Rosegrant et al., 2001). Under the AKST high and AKST high plus variants, the number of malnourished children in developing countries is projected to decline by 24 and 56%, respectively, from 104 million children under the baseline (Fig. 10). On the other hand, if investments slow more rapidly and supporting services degrade rapidly, then absolute childhood malnutrition lev-els could return to close to 2000 malnutrition levels at 137 million children in 2050 under the AKST low variation.

What are the implications for investment under these alternative policy variants? Investment requirements for developing countries in the baseline run for key sectors, including public agricultural research, irrigation, rural roads, education, and access to clean water, are calculated at nearly US$32 billion per year at 2008 prices. As Fig. 11 shows, the much better outcomes in developing country food security obtained under the AKST high plus variant can be achieved at estimated annual investment increases in the five key sectors of US$20 billion and are within reach if the political will and resources are made available.

an analysis of Food productionThe following trends are of importance in discussing the future of food and the role of technology. An enormous

Table 1. Assumptions for baseline and alternative Agricultural Knowledge, Science and Technology (AKST high, low) and infra-structure (AKST high plus) scenarios.

Parameter changes for growth rates Base 2050 AKST high 2050 AKST low 2050 AKST high plus 2050

Gross domestic product growth

3.06% yr−1 3.31% yr−1 2.86% yr−1 3.31% yr−1

Livestock numbers growth Base numbers growth of ani-mals slaughtered 2000–2050

Livestock: 0.74% yr−1

Milk: 0.29% yr−1

Increase in numbers growth by 20%

Increase in animal yield by 20%

Reduction in numbers growth by 20%

Reduction in animal yield by 20%

Increase in numbers growth by 30%

Increase in animal yield by 30%

Food crop yield growth Base yield growth rates 2000–2050:

Cereals: 1.02% yr−1

Roots and tubers: 0.35% yr−1

Soybean: 0.36% yr−1

Vegetables: 0.80% yr−1

Fruits: 0.82% % yr−1

Increase growth by 40% Reduce growth by 40% Increase growth by 60%

Irrigated area growth 0.06 Increase by 25%

Rainfed area growth 0.18 Decrease by 15%

Basin efficiency Increase by 0.15 by 2050Access to water Increase annual rate of

improvement by 50% relative to baseline level

Female secondary education Increase overall improvement by 50% relative to 2050 base-

line level

Figure 9. Cereal prices (US$ t−1) in 2050 per alternative Agricultural Knowledge, Science and Technology (AKST) scenarios.


productivity increase has occurred in highly effective and efficient farming systems. This improvement has resulted in unprecedented food security in the industrialized world but seems to lead to other problems as indicated below:

1. Industrialization of farming leads to alienation of consumers toward their food.

2. Strong integration in food chains, also internation-ally, and consequently strong interdependencies.

3. Because of the abundant food supply (in the industri-alized world), the role of the consumer has become leading and necessary. In other words, food systems are no longer supply driven but demand driven.

4. When food security is realized, food quality becomes of importance, which materializes in a strong atten-tion for the relation between food and health.

Food SystemsFood systems can be described as comprising four sets of activities: producing food, processing food, packaging and dis-tributing food, and retailing and consuming food (Ingram, 2008). The first activity of producing food is basically the production of raw materials in agriculture, horticulture, ani-mal husbandry, and aquatic production systems. The main

scientific disciplines involved are plant and animal breeding, agronomy, soil science, water management, phytopathology, and related disciplines. The main actors involved are farmers, seed companies, fertilizer and pesticides industry.

The second activity is about the processing of raw materials into food. This activity starts after harvesting and four subactivities can be distinguished (Van Boekel, 1998): stabilization, transformation, production of ingredients, and production of fabricated foods. Stabilization implies that measures are taken to prevent spoilage. The most important cause of spoilage is microbial activity, which is even dangerous as the microorganisms may be pathogenic and are a threat to human health. This threat concerns the very important aspect of food safety. Hence, many food technology activities are directed toward the prevention, or at least inhibition, of microbial growth. When microbial growth is prevented, chemical and biochemical reactions are the next cause of spoilage. This spoilage implies oxida-tion reactions and the so-called Maillard reaction, leading to desired changes such as browning and flavor compounds, and undesired changes such as loss of nutritive value and toxicological suspect compounds. Biochemical changes occur as a result of enzyme activity, which can lead to color,

Figure 10. Number of malnourished children, developing countries, projected 2000 to 2050 for alternative Agricultural Knowledge, Science and Technology (AKST) scenarios.

Figure 11. Annual investment requirements (2000–2050) for agriculture and complementary service sectors using alternative Agricultural Knowledge, Science and Technology (AKST) scenarios (US$ billion in 2008).


flavor, taste, and texture problems. Transformation implies that raw materials are changed into something different, such as milk into cheese, wheat into bread, or barley into beer. Production of ingredients is self-evident: sugar can be extracted from sugar beets (Beta vulgaris L.) and sugarcane (Saccharum officinarum L.), or protein and oil from soybeans. This activity can also imply production with the help of microorganisms or enzymes. Finally, production of fabri-cated foods implies that foods are composed or designed from several raw materials (sauces, desserts, pastry are some examples). Disciplines involved are food science and tech-nology and nutrition, actors involved are artisanal as well as industrial processors.

The third activity of packaging and distribution fol-lows immediately after transformation. Some authors would actually consider packaging to be part of pro-cessing. In any case, packaging is an essential element of food technology as it protects the food against all kinds of threats from the environment. These threats include microorganisms, insects, water, oxygen, and physical damage. Likewise, packaging technology can nowadays be actively used to preserve foods by applying controlled atmosphere and modified atmosphere. This implies that the gas atmosphere influences metabolic reactions of the food as well as of the microorganisms present in a desired direction. Furthermore, packaging also functions as infor-mation carrier for the consumer, such as nutritive value, presence of possible allergens, and any other relevant information, including advertising. Disciplines involved are food science, logistics, marketing, and actors are food processors, middlemen, and retailers.

The last but not least activity is about retailing and con-sumption. Retailers have quite some power these days, as they are able to influence the consumer directly by deter-mining what to offer to consumers. They seem to have appreciable governance in the food chain. Increasingly, glo-balizing markets are becoming important, as foods from all over the world may end up at the consumer’s plate. While this is not a bad development as such, this phenomenon has several institutional implications, such as access to mar-kets and a strong effect of rules and regulations that make it sometimes difficult for developing countries to comply with this development (Ruben et al., 2007). It also raises the question whether or not such developments are not enhancing sustainability problems. Actors involved are food processors, retailers, and consumers, and governments, to some extent, when regulation is involved.

Functions of FoodsFoods have many functions in society. First and foremost, they supply energy and nutrients that humans need to live. People need the macronutrients protein, fat, and carbo-hydrates. Next to that, the micronutrients vitamins and minerals are essential. Furthermore, nonnutrients such as

fiber, antioxidants, and other bioactive components are needed. However, not all proteins, fats, and carbohydrates are the same. Generally, animal proteins (meat, milk, eggs) are better from a nutritional point of view than plant proteins. Fats also differ, and polyunsaturated fatty acids, and especially the w-3/6 fatty acids, are preferred over the saturated ones. Carbohydrates that are absorbed in the gut only supply energy to the body. Dietary fibers can also be classified as complex carbohydrates that are not absorbed but are partly fermented in the colon.

Another function of food is to supply pleasure. Gener-ally, people enjoy eating and foods deliver stimuli to the senses (eyes, tongue, nose, ears). Childhood experiences appear to be very important to what people like in later years, and also cultural habits have a big influence.

Food is also a very important way to express social rela-tions. Hospitality is expressed by offering food and drinks to visitors and friends. It is also a way to distinguish oneself by (not) eating certain foods, usually dictated by religion.

the Future of FoodWe have discussed briefly the various aspects of food produc-tion, processing, distribution, and consumption. The key-words are food security (is there enough food), food safety (is the available food safe to consume), and food quality (is the food of such a quality that it can fulfill the need of the consumer). Humankind is capable to produce enough food for the whole world population; in other words, food secu-rity can be realized, in principle. In practice, however, this appears not to be possible due to socioeconomic and politi-cal problems. Food safety and food quality are manageable, again in principle. The question is now how future develop-ments can help in increasing food security, food safety, and food quality. A rapidly upcoming problem related to food production is sustainability: Are we able to produce food in such a way that also future generations are able to fulfill their needs of food without depleting Mother Earth?

Food and SustainabilityThe key sustainability issues in food production are optimal use of raw materials (with as little waste as possible while still satisfying consumer demands), efficient usage of water, energy, packaging materials, and processing aids, and eco-nomic efficiency in line with social and cultural values. A recent concern is that the production of meat and milk is contributing considerably to environmental problems. One reason is the emission of gases that are involved in climate change problems (especially methane). Another reason is the inefficient conversion of plant proteins in animal feed into animal proteins for human consumption (Aiking et al., 2006). This problem is complicated and the simple solu-tion is not to cut down animal production in developing countries, as the contribution of small-scale animal farming to livelihood: animals, which are a major source of food,


provide fuel and manure to be used as fertilizer in develop-ing countries. In the industrialized world, however, devel-opments of meat alternatives are a possible solution or part of a solution. It could also be of interest for upcoming coun-tries such as India and China.

Food and healthThe relationship between food, nutrition, and health is obvious on the one hand, but much is still unknown. First of all, an individual food cannot be called unhealthy or healthy, but diets can. In other words, it is the combination of several foods in a diet that determines what is healthy. In developing countries, health problems are related mainly to insufficient energy intake and lack of micronutrients. In the developed world, the main problem is overconsumption and too little fiber. Too many calories are taken in with a minimum of physical exercise, thus leading to obesity and ultimately to so-called “diseases of civilization” such as coronary heart disease and diabetes. It is not only food, by the way, but also lifestyle that determines the incidence of obesity. Incidentally, obesity is also increasing in certain populations in the developing world. Obviously, health is also related to food safety if microorganisms cause diarrhea with associated problems, and other food intoxications.

the role oF Food technologyHow could food technology help to alleviate some of the problems that we face? First of all, knowledge of what causes postharvest losses will help to tackle this problem. It is estimated that 30 to 40% losses occur in developing countries and, therefore, measures to avoid this will have a big effect on food security as well as on food safety (Eng-strom and Carlsson-Kanyama, 2004; Kader, 2005). One of the problems with postharvest spoilage is that micro-organisms produce toxins (such as the carcinogenic com-pound aflatoxin) that are really very dangerous to human health (Williams et al., 2004). Prevention of this would definitely help increase food security as well as safety (Ortiz et al., 2008).

Second, when raw materials are processed into foods, desired and undesired things happen. Desired effects are increased digestibility, increased food safety because of elimination of pathogens, and increased shelf life. Unde-sired effects are destruction of essential nutrients, and losses of resources (excessive waste). Because lack of micronutri-ents is a serious problem in the developing world (Kennedy et al., 2003), making micronutrients more bioavailable, possibly by adding them to food, and preventing losses of these compounds, would make a major contribution to alleviate inadequate nutrition.

Third, if food processing of local crops can be con-nected to demand of urban consumers—that is, by align-ing food processing to consumers’ wishes—this link would offer the opportunity to raise income and earn a

living for the local processors and producers. In doing so, it is essential that food safety and quality can be guaran-teed. Food technology can help in realizing this.

Fourth, institutional barriers to access markets (regionally and internationally) could be tackled by investing in quality and safety by technological measures, and to have knowledge about the products produced so that real safety problems can be distinguished from trade barriers in disguise.

A very important aspect with all preservation tech-nologies is packaging. It forms the barrier between the food and its environment, and it can protect the food from recontamination and other undesired influences from the environment (such as oxygen). Packaging has therefore a large effect on food quality as well as on food safety. Food technology can help in adjusting packaging technology to what is needed for a particular food.

FocuSing the agri-Food reSearch agenda oF the 21St centuryTable 2 highlights some issues related to food security and safety, the link between food and health, and sustainabil-ity of agro-ecosystems for both the industrialized and the developing world. In this regard, agri-food system research should focus on using better local crops for food produc-tion, reducing postharvest losses substantially, optimizing local processing in such a way that the nutritional value (especially bioavailability of micronutrients) is improved as well as the eating quality, linking local production and processing to urban consumers by delivering safe, nutri-tious, and good-quality food according to consumers’ demands, improving processing and storage or packaging to ensure food safety (i.e., absence of pathogens and con-taminating chemicals), dealing with institutional barriers for access to markets, and making more sustainable global food production systems.

concluSionSThis background article for the Science Forum 2009 has provided some scenarios of the future of food for the mid-21st century. This article has also briefly addressed the major developments in food production, especially the possible role of food technology, and based on this analy-sis, some issues for shaping a priority agri-food research agenda have been identified. Next to technological issues, it should be realized, however, that institutional barriers can be a major obstacle for the developing world to gain access to local, regional, and global markets. At the same time, it is also clear that technology could help to over-come these institutional barriers by making it possible to produce high-quality, safe, and nutritious food. If it is pos-sible to realize such a development, it should also be pos-sible to reduce poverty by linking urban food demand in the developing world to local food production.


acknowledgmentsThe authors thank the colleagues who kindly perused an early version of this manuscript, and Dr. Haruko Okusu (CGIAR Science Council Secretariat) for her kind editing of the final version of this manuscript.

referencesAgrimonde. 2008. Why a CIRAD-INRA foresight study on

world food and agricultural systems in 2050? Institut National de la Recherche Agronomique–Centre International de Recherche Agronomique pour le Développement, Paris.

Agrimonde. 2009. Scenarios and challenges for feeding the world in 2050. Available at http://www.fao.org/fileadmin/templates/wsfs/docs/expert_paper/Inra__Cirad__2009_-_ Agrimonde_Summary_Report__draft_.pdf (verified 21 Dec. 2009). Institut National de la Recherche Agronomique–Cen-tre International de Recherche Agronomique pour le Dével-oppement, Paris.

Ahammad, H., and R. Mi. 2005. Land use change modeling in GTEM: Accounting for forest sinks. ABARE Conf. Pap. 05.13. In Workshop on EMF 22: Climate Change Control Scenarios, Stanford, CA. 25–27 May 2005. Stanford Univ., Stanford, CA.

Aiking, H., J. De Boer, and J. Vereijken. 2006. Sustainable protein production and consumption: Pigs or peas. Springer, Dordrecht, Netherlands.

Carpenter, S.R., P.L. Pingali, E.M. Bennett, and M.B. Zurek (ed.) 2005. Ecosystems and human well-being: Scenarios. The millennium ecosystem assessment. Vol. 2. Island Press, Wash-ington, DC.

Chaumet, J.-M., F. Delpeuch, B. Dorin, G. Ghersi, B. Hubert, T. Le Cotty, S. Paillard, M. Petit, J.-L. Rastoin, T. Ronzon, and S. Treyer. 2009. Agriculture and food in the world of 2050—Scenarios and challenges for a sustainable develop-ment. Institut National de la Recherche Agronomique–Centre International de Recherche Agronomique pour le Développement, Paris.

Collomb, P. 1999. Une voie étroite pour la sécurité alimentaire d’ici à 2050. Economica, Paris.

Conway, G. 1997. The doubly green revolution—Food for all in the twenty-first century. Penguin Books, London.

De Jouvenel, H. 2000. A brief methodological guide to scenario building. Technol. Forecast. Soc. Change 65:37–48.

Eickhout, B., H. van Meijl, and A. Tabeau. 2006 Modeling agricul-tural trade and food production under different trade policies.

p. 61–76. In A.F. Bouwman et al. (ed.) Integrated modeling of global environmental change. An overview of IMAGE 2.4. Netherlands Environ. Assessment Agency (MNP), Bilthoven.

Engstrom, R., and A. Carlsson-Kanyama. 2004. Food losses in food service institutions. Examples from Sweden. Food Policy 29:203–213.

FAO. 2002. Agriculture mondiale: Horizon 2015/2030. Food and Agric. Organ. of the United Nations, Rome.

FAO. 2006. World agriculture: Towards 2030/2050. Interim rep. Food and Agric. Organ. of the United Nations, Rome.

Griffon, M. 2006. Nourrir la planète—Pour une révolution dou-blement verte. Odile Jacob, Paris.

IAASTD. 2008. Agriculture at a crossroads: Executive summary of the synthesis report. Int. Assessment of Agric. Knowledge, Sci., and Technol. for Dev. Island Press, Washington, DC.

Ingram, J.S.I. 2008. Food system concepts for the ESF/COST Forward Look on European food systems in a changing world. Work document ESF-COST Forward Look Work-shop, Brussels.

Kader, A.A. 2005. Increasing food availability by reducing post-harvest losses of fresh produce. Acta Hortic. 682:2169–2175.

Kennedy, G., G. Nantel, and P. Shetty. 2003. The scourge of “hid-den hunger”: Global dimensions of micronutrient deficien-cies. Food Nutr. Agric. 32:8–16.

Ortiz, R., T. Ban, R. Bandyopadhyay, M. Banziger, D. Bergvin-son, K. Hell, B. James, D. Jeffers, P. Lava Kumar, A. Menkir, J. Murakami, S.N. Nigam, H.D. Upadhyaya, and F. Wali-yar. 2008. CGIAR research-for-development program on mycotoxins. p. 415–424. In J.F. Leslie et al. (ed.) Mycotoxins: Detection methods, management, public health and agricul-tural trade. CABI Publishing, Wallingford, UK.

Rosegrant, M.W., X. Cai, and S.A. Cline. 2002. World water and food to 2025: Dealing with scarcity. Int. Food Policy Res. Inst., Washington, DC.

Rosegrant, M.W., M. Fernandez, and A. Sinha. 2009. Look-ing into the future for agriculture and KST. p. 370–376. In B.D. McIntyre et al. (ed.) IAASTD Global Rep. Island Press, Washington, DC.

Rosegrant, M.W., M.S. Paisner, S. Meijer, and J. Witcover. 2001. Global food projections to 2020: Emerging trends and alter-native futures. Int. Food Policy Res. Inst., Washington, DC.

Ruben, R., M. van Boekel, A. Van Tilburg, and J. Trienekens (ed.). 2007. Tropical food chains. Governance regimes for qual-ity management. Wageningen Academic Publ., Wagenin-gen, Netherlands.

Thornton, P.K., P.G. Jones, T. Owiyo, R.L. Kruska, M. Herrero,

Table 2. Overview of relevant issues for future of food and the role of technology.

Issue Developing world Developed worldFood security Is a big issue

Postharvest lossesPostprocessing packaging and storage

No issue anymore: tackled by industrialization of food production

Food safety Pathogenic microorganismsContaminating chemicals (pesticides, environmental

contaminants such as dioxin, PCBs†)

Mainly pathogenic microorganisms

Food and health MicronutrientsMacronutrients

Overconsumption leading to obesity and associated health problems

Sustainability Soil depletion and use of fertilizersWater qualityWaste of food

Fossil energy useWater use

Making better use of resourcesWaste of food

†PCBs, polychlorinated biphenyls.


P. Kristjanson, A. Notenbaert, N. Bekele, A. Omolo, with contributions from V. Orindi, B. Otiende, A. Ochieng, S. Bhadwal, K. Anantram, S. Nair, V. Kumar, and U. Kulkar. 2006. Mapping climate vulnerability and poverty in Africa—Report to the Department for International Development. Int. Livestock Res. Inst., Nairobi.

Thornton, P.K., R.L. Kruska, N. Henninger, P. Kristjanson, R.S. Reid, F. Atieno, A. Odero, and T. Ndegwa. 2002. Mapping poverty and livestock in the developing world. Int. Livestock Res. Inst., Nairobi.

United Nations. 2005. World population prospects: The 2004 revi-sion. United Nations, Population Division of the Dep. of Eco-nomic and Social Affairs of the United Nations Secretariat,

New York.Van Boekel, M.A.J.S. 1998. Developments in technologies for

food production. p. 87–116. In W.M.F. Jongen and M.T.G. Meulenberg (ed.) Innovation of food production systems. Wageningen Press, Wageningen, Netherlands.

Williams, J.H., T.D. Phillips, P.E. Jolly, J.K. Stiles, C.M. Jolly, and D. Aggarwal. 2004. Human aflatoxicosis in developing coun-tries: A review of toxicology, exposure, potential health conse-quences, and interventions. Am. J. Clin. Nutr. 80:1106–1122.

World Bank. 2008. World development report 2008: Agriculture for development. Int. Bank for Reconstruction and Dev., Washington, DC.


symposia

Malnutrition is the most chronic and pressing agricultural and human health problem of the 21st century. Undernu-

trition, micronutrient malnutrition, and unbalanced overnutri-tion with excessive consumption of carbohydrates and fats affect at least one third of the world’s population in a negative manner and impinge on both longevity and the quality of life with good health. Yet in different regions of the world, in developing and developed countries alike, a similar change in diet is occurring:

Relearning Old Lessons for the Future of Food—By Bread Alone No Longer: Diversifying Diets

with Fruit and Vegetables

John D. H. Keatinge,* Farid Waliyar, Ramni H. Jamnadas, Ahmed Moustafa, Maria Andrade, Pay Drechsel, Jacqueline d’A. Hughes, Palchamy Kadirvel, and Kartini Luther

aBsTRaCTDiversifying diets and agricultural enterpriseswith fruit and vegetables is a potent weaponinthecurrentglobalbattleagainstmalnutritionandpoverty.Agriculturalsciencecancontributesubstantiallytoenhancethedevelopmentpros-pectsandhealthofnotonlydisadvantagedandvulnerable individuals at oneendof the spec-trumbutalsothegrowthandequityofnationaleconomiesattheother.Moreover,withrelativelysimpleappliedresearch,newcropspeciesandtechnologiescanrapidlyenterthedevelopmentpathwaytobenefiteventhepoorestpeopleornations. More upstream research can help toguardfruitandvegetableproductionagainstthevagariesofpotentialclimaticuncertainty,whichis projected to become more prominent overfuture decades. However, historical and con-tinuingwidespreadunderinvestmentinfruitandvegetable researchanddevelopment from thenational to theglobal levelmayseverely com-promise the world’s ability to use such high-valuespecies forcropdiversificationandasamajorengineofdevelopmentgrowthtoensureglobalfoodandnutritionalsecurity.

J.D.H. Keatinge, J.d’A. Hughes, P. Kadirvel, and K. Luther, AVRDC – The World Vegetable Center, PO Box 42, Shanhua, Tainan, 74199 Tai-wan; F. Waliyar, ICRISAT, Patancheru 502 324, Andhra Pradesh, India; R.H. Jamnadas, ICRAF, United Nations Ave., PO Box 30677, Muth-aiga, Nairobi, 00100, Kenya; A. Moustafa, ICARDA-APRP, PO Box 13979, Dubai, U.A.E.; M. Andrade, CIP, Apartado 1558, Lima 12, Peru; Pay Drechsel, IWMI, 127, Sunil Mawatha, Pelawatte, Battaram-ulla, Sri Lanka. Received 21 Sept. 2009. *Corresponding author ([email protected]).

Abbreviations: AVRDC, Asian Vegetable Research and Development Center; CGIAR, Consultative Group on International Agricultural Research; CIP, Centro Internacional de la Papa, International Potato Center; CWANA, Central and West Asia and North Africa; GAP, Good Agricultural Practices; HIV/AIDS, human immunodeficiency virus/acquired immunodeficiency syndrome; ICARDA, International Center for Agricultural Research in the Dry Areas; ICRAF, World Agroforestry Centre; ICRISAT, International Crops Research Institute for the Semi-Arid Tropics; IITA, International Institute of Tropical Agriculture; IPPM, Integrated Production and Protection Management; IWMI, International Water Management Institute; NARES, National Agricultural Research and Extension Systems; QTL, quantitative trait loci; WHO, World Health Organization.




as incomes rise above a critical level, people are consum-ing more calories, typically from energy-dense foods such as cereals, meat, fats, and oils. The same negative trend toward carbohydrate-dominated diets is also being seen where poverty reduces incomes below the critical thresh-old for food security. In both cases the result is damaging to human health; the reasons for this have been known and understood for centuries and yet the problem of mal-nutrition persists. Innovative modern science both at the upstream and downstream ends of the research-to-devel-opment spectrum needs to be brought to bear globally to tackle this intransigent and highly undervalued problem.

The World Bank projects per capita income in devel-oping countries will grow at an annual rate of 3.4% from now through 2015. During that same period, the aver-age daily calorie intake in these countries is expected to increase by approximately 200 Kcal. Over the past 40 yr, net imports of meat, sugar, wheat, and vegetable oils by developing countries increased by 13 times and are expected to keep growing (FAO, 2004a).

Currently, most African countries have tuber/root or cereal-based diets with starchy staples making up about 60% of total food consumption (FAOSTAT; http:// faostat.fao.org/default.aspx [verified 31 Dec. 2009]). As per capita income increases in these countries, diets are likely to change in a way similar to those seen in past decades in other developing countries.

Daily calorie intake in European nations over the past 40 yr has increased by about 20%, from 2960 Kcal to 3340 Kcal. As populations in southern Europe, North Africa, and the Middle East have grown more affluent, diets previously rich in fresh fruit and vegetables now incorporate more meat and fat, are higher in calories, and contain greater amounts of high-glycemic carbohydrates (Schmidhuber, 2008). This is particularly an issue in the younger generations under 35 yr of age, where the les-sons of good nutritional understanding associated with the mid-20th century global conflict and austerity appear to have been forgotten. Moreover, with increasing urban-ization, diets across the world are becoming increasingly similar, characterized by a greater dependency on staple grains (particularly wheat and rice), increased consump-tion of meat, dairy products, edible oils, salt, and sugar, and a lower intake of dietary fiber (FAO, 2004a).

A balanced human diet is a diverse diet. Humans need to consume an adequate amount of calories, amino acids, fatty acids, minerals, vitamins, and fiber for good health. A balanced diet usually includes both plant and animal products and should include a diverse range of the five major food groups—grains, vegetables, fruits, dairy, and meat—to supply the nutrients necessary to sustain and sup-port healthy growth and activity. Dietary diversity enables people to meet their daily nutritional requirements. As the typical contents of diets worldwide converge and become

less diverse, the need to encourage and promote dietary diversification becomes more urgent and necessary.

RELEaRNiNG oLD LEssoNs: WHy DiVERsiFy DiETs WiTH FRUiT aND VEGETaBLEs?

Some efforts to increase nutrient intake have been made by biofortification and nutrient supplementation. Biofortifying staple crops is increasingly used as a comple-mentary strategy for combating micronutrient malnutri-tion among the poor in developing countries (Bouis, 2002). Maize (Zea mays L.) biofortified with β-carotene and rice (Oryza sativa L.) biofortified with iron or β-carotene are examples of efforts to increase micronutrient uptake in staple-heavy diets. Supplements are commonly available in developed countries and can be used to provide the required micronutrients among populations with nutrient deficiencies in developing countries but at a relatively high cost. It has been shown that administration of vitamin A capsules can reduce child mortality by up to 23% (S. Tan-umihardjo, personal communication, 2008).

Although supplementation and fortification programs have been successful in many countries, the majority of the world’s population lives in rural areas or peri-urban areas with limited access to, or the income to use, health ser-vices, vitamin or mineral supplements, and processed food. Rather, most poor people in rural communities still rely heavily on home-grown produce and inexpensive food pur-chased from local markets. In these situations, food-based approaches using locally available sources for diet diversi-fication are the most cost effective and quickest means of providing the essential micronutrients needed for health.

The importance of fruits and vegetables in the human diet has been recognized by the World Health Organiza-tion (WHO), which promotes and recommends the con-sumption of at least 400 g of fruit and vegetables per day to provide the necessary nutrients lacking in other food groups (World Health Organization, 2003a, b). Yet vegetables and fruit are the food groups mostly left out when a person’s capacity to obtain more energy-dense food increases.

Underconsumption of fruit and vegetables is among the top ten risk factors leading to micronutrient malnu-trition and is associated with the prevalence of chronic diseases (Ezzati et al., 2002; World Health Organization 2003b). Fruit and vegetables contain a range of macro- and micronutrients, including provitamin A, iron, and zinc, which contribute to the prevention of malnutrition disorders. Fruits and vegetables also are rich in bioactive phytochemicals that can reduce the risk of chronic dis-eases such as cancer.

Diets featuring a diversity of vegetables and fruit improve the health of hungry and poor people and afflu-ent consumers alike, as both groups can be malnour-ished. Consuming fruit and vegetables supplies multiple


quality fruits and vegetables to provide viable, practical, and sustainable solutions for these issues.

Nutrient ContentFruit and vegetables provide a range of vitamins and

nutrients necessary for human health. Low incomes and limited awareness of the importance of nutrition further constrain diet diversification. Modest improvements in the density of micronutrients in the most commonly con-sumed fruits and vegetables would increase the benefits of diet diversification. Plant breeding can develop fruit and vegetable lines with increased nutrient contents and con-centrations of bioactive compounds (Yang et al., 2007). Food preparation methods can be modified to increase the bioavailability of nutrients to the body (Yang et al., 2002). This is particularly relevant to vegetables where cooking methods can often substantially reduce the nutritional benefits of food, particularly the availability of vitamin C.

Some plant-based nutrients are not in a form that can easily be absorbed by the human body. The body’s absorp-tion of plant-based iron is generally lower than that of iron from meat and is greatly influenced by interactions with enhancers and inhibitors. For some vegetables such as amaranth (Amaranthus spp.) and crucifers (Brassicaceae family), cooking can double iron bioavailability and in some cases even increase it by ten times (Yang et al., 2002). Prolonged storage of cooked vegetables will reduce iron bioavailability. Vegetables with different levels of iron bio-availability can be cooked together to enhance the overall iron intake (Yang et al., 2002). In India, where mung-bean [Vigna radiata (L.) R. Wilczek], which is low in iron bioavailability, is a staple, the Asian Vegetable Research and Development Center (AVRDC) has designed recipes combining mungbean with tomato (Solanum lycopersicum L.) and pepper (Capsicum spp.), which are high in iron bio-availability, to increase the dish’s overall iron bioavailabil-ity (Subramanian and Yang, 1998 Bains et al., 2003).

The International Potato Center (CIP) promotes the orange-fleshed sweet potato [Ipomoea batatas (L.) Lam.], which contains high levels of provitamin A carotenoids (100–1,600 μg Retinol Activity Equivalent per 100 g), and the vitamin A formed on consumption is bioavailable (Hagenimana et al., 1999; Van Jaarsveld et al., 2005). Sweet potato is a familiar crop in sub-Saharan Africa, where it is grown for food security, thus farmers already understand sweet potato cultivation methods. This has eased the task of disseminating the nutrient-rich orange-fleshed vari-ety. However, the value of new orange types has had to be established as information on the importance of dietary vitamin A is lacking and innovative extension methods and educational campaigns, particularly among women’s groups, are being undertaken. This innovation has led to the “branding” of orange in society in Mozambique as a high-value, desirable product. Cars, market barrows and

essential nutrients simultaneously, compared to only one or two nutrients provided in supplements or biofortified food. The intrinsic, irreplaceable dietary fiber in fruit and vegetables also promotes good digestion, slows the rate at which sugar is absorbed into the bloodstream (USDA, 2005), and plays a role in reducing the incidence of cancer of the digestive tract (Park et al., 2007). Micronutrients are also important as they act as coenzymes necessary for the metabolism of macronutrients

Micronutrient malnutrition—also referred to as hid-den hunger—is seen in food-insecure households in much of the world and can be highlighted as a particular concern in sub-Saharan Africa. It has long been known that micro-nutrient deficiency inflicts anemia, IQ reduction, and blindness on millions of people. It damages the immune system, resulting in birth defects for thousands of chil-dren, complications during pregnancy and childbirth, and ultimately the deaths of an estimated four million children and mothers annually. More generally, it results in large-scale loss of energy, intellect, productivity, growth, and economic prospects for millions of people. Furthermore, overcoming or coping with malaria and human immu-nodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) can be facilitated with good nutrition, so the success of HIV/AIDS programs also depends in part on paying more attention to good and balanced nutrition.

Deficiencies of vitamin A, iron, zinc, and iodine are the four main micronutrient deficiencies affecting human health in Africa. Many Africans still consume insufficient amounts of these nutrients, even though they are only required in relatively small quantities. Fruit and vegetables provide substantial amounts of two of these micronutrients, vitamin A and iron, and also provide several other nutrients including vitamin C, vitamin E, phosphorous, calcium, and vitamin B complex. In addition, fruit and vegetables provide hundreds of naturally occurring substances (e.g., dietary fibers, easily digestible sugars, and antioxidants) that are not commonly available in food supplements and have a high nutrient and low energy density but protect against chronic health conditions (Nyambo et al., 2005). More-over, many of these micronutrients are not widely avail-able in other foods; almost 90% of vitamin C in the typical healthy diet comes from fruits and vegetables.

aGRiCULTURaL CoNsTRaiNTs NEEDiNG iNNoVaTiVE soLUTioNs

Diversifying diets through better access to fruit and vegetables is a challenge global agriculture cannot ignore. Inefficient production methods and a lack of well-adapted seed contribute to poor yields and environmental and health issues, while low quality standards and limited market access thwart opportunities for increasing farmers’ income. Research and development is a requirement to counter the bottlenecks to the availability of safe and high


stalls, shop walls, T-shirts, dresses, headgear, and other clothing have all been used to get the orange-fleshed sweet potato known and desired by adults and children. A recent campaign of this type is almost unknown in the agriculture field. However, knowledge of the nutritional value of sweet potato leaves as a vegetable and its easy means of prom-ulgation is also lacking in sub-Saharan Africa, although well-known in Southeast Asia. Germplasm of leafy sweet potatoes is being screened by AVRDC’s Regional Center for Africa in Tanzania for productivity, farmer acceptabil-ity, and nutrient quality. A “green” branding campaign may now be needed for such indigenous green leafy vegetables.

AVRDC has also developed high β-carotene tomato lines using a single β-carotene gene from wild tomato. The orange-fruited tomato lines contain 3.8 to 6.6 mg of β-carotene per 100 g fresh weight compared with 0.6 to 0.9 mg per 100 g in the common, red-fruited tomato. Heat tolerance and multiple disease resistance have also been bred into the lines. These lines are being promoted for home and school gardens, but this demands innovative research, extension, and education methods as it frequently involves the most disadvantaged sections of society. Nev-ertheless, the investment in a single packet of improved vegetable seed is an input within the reach of even the poorest farmer and its return in terms of either improved family nutrition or income can be substantive.

Integrated packages of vegetable seeds that provide a balanced source of micronutrients for families year round from only 100 m2 of land have been developed by AVRDC for different regions of the developing world. These “healthy gardening kits” are a central feature of ongoing farmer training programs in Africa, and over 35,000 such kits have been distributed through humani-tarian agencies to the victims of major disasters in Africa and Asia since 2000. They include seeds of locally adapted varieties of nutrient-rich, fast-growing vegetables and technical information in local languages on vegetable production, food preparation, and preservation methods. Vegetables have been selected that are commonly grown in many tropical and subtropical less-developed coun-tries, are nutritious, hardy, fast-growing with low input requirements, are and relatively free of pests and diseases. Many of these are indigenous vegetables.

Expanding the existing use of indigenous vegetables can be a very important means of improving family nutri-tion. In many cases, rural communities have a high rate of indigenous vegetable consumption, which may well average at least 50% of all vegetables consumed according to AVRDC national surveys in Asia and Africa. How-ever, lack of access to suitable seeds or planting material and knowledge of the nutritional benefits of these often “neglected” crops can be major factors in their failure to be sufficiently exploited in the battle against malnutrition. AVRDC and its partners are presently involved in a 10

yr effort to expand the availability of lines and varieties of these crops and to test whether they may have use-ful roles in areas beyond the current centers of produc-tion and diversity. Associated efforts for mainstreaming such crops in urban supermarkets are also ongoing. Such efforts include crops rich in vitamins and minerals such as amaranth, slippery cabbage [Abelmoschus manihot (L.) Medik.] and its relatives okra [A. esculentus (L.) Moench] and roselle (Hibiscus sabdariffa L.), drumstick tree (Mor-inga spp.), Malabar spinach (Basella alba L.), African egg-plants (Solanum aethiopicum L. and relatives), bitter gourd (Momordica charantia L.), African spider plant (Cleome gyn-andra L.), giant nightshade (Solanum scabrum Mill. and rela-tives), and many others.

Amaranth leaves have been a traditional food across Africa for centuries. New, improved lines of amaranth introduced by AVRDC, A. cruentus L., A. hybridus, and A. dubius Mart. ex Thell., were released through two local seed companies in East Africa. The new varieties can be harvested as a whole plant in less than a month and are sweeter and softer than the leaves of old varieties, faster to cook, and extremely nutritious. Besides providing the community with cheap, highly nutritious vegetables, heavy demand for the new varieties—known locally as “White Elma” and “Green Gina”—is helping these com-panies thrive and is contributing to the growth of a viable vegetable seed sector in the region.

In harmony with such efforts on vegetables, the World Agroforestry Centre (ICRAF) has made a broad range of potential introductions of indigenous fruit trees, which often allow harvestable fruit that can be enjoyed through-out a greater proportion of the year. This is in contrast to the condition common in West Africa today of a short term glut followed by many months of no mature fruit. To facilitate these introductions, propagation methods have also been developed for tree improvement and multiplica-tion for key African fruit tree species. The propagation methods developed employ the horticultural techniques of vegetative propagation (rooting of cuttings, grafting, and marcotting) that have the merits of perpetuating over generations desirable traits such as early fruiting, lower branching habit, and dwarfing, elements that contrib-ute to enhanced fruiting on many fruits species. In West Africa for example, these techniques are being applied to species such as Irvingia gabonensis (Aubry-Lecomte ex O’Rorke) Baill., Dacryodes edulis (G. Don) H. J. Lam, Cola spp., Garcinia kola Heckel, Prunus africana (Hook. f.) Kalk-man, Pausinystalia johimbe (K. Schum.) Pierre ex Beille, Annikia chlorantha, Ziziphus mauritiana Lam., and Vitellaria paradoxa C. F. Gaertn.. All these species make a very posi-tive contribution to the amelioration of malnutrition par-ticularly the supply of vitamins A and C.

For most rural communities in sub-Saharan Africa, seasonality of fruit ripening is a major cause of the low


availability and consumption of fruits. ICRAF stud-ies have demonstrated that because different fruit species (indigenous and exotic) ripen at different times, it may be possible to provide farming communities with fruit all year round. For example, in southern Africa, of the 40 indigenous fruits studied, at least one or more species ripened every month throughout the year. Based on such information and nutritional content of the specific fruits (focusing on the major micronutrients such as vitamins A and C and iron), site-specific fruit tree portfolios that pro-vide fruits all year round could be developed and tested.

Fruit tree portfolios thus comprise a number of fruit tree species with at least one species yielding fruit at any given time, ensuring fruit availability all year round. Based on nutritional analysis of the fruits, the selected species would be those that have fruits containing the essential micronu-trients for maximum nutritional and health benefits. There is a wide selection of indigenous and exotic fruit tree spe-cies in all regions of sub-Saharan Africa that can be used in developing of year-round fruit tree portfolios.

Such fruit tree portfolios, packaged with informa-tion on threshold nutritional needs of fruits at the family level, can contribute to raising the current low level of fruit consumption in sub-Saharan Africa and also ensure that the WHO standard amounts of fruit consumption for health and well-being are met. In addition, these year-round fruit supplies (fresh or processed) may contribute to mitigating nutritional crises and contribute to reduc-ing micronutrient deficiencies. Moreover, surplus fruit, either fresh or processed, could be an alternative source of income for farming communities.

Climate adaptationEnvironmental stress is a primary cause of crop losses,

reducing average yields for most major fruit and vegeta-ble crops by more than 50% (Boyer, 1982; Bray et al., 2000). The severity of environmental stress imposed on crops is influenced by climatic uncertainty and change. Extreme geophysical events such as more erratic rainfall patterns, unpredictable high temperature spells, reduced water availability, and the postulated shift in ecological and agro-economic zones will further reduce produc-tivity of fruit and vegetables (FAO, 2004b). Enhancing agro-ecosystem resilience to sustain fruit and vegetable production systems and increase productivity in the face of environmental uncertainty as the climate changes is a prudent and necessary course of action.

AVRDC has made significant contributions to the development of heat-tolerant tomato and Chinese cab-bage (Brassica rapa L.) lines and the subsequent release of adapted, tropical varieties worldwide. However, lower yields in heat-tolerant lines remain a concern. Research by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), the International Center for

Agricultural Research in the Dry Areas (ICARDA), and other institutions is in progress to breed drought-tolerant varieties that use water efficiently. Genetic factors underly-ing drought tolerance in Solanum chilense (Dunal) Reiche and S. pennellii Correll, two wild tomato species indigenous to arid and semiarid environments of South America, are being explored by ICRISAT and AVRDC with the aim of transferring drought tolerance into cultivated tomatoes. Exploration of a large set of representative germplasm from ICRISAT’s mandate crops such as pigeonpea [Cajanus cajan (L.) Millsp.] and pearl millet [Pennisetum glaucum (L.) R. Br.] has also revealed considerable variation in salinity toler-ance, another potential breeding objective. Marker-assisted selection is being used to screen for desired salt tolerance genes, and attempts to transfer quantitative trait loci (QTLs) and elucidate the genetics of salt tolerance are continuing. ICRAF is presently using conventional participatory fruit tree domestication to take advantage of high phenotypic variation within natural tree stands for selecting superior mother trees suitable for propagation under a range of pres-ent or potential environmental conditions.

Adapting agricultural management practices to the production challenges posed by climate variability and change such as limited rainfall and/or irrigation water, flooding, and salinity is vital for future harvests. ICARDA promotes protected agriculture in arid regions with con-trolled growth environments using less water and fertilizer and careful production timing to improve yield. AVRDC has also developed protected production systems for use with excessive rainfall and grafting to increase flood tol-erance. Management strategies have also been developed such as modification of fertilizer application, efficient delivery of water to roots through water-saving irrigation methods, and use of soil amendments to enhance nutrient uptake by plants and improve soil fertility.

Cultivation of indigenous vegetables, fruits, and other underutilized crops is being promoted as many of these plants naturally have higher levels of resilience to envi-ronmental stress. These underutilized crops can grow and produce adequately under poor soil conditions, for example in drought-prone or saline areas, and often are more resistant to pests and diseases. In many instances, indigenous vegetables can grow in environments unfit for other crops. They are immediate, strong candidates to withstand the adverse effects of a changing climate.

Since spatial distribution models are not available for the majority of tree species (including fruit tree species), and since information is not available on the ecological limits of most of these species, scientists at ICRAF are developing vegetation maps as useful proxies for the spatial distribution of individual species. By associating indigenous fruit tree species to specific vegetation types, a spatial decision sup-port tool was developed for the highlands of Kenya, allow-ing fruit tree species selection based on expected ecology


of sites and species. The same approach is currently being expanded to eastern Africa (Ethiopia, Kenya, Malawi, Uganda, Rwanda, Tanzania, and Zambia) and later glob-ally. At the same time, by modeling the spatial distribu-tion of specific vegetation types, predictions on the future suitable domains for individual fruit species can be made. Where point information is available on the suitability of individual tree species (even as presence-only data sets), it is also possible to combine such point information with the spatial distribution of vegetation types in which the species is known to occur to improve suitability maps and predic-tions under climate change scenarios. Such information is also required for vegetable species but is presently lacking.

pests and pathogensPests and diseases are not new problems in vegetable

and fruit production systems. However, in many countries the scope and breadth of phytosanitary precautions have not kept pace with the rising local demands for exotic veg-etables as well as global trade, leading to the inadvertent introduction of pests and diseases into new areas. Climate change has altered the dynamics of pest populations, crop resistance, and other factors, creating more complex chal-lenges for the management of pests and diseases. Even in a country such as Australia that practices the most stringent biosecurity measures, the occurrence of Tomato yellow leaf curl virus in urban Brisbane for the very first time in Australia is of significant concern given the devastating nature of the virus globally and the importance of the winter season tomato industry in Queensland (D. Persley, personal communication, 2009).

Integrated pest management and integrated produc-tion and protection methods have been applied to prevent pest infestations, control disease outbreaks, and reduce damage. These approaches include improved soil man-agement practices, use of biological control and biopes-ticides, and breeding varieties resistant to biotic stress. Integrated pest management research has been conducted at the International Institute for Tropical Agriculture (IITA), the International Centre of Insect Physiology and Ecology, ICRISAT, ICARDA, and AVRDC to control legume pod borers (Maruca vitrata, Helicoverpa armigera, etc.) in vegetable legumes. This integrated management method uses major parasitoids—such as the braconid wasp (Apanteles taragamae) and tachinid fly (Nemorilla maculosa)—along with Bacillus thuringiensis endotoxins, MaviMNPV (a pod borer-infecting nucleopolyhedrovirus discovered by AVRDC), and synthetic sex pheromones to control the legume pod borer in Southeast Asia and sub-Saharan Africa. Genetically modified brassicas to increase their resistance to diamondback moth (Plutella xylostella) and other secondary lepidopterans and genetically modified cowpea to increase resistance against pod borer (M. vit-rata) are under evaluation stage. Hopefully such vegetable

varieties will prove to be effective and achieve public acceptance as an alternative to the current biohazardous practices in many developing countries with chronic and injudicious excess spraying of often out-of-date, obsolete, inappropriate, or banned pesticides that are injurious to producers and consumers’ health.

Bacterial wilt (caused by Ralstonia solanacearum) can easily kill 100% of tomato plants in the fields and quickly destroy any hope of harvest from the cultivation. Graft-ing susceptible tomato varieties onto bacterial wilt-resistant root stock (e.g., tomato variety ‘Hawaii 7996’) is one effec-tive way to manage the disease. Scientists from the Institute of Agricultural Science of Vietnam learned the grafting technology from AVRDC and introduced the technology to farmers in the Lam Dong province from 2002 to 2004. By 2007, approximately 4000 ha of tomatoes in Lam Dong were cultivated with grafted seedling. The grafting pro-vided Lam Dong farmers with an additional US$6 million of annual profit (Vinh and Ngo, 2006).

Breeding for resistance is still the best defense against fruit and vegetable diseases. From 2006 to 2008, AVRDC researchers released 21 new lines of pepper in the Interna-tional Chili Pepper Nursery with resistance to one or more of the following diseases: Cucumber mosaic virus, Chilli veinal mottle virus, Potato virus Y, Tomato mosaic virus, bacterial wilt, and anthracnose. Research is in progress to fine-map and stack Tomato yellow leaf curl virus resistance genes Ty-1, Ty-2, and Ty-3 to study gene complementation and develop breeding lines with multiple resistance genes.

In the special case of protected agriculture, the warm and humid greenhouse environment is not only suitable for crop production but also is naturally suitable for the development and sustained presence of many pests and dis-eases. Likewise, overuse, misuse, and careless application of chemical pesticides can result in the accumulation of toxic substances on plants. Decreased pesticide use should thus be desirable as it results in less potential harm to human health and the environment and should help to prevent pesticide resistance building up in target organisms. To address such problems, ICARDA has developed and intro-duced an Integrated Production and Protection Manage-ment (IPPM) program to Central and West Asia and North Africa (CWANA) countries in the Arabian Peninsula, Yemen, Afghanistan, and Pakistan. IPPM is an approach to good management in protected agricultural structures and aims to provide improved means and ways of greenhouse production and crop protection with the use of safe proce-dures and less chemical pesticides. All factors that can have a direct effect on plant growth and the overall climate of greenhouses, and that may lead to the spread of pests and diseases, are taken into consideration in the IPPM technol-ogy approach. Irrigation and fertilization thus are impor-tant components of IPPM packages. In Yemen, protected agriculture technologies with an IPPM program were


introduced and successfully practiced by growers (Mukred et al., 2003). Vegetable plants were healthy and had low incidences of pests and disease (Dubaie et al., 2003). Apply-ing IPPM techniques in Oman also resulted in better yields (ICARDA-APRP, 2002).

production methodsFarmers’ limited access to improved, superior varieties

and lack of knowledge about the best cultivation and man-agement practices further constrain fruit and vegetable pro-duction. Understanding local production systems and the resources available to farmers within these systems is essen-tial to develop technological interventions that will max-imize returns to inputs, reduce seasonality, and promote year-round availability of fruit and vegetable products.

Improved production packages for fruit and vegetables have been developed by many agricultural research centers and typically include superior germplasm, balanced fer-tilization, seedling production, mulching, irrigation, and management of pests and diseases with integrated, sustain-able approaches. AVRDC proposes effective rhizosphere management practices to deliver water and nutrients effi-ciently and effectively to the zone surrounding the roots of plants, where complex interactions occur among the plant, the soil microorganisms, and the soil itself. Effec-tive rhizosphere management is one solution to maximize the productivity of low-input farming systems common in developing countries. Nutrient delivery methods such as starter solutions and rhizosphere soil packets of com-post, biomass-derived charcoal, clay, etc., and methods to enhance beneficial soil microbial communities are being evaluated to promote better soil and plant health.

The introduction of superior fruit tree germplasm from Asia into Africa is an important element of improving fruit production. Lack of access to germplasm has remained a key constraint for smallholder agroforestry adoption. By working with local commercial suppliers of seedlings, the ICRAF project will also address the development of smallholder “germplasm delivery systems” for tree species in Africa (Graudal and Lillesø, 2007). A recent example of the success of this approach has been the introduction by ICRAF and ICRISAT of improved Ziziphus mauritiana Lam. germplasm from Asia into Sahelian countries that is grafted onto local rootstocks. This crop, which is rich in vitamin C, grows in exceedingly harsh and dry condi-tions and is now being commonly marketed in Niger and Mali as “Pomme de Sahel.” Crucial to the implementation of such good production practices is the adaptation of the technologies to be suitable for various agro-ecological pro-duction environments. Promotion of improved technolo-gies through participatory research trials, training courses, and workshops will encourage dissemination of the adapted technologies to vegetable and fruit producers.

ICRAF is currently developing methods for land health surveillance to systematically identify land health problems and target preventive and rehabilitation pro-grams. Fruit trees are an attractive intervention for reha-bilitating degraded land profitably. Part of ICRAF’s land health surveillance will diagnose and assess the prevalence of soil nutritional constraints to the growth of fruit pro-duction. Infrared and X-ray spectroscopy provide tools that make large area assessments and monitoring feasible by analyzing large numbers of soil and leaf samples at low cost. ICRAF scientists are currently developing and dem-onstrating protocols for practical application of these tools in surveillance systems.

Soil is the most available and normal growing media for plants. The soil’s main functions are to pro-vide anchorage, nutrients, air, and water to plant rooting systems. However, soils may also pose serious limitations to plant growth. Plant diseases caused by soil organisms, unsuitable soil fertility, salt accumulation due to irriga-tion, unfavorable soil compaction, and poor drainage may cause substantial reductions in fruit and vegetable pro-ductivity. In greenhouses, poor soil management almost inevitably increases soil-borne pests such as nematodes and accumulation of salinity. ICARDA, through its research for development program in close collaboration with the National Agricultural Research and Extension Systems (NARES) of many countries in CWANA, has developed, enhanced, and then simplified a number of soilless production techniques for protected agriculture. Soilless culture (hydroponics), when employed on a large scale, can be very challenging and needs informed man-agement and sensitive measurement tools to be effective. However, ICARDA has been innovative and succeeded in enhancing and simplifying such techniques such that they are now suitable for application by small-scale grow-ers. The simplification and consequent lower investment needs are the key innovative features of this technology, making it more acceptable to farmers and improving its impact. A range of alternate hydroponic systems for growing cash crops, including cucumber (Cucumis sativus L.), tomato, pepper, musk melon (C. melo L.), green bean (Phaseolus vulgaris L.), and strawberry (Fragaria ´anannasa), were tested at research stations. Soilless culture has not only shown itself to be a practical solution for many soil problems but also has managed to significantly increase water use efficiency and productivity. Such techniques are now being transferred to growers in many countries.

The vertical soilless production system has been adopted by a number of small-scale growers for produc-tion of strawberries and green beans. Up to a five fold increase in yield has been recorded with a reduction of 98% in water use and 63% reduction in fertilizer con-sumption and a reduction of 59% in general production cost compared with traditional soil beds (Moustafa et al.,


2006). A study on the productivity and quality of cucum-ber was performed by ICARDA with three different soil-less production systems in the United Arab Emirates. The primary production data for cucumber and their compari-son with the normal soil bed at the research station show the significant superiority of soilless culture. Soilless sys-tems not only increase yields but also speed up production. Water productivity in the soilless system reached 81.6 kg m–3 while with soil beds it was only 31.3 kg m–3 (Moustafa et al., 2006), thus showing significantly increased water use efficiency.

Transportation and packagingFruit and vegetables are highly perishable commodi-

ties. Postharvest losses in vegetables from field to consumer have been reported to be as high as 50% in developing countries (Kader, 2003). Inappropriate postharvest han-dling and packaging and lack of access to adequate trans-portation infrastructure are some of the reasons for high postharvest losses. Postharvest mishandling and time lags due to the distance from farm to market can cause pro-duce quality and nutritional value to deteriorate, jeopar-dize food safety, and reduce the value of the products.

It is crucial to disseminate postharvest technolo-gies adapted to suit local conditions and materials and to enhance fruit and vegetable producers’ knowledge of good postharvest practices. Better postharvest handling methods can be as simple as harvesting leafy vegetables during the coolest time of day, using smooth instead of rough-surfaced containers to hold produce, and using sturdy containers to avoid crushing when stacked during transportation.

Farmers from the Kiensvay district, Cambodia, used to harvest their tomatoes around noon, let them sit in the sun while waiting for the collector to arrive around 5 pm, and then load them in bamboo baskets at about 200 kg fruits per basket onto cars or trucks. About 10 kg fruits per bas-ket were lost to physical damage. A training on postharvest technology, conducted by AVRDC, introduced AVRDC tomato line ‘CLN1462A’, which has high resistance against handling damage. Farmers in the area now harvests their tomatoes at mature green to breaker stage during the cooler part of the day and place the harvested fruits under a shaded area where the fruits are sorted based on maturity. Instead of bamboo basket, containers with smooth surfaces are used during harvest and hauling. The tomatoes are packed in plastic bags at 20 to 22 kg and careful handling and loading into the transport vehicles are practiced. With the higher price of CLN1462A and the avoided losses during posthar-vest, farmers increased their net income from $750 in 2006 to $1,750 in 2007 (AVRDC, 2008).

In Cambodia, Laos, and Vietnam, AVRDC has worked with tomato farmers to evaluate varietal differences in storage quality and resistance to mechanical damage during handling and transport, test the most appropriate

packaging, and improve storage practices. To simulate handling hazards, a “drop test” was conducted by drop-ping a container packed with tomatoes three times from a height of 1.5 m. Fruit of varieties with a higher level of firmness sustained the least damage when a grid polysty-rene crate was used with shredded paper as cushioning material (AVRDC, 2008). Precooling improved the stor-age behavior of tomato. Modified atmosphere packaging and evaporative cooling storage techniques were adapted and optimized for fresh tomato and chili (Capsicum frutescens L.). Low-cost solar driers were fabricated and tested for chili with considerable success. Opportunities for increased use of solar drying techniques in developing countries appear to be radically underexploited at present. Additionally, one of the reasons for the rapid penetration of the tomato varieties ‘Tengeru 97’ and ‘Tanya’, developed by AVRDC, which presently dominate the Tanzanian market, can be ascribed to their thicker skins and better transportability to market under local rough road conditions.

marketsDomestic and regional markets in developing countries

are growing rapidly and are substantially larger in volume and value than global export opportunities. Efficient mar-ket-oriented production and effective market interventions therefore offer significant opportunities for poverty alle-viation and increased consumption of fruit and vegetables for more vulnerable community members such as women and children in developing countries. However, domestic and regional market supply chains for vegetables tend to be inefficient, long, and complex (Koenig, 2008; J. Lenne and A. Ward, unpublished data, 2008). The major constraints include a high level of postharvest losses; high transac-tion costs due to inadequate infrastructure; nonexistent or inefficient market information systems; the low bargain-ing power of farmers; and large seasonal price and volume fluctuations due to undeveloped processing industries. AVRDC, ICRAF, ICRISAT, and ICARDA are address-ing these constraints by supporting farmer organizations, enhancing available market information systems to increase farmers’ leverage, identifying innovative approaches to link farmers and processors, strengthening farmers’ management capabilities and skills, and conducting research to develop or adapt low-cost processing technologies.

In Cameroon, ICRAF has conducted a value chain analysis to identify four important intervention areas for production and consumption of the indigenous fruit “njansang” [Ricinodendron heudelotii (Baill.) Heckel]: (i) the introduction of improved harvest and postharvest tech-niques that improve product quality; (ii) the grading of produce according to market-determined norms; (iii) the strengthening of farmer producer groups to allow direct negotiations with wholesalers; and (iv) the introduction of a range of services, including market information and


credit, to help farmers to negotiate higher prices and enable them to store produce for sale to avoid market gluts (Akinnifesi et al., 2008). Action in each of these areas has generated higher value for producers. For example, the prices received for njansang kernels increased by 31% as a result of collective sales. A similar approach to improve inefficient banana market chains has been successfully adopted in Kenya and Uganda.

Food ContaminantsDiversifying diets with fruit and vegetables becomes

counterproductive if the food itself contains contaminants detrimental to human health. Pesticide residues, micro-bial contaminants, heavy metals, and toxins are examples of harmful food contaminants that can, in extreme cases, cause death. A study conducted across Ghana’s cities showed that 33% of the lettuce sold in urban markets had residues of banned pesticides and nearly all lettuce showed pathogen levels exceeding common standards (Amoah et al., 2006). Good Agricultural Practices (GAP) to ensure safe produc-tion and postharvest management of vegetables and fruit should be emphasized to ensure produce is wholesome and free of contaminants. Such theoretical positions are easy to postulate but prove to be extraordinarily difficult to opera-tionalize in the developing world where lack of knowledge of the implications of injudicious activities such as over-spraying are common among small-scale growers.

Reducing pesticide use in fruit and vegetable culti-vation helps reduce the level of pesticide residue in crop harvests. Host plant resistance, biological control, sex pheromones, and mechanical controls are some alterna-tives to the use of pesticides. An integrated pest manage-ment strategy to control eggplant fruit and shoot borer developed by AVRDC involves proper field sanitation, prompt disposal of infected shoots throughout the season, installation of traps baited with sex pheromone, and with-holding insecticide use to allow proliferation of the pest’s natural enemies. This integrated pest management effort resulted in a 70% reduction of pesticide use in eggplant production in Bangladesh (Alam et al., 2006). Adapting and disseminating the technologies to other crop-produc-ing regions can lead to further reductions in pesticide resi-dues on vegetables and fruit.

Mycotoxin contamination is a widespread problem in developing countries in the tropics (Waliyar et al., 2003). Such toxins are a secondary metabolite produced by spe-cies of fungus, for example Aspergillus flavus and A. para-siticus. When ingested with contaminated food the toxins are particularly carcinogenic in humans and often result in liver cancer and other diseases, as they are mutagenic and immunosuppressive. The environment, farming practices, socioeconomic conditions, lack of awareness, inadequate monitoring skills, and processing facilities make the crops and food produced in tropical countries highly vulnerable

to mycotoxin contamination (Ortiz et al., 2005; Waliyar et al. 2005, 2008). Groundnuts (Arachis hypogaea L.) and groundnut-based products are particularly susceptible to aflatoxin contamination, although the contamination level in vegetables such as chili and nut crops such as pis-tachio (Pistacia vera L.) can also be very high. A recent market study in India by ICRISAT has shown that in some samples the level of aflatoxins in chili can be as high as 969 μg kg-1, which is a dramatically toxic concentra-tion given that the maximum tolerated level in food is 4 μg kg-1 in the European Union and 30 μg kg-1 in India (FAO, 2004c). Good Agricultural Practices can signifi-cantly reduce the aflatoxin contamination in many crops (Waliyar et al., 2008) and testing is now comparatively simple and cheap (Waliyar et al., 2009).

Information dissemination about mycotoxin con-tamination is critical, as most farmers and consumers are unaware of the problem. Once awareness has increased, preventative measures can be taken (Singh and Jauhar, 2006). Correct storage after harvest is also important and it is vital that crops are dried to a safe moisture level as quickly as possible.

Threats to human health from heavy metals are can also be a problem and are mainly associated with exposure to lead, cadmium, mercury, and arsenic. Elevated con-centrations of heavy metal in agricultural soil and foliar uptake from heavy metals in the atmosphere produced by smokestack and vehicle emissions are sources of heavy metal contamination on vegetables and fruit (Kachenko and Singh, 2006), as is compost composed of contami-nated urban waste. A significant proportion of contami-nation occurs during transport to market or at the point of sale. Urban and peri-urban vegetable production systems are particularly vulnerable.

Research conducted by the CIP-led Consultative Group on International Agricultural Research (CGIAR) challenge program “Urban Harvest” in Kampala, Uganda, showed that leafy vegetables should not be grown near roadsides and that the safest distance to conduct farming activities in the city is more than 30 m from the edge of the road. Vegetables should be washed thoroughly to remove heavy metals accumulated on the surface and their skins peeled where possible (CIP, 2007). Marshall et al. (2006) recommend washing vegetables twice in clean water to reduce heavy metal contamination. Planting gar-dens behind houses and other structures could help block some lead, cadmium, and zinc emitted from road vehicles, thus decreasing heavy metal contamination through foliar uptake. Regulation of vehicle emissions and the use of unleaded fuels should be part of the long-term solution to heavy metal contamination in urban and peri-urban vegetable production systems.

In urban areas, domestic users, industry, and commer-cial enterprises consume large volumes of fresh water and


in turn generate a large volume of wastewater, which in most developing countries is discharged back into natu-ral water bodies with no or little treatment. Jimenez and Asano (2004) and Keraita et al. (2008) suggested that up to 20 million ha are irrigated globally with untreated, partly treated, diluted, or treated wastewater. A recent global survey found that vegetables are the most common crops produced with diluted or raw wastewater (Raschid-Sally and Jayakody, 2007). This practice may harm human health due to pathogens (parasitic worms, protozoa, bac-teria, and viruses from fecal contamination), heavy met-als, pesticide residues, and fertilizers in the wastewater (Qadir et al., 2007). Farmers, consumers, and governmen-tal agencies in many countries are not fully aware of the hazards of irrigation with wastewater. The International Water Management Institute (IWMI) is presently heavily committed to addressing this problem in sensitive crops such as fruit and vegetables and is working on a range of health risk reducing options between farm and table.

To reduce microbial contamination from wastewater irrigation in lettuce production, IWMI studied ceasing irrigation before harvest. The study showed that during the dry season in tropical climates, the method can result in average daily reduction of 0.65 log units of thermo-tolerant coliforms (Keraita et al., 2007). However, irriga-tion cessation periods adversely affect the productivity and freshness of the crops, especially in hot climates such as West Africa, while it has more potential in cooler climates (e.g., in Ethiopia). Methods recommended by WHO (2006) to minimize wastewater contamination include primary water treatment, storing reclaimed water, wear-ing protective clothing to reduce exposure and washing hands and feet to prevent the spread of infection, farm-level wastewater management such as drip irrigation, and stopping irrigation before harvesting. Thorough washing, peeling, and cooking of vegetables are also important. The efficacy and adoption potential of these and other locally adapted safer practices is under study by IWMI and part-ners (Drechsel et al., 2008).

AVRDC and IWMI aim to incorporate more indig-enous vegetables in urban and peri-urban production sys-tems. Indigenous vegetables are easy to grow, need fewer external inputs such as water, and are often resistant to pests and diseases, needing less pesticide application in the production system. Indigenous vegetable production is expected in the long term to reduce the accumulation of pesticide residues and fertilizer in water bodies.

EXpECTED oUTComEs oF DiET DiVERsiFiCaTioN

As the adverse effects of climate change and other pro-duction constraints become more apparent, vegetable and fruit farmers will need tools to maintain or increase yield, nutritional quality, and profitability of safe products. Fruit

and vegetable farmers in developing countries are usually smallholders, thus often have fewer options and must rely heavily on resources available from their farms or within their communities. Technologies that are simple, affordable, and accessible must be the short-term outcomes of the effort to increase the resilience of fruit and vegetable production systems in less developed countries if the major problem of malnutrition in humans is to be addressed effectively. Crop diversification seems to be a simpler and more easily applied option than biofortification to address this issue. Neverthe-less, both approaches need to be seen as complementary as malnutrition is too complex and too overwhelming an issue not to use all possible resources and avenues in ensuring it is eliminated as early as possible.

Implementation of available, improved technologies should result in increased production levels of vegetables and fruits. Supported by good postharvest handling and efficient markets, such efforts are expected to reduce the seasonality of the produce to make it available year long. Eliminating or reducing food contaminants to safe levels and increas-ing the nutrient content and bioavailability should raise the quality of fruit and vegetables. The medium-term outcome of the effort to diversify diets should be an adequate quan-tity and quality of safe fruit and vegetables available year-round for human consumption.

Increasing awareness of the importance of nutritious diets and providing access to safe fruit and vegetables will be needed to attain the long-term outcome of diet diversi-fication: to have a healthy, balanced diet fulfilling the rec-ommended per capita consumption rate of 400 g of fruit and vegetables per day. However, effective development pathways are needed to promote sustainable diet diversi-fication. Stakeholders throughout the production system will benefit from capacity building; infrastructure and facilities will require installation, upgrading, and main-tenance. Dissemination and adoption efforts should be achieved in collaboration with local partners and with the participatory involvement of all stakeholders. NARES, the private sector, governmental and nongovernmental organizations, and community-based organizations are potential partners in the development and dissemination pathways of diet diversification with fruit and vegetables.

Fruit and vegetable production and marketing, how-ever, are knowledge-intensive areas of human behavior. Such knowledge can only be generated through supportive agricultural research and development in this very broad range of species. At present, investment in research and development in most of these species to overcome prob-lems of malnutrition and poverty is inadequate and will prevent their potential to be properly expressed. Broaden-ing the current myopic focus of key public sector agencies on only a very small number of food staple species might allow a more effective and equitable investment balance to be achieved.


CoNCLUsioNsThe need to diversify human diets with fruit and veg-

etables is an imperative we ignore at our own peril. The health benefits of consuming a well-balanced diet are pro-found and long-lasting; the impact on the well-being of individuals, families, communities, and economies in all countries and at all income levels cannot be underesti-mated. With nutritionally improved, pest-resistant fruit and vegetable varieties capable of thriving in increasingly unpredictable environments and with efficient and safe technologies adapted to the needs of farmers and consum-ers, the economic benefits of fruit and vegetable produc-tion and the nutritional advantages of diversified diets can extend deeply into societies all along the fruit and veg-etable value chains. Ensuring that smallholders and their families become part of this productive, healthy chain will demand substantive new investments in research and development, a strong commitment to partnerships and communication, and the desire to bring prosperity to the poor and health to all.

REFERENCEsAkinnifesi, F.K., R.R.B. Leaky, O.C. Ajayi, G. Sileshi, Z.

Tchoundjeu, P. Matakala, and F.R. Kwesiga. 2008. Indige-nous fruit trees in the tropics: Domestication, utilization and commercialization. CABI Publ., Wallingford, UK.

Alam, S.N., M.I. Hossain, F.M.A. Rouf, R.C. Jhala, M.G. Patel, L.K. Rath, A. Sengupta, K. Baral, A.N. Shylesha, S. Satpa-thy, T.M. Shivalingaswamy, A. Cork, and N.S. Talekar. 2006. Implementation and promotion of an IPM strategy for control of eggplant fruit and shoot borer in south Asia. Tech. Bull. 36. AVRDC Publ. 06-672. AVRDC – The World Vegetable Center, Shanhua, Taiwan.

Amoah, P., P. Drechsel, R.C. Abaidoo, and W.J. Ntow. 2006. Pes-ticide and pathogen contamination of vegetables in Ghana’s urban markets. Arch. Environ. Contam. Toxicol. 50:1–6.

Asian Vegetable Research and Development Center. 2008. Improving rural livelihoods through development of vege-table based postharvest technologies in Cambodia, Lao PDR and Viet Nam: Evaluation report. Project Final Report Asian Development Bank. AVRDC – The World Vegetable Center, Shanhua, Taiwan.

Bains, K., R.Y. Yang, and S. Shanmugasundaram. 2003. High-iron mungbean recipes for north India. AVRDC Publ. 03-562, AVRDC – The World Vegetable Center, Shanhua, Taiwan.

Bouis, H. 2002. A new tool for fighting micronutrient malnutri-tion. J. Nutr. 132:491S–494S.

Boyer, J.S. 1982. Plant productivity and environment. Science 218:443–448.

Bray, E.A., J. Bailey-Serres, and E. Weretilnyk. 2000. Responses to abiotic stresses. p. 1158–1249. In W. Gruissem, B. Buchan-nan, and R. Jones (ed.) Biochem. Mol. Biol. Plants. ASPP, Rockville, MD.

CIP. 2007. Impacts of urban agriculture: Highlights of urban har-vest research and development, 2003–2006. Comercial Grá-fica Sucre, Lima, Peru.

Drechsel, P., B. Keraita, P. Amoah, R. Abaidoo, L. Raschid Sally, and A. Bahri. 2008. Reducing health risks from wastewater

use in urban and peri-urban sub-Saharan Africa: Applying the 2006 WHO guidelines. Water Sci. Technol. 57:1461–1466.

Dubaie, A.H., A.R. Kadasi, and M. Al-Dhubhani. 2003. Inte-grated production and protection management in cucumber cultivation under plastic houses in Yemen terraces in winter season. ICARDA-APRP Annu. Rep. 2002–2003. ICARDA, Aleppo, Syria.

Ezzati, F., A.D. Lopez, A. Rodgers, S.V. Hoorn, and C.J.L. Mur-ray. 2002. Selected major risk factors and global and regional burden of disease. Lancet 360(9343):1347–1360.

FAO. 2004a. Food insecurity in urban future. Available at http://www.fao.org/newsroom/en/focus/2004/51786/article_51797en.html (verified 23 Dec. 2009). FAO, Rome.

FAO. 2004b. Proc. Reg. Conf. Asia and the Pacific, 27th, Beijing, China. 17–21 May 2004. FAO, Rome.

FAO. 2004c. Worldwide regulations for mycotoxins in food and feed in 2003. FAO Food Nutr. Pap. 81. FAO, Rome.

Graudal, L., and J.-P.B. Lillesø. 2007 Experiences and future pros-pects for tree seed supply in agricultural development support – based on lessons learnt in Danida supported programmes, 1965–2005. The Ministry of Foreign Affairs of Denmark, Copenhagen, Denmark.

Hagenimana, V., E.E. Carey, S.T. Gichuki, M.A. Oyunga, and J.K. Imungi. 1999. Carotenoid contents in fresh, dried and processed sweet potato products. Ecol. Food Nutr. 51:91–107.

ICARDA-Arabian Peninsula Regional Program. 2002. Strength-ening agricultural research and human resource development in the Arabian Peninsula. ICARDA-APRP, Aleppo, Syria.

Institute of Medicine of the National Academies. 2005. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids. Food and Nutri-tion Information Center. Available at http://www.nal.usda.gov/fnic/DRI/DRI_Energy/energy_full_report.pdf (veri-fied 30 Dec. 2009). The National Academies Press, Wash-ington, DC

Jimenez, B., and T. Asano. 2004. Acknowledge all approaches: The global outlook on reuse. Water 21:32–37.

Kachenko, A.G., and B. Singh. 2006. Heavy metals contamination in vegetables grown in urban and metal smelter contamina-tion sites in Australia. Water Air Soil Pollut. 169:101–123.

Kader, A. 2003. A perspective on postharvest horticulture (1978–2003). HortScience 38:1004–1008.

Keraita, B., B. Jimenez, and P. Drechsel. 2008. Extent and impli-cations of agricultural reuse of untreated, partly treated, and diluted wastewater in developing countries. CAB Rev. Per-spect. Agric. Vet. Sci. Nutr. Nat. Resour. 3:58.

Keraita, B., F. Konradsen, P. Drechsel, and R.C. Abaidoo. 2007. Reducing microbial contamination on wastewater-irrigated lettuce by cessation of irrigation before harvesting. Trop. Med. Int. Health. 12(Suppl.)2:8–14.

Koenig, T., J. Blatt, K. Brakel, K. Kloss, T. Nilges, and F. Woellert 2008. Market-driven development and poverty reduction: A value chain analysis of fresh vegetables in Kenya and Tanza-nia. SLE Publ. Ser. Humboldt University, Berlin, Germany.

Marshall, F., R. Agarwal, D.T. Lintelo, D.S. Bhupal, R.P.B. Singh, N. Mukherjee, et al. 2006. Heavy metal contamina-tion of vegetables in Delhi. Available at http://www.cseindia.org/programme/health/pdf/conf2006/toxins2_aggarwal2.pdf (verified 23 Dec. 2009).

Moustafa, A.T., S. Oraifan, A. Al Bakry, and A. Nejatian. 2006. Human and nature: Working together for sustainable devel-opment of drylands. p. 839–842. In Proc. Int. Conf. on


Dryland Development, 8th, Beijing, China. 26–28 Feb. 2006. ICARDA, Aleppo, Syria.

Mukred, A., A.T. Moustafa, A. Al Kirshi, and M. Al Dhubani. 2003. Introduction of protected agriculture into mountain terraces of Yemen. p. 504–511. In Proc. Int. Conf. on Devel-opment of Drylands, 7th, Tehran, Iran. 14–17 Sept. 2003. ICARDA, Aleppo, Syria.

Nyambo, A., A. Nyomora, C.K. Ruffo, and B. Tengnas. 2005. Fruits and nuts, species with potential for Tanzania. Tech. Handb, 34. Regional Land Management Unit, World Agro-forestry Center, Nairobi, Kenya.

Ortiz, R., T. Ban, R. Bandyopadhyay, D. Bergvinson, K. Hell, and D. Jeffers. 2005. CGIAR research-for-development program on mycotoxins. p. 44–45. Abstr. of Conf. on Reducing Impact of Mycotoxins in Tropical Agriculture with Emphasis on Health and Trade in Africa. Accra, Ghana. IITA and Myco-Globe.

Park, Y., A.F. Subar, V. Kipnis, F.E. Thompson, T. Mouw, A. Hol-lenbeck, M.F. Leitzmann, and A. Schatzkin. 2007. Fruit and vegetable intakes and risk of colorectal cancer in the NIH-AARP diet and health study. Am. J. Epidemiol. 166:170–180.

Qadir, M., B.R. Sharma, A. Bruggeman, R. Choukr-Allah, and F. Karajeh. 2007. Non-conventional water resources and opportunities for water augmentation to achieve food secu-rity in water scarce countries. Agric. Water Manage. 87:2–22.

Raschid-Sally, L., and P. Jayakody. 2007. Understanding the driv-ers of wastewater agriculture in developing countries. p. 425–457. IWMI, London, UK.

Schmidhuber, J. 2008. Food policies, consumption patterns, health concerns and the Mediterranean diet. Proc. of Conf. on Med-iterranean Products in the Global Market, , Calabria, Italy. 16–17 June 2008. Available at http://aic.ucdavis.edu/calmed/calmed4/agenda.html (verified 23 Dec. 2009).

Singh, R.J., and P.P. Jauhar. 2006. Genetic resources, chromo-some engineering, and crop improvement. CRC Press, Boca Raton, FL.

Subramanian, A.M., and R.Y. Yang. 1998. High-iron mungbean recipes from south Asia. AVRDC Publ. 98-480. AVRDC – The World Vegetable Center, Shanhua, Taiwan.

Van Jaarsveld, P.J., M. Faber, S.A. Tanumihardjo, P. Nestel, C.J. Lombard, and A.J. Benade. 2005. Beta-carotene-rich orange-fleshed sweet potato improves the vitamin A status of primary school children assessed with the modified-relative-dose–response test. Am. J. Clin. Nutr. 81:1080–1087.

Vinh, N.Q., and X.C. Ngo. 2006. Grafting and growing tomatoes by grafted seedlings to protect from bacterial wilt (Ralstonia solanacearum) on 3500–4,000 ha each year in Lam Dong Prov-ince, Viet Nam. IAS Research Highlights. Institute of Agri-cultural Sciences for Southern Viet Nam.

Waliyar, F., P.L. Kumar, A. Traore, B.R. Ntare, B. Diarra, and O. Kodio. 2008. Pre- and postharvest management of afla-toxin contamination in peanuts. p. 209–218. In J. Leslie, R. Bandyopadhyay and A. Visconti (ed.) Mycotoxins detection methods, management, public health and agricultural trade. CABI Publ., Wallingford, UK.

Waliyar, F., S.V. Reddy, T.Y. Reddy, K. Subramanyam, P.Q. Craufurd, and T.R. Wheeler. 2005. Management of afla-toxins in groundnut in southern India. Sum. Proc. Of Int. Peanut Conf. on Prospects and Emerging Opportunities for Peanut Quality and Utilization Technology, Bangkok, Thai-land. 9–15 Jan. 2005. Int. Crops. Res. Inst. for the Semi-Arid Tropics, Patancheru, India.

Waliyar, F., S.V. Reddy, K. Subramanyam, T.Y. Reddy, K. Rama-devi, and P.Q. Craufurd. 2003. Importance of mycotoxins in food and feed in India. Aspects Appl. Biol. 68:147–154.

Waliyar, F., S.V. Reddy, and P. Lava Kumar. 2009. Review of immunological methods for the quantification of aflatoxins in peanut and other foods. Peanut Sci. 36:54–59.

World Health Organization. 2003a. Fruit and vegetables promotion initiative. Rep. of the meeting, Geneva, Switzerland. 25–27 Aug. 2003. World Health Organization, Geneva, Switzerland.

World Health Organization. 2003b. Diet, nutrition and the pre-vention of chronic diseases. Report of a joint WHO/FAO expert consultation. WHO Tech. Rep. Ser. 916. World Health Organization, Geneva, Switzerland.

World Health Organization. 2006. WHO guidelines for the safe use of wastewater, excreta and greywater. p. 25–28. World Health Organization, Geneva, Switzerland.

Yang, R.Y., P.M. Hanson, and T.A. Lumpkin. 2007. Better health through horticulture –AVRDC’s approach to improved nutrition of the poor. Acta Hortic. (ISHS) 744:71–78.

Yang, R.Y., S.C.S. Tsou, and T.C. Lee. 2002. Effect of cooking on in vitro iron bioavailability of various vegetables. p. 130–142. In T.C. Lee and C.T. Ho (ed.) Bioactive Compounds in Foods: Effect of Processing and Storage. American Chemical Society, Washington, DC.


SympoSia

Key trends influencing agricultural research organizations include: First, pervasive computing and low-cost connec-

tivity is transforming the ways that science and development are conducted. Second, massive processing power is accessible through “clouds” of Internet and computing services, providing sharable tools, applications, and intelligently linked content and data. Local computing capacities are no longer a barrier to gain access to world-class computing services. Third, science can call on an increasing ability to collect, analyze, and reuse massive, distributed collections of data. Fourth, individuals and amateurs are increasingly able to create and manage sophisticated informa-tion and knowledge. This “democratization” of science and the Internet draws many more people and institutions into research and development processes.

In a review of current work on “data-intensive computing” for science (Hey et al., 2009), the editors argue that scientific break-throughs will increasingly be powered by advanced computing capabilities that help researchers manipulate and explore massive data sets. Almost everything about science is changing because of the impact of information and communication technologies.

Information and communication technologies (ICTs) are being applied to all parts of the research for development

Information and Communication Technologies—Opportunities to Mobilize Agricultural Science

for Development

Peter Ballantyne,* Ajit Maru, and Enrica M. Porcari

aBSTRaCTKnowledge, information, and data—and thesocial and physical infrastructures that carrythem—are widely recognized as key build-ing blocks for more sustainable agriculture,effective agricultural science, and productivepartnerships among the global research com-munity.Throughinvestmentsine-Scienceinfra-structure and collaboration on one hand, andrapiddevelopmentsindigitaldevicesandcon-nectivityinruralareas,thewaysthatscientists,academics, and development workers create,share,andapplyagriculturalknowledgeisbeingtransformedthroughtheuseofinformationandcommunicationtechnologies(ICTs).Thispaperexaminessometrendsandopportunitiesasso-ciatedwiththeuseoftheseICTsinagriculturalsciencefordevelopment.

P. Ballantyne, International Livestock Research Institute (ILRI), P.O. Box 5689, Addis Ababa, Ethiopia; A. Maru, Global Forum on Agricul-tural Research (GFAR), c/o FAO, Viale delle Terme di Caracalla, 00153 Rome, Italy; E. Porcari, CGIAR ICT-KM Program, c/o Bioversity Inter-national, Via dei Tre Denari 472/a, 00057 Maccarese, Italy. Received 23 Sept. 2009. *Corresponding author ([email protected]).

Abbreviation: ICTs, information and communication technologies.




continuum that connects agricultural science with agri-cultural and rural change.

On the one hand is “e-Science” (or e-Research), characterized by global collaboration and the next gen-eration of infrastructure that will enable it. On the other is “m-Agriculture”—that uses mobile digital devices such as phones, laptops, and sensors that put ICTs, connectiv-ity, and applications into the hands of rural communities. Between these extremes, ICTs are transforming agricul-tural extension (in the United States and the Philippines for example), facilitating the delivery of education and learning in universities and through open distance edu-cation (see the work of the Commonwealth of Learn-ing), helping to empower the rural poor in developing countries (see Heeks, 2009), and powering a wide array of agricultural finance, credit, market, weather, and other services delivered by public and private organizations (see www.e-agriculture.org and http://iaald.blogspot.com [verified 22 Dec. 2009] for examples).

Drawing on discussions and materials prepared for a workshop held in June 2009, this paper examines some trends and opportunities associated with the use of these ICTs in agricultural science for development. Significant trends include:

· Increasingly “ubiquitous” connectivity along value chains—As illustrated by Ballantyne (2009) and Gakuru et al. (2009), people are making use of a wide range of devices and platforms to access and share agricultural knowledge: from the Web to phones, radio, video, and text messaging. Most scientists already work in knowledge-rich environ-ments where good digital and social connectivity is the norm; farming communities, probably using dif-ferent devices from those we see today, will be far more connected than now. The widespread use of mobile phones by farmers and others to get market and weather information is well documented (CTA, 2009). Multiple connectivity paths widen the poten-tial reach of science; they also widen the potential to include the knowledge of rural communities into science.

· Increasingly “precise” applications and tools—ICTs and digital signatures or labels of various types are being used to track products from producer to consumer; to monitor local soil, weather, and market conditions; to tailor data and information services to the demands of a specific audience or individu-als (Mondal and Basu, 2009; Wolfert et al., 2009). Future applications will come in many shapes and sizes, to suit even the most specialized needs.

· Increasingly “accessible” data and informa-tion—Vast quantities of public data and information held by institutions and individuals are becoming visible, publicly accessible, and reusable at the click

of a device. Beyond the open-access movement that mainly focuses on scientific literature, there is a broader trend to make publicly funded data, soft-ware, and information more open (Hey et al., 2009; McLaren et al., 2009). Beyond shorter-term chal-lenges to make these data and information widely and openly accessible (through for instance the Coherence in Information for Agricultural Research for Devel-opment initiative, the information and communica-tion management initiative of the Global Forum on Agricultural Research, or the “Triple A” effort of the Consultative Group on International Agricultural Research), more intermediary skills and applications will be needed to help harvest, make sense of, and add value to the fast-growing layers of data and informa-tion that are becoming available.

· Increasingly “diverse” set of applications avail-able across digital clouds—The digital “identi-ties” of scientists and their collaborators are starting to give them access to a wide range of online tools and applications, accessible from any location and across different devices, enabling collaboration across boundaries as never before (Werth, 2009). As outlined by Porcari (2008), local firewalls and server configurations conditions will no longer be a con-straint to global sharing.

· Increasingly “interconnected” tools and knowl-edge bases—Hannay (2009) argues that the inter-connectedness of scientific data and information will be a key feature of future science. In agriculture, as illustrated by GTZ (2008), different communities are starting to connect and share their knowledge with each other, along research cycles and value chains and across disciplines. Increasing attention to innovation systems approaches in agricultural development, as discussed by Hall (2006), Kristjanson et al. (2009), and the World Bank (2007) among others, points to a much wider and more diverse involvement by differ-ent “actors” in science and research, such as farmers, traders, and politicians. As Waters-Bayer et al. (2006) argue, connecting these actors so they can interact enhances such innovation processes.

FRoNTiERS iN iCTs

Hardware and ConnectivityMoore’s Law that the number of transistors that can be placed inexpensively on an integrated circuit is growing exponentially, the number of transistors doubling approx-imately every 2 yr, has so far held. The same law can be applied to processing speeds of microprocessors, memory capacity, and the number of pixels that a digital camera can process. Memory storage capacities in magnetic and optical media have also increased exponentially and solid


The semantic Web and its related techniques and applica-tions (e.g., ontologies) currently work in this way, helping reshape machine-to-machine interaction and the way com-puters retrieve, manage, and share knowledge on the Web.

The science of pragmatics—the practical interpretation and use of signs by agents or communities within particu-lar circumstances and contexts—and going beyond con-ventional semantics, is now allowing ICTs to be used in much more supportive ways. This has been demonstrated in diverse areas such as health, scientific research, and busi-ness management in modeling, simulation, forecasting, and visualization, and has implications for agriculture. These potentials bring new challenges on how we understand this new pervasive computing landscape and how we can make use of collective and distributed forms of intelligence.

interactions with BiologyThe interaction of ICTs with biology, biotechnology, nan-otechnology, and new materials is enabling the develop-ment of high-quality information from diverse entities and sources and which is self-organizing. This self-organizing collective intelligence—living information—presents new frontiers in effective use and application. Continuous advances in ICTs and biology are enabling developments where the relationship between these two disciplines faces a paradigm shift: from ICTs that mimic biology to ICTs that use biology for information processing. Progress in synthetic biology—the study of the design and building of novel biological functions and systems—is bringing prog-ress in systematic design methodologies and manufacturing processes. The potential to interface ICTs with biological systems at the micro/nano scale is now emerging.

It can further be argued, even with current knowl-edge, that bio- and nanotechnology, material sciences, and ICTs will together define the core direction of agricultural science, research, and technology in the future—by hav-ing an impact on plant and animal breeding and improve-ment, agricultural production systems, risk management and aversion, sustainable use of natural resources, protect-ing the environment and agricultural market chains, and in agricultural innovation in general.

iCTs iN aGRiCULTURaL SCiENCE FoR DEVELopmENT

iCTs and agricultural productionAn area of application for ICTs is in improving, through better management, the efficiency and sustainability in using inputs—land, soil nutrients, feed and fodder, water, energy, pesticides, labor, and most importantly, informa-tion and knowledge—in agriculture. The ICTs also help reduce the negative effects of pests and disease and enable aversion and mitigation of risks such as from inclem-ent weather, droughts, floods, and long-term change in

state drives are already commercialized. Connectivity between computers and through the Internet has similarly increased in bandwidth. The rates at which data can be transmitted, both within buildings and across long dis-tances, grows without apparent limit and with ever reduc-ing costs. Parallel and Grid computing has demonstrated huge potentials of processing power available for use on the desktop of an average computer user, and this will be multiplied manyfold with memristors (already pro-totyped), photonic and quantum computers (still in the research phase). We are seeing a boom in handheld devices that interface with existing systems.

Ubiquitous Telecommunication infrastructureFlowing from the falling costs of all things digital, there has been a steady flow of investment into communications infrastructure around the world. Cell phone and broad-band (wired and wireless) Internet networks carrying both voice and data are being deployed in even the poor-est countries, and with time will expand to cover most rural areas. These systems are sophisticated; they increas-ingly allow agriculture and agricultural research to take improved connectivity for granted.

Utility or “Cloud” ComputingThe combination of progress in computing hardware, system software, and Internet communications has now enabled the construction of general-purpose data centers that can be reconfigured by command to support any soft-ware application in minutes. There are already data ser-vices that allow a user to have hundreds or thousands of computers at their command, and yet pay for them by the hour or minute, without owning or operating the hardware themselves. The costs are far less than even falling hardware prices would suggest because the cost of the data center can be shared among many users. In effect, the data center acts like a utility, providing as much computing as requested at just the times when needed. Since these data centers are shared over the Internet, they are sometimes called com-puting “in the cloud.” These “cloud” data centers are the natural repository for shared data sets, so that users in any location or institution can instantly access, analyze, and interpret public information goods without the need to move the data to their own facilities. This can enable a researcher in any location to work with data as well as with any other researcher, which can lead to new kinds of col-laboration and new sources of project direction.

Software and Content managementAn important frontier achieved through more complex processors, processing speeds, memory capacity and con-nectivity has been the development of agents, sensors, and devices such as radio-frequency identification tags (RFIDs) that are now reshaping how humans work and interact.


climate. Through innovation, ICTs continue to contribute to improving throughput of farming systems, increasing the quantity, quality, and marketability of outputs (e.g., food, energy, and biomaterials), supporting their market-ing and enabling their effective and efficient consump-tion by households and communities and their ultimate recycling. The ICTs helped pave the way for consumers to decide which products they can “responsibly” purchase, which seem to have higher food miles, and those whose production and safety can be traced all the way back to the fishpond.

For the small, resource-poor farmer and producers in economically developing countries, these applications of ICTs have not yet become mainstream. The economic returns from agriculture and access to affordable technol-ogy useful in small-farms operations are the main con-straints in more widespread use of ICTs in smallholder agricultural production.

iCTs and agricultural Science, Research, and Technology GenerationThe current application of ICTs in agricultural science, research, and technology generation can be clustered around:

· Data collection—Enabling collection of agricultural and environmental data from biological and environ-mental sources, with or without human interaction. The data are subsequently analyzed and manipulated to feed auxiliary applications or to conduct studies.

· Number crunching—Enabling management, shar-ing, and processing of large data sets, modeling and simulation, image processing, and visualization that contribute to plant and animal breeding and improvement, bioinformatics, agricultural meteo-rology, plant, animal, and zoonotic diseases epide-miology, farming systems research, market chain analysis and management, etc.

· Geospatial applications—Enabling data and infor-mation related to geography and space to be man-aged, processed, and visualized and contributing to land and water use planning, natural resources uti-lization, agricultural input supply and commodity marketing, poverty and hunger mapping, etc.

· Decision support and knowledge-based systems and robotics—Enabling data and information to be organized with added experiences of experts to mimic, multiply, and use expertise, especially in searching information and data semantically, in problem diagnosis, and in farm and agricultural pro-cess automation.

· Embedded ICTs in farm equipment and pro-cesses—Enabling greater efficiencies in farm equip-ment and agricultural processes and in what is termed “precision agriculture,” as also in agricultural products transport and marketing such as the use of

RFIDs, wireless Internet, and cellular telephony in labeling, traceability, and identity preservation.

· Connecting communities and enabling learn-ing—Using ICTs to connect communities of farm-ers, researchers, and all connected to agriculture. ICTs are already playing a significant role in con-necting scientists and researchers to communicate with each other and in scientific and technical publi-cations. Use of ICTs to connect farmers and produc-ers to new agricultural knowledge and technology and in problem resolution has been tested and found very useful, so much so that ICTs are now con-sidered to be transforming agricultural extension (examples are the Pinoy Farmers’ Internet project of the Open Academy for Philippine Agriculture, the “Digital Green” project in India, and the eXtension Initiative in the United States). Through enabling access to text, graphics, audio, and video objects in an integrated manner, ICTs have also helped educa-tion systems by broadening access to learning and by improving the quality of the classroom experience.

iCTs and agricultural innovationProgress in hardware, software, connectivity, and integra-tion of computing systems enables new forms of data gath-ering, both human assisted and automated. It is bringing new capacities for processing data and information dis-semination. In future, this process of connecting com-munities to new knowledge and information is likely to accelerate with advanced technologies that bring far greater processing powers, robust, reliable storage capac-ity, and connectivity.

This progress will also extend new forms of participa-tory science and research, extension, and learning to those within agricultural communities who are not yet included in these processes. Extension, as it is understood now as being “linear” from research to farmer through extension agents will be as in a network with pluralistic sources, formats, and users of information and knowledge (Gakuru et al., 2009; Gandhi et al., 2009). Learning will be ubiquitously avail-able and pervasive for all in an agricultural community. It will change the realm of agricultural science, where it will not only be the formally educated scientists who bring new technological innovations but whole communities who do so. All within an agricultural community will be producers and consumers of information and innovations.

The ICTs, as described above, can also enable more connected agricultural communities and it is envisaged that these connected communities will practice science and lead agricultural innovation. Participatory commu-nity-driven and -led agricultural innovation through extensive use of ICTs will lead to new technologies such as seeds, breeds, and animals customized to meet the specific needs of particular communities, participatory watershed


management, participatory plant and animal pest and dis-ease monitoring and surveillance, etc.

iCT TRENDS iNFLUENCiNG aGRiCULTURaL SCiENCE FoR DEVELopmENTIn general, the most significant impact of ICTs on agri-cultural technology generation will be in connecting and engaging communities in participatory agricultural inno-vation. Science will be able to come out of its “silos.” New agricultural processes and technologies to solve agricultural problems will emerge through innovation with user com-munities, thus eliminating many of the constraints facing agricultural science, research, and technology generation.

Other significant trends include:· Information and communication technologies,

devices, and software are becoming much cheaper and more affordable, even in rural areas.

· Connectivity is becoming more pervasive and “mobile”—people can connect and interact in real time with other people and data across a broad range of wireless, mobile, and other devices. And more and more of the devices are becoming smart and intel-ligent—capable of multiple operations.

· Geospatial and “neogeographic” functionalities, applications, and tools are spreading and becoming ubiquitous, offering pinpoint location and data col-lection and sharing possibilities.

· More and more services will be provided across the Internet through so-called “cloud” computing, obviating the need for sophisticated local ICT sys-tems and capabilities.

· As the quantity of data and information grows, new ways to organize, navigate, mine, share, visualize, and “mash” it up will emerge, creating new pos-sibilities and services.

· Digital applications and tools are being applied to enable and extend traditional “human” processes such as communication, collaboration, and analysis.

· Within agriculture and science, new thinking and approaches are emerging around “end user innova-tion,” focused on knowledge, value chains, innova-tion systems, etc.

· Scientists increasingly use and depend on ICTs in their daily work: computers, enhanced ICT literacy, and connectivity are part of the “basic” package.

iCT oppoRTUNiTiES FoR aGRiCULTURaL SCiENCE FoR DEVELopmENTOpportunities that agricultural science could gain from increased use of ICTs include:

· Through ICTs, the possibility to make agricultural research and development processes more inclusive,

enhancing communication among all agriculture stakeholders. As reported by the CGIAR “KS in Research” project (www.ks-cgiar.org [verified 22 Dec. 2009]), this provides greater potential for hori-zontal knowledge sharing among different stake-holders, increasing the likelihood of collaboration.

· Rural communities and farmers empowered through ICTs to enhance their own livelihoods and other opportunities. New types of ICT-enabled rural busi-nesses and entrepreneurs are emerging, providing a range of new services and livelihoods.

· Delivery of various ICT-enabled services to rural people: such as market access, access to international export markets through ICT traceability systems, mobile financial services, mobile extension services.

· Improved capabilities to create and store data and information; gaining rapid access to it.

· Enhanced two-way flow of timely, highly targeted, loca-tion-specific, and location-intelligent information.

· Increased possibilities for public and community to be lay data collectors; farmers and producers can con-tribute data directly to national and international initiatives. This will also facilitate stable and con-tinuous farm (field) data acquisition.

Realizing the potential of iCTs in agricultural Science for DevelopmentPriorities to realize these opportunities include:

· Improve communications infrastructure and band-width, investing in lower-cost hardware, software, and applications that connect science right along the development chain.

· Increase and improve formal education and training in information and communication sciences that contribute to innovation in the use of new ICTs in agriculture.

· Extend the generation and dissemination of data and information content as a “public good” that is widely accessible and is licensed to be easily reused and applied.

· Support applications that integrate data and informa-tion or foster the interoperability of applications and information systems, allowing safe and ethical access while protecting necessary rights.

· Encourage the effective uptake and use of data, infor-mation, and knowledge, particularly focusing on capacity-building dimensions necessary for the out-puts of science to have impacts.

· Support innovation in the workflows, processes, and tools used to create, share, publish, visualize, and connect the outputs of agricultural science and the people engaged in it.

· Promote open access to research documents and data, the use of open intellectual property licenses, and


adoption of open standards that facilitate interop-erability across systems. Alongside these changes, develop incentives and reward schemes that promote open sharing and open behavior.

Further ReadingThe Fate of Agriculture? (Balaji, 2009).EFITA Questionnaire on ICT Adoption Trends in Agriculture

(Gelb, 2009a).How Are Innovations Adopted—A Case Study of ICT (Internet)

for Agriculture (Gelb, 2009b).ICT Adoption Dangers and Pitfalls (Gelb, 2009c).The Impact of ICT on Agricultural Science and Scientific Rea-

soning (Gelb, 2009d).What are ICTs for Agriculture? (Gelb, 2009e).Semantic Interoperability (Keizer and Pesce, 2009).Changing the Emperor: ICT’s Transforming Agricultural Science,

Research and Technology (Manning-Thomas, 2009).Innovative IT Tools for Farm Management—Benefits and Benefi-

ciaries (Manukyan, 2009).ICTs Role in Improving Market Access for Small Scale Farmers

(Nichterlein, 2009).ICTs Transforming Agricultural Science, Research and Technol-

ogy Generation (Ninomiya, 2009).Ubiquitous Networks and Cloud Computing (Porcari, 2009).Managing Agriculture Knowledge: Role of Information and

Communication Technology (Rafea, 2009).Information and Communication Technologies for Climate

Change Adaptation, with a Focus on the Agricultural Sector (Sala, 2009).

acknowledgmentsThe authors acknowledge the contributions of speakers and resource persons to the Science Forum Workshop on “ICTs transforming agricultural science, research and tech-nology generation” (Wageningen, Netherlands, 16–17 June 2009): V. Balaji (ICRISAT), Ehud Gelb (Israel), Johannes Keizer (FAO), Nadia Manning-Thomas (IWMI), Arman Manukyan (Austria), Karin Nichterlein (FAO), Seishi Ninomiya (Japan), Kevin Painting (CTA), Valeria Pesce (GFAR), Ahmed Rafea (Egypt), Simone Sala (Italy), Sjaak Wolfert (Netherlands), and Stanley Wood (IFPRI).

ReferencesBalaji, V. 2009. The fate of agriculture? Thinkpiece for CGIAR

Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Ballantyne, P.G. 2009. Accessing, sharing and communicating agricultural information for development: emerging trends and issues. Inf. Dev. 25:260–271.

CTA. 2009. The many uses of mobiles. ICT Update 50. Available at http://ictupdate.cta.int/en/Feature-Articles/The-many-uses-of-mobiles (verified 21 Dec. 2009). Tech. Cent. for Agric. and Rural Cooperation (CTA), Wageningen, Neth-erlands.

Gakuru, M., K. Winters, and F. Stepman. 2009. Inventory of innovative farmer advisory services using ICTs. Forum for Agric. Res. in Africa, Accra.

Gandhi, R., R. Veeraraghavan, K. Toyama, and V. Ramprasad. 2009. Digital green: Participatory video and mediated instruc-tion for agricultural extension. Inf. Technol. Int. Dev. 5:1–15.

Gelb, E. 2009a. EFITA questionnaire on ICT adoption trends in agriculture. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Gelb, E. 2009b. How are innovations adopted—A case study of ICT (Internet) for agriculture. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural sci-ence, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Gelb, E. 2009c. ICT adoption dangers and pitfalls. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Gelb, E. 2009d. The impact of ICT on agricultural science and sci-entific reasoning. Thinkpiece for CGIAR Sci. Forum Work-shop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Gelb, E. 2009e. What are ICTs for agriculture? Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

GTZ. 2008. The participatory Web—New potentials of ICT in rural areas. Available at www.gtz.de/en/themen/laendliche-entwicklung/15081.htm (verified 21 Dec. 2009). GTZ, Esch-born, Germany.

Hall, A. 2006. New insights into promoting rural innovation: Learning from civil society organisations about the effective use of innovation in development. Available at http://www.research4development.info/PDF/Outputs/Misc_Crop/R8372FinalReport.pdf. UNU-MERIT, Maastricht.

Hannay, T. 2009. From Web 2.0 to the global database. Nascent. Available at http://blogs.nature.com/wp/nascent/2009/10/from _web_20_to_the_global_data.html (verified 21 Dec. 2009).

Heeks, R. 2009. The ICT4D 2.0 Manifesto: Where next for ICTs and international development? Dev. Informatics Working Pap. 42. Univ. of Manchester Inst. for Dev. Policy and Man-age., Manchester, UK.

Hey, T., S. Tansley, and K. Tolle. 2009. The fourth paradigm: Data-intensive scientific discovery. Available at http://research.microsoft.com/en-us/collaboration/fourthparadigm/ (veri-fied 21 Dec. 2009). Microsoft Research, Redmond, WA.

Keizer, J., and V. Pesce. 2009. Semantic interoperability. Think-piece for CGIAR Sci. Forum Workshop on “ICTs transform-ing agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Kristjanson, P., R.S. Reid, N. Dickson, W.C. Clark, D. Rom-ney, R. Puskur, S. MacMillan, and D. Grace. 2009. Link-ing international agricultural research knowledge with action for sustainable development. Proc. Natl. Acad. Sci. USA 106:5047–5052. doi: 10.1073/pnas.0807414106.

Manning-Thomas, N. 2009. Changing the emperor: ICT’s transforming agricultural science, research and technology. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Manukyan, A. 2009. Innovative IT tools for farm manage-ment—Benefits and beneficiaries. Thinkpiece for CGIAR


Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

McLaren, G., T. Metz, M. van den Berg, R.M. Bruskiewich, N.P. Magor, and D. Shires. 2009. Informatics in agricul-tural research for development. Adv. Agron. 102:135–147. doi:10.1016/S0065-2113(09)01004-9.

Mondal, P., and M. Basu. 2009. Adoption of precision agricul-ture technologies in India and in some developing countries: Scope, present status and strategies. Prog. Nat. Sci. 19:659–666. doi:10.1016/j.pnsc.2008.07.020.

Nichterlein, K. 2009. ICTs role in improving market access for small scale farmers. Thinkpiece for CGIAR Science Forum Workshop on ‘ICTs transforming agricultural science, research and technology generation,’ Wageningen, Nether-lands. 16–17 June 2009.

Ninomiya, S. 2009. ICTs transforming agricultural science, research and technology generation. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Porcari, E.M. 2008. Strategic technologies for the CGIAR in 2009. CGIAR ICT-KM Progr., Rome.

Porcari, E.M. 2009. Ubiquitous networks and cloud computing. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Rafea, A. 2009. Managing agriculture knowledge: Role of

information and communication technology. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs transforming agricultural science, research and technology generation,” Wageningen, Netherlands. 16–17 June 2009.

Sala, S. 2009. Information and communication technologies for cli-mate change adaptation, with a focus on the agricultural sector. Thinkpiece for CGIAR Sci. Forum Workshop on “ICTs trans-forming agricultural science, research and technology genera-tion,” Wageningen, Netherlands. 16–17 June 2009.

Waters-Bayer, A., L. van Veldhuizen, M. Wongtschowski, and C. Wettasinha. 2006. Recognizing and enhancing local innova-tion processes. In Innovation Africa Symp., Kampala, Uganda. 20–23 Nov. 2006. Available at www.innovationafrica.net/pdf/s6_waters-bayer_full.pdf (verified 21 Dec. 2009).

Werth, C. 2009. Number crunching made easy: Cloud computing is making high-end computing readily available to researchers in rich and poor nations alike. Newsweek, 2 May. Available at http://www.newsweek.com/id/195734 (verified 21 Dec. 2009).

Wolfert, J., N. Verdouw, C.M. Verloop, and A.J.M. Beulens. 2009. Organizing information integration in agri-food: A method based on a service-oriented architecture and living lab approach. In press. Comput. Electron. Agric. doi:10.1016/j.compag.2009.07.015.

World Bank. 2007. Enhancing agricultural innovation: How to go beyond the strengthening of research systems. Available at http://siteresources.worldbank.org/INTARD/Resources/EnhancingAgInnovationebook.pdf (verified 21 Dec. 2009). World Bank, Washington, DC.


symposia

The ClimaTe sysTem is Warming

Fossil fuel combustion and changes in the land use (includ-ing deforestation) have resulted in carbon dioxide (CO2)

accumulation in the atmosphere. The global atmospheric concen-tration of CO2 has increased from about 280 ppm (preindustrial) to 379 ppm in 2005. The current atmospheric concentration of carbon dioxide largely exceeds the natural range recorded over the last 650,000 yr (180 to 300 ppm) as determined from ice cores (IPCC, 2007a).

The accumulation of CO2 and other greenhouse gases is expected to cause observable climatic changes in the 21st cen-tury. The International Panel on Climate Change (IPCC) has been publishing assessment reports to governments since the early 1990s. The fourth report, published in 2007, concludes that the global temperature in the last 100 to 150 yr has increased 0.76 ± 0.19°C (IPCC, 2007a). The report also concludes that the warming of the climate system is unequivocal and evident from observed increases in global air and ocean temperatures as well as in melting of snow and ice and rising global sea level (IPCC, 2007a, b). The report includes several evidences of the impact that the observed global warming is already having on biological processes affecting agriculture, forestry, and human health.

Climate Risk Management for Adaptation to Climate Variability and Change

Walter E. Baethgen*

absTraCTThe warming of the climate system is evidentfrom observations of air and ocean tempera-turesaswellasinmeltingofsnowandiceandrisingsealevel.Measuresareneededtoreversethetrendsofincreasedaccumulationofgreen-housegases(GHG)intheatmosphere.Thetwomainpathstoreversethistrendare:(i)reducingGHG emissions through cleaner energy gen-eration and (ii) removing CO2 through carbonsequestration.Theagriculturalandforestrysec-torscanplayakeyrole inbothpaths.Carbonmarketswill likelyencourageincreasedcarbonsequestration. However, the implementationofcarbon-marketprojects forsmall farmers inleastdevelopedcountries is still amajorchal-lenge.Evenunderthemostoptimisticscenariosof future GHG emissions adaptive measuresareneededtoaddressimpactsofthewarmingdue topastandcurrentemissions. Integratingclimatechangeintodecisionmakingiscompli-catedby theuncertainty levelsofclimatesce-narios.Itisalsochallengedbya“doubleconflictofscales”:(i)climatescenariosareavailableforperiodsmuchfartherinthefuturethantheonestypically needed for decision making and (ii)spatial scales of the climate scenarios (globalto regional) are coarser than the ones oftenneeded for actual decision making (i.e., locallevel). Introducing the issue of climate changeinto policy and development agendas can befacilitatedbyconsideringthelonger-termvaria-tionsaspartof thecontinuumof total climatevariability(seasonstodecadestocenturies)andgenerating information at the temporal scalesthat are relevant and applicable for particu-lar decisions.

International Research Institute for Climate and Society (IRI), Colum-bia Univ., 61 Route 9W, Palisades, NY 10964. Received 23 Sept. 2009. *Corresponding author ([email protected]).

Abbreviations: CRM, Climate Risk Management; GHG, greenhouse gases; IPCC, International Panel on Climate Change; IRI, International Research Institute for Climate and Society; UNFCCC, United Nations Framework Convention for Climate Change.

Published in Crop Sci. 50:S-70–S-76 (2010). doi: 10.2135/cropsci2009.09.0526 Published online 8 Feb. 2010. © Crop Science Society of America | 677 S. Segoe Rd., Madison, WI 53711 USA



The IPCC report finally includes possible scenarios of future climate that would greatly affect agricultural pro-duction throughout the world. Research conducted since the late 1980s cited in the four IPCC’s assessment reports has been showing that crop yields in several regions of the world could be severely reduced under warmer conditions due to shorter crop growing seasons and increased pest and disease pressure (see for example Parry and Rosenz-weig, 1994; Parry et al., 2004, Rosenzweig and Iglesias, 1994, Baethgen and Magrin, 1995; Schneider et al., 2001). Moreover, agricultural systems that are already fragile under current climate conditions could become unsus-tainable in the future (e.g., northeast Brazil, the Sahel; Baethgen, 1997). Most of the conducted studies suggest that the most severe impacts on agriculture will be felt around the tropics where most of the least developed countries are located.

possible responses To ClimaTe Warming in The agriCulTural seCTor: miTigaTionEven though it is still difficult to determine how much of the global warming can be attributed to human activity, there is overwhelming agreement that measures should be taken to reverse the current trend of increased accumula-tion of greenhouse gases (GHG) in the atmosphere. In the jargon of the scientific, technical and policy communities working in the climate change arena, the actions oriented to reverse such trend is generally referred to as “mitiga-tion” of climate change. There are basically two paths to reverse the current trend of increased accumulation of GHG: (i) reducing GHG emissions through cleaner energy generation and (ii) removing CO2 through carbon “sinks” or carbon sequestration.

The agricultural sector in both developed and devel-oping countries can play an important role in helping to reverse the trend. Regarding the reduction of GHG there is increasing interest and growing opportunities to generate energy utilizing biofuels originated in the agricultural and forestry sectors and thus reduce the net emissions of GHG. Hence, several developing countries and some developed countries are investing considerable efforts to increase the generation of energy with crop-produced alcohol (e.g., sugarcane, sweet sorghum, maize), biodiesel produced with oil crops (e.g., oil palm, soybeans, sunflower), and with residues originated in the harvest of annual crops (e.g., rice husks) as well as in the forest harvest and forest industries (Baethgen and Martino, 2004). The observed increased interest in these alternative energy sources is probably rooted mainly in economical, financial, and geopolitical advantages of reducing the countries’ depen-dence on fossil fuels. However, this is a path that could lead to important reductions in net GHG global emissions (Sims, 2004) and generate important new alternatives

for the agricultural sector in the developing world (Hill et al., 2006). On the other hand, several researchers are expressing increased concerns with possible unintended and negative consequences that the production of biofuels could have on increasing world food prices and affecting the global food security (Von Braun, 2008).

A second possible path to reverse the current increased accumulation of GHG in the atmosphere is removing CO2 from the atmosphere. Soil science research evidenced that agricultural lands have the potential for removing 40,000 to 80,000 million tonnes of carbon over the next 50 to 100 yr (Kolshus, 2001; IPCC, 1992). Consequently, soil car-bon sequestration in agricultural lands alone might offset the effects of fossil fuel emissions and land use changes for 10 to 20 yr or longer (Post and Kwon, 2000; Lal, 2004).

The agricultural sector can help reduce the enhanced greenhouse effect by introducing agronomic practices that result in increased removal of CO2 from the atmo-sphere. Carbon dioxide, one of the most important gases that enhance the greenhouse effect, is produced when coal, oil, wood, and other carbon-based fuels are burned. Plants absorb CO2 and through photosynthesis convert it into dry matter (e.g., food, fiber, wood). Carbon fixed by plants can remain in the form of wood for several years and/or return to the soil as plant residues increasing the soil organic matter content. The enhanced carbon seques-tration strategy in the agricultural sector cannot be viewed as the permanent solution for the GHG emission problem, but it can be an excellent option for “buying time” and allow for the development and global adoption of new, clean, and safe energy sources.

Past and recent research has evidenced that reduction in atmospheric carbon dioxide content can be achieved by large-scale applications of land management practices (Lal, 2004). Others include: reduced or zero tillage, use of pas-tures (e.g., clovers, alfalfa) in rotation with annual crops, improved strategies to enhance fertilizer use efficiency, increased efficiency of animal feed and return of animal waste, establishment of forests and grasslands in former croplands and degraded soils. Most importantly, increasing sequestered carbon in the soils provides additional benefits to farmers such as improvement in soil fertility, water hold-ing capacity, and tilth as well as reduction in soil erosion.

Two other gases with greenhouse effect are also important in the agricultural sector: nitrous oxide (N2O) and methane (CH4). The importance of these gases derives from their warming potential which is much higher than that of the CO2; the global warming potential of meth-ane is about 20 times higher and that of nitrous oxide is 300 times higher than that corresponding to CO2 (IPCC, 2007a). Nitrous oxide is mainly produced through trans-formations in the soil of the nitrogen added as fertilizers and/or plant residues (Mosier et al., 1998). Methane in the agricultural sector is mainly produced as a result of


the warming that is already unavoidable due to past and current emissions. More realistic scenarios of GHG emis-sions and atmospheric concentrations impose more aggres-sive needs of the different socioeconomic sectors including the agricultural sector, to further develop adaptive strate-gies to the already changing climate.

ClimaTe Change adapTaTion and deCision making: unCerTainTies and ConfliCT of sCalesDecision makers (including those involved in policy) work-ing in the public and private sectors of developing countries typically confront the pressure to act in response to prob-lems that require immediate action. Moreover, the effect of such actions must also be evident during the usually short terms in which those decision makers operate (often a max-imum of 2–5 yr, sometimes up to 10 yr). Consequently, relatively lower priority is assigned to issues that deal with the long term, such as 50 to 100 yr in the future.

On the other hand, the scientific community has focused research on climate change and its impacts on societies mainly by proposing climate scenarios expected for the next decades (typically 70–100 yr in the future). This research approach and the communicated results have been crucial to raise the awareness of the general public on the climate change issue in both developed and developing countries. The research out-comes and the resulting increased public awareness have also contributed to current efforts to promote the use of cleaner energy sources, encourage practices that enhance carbon sequestration, and in general support actions conducive to reduce net GHG emissions.

At the same time, this research approach based almost exclusively in possible scenarios expected 70 to 100 yr in the future has also placed the issue of climate change as a problem that will affect societies in the future and in a timeframe that is far beyond the one in which policy-makers and decision makers operate. In addition, the pos-sible scenarios of future climate produced with the best available scientific methods include uncertainty levels that often impose further challenges to be considered in actual decision-making and planning activities. These uncer-tainties are partially due to limitations of the scientific knowledge included in the climate models that are used to produce the scenarios. Some uncertainty is just inher-ent in the climate system even with perfect knowledge and representation of the physical processes and climate feedbacks. Uncertainties are also the result of assumptions that need to be made about the characteristics of differ-ent socioeconomic scenarios used to estimate the GHG emission levels that drive the climate models. Thus the socioeconomic scenarios include a wide range of assump-tions dealing with trade, energy sources, technology transfer, etc., for the next 70 to 100 yr that inevitably embrace uncertainties.

the enteric fermentation occurring in the digestion pro-cess of ruminants and in flooded soils. A second important role of the agricultural sector contributing to reduce the greenhouse effect is therefore to reduce the emissions of these gases. Since the emission of both gases is the result of inefficiencies in the production system, a reduction of the emissions would also lead to better results for the farmers (higher nitrogen use efficiency and more efficient conver-sion of animal feed into milk, meat, and wool).

The international community has reacted to the increased evidences of global warming by signing the Kyoto Protocol (United Nations Framework Convention for Climate Change [UNFCCC]). The protocol introduced mechanisms to reduce the net greenhouse gas emissions. While Kyoto negotiations are stalled for the time being, the discussions in the UNFCCC have stimulated an impressive amount of activities all over the world, with the increasing involvement of governments, business people, and scien-tists. In contrast with the slow progress in the government negotiations of the Kyoto Protocol, the results of these side-track activities have been impressive, giving shape to the development of an international carbon market (examples of functioning markets are evident in Australia, Denmark, the United Kingdom, and the United States, among others) (Baethgen and Martino, 2004).

The development of mature carbon markets will prob-ably encourage the establishment of collaborative projects between industry and farmers which will lead the latter to adopt agronomic practices that will result in increased amounts of sequestered carbon and reduced emissions of GHG (such as N2O and CH4). This in turn will provide farmers with additional income as well as improve their production systems and natural resource base. On the other hand it will allow the industry to reduce their net GHG emissions during the process of adopting cleaner processes and energy sources. Important challenges, however, remain for the implementation of carbon-market oriented projects in the least developed countries, given the potential diffi-culties for small farmers (“atomized” carbon offer) to take advantage of the existence of such projects.

The need of soCieTies To adapT To a Changing ClimaTeEven under the most optimistic scenarios of globally coor-dinated actions to drastically reduce the net emissions of greenhouse gases (GHG) during the next decades, climate science confirms that warming is already unavoidable due to past emissions. As indicated by IPCC’s Fourth Assessment Report (IPCC, 2007a, b), granting that the atmospheric GHG concentrations remain at 2000 levels, the inertia of past emissions is estimated to cause some unavoidable warming and consequent changes in climate. Accordingly, even under this unrealistically optimistic scenario, adapta-tion will be necessary to address the impacts resulting from


Uncertainties are larger for establishing possible scenar-ios of rainfall as compared to those of global temperatures and even much larger for the climate scenarios at regional or local scales (as opposed to global scale) that are the ones typically needed for planning and decision making. With respect to the latter, some recent methods allow the down-scaling of global model runs to regional or even local lev-els. However, it should be noted that these methods do not reduce the uncertainty levels associated with the global sce-narios. In fact, these downscaled climate change scenarios can be viewed as scenarios with higher level of detail of the same (and often larger) level of uncertainty.

The challenge of effectively incorporating the infor-mation resulting from the research in climate change into decision making is thus complicated by the uncer-tainty levels as well as by the frequent “double conflict of scales.” On the one hand the temporal scales of climate change scenarios are much farther in the future than the ones often needed for decision making and planning. On the other hand the spatial scales of the climate scenarios that can be established with the currently best available tools and methods still have a larger spatial scale (global to regional level) than the ones often needed for actual deci-sion making (regional to local level).

a ComplemenTary approaCh To adapTaTion To ClimaTe ChangeThe earth’s climate system includes processes that cause variability at different temporal and spatial scales. Some processes are local and act in the short or immediate term (a few days) and cause the variability of “weather.” Other pro-cesses are affected by the interaction of the atmosphere with the oceans and the land surface. Some of these processes result in variations of climate at the scale of months to sea-sons (e.g., those processes affected by El Niño). Still other phenomena depend on natural and anthropogenic factors that affect the chemical composition of the atmosphere and cause variability of the climate at the scale of several decades to centuries. The latter includes the variability of climate that is commonly referred to as climate change.

All of these processes act simultaneously and result in the observed earth’s total climate variability. The magni-tude of climate variability at these temporal scales is differ-ent and the relative magnitude also varies among regions of the world. An example of the relative magnitude of climate variability at different time scales is shown in Fig. 1. The figure was constructed by partitioning the total variability observed in annual rainfall in the Sahel for the period 1900 to 2006. Panel (a) shows the rainfall variabil-ity at the long-term (linear trend in the last 100 yr, which is a crude representation of the man-made climate change signal), the scale that is usually called climate change. The second panel (b) shows the variations of rainfall measured at the decadal scale (after removing the linear trend) and

reveals decades when rainfall tended to be above average (e.g., the 1950s and the early 1960s) and decades when rainfall tended to be below average (e.g., the 1970s and 1980s). Finally, panel (c) shows the variability of rainfall in the year-to-year time scale that remains after removing the linear and the decadal trends. The figure shows the relative magnitude of the rainfall variability at these three temporal scales as measured by the percent of the total variance explained by each temporal scale. The proportion of total variance explained by the short-term (interannual) variability is three times greater than the corresponding to the long-term variability (climate change) and two times greater than that of the decadal variability.

In other regions of the world the relative magnitude of the variability of climate at the various temporal scales (long-term, decadal, and interannual) is different. For example in southeastern South America the long-term, linear trend in the observed changes in rainfall seems to be much larger that the multidecadal variability (data not shown).

The slow and persistent forcing of increasing GHGs is producing noticeable changes in the mean climate on which shorter-term variability is superimposed. Increas-ing GHGs may also change the magnitude of the short-term variability, for example by enhancing the strength of the hydrological cycle. Changes in the mean state, the variability, or both will alter the statistical distribution of climate and weather and will likely result in more frequent extreme events that can have devastating socioeconomic and environmental impacts. Consequently, an effective manner for assisting societies to be better prepared and adapt to possible climate change scenarios is by assisting them to cope better with current climate variability.

Thus, a possible approach to introduce the issue of “adaptation to climate change” into the policy and devel-opment agendas is to consider the longer-term variations (climate change) as part of the continuum of the total cli-mate variability, from seasons to decades to centuries, and generate information at the temporal scale that is relevant and applicable for the particular time frames or planning horizons of the different decisions. This approach allows considering climate change as a problem of the present (as opposed to a problem of the future) and aims to inform the decision-making, planning, and policy-making pro-cesses to reduce current and potential future vulnerabili-ties to climate variability and change.

One of the key premises of this approach to engage in adaptation to climate change is that improving year-to-year planning activities and decisions lead to societies that are better adapted to longer term climate change (Hansen et al., 2007). However, there are situations in the different socioeconomic sectors where important issues require fun-damentally different approaches and activities. Thus, several important decisions need information and climate projec-tions at temporal scales of 10 to 30 yr (e.g., transportation


infrastructure projects, water reservoir design, long-term business plans, etc.). Therefore, focus is also needed on climate risk management work for adaptation to “near-term” climate change, that is, 10 to 30 yr. These time frames require the consideration of “decadal climate variability,” which is still posing important scientific challenges and the climate sci-ence community is investing huge efforts in exploring ways to improve the ability to predict it. In the meantime much can be gained by interpreting and characterizing the decadal trends in the observed historic records and on methods for producing seasonal forecasts under a changing climatic base-line (as opposed to the “static” baseline).

An advantage of this approach is that it provides immediate assistance to the public and private sector: while it helps stakeholders to confront possible future cli-mate scenarios, it identifies immediate actions needed to manage the climate variability that is currently affecting societies. Furthermore, the impacts of the taken actions and interventions are also evident and verifiable in the short term making them more attractive to policymak-ers and decision makers. Research organizations such as the IRI (International Research Institute for Climate and Society) are focusing on this approach and labeling it Cli-mate Risk Management (CRM).

four pillars of iri’s ClimaTe risk managemenT approaChThe outcome of socioeconomic activities (agriculture and water resource management) as affected by climate can be represented by probabilistic curves (Fig. 2). Thus, a few years present very unfavorable climatic conditions (droughts, floods, and hurricanes) and the socioeconomic impacts are extremely negative (“Disasters”). The dam-age caused by these relatively infrequent years can be so large that planning is often designed with the priority of avoiding or minimizing such damage. For example, farm-ers often prefer to use “precautionary” technologies that do not have expectation of high yields but that reduce the chances of high losses in unfavorable conditions. In the water sector, managers often prefer very conservative strategies to minimize the chances of not being able to supply water for all intended uses in very dry years.

This type of strategies heavily influenced by the risk aver-sion of decision makers may be effective in reducing losses in extreme conditions but they also imply missing opportuni-ties that can be critical for development. Thus, near normal or favorable conditions (about two-thirds to the right of the curve in Fig. 2), which are much more frequent than the disastrous conditions, offer the possibility to for example optimize agricultural income through higher productivity. Near normal or favorable years are much more frequent than the disastrous years, and therefore the sum of missed oppor-tunities can have much greater impacts on the economies and on development. However, since the impacts of even a

single extreme negative event can be so devastating that deci-sion makers rightly adopt precautionary strategies to protect against those impacts. There is a cost associated with such an approach, namely the lost opportunities of the more frequent favorable years that could be captured if we could protect against the negative extremes.

The CRM approach as understood in the IRI seeks managing the entire range of climate-related risks: from the very unfavorable conditions (extreme left area of Fig. 2) up to the “risk of missing opportunities” (e.g., about two-thirds of the area to the right of Fig. 2). Managing the entire risk range is based on four main pillars:

Fig. 1. Partition of the total observed rainfall variability in the Sahel. Rainfall is expressed as anomalies (i.e., deviations from the mean annual rainfall of 1900–2006). (a) long-term variability (linear trend), (b) decadal variability (after removing the linear trend), (c) inter-annual variability (after removing the linear and decadal trends).


1. Identify vulnerabilities and potential opportu-nities due to climate variability and/or change for a given water, agriculture, or health system. This pro-cess begins with stakeholder analysis, as they iden-tify their climate challenges, and then proceeds with modeling of the system being analyzed to identify other vulnerabilities and/or opportunities that stake-holders may not identify.

2. Quantify uncertainties in “climate informa-tion” to reduce uncertainties in using that informa-tion. A better understanding of the climate aspects of these vulnerabilities, challenges and opportunities, such as predictability, expected recurrence, and pos-sible long-term changes requires: (i) understanding climate variability at different time scales and assess-ing the socioeconomic impacts observed in the past, (ii) monitoring the present conditions of relevant environmental factors (climate, vegetation, water, diseases, etc.), and (iii) providing the best possible climate information of the future, from seasons to decades depending on the relevance for different decisions and activities1.

3. Identify technologies and practices that opti-mize results in normal or favorable years as well as technologies and practices that reduce vulnerabilities to climate variability and change (examples in agri-culture include crop diversification, crop rotations,

improved tillage systems, increased water soil stor-age, improved crop water use efficiency, drought-resistant cultivars).

4. Identify interventions, institutional arrange-ments and best practices that reduce exposure to climate vulnerabilities and enable the opportunistic exploitation of favorable climate conditions. Exposure reduction can be achieved through, for example: (i) improved early warning and response to crisis (e.g., improved emergency systems) and (ii) transferring portions of the existing risks (e.g., different modali-ties of rural insurance, supervised or differential credit programs, etc.). Risk-transfer instruments require efforts to characterize and quantify the different risk levels (“Disasters,” “Harm,” etc.) which vary for dif-ferent production systems and for different regions of the world. Such characterization and quantification of risk levels is in turn a key input for institutions that design insurance (and reinsurance) policies.

Typically a portfolio of approaches will be necessary, for example, insurance covering extreme negative events, diversification covering moderately negative events, and forecast or scenario use and access to the means to capture good year opportunities (e.g., fertilizer, improved crop seeds in agriculture) to take advantage of favorable climate conditions given the downside risk is covered by other parts of the portfolio (Fig. 2).

An advantage of this approach is that it provides imme-diate assistance to the public and private sector: while it helps stakeholders to confront possible future climate sce-narios, it identifies immediate actions needed to manage the climate variability that is currently affecting societies. Furthermore, the impacts of the taken actions and inter-ventions are evident and verifiable in the short term making them more attractive to policymakers and decision makers.

International agencies and development banks are increasingly adopting this approach as a means to effec-tively incorporate adaptive measures into policies and development plans. For example, Mr. Warren Evans, Environment Director at the World Bank, addressing the Global Environmental Fund Assembly in Cape Town (August, 2006, Cape Town) stated that: “adaptation to cli-mate risks needs to be treated as a major economic and social risk to national economies, not just as a long-term environment prob-lem. By enhancing climate risk management, development insti-tutions and their partner countries will be able to better address the growing risks from climate change and, at the same time, make current development investments more resilient to climate variabil-ity and extreme weather events.”

Fig. 2. Example of Climate Risk Management for a rainfed agricultural production system. Adequate technologies (increased soil water storage, drought resistant cultivars, crop–pasture rotations, diversification) provide protection against below average years (“Harm” to “Normal”). However, in extremely unfavorable years, improved technology is not sufficient, and so different modalities of insurance can cover this residual risk. Finally, in above average years, additional income can be generated given that a skillful forecast is provided and that protections are in place if the forecast leads to an ex-post “incorrect” decision.

1 The temporal scale (seasons, years, or decades) of the information that is needed is defined by the needs of the stakeholders demanding it. Farmers usually demand information at seasonal to inter-annual scales; development banks, foresters, and water reservoir builders may be interested in the likelihood of decades with frequent droughts or floods; and national authorities negotiating in the UNFCCC may require climate scenarios for the next 50 or more years.


referencesBaethgen, W.E. 1997. Vulnerability of the agricultural sector of

Latin America to climate change. Clim. Res. 9:1–7.Baethgen, W.E., and D.L. Martino. 2004. Mainstreaming climate

change responses in economic development of Uruguay. In OECD global forum on sustainable development. Environ. Directorate. Environ. Policy Committee. ENV/EPOC/GF/SD/RD(2004)2. Paris, France.

Baethgen, W.E., and G.O. Magrin. 1995. Assessing the impacts of climate change on winter crop production in Uruguay and Argentina using crop simulation models. p. 207–228. In C. Rosenzweig et al. (ed.) Climate change and agriculture: Analysis of potential international impacts. ASA Spec. Publ. 59. ASA, Madison WI.

Hansen, J.W., W. Baethgen, D. Osgood, P. Ceccato, and R.K. Ngugi. 2007. Innovations in climate risk management: Pro-tecting and building rural livelihoods in a variable and chang-ing climate. J. Semi-Arid Trop. Agric. Res 4(1):1–38.

Hill, J., E. Nelson, D. Tilman, S. Polasky, and D. Tiffany. 2006. Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proc. Natl. Acad. Sci. USA 103:1206–1210.

IPCC. 1992. Climate change: The IPCC response strategies. WMO/UNEP, Island Press, Covelo, CA.

IPCC. 2007a. Intergovernmental panel on climate change, work-ing group I contribution of to the fourth assessment report, climate change 2007: The physical science basis, summary for policymakers. http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-spm.pdf (verified 21 Jan. 2010).

IPCC. 2007b. Intergovernmental panel on climate change, work-ing group II contribution to the fourth assessment report, climate change 2007: Climate change impacts, adaptation and vulnerability, summary for policymakers http://www.

ipcc-wg2.gov/AR4/website/spm.pdf (verified 21 Jan. 2010).Kolshus, H.H.. 2001. Carbon sequestration in sinks: An overview

of potential and costs. CICERO Working Paper 2001. Cent. for Int. Clim. and Environ. Res., Oslo, Blindern, Norway.

Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food security. Science 304:1623–1627.

Mosier, A.R., J.M. Duxbury, J.R. Freney, O. Heinemeyer, and K. Minami. 1998. Assessing and mitigating N2O emissions from agricultural soils. Clim. Change 40:7–38.

Parry, M.L., and C. Rosenzweig. 1994. Potential impact of cli-mate change on world food supply. Nature 367:133–138.

Parry, M.L., C. Rosenzweig, A. Iglesias, M. Livermore, and G. Fischer. 2004. Climate change effects of climate change on global food production under SRES emissions and socio-eco-nomic scenarios. Glob. Environ. Change 14:53–67.

Post, W.M., and K.C. Kwon. 2000. Soil carbon sequestration and land-use changes: Processes and potential. Glob. Change Biol. 6:317–327.

Rosenzweig, C., and A. Iglesias (ed.). 1994. Implications of climate change for international agriculture: Crop modeling study. USEPA 230-B-94-003, Washington, DC.

Schneider, S., J. Sarukhan, J. Adejuwon, C. Azar, W.E. Baethgen, C. Hope, R. Moss, N. Leary, R.Richels, and J.P. van Yper-sele. 2001. Overview of impacts, adaptation, and vulnerabil-ity to climate change. pp 75–103. In J.J. McCarthy, et al. (ed.) Climate change 2001: Impacts, adaptation and vulnerability. Contribution of working group II to the third assessment report of the intergovernmental panel on climate change (IPCC). Cambridge Univ. Press, Cambridge.

Sims, R.E.H. 2004. Renewable energy: A response to climate change. Sol. Energy 76:9–17.

Von Braun, J. 2008. Food prices, biofuels and climate change. Int. Food Policy Res. Inst., Washington, DC.


symposia

The cereals provide fundamental contributions to the human and animal nutrition. Although conventional breeding

approaches manipulating genetic variation have been very suc-cessful in improving the agronomic properties of these crops, the next waves of crop improvement will require much greater knowledge of gene function. However, the process of identifying genes with specific functions is very slow and complicated outside of the few model plant species.

In model plants such as Arabidopsis and rice (Oryza sativa L.), many tools of molecular biology have been developed to greatly improve the process of isolating genes. Two tools that have sig-nificantly accelerated the speed with which gene function can be determined in model plants are T-DNA knockout libraries and T-DNA activation libraries (Weigel et al., 2000). Here, large collections of plants have been generated that contain insertions of T-DNAs. In knockout libraries the T-DNAs interrupt and thereby inactivate the genes they have inserted into, while in

Rapid Determination of Gene Function by Virus-induced Gene Silencing in Wheat

and Barley

Cahid Cakir, Megan E. Gillespie, and Steven R. Scofield*

abstractThecereal cropsare essential components tothe human and animal food supply. Solutionsto many of the problems challenging cerealproduction will require identification of genesresponsible for particular traits. Unfortunately,theprocessofidentifyinggenefunctionisveryslow and complex in crop plants. In wheat(Triticum aestivum L.) and barley (Hordeum vulgare L.), this process is made very difficultby the very large size and complexity of theirgenomes and the difficulty with which thesecropscanbegeneticallytransformed.Addition-ally,thepolyploidyofwheatgreatlycomplicatesany approach based on mutational analysisbecause functional, homeologousgenes oftenmask genetic mutations. Virus-induced genesilencing (VIGS) is an important new tool thatovercomesmanyoftheseobstaclesandprom-isestogreatlyfacilitatetheassessmentofgenefunction.AVIGSsystembasedonbarleystripemosaicvirus (BSMV)has recentlybeendevel-oped foruse inwheatandbarley.TheBSMV-VIGSsystemallowsresearcherstoswitch-offor“knockdown” the expression of chosen genesso that the gene’s function may be inferredbasedontheknockoutphenotypes.ThisarticledescribesthecharacteristicsoftheBSMV-VIGSsystem, relates examples of its application forfunctional genomics in wheat and barley, anddiscusses the strengths and weaknesses ofthis approach.

C. Cakir, USDA-ARS, Cropping Systems Research Lab., 3810 4th St., Lubbock, TX 79415; M.E. Gillespie and S.R. Scofield, Dep. of Agron-omy, Purdue Univ., 915 West State St., West Lafayette, IN 47906; S.R. Scofield, USDA-ARS, Crop Production and Pest Control Research Unit, 915 West State St., West Lafayette, IN 47907. Received 2 Oct. 2009. *Corresponding author ([email protected]).

Abbreviations: Bgh, Blumeria graminis f. sp. hordei; BLN1, Blufensin1; BSMV, barley stripe mosaic virus; CCR1, Cochliobolus carbonum race 1; dsRNA, double-stranded RNA; HR, hypersensitive response; NB-LRR, nucleotide binding–leucine-rich repeat; PDS, phytoene desaturase; QTL, quantitative trait locus; R-genes, resistance genes; RNAi, RNA interference; siRNA, small interfering RNA; T-DNA, transferred DNA; VIGS, virus-induced gene silencing.




activation libraries the T-DNAs are engineered to have enhancer elements near their borders, which activate the expression of genes near the T-DNA insertion site. Once plants have been identified with the desired phenotypes, isolation of the relevant gene is readily accomplished by finding the genomic location of the T-DNA that cosegre-gates with the mutant phenotype. Unfortunately, none of these tools exist for wheat (Triticum aestivum L.) and barley (Hordeum vulgare L.) because these plants can only be trans-formed with very low efficiency and the genomes are so large that the number of transformants needed is too great. An additional complication for determining gene func-tion through the analysis of loss-of-function mutations in wheat is the fact that all cultivated varieties are polyploid and, therefore in most cases, expression of homeologous genes will mask loss-of-function phenotypes.

Virus-induced gene silencing (VIGS) is a tool for the rapid assessment of gene function that overcomes many of the limitations present in cereal crops. Recently a system for VIGS has been developed for use in wheat and barley. This article will discuss the properties of this system and how it is being used for functional genomics in wheat and barley. It will also discuss the strengths and weaknesses of this approach and what might be done to improve its utility.

What is ViGs?Virus-induced gene silencing is a very useful research tool for rapid creation of gene knockdown phenotypes that can be used to assess plant gene function (Baulcombe, 1999; Kumagai et al., 1995; Ratcliff et al., 1997). The biologi-cal principles on which VIGS is based were uncovered as molecular biologists studied the consequences of virus infection in plants (Lindbo and Dougherty, 1992). Through this work, it was discovered that many RNA viruses acti-vate a conserved, RNA-based plant antiviral defense response, which targets the RNA produced by infecting viruses for sequence-specific degradation (Ratcliff et al., 1997). This RNA-based plant defense response is triggered when double-stranded RNA (dsRNA) accumulates within cells, as occurs during the replication of RNA viruses. All the sequence within the dsRNA becomes targeted by the host defense system for sequence-specific degradation. This process is exploited in VIGS to permit researchers to down-regulate or “knockdown” the expression of plant genes of their choosing by infection with engineered viruses. By inserting a fragment of transcribed sequence from a plant gene, which the researcher wishes to silence, into the VIGS construct, transcripts of the gene-of-interest are targeted to undergo homology-dependent degradation, thereby caus-ing the gene to be silenced.

Several aspects of VIGS make it a particularly useful tool for plant functional genomics. (i) It is a rapid experi-mental process. In most instances, the knockdown pheno-type of a gene-of-interest is generated within 1 to 2 mo

of identifying the candidate sequence. This is far quicker than what is possible through the production and analysis of knockout mutant or stably transformed RNA interfer-ence (RNAi) plants. (ii) VIGS does not require full-length cDNA sequences to function, so experiments can be ini-tiated without complete gene sequence information. (iii) VIGS is initiated by infecting plants with a viral construct, so it is possible to observe the effect of transient silencing of genes that would have homozygous lethal phenotypes in conventional mutant analyses. (iv) VIGS can be partic-ularly useful for research in polyploid plants such as wheat because gene silencing occurs through homology-depen-dent RNA-mediated gene silencing and, therefore, any genes sharing at least : 85% sequence identity are likely to be down-regulated. In this way, knockdown phenotypes can be observed because the closely related homeologous genes present in polyploids are likely to be silenced as well. (v) VIGS is initiated by virus infection and so can be performed on species that are difficult to transform for stable RNAi studies.

In principle VIGS should be a very useful tool for research in any plant species. However, a major limitation to its widespread adoption is the lack of viral vectors that generate useful gene silencing in different plant species. Initially, VIGS was almost exclusively performed in Nicoti-ana benthamiana (Karel Domin) using vectors derived from tobacco mosaic virus (Kumagai et al., 1995), potato virus X (Ratcliff et al., 1997), and tobacco rattle virus (Liu et al., 2002; Ratcliff et al., 2001). In recent years, new proto-cols and vectors have expanded the list of dicotylendonous plants in which VIGS can be employed, for example, Ara-bidopsis (Burch-Smith et al., 2006) and potato (Solanum tuberosum L.) (Brigneti et al., 2004; Faivre-Rampant et al., 2004), but it was not until recent experimentation with barley stripe mosaic virus (BSMV)–based vectors that VIGS became an option for functional genomics research in wheat (Scofield et al., 2005) and barley (Hein et al., 2005; Holzberg et al., 2002).

a ViGs system based on barley stripe mosaic VirusBarley stripe mosaic virus is a single-stranded RNA virus of the Hordeivirus genus. Its genome is tripartite, comprised of the a, b, and g RNAs. Petty et al. (1989) synthesized cDNAs of the three RNAs and cloned them each into DNA plasmids such that infectious BSMV RNAs could be synthesized from the three plasmids by in vitro transcrip-tion of capped RNAs. Barley stripe mosaic virus infection is initiated by mixing together in vitro transcripts from the a, b, and g DNA plasmids and rub-inoculating them onto susceptible host plants.

Holzberg et al. (2002) first demonstrated that BSMV could trigger useful VIGS in barley plants. In this ini-tial study, it was demonstrated that BSMV constructs


barley are each controlled by nucleotide binding–leucine-rich repeat (NB-LRR) class resistance genes (R-genes). The NB-LRR class of R-genes is the most extensively studied class of R-genes. The NB-LRR proteins are known to either directly or indirectly detect the pres-ence of elicitors expressed by avirulent pathogens and initiate a signal-transduction process that results in the hypersensitive response (HR) and activation of patho-gen defense responses. These BSMV-VIGS studies each demonstrated that resistant genotypes became susceptible when infected with BSMV-VIGS constructs designed to silence the wheat Lr21 and barley Mla13 R-genes (see Fig. 2). Resistant genotypes infected with control constructs remained resistant, indicating that BSMV infection did not disturb plant physiology sufficiently to interfere with these R-gene responses. Both studies also tested whether the Lr21- and Mla13-resistant responses were dependent on the expression of the RAR1, SGT1, or HSP90 genes, which have been found to be essential in other NB-LRR resistance pathways of Arabidopsis, tobacco (Nicotiana taba-cum L.), tomato (Lycopersicon esculentum Mill.), and barley. The BSMV-VIGS experiments showed that Lr21- and Mla13-mediated resistance were abrogated when RAR1, SGT1, and HSP90 were silenced, indicating their essen-tial role in these resistance pathways.

Two published studies have employed BSMV-VIGS in the final stages of map-based cloning to functionally confirm that the correct gene had been isolated. BSMV-VIGS was used to functionally confirm that the wheat Lr1 leaf rust R-gene (Cloutier et al., 2007) and the barley Rpg5 stem rust R-gene (Brueggeman et al., 2008) had been iso-lated after chromosome-walking procedures. Before the availability of a suitable VIGS system, confirmation would have been performed by the time-consuming step of transforming a susceptible genotype with a construct that would express the candidate R-gene. In fact, transforma-tion was required to unambiguously identify Lr1 because it was found that several highly homologous NB-LRR genes were present at the Lr1 locus. The VIGS analy-ses performed in this study were unable to discriminate between the highly homologous R-gene-like sequences at this locus (Cloutier et al., 2007).

Microarray studies have generated large lists of differ-entially expressed genes during compatible and incompat-ible interactions. The great challenge now is to sort through these lists to determine which genes have causal roles in the outcome of the plant–pathogen interactions. Several recent studies have employed BSMV-VIGS to test if genes implicated in disease resistance by induction of transcrip-tion during resistance may have bona fide functions in resistance. Zhou et al. (2007) took this approach to test if three wheat receptor-like kinase genes (TaRLK 1, 2, and 3), which were up-regulated during incompatible reac-tions with stripe rust, had functional roles in resistance.

engineered to carry 0.19- to 1.4-kb inserts of the barley phytoene desaturase (PDS) gene would cause down-reg-ulation of PDS expression. Phytoene desaturase expres-sion is essential for the synthesis of carotenoid pigments, which protect chlorophyll from photolysis. Tissue in which PDS had been down-regulated can therefore be visualized by the appearance of photobleaching where chlorophyll had undergone photolysis (see Fig. 1). Typi-cally in BSMV-VIGS studies, a 120- to 500-bp fragment, representing a portion of a transcribed sequence from a plant gene, is inserted into the g RNA plasmid at restric-tion sites immediately 3¢ to the stop codon of the gb gene (Holzberg et al., 2002). The 120-bp minimum size for the plant gene fragment is based on the observation that host insert sequences <120 bp are significantly less effective in BSMV-VIGS (Bruun-Rasmussen et al., 2007; Scofield et al., 2005). The upper size limit of 500 bp is less well defined, but reflects the fact that all sequences inserted into plant viral vectors are unstable as the virus replicates (Pogue et al., 2002), and larger fragments may be lost with greater frequency (Bruun-Rasmussen et al., 2007; Cakir and Scofield, 2008).

Unlike genetic mutations abolishing gene function or gene silencing in transgenic plants expressing RNAi constructs, silencing in VIGS occurs transiently. The temporal and spatial patterns of gene silencing have been analyzed for BSMV-VIGS in wheat and barley seedlings. When BSMV infection is initiated on the second leaf, BSMV moves systemically into the third leaf and signifi-cant silencing can be detected there 3 d postinoculation (dpi) and will persist until at least 21 dpi (Hein et al., 2005; Scofield et al., 2005).

the utility of bsmV-ViGs for functional Genomics in Wheat and barleyMuch of the initial use of the BSMV-VIGS system has been in the functional analysis of disease resistance path-ways in wheat and barley. The general design of these experiments is to initiate silencing of a candidate gene in a wheat genotype that is normally resistant to a pathogen of interest, challenge the silenced plants with this pathogen, and then assess whether or not the plants are still resis-tant to the pathogen. Conversion to susceptibility of plants infected with the viral constructs silencing the candidate gene, while plants infected with control viral constructs remain resistant, is strong evidence for the candidate gene having an essential function in the resistance pathway.

The first published studies employing BSMV-VIGS demonstrated the system’s utility in the functional dis-section of two gene-for-gene disease resistance pathways from wheat (Scofield et al., 2005) and barley (Hein et al., 2005). Lr21-mediated resistance to leaf rust in wheat and the Mla13-mediated resistance to powdery mildew in


Figure 1. Silencing phytoene desaturase (PDS) in the leaves of hexaploid wheat by barley stripe mosaic virus–virus-induced gene silencing (BSMV-VIGS). The first and second leaves of wheat plants were inoculated with a control construct BSMV:00, which carries no plant gene sequence, and BSMV:PDS, which carries a 185-bp fragment of the PDS gene. The white photobleached areas result from degradation of chlorophyll when PDS expression is silenced. (A higher-magnification image of photobleaching can be seen at the bottom left image in Fig. 2.)

Figure 2. Using barley stripe mosaic virus–virus-induced gene silencing (BSMV-VIGS) to identify genes functioning in Lr21-mediated resistance to leaf rust. All plants were infected with the indicated BSMV constructs 7 d after germination and then spray-inoculated with the leaf rust fungus Puccinia triticina 8 d after viral infection. The photographs were taken 10 d after inoculation with the leaf rust fungus. R, WGRC7 genotype expressing Lr21-mediated resistance; S, the susceptible genotype Wichita, which lacks Lr21. Left column photographs: Infection with control constructs does not alter resistance or susceptibility. Right column photographs: BSMV constructs silencing Lr21, RAR1, SGT1, and HSP90 abrogate Lr21-mediated resistance. (Scofield et al., 2005).


Using VIGS constructs designed to specifically silence each TaRLK gene or constructs that target the silencing of all three genes, they observed that all three TaRLKs contrib-ute to stem rust resistance (Zhou et al., 2007).

Two recent publications from the Wise laboratory have also used BSMV-VIGS to test for functional roles in resis-tance of genes first implicated through differential expres-sion detected in microarray analysis of barley–powdery mildew interactions (Hu et al., 2009; Meng et al., 2008). Both of these studies are focused on identifying the genetic pathways of basal defense, which functions independently of R-genes. Transcription of a small family of genes encod-ing barley-specific peptides was found to be up-regulated during interactions with the powdery mildew fungus, Blu-meria graminis f. sp. hordei Speer (Bgh). Silencing this gene family, called Blufensin1 (BLN1), by BSMV-VIGS results in enhanced resistance that is independent of Mla R-genes. When BLN1 was silenced in plants expressing the Mlo sus-ceptibility factor, they also observed increased resistance. Taken together, they conclude that BLN1 functions in basal defense as a negative regulator of resistance to pen-etration by Bgh. The second study examines three genes, chorismate synthase, anthranilate synthase, and chorismate mutase, which function in the aromatic amino acid biosyn-thetic pathway and are coordinately induced from 0 to 16 h after infection in compatible and incompatible interactions with Bgh (Meng et al., 2008). Silencing each of these genes by BSMV-VIGS results in increased penetration by Bgh, while overexpression causes increased resistance, strongly suggesting a function for these genes in resistance to pen-etration by Bgh.

When an entire species is resistant to a pathogen, it is termed nonhost resistance. This is one of the least understood areas of plant disease resistance and likely results from a wide variety of mechanisms. Johal and Briggs (1992) iso-lated the first nonhost resistance gene from plants, the HM1 gene of maize (Zea mays L.). All maize plants are resistant to the pathogen Cochliobolus carbonum Nelson race 1 (CCR1), except mutants in which the HM1 and HM2 loci are nonfunctional. The hm1/hm2 mutants are highly susceptible to CCR1, which kills the mutant plants at any stage of their growth. Johal and Briggs (1992) found that HM1 encodes a reductase enzyme, which detoxifies the toxin produced by CCR1 that is absolutely required for pathogenesis. They found that HM1 homologs are only found in grass species, and as no other grasses are hosts to CCR1, it begged the question: Do HM1 genes pres-ent in other grasses provide nonhost resistance to CCR1? This hypothesis was tested by employing BSMV-VIGS to silence the six copies of HM1 present in barley (Sindhu et al., 2008). It was found that silencing the barley HM1 genes conferred very strong susceptibility to CCR1. This result is quite sobering to contemplate. The grass family provides the most important crops for human survival,

and they are all being protected against this devastating pathogen by the HM1 gene, which arose uniquely in the grass lineage approximately 40 million yr ago (Sindhu et al., 2008). Understanding how this system has provided such durable resistance will be very important for future efforts to engineer disease resistance.

While most of the published applications of BSMV-VIGS have addressed the functional genomics of disease resistance pathways, there are a few studies investigating other biological systems. Aside from the studies charac-terizing BSMV-VIGS, where plant genes are silenced which give photobleaching phenotypes, there are two publications investigating the silencing of genes involved in cell wall biosynthesis. Oikawa et al. (2007) employed BSMV-VIGS to characterize the function of the P23k gene that is unique to monocots. Silencing P23k in bar-ley resulted in asymmetrically shaped leaves with frequent cracks along the margins. These results, together with the finding that P23k expression is induced in the vascular bundles, supports their assertion that this gene is involved in the synthesis of cell wall polysaccharides and second-ary wall formation. Held et al. (2008) utilized BSMV-VIGS to observe the consequences of silencing the barley cellulose synthase, HvCesA. Surprisingly, the HvCesA VIGS experiment resulted in not only down-regulation of HvCesA but also a number of nontarget cellulose syn-thase–like (Csl) genes. Investigation of the basis of this effect identified the existence of naturally occurring anti-sense transcripts of HvCesA. The antisense and sense transcripts form dsRNAs, which are then processed by a dicer enzyme to form small interfering RNAs (siRNAs). These siRNAs accumulate late in leaf development as cel-lulose synthesis decreases, consistent with their acting to coordinately down-regulate other Csl genes during cell wall biosynthesis. This result points not only to the power of VIGS, but also to the complexities of plant genomes that will be encountered as this technique is applied.

strenGths, Weaknesses, and future directionsGenetic improvement of cereal crops is crucial to meeting the rapidly increasing production requirements for world food supply. Genetic experimentation in wheat and barley has been greatly impeded by the size and complexity of their genomes and also aspects of their biology that prevent the easy implementation of the many advanced technolo-gies developed in model plants. The examples discussed here make it clear that BSMV-VIGS is opening many new avenues for functional genomics in wheat and barley. The ability to generate knockdown phenotypes without hav-ing to perform the difficult and time-consuming process of transformation and regeneration is a highly significant advantage, as is the ability to silence all copies of a gene present in complex, polyploid genomes.


However, BSMV-VIGS is certainly not without limi-tations and experimental complications. Perhaps the big-gest difficulty in employing any VIGS systems is variation in the extent with which silencing phenotypes develop. In the context of performing functional genomics assays, an optimal VIGS system would generate silencing over a predictable and sufficiently large area of the plant so that the expected phenotype can be easily recognized. VIGS results from a complex interaction between the plant and virus in which the size of the area exhibiting silencing is determined by the balance between the replication, move-ment, and pathogenicity of the virus and the strength of the silencing response mounted by the host. Movement of the virus and spreading of the silencing signals is driven by the source–sink relationships within the plant, so care-ful attention to plant growth is crucial for obtaining uni-form results. The addition of a pathogen assay to the VIGS experimental system adds a third organism and an addi-tional source of variation.

Very careful attention to controls is essential to suc-cessful VIGS experiments. Without question, infection by the virus has great effects on the physiological state of the host. Therefore, it is essential to run ample numbers of controls in which plants are infected with viral constructs that do not target plant genes for silencing. The control plants must be carefully assessed to ensure that there is no perturbation of the experimental phenotype to be scored in the experiment. Dealing with this variation and the large number of control and experimental plants required to manage the variation are the greatest factors limiting throughput in this experimental system.

Another essential element of VIGS analysis is con-firmation that the target gene has been silenced. This is typically done by quantitative reverse transcription poly-merase chain reaction (Bustin, 2005). This determination is particularly important when a VIGS experiment does not yield the expected phenotype. Finally, the possibility of “off-target” silencing has been reported in VIGS and RNAi experiments. Here a gene other than the intended target becomes silenced and is the true cause of the knock-down phenotype (Xu et al., 2006). A very strong way to refute this possibility is to perform another round of VIGS in which the same gene is targeted by VIGS constructs carrying fragments from the same gene, but which have no homology to the previous VIGS targeting sequence (Scofield et al., 2005). Generation of a similar knockdown phenotype using fragments of the same gene, but which have no sequence overlap, provides very strong evidence against “off-target” silencing.

All of the BSMV-VIGS studies to date have screened a small number of candidate sequences that were selected based on the results of other experimentation. The ultimate value of VIGS, however, would come when screens can be performed with sufficient throughput so that the function

of genes without a priori experimental support could be identified. There is one published account of a high-throughput VIGS assay in which 4992 unique sequences from a normalized tobacco cDNA library were screened for function in the hypersensitive response mediated by the Pto R-gene (Lu et al., 2003). This screen led to identifica-tion of 79 cDNAs whose expression was required for the Pto-mediated HR. Of these, six were then demonstrated to be essential for disease resistance. Screening this large num-ber of cDNAs was only possible because the potato virus X (PVX) VIGS system had been engineered so that VIGS could be initiated by injecting cultures of Agrobacterium-carrying T-DNA constructs that then expressed infectious PVX RNA. Additionally, the screening for Pto-mediated HR was performed by transiently infecting the silenced plants with Agrobacterium-carrying T-DNA constructs that transiently expressed the AvrPto elicitor in planta, thereby activating the Pto-mediated HR.

Efforts to adapt the BSMV system for more high-throughput applications are just beginning. The first-gen-eration BSMV-VIGS system requires synthesis of capped in vitro transcripts from the three plasmids carrying the BSMV genomic RNAs. This is both time consuming and expensive. A recent improvement that eliminates the need for in vitro transcription was developed in the Wise labo-ratory. In this system, plasmids have been constructed so that each BSMV RNA can be transcribed in planta from the cauliflower mosaic virus 35S promoter (CaMV35S), ribozyme processing sequences are cloned at the 3¢ end of the viral RNAs to generate the correct 3¢ termini. In this system, infection is initiated by mixing the three BSMV plasmids together and then biolistically bombarding the mixture into plants (Hu et al., 2009; Meng et al., 2008). Very recently, progress toward a T-DNA–based BSMV-VIGS system was reported by Jackson et al. (2009). Clearly these improvements should greatly increase the utility of BSMV-VIGS.

The majority of published studies employing BSMV-VIGS, and other VIGS systems as well, are investigating disease resistance pathways. It is not entirely clear why this is the case, although it should be noted that most of VIGS systems have been developed by research groups already studying plant–pathogen interactions. However, activation of disease resistance responses happens very quickly and utilizes signaling components that tend to be of low abundance in plant cells. Such pathways may be particularly well suited for analysis by the transient silenc-ing characteristic of VIGS. Transient silencing may be less effective in interrupting the function of pathways involv-ing proteins that are expressed at high levels or have long half-lives.

A substantial amount of experience has been gained using BSMV-VIGS to dissect disease resistance pathways of wheat and barley. Cereal research groups around the world


are rapidly adopting this technique. The increasing user base should lead to improvements in this technology and should increase its effectiveness for the analysis of genetic traits.

dedicationThis manuscript is dedicated to the memory of Profes-sor Mike Gale FRS, who enriched plant genetics with his many significant scientific contributions, enthusiasm, generosity, and keen sense of humor.

acknowledgmentsThe authors are very appreciative of the contributions of Amanda Brandt, Li Huang, Bikram Gill, and Guri Johal. This work was supported by U.S. Department of Agriculture, Agri-cultural Research Service Current Research Information Sys-tem (project no. 3602-21220-010) and the U.S. Wheat and Barley Scab Initiative (project no. FY09-SC-005).

referencesBaulcombe, D.C. 1999. Fast forward genetics based on virus-

induced gene silencing. Curr. Opin. Plant Biol. 2:109–113.Brigneti, G., A.M. Martin-Hernandez, H. Jin, J. Chen, D.C.

Baulcombe, B. Baker, and J.D. Jones. 2004. Virus-induced gene silencing in Solanum species. Plant J. 39:264–272.

Brueggeman, R., A. Druka, J. Nirmala, T. Cavileer, T. Drader, N. Rostoks, A. Mirlohi, H. Bennypaul, U. Gill, D. Kudrna, C. Whitelaw, A. Kilian, F. Han, Y. Sun, L. Gill, B. Steffenson, and A. Kleinhofs. 2008. The stem rust resistance gene Rpg5 encodes a protein with nucleotide-binding-site, leucine-rich, and protein kinase domains. Proc. Natl. Acad. Sci. USA 105:14970–14975.

Bruun-Rasmussen, M., C.T. Madsen, S. Jessing, and M. Albrecht-sen. 2007. Stability of barley stripe mosaic virus–induced gene silencing in barley. Mol. Plant Microbe Interact. 20:1323–1331.

Burch-Smith, T.M., M. Schiff, Y. Liu, and S.P. Dinesh-Kumar. 2006. Efficient virus-induced gene silencing in Arabidopsis. Plant Physiol. 142:21–27.

Bustin, S.A. 2005. Real-time, fluorescence-based quantitative PCR: A snapshot of current procedures and preferences. Expert Rev. Mol. Diagn. 5:493–498.

Cakir, C., and S.R. Scofield. 2008. Evaluating the ability of the barley stripe mosaic virus–induced gene silencing system to simultaneously silence two wheat genes. Cereal Res. Com-mun. 36:217–222.

Cloutier, S., B.D. McCallum, C. Loutre, T.W. Banks, T. Wicker, C. Feuillet, B. Keller, and M.C. Jordan. 2007. Leaf rust resis-tance gene Lr1, isolated from bread wheat (Triticum aestivum L.) is a member of the large psr567 gene family. Plant Mol. Biol. 65:93–106.

Faivre-Rampant, O., E.M. Gilroy, K. Hrubikova, I. Hein, S. Mil-lam, G.J. Loake, P. Birch, M. Taylor, and C. Lacomme. 2004. Potato virus X–induced gene silencing in leaves and tubers of potato. Plant Physiol. 134:1308–1316.

Hein, I., M. Barciszewska-Pacak, K. Hrubikova, S. Williamson, M. Dinesen, I.E. Soenderby, S. Sundar, A. Jarmolowski, K. Shirasu, and C. Lacomme. 2005. Virus-induced gene silenc-ing–based functional characterization of genes associated with powdery mildew resistance in barley. Plant Physiol.

138:2155–2164.Held, M.A., B. Penning, A.S. Brandt, S.A. Kessans, W. Yong,

S.R. Scofield, and N.C. Carpita. 2008. Small-interfering RNAs from natural antisense transcripts derived from a cel-lulose synthase gene modulate cell wall biosynthesis in barley. Proc. Natl. Acad. Sci. USA 105:20534–20539.

Holzberg, S., P. Brosio, C. Gross, and G.P. Pogue. 2002. Barley stripe mosaic virus–induced gene silencing in a monocot plant. Plant J. 30:315–327.

Hu, P., Y. Meng, and R.P. Wise. 2009. Functional contribution of chorismate synthase, anthranilate synthase and chrosimate mutase to penetration resistance in barley–powdery mildew interactions. Mol. Plant Microbe Interact. 22:311–320.

Jackson, A.O., H.-S. Lim, J.N. Bragg, U. Ganesan, and M.Y. Lee. 2009. Hordeivirus replication, movement and pathogenesis. Annu. Rev. Phytopathol. 47:385–422.

Johal, G.S., and S.P. Briggs. 1992. Reductase activity encoded by the HM1 disease resistance gene in maize. Science 258:985–987.

Kumagai, M.H., J. Donson, G. della-Cioppa, D. Harvey, K. Han-ley, and L.K. Grill. 1995. Cytoplasmic inhibition of carot-enoid biosynthesis with virus-derived RNA. Proc. Natl. Acad. Sci. USA 92:1679–1683.

Lindbo, J.A., and W.G. Dougherty. 1992. Untranslatable tran-scripts of the tobacco etch virus coat protein gene sequence can interfere with tobacco etch virus replication in transgenic plants and protoplasts. Virology 189:725–733.

Liu, Y., M. Schiff, R. Marathe, and S.P. Dinesh-Kumar. 2002. Tobacco Rar1, EDS1 and NPR1/NIM1 like genes are required for N-mediated resistance to tobacco mosaic virus. Plant J. 30:415–429.

Lu, R., I. Malcuit, P. Moffett, M.T. Ruiz, J. Peart, A.J. Wu, J.P. Rathjen, A. Bendahmane, L. Day, and D.C. Baulcombe. 2003. High throughput virus-induced gene silencing impli-cates heat shock protein 90 in plant disease resistance. EMBO J. 22:5690–5699.

Meng, Y., M. Moscou, and R.P. Wise. 2008. Bluefinsin1 nega-tively impacts basal defense in response to barley powdery mildew. Plant Physiol. 149:271–285.

Oikawa, A., A. Rahman, T. Yamashita, H. Taira, and S. Kidou. 2007. Virus-induced gene silencing of P23k in barley leaf reveals morphological changes involved in secondary wall formation. J. Exp. Bot. 58:2617–2625.

Petty I.T., B.G. Hunter, N. Wei, and A.O. Jackson. 1989. Infec-tious barley stripe mosaic virus RNA transcribed in vitro from full-length genomic cDNA clones. Virology 171:342–349.

Pogue, G.P., J.A. Lindbo, S.J. Garger, and W.P. Fitzmaurice. 2002. Making an ally from an enemy: Plant virology and the new agriculture. Annu. Rev. Phytopathol. 40:45–74.

Ratcliff, F., B.D. Harrison, and D.C. Baulcombe. 1997. A similar-ity between viral defense and gene silencing in plants. Science 276:1558–1560.

Ratcliff, F., A.M. Martin-Hernandez, and D.C. Baulcombe. 2001. Technical advance. Tobacco rattle virus as a vector for analysis of gene function by silencing. Plant J. 25:237–245.

Scofield, S.R., L. Huang, A.S. Brandt, and B.S. Gill. 2005. Devel-opment of a virus-induced gene-silencing system for hexaploid wheat and its use in functional analysis of the Lr21-mediated leaf rust resistance pathway. Plant Physiol. 138:2165–2173.

Sindhu, A., S. Chintamanani, A.S. Brandt, M. Zanis, S.R. Sco-field, and G.S. Johal. 2008. A guardian of grasses: Specific origin and conservation of a unique disease-resistance gene in the grass lineage. Proc. Natl. Acad. Sci. USA 105:1762–1767.


Weigel, D., J.H. Ahn, M.A. Blazquez, J.O. Borevitz, S.K. Chris-tensen, C. Fankhauser, C. Ferrandiz, I. Kardailsky, E.J. Malancharuvil, M.M. Neff, J.T. Nguyen, S. Sato, Z.Y. Wang, Y. Xia, R.A. Dixon, M.J. Harrison, C.J. Lamb, M.F. Yanof-sky, and J. Chory. 2000. Activation tagging in Arabidopsis. Plant Physiol. 122:1003–1013.

Xu, P., Y. Zhang, L. Kang, M. Roossinck, and K. Mysore. 2006. Computational estimation and experimental verification of

off-target silencing during posttranscriptional gene silencing in plants. Plant Physiol. 142:429–440.

Zhou, H., S. Li, Z. Deng, X. Wang, T. Chen, J. Zhang, S. Chen, H. Ling, A. Zhang, D. Wang, and X. Zhang. 2007. Molecular analysis of three new receptor-like kinase genes from hexa-ploid wheat and evidence for their participation in the wheat hypersensitive response to stripe rust fungus infection. Plant J. 52:420–434.


SYMPOSIA

Cereals, dominated by wheat, rice, and maize, provide approximately 50% of human food calories directly and con-

siderably more indirectly via feed grains (Tweeten and Thomp-son, 2008). Over the last 20 yr, a period chosen to best estimate current rates of progress without influence of earlier periods, the linear rates of yield change for the world (Fig. 1) have been 25 kg ha–1 yr–1 (wheat), 38 kg ha–1 yr–1 (rice), and 80 kg ha–1 yr–1 (maize). With the exception of maize in some regions, there is no evidence for exponential growth in yield. In fact, relative rates of yield increase are declining and, expressed relative to predicted yield in 2007, are 0.9% yr–1 for wheat, 0.9% yr–1 for rice, and 1.6% yr–1 for maize. Even if these relative rates could be main-tained, various studies suggest they would not prevent real price rises for the three cereals, in the face of projected demand growth to 2050 (Tweeten and Thompson, 2008). Thus there is little doubt that the world needs to continue increasing cereal yields.

In this paper we focus on factors determining current rates of yield progress in several key situations (or case studies) and consider

Breeding and Cereal Yield Progress

R. A. (Tony) Fischer* and Gregory O. Edmeades

ABSTRACTThispaperreviewsrecentprogressinwheat(Triticum aestivum L.), rice (Oryza sativa L.), and maize (Zea mays L.) yields resulting from substantial breedingefforts in mostly favorable environments and exam-ines its physiological basis. Breeding and improvedagronomyliftpotentialyield(PY),namelyyieldwiththebestvarietyandmanagementintheabsenceofman-ageableabioticandbioticstresses,andPYincreaseis a key component of progress in farm yield (FY),the other component being closure of the PY to FYgap.Changes inPYandFYarereviewedforseveralkeyproduction regions, namely theUnitedKingdomandtheYaquiValleyofMexicoforwheat,JapanandCentralLuzoninthePhilippinesforrice,andIowaandbrieflysub-SaharanAfrica formaize.ThePYgrowthrateshavefallenandarecurrentlygenerallynomorethan1%perannumandusuallymuch less.The tra-jectoryofFYwithtimeoftencloselyparallelsPY,but,especiallyindevelopingcountries,thereremainlargeyieldgaps.Inatleastoneinstance(maizeinIowa)thegapbetweenPYandFYappearstobeclosingrapidly.Current genetic progress is linked to increased bio-massaccumulation,andthiswillremainthewayfor-wardinthefuturegiventhelimitstoincreasedharvestindex (HI). There is evidence that recent progress isrelatedtoincreasedphotosynthesis(e.g.,greaterradi-ationuseefficiency(RUE)at thecanopy leveland/ormaximumphotosyntheticratePmaxatsaturating irra-dianceat the leaf level)beforeandaroundanthesis.There isno theoretical reasonwhy this trendcannotcontinue,especiallygiventhevastgeneticresourcesalready found within each crop species. However, itwillnotbeeasilyorcheaplyaccomplished,sopros-pectsforhigherratesofpotentialyieldgrowthappeartobelimited,notwithstandingnewmoleculartoolsandclaimstothecontrary.Closingtheyieldgap,therefore,becomesmoreimportant.Manyfactorsareinvolved,but breeding can also help farmers achieve thisthrough,forexample,improvedhostplantresistance.

R.A. Fischer, CSIRO Plant Industry, ACT, Australia; G.O. Edmeades, 43 Hemans St., Cambridge 3432, New Zealand. Received 2 Oct. 2009. *Corresponding author ([email protected]).

Abbreviations: AY, attainable yield; CIMMYT, International Maize and Wheat Improvement Center; DM, dry matter; DS, dry season; FY, farm yield; GM, genetically modified; HGCA, Home Grown Cereals Authority; HI, harvest index; IRRI, International Rice Research Institute; LAI, leaf area index; NPT, new plant type; NPT2, new plant type, second generation; Pmax, light-saturated photosynthetic rate; PY, potential yield; PYW, water-limited potential yield; RUE, radiation use efficiency; SLN, specific leaf nitrogen; TE, transpiration efficiency; Tmin; minimum daily temperature; WS, wet season.




the physiological bases of this progress. We consider also the prospects for continued yield growth, in particular that resulting from plant breeding, which we believe is becom-ing a proportionally larger component of yield growth. Possible yield changes due to shifts in cropping regions or proportions irrigated, changes in cropping intensity, or climate change (apart from CO2 increase itself ) are not discussed, although these are also factors that need to be considered for a complete understanding of cereal yield changes. For example, wheat is likely to be gradually dis-placed from irrigated areas by higher value crops (fruits, vegetables, sugarcane, and forages). Finally, as we look at yield progress, it needs to be realized that CO2 increase itself (current rate of change about 2 μmol mol–1 yr–1) will be currently adding about 0.3% annually to the yields of C3 crops such as wheat and rice (Tubiello et al., 2007), assum-ing a relative responsiveness of yield to CO2 change of 0.7, a number likely to decline with further CO2 increase. The case studies presented here are, with the exception of maize in sub-Saharan Africa, favorable situations that have been the target of substantial agricultural research and develop-ment, in particular crop breeding, and should therefore illustrate what is achievable with such investments and what may lie ahead. The case studies are also chosen to represent major world agro-ecologies for the crops concerned.

YIELD DEFINITIONSAlong with farm yield (FY) (Fig. 1), attainable yield (AY) and potential yield (PY) are useful concepts (Loomis and Connor, 1992; van Ittersum and Rabbinge, 1997). Attain-able yield has come to mean the yield a skillful farmer should reach when taking prudent account of economics and risk; it has complications because farmers vary, as do farm gate economics. The difference between FY and AY has been defined as the exploitable yield gap.

Our definition of PY builds on that of Evans (1993) and refers to the yield of an adapted cultivar when grown with the best management and without natural hazards such as hail, frost, or lodging, and without water, nutrient, or biotic stress limitations (water stress being eliminated by full irrigation or ample rainfall). “Yield” is used as a noun throughout, and for PY, it is the yield of the best cultivar available at a location, usually representative of a cropping region in terms of the natural resource base of the environment (photoperiod, solar radiation, temperature, vapor pressure regime, and soil type). The natural resource base cannot be readily changed by the manager, but the definition becomes blurred when includ-ing activities such as land leveling, tile drainage, or liming, which are long-term management investments that improve the natural resource base. PY is usually determined from carefully managed field experiments with the best cultivars, which in turn can be used to calibrate crop simulation mod-els for PY prediction across time, space, and management options. Modeled PY must be based on the best available cultivar(s), meaning models should be regularly updated to match breeding progress in PY.

Usually PY progress is measured in side-by-side com-parisons of historic sets of cultivars executed with modern agronomy and with protection from biotic stresses. As such, this measure includes that component of the progress that derives from positive cultivar × agronomy interactions. In fact this interplay of genetic improvement and agronomic improvement has been an important component of crop productivity progress and it cannot be attributed to breed-ing or agronomy alone (Evans, 1993; Fischer, 2009). Effec-tive protection from biotic stresses is critical as breakdown in resistance to biotropic pathogens in older varieties can other-wise lead to overestimated rates of PY progress. The abiotic environment may also change over the time span of variet-ies compared (e.g., climate or soil change, ozone increase, or CO2 increase) in which case apparent breeding progress in gain yield could include adaptation to these changes. For rain-fed cropping it is also useful to define a water-limited potential yield (PYW), where W refers to an amount of crop evapotranspiration that is notably less than potential evapo-transpiration in the environment in which the crop is grown.

Because of constrained and variable farmer circum-stances there is usually an FY to AY gap and, because of economics, always an AY to PY gap; here we express these gaps as a percent of FY, thus providing a number more rel-evant to any discussion of increasing supply to meet a grow-ing global demand. It should be noted that plant breeding increases PY (and PYW), but breeding can also help to close the yield gap between PY and FY.

WHEAT YIELDUnited KingdomThe United Kingdom represents well-watered winter wheat around the world and is one of the more favorable

Fig. 1. World yields for wheat, rice, and maize vs. time, 1988 to 2007. Source: FAOSTAT. 2009. (Available at http://faostat.fao.org/site/339/default.aspx [verified 23 Dec. 2009]).


points strongly to greater Pmax (light-saturated photosyn-thetic rate) in the more recent cultivars. United Kingdom scientists are confident of further genetic progress in PY, predicting it will reach 17.4 t ha–1 by 2050, although FY, constrained by water and economic considerations, is pre-dicted to average 13 t ha–1 (Sylvester-Bradley et al., 2005; R. Sylvester-Bradley personal communication, 2009). This prediction relies on a reasonable assumption of RUE at 1.4 g dry matter (DM) MJ–1 intercepted (total) solar radia-tion. It also includes novel but physiologically sound calcu-lations of the dry matter that must be invested in stems to minimize lodging risk (Berry et al., 2007), thereby limit-ing harvest index (HI) to a value somewhat below the oft-quoted maximum estimate of 0.62 (Austin, 1982). It should be noted, however, that these predicted yields imply linear progress in PY and FY of 167 and 119 kg ha–1 yr–1, respec-tively—around 2.5 times the current rates.

Yaqui Valley, MexicoThe Yaqui Valley irrigation area in northwest Mexico (27o N lat) normally grows between 150,000 and 180,000 ha of irrigated spring wheat in each winter season (Novem-ber–April) and is well representative of irrigated wheat in the developing world. Good records go back to the 1950s, when the wheat crop became the breeding target of the Interna-tional Maize and Wheat Improvement Center’s (CIMMYT) Wheat Program and its predecessor organization. Average farm yields increased dramatically in the first 30 yr, from 1.4 t ha–1 in 1950 to almost 5 t ha–1 in 1980 (>4% growth per annum; Fig. 3); semidwarf cultivars first appeared in 1962 and in less than 5 yr occupied the whole area, while N fertil-izer application rose from zero to around 175 kg N ha–1 in 1980 (Bell et al., 1995). Fischer (2008) argued that although the rate of progress has clearly slowed with time in the period

wheat growing environments: it is dominated by winter wheat planted in September to October and harvested in August the following year, thus occupying the land for 10 to 11 mo. The United Kingdom delivers one of the high-est national farm wheat yields, averaging over 8 t ha–1 (15% moisture) across 2 million ha (Fig. 2). Given the competi-tive private breeding sector, the substantial private and pub-lic sector crop management research, sophisticated farmers, and production that is not nowadays distorted greatly by price subsidies, this level of FY is probably close to AY. In the United Kingdom, fertilizer rates are high (the Brit-ish Survey of Fertilizer Practice lists this as 190 kg N, 31 kg P2O5, and 39 kg K2O per hectare in 2007), though the rate has been steady for the last 25 yr. Since 1989 yield has increased at a rate of 53 kg ha–1 yr–1 or 0.7% of the cur-rent yield level (Fig. 2). An excellent and extensive system of national trials is conducted by the Home Grown Cereals Authority (HGCA): with complete disease and pest con-trol, these trials, which are dominated by the latest culti-vars, averaged 10.4 t ha–1 in 2004 to 2008. These HGCA protected trial yields can be considered a good measure of the current PY, since yield is little affected by water deficits (Sylvester-Bradley et al., 2005). The yield gap between FY and PY is currently 30% of average FY. This leaves little scope for gap closing through breeding, although greater host plant resistance could substitute for the biocides cur-rently used to keep this gap small.

The system of HGCA trials also permits an accurate estimate of the relative rate of PY progress from breeding. Yields of cultivars released over the last 20 yr have increased linearly with time at a rate of 61 kg ha–1 yr–1 (or 0.6% of 2008 PY) with no sign of slowing. Although PY is plotted against the year of cultivar release (Fig. 2), it is reasonable to assume that the best cultivars moved quickly to occupy substantial portions of the area planted once released. Thus the relative rate of FY progress in the United Kingdom is similar to that from breeding over the last 20 yr (Fig. 2), and observers suggest that recent FY progress is dominated by genetic improvement (e.g., British Society of Plant Breeders, 2008); recent analysis of the HGCA wheat yields over the last 50 yr strongly supports the view that over 90% of progress since the 1982 is derived from breeding alone (i.e., independent of variety by agronomy interaction [I. Mackay, personal communication, 2009]).

A recent study of progress in U.K. winter wheats (1972–1995) under very well-managed conditions at Not-tingham (with yields to 11.4 t ha–1) showed, as have most wheat studies, that yield progress was associated closely with an increased number of grains per square meter. The increase is also associated with greater crop growth rate and radiation use efficiency (RUE) in the period leading up to flowering and with increased water soluble carbohydrates in the stem at flowering (Shearman et al., 2005). Although there are other less likely explanations, the increase in RUE

Fig. 2. Farm yield vs. time (lower line) and potential wheat yield (upper line) vs. year of release in the United Kingdom; yields at 15% moisture. Potential yields were obtained under fungicide protection and are averaged over the period 2004 to 2008. Sources: FAOSTAT. 2009.(Available at http://faostat.fao.org/site/339/default.aspx [verified 23 Dec. 2009]); HGCA Recommended Lists 2004–08.


1950 to the present, with significant curvilinearity in yield vs. time, current yield progress is best represented by the linear slope of yield vs. time over the last 30 yr (Fig. 3; 49 kg ha–1 yr–1 or about 0.8% per annum of the current aver-age yield of 6 t ha–1). However, it has been pointed out that secular weather changes in the Yaqui Valley, namely warm-ing (Bell and Fischer, 1994) and later cooling (Lobell et al., 2005), can confound the interpretation of yield change over time. Allowing for the recent steady decline in minimum daily temperature (Tmin, 0.07°C per year over 1979–2008, R2 = 0.38) and an apparent FY responsiveness to Tmin, calcu-lated using the first difference method to be 377 kg ha–1 °C–1 (R2 = 0.43), suggests therefore that technological progress was less than the FY numbers indicate in Fig. 3 but still posi-tive at 23 kg ha–1 yr–1 or 0.4% per annum.

Potential yield progress in the Yaqui Valley has been measured many times in sets of historic and new cultivars grown side by side at the centrally-located Centro de Inves-tigaciones Agricolas del Noroeste (CIANO) research station, using optimal agronomy and frequent fungicide applications for complete disease control; yield is regressed against year of release in Fig. 3, and rapid adoption of the best cultivars can be assumed. Clearly an important factor driving up FY since 1950 has been the initial rapid increase in PY (>1% per annum) of the cultivars released and quickly adopted (Fischer and Wall, 1976; Bell et al., 1995). However, PY progress has slowed in the last 30 yr to around 23 kg ha–1 yr–1 or 0.3% per annum (Fig. 3). Thus it would seem that the relative rates of progress in FY and PY are about the same. Currently FY and PY are at 6 and 9 t ha–1, respectively, and the gap between them is fairly steady at about 50% of FY. Much has been

learned about variation between yields in farmers’ fields using ground surveys and, more recently, satellite imagery. About one third of this yield gap is readily exploitable with bet-ter agronomy (earlier sowing and improved water and weed management) (Lobell and Ortiz-Monasterio, 2008; Lobell et al., 2005; Ortiz-Monasterio and Lobell, 2007).

Of great relevance here is the physiology of the more recent PY increase shown in Fig. 3. Using six key semi-dwarf bread wheat cultivars spanning 1962 to 1988 and grown with full foliar disease protection, Sayre et al. (1997) showed that PY progress of 0.8% per annum over the period was associated with several traits (Table 1), includ-ing increased grains per square meter (r2 = 0.71, p < 0.01) and increased HI (r2 = 0.66, p < 0.05) as seen in earlier studies of tall and short wheats (e.g., Fischer and Wall, 1976). Physiological measurements over 3 yr (1993–1995) revealed that the yield progress was also associated with sto-matal conductance (r2 = 0.88, p < 0.01), which increased by 63% over the release period, and with Pmax (r

2 = 0.72, p < 0.01), which increased 23% over the same time interval (Fischer et al., 1998); these correlations were found both before anthesis as well as after. A parallel unpublished study of seven semidwarf durum cultivars released between 1967 and 1989 found similar associations among yield, grains per square meter, and physiological activities (Table 1). In both studies, there was no correlation between grain yield and plant height or days to flower. Attempts to relate grain number and yield to crop growth rate and RUE in the three week period just before flowering (as in Shearman et al., 2005) found significant correlations in only one out of the 3 yr (Fischer et al., 1998), though RUE estimates had large errors.

The above results strongly suggest that progress for PY is related to increased stomatal conductance and provide some support for an accompanying increase in Pmax; some earlier spring wheat studies showed clearly increased Pmax (e.g., Shimshi and Ephrat, 1975; Watanabe et al., 1994). It is interesting that in both bread wheat and durum wheat the morphological changes in the flag leaf seen with this breed-ing progress at CIMMYT (e.g., smaller, more erect, higher N per unit area, and higher chlorophyll concentration) were not unlike those reported for winter wheat progress in the United Kingdom (Shearman et al., 2005). CIMMYT has reported some success in the use of stomatal conductance as an early generation selection criterion for PY (Condon et al., 2005). Although this trait can be remotely sensed quickly and cheaply, it appears that there has been insuf-ficient support to properly validate this selection strategy.

RICE YIELDJapanJapan has a long and rich history of research in rice breed-ing, agronomy, and physiology, and well represents higher latitude japonica rice grown under the summer monsoon.

Fig. 3. Yield of wheat vs. time in the Yaqui Valley, Mexico, and potential yield (upper line) of bread and durum wheat varieties versus year of release; yields at 12% moisture. Potential yields were determined in experiments over the period 1990 to 2005 with fungicide protection and standardized against a common check, ‘Siete Cerros 66’, with an average protected potential yield of 7.0 t ha-1. Regressions refer to the last 30 yr data only. Source: K.D. Sayre, personal communication, 2009.


Japan had its “green revolution” in rice in the 1950s to 1970s, resulting in rapid yield growth from new cultivars, fertilizer, and improved crop management through mech-anization (Horie et al., 2005). National yield increased by about 50% in this period and reached about 6 t ha–1 in 1975 (all rice yields are for rough or paddy rice, although the Japanese literature often quotes brown rice yields). It should be noted that winning yields in farmer contests peaked at over 12 t ha–1 in the 1950s and 1960s, more than double the national yield, highlighting what can be achieved with certain specific management techniques and presumably fortuitous weather conditions (Horie et al., 2005). Since then yield has grown only slowly (Fig. 4), now approaching 6.5 t ha–1, and rice area has actually declined around 1.5% annually to reach 1.7 million ha today. Area decline reflects less favorable policies for rice and may be part of the reason for the slow yield growth of only 0.4% per annum relative to present FY (Fig. 4). Another reason for slow growth in yields is increased atten-tion paid to better rice quality, which requires reduced N fertilization levels and absorbs much of the breeding effort (T. Horie, personal communication, 2009).

An increase in rice PY in Japan appeared to occur in the early 1990s. The cultivar ‘Takanari’, released in 1990, has been widely assessed in central and southern Honsu, giving an average yield of 10.5 t ha–1, 36% more than ‘Nip-ponbare’, a landmark cultivar released in 1963 (San-oh et al., 2004; Horie et al., 2005; Takai et al., 2006; Katsura et al., 2007; Taylaran et al., 2009; H. Yoshida, personal communication, 2009); the yield advance was seen even when crop protection was clearly specified (e.g., Takai et al., 2006; Katsura et al., 2007). Both varieties are conven-tional cultivars (inbreds), yet Takanari yielded as well as one of the best current Chinese hybrids (Katsura et al., 2007). Needless to say, Takanari is not a good quality food rice, and more recently breeders have produced other very high yielding feed-type rices such as ‘Hokuriku 193’ in 2007 (Goto et al., 2009) and ‘Momiroman’ in 2008 (Yoshinaga et al., 2009), as well as a reasonable food quality variety, ‘Akita 63’, which has recorded an average yield of 11.1 t ha–1 over 3 yr (Mae et al., 2006). Between Nippon-bare and Takanare, PY increased at around 100 kg ha–1 yr–1 as shown in Fig. 4; this is close to 1.0%. It is not possible, however, to get an accurate measure of the more recent PY progress, but it seems reasonable that progress has con-tinued and that the current potential is at least 11 t ha–1, giving a yield gap of 70%. We suggest that this large gap and the apparent stagnation of Japanese yields is the con-sequence of the overriding emphasis on producing excel-lent food quality rice for the limited home market. All the new high-yielding varieties respond to higher N levels, but they have high N utilization efficiency (e.g., Katsura et al., 2007), and new agronomic techniques can lift N recov-ery efficiency substantially (Horie et al., 2005). It is not

known whether even higher yields could be obtained with contest-winning management, since contests have been discontinued, but there is some evidence to support this (e.g., San-oh et al., 2004).

The physiology of the high PY cultivar, Takanari, rela-tive to its lower yielding predecessors has been studied in detail at Kyoto. It had the highest filled spikelet number per square meter and the highest crop growth rate during the late reproductive period (just before heading). These were associated with a higher RUE (2.11 g MJ–1 total intercepted solar radiation) and higher nonstructural carbohydrate con-tent at heading (Takai et al., 2006). The parallel between this result and that of Shearman et al. (2005) with winter wheats in the United Kingdom is clear. Katsura et al. (2007) continued studies in Kyoto and confirmed these results when

Table 1. Correlation between physiological traits and grain yield in CIMMYT bread wheat and durum wheat semi-dwarf cultivars. Source: bread wheat: Fischer et al., 1998; durum wheat: R.A. Fischer, unpublished data, 1997.

Bread wheat Durum wheatNumber of varieties 8 7

Release period 1962–88 1967–1989

Independent variable Association with grain yield (r)

Year of release 0.97** 0.66*

Total dry weight 0.07 0.79**

Harvest index 0.66* 0.88**

Grains per square meter 0.71** 0.98**

Kernel weight –0.01 –0.4

Days to anthesis 0.03 –0.23

Stomatal conductance 0.88** 0.79**

Pmax 0.72** 0.52

dC13 0.50* 0.61*

Canopy temperature depression 0.58* na

* Significant at the 0.05 probability level.

** Significant at the 0.01 probability level.

Fig. 4. Japan’s rice yield vs. time and potential yield for varieties Nipponbare and Takanari; yields for rough rice at 14% moisture. Sources: FAOSTAT. 2009. (Available at http://faostat.fao.org/site/339/default.aspx [verified 23 Dec. 2009]); for potential yield see text.


comparing Takanari with Nipponbare, again recording a very high RUE in the preheading stage with Takanari (1.96 g MJ–1); they also reported a tendency for Takanari to take up more soil nitrogen, although specific leaf nitrogen (SLN) was not superior. The same authors confirmed that the increased crop growth rate of Takanari was due to greater daytime canopy photosynthesis and not due to any genetic differ-ences in respiratory parameters (Katsura et al., 2009). At the same time Ohsumi et al. (2007) confirmed a higher Pmax in Takanari compared with other high yielding cultivars and showed it to be associated with higher stomatal conductance rather than higher SLN, although ontogenetic variation in Pmax was related to SLN. Ohsumi and colleagues concluded that simultaneous improvement in stomatal conductance and in SLN are essential for breeding cultivars with higher Pmax (and yield). Physiological data on the newer high-yielding varieties does not appear to be available.

Central Luzon, the PhilippinesCentral Luzon is a major rice growing plain in the Phil-ippines, with the International Rice Research Institute (IRRI) located just to the south and the Philippine Rice Research Institute (PhilRice), the national research center, located in its northeast corner. This region has been the target of intensive rice breeding efforts for at least 50 yr, grows about 1.2 million ha of rice, and is representative of tropical irrigated indica rice growing environments. About two thirds of this rice area is planted to the wet season (WS) crop and the balance to dry season (DS) rice under irriga-tion. Farm yields have been surveyed from time to time by IRRI since 1966. At that time only traditional culti-vars were grown, fertilizer use was negligible, and cropping intensity only 110% for an average yield of 2.5 t ha–1. Farm yield for survey farms over the last 30 yr period when only modern cultivars were grown is shown in Fig. 5 (Estudi-llo and Otsuka, 2001; P. Moya, personal communication,

2009). Farm yield progress is slow but steady at 0.6% per annum from both wet- and dry-season rice. The surveys showed fertilizer use rising from around 70 kg ha–1 (ele-mental N+P+K) in the early 1980s to 150 kg ha–1 now, while rice cropping intensity has remained steady at 150%.

Current estimates of PY of 6 and 9 t ha–1 for WS and DS, respectively, are derived from yields of inbred varieties in current very well managed irrigated trials at IRRI in the absence of significant diseases, pests, and weeds (e.g., Yang et al., 2007); Peng et al. (1999) declared a DS PY of 10 t ha–1 but yields appear to have declined since then (see Table 2). Despite the rice climate in Central Luzon being slightly more favorable than at IRRI (S. Peng, personal communica-tion, 2009), the WS yield gap is therefore 60% of FY, and the gap is even larger in the DS (100%). These gaps do not reflect significant current use of older disease-susceptible cultivars, because cultivar turnover is rapid in the region. More recent wet-season surveys closer to IRRI (Laguna Province) sug-gest that the average yield of the top one third of farmers is approaching 6 t ha–1. However, the average of all farmers is considerably lower at 4.4 t ha–1 in Laguna, with a 0.4% annual increase over the last 30 yr, and 4.0 t ha–1 with no trend in Nueva Ejica Province (close to PhilRice). These data suggest yield gaps of 36% and 50%, respectively, or somewhat less than the 60% observed by the Central Luzon WS survey.

Changes in PY are difficult to assess in the current trials at IRRI, because older cultivars such as ‘IR8’, the first tropi-cal semidwarf cultivar released in 1966, cannot now be easily protected from pests and diseases or may not be well adapted to recent unfavorable changes in the natural resource base, including the climate (Peng et al., 1999, 2000). The latter possibility is suggested by the fact that even when, with pro-tection and in the DS there is no obvious disease or pest dam-age, IR8 yields no more than 8 t ha–1 or 1 to 2 t ha–1 below its yields in the late 1960s (S. Peng, personal communica-tion, 2009). Progress in the PY of inbreds since the release of ‘IR72’ in 1988 seems also to have been very slow (see Table 2). Therefore, in accord with Peng et al. (2000), Fig. 5 shows no progress in PY, even if maintaining PY in the face of a deteriorating physical environment would clearly constitute genetic progress. It should be noted that breeders have made good progress on other fronts such as disease and insect resis-tance, earliness, and grain quality.

Since the early 1990s IRRI breeders have made a con-certed effort to boost rice PY by design, breeding first for the new plant type (NPT) ideotype and more recently for a second generation of NPT products (NTP2). NPT2 cul-tivars perform better than the original NPTs, but barely better than the best cultivars of the same vintage coming from the conventional inbred breeding program (Yang et al., 2007; Table 2). This is a disappointing outcome for physiological plant breeding, especially since the develop-ment of IR8 itself came from ideotype breeding (Jennings, 1964). However, NPT thinking has spilled over into China

Fig. 5. Farm rice yield vs. time of harvest in Central Luzon in the wet and dry seasons, and potential yield at IRRI; yields for rough rice at 14% moisture. Source: see text.


where it appears to be an important factor in the design of the very high yielding hybrids now in current use (Peng et al., 2008). Yields in the irrigated eastern China lowlands, under a climate similar to the warm cloudy Philippines WS, are now surpassing 12 t ha–1. Some of this technology has flowed back to IRRI from China, and the best Philip-pine F1 tropical hybrids are now yielding 11 to 14% more than the best inbreds in the DS at IRRI, though they show little advantage in the WS (Yang et al., 2007; Table 2). A more recent study of hybrids and inbreds carefully matched for phenology found 18% (DS) and 14% (WS) yield advan-tages for the hybrids (Bueno and Lafarge, 2009).

Retrospective physiology on recent PY progress in trop-ical rice is also hampered by concerns over the validity of side-by-side comparisons of historic sets of cultivars. There is some evidence that increased HI was important from 1966 to 1980 as cultivars became shorter and earlier, but since 1980 total biomass and a slightly longer duration seem to be asso-ciated with the highest yielding inbreds (Peng et al., 2000). Changes in filled spikelets per square meter and seed weight were not consistent, and the decline in yield of IR8 seems to be related to falling HI and lower filled spikelet percentage (Peng et al., 1999). Yang et al. (2007) recently compared the best of the inbreds derived conventionally and from NPT2s with IRRI hybrids (Table 2). Hybrids were superior in the DS by virtue of both greater biomass and greater HI, more spikelets per square meter, and heavier seeds but have a lower proportion of filled spikelets. Bueno and Lafarge (2009) also found higher biomass and HI with the hybrids. Little data on leaf-level physiology has been reported lately, although the earlier study of conventional inbreds found no difference among cultivars released between 1966 and 1995 for crop growth rates in the critical panicle initiation-to-flowering stage, while Peng et al. (1999) reported lower Pmax in a hybrid compared to IR72. Later Peng et al. (2008) note that the best hybrid “super” rices in China appear to have higher bio-mass and Pmax around heading, higher specific leaf weight, and higher leaf chlorophyll than older hybrid check culti-vars, which they outyielded by 10 to 20%. In addition, the final three leaves of super rice hybrids are very erect, like the NPT rices, but, unlike modern wheats, the leaves are long and reach well above the panicle. Zhang et al. (2009) recently confirmed the superior yield of the hybrid “super” rices in eastern China but could not explain this via increased RUE.

MAIZE YIELDSIowa State, United StatesThe state of Iowa grows 5.3 million ha of maize under very favorable rain-fed conditions in the heart of the U.S. Corn Belt. This region represents the majority of tem-perate maize environments globally. The Corn Belt, and Iowa in particular, is a major battleground for the large maize breeding and biotechnology transnational compa-nies, Monsanto, Pioneer-DuPont, and Syngenta, who have

invested heavily in maize improvement, especially over the past 20 yr. Global expenditures on maize improvement, widespread testing, and refining farm-level agronomy for the new hybrids by these three companies alone are cur-rently estimated to exceed $3 million per day. Maize yield growth in the United States has been strong and steady, while in Iowa it is nothing short of spectacular (Fig. 6). The Iowa average yield is currently 10.5 t ha–1, with a linear slope of 214 kg ha–1 yr–1 or 2.0% per annum of the cur-rent yield; progress has also accelerated since 1990. What is behind this impressive growth, and how much is due to breeding?

The changes from 1961 through 1990 were largely due to increases in fertilizer use, chemical weed control, and higher plant densities, coupled with the use of hybrids that could respond to fertilizers and tolerate crowding (Cardwell, 1982). Nitrogen use, a key factor driving yield increases in the 1960s, has stabilized since the late 1970s at around 140 to 160 kg N ha–1, meaning fertilizer use effi-ciency has increased notably since then (Fig. 7). Irrigated land in Iowa has only increased from 0.5% of the area in 1997 to 0.7% in 2007 and has not been a factor in pro-ductivity increases. Precision farming, where input levels are varied in response to within-field fertility variation, may have led to small improvements in yield (Cassman, 1999), and more uniform spacing between plants because of improved machinery has also contributed to a minor rise in yields. One key factor recently contributing to the rapid increase in yields in Iowa has been earlier planting. Maize crops in Iowa today are planted on average 12 d ear-lier than in 1979, thus allowing the crop to capture more radiation and to fill grain under a more favorable tempera-ture regime. Kucharik (2008) estimates that early planting may account for half of the annual increase in grain yields in Iowa in the last 30 yr; this has been facilitated by zero tillage, improved seed fungicide dressing, hybrids with herbicide resistance and greater early cold tolerance, and

Table 2. Grain yield (14% moisture), biomass, harvest index, and duration of recent inbred, hybrid, and second genera-tion new plant type (NPT2) rice cultivars, compared to IR72 released in 1988. Trials were grown at IRRI and averaged for 2003 and 2004 (from Yang et al., 2007).

IR72Inbreds Hybrids NPT2(n = 3,4) (n = 5) (n = 5)

Dry season

Grain yield† 8.29b 8.73b 9.55a 8.38b

Biomass, g m–2 1726 1722 1792 1691

Harvest index 0.45 0.46 0.49 0.44

Growth duration, d 117 117 116 118

Wet season

Grain yield† 5.60ab 6.00a 5.98a 5.40b

Biomass, g m–2 1339 1413 1317 1409

Harvest index 0.36 0.38 0.41 0.37

Growth duration, d 113 122 113 121† Yields followed by different letters significantly different at P < 0.05.


the use of large high-speed planters. A second significant factor is the continuing rise in plant densities in the Corn Belt at a rate of about 1000 plants ha–1 yr–1 over the last 25 yr, and this trend to higher densities continues despite the rising price of seed. Improved plant-to-plant uniformity arising from the greater use of precision planters has also

resulted in a small increase in yield (Liu et al., 2004), and it is likely that reduced plant-to-plant variation observed in newer vs. older hybrids has had a similar effect (Edmeades et al., 2006). Finally, a recent factor is the increasing use of foliar fungicide in disease-prone situations (maize after maize, conservation tillage, and/or high temperature and humidity), even when disease is not very evident.

Older research on the sources of productivity gains in the Corn Belt have suggested that about 40 to 50% of increased yields can be attributed to genetic improvement (Cardwell, 1982; Duvick, 2005). In recent years, however, as mean yield levels rise, the easy gains from better management practices alone such as applied N and improved weed control often are fully exploited, and the proportion of gain due to improved genetics that address genotype × management practice and yield potential rises (Edmeades and Tollenaar, 1990). Thus the full benefits of changed growing practices are only seen when complemented with cultivars developed to exploit those practices—the ubiquitous positive genotype by man-agement interaction—and that complementation becomes increasingly important as yield levels rise. An important example is the optimum plant density of maize hybrids devel-oped over the past 70 yr. This has risen substantially, exploit-ing the positive interaction between improved hybrids and plant density; it now averages around 80,000 plants ha–1 in Iowa. A second example relates to date of planting. Genetic improvements in tolerance to cold and waterlogged soils have played an important part in allowing the earlier planting of maize and the expansion of conservation tillage practices that themselves favor earlier planting. The growing importance of genetic improvement makes an interesting parallel with wheat in the United Kingdom, mentioned earlier, but the continuing contribution of genotype by agronomic manage-ment interaction appears to be greater with maize in Iowa.

It has not been easy to get a measure of maize PY and its progress in Iowa. First, hybrids must be tested at their optimum density. Studies of Pioneer hybrids released in each decade from 1930 to 2002 at their best density showed remarkably linear yield growth of 79 kg ha–1 yr–1, or 1.5% of mean yields of 1930 hybrids, but only 0.8% of 2002 hybrid yields, which were 10.2 t ha–1 (Cooper et al., 2004). This rate of yield growth is about 66% of the increase in Iowa yields over the same period. A later look at this Pioneer hybrid set with releases to 2007 showed yield had almost reached 12 t ha–1 and was increasing at 116 kg ha–1 yr–1 or 1.0% per annum (Hammer et al., 2009; see Fig. 8). The yield of the best three hybrids for each maturity class in each of the five districts in Iowa Crop Improvement Association trials in 2007 and 2008 averaged 11.8 t ha–1 and could be consid-ered another valid estimate of PY. However it is not possible to measure genetic progress with these trials, and all of the above PY estimates seem low for a region where FY averages 10.5 t ha–1. Thus it is important to note that the same Pioneer hybrids as used by Cooper et al. (2004), when grown under

Fig. 6. Maize grain yield vs. year of harvest for (a) the United States and (b) the state of Iowa; yields at 15.5% moisture. Linear rates of yield increase are shown separately for 1961 to 2008 period. Split line regression resulted in a significantly improved fit for the Iowa yield data. Source: http://www.nass.usda.gov/Data_and_Statistics/Quick_Stats/ (verified 1 Jan. 2010).

Fig. 7. Nitrogen applied to maize in Iowa (upper curve) and the ratio of grain yield to N applied (lower curve), 1964 to 2005. Linear rates of increase for each variable are shown for 1990 to 2005 periods. Source: http://www.ers.usda.gov/Data/FertilizerUse/ (verified 23 Dec. 2009).


irrigation in a very favorable Chilean environment, showed PY progress at around 200 kg ha–1 yr–1 or 1%, with the high-est yields close to 20 t ha–1 (Campos et al., 2004).

Near Iowa, in Nebraska, higher yields have been reported in experiments to calibrate a maize simulation model (Hybrid-Maize [University of Nebraska, Lincoln, NE; Yang et al., 2004]), where, with irrigation, around 100,000 plants ha–1 and 225 to 298 kg N ha–1, yields of the latest hybrids in 1999 to 2002 averaged about 16 t ha–1 (Yang et al., 2004). Annual crop contest winners report even higher yields: for rain-fed maize in Iowa, contest-winning yields are approach-ing 17 t ha–1 (R.W. Elmore, personal communication, 2009), and for irrigated maize in Nebraska, contest-winning yields have averaged 18.8 t ha–1 over the period 1984 to 2002 with little apparent increase over time (Cassman et al., 2003). Contest yields for rain-fed maize in Nebraska over the same period have risen steadily, reaching 15 t ha–1 in 2002 (Cass-man et al., 2003). Competition-winning yields arise from excellent management and very favorable and often unpre-dictable genotype × environment interactions, so as an esti-mate of overall PY for the region they must be treated with caution. Thus, as with contest-winning rice yields in Japan in the 1960s, we are forced to regard maize contest yields as not being representative of PY for the whole state.

It would be good to have data from other breeding sources, but on the basis of the published Pioneer data, we conclude that the breeders are still increasing maize PY in Iowa but at only 1% per annum. Regarding the current PY for rain-fed Iowa we conclude that the best estimate is the average yield simulated by Hybrid-Maize above using 20 yr weather from Ames, Iowa; that was 15.5 t ha–1 (Gras-sini et al., 2009); it may be less than that of irrigated corn in Nebraska, because of subtle water stress (see also Ham-mer et al. [2009], below). It can now be calculated that the yield gap in Iowa is moderate (around 50%) and closing quickly as suggested in Fig. 8—not surprising given mod-ern farmers, attractive prices, and a substantial investment in extension by private seed companies. Also, because of the uncertainty about contest-winning yields, we cannot agree with Cassman et al. (2003) that progress in maize FY in the Corn Belt may be approaching a limit, as reflected in Nebraska contest-winning irrigated PY; we do believe contest-winning crops are worthy of further careful study. Also, we agree that the rate of return to research on increas-ing maize yields in the Corn Belt appears to be declining, since research investments increased substantially in real terms during the 1970s through 1994 (Duvick and Cass-man, 1999) while national yield increases have remained essentially linear with time through 2008 (Fig. 6(a)).

Significant changes in many traits have occurred as yields have increased in temperate maize, pointing to the physiological bases of improved PY. The late Don Duvick led the systematic evaluation of these changes, summarized in Duvick (2005) (Table 3). Most of the trait changes apart from

yield, lodging resistance, speed of dry down, and disease and insect resistance were not subject to direct selection pressure. The changes, however, can be better understood when solar radiation capture and competition for light are considered. In the 1930s hybrids were strongly single eared, tended to tiller, and had large branching tassels. They were well adapted to the low density and wide rows required for interrow weed control using animals, and light interception was relatively low. As a consequence they were sensitive to light competi-tion, which caused individual plants to go barren. Breed-ing has reduced this sensitivity to such an extent that much higher densities are now tolerated and little light is wasted. A marked increase in leaf erectness and a reduction in tassel size is parallel to the changes due to selection in wheat and rice.

Tollenaar and coworkers, evaluating a smaller set of early-maturing temperate hybrids in Ontario, Canada, have also identified a number of underlying changes in plant archi-tecture and function that have accompanied selection. Their studies have confirmed most of the changes shown in Table 3 (Tollenaar and Lee, 2006) but showed that selection has resulted in greater crop growth rate, especially during grain filling, arising from somewhat greater leaf area index (LAI), prolonged staygreen, and more erect leaves. Pmax did not appear to change, although its rate of decline during grain filling was slower with newer hybrids (better functional stay-green) and it suffered less depression after cool nights in mod-ern hybrids. Increased grain number of modern hybrids was related to greater dry matter accumulation around silking, to increased dry matter partitioning to the ear, and to greater ear fitness resulting in more kernels per unit ear dry weight (Tollenaar and Lee, 2006; Echarte and Tollenaar, 2006). More kernels, amounting to a greater grain-filling sink, may also be increasing photosynthesis and dry weight accumula-tion during this period through feedback mechanisms.

Fig. 8. Iowa maize farm yield vs. time and potential yield (at optimum density) vs. time of hybrid release; yields at 15.5% moisture. Sources: http://www.nass.usda.gov/Data_and_Statistics/Quick_Stats/ (verified 1 Jan. 2010); Hammer et al. (2009).


Breeders’ trials conducted in Argentina show similar PY progress as in Iowa (I. Colonna, personal communi-cation, 2009). For example, with a set of maize hybrids released between 1965 and 1997 there was an overall yield increase of over 60%, closely associated with increased grain number, and irrigated yield levels with the best hybrid of 17 t ha–1 (Luque et al., 2006). Results from extensive research on the physiology of yield progress also showed that selection resulted in increases in crop growth rate and dry matter allocation to the ear during the criti-cal period 1 wk before silking to 20 d after. It was during this period that the difference in favor of the more modern hybrids began to appear, and it was from this period that differences in kernel number and ultimately the progress in yield were derived. In addition RUE during this criti-cal period showed a significant increase in response to year of hybrid release, while canopy extinction coefficient showed no clear tendencies (Luque et al., 2006). This is highly suggestive of an increase in Pmax and establishes a clear parallel with PY progress in wheat and rice.

There is ample evidence to show good progress in maize yield under conditions of less water supply, as in dry years in Iowa (Duvick and Cassman, 1999) or under restricted irriga-tion where sometimes, especially with stress at flowering, the relative progress was actually greater under the less favorable conditions (Campos et al., 2004). Several key traits indicate increased water stress tolerance in modern hybrids, such as a reduced anthesis-silking interval and better tolerance of oxidative stress, as pointed out by Tollenaar and Lee (2006). These authors suggested that the modern hybrids are not only more resistant to high density and drought stresses but also to multiple minor stresses such as short dry spells, hot or dry windy days, cool nights, sudden radiation changes with cloud passage, oxidative herbicides, and brief waterlogging.

Hammer et al. (2009) have recently proposed that mod-ern hybrids acquire more soil water through deeper roots, which could explain part of their improved density and drought tolerance. They argue also that progress is driven also by greater transpiration at a constant transpiration efficiency (TE), amounting to a TE of 44 kg DM ha–1 mm–1 (4.4 mg g–1) at a mean vapor pressure deficit of 2 kPa. At prevailing HI values, this equates to 45 mm of transpiration per ton of grain or more than 600 mm crop evapotranspiration for a 12 t ha–1 yield. A fixed TE could be seen to contrast with the notion that increased biomass is related to greater RUE and photosynthetic rate. Such changes, however, can occur with-out a change in TE, depending on changes in stomatal vs. internal leaf conductance and on the extent to which canopy gas exchange is coupled to the atmosphere. Cassman’s group in Nebraska recently presented simulations and measurements that supported a higher TE and hence a steeper response of yield to water use, amounting to only 27 mm evapotranspira-tion per ton of grain (Grassini et al., 2009). Currently there is not enough information to resolve the important issues of change in rooting depth, TE, and the extent to which water supply may ultimately limit rain-fed maize yield in Iowa.

We cannot leave the Corn Belt without comment on the likely impact on yield of widespread adoption of GM (geneti-cally modified) hybrids, commencing in 1996 and currently approaching 90% adoption in Iowa. Although the traits involved (herbicide resistance and insect resistance) do not increase PY per se, many observers point to yield increases at the farm level when GM hybrids are deployed. This is because weed control is generally better, timely planting is facilitated, and insect control is improved, particularly with respect to corn root worm. Damage from corn root worm was often not fully appreciated before transgenic sources of resistance became available, and the resulting complete root systems are

Table 3. Changes in yield and associated traits in Pioneer maize hybrids released between 1930 and 2002 (Duvick, 2005).

Trait Change in trait Comment Yield at optimum density Linear increase Around 80 kg ha–1 yr–1

Optimum density Increase to >80,000 plants ha–1 Started at 30,000 plants ha–1

Biomass Steady increase

Harvest index Slight increase, now 50–55% Older hybrids may go barren at high density

Kernel number per square meter Steady increase More plants ha–1, less barrenness

Kernel weight Small increase

Grain protein Consistent decline Consistent rise in starch percentage

Days to flowering Unchanged No change in leaf number

Grain fill duration Longer

Grain dry down Faster

Plant height No trend

Ear height Slight reduction

Root lodging Large reduction Not completely eliminated

Canopy Much more erect Maximum leaf area index (LAI) generally unchanged

Staygreen Markedly increased Little change under terminal drought

Leaf rolling More More obvious with upright leaves

Tassel size Much reduced

Anthesis-silking interval Reduced to near zero Indicator of stress tolerance at flowering


now seen to contribute to drought tolerance. Part of the rapid farm level yield progress (Fig. 6) could reflect this.

Sub-Saharan AfricaMaize in sub-Saharan Africa occupies lowland, mid-alti-tude, and highland tropical and sub-tropical zones with moderate to good rainfall; these are the key agro-ecologies of maize in most developing countries. Notwithstanding its critical importance for feeding poor people, we give lim-ited attention to the region because it is covered in another paper in this issue (Keating et al., 2010), and because it is clear that FY is constrained by many factors besides prog-ress in PY and PYW. These include low soil fertility, low fertilizer use, inadequate plant density, weeds, poor tillage, and labor shortages, as well as serious infrastructural and institutional constraints limiting the adoption of improved technologies for maize (Keating et al., 2010). Improved open-pollinated cultivated and hybrids show moderate PY and PYW gains (e.g., Bänziger et al. [2006] in southern Africa), but improved management, especially soil fertility, brings larger gains (Keating et al., 2010), and combining improved germplasm with improved agronomy delivers even greater gains. For illustration we have taken yields from on-farm trials with improved cultivars and best-bet management to show AYs across the region (Table 4); FY is much lower and is growing only slowly; there is no doubt that the PY–FY yield gap is at least 200%.

SYNTHESIS AND CONCLUSIONSThis paper has traced progress in breeding for PY in favor-able or irrigated environments that have attracted substan-tial research and development investment in breeding and agronomy (sub-Saharan Africa is the exception to this). We have shown that progress in FY closely tracks progress in PY. Potential yield and FY progress are summarized in Table 4, but it should be noted that PY progress was measured under modern agronomy and therefore reflects genetic progress per se plus that derived from any positive cultivar by management interactions.

None of the regions shown in Table 4 recorded gains in PY exceeding 1% per annum, although some uncertainty surrounds the estimate for PY and gains for rice in Japan and for maize PY in Iowa. Breeding progress in PYW in indus-trialized countries (e.g., wheat in Australia, Great Plains of the United States) seems to be steady at about 0.5% per annum (Fischer, 2009; Fischer et al., 2009). These relative rates augur poorly for the future, for FY tends to follow PY progress, and all the rates of progress are below those needed to prevent real price rises (Tweeten and Thompson, 2008). There is a ray of hope in the rapid maize FY progress in Iowa, but worrying also because it amounts to notable yield gap closing and may become exhausted. Higher yield gaps persisted in developing countries despite major research and development efforts (in the Yaqui Valley of Mexico and in

the Philippines). Even higher yield gaps are to found in most rain-fed situations with limited rainfall (data not shown) and with maize in sub-Saharan Africa (Fischer et al., 2009; Lobell et al., 2009). Clearly the cereal yield growth that the world needs must, to a significant extent, derive from a focus on closing these large yield gaps, though this chal-lenge has not been the major focus of this paper. Suffice to say that breeding can also help farmers close yield gaps by improving, for example, disease and pest resistance, nutri-ent extraction, tolerance to manageable soil toxicities, weed competitiveness, or adaptation to conservation tillage.

The physiology of recent past progress in PY may point the way to future gains. Harvest index is approaching 0.5 and increasingly PY progress is associated with increasing yields of biomass, especially for maize. Progress in both HI and biomass was associated with more efficient partition-ing of dry matter to, and utilization within, the growing reproductive structures around flowering. This has led to a greater number of grains per square meter, a change that has always accompanied PY increase—with the exception of tropical rice—and to greater sink strength during grain fill. Effective staygreen during grain filling was an impor-tant contributor to PY increases in maize. In all three crops and considering only field plot data, there was good evi-dence that two or all three of the following—crop growth rate, RUE, and Pmax—were higher around flowering in the highest yielding cultivars; there have been numerous earlier reports of increases during grain filling (e.g., Evans, 1993), but this may simply reflect increased sink strength.

Prospects for further PY progress must build on this physiological understanding, and the study by Sylvester-Bradley et al. (2005) on wheat yield prospects in the United Kingdom does just this. There appears to be only limited scope for lifting HI when it is already at 50%; for exam-ple, even with the best designed wheat plants, HI seems unlikely to exceed 0.60 or even 0.55 if the risk of lodging is to be kept at a reasonable level (Berry et al., 2007). More biomass production, therefore, must be the main way for-ward, and this usually means greater crop growth rate, since extensions of crop duration have other limitations. Greater crop growth rate is a question of RUE because most crops grown under high yield potential conditions intercept >95% of the incident solar radiation for much of their life cycles. Under such conditions, RUE is a function of Pmax and of light distribution in the canopy, but light distribu-tion appears to already have already been optimized in the erect-leaf canopies of most modern cereal cultivars.

Evidence to support increased photosynthetic rate and crop growth rate is provided by CO2 fertilization experi-ments that show yield responds to greater crop growth rate at key stages of development, at least in C3 crops, in which Pmax responds more strongly to CO2 increase above current ambient.. It is noteworthy that Pmax rates in modern wheat cultivars in the field are clearly below the highest rates of


Pmax found in the wild relatives of wheat (Evans, 1993). And although Evans (1993) gives reasons why Pmax may not have changed initially with breeding, and why selection for Pmax itself has not delivered higher yields, there seems no strong physiological reasons why it cannot be pushed higher with-out insurmountable trade-offs. Finally, we are well below the theoretical photosynthetic limit, which is commonly taken from the response of photosynthesis to radiation at low levels of radiation; according to Zhu et al. (2008), for C3 crops this amounts to the capture (net of respiration) of 4.6% of intercepted total solar radiant energy as carbohy-drate energy or about 2.7 g DM MJ–1 intercepted total solar radiation. The highest recorded conversion rates for the full crop life cycle are around half of this; for shorter periods 2 g DM MJ–1 have been recorded. The corresponding theoreti-cal limit for C4 crops is 6.0% or 3.5 g DM MJ–1 of total inci-dent solar energy (Zhu et al., 2008). Actual rates found in maize fall well short of this. Lindquist et al. (2005) reported one of the highest conversion rates for maize, namely 1.8 g DM MJ–1, over the crop cycle. It may take more leaf N or it may not, but RUE must remain the focus of any major yield initiative if we are to improve, complement, or even maintain the rates of progress currently derived largely from conventional breeding. Surely the easiest way forward is to look at the natural variation we have within our crop species and their close relatives while bearing in mind the complexity of source-sink relationships in modern culti-vars. The availability of untapped genetic diversity within each crop species for almost every trait is considerable; its efficient utilization remains a major breeding challenge. A related way forward could be to explore and exploit the apparent differences observed among modern cultivars in response to environmental changes, for example, CO2 increase experienced over the past 40 yr (Ziska, 2008).

This paper does not have space to discuss in depth several other important issues surrounding breeding. Increased PY also results in increased resource use efficiency, not only for solar radiation but also water, nitrogen (Fig. 7), phosphorus,

energy, and labor. Breeding offers some scope to counteract the negative effects of higher temperature on yield, which are expected to accompany climate change in most current crop-ping locations. The breeding resources diverted from yield breeding to maintenance of disease resistance and improve-ment of grain quality are substantial, especially in wheat and rice, and may account for a significant part of the differences in rate of global yield increase between maize and the two self-pollinated cereals. The advent of new tools including molecular markers, genomic selection, association mapping, marker-aided recurrent selection, bioinformatics, biometrics, robotics, and remote sensing are beginning to aid breeding for yield. These specialized techniques, as they mature, have considerable potential to increase rates of progress and may even reduce the unit cost of yield gains.

The possibilities for genetic engineering to increase PY per se is obviously an important subject, but skepticism has been expressed about many of the claims and assump-tions made by its proponents with respect to progress at the crop level in the medium term (e.g., Fischer, 2008). At the very least there needs to be much more attention to the physiology that links processes at the molecular level to crop performance in the field, a recurring theme in recent crop science literature (e.g., Edmeades et al., 2004; Sinclair and Purcell, 2005; Spiertz et al., 2007). Recent develop-ments suggest there is credible evidence of field progress in PYW through the use of transgenics (Castiglioni et al., 2008; Zhang, 2009); in general this seems to be related to countering the well-known sensitivity of seed number to drought around flowering in cereals, although in the case of maize it appears to have a positive effect on drought-affected yields throughout the crop lifecycle. The next few years must clearly validate the success of these claims of genetic engineering for increased PY and PYW and reveal the underlying physiological mechanisms if we are to place any reliance on this approach to improving yield per se.

Finally, a strong and competitive plant breeding and seed industry, fostered by hybrid cultivars and other forms

Table 4. Summary of rates of recent progress in yield expressed relative to measured or predicted yield in 2007 or 2008. See text for sources.

Region and Period

Potential yield (PY)

Farm yield (FY)

Yield gap 2007 Change Progress 2007 Progress % of FYt ha–1 kg ha–1 yr–1 % per annum t ha–1 % per annum

Wheat

United Kingdom, 1989–2008 10.4 61 0.6 8 0.7 30

Yaqui Valley, 1979–2008 9 23 0.3 6 0.4 50

Rice

Japan, 1978–2007 11 104 0.9 6.5 0.4 70

Central Luzon, 1978–2007 wet season 6 0 0 3.8 0.6 58

Central Luzon 1978–2007 dry season 9 0 0 4.5 0.6 100

Maize

Iowa, 1990–2008 15.5 116 1.0 10.5 2.0 48

Sub-Saharan Africa 1989–2007 4† ? ? 1.6 0.8 >200

† Attainable yields with best bet management averaged for Malawi, Ethiopia, Nigeria, Uganda, Mali, and Mozambique (Source: C. Dowswell of Sasakawa Global 2000, personal communication, 2009).


of cultivar protection, appears to be a major factor in the remarkable progress in maize yields in the United States and surely has lessons for elsewhere.

ACKNOWLEDGMENTSWe are grateful to Derek Byerlee for his thoughtful insights and discussion that led to this paper. We thank the many individuals who provided data on estimates of yield by region for each crop.

REFERENCESAustin, R.B. 1982. Crop characteristics and the potential yield of

wheat. J. Agric. Sci. (Cambridge) 98:447–453.Bänziger, M., P.S. Sentinela, D. Hodson, and B. Vivek. 2006.

Breeding for improved stress tolerance in maize adapted to southern Africa. Agric. Water Manage. 80:212–224.

Bell, M.A., and R.A. Fischer. 1994. Using yield prediction models to assess yield gains: A case study for wheat. Field Crops Res. 37:149–160.

Bell, M.A., R.A. Fischer, D. Byerlee, and K.D. Sayre. 1995. Genetic and agronomic contributions to grain yields. Field Crops Res. 44:55–65.

Berry, P.M., R. Sylvester-Bradley, and S. Berry. 2007. Ideotype design for lodging resistant wheat. Euphytica 154:165–179.

British Society of Plant Breeders. 2008. New varieties underpin growth in UK cereal yields – NIAB study. p. 1–2. In Brit-ish Society of Plant Breeders Newsletter, Nov. 2008. British Society of Plant Breeders Ltd., Ely, UK.

Bueno, C.S., and T. Lafarge. 2009. Higher crop performance of rice hybrids than of elite inbreds in the tropics: 1. Hybrids accumulate more biomass during each phenological phase. Field Crops Res. 112:229–237.

Campos, H., M. Cooper, J.E. Habben, G.O. Edmeades, and J.R. Schussler. 2004. Improving drought tolerance in maize: A view from industry. Field Crops Res. 90:19–34.

Cassman, K.G. 1999. Ecological intensification of cereal produc-tion systems: Yield potential, soil quality, and precision agri-culture. Proc. Natl. Acad. Sci. USA 96:5952–5959.

Cassman, K.G., A. Dobermann, D.T. Walters, and H. Yang. 2003. Meeting cereal demand while protecting natural resources and improving environmental quality. Annu. Rev. Environ. Resour. 28:315–358.

Cardwell, V.B. 1982. Fifty years of Minnesota corn production: Sources of yield increase. Agron. J. 74:984–990.

Castiglioni, P., D. Warner, R.J. Bensen, D.C. Anstrom, J. Har-rison, M. Stoecker, M. Abad, G. Kumar, S. Salvador, R. D’Ordine, S. Navarro, S. Back, M. Fernandes, J. Targolli, S. Dasgupta, C. Bonin, M. Luethy, and J.E. Heard. 2008. Bacte-rial RNA chaperones confer abiotic stress tolerance in plants and improved grain yield in maize under water-limited con-ditions. Plant Physiol. 147:446–455.

Condon, A.G., M.P. Reynolds, G.J. Rebetzke, M. van Ginkel, R. Richards, and G. Farquhar. 2007. Using stomatal aperture-related traits to select for high yield potential in bread wheat. p. 617–624. In H.T. Buck et al (ed.) Wheat Production in Stressed Environments. Proc. 7th Int. Wheat Conference, Mar del Plata, Argentina, 27 Nov–2 Dec, 2005. Springer, New York, NY.

Cooper, M., O.S. Smith, G. Graham, L. Arthur, L. Feng, and D.W. Podlich. 2004. Genomics, genetics, and plant breeding: A private perspective. Crop Sci. 44:1907–1913.

Duvick, D.N. 2005. The contribution of breeding to yield

advances in maize (Zea mays). Adv. Agron. 86:83–145.Duvick, D.N., and K.G. Cassman. 1999. Post-green revolution

trends in yield potential of temperate maize in the north-cen-tral United States. Crop Sci. 39:1622–1630.

Echarte, L., and T. Tollenaar. 2006. Kernel set in maize hybrids and their inbred lines exposed to stress. Crop Sci. 46:870–878.

Edmeades, G.O., M. Bänziger, H. Campos, and J. Schussler. 2006. Improving tolerance to abiotic stresses in staple crops: A ran-dom or planned process? p. 293–309. In K.R. Lamkey and M. Lee (ed.) Plant breeding. The Arnel R. Hallauer International Symposium. Blackwell, Ames, IA.

Edmeades, G.O., G.S. McMaster, J.W. White, and H. Campos. 2004. Genomics and the physiologist: Bridging the gap between genes and crop response. Field Crops Res. 90:5–18.

Edmeades, G.O., and T. Tollenaar. 1990. Genetic and cultural improve-ment in maize production. p. 164–180. In S.K. Sinha et al (ed.) Proc. Intl. Congr. Plant Physiol. Vol. 1. IARI, New Delhi, India.

Estudillo, J., and K. Otsuka. 2001. Has the Green Revolution ended? A review of long-term trends in MV adoption, rice yields, and rice income in central Luzon, 1966–99. Jpn. J. Rural Econ. 3:51–64.

Evans, L.T. 1993. Feeding the ten billion. Cambridge Univ. Press, Cambridge, UK.

Fischer, R.A. 2008. Improvements in wheat yield: Farrer, physiol-ogy and functional genomics. Agric. Sci. NS 1/08:6–18.

Fischer, R.A. 2009. Farming systems of Australia: Exploiting the synergy between genetic improvement and agronomy. p. 23–54. In V. Sadras and D. Calderini (ed.) Crop physiology: Applications for genetic improvements and agronomy. Else-vier, Amsterdam, The Netherlands.

Fischer, R.A., D. Byerlee, and G.O. Edmeades. 2009. Can tech-nology deliver on the yield challenge to 2050? Paper prepared for expert meeting on “How to Feed the World in 2050.” 24–26 June 2009. FAO, Rome.

Fischer, R.A., D. Rees, K.D. Sayre, Z.-M. Lu, A.G. Condon, and A. Larque-Saavedra. 1998. Wheat yield progress associated with higher stomatal conductance and photosynthetic rate, and cooler canopies. Crop Sci. 38:1467–1475.

Fischer, R.A., and P.C. Wall. 1976. Wheat breeding in Mexico and yield increases. J. Aust. Inst. Agric. Sci. 42:139–148.

Goto, A., H. Sasahara, A. Shigemune, and K. Miura. 2009. Hokuriku 193: A new high-yielding indica rice cultivar bred in Japan. JARQ 43:13–18.

Grassini, P., H. Yang, and K.G. Cassman. 2009. Limits to maize productivity in western Corn Belt: A simulation analysis for fully irrigated and rainfed conditions. Agric. For. Meteorol. 149:1254–1265.

Hammer, G.L., Z. Dong, G. McLean, A. Doherty, C. Messina, J. Schussler, C. Zinselmeier, S. Paszkiewicz, and M. Cooper. 2009. Can changes in canopy and/or root system architecture explain historical maize yield trends in the U.S. Corn Belt? Crop Sci. 49:299–312.

Horie, T., T. Shiraiwa, K. Homma, K. Katsura, S. Maeda, and H. Yoshida. 2005. Can yields of lowland rice resume the increases that they showed in the 1980s? Plant Prod. Sci. 8:259–274.

Jennings, P.R. 1964. Plant type as a rice breeding objective. Crop Sci. 4:13–15.

Katsura, K., S. Maeda, T. Horie, and T. Shiraiwa. 2007. Analysis of yield attributes and crop physiological traits of Liangyoupeiju, a hybrid rice recently bred in China. Field Crops Res. 103:170–177.

Katsura, K., S. Maeda, T. Horie, and T. Shiraiwa. 2009. Estima-tion of respiratory parameters for rice based on long-term and intermittent measurement of canopy CO2 exchange rates in


the field. Field Crops Res. 111:85–91.Keating, B., P. Carberry, P. Bindraban, S. Asseng, and H. Meinke.

2010. Eco-efficient agriculture: Concepts, challenges and opportunities. Crop Sci. 50(2):109–119.

Kucharik, C.J. 2008. Contribution of planting date trends to increased maize yields in the central United States. Agron. J. 100:328–336.

Lindquist, J.L., T.J. Arkebauer, D.T. Walters, K.G. Cassman, and A. Dobermann. 2005. Maize radiation use efficiency under optimal growth conditions. Agron. J. 97:72–78.

Liu, W., M. Tollenaar, G. Stewart, and W. Deen. 2004. Impact of planter type, planting speed, and tillage on stand uniformity and yield of corn. Agron. J. 96:1668–1672.

Lobell, D.B., K.G.Cassman, and C.B. Field. 2009. Crop yield gaps: Their importance, magnitudes, and causes. Annu. Rev. Envi-ron. Res. 34: 4.1–4.26.

Lobell, D.B., and J.I. Ortiz-Monasterio. 2008. Satellite monitor-ing of yield responses to irrigation practices across thousands of fields. Agron. J. 100:1005–1012.

Lobell, D.B., J.I. Ortiz-Monasterio, G.P. Asner, P.A. Matson, R.L. Naylor, and W.P. Falcon. 2005. Analysis of wheat yields and climate trends in Mexico. Field Crops Res. 94:250–256.

Loomis, R.S., and D.J. Connor. 1992. Crop ecology: Productivity and management in agricultural systems. Cambridge Univ. Press, Cambridge, UK.

Luque, S.F., A.G. Cirilo, and M.E. Otegui. 2006. Genetic gains in grain yield and related physiological attributes in Argentine maize hybrids. Field Crops Res. 95:383–397.

Mae, T., A. Inaba, Y. Kaneta, S. Masaki, M. Sasaki, M. Aizawa, S. Okawa, S. Hasegawa, and A. Makino. 2006. A large-grain cultivar, Akita 63, exhibits high yields with high physiologi-cal N-use efficiency. Field Crops Res. 97:227–237.

Ohsumi, A., A. Hamasaki, H. Nakagawa, H. Yoshida, T. Shi-raiwa, and T. Horie. 2007. A model explaining genotypic and ontogenetic variation of leaf photosynthetic rate in rice (Oryza sativa) based on leaf nitrogen content and stomatal con-ductance. Ann. Bot. (Lond.) 99:265–273.

Ortiz-Monasterio, J.I., and D.B. Lobell. 2007. Remote sensing assessment of yield losses due to sub-optimal planting dates and fallow period weed management. Field Crops Res. 101:80–87.

Peng, S., K.G. Cassman, S.S. Virmani, J. Sheehy, and G.S. Khush. 1999. Yield potential trends of tropical rice since the release of IR8 and the challenge of increasing rice yield potential. Crop Sci. 39:1552–1559.

Peng, S., G.S. Khush, P. Virk, Q. Tang, and Y. Zou. 2008. Prog-ress in ideotype breeding to increase rice yield potential. Field Crops Res. 108:32–38.

Peng, S., R.C. Laza, R.M. Visperas, A.L. Sanico, K.G. Cassman, and G.S. Khush. 2000. Grain yield of rice cultivars and lines developed in the Philippines since 1966. Crop Sci. 40:307–314.

San-oh, Y., Y. Mano, T. Ookawa, and T. Hirasawa. 2004. Com-parison of dry matter production and associated character-istics between direct-sown and transplanted rice plants in a submerged paddy field and relations to planting patterns. Field Crops Res. 87:43–58.

Sayre, K.D., S. Rajaram, and R.A. Fischer. 1997. Yield potential prog-ress in short wheats in northwest Mexico. Crop Sci. 37:36–42.

Shearman, V.J., R. Sylvester-Bradley, R.K. Scott, and M.J. Foul-kes. 2005. Physiological processes associated with wheat yield progress in the UK. Crop Sci. 45:175–185.

Shimshi, D., and J. Ephrat. 1975. Stomatal behavior of wheat

cultivars in relation to their transpiration, photosynthesis and yield. Agron. J. 67:326–331.

Sinclair, T.R., and L.C. Purcell. 2005. Is a physiological perspective relevant in a “genocentric” age? J. Exp. Bot. 56:2777–2782.

Spiertz, J.H.J., P.C. Struik, and H.H. van Laar (ed.). 2007. Scale and complexity in plant systems research: Gene-plant-crop relations. Springer, New York, NY.

Sylvester-Bradley, R., J. Foulkes, and M. Reynolds. 2005. Future wheat yields: Evidence, theory and conjecture. p. 233–260. In R. Sylvester-Bradley and J. Wiseman (ed.) Yields of farmed species. Nottingham Univ. Press, Nottingham, UK.

Takai, T., S. Matsuura, T. Nishio, A. Ohsumi, T. Shiraiwa, and T. Horie. 2006. Rice yield potential is closely related to crop growth rate during late reproductive period. Field Crops Res. 96:328–335.

Taylaran, R.D., S. Ozawa, N. Miyamoto, T. Ookawa, T. Moto-bayashi, and T. Hirasawa. 2009. Performance of a high-yield-ing modern rice cultivar Takanari and several old and new cultivars grown with and without chemical fertilizer in a sub-merged paddy field. Plant Prod. Sci. 12:365–380.

Tollenaar, M., and E.A. Lee. 2006. Dissection of physiologi-cal processes underlying grain yield in maize by examining genetic improvement and heterosis. Maydica 51:399–408.

Tubiello, F.N., J.S. Amthor, K.J. Boote, M. Donatelli, W. Easter-ling, G. Fischer, R.M. Gifford, M. Howden, J. Reilly, and C. Rosenzweig. 2007. Crop response to elevated CO2 and world food supply. Eur. J. Agron. 26:215–223.

Tweeten, L., and S.R. Thompson. 2008. Long-term agricultural output supply-demand balance and real farm and food prices. Working Paper AEDE-WP 0044-08, Ohio State University, Columbus, OH.

van Ittersum, M.K., and R. Rabbinge. 1997. Concepts in pro-duction ecology for analysis and quantification of agricultural input-output combinations. Field Crops Res. 52:197–208.

Watanabe, N., J.R. Evans, and W.S. Chow. 1994. Changes in the photosynthetic properties of Australian wheat cultivars over the last century. Aust. J. Plant Physiol. 21:169–183.

Yang, H.S., A. Dobermann, J.L. Lindquist, D.T. Walters, T.J. Arkebauer, and K.G. Cassman. 2004. Hybrid-Maize – a maize simulation model that combines two crop modeling approaches. Field Crops Res. 87:131–154.

Yang, W., S. Peng, R.C. Laza, R.M. Visperas, and M.L. Dionisio-Sese. 2007. Grain yield and yield attributes of new plant type and hybrid rice. Crop Sci. 47:1393–1400.

Yoshinaga, S., Y. Arai (San-oh), T Ishimaru, A. Ohsumi and M. Kondo. 2009. Characteristics of dry matter production and grain filling in new rice cultivar “Momiroman” suitable for grain feed use. Japan. J. Crop Sci. 227: 212-213 (in Japanese).

Zhang, Q. 2009. Strategies for developing super green rice. Proc. Natl. Acad. Sci. USA 104:16402–16409.

Zhang, Y., Q. Tang, Y. Zou, D. Li, J. Qin, S. Yang, L. Chen, B. Xia, and S. Peng. 2009. Yield potential and radiation use effi-ciency of “super” hybrid rice grown under subtropical condi-tions. Field Crops Res. 114:91–98.

Zhu, X.-G., S.P. Long, and D.R. Ort. 2008. What is the maxi-mum efficiency with which photosynthesis can convert solar energy into biomass? Curr. Opin. Biotechnol. 19:153–159.

Ziska, L.H. 2008. Three-year field evaluation of early and late 20th century spring wheat cultivars to projected increases in atmospheric carbon dioxide. Field Crops Res. 108:54–59.


SYMPOSIA

Genetic variability hidden in our plant and animal genomes represents an international treasure important to everyone

living on this planet. Natural selection of this variability is the basis for the evolution of all living things, as proposed by Darwin in Origin of the Species 150 yr ago (Darwin, 1859). Selection of phenotypic traits at the hands of humankind for the past 10,000 yr now allows the production of enough food for 6.5 billion people. No one knows if this hugely successful past can foretell the future. Fortunately, undiscovered variability lurks in every genome to an extent that we could not even contemplate before the advent of genomics, the main theme of this article. Genetic variability is worthless in agriculture if not utilized; once discovered the variability needs to be understood and incorporated into breed-ing programs. Rapid means to detect the variability, identify the gene function(s), and progress through a breeding program is key to breaking yield barriers. With the past and modern advances in genetics and the advent of transgenic technology, the circle is complete in making available the wonderful arrays of variability manifested in the natural world.

Yield barriers can be viewed from at least two perspectives. A yield barrier may exist because increases in production are not keeping pace with the increases in demand. Another viewpoint is that a yield barrier is present when physiological maxima have

Mobilizing Science to Break Yield Barriers

Ronald L. Phillips*

ABSTRACTYieldbarriersmustbebroken.Thediminishedstockofstaple foods,highergrainprices,andincreases inproduction failing tokeepupwithdemand,coupledwith80millionpeoplebeingaddedtotheworldpopulationeveryyear,sug-gests that we are on a collision course withfamineunlessgreaterinvestmentsaremadeinresearchanddevelopment,aswellaseducation.Geneticimprovementofstapleshasaccountedformorethanhalfofthepastincreasesinyields.Fortunately, a revolution in genetic knowledgeis co-evolving with the increased demand forfood,feed,fiber,andfuel.Utilizinggeneticdiver-sity has been a mainstay of past productionimprovementsHigh throughputDNAsequenc-ing, the related bioinformatics, and a cascadeofgenetic technologiescannowbeemployedto detect previously hidden genetic variability,tounderstandgenefunctions,tomakegreateruseofaccessionsingermplasmbanks,andtomakebreedingschemesmoreefficacious.Theinvolvementofoutstandingscientistswhocanbring interdisciplinary ideas to the question ofhowtobreakyieldbarriersmustbepartofthestrategy. Educational programs at all levels,evenhighschool,shouldemphasizetheoppor-tunities in international agriculture to build acadreofdedicatedscientistsforthefuture.

Regents Professor and McKnight Presidential Chair in Genomics, Dep. of Agronomy and Plant Genetics, Univ. of Minnesota, 1991 Upper Buford Cir., St. Paul, MN 55108. Received 21 Sept. 2009. *Corre-sponding author ([email protected]).

Abbreviations: MAS, marker-assisted selection; QTL, quantitative trait locus.




been reached. Most technologies reported herein apply directly to the first perspective and indirectly to the sec-ond. Transferring C4 photosynthesis, for example, would apply directly to the second viewpoint.

The importance of breaking yield barriers is paralleled by the rate of the world population increase. The world pop-ulation is expanding at a rate of about 1 billion people every 12 to 14 yr, or about 80 million people per year. T.J. Hig-gins, the Commonwealth Scientific and Industrial Research Organization’s (CSIRO) deputy chief of plant industries, recently said that “population and rising wealth could mean an extra 10 billion tons of food consumed each year by 2025” (McKenzie, 2009). Furthermore, the effect of population increase is much more pronounced in the developing world where applications of some of the new technologies are just commencing to be applied to food production.

The objective of this meeting as stated in the Science Forum 2009 announcement was as follows: “The Forum intends to stimulate provocative and challenging discus-sion, with a forward-looking perspective on research and partnership needs to increase the resilience and the productivity of agricultural and natural resource systems (CGIAR, 2009).”

This report principally focuses on technology and the progress expected from these technologies. Bruce Alberts, the former President of the National Academy of Sciences and current editor-in-chief of Science magazine, recently stated in an editorial on breakthroughs of the year: “The scien-tists who achieved each of this year’s breakthroughs exploited techniques and instrumentation that were unimaginable when I began my life as a scientist in the 1960s. To men-tion only a few: computational speeds and methods, detec-tors, telescopes, DNA sequencers, and recombinant DNA technologies. These new technologies are created from the knowledge of the natural world generated by previous scien-tific and technical advances. Therefore, the more we know, the more we can discover, and the pace of scientific discovery constantly accelerates (Alberts, 2008).” It seems clear that we must embrace traditional approaches but complement them with new technologies to maximize the prospect of breaking the yield barrier in food staples.

High food prices have left millions of people hungry, with this group of people now accounting for 14% of the world’s population. Oxfam reported “current severe food shortages in Afghanistan, Ethiopia, Kenya, Mozambique, and Zimbabwe are evidence that the global food crisis is far from over. Even before recent price rises, there were over 850 million people classified as undernourished. Now, there are nearly a billion, as a result of the price rises, alongside other factors such as political instability and conflict (Oxfam International, 2009).”

Nearly two-thirds of the world’s undernourished people live in Asia and one-third in Sub-Saharan Africa. Speak-ing on behalf of the developing countries, the Bangladesh

minister of Agriculture, C.S. Karim, indicated that “we need to find more powerful gene revolution tools for food, feed, medicine, renewable energy, and other human needs” (ISAAA, 2008). It is encouraging that one of the world’s largest agricultural companies, Monsanto, has reconfirmed that they believe their goal of doubling crop yields by 2030 is attainable (Wanzek, 2008). Although much more than technology is needed to increase food production, includ-ing credit, availability of inputs such as nutrients and water, markets, and other externalities, about half of the increases in productivity in major crops over the past several years have been due to the genetic improvement of these crops. Thus, this article will focus on genetic technologies believed to be important for the future of agriculture. Employment of new agronomic practices is also essential and may need to be coupled with the breeding procedure for maximum impact.

History shows that as the population increased, agri-cultural practices and technology allowed more food to be produced on less land. The assumption is made that breaking the yield barrier will again mean that more food be produced on less land (Table 1).

Maize provides a good case study to show how plant breeding technology has allowed a 1% or better gain per year. Figure 1 (used with permission of A. Forrest Troyer; E. Wellin and F. Troyer, unpublished data, 2008) shows the steady improvement in corn yields since new technol-ogies were introduced. Note that the slope increases with the advent of each new technology– from open-pollinated populations, to double crosses, to single crosses, to bio-tech hybrids.

Productivity increases for the important staples rice and wheat, however, are not nearly as impressive as for corn. The average increase in rice production in the 1980s was 3.1% per year but in the 1990s it decreased to 1.4% per year. Even worse, the 2000s have seen increases of only 0.8% per year. Likewise, wheat production increased at a rate of 2.9% per year in the 1980s but only 0.9% in the 1990s and 0.4% in the 2000s (Ziska, USDA-ARS, Beltsville, MD, personal communication, 2009). Why is this the case? Pardey et al. (2006) argues that reduced public investment in agricul-tural research may be a principal factor. The private sector investment in maize may account for its success.

As stated by Cliff Weil ( Johal et al., 2008): “The suc-cess of a breeding program depends on having adequate diversity in the germplasm. However, as advanced breed-ing stocks and materials are generated, one casualty is the diversity itself. As a result, breeding programs in many crop species have reached a point of diminishing returns and it is feared that unless new diversity is infused into the breeding germplasm, we face catastrophic reductions in crop productivity if the climate turns adverse. Although some scientists favor transgenic approaches, a ‘back to nature’ approach to genetic diversity may prove faster and more effective. Wild and exotic relatives of crop plants


the human genome cost $3 billion and took 13 yr. Two years ago, James Watson’s genome was sequenced in 2 mo for $2 million. Several plant species now have been sequenced such as Arabidopsis, rice, poplar, grapes, sor-ghum, maize, soybean, etc. (see http://www.jgi.doe.gov [verified 14 Dec. 2009]). The sequencing of Arabidopsis required $70 million and about 7 yr. Maize, which has about 2300 Mb of DNA as opposed to the much smaller Arabidopsis 140 Mb genome, will cost half as much as Arabi-dopsis. New sequencing procedures now can give sequenc-ing reads of 400 to 500 bases or longer and generate more than 1 million sequencing reads per 10-h instrument run (see http://www.454.com [verified 14 Dec. 2009]) Newer technologies such as the Single Molecule Real Time (SMRT) DNA sequencing technology (see http://

hold a wealth of alleles that, if we can find them, can help break yield barriers and enhance tolerance to stresses.”

A 2009 National Research Council Report entitled “Emerging technologies to benefit farmers in Sub-Saharan Africa and South Asia” reports on 60 emerging technolo-gies and lists 18 as priorities for immediate development (National Research Council, 2009). Technologies con-sidered the most important are those that “(i) manage the natural resource base supporting agriculture; (ii) improve the genetic characteristics of crops and animals; (iii) reduce biotic constraints (such as disease, pests, and weeds); and (iv) provide affordable, renewable energy for farmers.” This article supports those opinions, but due to space limitations obviously cannot directly touch on all of them. The analy-sis presented here was deliberately prepared before read-ing the NRC report to present an independent viewpoint on breaking the yield barrier. The two reports are both supportive and complementary of each other. This article perhaps provides more emphasis on breeding schemes, vari-ability in parental materials, and investing in human capital.

VARIABILITYDetection of variability depends in large part on

progress in DNA sequencing. The initial deciphering of

Table 1. Historical information on the amount of land required to provide food for the size of the population at different times (cf. CropLife Intl. 2007–2008 Annual Report).

Date Population Land/Person1960 3.0 billion 4.3 ha (10.8 ac)

1980 4.4 billion 3.0 ha (7.5 ac)

2000 6.0 billion 2.2 ha (5.5 ac)

2020 7.5 billion 1.8 ha (4.5 ac)

2050 9.2 billion ???

Fig. 1. Average U.S. corn yields and kinds of corn, Civil War to 2007; periods dominated by open-pollinated varieties, four-parent hybrids, two-parent hybrids, and genetically modified hybrids are shown. b values (regressions) are yield gain per year (kg/bu). USDA data compiled by E. Wellin and F. Troyer.


www.pacificbiosciences.com [verified 14 Dec. 2009]) may allow the sequencing of Arabidopsis for $70. About 15 Gb of DNA may be sequenced in a 7-d run in the future at a cost of about $0.50 per million bases.

If understanding and utilizing diversity is one of the keys to breaking the yield barrier, why don’t we think big and sequence the entire rice germplasm collection of 100,000 accessions in the International Rice Research Institute (IRRI) genebank? At $0.50 cost per million bases and rice with 430 Mb of DNA, the cost would be $200 to $300 per accession, or a total of $20 to $30 million. At first, this bold proposal may appear unrealistic. However, consider the fact that there is already a 1000 Genomes Project for humans and a proposed 1001 Genomes Project for Arabidopsis. The goal of the 1001 Genomes Project is to discover whole-genome sequence variation in Arabidopsis. Although such information in Arabidopsis will provide important param-eters for understanding variability in other species, much of the information is species- and strain-specific.

As scientists, we must remember that technology advances—especially in genetics—faster than most of us would ever contemplate. Even with the advances in genomics and other “omics,” some still complain that the new technology has not solved many problems. This will no doubt be a fading viewpoint as agricultural applications become apparent, especially in rice. With regard to DNA sequencing, the sequencing of the 5386-base bacteriophage PhiX174 in 1977 was a highlight in science. Considering the technology used to sequence PhiX174, completion of the E. coli chromosome would have required a thousand years and the human genome would have taken a mil-lion years (Stein, 2008). Resequencing is a rapidly grow-ing activity and will be extremely enhanced as the next generation sequencing comes along (Liszewski, 2009). In resequencing, the sequence of many lines are compared to the sequence of the reference genome to find differences that provide a multitude of useful new molecular genetic markers. Pioneer Hi-Bred International has resequenced more that 600 key lines for 10,000 sequences (Mark Coo-per, Pioneer Hi-Bred Intl. personal communication, 2009).

Sorting out useful information is a challenge when dealing with such large amounts of data. However, bioinformatics approaches are becoming more and more user friendly. A new web-based program called TAR-GeT (formerly called TATE) will automatically identify and list gene family members from a genomic database and show a phylogenetic tree. For example, TARGeT has been used to find the homologs of the ascorbate peroxi-dase gene family in maize, rice, and sorghum. TARGeT can find the known homologs in this gene family in the 430 Mb genone of rice and draw a phylogenetic tree—and it takes only 75 s! (Han et al., 2009). Twenty-eight homo-logs were found across all three genomes (totaling thou-sands of megabases)—and that only took about 10 min.

For maize, if the genetic map location of a gene is known, the Maize GDB Genome Browser will show the sequence in that region using the Locus Lookup tool (Sen et al., 2009). For the waxy (wx1) locus, for example, the tool shows that this mutation is between 28,694,400 and 28,699,300 on chromosome 9. The gene thus is targeted to a 4900 base pair region of the genome.

Many genomics-related software programs are available for testing a wide variety of questions (National Research Council, 2008; see http://www.nap.edu [verified 14 Dec. 2009]). A program called MAGIC (Mutant-assisted gene identification and characterization) searches for genes that either enhance or diminish certain traits (Weil, 2008). This approach utilizes the introgression of a mutant into various backgrounds and identifies QTLs (quantitative trait loci) that modify the expression of that mutant trait.

Beyond identifying genes and their known functions, databases exist that assist in understanding various plant metabolic pathways. The PMN (Plant Metabolic Net-work) database (Available at http://www.plantcyc.org/ [verified 14 Dec. 2009]) helps in understanding the chem-ical reactions that make up metabolic pathways, such as the conversion of carbon dioxide, transportation of nutri-ents, and responses to the environment (Weil, 2008).

Natural sequence variability in crop and animal genomes is vast. The resequencing approach identifies sin-gle nucleotide polymorphisms (SNPs) between and among different strains. It is not uncommon now to find a million differences that can be used as molecular genetic markers and identified by simple PCR (polymerase chain reaction) procedures. SNPs can be used in association mapping or other mapping approaches. Thus, the variation that can be used to locate the portion of a genome associated with the expression of a specific trait (Buckler et al., 2008) is read-ily available once the resequencing is accomplished.

Diversity is a key feature to plant improvement. The extensive germplasm collections maintained around the world are important in protecting future food supplies. They need to be maintained in viable condition, evalu-ated for traits of interest, and rejuvenated at appropriate times. But even with the proper care of these materials, they are generally difficult to use due to the fact that a cross brings in all of the deleterious factors along with the genetic factor of interest. Crosses between adapted variet-ies and exotic accessions require a long time to derive use-ful genetic material; this has deterred plant breeders from using exotic germplasm (Bernardo, 2009). Procedures need to be developed that allow the transfer of much less than the whole genome from the germplasm accession. Transgenic technology as practiced today only inserts one or a few genes at a time. Expansions of the technol-ogy are needed to allow the transfer of at least the “selec-tive sweep” of genes associated with a particular region. Of course, the extensive availability of SNPs allows the


monitoring of segments of the genome to facilitate the introgression of chromosome blocks. The identification of introgressed markers followed by the subsequent compari-son with the whole-genome sequence will identify the chromosomal regions that are now transferred from one strain to another (Stupar, Univ. Minnesota, personal com-munication, 2009).

Breaking the yield barrier can be achieved in two ways: Increasing “attainable yield” such as from insect resistance, etc., and from “potential yield” such as raising the base yield due to changes in physiological processes. Transgenics thus far appear to have raised operational yield. Interestingly, improvements in the heterotic mechanism(s) with traditional breeding have not been documented; much of the yield improvements have been due to increased inbred yield (Duvick, 1999). Plant pop-ulation density, leaf uprightness, lodging resistance, and “stay green” clearly have been important. To date, manip-ulation to increase photosynthesis does not appear to be involved in raising yields.

GENE FUNCTIONSystems for tagging genes are now quite robust.

About 250,000 T-DNA insertions are available in Ara-bidopsis (Phillips et al., 2004). This reflects an insertion about every 500 bp in the genome. According to a report by Eric Vollbrecht (Vollbrecht et al., 2008), about 2000 maize lines now exist that have a uniquely located endog-enous Ds (McClintock’s Disassociation element) inser-tion, such that 85% of the genetic map space is within 4 cM or less of a placed Ds. Given the propensity of Ds for local transposition, targeted disruption of most corn genes is now possible (see http://www.plantgdb.org/prj/AcDsTagging/ [verified 14 Dec. 2009]).

Testing for expression of genes across the genome in one experiment has become possible for many species. A tremendous number of sequences can be assayed for expres-sion. The various DNA chips manufactured by Affyme-trix (http://www.affymetrix.com [verified 14 Dec. 2009]), for example, for Arabidopsis, barley, citrus, cotton, maize, and Medicago allow thousands of genes to be assayed for expression. Nimblegen (http://www.nimblegen.com [veri-fied 14 Dec. 2009]) has an array for maize with 2 million probes. Of course, expression depends on the environment, so the conditions of plant growth are important and, ulti-mately, should reflect the targeted field situation.

BREEDINGPrecision phenotyping also will be a key to break-

ing the yield barrier. High throughput genotyping is of little value if the phenotype is not accurately measured. The measurement of yield requires sophisticated experi-mental designs involving large populations, replication across time and space, and appropriately adjusting for mois-ture and other factors. Whether the trait is a visible phe-notype or biochemical in nature, the accuracy of the data largely determines the validity of the interpretations. High throughput phenotyping even in the field will be increas-ingly common (Montes et al., 2007). What might be called “phenotype science” needs to be expanded in theory and practice and emphasized in all programs across the world.

Resurgence of doubled haploid breeding has occurred in perhaps 80% of the corn companies. With the new high throughput genotyping platforms and the exten-sive phenotyping efforts, the application of these technolo-gies is even more cost effective with doubled haploids since the material is true breeding. Other factors making the doubled haploid breeding method attractive include the development of an inducer line called RWS that generates

Fig. 2. Breeding approach using genomic selection which is shorter than a conventional program largely by eliminating the phenotypic evaluation of parents for the next cycle (GEBV, genomic estimated breeding value). Used with permission from Jean-Luc Jannink (Cornell University; J.-L. Jannink, unpublished data, 2008).


an average of over 8% haploids on an ear, more efficient chromosome doubling methods utilizing nitrous oxide, new genetic markers, etc. (Touraev et al., 2008), as well as the efficiency of introgressing transgenes—especially stacked transgenes (Rober et al., 2005). Training in the doubled-haploid method of breeding may be important in breaking yield barriers. Different systems of producing hap-loids exist with different crops; haploids can be produced at high efficiency at some point in the life cycle of different species. Of course, haploids also can be valuable in poly-ploid plants, such as the tetraploid potato where a haploid is essentially a diploid with much simpler genetics than the tetraploid parent (Mendiburu and Peloquin, 1977).

Genomic selection, where high-density marker information—if not whole genome sequences—allow the prediction of breeding values, is being considered by several plant breeders. All marker information is incor-porated into the prediction model (Heffner et al., 2009; Bernardo, 2009). Although only based on simulations at this point, the correlation of the estimated breeding value and the true breeding value may be as high as 0.85. This approach may decrease the need for extensive phenotyp-ing in certain portions of the breeding process. Genomic selection appears to be effective even for low heritabil-ity traits controlled by many genes. The marker-assisted selection (MAS) strategies currently employed seem to work best for genes with major effects and, thus, are not very efficient for most traits of interest which are gen-erally polygenic. Schaeffer (2006) estimated that the rate of gain could be increased two-fold and reduce the cost (perhaps by 92%) of progeny tests. Several recent papers on sequencing and mapping indicate that such strategies will be increasingly available for agricultural animal spe-cies (The Bovine Genome Sequencing and Analysis Con-sortium et al., 2009; The Bovine hapMap Consortium, 2009; Chessa et al., 2009).

The purpose of phenotyping in this breeding approach is to estimate, or re-estimate the effects of the various markers or sequences (Fig. 2).

Biotech varieties created through transgenic tech-nologies theoretically allow the capturing of any genetic variability of interest no matter its source. It seems pru-dent in regard to breaking the yield barrier to advocate both modern conventional breeding schemes and trans-genic approaches even though genetic engineering is not totally acceptable around the world. Many traits can be approached by both technologies, such as the extremely important trait of drought tolerance. For example, the WEMA (Water Efficient Maize for Africa) program hopes to have their first tolerant varieties available in 6 to 7 yr, with drought tolerant transgenic varieties to be released commercially in the U.S. in 2012 and in Sub-Saharan Africa by 2017 ( James, 2008). This work is supported by the Bill and Melinda Gates Foundation ($42 million)

and the Howard G. Buffett Foundation ($5 million) and reflects a highly collaborative project led by the Afri-can Agricultural Technology Foundation and involving CIMMYT, Monsanto, farmer groups and seed compa-nies in Kenya, Mozambique, South Africa, Tanzania, and Uganda. Many of the private sector companies are also working toward similar goals (Edmeades, 2008). Mon-santo recently submitted a request to USDA for release of the first corn tolerant to drought by transgenic technol-ogy. Their field trials indicate a 6 to 10% yield advantage across a range of genetic backgrounds under a stress that reduces yields by 50% (AgBioView, 2009).

Another example where a dual approach is prudent is in regard to submergence-tolerant rice. Flooding causes the loss of 4 million tons of rice each year—enough to feed 30 million people. Normal rice cannot tolerate com-plete flooding more than about 3 d, whereas rice with the sub1 gene from indica rice can survive 2 wk or more. This gene was identified about 13 yr ago and has now been incorporated by MAS into several megavarieties by IRRI and into local varieties by national programs. IRRI is able to transfer the gene into other varieties in 2.5 yr with the assistance of MAS. Trials by IRRI in Bangla-desh, Vietnam, Cambodia, and India have given positive results– saving the crop and providing income. This gene has been cloned (Xu et al., 2006) and could be transferred subsequently by transgenic technology.

Golden rice is an example where genetic variability for a trait of interest was not available in the target spe-cies, namely rice. Even though the rice germplasm bank is very comprehensive with over 100,000 entries, rice with higher levels of β-carotene did not exist. However, Potrykus and Beyer (Ye et al., 2000) were able to uti-lize genes in the daffodil and the Erwinia bacterium to complete the carotenoid pathway in rice. Golden rice 2 has the yellow endosperm gene from maize incorporated and gives a much higher carotenoid level. IRRI antici-pates releasing Golden rice 2 in 2012 (Aguiba, 2009). The Rockefeller Foundation will provide funding to help guide golden rice through national regulatory approval processes in Bangladesh, India, Indonesia, and the Philip-pines (Rodin, 2008).

Post-harvest preservation of the agricultural prod-uct can alleviate the need to break the yield barrier. The African Journal of Food, Agriculture, Nutrition, and Development (Oniang’o, 2009) reported a story in the local press that $8 million worth of maize was destroyed by court order due to high levels of aflatoxin. Given that about 10 million people in Kenya were chronically short of food (out of 36 million), such loss of food can be devas-tating– and can be prevented.

Genetics can play a role in reducing postharvest losses. Interesting examples relate to aflatoxin. For example, in groundnut, stilbene phytoalexin is produced in response


to fungal infection (Sobolev et al., 2006). The aflatoxin gene cluster of Aspergillus has been identified and stud-ied extensively (Carbone et al., 2007). The lipoxigenase enzymes (LOXs) also are believed to play a role in Asper-gillus infection (Tsitsigiannis et al., 2005).

In general, crop management and storage structures and practices are important in avoiding dangerous afla-toxin levels.

Frontier projects with high risk but high reward—and usually requiring partnerships as do most international agriculture projects—need to be part of the portfolio of approaches to break the yield barrier. Crop species differ in the efficiency of the photosynthetic process. In C3 plants, the direct carboxylation of ribulose-1,5-bisphosphate is achieved via the enzyme RUBISCO. This enzyme causes the loss of fixed CO2 via photorespiration, which can decrease the photosynthetic potential by 40% (Matsuoka et al., 2001). C4 plants concentrate CO2 in the bundle sheath cells—the locus of RUBISCO—therefore preventing the loss via photorespiration of previously fixed carbon.

Nineteen families of flowering plants have C4 pho-tosynthesis; the evidence indicates that C4 photosynthesis has evolved naturally over 50 times (Sage and Sage, 2007). This and the finding of species with intermediate forms of photosynthesis inspired the creation of a global consor-tium to work toward the development of a C4 rice that is expected to be much more efficient. The Bill & Melinda Gates Foundation has provided a grant of $11 million in support of the consortium led by IRRI; the goal is to develop rice plants that can produce 50% more yield using fewer inputs such as nitrogen fertilizer and water with much greater efficiency. This is a high risk long-term project requiring at least 10 yr (Sheehy et al., 2007).

An interesting approach is via the crossing of C3 and C4 species. The generation of chromosome addition lines by crossing oats by corn followed by embryo rescue (Phil-lips and Rines, 2009) represents one approach for trans-ferring C4 characteristics to a C3 species. A complete set of oat–corn addition lines have been produced where each corn chromosome (1 through 10) is present individually in an oat background. Over 650 radiation hybrid lines have been produced from these addition lines via irradiation pro-viding lines that have only a segment of a corn chromo-some in the oat background. Studies have shown that genes for key enzymes in photosynthesis are present in different addition lines as expected from earlier mapping data. These genes from corn express both the RNA and the protein in the oat background. CO2 compensation point analysis of the individual addition lines for corn chromosomes 6 and 9 was normal; however, the double addition of 6 plus 9 had a significantly lower compensation point but it was still more like oat than corn (Kowles et al., 2008).

Duplications arising from ancient tetraploidy can be observed in crop species such as corn and rice even though

their behavior is as diploids. One hundred fifty years ago in Origin of the Species, Charles Darwin said “It is not the stron-gest species that survive or the most intelligent, but the ones who are most responsive to change.” Could it be that dupli-cate genes and chromosome regions or whole genomes pro-vide that ability to be responsive to change? The original function of the gene can be maintained while the duplicate copy is free to change and provide a new or related func-tion. Or the expansion of a repeated genic region can lead to phenotypic changes. Triplet repeat expansions underlie many human genetic disorders and phenotypic variation in microbes as well as plants (Sureshkumar et al., 2009). Or do highly homologous duplicate genes produce products that interact to increase productivity? Since duplicate genes exist in all of our agricultural species, do these interact to provide hybrid vigor in outcrossing species or simply increases in productivity in selfing species?

De novo variation allows variation to appear in progenies that is not present in the parents. Today, we are aware of several mechanisms by which this can occur; these include point mutations, intragenic recombination, transposable elements, epigenetic variation, gene amplifi-cation, and others (Rasmusson and Phillips, 1997).

Understanding epigenetic variation may lead to a better understanding of intrinsic yield. The complexity of pheno-types in crop plants and the interactions with the environ-ment (G x E interactions) probably cannot be explained by structural genetics alone but must be considered together with DNA alterations occurring through potentially revers-ible changes such as histone modifications, methylation, and imprinting. A genomic methylation dataset will be possible reflecting various tissues and developmental times. A NIH Epigenome Roadmap project has been funded– involving four U.S. genome centers– to map epigenetic sites in about 100 cell types (Stein, 2009).

How the genetic background is manifested in terms of variability in a particular trait is not understood, yet the phenomenon is common and can reflect major differ-ences in expression of the phenotype. For example, early flowering in maize is considered a highly heritable trait but shows extensive genetic background effects. The vgt1 gene of maize had a 10-d effect on maturity in the mate-rial utilized for QTL analysis and gene cloning (Phillips et al., 1993). The effect can be absent in certain backgrounds and much greater than 10 d in others. The responsible genic segment is a noncoding sequence acting on a flow-ering gene 70 kb distant (Salvi et al., 2007).

How noncoding sequences are involved in G x E interactions is yet to be learned; however, any informa-tion on the molecular basis of G x E interactions will be important in breaking yield barriers.

Double-stranded break-enhanced gene targeting allows specific genes to be modified in specific ways. Endonucle-ases can be designed to produce double-stranded breaks at


predetermined sites. Such enzymes can cause a single cleav-age within a genome as large as the human genome– which is the same size as that of corn. Also, the genome of a patho-gen can be the target without affecting the host genome. Precision Biosciences (http://www.precisionbiosciences.com [verified 14 Dec. 2009]) has the “directed nuclease editor” that enables the production of such “homing endo-nucleases.” Zinc-finger nucleases are increasingly being proposed as a more reliable way of modifying genes, such as for the generation of herbicide-resistant plants (Townsend et al., 2009). A Zinc-Finger Consortium is making the technology openly available to researchers at no cost.

Learning from the past should help us enhance yields in the future. Considering the response of corn yield versus time, one can consider where different coun-tries are on that graph. India currently produces 2000 kg/ha (31 bu/ac), which is the same point the U.S. corn yields were at in about 1930. South Africa reflects about 1940 technology at 2500 kg/ha (40 bu/ac), Brazil is at 1950 yields, and China is at 1960 U.S. production levels. Even different states in the U.S. differ in their average yields (Fig. 3). The question is: Can we apply what has been learned in the improvement of corn yields in the U.S. to developing countries and greatly speed up the breaking of their current yield barrier? Does the answer lie in science, education, input availability, credit, workforce, politics, social structure, or other factors that impact food security?

FUTUREInvesting in human resources to capture new

ideas and enthusiastic dedication is perhaps one of the

most important approaches to breaking the yield bar-rier. Upon learning of the needs in developing countries, today’s students recognize the extensive opportunities that await them to make a difference. Because many students cannot spend extended time away from their current stud-ies, a program was developed to provide short-term expo-sure to international agriculture via IRRI (Phillips et al., 2008). A course on rice biology called ‘Rice Research to Production’ also is offered, funded by the National Sci-ence Foundation (see http://beta.irri.org/training/home.php [verified 14 Dec. 2009]). Another recent opportunity is the Monsanto Beachell-Borlaug International Scholars Program offering opportunities for Ph.D.-level training in rice and wheat breeding, connecting developed and developing countries (see http://www.monsanto.com/responsibility/sustainable-ag/produce_more/beachell_borlaug/goals.asp [verified 14 Dec. 2009]). Programs such as these are needed to bring fresh ideas to the yield barrier issue. Even starting younger at the high school level can be quite effective. The World Food Prize Global Youth Insti-tute identifies high school students who then spend a sum-mer at an international agricultural research institute (see http://www.worldfoodprize.org [verified 14 Dec. 2009]). There is little question that the experience turns their lives around causing them to have an intense interest in food production and poverty in developing nations.

The extensive interactions of the CGIAR centers with advanced research institutes and national programs (NARES) are highly collaborative and have paid huge dividends. These partnerships together with appropriate funding mechanisms such as the Challenge Grants (see

Fig. 3. U.S. corn yield improvement over the past century and more. Used with permission from Geoff Graham (Pioneer HiBred Intl.; G. Graham, unpublished data, 2009).


http://www.cgiar.org/impact/challenge/index.html [ver-ified 14 Dec. 2009]) need to be fostered, based on many years of positive experiences.

A combined effort of the Rockefeller Foundation and the Bill and Melinda Gates Foundation has established the Alliance for a Green Revolution in Africa (AGRA). Train-ing will be a major component, with the program expecting to train approximately 120 Ph.D.-level plant breeders over the next decade. For example, AGRA recently announced a new partnership with the University of Ghana, Legon, and the strengthening of a program piloted at the University of KwaZulu-Natal in South Africa to create a critical mass of breeders to alleviate Africa’s food deficit (see http://www.rockfound.org/about_us/news/2007/0919agra_pr.shtml [verified 14 Dec. 2009]).

To break yield barriers, truly interdisciplinary approaches need to be implemented, with many to be found outside of the traditional agricultural institutions. How do we make outstanding scientists in complementary fields aware of the opportunities? Publishing “invitational articles” in Science and other journals is one approach indi-cating the need and the rewards of such research (Phillips, 1997). Of course, there needs to be the prospect of longer-term funding.

In an April 2008 editorial in Science, Nina Federoff, senior scientific advisor to the U.S. Secretary of State, wrote: “A new Green Revolution demands a global com-mitment to creating a modern agricultural infrastructure everywhere, adequate investment in training and modern laboratory facilities, and progress toward simplified regu-latory approaches that are responsive to accumulating evi-dence of safety. Do we have the will and the wisdom to make it happen?” (Federoff, 2008).

Bill Gates suggested in a speech to the 2008 Davos World Economic Forum that we need a new business model. The model should include the motivation to help humanity and development driven by the profit motive. He called this “Creative capitalism”– coupling idealism and an altruistic desire to help others (Federoff, 2009).

Former U.S. Vice President Al Gore said at the 2009 American Association for the Advancement of Science meetings that: “There is nothing more powerful than an idea whose time has come”. He also said that: “If you want to go quietly, go alone. If you want to go far, go together.”

Pandit Jawaherlel Nehru said in a famous statement that “Everything else can wait, but not agriculture.” Let us not wait any longer but promptly move ahead in help-ing those in need. The theme in this article is that agricul-tural applications follow the biology. We are generating, for example, genome instruction books in regard to plants, animals, and microbes—how will we use them in agricul-ture around the world? Remember, to answer a question, the question must first be asked.

ReferencesAgBioView. 2009. Biotech frontier; World’s first drought-tolerant

corn; Biotech surge; Seeds of Promise; Enabling bio-innova-tions. 10 Mar. 2009. Available at http://www.agbioworld.org/newsletter_wm/index.php?caseid=archive&newsid=2873 (verified 14 Dec. 2009).

Aguiba, M.M. 2009. Fortified with Vitamin A: RP may be first to okay ‘Golden’ rice. Available at http://www.bic.searca.org/news/2009/feb/phi/15b.html (verified 14 Dec. 2009). Seameo Searca Biotechnology Information Center, Laguna, Philippines.

Alberts, B. 2008. Celebrating a year of science. Science 322:1757.Bernardo, R. 2009. Genomewide selection for rapid introgression

of exotic germplasm in maize. Crop Sci. 49:419–425.Buckler, E.S., J. Yu, J.B. Holland, and M.D. McMullen. 2008.

Genome-wide complex trait dissection through nested asso-ciation mapping. Genetics 178:539–551.

Carbone, I., J.H. Ramirez-Prado, J.L. Jakobek, and B.W. Horn. 2007. Recombination, balancing selection, and adaptive evo-lution in the aflatoxin gene cluster of Aspergillus parasiticus. Mol. Ecol. 16:4401–4417.

CGIAR. 2009. About Science Forum 2009. Available at http://www.scienceforum2009.nl (verified 14 Dec. 2009). CGIAR, Washington, DC.

Chessa, B., F. Pereira, F. Arnaud, et al. 2009. Revealing the his-tory of sheep domestication using retrovirus integrations. Sci-ence 324:532–536.

Darwin, C. 1859. Origin of the species. John Murray, Albemarle Street, London.

Duvick, D. 1999. Heterosis: Feeding people and protecting natural resources. p. 19–29. In J.G. Coors and S. Pandey (ed.) The genetics and exploitation of heterosis in crops. ASA, CSSA, Madison, WI.

Edmeades, G.O. 2008. Drought tolerance in maize: An emerg-ing reality. Companion document to Executive Summary, ISAAA Briefs 39.

Federoff, N. 2008. Seeds of a perfect storm. Science 320:425.Federoff, N. 2009. Science diplomacy in the 21st century. Cell

136:9–11.Han, Y., J.M. Burnette, and S. Wessler. 2009. TARGeT– a web-

based pipeline for retrieving and characterizing gene and transposable element families from genomic sequences. Nucleic Acids Res., 1–13 doi:10.1093/nar/gkp295.

Heffner, E.L., M.E. Sorrells, and J.-L. Jannink. 2009. Genomic selection for crop improvement. Crop Sci. 49:1–12.

ISAAA. 2008. Crop Biotech Update, Dec. 12, 2008. International Service for the Acquisition of Agri-Biotech Applications, Ithaca, NY.

James, C. 2008 Global status of commercialized biotech/GM crops: ISAAA Brief 39- Executive Summary.

Johal, G.S., P. Balint-Kurti, and C.F. Weil. 2008. Mining and harnessing natural variation: A little MAGIC. Crop Sci. 48:2066–2073.

Kowles, R.V., M.D. Walch, J.M. Minnerath, C.J. Bernacchi, A.O. Stec, H.W. Rines, and R.L. Phillips. 2008. Expression of C4 photosynthetic enzymes in oat–maize chromosome addition lines. Maydica 53:69–78.

Liszewski, K. 2009. 2009. Honing in on targeted resequencing. Genet. Eng. News 29(4), 15 Feb. 2009.

Matsuoka, R.C., R.T. Furbank, H. Fukayama, and M. Miyao. 2001. Molecular engineering of C4 photosynthesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52:297–314.


McKenzie, D. 2009. GM key to food shortage. www.agbioworld.org 3–16–09.

Mendiburu, A.O., and S.J. Peloquin. 1977. Bilateral sexual poly-ploidization in potatoes. Euphytica 26:573–583.

Montes, J.M., A.E. Melchinger, and J.C. Reif. 2007. Novel throughput phenotyping platforms in plant genetic studies. Trends Plant Sci. 12:433–436.

National Research Council. 2008. Achievements of the National Plant Genome Initiative and new horizons for plant biology. The National Academies Press, Washington, DC.

National Research Council. 2009. Emerging technologies to benefit farmers in sub-Saharan Africa and South Asia. The National Academies Press, Washington, DC.

Pardey, P.G., N. Beintema, S. Dehmer, and S. Wood. 2006. Agri-cultural Science and Technology Indicators Initiative. Intl. Food Policy Res. Inst., Washington, DC.

Oniang’o, R. 2009. Editorial - The never-ending story of Africa’s hunger tragedy: where lies food safety? Afr. J. Food Agric. Nutr. Dev. 9(1).

Oxfam International. 2009. Available at http://www.oxfam.org (verified 14 Dec. 2009.

Phillips, R.L. 1997. Opportunities await. Proc. Natl. Acad. Sci. USA 94:1609.

Phillips, R.L., T.S. Kim, S.M. Kaeppler, S.N. Parentoni, D.L. Shaver, R.E. Stucker, and S.J. Openshaw. 1993. Genetic dis-section of maturity using RFLPs. p. 135–150. In Proc. 47th Annu. Corn & Sorghum Industry Res. Conf.

Phillips, R.L., H. Leung, and R.P. Cantrell. 2004. An interna-tional genetic platform for the assessment of gene function in rice. Int. Crop Sci. Congr. Available at http://www.crop-science.org.au/ (verified 14 Dec. 2009).

Phillips, R.L., N.P. Magor, D. Shires, H. Leung, S.R. McCouch, and D. McIntosh. 2008. Student opportunity: Short-term exposure to international agriculture. Rice 1:11–15.

Phillips, R.L., and H.W. Rines. 2009. Genetic analyses with oat-maize addition and radiation hybrid lines. p. 523–538. In J.L. Bennetzen and S. Hake (ed.) Maize handbook– Vol. II: Genet-ics and genomics. Springer Science + Business Media LLC.

Rasmusson, D.C., and R.L. Phillips. 1997. Plant breeding prog-ress and genetic diversity from de novo variation and elevated epistasis. Crop Sci. 37:303–310.

Rober, F.K., G.A. Gordillo, and H.H. Geiger. 2005. In vivo hap-loid induction in maize– performance of new inducers and significance of doubled haploid lines in hybrid breeding. Maydica 50:275–283.

Rodin, J. 2008. Rockefeller Foundation to fund golden rice’s regu-latory approval effort. Available at http://www.gmo-compass.org/eng/news/387.rockefeller_foundation_fund_golden_rices_regulatory_approval.html (verified 14 Dec. 2009).

Sage, R., and T.L. Sage. 2007. Learning from nature to develop strategies for the directed evolution of C4 rice. p. 195–216. In Sheehy et al. (ed.) Charting new pathways to C4 rice. World Scientific Publishing.

Salvi, S., G. Sponza, M. Morgante, D. Tomes, X. Niu, K.A. Fen-gler, R. Meeley, E.V. Ananiev, S. Svitashev, E. Bruggemann, R. Li, C.F. Hanley, S. Radovic, G. Zaina, J.-A. Rafalski, S.V. Tingey, G.-H. Miao, R.L. Phillips, and R. Tuberosa. 2007. Conserved noncoding genomic sequences associated with

flowering-time quantitative trait locus in maize. Proc. Natl. Acad. Sci. USA 104:11376–11381.

Schaeffer, L.R. 2006. Strategy for applying genome-wide selec-tion in dairy cattle. J. Anim. Breed. Genet. 123:218–223.

Sen, T.Z., C.M. Andorf, D.A. Campbell, M.L. Schaeffer, L.C. Harper, and C.J. Lawrence. 2009. The maizeGDB browser. p. 28. In Proc. 51st Maize Genetics Conf.

Sheehy, J.E., et al. (ed.) 2007. Charting new pathways to C4 rice. World Scientific Publishing.

Sobolev, V., B.W. Horn, T.L. Porter, S.T. Deyrukp, and J.B. Gloer. 2006. Production of stilbenoids and phenolic acids by the peanut plant at early stages of growth. J. Agric. Food Chem. 54:3505–3511.

Stein, R.A. 2008. 2008. Next generation sequencing update. Genet. Eng. News 28(15), 1 Sept. 2008.

Stein, R.A. 2009. 2009. Epigenomics rises as key research tool. Genet. Eng. News 29(3), 1 Feb. 2009.

Sureshkumar, S., M. Todesco, K. Schneeberger, R. Harilal, S. Balasubramanian, and D. Weigel. 2009. A genetic defect caused by a triplet repeat expansion in Arabidopsis thaliana. Sci-ence 323:1060–1062.

The Bovine Genome Sequencing and Analysis Consortium, C.G. Elsik, R. L. Tellam, and K.C. Worley. 2009. The genome sequence of Taurine cattle: A window to ruminant biology and evolution. Science 324:522–528.

The Bovine hapMap Consortium. 2009. Genome-wide survey of SNP variation uncovers the genetic structure of cattle breeds. Science 324:528–532.

Touraev, A., B.P. Forster, and S.M. Jain (ed.). 2008. Advances in hap-loid production in higher plants. Springer Verlag, New York.

Townsend, J.A., D.A. Wright, R.J. Winfrey, F. Fu, M.L. Maeder, J.K. Joung, and D.F. Voytas. 2009. High-frequency modifi-cation of plant genes using engineered zinc-finger nucleases. Nature (29 April) doi:10.1038/nature07845.

Tsitsigiannis, D.I., S. Kunze, D.K. Willis, I. Feussner, and N.P. Keller. 2005. Aspergillus infection inhibits expression of pea-nut 13SHPODE-forming seed lipoxygenases. Mol. Plant Microbe Interact. 18:1081–1089.

Vollbrecht, E., K. Ahern, J. Schares, P. Deewatthanawong, R. Weeks, L. Xu, T. Kubinec, E. Unger-Wallace, M. Peto, M. Muszynski, J. Duvick, V. Brendel, and T. Brutnell. 2008. Properties of transposition and uses in regional mutagenesis of the Dissociation (Ds) transposable element. p. 23. 51st Maize Genetics Conf. abstract.

Wanzek, T. 2008. 2008. Biotech crops directly tied to higher yields. Genet. Eng. News 26(16), 15 Sept. 2008.

Weil, C.F. 2008. Mining and harnessing natural variation: A little MAFIX. Crop Sci. 48:2066–2073.

Xu, K., X. Xu, T. Fukao, P. Canlas, R. Maghirang-Rodriguez, S. Heuer, A.M. Ismail, J. Bailey-Serres, P.C. Ronald, and D.J. Mackill. 2006. Sub1A is an ethylene-response-factor-like gene that confers submergence tolerance to rice. Nature 442:705–708.

Ye, X., S. Al-Babili, A. Klöti, J. Zhang, P. Lucca, P. Beyer, and I. Potrykus. 2000. Engineering the provitamin A (beta-caro-tene) biosynthetic pathway into (carotenoid-free) rice endo-sperm. Science 287:303–305.


symposia

Efficiency in use of natural resources has been central to agri-cultural practice for over 10,000 yr. Ever since humans inter-

vened in natural ecosystems to gather food, there has been interest in raising the efficiency of agro-ecosystems. If efficiency is simply the level of output per unit of input, “eco-efficiency” targets this simple notion toward the production of food and fiber products relative to the ecological resources used as inputs, mainly land, water, nutrients, energy, or biological diversity. Such focus should not be considered in isolation of the critical human and economic dimensions of labor and capital nor ignoring outputs such as environmental loads on wider ecosystems—nutrient, salt, acid, or sediment losses to terres-trial, aquatic, or marine ecosystems, greenhouse gas emissions to the atmosphere—or other ecosystem services that might be positively or negatively influenced by agricultural practice. It is generally consid-ered that the term eco-efficiency was coined by the World Business Council for Sustainable Development around the 1992 Rio Earth Summit (WBCSD, 2000) to provide a concept for the engagement of the private sector. During the past decade the concept has been

Eco-efficient Agriculture: Concepts, Challenges, and Opportunities

Brian A. Keating,* Peter S. Carberry, Prem S. Bindraban, Senthold Asseng, Holger Meinke, and John Dixon

aBsTRaCTEco-efficiencyinthesimplestoftermsisaboutachieving more with less—more agriculturaloutputs,intermsofquantityandquality,forlessinputofland,water,nutrients,energy,labor,orcapital.Theconceptofeco-efficiencyencom-passes both the ecological and economicdimensions of sustainable agriculture. Socialand institutional dimensions of sustainability,while not explicitly captured in eco-efficiencymeasures,remaincriticalbarriersandopportu-nitiesonthepathwaytowardmoreeco-efficientagriculture. This paper explores the multidi-mensionality of the eco-efficiency concept asit applies to agriculture across diverse spatialandtemporalscales,fromcellularmetabolismsthrough to crops, farms, regions, and eco-systems. These dimensions of eco-efficiencyare integrated through the presentation andexploration of a framework that explores anefficiency frontierbetweenagriculturaloutputsand inputs, investment, or risk. The challengeforagricultureinthecomingdecadeswillbetoincreaseproductivityofagriculturallandsinlinewiththeincreasingdemandsforfoodandfiber.Achieving such eco-efficiency, while address-ingriskandvariability,willbeamajorchallengefor futureagriculture.Often, riskwillbeacriti-calissueinfluencingadoption;itneedsexplicitattentioninthediagnosisandinterventionstepstoward enhancing eco-efficiency. To ensurefood security, systems analysis and modelingapproaches, combined with farmer-focusedexperimentationandresourceassessment,willprovide the necessary robust approaches toraisetheeco-efficiencyofagriculturalsystems.

B.A. Keating, CSIRO Sustainable Agriculture Flagship, St Lucia, QLD, Australia; P.S. Carberry, CSIRO Sustainable Agriculture Flag-ship/APSRU, Toowoomba, QLD, Australia; P.S. Bindraban, ISRIC-World Soil Information, Wageningen University and Research Centre, Wageningen, The Netherlands; S. Asseng, CSIRO Sustainable Agricul-ture Flagship, Floreat, WA, Australia; H. Meinke, Centre for Crop Sys-tems Analysis, Wageningen University, Wageningen, The Netherlands; J. Dixon, Australian Centre for International Agricultural Research, ACT, Australia. Received 13 Oct. 2009. *Corresponding author ([email protected]).

Abbreviations: GM, genetically modified; NGO, nongovernmental organization; WUE, water use efficiency.




applied more directly to agriculture and specifically eco-effi-cient agriculture for the rural poor (CIAT, 2009).

Yield per unit land area is the simplest and most widely used eco-efficiency measure for field crops. However, there are inevitably multiple efficiency measures at play at the same time, such as water use efficiency (yield per unit of water used, e.g., rainfall, stored soil moisture, and/or irrigation), nutrient use efficiency (yield per unit nutri-ent uptake or nutrient supplied), radiation use efficiency (biomass produced per unit radiation intercepted), labor efficiency (production per unit labor invested), return on capital (profit as a fraction of capital invested), and so on.

Even within these simple ratios, there are multiple ways that efficiency can be measured. For example, nutrient use efficiencies can be defined in a number of different ways: crop yield per unit of nutrient applied (partial factor pro-ductivity); crop yield increase per unit of nutrient applied (agronomic efficiency); nutrient in harvested crop per unit of nutrient applied (partial nutrient budget); or increase in above-ground crop uptake per unit of nutrient applied (recovery efficiency). Additionally, nutrient use efficiency calculations may only consider nutrient inputs derived from chemical fertilizers or may include contributions from the mineralization of nutrients in soil organic matter, crop resi-dues, or manures and over several crop cycles. Such calcula-tions could also be undertaken in terms of economic return per cost invested to quantify the financial benefits to farm-ers of enhancing nutrient use efficiency.

Broadly, eco-efficiency can cover the interrelation-ships and trade-offs among a host of production, con-servation, economic, and social values at landscape scale (e.g., Groot et al., 2007). In this context, some efficiency measures can relate two or more system outputs, such as harvested product per unit nutrient lost by leaching or per unit of habitat loss or biodiversity impact or value per unit greenhouse gas emissions.

In this paper we first review the eco-efficiency concept as it applies to agriculture today, its multidimensionality, and its relevance over diverse spatial and temporal scales. Second, we examine contemporary food production challenges through the eco-efficiency lens. What role will efficiency of natu-ral resource use need to play in addressing emerging global challenges around food security for our growing population? Finally, we look for guidance for agricultural science. How can we better diagnose the nature of resource use inefficiency in the diverse agricultural systems of the world, and what agro-technical solutions are likely to have greatest benefit for human development and food security?

mULTiDimENsioNaLiTy oF ECo-EFFiCiENCy

Eco-efficiency is invariably influenced by multiple factors interacting on growth and development processes in nonlinear and nonadditive ways. In a classic paper on

“Resource Use Efficiency in Agriculture,” C.T. de Wit (1992) explores the interactive nature of different produc-tion factors in terms of Liebig’s Law of the Minimum, Liebscher’s Law of the Optimum, and Mitscherlich’s Law of Constant Activity. All these yield response functions reflect, ceteris paribus, decreasing returns to increases in the supply of one production factor. While these response curves differ in the way multiple growth factors interact to determine growth or yield, de Wit concluded that Lieb-scher’s Law of the Optimum best described the observed growth responses: “It may be concluded, with some reservations regarding the control of pests, diseases and weeds, that no produc-tion resource is used with any less efficiency and that most produc-tion resources are used more efficiently with increasing yield level due to further optimizing of growing conditions.”

While de Wit (1992) argues that the totality of resources are utilized most efficiently when their sup-plies are all close to yield-optimizing levels, the reality of a response curve for any single factor is that the highest increments in output are achieved for the first increments in inputs and efficiency declines thereafter. The phenom-enon of decreasing efficiencies with increasing inputs is well illustrated for yield response to N fertilization. At a global scale, N fertilizer use on cereals increased by sev-enfold between 1960 and 1995, cereal yields more than doubled, and the N fertilizer efficiency (partial factor productivity: cereal yields divided by N fertilizer inputs) declined from over 70 to around 25 kg grain per kg N (Tilman et al., 2001).

Eco-efficiency can also be examined at different spa-tial scales. At the cellular level, the efficiency with which solar radiation drives the transformation of atmospheric CO2 to carbohydrates can be expressed in terms of pho-tosynthetic rates, which in turn are influenced by ambient CO2 concentrations, climatic conditions (such as light and temperature), and genetic factors (such as photosynthetic metabolic pathways) as well as other growth factors (such as water or nutrient stress that directly influence the effec-tive functionality of the cellular photosynthetic appara-tus). At the crop level, dry matter production is frequently expressed as the product of radiation intercepted and “radiation use efficiency”—the latter being a key compo-nent of many crop simulation models—and an efficiency function that is influenced by other environmental stresses (Sinclair and Muchow, 1999). Also at the crop level, har-vested yield can be interpreted in terms of efficiency of water transpired, water lost via evapotranspiration, or water supplied as rain or irrigation (Sinclair et al., 1984). Likewise nutrient use efficiencies frequently express yield per unit of nutrient taken up or supplied (Baligar et al., 2001; Asseng et al., 2001).

At the farm level, eco-efficiency might be represented in terms as diverse as the food output per unit labor, the bio-diversity benefits provided by retention of natural habitat


by economic, social, and institutional factors. One of the most commonly analyzed gaps is the yield gap between typical farms and best farms or research stations (Even-son et al., 1996). Through successful agricultural research even greater efficiencies are continually sought for a given level of input (the third highest curve in Fig. 1). Generally such production or efficiency functions exhibit the dimin-ishing returns curve displayed in Fig. 1.

In most agricultural output–input relationships, how-ever, there is a probability distribution of responses driven by sources of variability, primarily climate but also due to the inherent diversity in biological systems and different management across farms and years. The higher the climate variability the more pronounced will be the risk dimension of any eco-efficiency enhancing strategy. One approach to exploring relationships between risks and returns has been the concept of “mean-variance (E–V) space” (Anderson et al., 1977; Parton and Carberry, 1995). For any system, a scatter plot of expected or mean economic returns against their associated variance over a range of input levels will elicit an efficiency frontier of outermost points where mean returns can be maximized for any given level of variance in returns. The statistical variance (or standard deviation) in returns is used as a surrogate for risk.

An eco-efficiency diagnosis framework, based on the return–risk approach, is presented in Fig. 2. This frame-work provides for the notion of an “efficiency frontier” whereby the return with existing knowledge and tech-nology is maximized for different levels of risk, defined as the variance of output. The curve in Fig. 2 joining points C–D–A is a styled example of such a frontier based on cur-rently known technologies and practices adapted to local circumstances. This curve shows returns (profit or yields) approaching a plateau while risks of increased investment (due to variability or some other measure of risk such as rates of crop failure, rates of economic loss, or in some cases level of environmental damage) continue to rise. Note that is it not generally possible to map input level and risk one-on-one because of system nonlinearity and thresholds.

Based on both the input–output space in Fig. 1 and the return–risk space in Fig. 2, we can explore three spe-cific pathways to address productivity improvement:

(i) Moving along the efficiency frontier (moving from C to D toward A) but with associated increase in inputs and riskiness;

(ii) Addressing system inefficiencies through best prac-tice for a certain level of input and risk (moving from B to D); and

(iii) Breakthrough technologies or practices to redefine a new efficiency frontier (from D to F onto a new frontier C–F).

These simple input–output and return–risk frame-works can be employed to diagnose the most promising

per unit food production, or the aggregate food output per unit water or fertilizer applied. Regional level analy-ses might target differences in the relationships between agricultural inputs and outputs associated with a diverse range of impacts on the natural resource base. At the global scale, the ultimate measure of eco-efficiency of agriculture emerges from the complex interrelationships among global food supplies to meet human needs, avoidance of land deg-radation to sustain long-term productivity, provision of ecosystem services beyond agriculture such as biodiversity conservation, healthy waters, and atmosphere.

Farming system eco-efficiency can vary with time. A farming system that is mining soil nutrient reserves or depreciating the productive capacity of soils through phys-ical or chemical degradation may appear highly efficient at the outset but progressively deteriorates as degradation intensifies. Agro-ecological systems are generally nonlin-ear and subject to relatively rapid change when thresholds are breached. Classic examples of such thresholds include:

· rising water tables or irrigation leading to salinization in dryland (McFarlane and George, 1992) or irri-gated systems, often long after shifts in water balance first occurred;

· acidification of soils due to nitrate leaching and cation removal in harvested products; this can reach thresh-olds where plants can no longer adapt, thus further intensifying degradation (Tang et al., 2000); and

· a farming system that depletes soil nutrients (De Jager et al., 2001) or exposes the soil to degradation via wind or water erosion may pass a soil health or fertil-ity threshold that quickly switches the system from profit to financial loss.

aN EFFiCiENCy FRoNTiER FRamEWoRKEffective application of the eco-efficiency concept

requires an understanding of the production functions that relate agricultural outputs to the level of resource and other inputs (Dillon, 1977). Figure 1 illustrates three production functions that relate production to inputs at any spatial or temporal scale. The lowest production func-tion depicts the current efficiencies observed on farms in a particular agro-ecological or farming system setting and may represent the performance of the best managed farms across a range of input usage. While efficiency is the ratio of the output achieved to the input applied, the slope of the function represents the marginal efficiency gain from moving along the production function. The second higher function represents the achievable efficien-cies through the deployment of the best known technol-ogies for that setting. This function is a styled example of such a frontier based on currently known technolo-gies and practices adapted to local circumstances. The gap between the observable farm efficiencies and those obtainable with known technologies is caused in the main


pathways for intervention into agricultural systems to improve their performance.

EXampLEs oF CURRENT ECo-EFFiCiENCiEs iN aGRiCULTURE

soil Fertility in africaIn Africa, there is compelling evidence that food pro-

duction systems are constrained by overwhelmingly low soil fertility (McCown et al., 1992; Giller et al., 2006; Sanginga and Woomer, 2009) and unless ways are found to relieve this constraint, eco-efficient use of other natural and human resources will remain low. This is consistent with de Wit’s (1992) conclusion that resource use efficiency is maximized when the level of all inputs is close to their optima and confirmed by Bindraban et al. (2008), who have analyzed temporal trends in crop yields in Africa.

There would appear to be two linked pathways for action immediately with soil fertility and smallholder agri-culture in Africa. One path is to move along the efficiency frontier (from C to D in Fig. 2) through farmers taking on additional risk in terms of nutrient inputs (from fertilizers, manures, legumes, or a combination of all three), with the prospect of higher long-term returns but also higher variability of returns. Another path is to identify agro-nomic practices that currently cause crop performance to fall below the technically feasible efficiency frontier (i.e., moving from B to D in Fig. 2).

In terms of increasing returns while taking on greater (but acceptable) risk, the work of Keating et al. (1991), McCown et al. (1992), Dimes et al. (2003), Robertson et al. (2005), and Twomlow et al. (2008) illustrate the pos-sibilities arising from small amounts of carefully targeted fertilizer inputs. Some important facts emerge from this set of interrelated studies conducted on African smallholder farms, supplemented with modeling analyses over the last

25 yr. First, the soil fertility constraint is often a dominant ecological factor constraining productivity at any given level of moisture availability. In Africa, N is almost uni-versally limiting and phosphorus and other nutrients are limiting in many situations. Within-farm variability is high (Giller et al., 2006) and many farms have only small areas of higher nutrient supply, usually associated with household areas and livestock containment yards.

Rainfall variability is often extreme, yet seemingly attractive responses to fertilizer are still possible. In particu-lar, there are high returns possible from small amounts of targeted fertilizer additions (Twomlow et al., 2008). Some gains in reducing risk associated with fertilizer inputs have been identified through the use of various forms of climate information, including seasonal forecasts (Keating et al., 1991; McCown et al., 1992), but a major hurdle remains in the lack of access to fertilizer by smallholder farmers. Poor agronomic practices (such as late sowing, weed com-petition, and suboptimal plant populations) reduce fertil-izer use efficiency. Nevertheless, significant gains in maize yield (50–80%) can be achieved from open pollinated vari-eties and small inputs of fertilizer by smallholder farmers (Twomlow et al., 2008). Further modest gains are possible using improved maize varieties (Bänziger et al., 2000). However the African “green revolution” will need to be first a revolution based on improved soil fertility, albeit sup-plemented and reinforced by new germplasm development and agronomy for improved water management.

Despite what appears to be a strongly attractive tech-nical opportunity to significantly raise yields on Afri-can smallholder farms, large-scale uptake and growth in fertilizer use remains painfully slow. Recently a major government intervention in Malawi provided support for fertilizer use and yields responded accordingly, but

Fig. 1. Production functions that relate agricultural outputs to the level of inputs for observed farm performance (-), for current best technologies (-), and for foreseen new technologies (--).

Fig. 2. A stylized return–risk framework for the diagnosis of system state and opportunities to enhance eco-efficiency. Points A, C, and D are representative points on the efficiency frontier for current best technologies (-) and point F is a specific point on the efficiency frontier for a hypothesized new technology (--). Point B represents a position below the efficiency frontier.


institutional and budgetary challenges around such inter-ventions do not confer confidence that they are sustain-able (Ricker-Gilbert et al., 2009). The small but growing role of private and nongovernmental organization (NGO) sector networks in input/output markets holds promise for breaking the cycle of low inputs, low outputs, and food insecurity (Rusike and Dimes, 2004). Government policy needs to evolve to be supportive of this emerging sec-tor and there is a role for the research and development community to consider how they will build the two-way information flows from research knowledge and technol-ogies to application in a local context.

While the technical challenges around soil fertility in African smallholder agriculture are reasonably well defined by the work referenced above and other reports (World Bank, 2008), the solutions may lie outside the farm household at the level of input/output markets and the institutions governing them. In developed country agriculture, strategies to enable farmers to better under-stand and manage risks (in particular climate-related risks) associated with input use have been shown to greatly assist in this transition (Hochman et al., 2009b). In developing country agriculture, particularly in Africa, this pathway is more complex than simply improved knowledge support-ing a farmer’s decision to take on higher risks (Meinke et al., 2006). In many situations, the input/output markets are poorly developed or distorted to the point that even very modest inputs that could be highly effective have not been possible (Rohrbach, 1998). The emergence of private sector and/or NGO sector markets for agri-inputs and -outputs holds great promise for African smallholder agriculture to break out of the vicious cycle of soil-fertil-ity-constrained low productivity.

Nitrogen Use in the United states and ChinaRecent U.S. experience demonstrates that signifi-

cant improvements in fertilizer use efficiency are possible. For U.S.–grown maize, the ratio of crop yield per unit of applied N fertilizer (i.e., the partial factor productivity for N fertilizer) has increased by 36%, from 42 kg kg-1 in 1980 to 57 kg kg-1 in 2000 (Cassman et al., 2002). The drivers of this improvement would appear to reside in the combination of continuing yield improvement through technological change and flat N fertilization rates since the concerns over environmental consequences of excess fertilizer use emerged in the 1980s. Such changes are clear evidence that agricultural research has continually created new efficiency frontiers (from D to F in Fig. 2).

In contrast in China, a problem of low eco-efficiency is generated through excess use of fertilizers, particularly N, on cereal crops, leading to heavy costs to the economy and environment (Liu and Diamond, 2005; Ju et al., 2009). The national average in N fertilizer use efficiency in China is around 20 kg kg-1 N (Zhu and Chen, 2002; Ju et al., 2009).

From 1977 to 2005, grain production in China increased by 71% (from 283 to 484 million tonnes). Over this same period, N fertilizer application increased by 271% (from 7.07 to 26.21 million tonnes) ( Ju et al., 2009) encouraged by strong policy support. This overuse of N fertilizer is fur-ther exacerbated by high levels of organic N inputs from the atmosphere, water supplies, animal manures, and legume sources. Nitrogen inputs from atmospheric and irrigation water sources in the range of 89 to 104 kg N ha-1 yr-1 have been measured in two study regions ( Ju et al., 2009). Here, inefficiencies in system performance have widened the gap between farmer practice and the attainable efficiency fron-tier (between B and D in Fig. 2).

The excess N applied to cereals in China may be in the order of 11.8M tonnes. This figure is based on an aver-age N rate of 250 kg N ha-1, a cereal cropping area of 104 M ha, and rates of overfertilization in the range of 30 to 60% (average 45% used in these estimates) ( Ju et al., 2009). The solution to the overuse of N fertilizer in China might be a reform of fertilizer pricing policies or other institutional arrangements (OECD, 2003; Zhu et al., 2006). This appears to be an area under active policy review with the recent cessation of price controls on fer-tilizers, despite retention of production subsidies, and the expanded efforts to improve fertilizer use efficiency (The Economist, 2008).

It is sobering to reflect on the thought that if this “excess” N fertilizer used in China was instead used in sub-Saharan Africa, it represents an average annual N fertiliza-tion rate of 68 kg N ha-1 over 174 million ha of cropland. If nutrients of this magnitude were available (and rebal-anced to address other nutrient needs), the studies refer-enced above (McCown et al., 1992; Twomlow et al., 2008) would suggest a potential doubling of cereal production in sub-Saharan Africa. In other words, fertilizer use of this order could lay a very solid foundation for Africa’s “green revolution.” Of course there is no direct link between over-fertilization in China and underfertilization in Africa, but the comparison places the African challenge in perspective. Modest fertilizer inputs, carefully targeted and matched with other soil enrichment strategies involving legumes and manure, are not beyond the realms of possibility for Africa.

Water Use in australiaIn many environments, water supply is a major source

of variability in crop yields (Ritchie, 1983). Total water use or total evapotranspiration by a crop can vary sub-stantially due to variable soil water and rainfall. It can also vary as a result of variation in crop transpiration result-ing from management such as nutrient supply (Shepherd et al., 1987) and sowing time (Connor et al., 1992) or from use of different species (Gregory et al., 1992) or cul-tivars (Richards and Townley-Smith, 1987). The seasonal water use of a crop consists of both crop transpiration and


soil evaporation (French and Schultz, 1984), with the lat-ter varying between 14% (Angus et al., 1980) and 75% (Cooper et al., 1987) of total water use. The ratio between grain yield and evapotranspiration, the generally accepted definition of water use efficiency (WUE) for grain pro-duction, can be an important parameter defining the pro-ductivity of crops in water-limited environments (Fischer and Turner, 1978; Tanner and Sinclair, 1983).

In the dryland wheat systems of southern Austra-lia, on-farm monitoring and systems simulation model-ing have been effectively deployed to benchmark farms and fields relative to a rainfall-driven efficiency frontier (Hochman et al., 2009a; Carberry et al., 2009). The uniqueness of these studies lies in their effort to directly measure accurately the soil water resource available to commercially-managed crops and relate this to both actual crop performance and management.

Hochman et al. (2009a) measured the actual WUE (in kg grain ha–1 mm–1) of 334 commercial wheat crops in Australia and compared their performance to the APSIM model (Keating et al., 2003). They found an average WUE of 15.2 kg grain ha–1 mm–1 for harvested crop yields while APSIM simulated a higher WUE of 16.9 kg grain ha–1 mm–1 for these same crops. The differential is attributed to factors influencing the actual crop that are not simu-lated by APSIM (e.g., weeds, pests, diseases, and impacts of severe weather). APSIM was used to explore if these crops could have achieved higher WUE through changed management, such as higher sowing rates, earlier sowing time, or high input levels. A maximum WUE of 21.4 kg grain ha–1 mm–1 is suggested as the potential WUE for this dataset of crops. Their study demonstrates opportunities for these farmers to improve WUE using two pathways of the stylized return–risk framework introduced earlier (Fig. 2). Clearly there is some opportunity to further close the average “yield gap” of 1.7 kg grain ha–1 mm–1 and move crop yields up onto the efficiency frontier by better con-trolling biotic stresses. And second, there are options for farmers to move along the efficiency frontier and increase WUE by 4.5 kg grain ha–1 mm–1 with a yield-maximizing strategy. In this latter case, each suggested intervention would require farmers to increase their investment risk either by increasing inputs (seed and/or fertilizer) or the chance of crop failure (sowing early increases the risk of frost damage during grain filling).

In fact, the study of Carberry et al. (2009) suggests that the better farmers in Australia are quite close to the efficiency frontier for crop yields and so are performing at their best for the level of investment risk they are willing to accept under current market circumstances. For over 700 commercial crops grown under a wide range of environ-mental and agronomic conditions from all cropping regions of Australia, APSIM simulations showed close agreement to measured commercial yields. This shows that the best

Australian farmers have a very low yield gap, and it appears that most of their crops are not seriously affected by yield-reducing factors such as weeds, pests, and diseases or poor agronomic management. This study suggested that the sup-ply of water and N can account for most of the variation in crop yield, while farmers are mostly controlling other yield-limiting factors such as weeds and diseases. If true, then the only pathways for these farmers to increase pro-duction are either to move along the efficiency frontier (and so increase their investment risk) or to adopt new technolo-gies that generate a new efficiency frontier.

paTHWays To impRoVE ECo-EFFiCiENCy

Narrowing yield GapsThe term “yield gap” has often been used to describe

the difference between actual yields recorded on farmer fields and the yields that are possible with known tech-nologies and practices (identified via farm demonstration plots or a combination of on-farm experiments and simu-lation modeling). Yield gaps are pertinent to this discus-sion on eco-efficiency because they are indicative of the eco-inefficiency that persists in different food production systems of the world. One well-known study of rice yield gaps can be found in Evenson et al. (1996), which brought together a series of assessments of country and regional level constraints (as well as priority setting) for South and East Asia. Wheat yield gaps and production constraints have been documented by Kosina et al. (2007).

In terms of the simple input–output and return–risk frameworks, yield gaps would generally represent the differ-ence between the current status and an improved situation (often the climate and genetic potential). They represent what is possible if farmers were operating on the eco-effi-ciency frontier and were insensitive to risk (point A in Fig. 2). Examples of yield gaps measured for maize cropping in Africa and for agricultural production more generally in various parts of the world are estimated to be in the 40 to 80% range (World Bank, 2008; Comprehensive Assessment of Water Management in Agriculture, 2007).

While yield gaps are generally large in Africa (and to a lesser degree South Asia), they have narrowed in other parts of the world such as China and in wheat and maize production areas of western Europe and North America (World Bank, 2008). The lack of inputs, in particular irri-gation and nutrients, is the conspicuous explanation for the continuing large yield gaps in Africa (World Bank, 2008).

The World Development Report (World Bank, 2008) notes that exploitable yield gaps are especially high in medium- to high-potential areas and countries (i.e., in the higher rainfall areas of Africa) but that closing the gap is a matter not just of transferring known technologies and practices to farmers, but involves “putting in place the


institutional structures—especially well-functioning input and output markets, access to finance, and ways to manage risks—that farmers need to adopt the technology” (World Bank, 2008).

When viewing these remarks of the World Develop-ment Report on institutional arrangements, our return–risk framework suggests to explicitly install institutional mechanisms to deal with risk management and reduction both in agro-technical as well as socio-economic terms when aiming to close yield gaps.

Transformational ChangeEven the best farmers in a developed country context

do not necessarily choose to operate at the point of maxi-mum returns (i.e., point A in Fig. 2) because of the greater exposure to downside risk associated with the higher lev-els of investment. Under such circumstances, the only real option for these farmers to increase productivity is to break away from the current efficiency frontier by creat-ing new production systems or practices that can increase returns for little added risk (i.e., moving from D to F in Fig. 2).

Transformational technologies that improve the effi-ciency of resource use will generate new frontiers of return relative to investment risk. Historically, the large step gains in agricultural productivity have resulted from an increased use of inorganic fertilizer inputs and the devel-opment of irrigation infrastructure on previously rain-fed cropping lands. Their introduction to local farming sys-tems created new efficiency frontiers, and typically the fol-lowing decades were needed for farmers to improve their practices in adapting irrigation and fertilizer technologies to approach these new productivity benchmarks (path-way B to D in Fig. 2). And as farmer practice improved, more sophisticated technologies could be introduced to continue the steady increases in the efficiency of nutri-ent and water use. For instance, fundamentally changing the way rice is grown can lead to drastic water savings. Among the promising new management practices are AWD (alternate wetting and drying, whereby the paddy is allowed to dry, but irrigation water is reapplied before water limitations start to impact on yields) and the aerobic rice systems, where specially developed rice varieties are grown in well-drained soil like dryland crops. This can save up to half of the normal water requirement, while with good management yields between 4 and 6 t ha-1 can be routinely achieved (Bouman et al., 2005; Bindra-ban et al., 2006). Other examples include biodegradable mulches to restrict soil water loss from evaporation or weeds (Shogren, 2000), nitrification inhibitors to reduce fertilizer losses from leaching or denitrification (Shoji et al., 2001), removing subsoil constraints such as hard pans (Wong and Asseng, 2007), or ameliorating subsoil acidity (Tang et al., 2003).

Plant and animal breeding has been a traditional route for the creation of new efficiency frontiers in agri-culture. Clearly, continued breeding effort is important for improved adaptation to current and future climates or enhanced resistance to biotic and abiotic stresses. Yet, for Africa, average yield gains of 60% are found with small inputs of fertilizer N used on open-pollinated variet-ies, compared to 20% gains from improved hybrid seeds (Twomlow et al., 2008). Both hybrid seed and small inputs of fertilizer N combine positively to raise the yield gains toward 80%, but these smallholder farming systems are most in need of strategies to improve soil fertility, not new plant varieties. Hence, breeding alone is unlikely to provide the transformational breakthrough required by the current low productivity, low eco-efficiency, and food insecure farming systems of the most food insecure part of the world, particularly in Africa (Twomlow et al., 2008). Hence, Meinke et al. (2009) call for “adaptation science” as a solution-oriented, scientific endeavor to facilitate adaptation actions. In contrast to conventional, disciplin-ary-based science, adaptation science analyses the prob-lems in a broadly participatory mode without a predefined disciplinary lens. Adaptation science differs from science for adaptation by testing alternative solutions (often via simulation models) and developing adaptation pathways or processes as opposed to generating more data.

Gene technologies are often viewed as the source of the next generation of breakthrough innovations in agriculture. It is true that the genetically modified (GM) cotton story can be viewed as a breakthrough story in the developed world and increasingly in the developing world. In reviewing the literature on GM crops, Brookes and Barfoot (2005) reported yield increases as high as 50% for GM insect-resistant cotton in India and 10% for GM herbicide-tolerant canola in Canada. There were also a number of examples of no yield benefits, notably for GM herbicide-tolerant soybean in the United States and Argentina (Bindraban et al., 2009a) or for GM insect-resistant cotton in Australia (Brookes and Barfoot, 2005). Nevertheless, the proponents of GM crops actively pro-mote such technology as yield frontier breaking and sci-ence should indeed be in pursuit of such outcomes.

ECo-EFFiCiENCy aND CURRENT CHaLLENGEs

Food securityFood security is defined as a state when “all people, at

all times, have physical and economic access to sufficient, safe and nutritious food for a healthy and active life” (FAO, 2008). FAO (2008) have estimated that 923M people were undernour-ished in 2007—75M more than in 2005—highlighting the immediate food security challenge. Both supply and demand side perspectives are important.


The World Development Report (World Bank, 2008) estimated that cereal production will have to increase by nearly 50% and meat production by 85% from 2000 to 2030 to meet projected increases in food demand. Look-ing forward to 2050, this food security challenge will grow with a world population increase of approximately 35% (6.7 billon in 2008 to 9 billion in 2050). If population growth is the only force acting, an increase in food pro-duction of 35% over the next 40 yr can be expected. Eco-nomic development trajectories (including changes in diet preferences toward greater protein intake) combined with population growth suggest an increase in food demand of the order of 75% between 2010 and 2050. Nonfood uses of agricultural products and lands (such as for bio-fuels, carbon bio-sequestration, and urban development) are also growing. The demand from agriculture to supply the required amounts of biomass for bio-energy heavily depends on desired blending targets and can put a claim on many hundreds of millions of hectares of land (Bindra-ban et al., 2009b). While uncertainty is high, these forces together suggest a likely 75 to 100% increase in demand for agricultural products over the next 40 yr.

Significant production increases have been achieved in past years (e.g., Angus, 2001), but they came via a com-bination of increased deployment of natural resources (land, water, nutrients, and energy) and human resources (labor and capital) as well as via technological advances (varieties and agronomic practices). Globally, average yields have climbed steadily for all major cereals, at least since the 1960s (World Bank, 2008). Worryingly, the yield growth rate of wheat and rice has declined markedly since 1995 (Dixon et al., 2009). Since 1961, around 78% of the increase in production has come from increases in yields, except in Africa where about 60% of the gains have come from expanding the area of cultivation (based on Pardey et al., 2007). The yield expansion pathway of Asia contrasts dramatically with the area expansion pathway of Africa over the last 40 yr (World Bank, 2008).

The yield and production increases of the last 50 yr have been achieved with a significant cost to the natural resource base (degraded soils and ecosystem impacts) as well as the global atmosphere, that is, 31% of global green-house gas emissions can be attributed to agriculture and forestry (including land clearing; IPCC, 2007).

How will the world achieve the production challenge of the next 40 to 50 yr? There appears to be little alterna-tive answer to this question other than via productivity gains—in other words, eco-efficiency gains in terms of the key natural resources of land, water, nutrients, and energy. Fischer et al. (2005) reported on the likely sources of growth in agricultural production over the 2000 to 2050 period and estimated only 7 to 12% of the projected growth in food production is likely to come from expan-sion in the areas of arable land. The dominant source of

production growth will need to come from intensification of agriculture, either through increased cropping intensity (13–15%) or, more significantly, yield increases (75–76%) (Fischer et al., 2005, 2007).

While yield gains dominate the future prospects for a secure world food system, the rates of gain in crop yields are slowing in developing countries (World Bank, 2008), although the same phenomenon is not apparent in U.S. maize yield data (Cassman et al., 2006).

Climate Variability and ChangeUnder the current climate, agricultural production is

already significantly affected by climate variability, espe-cially in semiarid regions. Changes in climate variability with future climate change are still uncertain (Nicholls and Alexander, 2007), but rainfall variability typically increases as mean annual rainfall decreases (Nicholls and Wong, 1990). Improving the management of climate variability will have immediate benefits to improving eco-efficiency and could be a positive step toward adapting to climate change if the future climate becomes more variable (Meinke et al., 2007). For example, in water-limited environments, seasonal rain-fall variability is one of the most important factors for fluctu-ations in agricultural production and risk. Perceptions about climatic risk and uncertainty of rainfall in the forthcoming season have led to the development of conservative, low-input management approaches, which aim to reduce the losses in poor rainfall seasons. However, such low-input approaches usually fail to capitalize on the upsides of climatic variability, that is, the good rainfall seasons (Meinke and Stone, 2005; Twomlow et al., 2008).

Crop models can assist to quantify the season- and site-specific outcomes of agricultural interventions (Mat-thews and Stephens, 2002; Whitbread et al., 2010) and, when integrated with long-term historical weather data, allow retrospective analysis of the potential value of cli-mate forecasts for a particular decision issue (Meinke and Stone, 2005; Hansen, 2002). Forecasting systems linked with crop simulation models have been employed to estab-lish optimized management strategies for improved risk management and enabled farmers to better tailor manage-ment decisions to the season and consequently improve eco-efficiency (Hammer et al., 1996; Hansen, 2005). Cur-rently, global efforts to provide a much improved, contin-uous weather and climate service for agriculture and other climate sensitive sectors are expected to provide “action-able climate knowledge” for better tactical and strategic management of crops and cropping systems at intersea-sonal, seasonal, and decadal time scales.

Climate has changed significantly over the last cen-tury and proactively planned adaptation action is now imperative (Howden et al., 2007; Meinke et al., 2009; Wassmann et al., 2009). Faced with such complexity and uncertainty, it would appear the best strategy is to work


with the well-established efficiency constraints and test the diagnosis and solution pathways for resilience and contin-ued relevance under a range of climate change scenarios.

CoNCLUsioNsThe term eco-efficiency has emerged relatively

recently but efforts to increase the level of desired outputs from inputs of natural resources and human endeavor have characterized agriculture evolution for over 10,000 yr. What is new is the magnitude of the efficiency challenge global food systems are facing. The agricultural revolution over the next 40 yr has to be the eco-efficiency revolution, with 50 to 100% increases in the efficiency with which scarce resources of land, water, nutrients, and energy are used. Importantly, this greater output and efficiency has to be achieved without further greenhouse gases emissions while maintaining or restoring land, water, biodiversity, and agro-ecosystems.

Eco-efficiency is a “multi-factor” issue that—as shown by C.T de Wit from first principles—is maximized when all production factors approach their optimum (Law of the Optimum). However, in the case of Africa and in terms of the journey toward such optima, it is clear the critical first steps lie in getting small improvements in soil fertility in place and these first increments in fertility can produce both efficiency and production gains.

The imperative needs to be accompanied with mitigat-ing measures to reduce risks associated with increased invest-ments. Agricultural research is the engine for developing the technologies and practices that can lift eco-efficiency while keeping risks low by lifting the efficiency frontier. Social and economic systems and institutions are often the necessary “gears” whereby improved technologies and practices can be embedded with constant or declining production risks. Food security involves both closing the yield gap (particu-larly in Africa) and breakthrough technologies and practices to raise yield potential and increase efficiency of resource use. In terms of how agricultural research approaches these dual challenges, there is a critical role for more effective “systems diagnosis” of constraints and opportunities for technological and policy intervention. This need is more important than ever, given the poor adoption of existing technologies, the multiple impacts of climate change, and the decreasing avail-ability of resources per person.

In this paper we have referred to examples where appropriately configured and validated simulation mod-els, in conjunction with field experimentation and farmer engagement, have proven very useful in the diagnosis phase of research and development aimed at raising eco-efficiency in farming systems. Such approaches can be used to explore the likely nature of single and multifactor response surfaces and to explore opportunities for inter-ventions including new farming practice, breeding strate-gies, and agricultural infrastructure and market policies.

REFERENCEsAnderson, J.R., J.L. Dillon, and J.B. Hardaker. l977. Agricultural

Decision Analysis. Iowa State Univ. Press, Ames, IA.Angus, J.F. 2001. Nitrogen supply and demand in Australian agri-

culture. Aust. J. Exp. Agric. 41:277–288.Angus, J.F., H.A. Nix, J.S. Russell, and J.E. Kruizinga. 1980.

Water use, growth and yield of wheat in a sub-tropical envi-ronment. Aust. J. Agric. Res. 31:873–886.

Asseng, S., N.C. Turner, and B.A. Keating. 2001. Analysis of water- and nitrogen-use efficiency of wheat in a Mediterra-nean climate. Plant Soil 233:127–143.

Baligar, V.C., N.K. Fageria, and Z.L. He. 2001. Nutrient use effi-ciency in plants. Commun. Soil Sci. Plant Anal. 32:921–950.

Bänziger, M., G.O. Edmeades, D. Beck, and M. Bellon. 2000. Breeding for drought and nitrogen stress tolerance in maize: From theory to practice. CIMMYT, Mexico City, Mexico.

Bindraban, P.S., A.C. Franke, D.O. Ferraro, C.M. Ghersa, L.A.P. Lotz, A. Nepomuceno, M.J.M. Smulders, and C.C.M. van de Wiel. 2009a. GM-related sustainability: Impacts, risks and opportunities of soy production in Latin America. Plant Res. Int. Rep. 259. Plant Research International, Wageningen, The Netherlands.

Bindraban, P.S., E. Bulte, and S. Conijn. 2009b. Can biofuels be sustainable by 2020? Agric. Syst. 101:197–199.

Bindraban, P.S., H. Löffler, and R. Rabbinge. 2008. How to close the ever widening gap of Africa’s agriculture. Int. J. Technol. Glob. 4:276–295.

Bindraban, P.S., H. Hengsdijk, W. Cao, Q. Shi, T.M. Thiyagara-jan, W. Van der Krogt, and I.P. Wardana. 2006. Transforming inundated rice cultivation. Water Resour. Dev. 22:87–100.

Bouman, B.A.M., S. Peng, A.R. Castañeda, and R.M. Visperas. 2005. Yield and water use of tropical aerobic rice systems. Agric. Water Manage. 74:87–105.

Brookes, G., and P. Barfoot. 2005. GM crops: The global eco-nomic and environmental impact: the first nine years 1996–2004. AgBioForum 8:187–196.

Carberry, P.S., Z. Hochman, J.R. Hunt, N.P. Dalgliesh, R.L. McCown, J.P.M. Whish, M.J. Robertson, M.A. Foale, P.L. Poulton, and H. van Rees. 2009. Re-inventing model-based decision support with Australian dryland farmers: 3. Rel-evance of APSIM to commercial crops. Crop Pasture Sci. 60:1044–1056.

Cassman, K.G., A. Dobermann, and D. Walters. 2002. Agroeco-systems, nitrogen-use efficiency, and nitrogen management. Ambio 31:132–140.

Cassman, K., V. Eidman, and E. Simpson. 2006. Convergence of energy and agriculture: Implications for research and policy. CAST Report. Council for Agricultural Science and Tech-nology, Ames, Iowa.

Comprehensive Assessment of Water Management in Agriculture. 2007. Water for food, water for life: A comprehensive assess-ment of water management. International Water Manage-ment Institute, London, UK.

CIAT. 2009. CIAT strategic directions. CIAT, Cali, Columbia.Connor, D.J., S. Theiveyanathan, and G.M. Rimmington. 1992.

Development, growth, water-use and yield of a spring and a winter wheat in response to time of sowing. Aust. J. Agric. Res. 43:493–516.

Cooper, P.J.M., P.J. Gregory, J.D.H. Keatinge, and S.C. Brown. 1987. Effects of fertilizer, variety and location on barley pro-duction under rainfed conditions in northern Syria. 2. Soil


water dynamics and crop water use. Field Crops Res. 16:67–84.De Jager, A., D. Onduru, M.S. van Wijk, J. Vlaming, and G.N.

Gachini. 2001. Assessing sustainability of low-external-input farm management systems with the nutrient monitoring approach: A case study in Kenya. Agric. Syst. 69:99–118.

de Wit, C.T. 1992. Resource use efficiency in agriculture. Agric. Syst. 40:125–151.

Dillon, J.L. 1977. An analysis of response in crop and livestock production (2nd edition), Pergamon Press, Oxford, UK.

Dimes, J., S. Twomlow, and P. Carberry. 2003. Application of APSIM in small holder farming systems in the semi-arid tropics. p. 85–99. In T. Bontkes and M. Wopereis (ed.) Deci-sion support tools for smallholder agriculture in sub-Saharan Africa: A practical guide. International Center for Soil Fer-tility and Agricultural Development (IFDC), Muscle Shoals, AL, USA, and Technical Centre for Agricultural and Rural Cooperation (CTA), Wageningen, The Netherlands.

Dixon, J., H.-J. Braun, P. Kosina, and J. Crouch (ed.). 2009. Wheat facts and futures – 2007. CIMMYT, Mexico City, Mexico.

Evenson, R.E., R.W. Herdt, and M. Hossain (ed.). 1996. Rice research in Asia: Progress and priorities. CABI Publ., Wall-ingford, UK.

FAO. 2008. The state of food insecurity in the developing world. FAO, Rome.

French, R.J., and J.E. Schultz. 1984. Water use efficiency of wheat in a Mediterranean-type environment. I. The relation between yield, water use and climate. Aust. J. Agric. Res. 35:743–764.

Fischer, G., M. Shah, F.N. Tubiello, and H. van Velhuizen. 2005. Socio-economic and climate change impacts on agriculture: An integrated assessment, 1990–2080. Philos. Trans. R. Soc. B Biol. Sci. 360:2067–2083.

Fischer, G., F.N. Tubiello, H. van Velthuizen, and D.A. Wiberg. 2007. Climate change impacts on irrigation water require-ments: Effects of mitigation, 1990–2080. Technol. Forecast. Soc. Change 74:1083–1107.

Fischer, R.A., and N.C. Turner. 1978. Plant productivity in the arid and semiarid zones. Annu. Rev. Plant Physiol. 29:277–317.

Giller, K.E., E.C. Rowe, N. de Ridder, and H. van Keulen. 2006. Resource use dynamics and interactions in the tropics: Scal-ing up in space and time. Agric. Syst. 88:8–27.

Gregory, P.J., D. Tennant, and R.K. Belford. 1992. Root and shoot growth, and water and light use efficiency of barley and wheat crops grown on a shallow duplex soil in a Mediterranean-type environment. Aust. J. Agric. Res. 43:555–573.

Groot, J.C.J., W.A.H. Rossing, A. Jellema, D.J. Stobbelaar, H. Renting, and M.K. van Ittersum. 2007. Exploring multi-scale trade-offs between nature conservation, agricultural profits and landscape quality—A methodology to support discussions on land-use perspectives. Agric. Ecosyst. Environ. 120:58–69.

Hammer, G.L., D.P. Holzworth, and R. Stone. 1996. The value of skill in seasonal climate forecasting to wheat crop manage-ment in a region with high climatic variability. Aust. J. Agric. Res. 47:717–737.

Hansen, J.W. 2002. Realizing the potential benefits of climate pre-diction to agriculture: Issues, approaches, challenges. Agric. Syst. 74:309–330.

Hansen, J.W. 2005. Integrating seasonal climate prediction and agricultural models for insights into agricultural practice. Philos. Trans. R. Soc. 360:2037–2047.

Hochman, Z., D.P. Holzworth, and J.R. Hunt. 2009a. Potential to improve on-farm wheat yield and WUE in Australia. Crop

Pasture Sci. 60:708–716.Hochman, Z., H. van Rees, P.S. Carberry, J.R. Hunt, R.L.

McCown, A. Gartmann, D. Holzworth, S. van Rees, N.P. Dalgliesh, W. Long, A.S. Peake, P.L. Poulton, and T. McClel-land. 2009b. Re-inventing model-based decision support with Australian dryland farmers. 4. Yield Prophet® helps farmers monitor and manage crops in a variable climate. Crop Pasture Sci. 60:1057–1070.

Howden, S.M., J.F. Soussana, F.N. Tubiello, N. Chhetri, M. Dun-lop, and H. Meinke. 2007. Adapting agriculture to climate change. Proc. Natl. Acad. Sci. USA 104:19691–19696.

IPCC. 2007. Intergovernmental Panel on Climate Change. Fourth Assessment Report. Available at http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_syr_spm.pdf (verified 23 Dec. 2009).

Ju, X.T., G.X. Xing, X.P. Chen, S.L. Zhang, L.J. Zhang, X.J. Liu, Z.L. Cui, B. Yin, P. Christie, Z.L. Zhu, and F.S. Zhang. 2009. Reducing environmental risk by improving N manage-ment in intensive Chinese agricultural systems. Proc. Natl. Acad. Sci. USA 106:3041–3046.

Keating, B.A., D.C. Godwin, and J.M. Watiki. 1991. Optimizing nitrogen inputs in response to climatic risk. p. 329–358. In Muchow and J.A. Bellamy (ed.) Climatic risk in crop produc-tion: Models and management for the Semiarid Tropics and Subtropics. CABI Publ., Wallingford, UK.

Keating, B.A., P.S. Carberry, G.L. Hammer, M.E. Probert, M.J. Robertson, D. Holzworth, N.I. Huth, J.N.G. Hargreaves, H. Meinke, Z. Hochman, G. McLean, K. Verburg, V. Snow, J.P. Dimes, M. Silburn, E. Wang, S. Brown, K.L. Bristow, S. Asseng, S. Chapman, R.L. McCown, D.M. Freebairn, and C.J. Smith. 2003. An overview of APSIM, a model designed for farming systems simulation. Eur. J. Agron. 18:267–288.

Kosina, P., M.P. Reynolds, J. Dixon, and A. Joshi. 2007. Stake-holder perception of wheat production constraints, capacity building needs, and research partnerships in developing coun-tries. Euphytica 157:475–483.

Liu, J., and J. Diamond. 2005. China’s environment in a global-izing world. Nature 435:1179–1186.

Matthews, R.B., and W. Stephens. 2002. Crop-soil simulation models: Applications in developing countries. CABI Publ., Wallingford, UK.

McCown, R.L., B.A. Keating, M.E. Probert, and R.K. Jones. 1992. Strategies for sustainable crop production in semi arid Africa. Outlook Agric. 21:21–31.

McFarlane, D.J., and R.J. George. 1992. Factors affecting dryland salinity in two wheat belt catchments in western Australia. Aust. J. Soil Res. 30:85–100.

Meinke, H., and R.C. Stone. 2005. Seasonal and inter-annual cli-mate forecasting: The new tool for increasing preparedness to climate variability and change in agricultural planning and operations. Clim. Change 70:221–253.

Meinke, H., R. Nelson, P. Kokic, R. Stone, R. Selvaraju, and W. Baethgen. 2006. Actionable climate knowledge – From analysis to synthesis. Clim. Res. 33:101–110.

Meinke, H., M.V.K. Sivakumar, R.P. Motha, and R. Nelson. 2007. Preface: Climate predictions for better agricultural risk management. Aust. J. Agric. Res. 58:935–938.

Meinke, H., S.M. Howden, P.C. Struik, R. Nelson, D. Rodriguez, and S.C. Chapman. 2009. Adaptation science for agricultural and natural resource management – Urgency and theoretical basis. Curr. Opin. Environ. Sustain. 1:69–76.

Nicholls, N., and L. Alexander. 2007. Has the climate become


more variable or extreme? Progress 1992–2006. Prog. Phys. Geogr. 31:77–87.

Nicholls, N., and K.K. Wong. 1990. Dependence of rainfall vari-ability on mean rainfall, latitude, and southern oscillation. J. Clim. 3:163–170.

OECD. 2003. China in the world economy: An OECD statistical and economic survey. OECD Publ., Paris, France.

Pardey, P., J. James, J. Alston, P. Glewwe, E. Binenbaum, T. Hur-ley, and S. Wood. 2007. Science, Technology and Skills. Background Paper for the World Development Report 2008. University of Minnesota, Minneapolis, MN.

Parton, K.A., and P.S. Carberry. 1995. Stochastic efficiency and mean-standard deviation analysis: Some critical issues. Aust. J. Agric. Res. 46:1487–1491.

Richards, R.A., and T.F. Townley-Smith. 1987. Variation in leaf area development and its effect on water use, yield and harvest index of droughted wheat. Aust. J. Agric. Res. 38:983–992.

Ricker-Gilbert, J., T.S. Jayne, and J.R. Black. 2009. Does subsidizing fertilizer increase yields? Evidence from Malawi. Agricultural & Applied Economics Association 2009 AAEA & ACCI Joint Annual Meeting, 26–29 July 2009, Milwaukee, WI. Available at http://ageconsearch.umn.edu/bitstream/49532/2/42422.pdf (verified 23 Dec. 2009).

Ritchie, J.T. 1983. Efficient water use in crop production: Discus-sion on the generality of relations between biomass produc-tion and evapotranspiration. p. 29–44. In H M Taylor, W.R. Jordan, and T.R Sinclair (ed.) Limitations to efficient water use in crop production. ASA, Madison, WI.

Robertson, M.J., W. Sakala, T. Benson, and Z. Shamudzarira. 2005. Simulating response of maize to previous velvet bean (Mucuna pruriens) crop and nitrogen fertilizer in Malawi. Field Crops Res. 91:91–105.

Rohrbach, D.D. 1998. Developing more practical fertility man-agement recommendations. p. 237–244. In S.R. Waddington, H.K. Murwira, J.D.T. Kumwenda, D. Hikwa, and F. Tagwira (ed.) Soil fertility research for maize-based farming systems in Malawi and Zimbabwe. The Soil Fertility Network/CIM-MYT, Harare, Zimbabwe

Rusike, J., and J.P. Dimes. 2004. Effecting change through private sector client services for smallholder farmers in Africa. In “New directions for a diverse planet” 4th International Crop Science Congress, 26 Sept.–1 Oct. 2004, Brisbane, Australia [CD-ROM]. Available at http://www.cropscience.org.au/icsc2004/symposia/4/6/997_rusikej.htm (verified 23 Dec. 2009).

Sanginga, N., and P.L. Woomer. 2009. Integrated soil fertility management in Africa: Principles, practices and develop-mental process. Tropical Soil Biology and Fertility Institute, CIAT, Nairobi, Kenya.

Shepherd, K.D., P.J.M. Cooper, A.Y. Allan, D.S.H. Drennan, and J.D.H. Keatinge. 1987. Growth, water use and yield of barley in Mediterranean-type environments. J. Agric. Sci. (Cam-bridge) 108:365–378.

Shogren, R.L. 2000. Biodegradable mulches from renewable resources. J. Sustain. Agric. 16:33–47.

Shoji, S., J. Delgado, A. Mosier, and Y. Miura. 2001. Use of con-trolled release fertilizers and nitrification inhibitors to increase

nitrogen use efficiency and to conserve air and water quality. Commun. Soil Sci. Plant Anal. 32:1051–1070.

Sinclair, T.R., and R.C. Muchow. 1999. Radiation use efficiency. Adv. Agron. 65:215–265.

Sinclair, T.R., C.B. Tanner, and J.M. Bennett. 1984. Water-use efficiency in crop production. Bioscience 34:36–40.

Tang, C., S. Asseng, E. Diatloff, and Z. Rengel. 2003. Modelling yield losses of aluminium-resistant and aluminium-sensitive wheat due to subsurface soil acidity: Effects of rainfall, liming and nitrogen application. Plant Soil 254:349–360.

Tang, C., C. Raphael, Z. Rengel, and J.W. Bowden. 2000. Under-standing subsoil acidification: Effect of nitrogen transforma-tion and nitrate leaching. Aust. J. Soil Res. 38:837–849.

Tanner, C.B., and T.R. Sinclair. 1983. Efficient water use in crop production: Research or re-search? p. 1–27. In H.M. Taylor, W.R. Jordan, and T.R. Sinclair (ed.) Limitations to efficient water use in crop production. ASA, Madison, WI.

The Economist. 2008. China ends fertilizer price controls. Avail-able at http://www.rsc.org/chemistryworld/Issues/2009/March/ChinaEndsFertiliserPriceControls.asp (verified 23 Dec. 2009).

Tilman, D., J. Fargione, B. Wolff, C. D’Antonio, A. Dobson, R. Howarth, D. Schindler, W.H. Schlesinger, D. Simberloff, and D. Swackhamer. 2001. Forecasting agriculturally driven global environmental change. Science 292:281–284.

Twomlow, S., D. Rohrbach, J. Dimes, J. Rusike, W. Mupangwa, B. Ncube, L. Hove, M. Moyo, N. Mashingaidze, and P. Mahposa. 2008. Micro-dosing as a pathway to Africa’s Green Revolution: Evidence from broad-scale on-farm trials. Nutr. Cycling Agroecosyst. 10.1007/s10705–008–9200–4.

Wassmann, R., S.V.K. Jagadish, K. Sumfleth, H. Pathak, G. Howell, A. Ismail, R. Serraj, E. Redona, R.K. Singh, and S. Heuer. 2009. Regional vulnerability of climate change impacts on Asian rice production and scope for adaptation. Adv. Agron. 102:91–133.

Whitbread, A., M. Robertson, P. Carberry, and J. Dimes. 2010. Applying farming systems simulation to the development of more sustainable smallholder farming systems in Southern Africa. Eur. J. Agron. 32:51–58.

World Bank. 2008. World Development Report 2008: Agricul-ture for Development. The World Bank, Washington, DC.

World Business Council for Sustainable Development. 2000. Eco-efficiency—Creating more value with less impact. WBCSD. Earthprint, Hertfordshire, UK. Available at http://www.wbcsd.org/web/publications/eco_efficiency_creating_more_value.pdf (verified 23 Dec. 2009).

Wong, M.T.F., and S. Asseng. 2007. Yield and environmental ben-efits of ameliorating subsoil constraints under variable rainfall in a Mediterranean environment. Plant Soil 297:29–42.

Zhu, Z.L., and D.L. Chen. 2002. Nitrogen fertilizer use in China – Contributions to food production, impacts on the environ-ment and best management strategies. Nutr. Cycling Agro-ecosyst. 63:117–127.

Zhu, Z.L., D. Norse, and B. Sun. 2006. Policy for reducing non-point source pollution from crop production in China. p. 135–299. China Env. Sci. Press, Beijing, China.


symposia

The 20th century witnessed a drastic change in global resource use. Between 1900 and 2000, there occurred an

increase in world population by a factor of 3.8, urban popula-tion by 12.8, industrial output by 35, energy use by 12.5, oil production by 300, water use by 9, irrigated area by 6.8, fertil-izer use by 342, fish catch by 65, organic chemical production by 1000, car ownership by 7750 times, and atmospheric CO2 by 30% (Ponting, 2007). World population of 6.8 billion in 2009 is projected to increase to 8.1 billion by 2030 and to 9.2 billion by 2050 (United Nations, 2008; Table 1). Almost all the future increase in population will occur in developing countries where soil and water resources are already under great stress. In 2006, food-insecure population of 854 million comprised 300 million in South Asia (SA), 220 million in East Asia, 200 million in sub-Saharan Africa (SSA), 50 million in Latin America and the Carib-bean (LAC), and 41 million in Near East and North Africa (FAO, 2005, 2008). By 2009, food-insecure population increased to 1020 million because of surge in prices of food staples since 2008 (FAO, 2009). Of this, 645 million food-insecure people are in the Asia/Pacific region, 265 million in SSA, 53 million in LAC, 43 million in Middle East/North Africa, and 16 million in the developed countries (FAO, 2009).

Cereal production will have to be increased from 2012 ´ 106 Mg in 2005 to 3009 ´ 106 Mg by 2050. The required increase

Enhancing Eco-efficiency in Agro-ecosystems through Soil Carbon Sequestration

R. Lal*

abstractGlobal cereal production must be increasedby~50%by2050.Cropyields insub-SaharanAfricaandSouthAsiahaveeitherstagnatedordeclinedsincethe1990sbecauseofthewide-spreaduseofextractivefarmingpracticesandproblemsofsoilandenvironmentaldegradation.Yield potential of improved varieties and elitegermplasmisnotrealizedbecauseofsoildeg-radation.Theconceptofeco-efficiencyimpliesefficient and sustainable use of resources inagronomic production and soil management.However, it is not enough to merely minimizetheenvironmentalimpact.Itisalsoimportanttomaximizeagronomicproductionwhileenhanc-ing ecosystem services. Most degraded anddepleted soils of agro-ecosystems contain alower soil organic carbon (SOC) pool than inthose under natural ecosystems. Thus, restor-ingtheSOCpool isessential to improvingsoilquality, increasingeco-efficiency,andenhanc-ing numerous ecosystem services. Increasingthe SOC pool in the root zone can enhanceagronomic production (kg grains ha−1 Mg C−1)attherateof200to300formaize(Zea maysL.),30 to60 forbean (Phaseolis vulgaris L.),20 to40forwheat(Triticum aestivumL.),20to50forsoybean [Glycine max (L.)Merr.],and20 to50forrice(Oryza sativaL.).Notallimprovedman-agementpracticesareapplicabletoallsoilandecological conditions. However, no-till farmingalong with application of crop residue mulch,manuring, legume-based complex rotations,andintegratednutrientmanagementshouldbeapplicableundermostconditions.Global foodinsecurity,affecting1.02billionpeoplein2009,can only be alleviated by improving soil qual-ity and eco-efficiency through restoration ofdegraded/depletedsoils.

Carbon Management and Sequestration Center, The Ohio State Univ., Columbus, OH 43210. Received 5 Jan. 2010. *Corresponding author ([email protected]).

Abbreviations: INM, integrated nutrient management; SOC, soil organic carbon; SA, South Asia; SSA, sub-Saharan Africa.




in global food production will have to be achieved despite the increasing temperatures, decreasing rainfall effective-ness, increasing frequency of extreme events, degrading soils, declining and polluting water resources, increasing demands on energy, and regressive change to more frag-ile and harsh environments. The Green Revolution tech-nology, growing input-responsive varieties on favorable soils under irrigated conditions, progressively increased global average grain yields of cereals to 1.77 Mg ha−1 in 1970, 2.16 Mg ha−1 in 1980, 2.75 Mg ha−1 in 1990, and 3.06 Mg ha−1 in 2000. Even if dietary preferences do not change, meeting the food demands simply because of the increase in world population will necessitate increasing average cereal yield to 3.60 Mg ha−1 (+18% compared with 2000) by 2025 and 4.30 Mg ha−1 (+41%) by 2050. With change in dietary preferences, toward a more animal-based diet, the world average cereal yields must be increased to 4.40 Mg ha−1 (+44%) by 2025 and to 6.0 Mg ha−1 (+96%) by 2050 (Wild, 2003). The objective of this article is to describe the importance of using the eco-efficiency con-cept in enhancing agronomic production while restoring degraded soils and improving the environment.

Data source anD methoDsThis article is based on collation, analyses, review,

and synthesis of the data published in the literature. Rel-evant articles containing data on agronomic yields under a range of farming techniques in relation to rainfall and soil properties were selected as appropriate examples of eco-efficient production systems. Being from diverse sources, data were converted to metric units. For example, data on crop yields were changed to kg ha−1 or Mg ha−1, and those on soil organic matter (SOM) were converted to soil organic carbon (SOC; 58% C in SOM) to % on mass basis, or to Mg ha−1 for specific depths if soil bulk density was known. All graphs included in this article are based on recalculations and redrawing of the secondary data thus generated. Regression equations were computed relating crop yields to SOC concentration or pool. Rather than a

comprehensive review, this article specifically cites some examples from diverse ecoregions and for eco-efficient systems, and the impact of SOC concentration and pool on agronomic yield and productivity. Readers are referred to other review articles (e.g., Bindraban et al., 2008; Wilkins, 2008; Keating et al., 2009) for additional details.

resource constraints to achieving Food security

Agriculturally suitable soils are limited, unequally dis-tributed among ecoregions/biomes, and most are already being used for agricultural production. The remaining soils are either located in ecologically sensitive ecoregions (e.g., tropical rainforest, peat lands) or are marginal (e.g., too shallow, stony, steep, dry, or wet) for agricultural land use. Consequently, per capita cropland area declined from 0.536 ha in 1900 to 0.216 ha in 2000, and is projected to decline to 0.182 ha by 2050. Similarly, per capita irrigated land area peaked at 0.046 ha in 1990, and is projected to decline to 0.034 ha by 2050 (Table 1). Approximately 7000 km3 of water is used annually in crop production, corresponding to about 2750 L−1 person−1 d−1 for the world population of 7 billion in 2010 and increasing at the rate of 6 million mo−1. Additional resources required by 2050 for medium-level population projection include 200 Mha of land and 1002 km3 of water per year for expanding crop-land area, and 80 Mha of land and 398 km3 of water per year for expanding grazing land (Falkenmark et al., 2009). Water scarcity is also closely linked with the energy cri-sis (Kahrl and Roland-Holst, 2008). Change in dietary preference to meat-based diet may drastically increase the demand for already scarce water resources (Table 2). Pro-duction of 100 kg of protein equivalent from different food sources requires 0.6 ha of land for ruminant meat, 0.36 ha for pork, 0.25 ha for grain legumes (pulses), and 0.10 ha for milk (Stehfest et al., 2009). Similarly, kilograms of grains (and kcal of energy) required to produce 1 kcal of protein is 21 kg (57 kcal) for lamb, 13 (40) for beef cattle, 11 (39) for eggs, 5.9 (14) for swine, 3.8 (10) for turkey, 2.3 (4) for boil-ers, and 0.7 (14) for poultry (Pimentel and Pimentel, 2003).

Soil, water, and nutrients are also required for meet-ing the projected demands for biofuels. Conversion of tropical rainforest and peatlands for soybean [Glycine max (L.) Merr.] or oil palm (Elaeis guineensis Jacq.) production can drastically deplete the ecosystem C pool, and create a long-term ecosystem C debt (Farigone et al., 2008). Because corn (Zea mays L.)–based ethanol production has escalated food prices (Rosegrant, 2008), using biomass for producing cellulosic ethanol (second generation) is being widely considered. Biomass required to meet the projected demand for bioethanol in the United States is about 1 billion Mg yr−1 (Somerville, 2006). However, residue retention on croplands is essential to soil and water conservation, C sequestration, nutrient cycling, and

Table 1. World population and per capita cropland and irri-gated land area (calculated from Postel, 1999; FAO, 2002; Lal, 2007; Stewart, 2009; Falkenmark et al., 2009; Bruin-sma, 2009).

Year Population Cropland area Irrigated land area

Total Per capita Total Per capita billions Mha ha Mha ha

1900 1.5 805 0.536 40 0.027

1950 2.5 1170 0.468 94 0.038

1970 3.9 1300 0.333 169 0.043

1980 4.5 1333 0.296 211 0.047

1990 5.2 1380 0.265 239 0.046

2000 6.3 1360 0.216 277 0.044

2030 8.1 1648 0.203 300 0.037

2050 9.2 1673 0.182 318 0.034


entire life-cycle to a level at least in line with Earth’s esti-mated carrying capacity” (WBCSD, 1997; Day, 1998). The strategy is to produce more food without using more land, water, and energy-based inputs. The key to enhancing and preserving ecosystem services of soil and natural resources lies in enhancing eco-efficiency and their resilience to both natural and anthropogenic perturbations. Soils must have the capacity to restore their physical, chemical, and biologi-cal quality through enhancing restorative processes.

causes of Low agronomic production in Developing countries

Long-term and widespread use of extractive farm-ing practices, leading to negative nutrient (N, P, K, Ca, Mg, Zn, Cu) and C budgets in soil and ecosystems and the attendant decline in soil quality, are among important causes of low and declining agronomic production in SA and SSA. Decline in soil quality has numerous feedbacks which exacerbate the process of soil degradation, such as those caused by accelerated erosion, increase in intensity and frequency of drought, and high risks of desertification. Soil degradation is also a reason for inadequate human nutri-tion (Lal, 2009b). Consequently, there is a strong decline in the rainfall effectiveness (i.e., the percentage of rain uti-lized by plants as evapotranspiration; also called green water) and the use efficiency of inputs (e.g., fertilizer, irrigation, energy). The data in Fig. 1 from Machakos, Kenya, indicate the adverse impact of soil degradation on regressive decline in grain yield of maize. The grain yield of 1.3 Mg ha−1 in 1987 (Year 21) declined to <400 kg ha−1 after 1994 (Year 28) even with good rainfall of 816 mm (1994) and 1020 mm (1997). Decline in grain yield of maize even in sea-sons with high rainfall (Fig. 2) may be because of severe soil degradation and reduction in the rainfall effectiveness (kg ha−1 yr−1). The yield potential of improved varieties and elite germplasm is not realized when soils are degraded and crops are grown under sub-agronomic/edaphic conditions.

other ecosystem services. Use of crop residues and other agricultural coproducts can exacerbate the already severe problems of soil degradation and desertification (Bai et al., 2008). Despite the increases in food demand, the rate of annual increase in crop yields has either been negative or declining, especially in rainfed agriculture in SA and SSA. Crop yields stagnated in the range of 750 kg ha−1 and 1000 kg ha−1 in SSA between 1960 and 2005 (Hazell and Wood, 2008). In India, the grain yield of pigeonpea [Cajanus cajan (L.) Millsp.] decreased by 0.5% yr−1 between 1986 and 2006. The yields of chickpea (Cicer arietinum L.) and groundnut (Arachis hypogaea L.) increased at the rate of 0.7 and 1.1% yr−1, respectively (Singh et al., 2009). The yield of groundnut in India declined from 1030 kg ha−1 in 1992–1996 to 1020 kg ha−1 in 2002–2006 (Singh et al., 2009). The problem of yield stagnation or decline is even more severe in SSA (Table 3). Over the 30-yr period end-ing in 2005, yields of most crops in SSA hardly increased, at the rate of merely <1% yr−1 (Table 3).

In view of the increasing population and scarcity of natural resources, there are three options for meeting the growing demand for food, fuel, fodder, and other agri-cultural products. One, loosening or breaking the grip of agrarian stagnation on rural communities in SSA, SA, and elsewhere in developing countries by replacing extractive farming practices with scientifically proven technologies. It is important to recognize that agro-ecosystems are sus-tainable in the long term only if the outputs of all com-ponents harvested are balanced by inputs into the system (Lal, 2009a). The negative nutrients and carbon budgets of agro-ecosystems must be changed to positive balances to set in motion soil restoration trends. Productivity and ecosys-tem services of degraded/desertified soils must be restored. Soil restoration also increases C sink capacity which must be filled for increasing ability of ecosystems toward adap-tation and mitigation of climate change. Furthermore, C sink capacity of terrestrial ecosystems is decreasing (Can-adell et al., 2007), probably because of the increasing prob-lem of land degradation and desertification. Two, soil and other natural resources must be managed to enhance their resilience (Walker and Salt, 2006). Three, production from agro-ecosystems must be increased on the basis of per unit area and input of external resources (fertilizers, irrigation, energy) by improving eco-efficiency of production sys-tems. The term eco-efficiency was first proposed by the World Business Council for Sustainable Development (WBCSD) in 1992. It implies creating more goods and services while using fewer resources and creating less waste and pollution (WBCSD, 1997). The 1992 Earth Summit endorsed eco-efficiency to implement Agenda 21. The WBCSD defines eco-efficiency as a strategy to produce “competitively priced goods and services that satisfy human needs and bring quality of life while progressively reducing environmen-tal impacts of goods and resource intensity throughout the

Table 2. Water requirements per kilogram of different agricul-tural products (Clay, 2004).

ProductWater requirement

by areaWater requirement

by weight

106 L ha−1 103 L kg−1

Potato (Solanum tuberosum L.)

3.50–6.25 0.5–1.5

Wheat 4.50–6.50 0.9–2.0

Rice 5.0–9.5 1.9–5.0

Sorghum NA† 1.1–1.8

Soybeans 4.50–8.25 1.1–2.0

Sugarcane 10.0–15.0 1.5–3.0

Chicken NA 3.5–5.7Cotton (Gossypium hirsutum L.)

5.5–9.5 7.0–29.0

Beef NA 15.0–70.0

Shrimp 10.0–100.0 1.0–300.0†NA, not available.


Heinrich and Rusike (2003) reported sorghum [Sorghum bicolor (L.) Moench] grain yield of 330 kg ha−1 with tra-ditional crop varieties and no input, 440 kg ha−1 (+33%) with improved crop varieties and no input, and 740 kg ha−1

(+124%) with improved varieties and recommended agro-nomic management. Similar experiments at IITA in Nige-ria (Lal, 1987) and CIMMYT in Mexico (Govaerts et al., 2009) have shown that optimal soil management (conserva-tion agriculture with mulch cover, and integrated nutrient management [INM]) is essential to achieving the agro-nomic potential of improved varieties. Genetic engineering can also help in utilizing water (Somerville and Briscoe, 2001), but it is not a substitute for good soil quality.

There is a wide range of indicators that can be used to assess the impact of technological options on quality of soil and the environment (Bastida et al., 2008), and evaluating differences in sustainability efficiency (Van Passel et al., 2007). Metabolic quotient (qCO2), defined as respiration to microbial biomass ratio, an indicator of the soil biological quality, is strongly influenced by the SOC concentration–pool and its quality. An optimal level of the SOC pool, which enhances soil biodiversity, is also sustained by eco-agriculture (Scherr and McNeely, 2008).

Table 3. Grain yields at rainfed crops in Africa (recalculated from Singh et al., 2009).

Region CropGrain yield Increase

over 30 yr1971–1975 2001–2005––––––––– kg ha−1 ––––––––– % yr−1

West Africa Millet 600 770 0.9

Sorghum 690 900 1.0

Maize 860 1230 1.4

Central Africa Millet 610 560 −0.3

Sorghum 660 880 1.1

Maize 790 930 0.6

East Africa Millet 1170 1380 0.6

Sorghum 800 1000 0.8

Southern Africa Maize 1810 2840 1.9

Sorghum 1270 1930 1.7

Figure 1. Decline in maize grain yield with continuous cultivation with subsistence farming in the Machakos and Makueni districts of Kenya (redrawn from Singh et al., 2009).

Figure 2. Effect of rainfall on maize grain yield under subsistence farming in the Machakos and Makueni districts of Kenya (redrawn from Singh et al., 2009).


The threshold level of SOC concentrations in the root zone for most upland soils in the tropics (e.g., Oxisols, Ultisols, Alfisols) is ~1.1% (Aune and Lal, 1997).

eco-efficiency of agro-ecosystemsSoil resources of good agronomic quality are limited,

yet essential for the production of food, feed, fiber, and fuel. There is also a scarcity of water and energy needed for producing the required amount of biomass. Thus, the con-cept of eco-efficiency is important to producing more and more from less and less. Eco-efficiency is related to both “ecology” and “economy,” and denotes both efficient and sustainable use of resources in farm production and land management (Wilkins, 2008). Eco-efficiency is increased by those farming systems that increase agronomic production by using less resources through reduction in losses of input, and sustaining and enhancing the production potential of land. Yet it is not enough to develop agricultural practices that merely minimize the adverse environmental impact. Because of the increasing population and rising standards of living, it is essential to develop those agricultural practices that maximize agricultural production while also enhanc-ing ecosystem services (Firbank, 2009). Considering the vast amount of capital required for improving agronomic production in developing countries (Schmidhuber et al., 2009) in a changing climate (Schmidhuber and Tubiello, 2007), feeding of 9.2 billion people by 2050 (Evans, 2009) necessitates identification, development, validation, and use of eco-efficient agro-ecosystems.

There exists a strong relationship between agronomic production and the SOC pool, especially in low-input agri-culture (none or low rate of fertilizer input). An optimal level of the SOC pool is an essential determinant of soil quality because of its positive impact on (i) soil structure and aggregation, (ii) water retention, (iii) nutrient reten-tion, (iv) biotic activity including the microbial biomass, (v) erosion control, (vi) nonpoint-source pollution abatement, (vii) sedimentation reduction and control of hypoxia, (viii) C sequestration, (ix) increase in use efficiency of input, and (x) increase in biomass production. Increase in aggregation and available water capacity are among important benefits of SOC (Emerson, 1995; Huntington, 2003).

More (1994) assessed the yield response of wheat (Triti-cum aestivum L.) and rice (Oryza sativa L.) to increase in the SOC pool in the root zone of a Vertisol in Central India. These and other data show that positive effects on crop yields are generally more at lower than at higher magnitude of the SOC pool, and at lower than higher level of external inputs such as fertilizers and other soil amendments. The data in Fig. 3 from Mandan, ND, show that yield response of wheat to increase in the SOC pool is more at low than at high level of N input. The data in Fig. 4 from Austra-lia clearly demonstrate that yield of wheat decreased with reduction in the SOC pool and increased with increase in

the SOC pool. The data in Fig. 5 for a Russian Chernozem show a linear increase in crop yield with increase in the SOC concentration. There occurred a logarithmic increase in maize grain yield in Thailand with increase in the SOC concentration (Fig. 6). The slope of regression equations of these graphs relating crop yields to SOC concentration–pools in Fig. 3 to 6 indicates increase in the yield per unit increment in the SOC amount. In this regard, the data in Fig. 4 from Australia indicate the decline in yield of wheat with depletion of the SOC pool and increase in wheat yield with accretion of the SOC pool. Wheat grain yield of ~2.75 Mg ha−1 was obtained with a steady-state level of the SOC pool. Synthesis of data from several experiments show that increase of the SOC pool in the root zone by 1 Mg C ha−1 yr−1 can cause increase in grain yield (kg ha−1 Mg−1 C) of food crops in a developing country by 200 to 300 for maize,

Figure 3. Response of wheat grown in Mandan, ND, to soil organic carbon pool at 30-cm depth for three rates of nitrogen application (redrawn from Bauer and Black, 1994).


20 to 40 for wheat, 20 to 50 for rice, 80 to 140 for sorghum, 30 to 70 for millet [Pennisetum glaucum (L.) R. Br.], 30 to 60 for bean (Phaseolus vulgaris L.), and 20 to 50 for soybeans (Lal, 2006a).

Most agro-ecosystems contain lower SOC pools than their natural counterparts (Lal, 2004a) because of land mis-use, soil mismanagement, and the attendant depletion of the SOC pool. Restoration of the SOC pool to above the criti-cal/threshold level is essential to enhancing soil quality and improving agronomic productivity (Lal, 2006a). Therefore,

the benefits of increase in SOC concentration and pool are high for depleted soils of the developing countries (Lal, 2006a). As a corollary, the detrimental effect of the loss of the SOC pool on agronomic productivity would be higher in coarse-textured than in fine-textured soils. Eco-effi-ciency of agro-ecosystems can be enhanced through adop-tion of farming–cropping systems that observe the basic laws of sustainable soil management (Lal, 2009b), and those that create positive C and nutrient budgets in the root zone. Sustainability of the rice–wheat system of SA also depends

Figure 4. Effects of changes in soil organic carbon (DSOC) pool in the root zone on grain yield of wheat in Australia (redrawn and recalculated from Farquharson et al., 2003).

Figure 5. Effects of soil organic carbon concentration in the root zone on agronomic production for a Russian Chernozem (recalculated and redrawn from Ganzhara, 1998).


on management of soil quality (Duxbury, 2002), and adopt-ing water-saving technologies such as growing aerobic rice (Bouman et al., 2007). The goal is to adapt agriculture to climate change (Howden et al., 2007; Batisti and Naylor, 2009), through restoration of soil quality by improving the quantity and quality of the SOC pool (Lal, 2004a). Eco-efficient systems, with numerous co-benefits and enhanced ecosystem services, will be the drivers of change in global agriculture (Hazell and Wood, 2008).

eco-efficient production systems for advancing Global Food security

The eco-efficiency concept, although scale-neutral and useful in enhancing the profit margin and environ-ment quality of large-scale commercial farming in North America/Western Europe and Australia, is especially rel-evant to improving production of resource-poor and small landholders of SSA and SA. The strategy is to minimize soil erosion, conserve water in the root zone, recycle plant nutrients, create positive budgets of C and plant nutrients, optimize soil temperature and moisture regimes, and mini-mize losses of water and nutrients from the ecosystem. The goal is to build on the indigenous systems (e.g., Ngoro in Tanzania [Malley et al., 2004]; half-moon soil and water conservation systems in Burkina Faso [Zougmoré et al., 2003]; use of manuring and biosolids in SSA [Pieri, 1992; Gicheru et al., 2004; Kapkiyai et al., 1999], among others). Enhancement of the SOC pool (Nandwa, 2001; Batino and Buerkert, 2001) is essential to sustaining agronomic production in depleted and degraded soils of SSA, SA, and elsewhere in the tropics. The beneficial impacts of long-term manuring (Kanchikerimath and Singh, 2001; Gong et al., 2009), legume cover cropping (Venkateswarlu et al., 2007), conservation agriculture (no-till, mulch farming, INM, and complex rotations) (Lal, 1976, 1981; Diaz-Zorita

et al., 1999, 2002; Diaz-Zorita and Grosso, 2000; Govaerts et al., 2009), and biodegradable mulches (Shogren, 2000) have been proven under diverse agro-ecosystems. Recy-cling crop residues as mulch is important to maintaining soil quality (Wilhelm et al., 2004; Powell and Unger, 1998), even when there are competing demands (biofuel) for this precious resource. The SOC pool can also be enhanced by complex farming systems including agroforestry (Zhukov et al., 2002). Furthermore, the impact of these production systems can be quantified through use of soil/land quality indicators (Bindraban et al., 2008; Bastida et al., 2008) based on key soil properties. A quantitative approach of assessing the productivity and ecological contributions (Dalsgaard and Oficial, 1997) is relevant to small-holder farmers of the tropics. Measuring sustainable efficiency (Van Passel et al., 2007) through application of life cycle assessments (Naray-anawamy et al., 2005) is an important tool to measure eco-efficiency of a range of agro-ecosystems (Lal, 2004b).

Among several examples of eco-efficient production systems, two described below are relevant to a wide range of soils of the tropics and subtropics. An 18-yr experiment conducted on a Humic Nitisol (Kikuyu Red Clay) under maize–bean in East African Highlands provides impor-tant data on eco-efficient production systems. Kapkiyai et al. (1999) reported from the long-term study in Kenya that total crop yield of maize and beans ranged from 1.4 Mg ha−1 yr−1 in traditional systems without external input to 6.0 Mg ha−1 yr−1 when stover was retained as mulch along with application of fertilizers and manure. The SOC pool to 15-cm depth increased from 23.6 Mg ha−1 in traditional systems to 28.7 Mg ha−1 in improved management, with average SOC sequestration rate of 280 kg ha−1 yr−1 for the 18-yr period. Another long-term experiment on integrated watershed management was conducted on Vertisols in central India for which Wani

Figure 6. Effects of soil organic carbon concentration in the root zone on grain yield of maize grown in northeastern Thailand (redrawn from Petchawee and Chaitep, 1995).


et al. (2003a, 2009) reported the data from a 30-yr study (Fig. 7). Improved systems of soil/crop/water management produced a 30-yr average grain yield of 5.1 Mg ha−1 com-pared with 1.1 Mg ha−1 with the traditional system (Fig. 7a). There was a decline in the ratio of grain yield with improved to traditional system over time (Fig. 7b) because the productivity of the improved system is reaching the maximum potential yield. The ecological potential of this ecoregion is 7 Mg ha−1 yr−1, indicating scope for addi-tional improvement in agronomic production. The rate of SOC sequestration with the improved system was 330 kg ha−1 yr−1, similar to that of the study in Kenya. A large gap (1–3 Mg ha−1 in India, Thailand, Vietnam, and Kenya) that exists in on-farm vis-à-vis on-station (research) yields (Table 4) can be bridged by the adoption of those technol-ogies that enhance eco-efficiency, increase the SOC pool, improve soil quality, conserve water in the root zone, and create positive C and nutrient budgets.

climate change mitigation by eco-efficient agro-ecosystems

World grain production is approximately 2500 million Mg annually. More than 54% of this is produced in devel-oping countries, and the rate of increase must be higher in developing than developed economies, in the range of 1.5 to 2.25% per annum. These high rates of increase in grain production have to be realized despite the severe problems of soil degradation, especially those caused by strong depletion of the SOC pool. Enhancement of the SOC pool is essential to improving soil quality, and increasing the eco-efficiency of production systems. Increasing the SOC pool is a major challenge (Schlesinger, 1999), especially in developing coun-tries where crop yields are low, climate is harsh, and there is a scarcity of water and nutrients (Lal, 2009c). Yet, with adoption of recommended management practices (RMPs), high rates of SOC sequestration have been reported. For an irrigated Vertisol in Central Mexico, Follett et al. (2005) reported a sequestration rate of 1.0 to 1.9 Mg C ha−1 yr−1. Follett and colleagues also reported a significant correlation

Figure 7. Changes in grain yields of sorghum and pigeonpea grown on Vertisols in central India under (a) traditional and improved systems of management, and (b) ratio of grain yield under improved to traditional systems of management (redrawn from Wani et al., 2008).


between the aboveground crop residue C produced and the amount of SOC sequestered. For sodic soils in northern India, recalculation of the data on the effects of agroforestry practices on the soil C pool indicated a rate of increase of 2 to 3 Mg C ha−1 yr−1 (Garg, 1998). These and other data (Lal, 2001) show that high rates of SOC sequestration can be achieved with restoration of degraded soils and adoption of RMPs. If that were the case, food grain production in developing countries can be increased by 24 to 39 (32 ± 11) million Mg yr−1 through improving soil quality by restoring the depleted SOC pool at the rate of 1 Mg C ha−1 yr−1 (Lal, 2006a). Similar to food grains, improvements in soil quality through SOC sequestration can also enhance production of roots and tubers, such as cassava (Manihot esculenta Crantz), yam (Dioscorea rotundata Poir.), sweetpotato [Ipomoea batatas (L.) Lam.], and taro [Colocasia esculenta (L.) Schott] by as much as 7 to 11 × 106 Mg yr−1 in developing countries (Lal, 2006b). Restoration of the SOC pool would involve widespread adoption of eco-efficient production systems based on mulch farming, retention and recycling of crop residues, and use of manure and other biosolids. These production systems also have the capacity to offset anthropogenic emission of CO2 by 0.6 to 1.2 Pg C yr−1 in croplands (Lal, 2004a) and 0.4 to 0.7 Pg C yr−1 through desertification control (Lal et al., 1999). Technical potential of carbon sequestration through reclama-tion of salt-affected soils is 0.4 to 1.0 Pg C yr−1 (Lal, 2010).

concLusionsThe schematic in Fig. 8 shows basic principles of

managing soil properties and processes for enhancing

Table 4. Potential and farmer yield under rainfed conditions (recalculated from Rockström et al., 2007).

Country CropYield

Potential:FarmerPotential Farmer––––– Mg ha−1 –––––

India

Soybean 2.19 0.94 2.3

Groundnut 2.69 1.06 2.5

Sorghum (summer) 3.50 1.19 2.9

Sorghum (winter) 1.44 0.63 2.3

Pearl millet 1.94 0.69 2.8

Pigeonpea 1.50 0.56 2.7

Chickpea 1.82 0.75 2.4

Northeastern Thailand

Soybean 1.94 1.19 1.6

Groundnut 1.69 1.31 1.3

Maize 4.69 2.38 2.0

Sunflower 1.56 1.44 1.1

Vietnam

Soybean 2.19 1.31 1.7

Groundnut 3.81 1.56 2.4

Maize 5.19 3.31 1.6

Kenya

Maize 4.25 1.25 3.4

Figure 8. Managing soil properties and processes for enhancing eco-efficiency in production systems.


eco-efficiency of production systems. Three components to be judiciously managed are C, H2O, and nutrients. The strategy of C sequestration in soils is by enhancing for-mation of stable microaggregates, translocation of C deep into the subsoil, and formation of recalcitrant substances through enhancement of protective mechanisms (e.g., physical, chemical, and biological), and manage the cou-pled cycling of C with H2O, N, P, and S. The objective of management is to create positive soil and ecosystem C budgets through conservation agriculture, cover cropping, mulch farming, agroforestry techniques, and use of bio-char and other amendments. Similar to C, there is a strong need to conserve and recycle water and enhance its use efficiency. Understanding and managing the hydrologi-cal cycle is needed at the landscape and watershed scales to maximize the green water component and minimize losses by runoff and evaporation. Use of drip subirrigation, drought-tolerant crops, and remote-sensing techniques to predict the onset of drought are among the modern inno-vations. Soils depleted of their nutrient reserves must be managed for creating positive budgets through adoption of the INM approaches. It is also important to apply both macro- (N, P, K, Ca, Mg) and micro- (Zn, Cu, Mo, Fe) nutrients. Using biofertilizers and nano-enhanced materi-als can reduce losses while improving the use efficiency of external inputs. The choice of eco-efficient technologies must be based on two criteria: (i) minimize the adverse environmental impact, and (ii) maximize the agronomic production. With the world population projected to reach 9.2 billion by 2050, it is not enough to merely minimize the environmental impact. Agronomic production must also be increased, for which improvement of the SOC pool is an important determinant. As Charles Mackay (1814–1889) stated, “In nature nothing dies. From each sad remnant of decay, some forms of life arise….”

referencesAune, J.B., and R. Lal. 1997. Agricultural productivity in the

tropics and critical limits of properties of Oxisols, Ultisols and Alfisols. Trop. Agric. 74:96–103.

Bai, Z.G., D.L. Dent, L. Olsson, and M.E. Schaepman. 2008. Proxy global assessment of land degradation. Soil Use Man-age. 24:223–234.

Bastida, F., A. Zsolnay, T. Hernandez, and C. Garcia. 2008. Past, present and future of soil quality indices: A biological per-spective. Geoderma 147:159–171.

Batino, A., and A. Buerkert. 2001. Soil organic carbon manage-ment for sustainable land use in Sudano-Sahelian West Africa. Nutr. Cycling Agroecosyst. 61:131–142.

Batisti, D.S., and R.L. Naylor. 2009. Historical warnings of future food insecurity with unprecedented seasonal heat. Science 323:240–244.

Bauer, A., and A.L. Black. 1994. Quantification of the effect of soil organic matter contents on soil productivity. Soil Sci. Soc. Am. J. 58:185–193.

Bindraban, P.S., H. Löffler, and R. Rabbinge. 2008. How to close the ever widening gap of Africa’s Agriculture. Int. J. Technol.

Glob. 4:276–295.Bouman, B.A.M., R.M. Lampayan, and T.P. Toung. 2007. Water

management in irrigated rice. IRRI, Los Baños, Philippines.Bruinsma, J. 2009. The resource outlook to 2050: By how much

do land, water and crop yields need to increase by 2050. In Expert Meeting on How to Feed the World by 2050, Rome, Italy. 24–26 June 2009. FAO, Rome.

Canadell, J.G., D.E. Pataki, and L.F. Piteeka (ed.) 2007. Terrestrial ecosystems in a changing world. Springer-Verlag, Berlin.

Clay, J. 2004. World agriculture and the environment; a com-modity-by-commodity guide to impacts and practices. Island Press, Washington, DC.

Dalsgaard, J.P.T., and R.T. Oficial. 1997. A quantitative approach for assessing the productive performance and ecological con-tributions of small holder farms. Agric. Syst. 55:503–533.

Day, R.M. 1998. Beyond ecoefficiency: Sustainability as a driver for innovation. World Resources Inst., Washington, DC.

Diaz-Zorita, M., D.E. Buschiazzo, and N. Peinemann. 1999. Soil organic matter and wheat productivity in the semi-arid Argentina Pampas. Agron. J. 91:276–279.

Diaz-Zorita, M., G.A. Duarte, and G.H. Grove. 2002. A review of no-till systems and soil management for sustainable crop pro-duction in the sub-humid and semi-arid Pampas of Argentina. Soil Tillage Res. 65:1–18.

Diaz-Zorita, M., and G.A. Grosso. 2000. Effects of soil texture, organic matter and wheat productivity in the semi-arid Argentina Pampas. Soil Tillage Res. 54:121–126.

Duxbury, J.M. 2002. Long-term yield trends in the rice-wheat cropping system: Results from experiments in Northwest India. J. Crop Prod. 3:27–32.

Emerson, W.W. 1995. Water retention, organic carbon and soil texture. Aust. J. Soil Res. 33:241–251.

Evans, A. 2009. The feeding of the nine billion: Global food secu-rity for the 21st century. Chatham House, London.

Falkenmark, M., J. Rockstrom, and L. Karlberg. 2009. Present and future water requirements for feeding humanity. Food Sec. 1:59–69.

FAO. 2002. FAOSTAT statistics. FAO, Rome.FAO. 2005. The state of food-insecurity in the world: Key to

achieving the millennium development goals. FAO, Rome.FAO. 2008. The state of food insecurity in the world. FAO, Rome.FAO. 2009. Food security and agricultural mitigation in develop-

ing countries: Options for capturing synergies. FAO, Rome.Farigone, J., J. Hill, D. Tilman, S. Polasky, and P. Hawthorne.

2008. Land cleansing and biofuel carbon debt. Science 319:1235–1238.

Farquharson, R.J., G.D. Schwenke, and J.D. Mullen. 2003. Should we manage soil organic carbon in Vertisols in the northern grains region of Australia. Aust. J. Exp. Agric. 43:261–270.

Firbank, L. 2009. It’s not enough to develop agriculture that mini-mizes environmental impact. Int. J. Agric. Sust. 7:151–152.

Follett, R.F., J.G. Castellanos, and E.D. Buenger. 2005. Carbon dynamics and sequestration in an irrigated Vertisol in Central Mexico. Soil Tillage Res. 83:148–158.

Ganzhara, N.F. 1998. Humus, soil properties and yield. Eurasian Soil Sci. 31:738–745.

Garg, V.K. 1998. Interaction of tree crops with a sodic soil envi-ronment: Potential for rehabilitation of degraded environ-ments. Land Degrad. Dev. 9:81–93.

Gicheru, P., C. Gachene, J. Mbuvi, and E. Mare. 2004. Effects of soil management practices and tillage systems in surface soil water conservation and crust formation on a sandy loam in


semi-arid Kenya. Soil Tillage Res. 74:173–184.Gong, W., X.-Y. Yan, J.-Y. Wang, T.-X. Hu, and Y.-B. Gong.

2009. Long-term manuring and fertilizer effects on soil organic carbon pools under a wheat-maize cropping system in North China Plain. Plant Soil 314:67–76.

Govaerts, B., K.D. Sayre, B. Goudeseune, P. De Corte, K. Lichter, L. Dendooven, and J. Deckers. 2009. Conservation agricul-ture as a sustainable option for central Mexican Highlands. Soil Tillage Res. 103:222–230.

Hazell, P., and S. Wood. 2008. Drivers of change in global agri-culture. Philos. Trans. R. Soc. London Ser. B 363:495–515.

Heinrich, G.M., and J. Rusike. 2003. Increasing productivity final report SM.PIR1.2. SADC/ICRISAT Sorghum and Millet Improve-ment Progr. Final Rev. and Reporting Workshop: Achievements and Future Directions for Sorghum and Millet R&D in Southern Africa, Bulawayo, Zimbabwe. 25–26 Nov. 2003.

Howden, S.M., J.-F. Soussana, F.N. Tubiello, N. Chhetri, N. Dunlop, and H. Meike. 2007. Adapting agriculture to climate change. Proc. Natl. Acad. Sci. USA 104:19691–19696.

Huntington, T.G. 2003. Available water capacity and soil organic matter. In R. Lal (ed.) Encyclopedia of soil sciences. Vol. 1. CRC Press, Boca Raton, FL.

Kahrl, F., and D. Roland-Holst. 2008. China’s water–energy nexus. Water Policy 10:51–65.

Kanchikerimath, M., and D. Singh. 2001. Soil organic matter and biological properties after 26 years of maize-wheat-cowpea cropping as affected by manure and fertilization in a Cam-bisol in semi-arid region of India. Agric. Ecosyst. Environ. 86:155–162.

Kapkiyai, J.J., N.K. Karanja, J.N. Qureshi, P.C. Smithson, and P.L. Woomer. 1999. Soil organic matter and nutrient dynam-ics in a Kenyon Nitisol under long-term fertilizer and organic input management. Soil Biol. Biochem. 31:1773–1782.

Keating, B., P. Carberry, P. Bindraban, S. Asseng, and H. Meike. 2009. Eco-efficient agriculture: Concepts, challenges and opportunities. Background paper to the CGIIAR 2009 Sci-ence Forum Workshop, Wageningen, Netherlands. 14–17 June 2009.

Lal, R. 1976. No-tillage effects on soil properties under different crops in Western Nigeria. J. Sustain. Agric. 11:135–140.

Lal, R. 1981. Soil erosion problems on Alfisols in Western Nige-ria. VI. Effects of erosion on experimental plots. Geoderma 25:215–230.

Lal, R. 1987. Managing soils of Sub-Saharan Africa. Science 236:1069–1076.

Lal, R. 2001. World cropland soils as source and sink for atmo-spheric carbon. Adv. Agron. 71:145–191.

Lal, R. 2004a. Soil carbon sequestration impacts on global climate change and food security. Science 304:1623–1627.

Lal, R. 2004b. Carbon emission from farm operations. Environ. Int. 30:981–990.

Lal, R. 2006a. Enhancing crop yields in developing countries through restoration of the soil organic carbon pool in agricul-tural lands. Land Degrad. Dev. 17:197–209.

Lal, R. 2006b. Managing soils for feeding a global population of 10 billion. J. Sci. Food Agric. 86:2273–2284.

Lal, R. 2007. Anthropogenic influences on world soils and impli-cations to global food security. Adv. Agron. 93:69–93.

Lal, R. 2009a. Soil degradation as a reason for inadequate human nutrition. Food Sec. 1:45–57.

Lal, R. 2009b. Ten tenets of soil management. J. Soil Water Con-serv. 64:20A–21A.

Lal, R. 2009c. Challenges and opportunities in soil organic matter research. Eur. J. Soil Sci. 60:158–169.

Lal, R. 2010. Carbon sequestration in saline soils. J. Soil Salinity Water Qual. 1(1&2):30–40.

Lal, R., H.M. Hassan, and J. Dumanski. 1999. Desertification con-trol to sequester carbon and mitigate the greenhouse effect. p. 83–151. In N.J. Rosenberg et al. (ed.) Carbon sequestration in soils: Science, monitoring and beyond. Battelle Press, Colum-bus, OH.

Malley, Z.J.U., B. Kayombo, T.J. Willcocks, and P.W. Mtakwa. 2004. Ngoro: An indigenous, sustainable and profitable soil, water and nutrient conservation systems in Tanzania for slop-ing land. Soil Tillage Res. 77:47–58.

More, S.D. 1994. Effect of farm wastes and organic manures on soil properties, nutrient availability and yield of rice-wheat grown in Sodic Vertisol. J. Indian Soc. Soil Sci. 42:253–256.

Nandwa, S.M. 2001. Soil and organic carbon management for sustainable productivity of cropping and agroforestry systems in Eastern and Southern Africa. Nutr. Cycling Agroecosyst. 61:143–158.

Narayanawamy, V., R. Van Berkel, J. Altham, and M. McGregor. 2005. Application of life cycle assessment to enhance eco-effi-ciency of grains supply chains. p. 1–16. In Proc. Aust. Conf. on Life Cycle Assessment, 4th, Sydney, Australia. 23–25 Feb. 2005.

Petchawee, S., and W. Chaitep. 1995. Organic matter manage-ment for sustainable agriculture. p. 21–26. In R.D.B. LeFroy et al. (ed.) Organic matter management in upland systems in Thailand. ACIAR, Canberra.

Pieri, C.M.M.G. 1992. Fertility of soil: A future of farming in West African savannah. Springer Verlag, Berlin.

Pimentel, D., and M. Pimentel. 2003. Sustainability of meat-based and plant-based diets and the environment. Am. J. Sustain. 78:6605–6638.

Ponting, C. 2007. A new green history of the world: The envi-ronment and collapse of great civilizations. Penguin Books, New York.

Postel, S. 1999. Pillar of sand: Can the irrigation miracle last? Nor-ton, New York.

Powell, J.M., and P.W. Unger. 1998. Alternatives to crop residues for sustaining agricultural productivity and natural resource conservation. J. Sustain. Agric. 11:59–84.

Rockström, J., N. Hatibu, T. Owens, and S.P. Wani. 2007. Manag-ing water in rainfed agriculture. p. 315–348. In D. Molden (ed.) Water for food, water for life: A comprehensive assessment of water management in agriculture. IWMI/Earthscan, London.

Rosegrant, M. 2008. Biofuels and grain prices: Impacts and policy responses. 7 May 2008. IFPRI, Washington, DC.

Scherr, S., and J.A. McNeely. 2008. Biodiversity conservation and agricultural sustainability: Towards a new paradigm of ‘ecoagriculture’ landscapes. Phil. Trans. R. Soc. London B 363:477–494.

Schlesinger, W.H. 1999. Carbon sequestration in soils. Science 284:2095.

Schmidhuber, J., J. Bruinsma, and G. Boedeker. 2009. Capi-tal requirements for agriculture in developing countries to 2050. In Expert Meeting on How to Feed the World by 2050, Rome, Italy. 24–26 June 2009. FAO, Rome.

Schmidhuber, J., and F.N. Tubiello. 2007. Global food security under climate change. Proc. Natl. Acad. Sci. USA 104:19703–19708.

Shogren, R.L. 2000. Biodegradable mulches from renewable resources. J. Sustain. Agric. 16:33–47.

Singh, P., P.K. Aggarwal, V.S. Bhatia, M.V.R. Murty, M. Pala,


T. Owens. B. Benli, K.P.C. Rao, and S.P. Wani. 2009. Yield gap analysis: Modeling of achievable yields at farm level. p. 81–123. In S.P. Wani et al. (ed.) Rainfed agriculture: Unlock-ing the potential. CAB Int., Wallingford, UK.

Somerville, C. 2006. The billion for biofuel vision. Science 312:1277.

Somerville, C., and J. Briscoe. 2001. Genetic engineering and water. Science 299:2217.

Stehfest, E., L. Bouwman, D.P. van Vuuren, M.G.J. den Elzen, B. Eickhout, and P. Kabat. 2009. Climate benefits of changing diet. Clim. Change doi 10.1007/s10584-008-9534-6.

Stewart, B.A. 2009. Advances in water science management and conservation. p. 301–316. In W.G. Pond et al. (ed.) Adequate food for all: Culture, science and technology of food in the 21st century. CRC Press, Boca Raton, FL.

United Nations. 2008. World population prospects: The 2008 revision, population database. United Nations Population Division, New York.

Van Passel, S., F. Nevens, E. Mathija, and G. Van Huylenbroeck. 2007. Measuring farm sustainability and explaining differ-ences in sustainability efficiency. Ecol. Econ. 62:149–161.

Venkateswarlu, B., Ch. Srinivasrao, G. Ramesh, S. Venkates-warlu, and J.C. Katyal. 2007. Effects of long-term cover crop incorporation on soil organic carbon, microbial biomass, nutrient build up and grain yield of sorghum/sunflower under rainfed conditions. Soil Use Manage. 23:100–107.

Walker, B., and D. Salt. 2006. Resilience thinking: Sustaining ecosystems and people in a changing world. Island Press,

Washington, DC.Wani, S.P., P. Pathak, L.S. Jangawad, H. Eswaran, and P. Singh.

2003a. Improved management of Vertisols in the semi-arid tropics for increased productivity and soil carbon sequestra-tion. Soil Use Manage. 19:217–222.

Wani, S.P., T.K. Sreedevi, J. Rockström, and Y.S. Ramakrishna. 2009. Rainfed agriculture—Past trends and future prospects. p. 1–35. In S.P. Wani et al. (ed.) Rainfed agriculture: Unlock-ing the potential. CAB Int., Wallingford, UK.

WBCSD. 1997. Exploring sustainable development. Global sce-narios 2000–2050. World Business Counc. for Sustainable Dev., Geneva.

Wild, A. 2003. Soils, land and food: Managing the land during the 21st century. Cambridge Univ. Press, Cambridge, UK.

Wilhelm, W.W., J.M.F. Johnson, J.L. Hatfield, W.B. Voorhees, and D.R. Linden. 2004. Crop and soil productivity response to corn residue removal: A literature review. Agron. J. 96:1–17.

Wilkins, R.J. 2008. Eco-efficient approaches to land management: A case for increased integration of crop and animal produc-tion systems. Phil. Trans. Roy. Soc. Lond. 363:517–525.

Zhukov, A.I., L.V. Sorokina, and V.V. Mosalvela. 2002. Soil organic carbon as affect by afforestation with Eucalyptus and Pinus in the Cerrado region of Brazilian forestry. Ecol. Man-age. 166:285–294.

Zougmoré, R., Z. Zida, and N.F. Kambou. 2003. Role of nutri-ent amendments in the success of half moon in soil and water conservation practice in semi-arid Burkina Faso. Soil Tillage Res. 71:143–149.


symposia

Ensuring the security of food systems is one of the greatest challenges facing humanity today and will require “every

ounce of our collective ingenuity to innovate and apply inte-grated and holistic ways of working with nature and natural resources” (El-Ashry, 2002, p. 7). Key drivers such as climate, human population growth, recent commodity price increases, world market volatility, loss of land to non-food production, and the degradation of natural resources and concomitant impairment of ecosystem services are likely to intensify the challenge of food security into the future (Millenium Ecosystem Assessment, 2005; Alcamo et al., 2005; Schmidhuber and Tubiello, 2007; Koning et al., 2008; FAO, 2008).

Agricultural eco-efficiency is promoted as a means of increas-ing primary production and improving food security ( Jansen, 2000). The dominant means for increased eco-efficiency is more intensive use of economic and environmental resources (Gregory et al., 2002). Wilkins (2008) notes that the two ways of achieving intensification include altering the management of individual crop and livestock (and fisheries) enterprises, and altering the land-use system itself. As used here, intensification refers to the integration

More than Eco-efficiency is Required to Improve Food Security

S. E. Park,* S. M. Howden, S. J. Crimp, D. S. Gaydon, S. J. Attwood, and P. N. Kokic

aBsTRaCTAgricultural eco-efficiency is promoted as ameans of increasing agricultural productionand improving the security of food systems inresponse to climate change. The rationale isthat economic and environmental resourceswillbeusedmoreefficiently,enablingincreasedamountsoffoodtobeproducedfromthesameamountorfewerinputs.Weused(i)aquantitativeliterature analysis to examine current usage oftheeco-efficiencyconcepttoassessstrategiesaimedat improvingfoodsecurityunderclimatechange, and (ii) a wheat (Triticum aestivum L.)simulationexperimenttoconsiderpossibletrad-eoffsbetweeneconomicbenefitsofagriculturalintensification,environmentalperformance,andsocialimpacts.Twoissueswerehighlightedfromthis.First,therelationshipbetweeneconomicandenvironmental outcomes is regularly assumed,leadingtopotentiallyerroneousconclusionsandunintended outcomes. Second, the lack of anyconsideration for thesocialdimensionsof foodsecurityignoresvariabilityinincomesgeneratedfromagriculturalproduction,andthepotentialforreducedquantitiesoffoodtobeproducedasarational response tomaximizinggrossmargins.Wesuggesttheeco-efficiencyconceptexplicitlyincludesocialaswellaseconomicandenviron-mentalcriteriaifitistoavoidpoorratesofuptakeofeco-efficiencytechnologies,thepromotionofpractices that reduce theeffectivenessof hun-ger-reduction efforts, and unintended environ-mentaldegradation.

S.E. Park, S.M. Howden, S.J. Crimp, and P.N. Kokic, CSIRO Climate Adaptation Flagship, GPO Box 284, Canberra, ACT 2601, Australia; D.S. Gaydon, CSIRO Sustainable Ecosystems, 306 Carmody Rd., St Lucia, QLD 4067, Australia; S.J. Attwood, Australian Centre for Sus-tainable Catchments, Univ. of Southern Queensland, Toowoomba, QLD 4350, Australia. Received 2 Oct. 2009. *Corresponding author ([email protected]).

Abbreviations: APSIM, Agricultural Production Systems Simulator; PAWC, plant-available water capacity.




of livestock and crop production to increase recycling of resources and minimize resource loss; and second, to the use of technologies that directly reduce dependence on nonrenewable resources or improve the efficiency with which they are utilized (e.g., precision agriculture). A third method of achieving agricultural intensification may focus on modifying the regional agricultural system to optimize production for current climatic conditions. In this context, the term agricultural intensification is not lim-ited to production systems characterized by high inputs of capital or labor, or heavy reliance on technologies sourced externally to the system, such as pesticides and chemi-cal fertilizers. Rather, it extends the definition to include increased production resulting from the spatial and tem-poral concentration of sustainably sourced resources from within the production system.

In all three of the above examples of intensification, the aim is to move the system toward the production effi-ciency frontier as discussed by Keating et al. (2010). How-ever, increasing amounts of food production has failed to significantly reduce global levels of chronic undernutri-tion and the absolute number of hungry people over the past few decades (FAO, 2006), due to the lack of a simulta-neous consideration of the availability (encompassing both the quantity and quality of food), accessibility, stability of food supplies, and their utilization (FAO, 2000). Collec-tively these four key dimensions of food security relate to the availability of sufficient quantities of food of appropri-ate quality, accessed by individuals at all times and utilized through adequate diet, clean water, sanitation, and health care to reach a state of nutritional well-being where all physiological needs are met (Garrett, 1995; FAO, 1996). When viewed in this way, nonfood inputs of an economic, environmental, and social nature are shown to be integral to food security.

Ongoing anthropogenic global warming may have both positive and negative impacts on all four aspects of food security (Schmidhuber and Tubiello, 2007). For example, enhanced concentrations of atmospheric CO2, higher temperatures, and changes in the patterns of pre-cipitation suggest changes in potential yield ( Jones and Thornton, 2003; Parry et al., 2004; Liu et al., 2008; Lobell et al., 2008). While some yield benefits may be predicted for the short term, projected increases in cli-matic variability and the volatility of extreme weather events, and resultant increases in the range of crop pests and diseases, may negate many gains and disrupt the sta-bility of food supply, particularly in developing countries (Rosenzweig and Parry, 1994). Changes in the quality of production (Manderscheid et al., 1995; Kimball et al., 2001) and increased production and energy requirements postharvest (Gleadow et al., 2009) will further challenge the availability of a stable supply of food.

Maintaining the parity of consumer purchasing power offers potential to improve food security. However, the resultant complexity of the interactions between the socioeconomic and environmental drivers on the capacity to buy food is likely to result in geographically varying outcomes under climate change (Liu et al., 2008). As a further aspect of food security, accessing sufficient clean water to enable effective utilization of the nutritional value of food is likely to be further challenged in regions where current access to supplies of good-quality water are already poor and projections of precipitation sug-gest ongoing reductions (Stige et al., 2006). In addition, projections of an increase in the incidences of food- and water-borne diseases (IPCC, 2007) suggest further reduc-tions in the potential for the nutritional value of food to be fully utilized by those presently most vulnerable. While this brief summary highlights the varying impacts of cli-mate change on food security, it is generally considered that negative effects will dominate already food-insecure livelihoods in the low latitudes (Stern, 2007).

Despite the recognition that numerous social drivers underpin the capacity for food security, the inclusion of an assessment of social outcomes within the eco-efficiency concept has remained on the periphery for over a decade (Schmidheiny and The Business Council for Sustainable Development, 1992; Zoebl, 1996; Jansen, 2000), with lit-tle progress made to formalize it within a broader assess-ment of efficiency. The aim of this paper is therefore to provide a coherent qualitative and quantitative argument for the simultaneous consideration of economic, environ-mental, and social outcomes resulting from the implemen-tation of technologies and practices. As one of the most significant challenges facing global society today, we use the context of developing and implementing sustainable adaptation response strategies to improve food security under a changing climate, to illustrate the imperative for an explicit triple-bottom-line approach to assessing effi-ciency. We do this qualitatively and quantitatively by first reviewing the current usage of the term eco-efficiency in the scientific literature domain to assess the extent to which the concept is being applied to the development of climate change adaptation strategies. We also assess the level of consideration given not only to economic and environ-mental criteria, but also to social aspects such as risk man-agement. Second, we use a crop growth simulation model to examine the economic, environmental, and social out-comes resulting from a change in agricultural manage-ment practice. This exercise highlights the potential for conflicting and perverse economic, environmental, and social outcomes to be overlooked when only economic and environmental efficiency is considered.


exhibits summer-dominant rainfall. Soil at the Katanning site was a shallow sandy duplex (Isbell 2002) with a wheat plant-available water capacity (PAWC) of 56 mm. Wheat root growth at this site was allowed to reach a maximum depth of 1000 mm in simulations. Two soils were simulated at the Dalby location to explore the influence of soil water holding capacity on model output. One of the soils was a Dalby black Vertisol (Isbell, 2002) with a wheat PAWC of 336 mm, in which the wheat roots were allowed to reach a maximum depth of 1800 mm. The second soil was a self-mulching black earth (Lever Gully soil) (Isbell, 2002) with a wheat PAWC of 185 mm, in which roots were allowed to reach a maximum depth of 1300 mm. Details of physical and chemical characteristics for all three soils were obtained from the APSoil Database (Dalgliesh et al., 2006). This effectively provided three sites for the simulations, which are referred to as Katanning, Dalby 1 (336-mm PAWC), and Dalby 2 (185-mm PAWC).

After parameterization, APSIM was used to simulate the growth of wheat using daily climate data for the period 1900 to 2004 obtained from the Queensland Department of Environ-ment and Resource Management SILO patched point data sets ( Jeffrey et al., 2001). The simulations were set to sow wheat on a yearly basis on Day 152 after soil water had been reset to a certain level (as indicated in Table 2), with different simulations representing historically dry (fifth percentile), semidry (25th percentile), or average starting soil-moisture conditions for a continuous wheat rotation (with a weedy summer fallow). For all sites soil N was reset to 22 kg ha−1 in the wheat rooting zone just before sowing.

The wheat varieties simulated at the Katanning and Dalby sites were ‘Kulin’ and ‘Hartog’, respectively, reflecting those commonly used in the regions. In all cases wheat was planted at 140 plants m−2 and sown to a depth of 30 mm, again reflecting regional practice. Harvesting occurred when the crop reached “harvest-ripe” stage as defined by the APSIM-Wheat model. A grazing event was simulated by removing 75% of the crop residues from the field immediately after harvest.

At each site, and for each of the soil water conditions in Table 2, a matrix of fertilizer strategies was applied. Fertilizer rates (kg N ha−1 as NO3) were set at 0, 25, 50, 75, 100, 150, 200, and 300 and applied as a single application at the time of sowing. Average gross margins for the 105-yr period were estimated using the model output, the production costs listed in Table 3, and the grain price relationship in Fig. 1. Average gross margins were calculated for two N fertilizer prices (A$1 kg−1 and A$2 kg−1) to explore the impact of differences in N fertilizer cost (or increased application rates used in response to soil degradation [Koning et al., 2008]), on the production incentives for wheat growers and more broadly, the potential implications for national and global food supply.

Simulated data for the three environmental indicators of N leaching, surface water runoff, and drainage from the bot-tom of the soil profile (henceforth referred to as drainage), were normalized across the eight N application rates, with 0 being attributed to the worst performance in terms of environmental outcomes and 1 to the best performance. Correlation coeffi-cients were computed to measure the strength of association between the three environmental indicators and gross margin.

maTERiaLs aND mETHoDs

Quantitative Literature analysisA quantitative literature analysis (also referred to as a simple meta-analysis) was undertaken by identifying published litera-ture relating to eco-efficient agricultural practices and food production activities up to 1 Apr. 2009 using the Internet-based scientific literature search engine Web of Knowledge (http://apps.isiknowledge.com [verified 26 Dec. 2009]). Keywords used in searching the database were divided into variations on the terms eco-efficiency and agricultural production (Table 1). Each of the variations relating to the term eco-efficiency was paired with an agricultural term to produce 32 search criteria. From these searches we identified 18 studies relating to eco-efficient agricultural and food production systems that were (i) based on a case study, that is, not a purely theoretical or conceptual consideration; and (ii) focused on primary production or food processing of either a crop, animal, or fish species.

To determine if there was any statistical evidence that case studies with a climate change rationale tended to include consideration of economic, environmental, or social outcomes resulting from the application of a new technology or practice, the 18 research and conference papers found from the review of literature were subjected to a simple meta-analysis. This analy-sis was undertaken by assessing (i) the extent to which climate change was the objective of the study, and (ii) the extent to which analysis of economic, environmental, or social efficiency was detailed. As there was only one paper from the review that showed climate change to be the primary objective of the study, for the purpose of statistical analysis this was grouped together with papers citing climate change as a secondary or minor objective. A (nonparametric) two-sample Wilcoxon rank-sum test (Lehmann, 1975) was then performed against a one-sided alternative: that is, those papers without a climate change rationale (i.e., scored 0) were likely to include less assessment of economic, environmental, and social outcomes than those papers with a climate change rationale. This is equivalent to the Mann–Whitney test.

Wheat simulation ExperimentTo consider possible tradeoffs between economic, environmen-tal, and social outcomes, the Agricultural Production Systems Simulator (APSIM) (v. 3.6) (Keating et al., 2003) was used to simulate the growth of wheat (Triticum aestivum L.) and three key indicators of environmental performance, namely the amount of N leached below the root zone, surface water runoff, and drainage of water from the base of the soil profile. The wheat module in APSIM has already been validated across a wide range of locations and management practices (Asseng et al., 2000; Wang et al., 2003; Verburg and Bond, 2003; Lilley et al., 2003, 2004), and thus provided confidence that standard APSIM wheat parameter values would adequately simulate growth and environmental performance for this study.

Simulations were undertaken for two diverse locations in the Australian Wheat Belt. The sites were Katanning in Western Australia (33°41¢23.9994² S, 117°33¢36²E), which exhibits a winter-dominant rainfall pattern, and Dalby in Queensland (27°10¢47.9994²S, 151°15¢35.9994²E), which


REsULTs

Quantitative Literature analysisA total of 18 journal and conference papers were iden-tified as utilizing the eco-efficiency concept (Table 4). These studies cover either primary production and food processing practices, or renewable energy products and inputs such as fertilizers and pesticides. In just over half (56%) of the studies, climate change is cited as the ratio-nale for determining the eco-efficiency of a technology or practice. Half of the studies in the review measured eco-nomic impact, 44% assumed the economic outcome, and only one study omitted to consider the economic impacts resulting from a technology or practice. In contrast, all

but one of the studies used empirical data to show envi-ronmental benefits. Overall, only 50% of the studies mea-sured both economic and environmental impacts, thereby satisfying the definition of an eco-efficiency assessment.

In terms of assessing the social impact of a technol-ogy or practice, only one of the studies included this in their assessment; 39% assumed either a positive or negative impact, and 54% failed to give it any consideration. Across the range of studies included in this review, only one mea-sured triple-bottom-line outcomes in terms of economic, environmental, and social impacts.

Although there is some visual evidence for an associa-tion between a climate change rationale and economic, environmental, or social assessment (Fig. 2), this rela-tionship is not statistically significant. When a combined score was constructed by summing the individual scores for economic, environmental, and social content (Fig. 3), there seems to be slightly stronger visual evidence for studies with a climate change rationale to also contain an economic, environmental, and social assessment compo-nents (i.e., a combined score of 6). Again, this difference is not statistically significant with the Wilcoxon statistic only providing a P value of 12%.

Wheat simulation ExperimentWheat yields produced for the eight N fertilizer treatments simulated for the Katanning and Dalby sites are shown in Fig. 4. The solid lines indicate that when fertilizer costs were A$1 kg N−1, maximum gross margin was achieved at Katanning by applying 150 kg N ha−1 and producing a yield of approximately 2500 kg ha−1. However, when the cost of fertilizer doubled to A$2 kg N−1, maximum gross margin was achieved at a reduced N fertilizer application rate of 100 kg N ha−1 and approximately 20% lower yield. A similar, although smaller (10%) reduction in wheat yield also occurred at each of the two Dalby sites when the cost of N fertilizer was doubled.

Figure 5 shows that mean maximum economic effi-ciency (as measured by gross margin) can be achieved

Table 1. Eco-efficient and agricultural production terms used in the literature search. Terms were used in a paired fashion. The wildcard facility (*) was used in the search.

Eco-efficient terms Agricultural production termsecoefficien* agri*

eco-efficien* crop*

farm*

fertiliz*

cattle

livestock

herbicide

intercrop*

land use

organic*

pasture

pesticide

till*

monoculture

permaculture

cultiva*

Table 2. Soil water reset levels for wheat grown at three sites in Australia.

Statistic Katanning Dalby 1 Dalby 2–– mm water over wheat rooting depth ––

5th percentile 36 80 30

25th percentile 43 149 76

Average 54 220 117

Table 3. Operational costs included in gross margin calculations.†

Operational cost Annual expenditureA$ ha−1

N fertilizer (i) 1.00(ii) 2.00

Cultivation 7.71

Sowing 38.32

Extra field operation 13.32

Herbicide 37.89

Harvesting 32.00

Levies 3.75

Crop insurance 7.69†Source: NSW Department of Primary Industries (2005).

Figure 1. Relationship between grain protein (expressed as N) and grain price (A$ t−1).


by applying between 100 and 200 kg N ha−1 for wheat grown at Katanning, 150 and 300 kg N ha−1 at the Dalby 1 site, and 75 to 150 kg N ha−1 at the Dalby 2 site. At all three sites, increasing rates of N generally resulted in increased variability in annual gross margin, and, hence, the amount of risk faced by primary producers. While the greatest mean gross margins may be achieved at the Dalby 1 site, these profits are also the most risky of the three sites in terms of annual fluctuations.

Figure 5 also shows the performance of the three environmental indicators (leaching below the root zone, surface water runoff, and drainage of water from the base of the soil profile) for the three highest gross margin esti-mates at each site. Drainage was the worst (highest nor-malized score) outcome across the three environmental indicators for the three highest gross margin N application rates simulated at the Katanning site, with leaching being the best (lowest normalized score). While the relative

Table 4. Summary details of literature found in the search, including the extent to which climate change was the objective of the study categorized as either none (0), a secondary or minor objective (1), or the primary objective (2) and the extent to which analysis of economic, environmental or social efficiency were detailed categorized as either absent (0), inferred or based on first principles (1), or explicitly measured or tested (2).

Publication Climate change rationale Economic Environmental SocialBasset-Mens et al. (2009) 1 1 2 0

Catarino et al. (2007) 1 2 2 0

Chen and Sun (2007) 0 1 2 0

De Jonge (2004) 1 1 2 0

Ingaramo et al. (2009) 0 0 2 0

Kim and Dale (2005) 1 2 2 0

Kim and Dale (2008a) 2 2 2 0

Kim and Dale (2008b) 1 2 2 1

Lozano et al. (2009) 0 2 2 0

Mouron et al. (2006) 1 2 2 1

Nevens et al. (2006) 0 1 2 0

Ngoc and Schnitzer (2008) 0 2 2 1

Pelletier et al. (2008) 1 1 2 1

Reith and Guidry (2003) 0 1 1 0

Shoichi (2007) 0 2 2 1

Swanston and Newton (2005) 0 1 2 1

Van Passel and Nevens (2007) 1 2 2 2

Wilkins (2008) 1 1 2 1

Figure 2. Association between a climate change rationale and the economic, environmental, and social outcomes from the application of a new technology or practice. Coding detailed in Table 4 caption. Prop.=proportion.

Figure 3. Association between a climate change rationale and a combination simple addition of coding scores (1–6) of the economic, environmental, and social outcomes from the application of a new technology or practice. Prop. = proportion.


performance of the environmental indicators remained the same across the three highest gross margin N appli-cation rates at Katanning, their absolute values changed, with an overall worst environmental outcome resulting at the highest of the three levels of N (200 kg N ha−1).

Similarly, at the Dalby 1 site, the relative performance of the three environmental indicators remained the same across the three highest gross margin N application rates. However, while leaching and drainage reduced with increasing rates of N fertilizer, runoff increased. At the second Dalby site, leaching increased, drainage decreased, and runoff remained almost the same as N rates were increased, resulting in a change in the relative rankings of the environmental indicators.

The correlation coefficients shown in Table 5 indicate a generally negative relationship between the three envi-ronmental indicators and gross margin at all three sites with the exception of that between leaching and gross margin at the Katanning site, which increased with the application of higher rates of N fertilizer.

DisCUssioNAdvancing the assessment of a technology or practice from a single-discipline perspective to both economic and envi-ronmental outcomes, as advocated in the eco-efficiency concept, is a positive step toward identifying win-win strategies for sustainable development. The wheat simula-tions in this study illustrate the potential consequences of a single-discipline perspective on the issue of food secu-rity. For example, in the face of an increase in the price of N fertilizer, the rational response anticipated from pri-mary producers using a purely economic criterion is to reduce yields in an attempt to maintain maximum gross margins. Given the global nature of markets for key agri-cultural inputs such as N, it is reasonable to hypothesize a resulting global-wide reduction in food production, a concomitant drop in global food supply, and increasing volatility in food prices.

In the above example, the private benefits gained by individual primary producers are both rational and effi-cient from an economic perspective, but fail to recognize the wider potential public costs resulting from a reduc-tion in the availability of food and from lower levels of consumer purchasing power. Including a second disciplin-ary perspective in the assessment clearly offers increased potential for win-win outcomes and improvements in food security. Indeed, the asymptotic yield curves pro-duced from the wheat simulations in response to increas-ing applications of N fertilizer suggest maximum yields could be produced at a private cost that is close to maxi-mum gross margin. Such a strategy would simultaneously deliver both private and public benefits.

In addition, the dual consideration of both economic and environmental outcomes also offers some poten-tial for identifying and minimizing unintended perverse incentives and maladaptations. As shown at the Katan-ning site, increasing levels of N leached from the soil pro-file are predicted to result from the pursuit of maximum profit margins. Identifying the tradeoffs between economic and environmental efficiency provides opportunities for optimizing outcomes in the face of competing demands. However, the seemingly unintended incentive for primary producers to maximize economic efficiency at the expense of environmental performance also overlooks the associ-ated increases in the variability of profit margins. Indeed, while the greatest mean gross margins may be achieved at the Dalby 1 site, these profits are also the most risky of the three wheat-growing locations in terms of mean annual

Figure 4. Gross margin (A$) and wheat yield N response curves for applications of N fertilizer at rates of 0, 25, 50, 75, 100, 150, 200, and 300 kg ha−1 for wheat grown at Katanning and two sites at Dalby, Australia. The solid line indicates the maximum mean annual gross margin at a N fertilizer cost of A$1 kg−1. The dotted line indicates the maximum mean annual gross margin at a N fertilizer cost of A$2 kg−1.


fluctuations. Increasingly variable incomes are associated with increased levels of risk and poverty for farmers (Rav-allion, 1988) and negatively impact sustainable food supply. The broader consideration of efficiency (i.e., reduced levels of risk) provided by considering social outcomes in addition to economic and environmental outcomes offers a further advancement on assessing the efficiency of technologies and practices to address food security.

This study has shown that the eco-efficiency concept is presently being used to assess the potential for technolo-gies and practices to improve food security under a chang-ing climate. In the majority of cases, the assessments are limited to only economic and environmental outcomes. By showing the potential for such studies to result in unin-tended perverse social and environmental outcomes and

incentives for primary producers, we argue that a broader assessment of efficiency would be provided by additional consideration of social outcomes, such as variability in gross margin and the risk profile of primary producers. While this proposition is not new to the eco-efficiency

Figure 5. Estimates of mean gross margin (A$) and risk (expressed as variance in yield t ha−1) for wheat grown at Katanning and two sites at Dalby, Australia, for a range of fixed N fertilizer application rates 0, 25, 50, 75, 100, 150, 200, and 300 kg ha−1. Performance of key environmental indicators, that is, leaching, drainage, runoff) also shown for the three highest mean gross margin estimates normalized across the eight N application rates (where 0 indicates the worst performance and 1 indicates the best performance in terms of environmental outcomes).

Table 5. Correlation statistics (r) for the relationship between gross margin and the three indicators of environmental per-formance at three sites in Australia: the amount of nitrogen leached below the root zone, water runoff, and crop drainage.

Gross marginSite Leaching Drainage Runoff

Katanning 0.63 -0.81 -0.82Dalby 1 -0.99 -0.98 -0.72Dalby 2 -0.22 -0.50 -0.89


debate (Schmidheiny and The Business Council for Sus-tainable Development, 1992; Jansen, 2000), the lack of a coherent qualitative and quantitative argument for the simultaneous consideration of economic, environmental, and social outcomes is addressed here for the first time.

A secondary concern regarding the use of the eco-effi-ciency concept highlighted by this study is the readiness of many authors to assume the nature of the relationship existing between economic and environmental variables. For example, the assumption has been made in a num-ber of the reviewed studies that improved environmen-tal outcomes resulting from the adoption of a practice or technology will lead to positive economic benefits (e.g., Chen and Sun, 2007; de Jonge, 2004; Nevens et al., 2006). The experimental results demonstrate a generally nega-tive relationship between increasing N fertilizer applica-tion rates and gross margin, with the exception of wheat growth at Katanning. Differences in this relationship have resulted from the diversity in biophysical conditions and crop management across the sites. Studies that empirically measure environmental benefits and infer concomitant economic benefits, or vice versa, must therefore be con-sidered with some caution.

The four dimensions of food security, (i.e., availabil-ity, access, supply stability, and utilization) move beyond the simple quantification of food supply to provide a com-prehensive appraisal of the requirements for individuals to reach a state of nutritional well-being in which all physio-logical needs are met. These dimensions focus simultane-ously on economic conditions and drivers (e.g., consumer purchasing power, agricultural labor supply), as well as environmental (e.g., biophysical conditions impact-ing food quality and the continuity of supply) and social (adequate diet, clean water, sanitation, and health care) considerations. We have demonstrated that it is therefore of little benefit to consider potential advances in any one aspect of food security as reducing the number of people at risk of hunger, if the advances are achieved in isolation of sufficient capacity in all dimensions of food security.

Broadening the assessment of efficiency to include social outcomes may more fully capitalize on technologi-cal advances that increase production beyond the present production and efficiency frontiers (Keating et al., 2010; de Wit, 1992), and provide an evidence base to support the delivery of deliberate benefits (Giller et al., 2009). Increased understanding of the social outcomes result-ing from a technology change or practice improvement may also contribute to addressing the historically low levels of update and ongoing use observed in the fields of agricultural and development practice (Doss, 2001; Moser and Barrett, 2003). This analysis similarly sug-gests that the assessment of the efficiency of a technology or practice to improving food security in the face of cli-mate change. It also requires broader consideration of the

socioeconomic and biophysical environment necessary to facilitate its adoption (sensu Howden et al., 2007). While it is suggested that socioeconomic developments such as population growth and energy use have relatively greater importance on the number of people at risk of hunger compared to that of climate change (Tubiello and Fischer, 2007), clearly it is important to consider all key factors impacting the ability of an individual to have adequate quantities of food of appropriate quality, accessible at all times, and with the capacity to be utilized sufficiently to reach a state of nutritional well-being in which all physi-ological needs are met (Garrett, 1995).

REFERENCEsAlcamo, J., D. van Vuuren, C. Ringler, W. Cramer, T. Masui,

J. Alder, and K. Schulze. 2005. Changes in nature’s balance sheet: Model-based estimates of future worldwide ecosystem services. Ecol. Soc. 10(2): Article 19.

Asseng, S., H. van Keulen, and W. Stol. 2000. Performance and application of the APSIM N_wheat model in the Nether-lands. Eur. J. Agron. 12:37–54.

Basset-Mens, C., S. Ledgard, and M. Boyes. 2009. Eco-efficiency of intensification scenarios for milk production in New Zea-land. Ecol. Econ. 68:1615–1625.

Catarino, J., E. Mendonça, A. Picado, A. Anselmo, J. Nobre da Costa, and P. Partidário. 2007. Getting value from wastewa-ter: By-products recovery in a potato chips industry. J. Clean. Prod. 15:927–931.

Chen, H.Z., and F.B. Sun. 2007. Novel bioconversion of wheat straw to bio-organic fertilizer in a solid-state bioreactor. Bio-syst. Eng. 30:99–105.

Dalgliesh, N., G. Wockner, and A. Peake. 2006. Delivering soil water information to growers and consultants. In Ground breaking stuff. Proc. Agron. Conf. 2006, 13th, Perth, Australia. 10–14 Sept. 2006. Available at http://www.regional.org.au/au/asa/2006/poster/water/4552_dalglieshn.htm#TopOfPage (verified 18 Jan. 2010).

de Jonge, A.M. 2004. Eco-efficiency improvement of a crop pro-tection product: The perspective of the crop protection indus-try. Crop Prot. 23:1177–1186.

de Wit, C.T. 1992. Resource use efficiency in agriculture. Agric. Syst. 40:125–151.

Doss, C.R. 2001. Designing agricultural technology for African women farmers: Lessons from 25 years of experience. World Dev. 29:2075–2092.

El-Ashry, M.T. 2002. Ecological sustainability and food security: Challenges and opportunities. CGIAR, Manila, Philippines.

FAO. 1996. World food summit: Synthesis of the technical background documents. Available at http://www.fao.org/docrep/003/W2537E/w2537e00.HTM (verified 24 Dec. 2009). Food and Agric. Organ. of the United Nations, Rome.

FAO. 2000. The state of food and agriculture 2000. Available at http://www.fao.org/docrep/x4400e/x4400e00.HTM (veri-fied 24 Dec. 2009). Food and Agric. Organ. of the United Nations, Rome.

FAO. 2006. FAOSTAT: FAO statistical databases. Food and Agric. Organ. of the United Nations, Rome.

FAO. 2008. The state of food insecurity in the world 2008. Food and Agric. Organ. of the United Nations, Rome.

Garrett, J.L. 1995. A 2020 vision for food, agriculture and the


environment in Latin America. Food Agric. and the Environ. Discussion Pap. 6. Int. Food Policy Res. Inst., Washington, DC.

Giller, K.E., E. Witter, M. Corbeels, and P. Tittonell. 2009. Con-servation agriculture and smallholder farming in Africa: The heretics view. Field Crop Res. 144:23–34.

Gleadow, R.M., J.R. Evans, S. McCaffery, and T.R. Cavagnaro. 2009. Growth and nutritive value of cassava Manihot esculenta Cranz. are reduced when grown in elevated CO2. Plant Biol. 11:76–82.

Gregory, P.J., J.S.I. Ingram, R. Andersson, R.A. Betts, V. Brovkin, T.N. Chase, P.R. Grace, A.J. Gray, N. Hamilton, T.B. Hardy, S.M. Howden, A. Jenkins, M. Meybeck, M. Olsson, I. Ortiz-Monasterio, C.A. Palm, T.W. Payn, M. Rummukainen, R.E. Schulze, M. Thiem, C. Valentin, and M.J. Wilkinson. 2002. Environmental consequences of alternative practices for intensifying crop production. Agric. Ecosyst. Environ. 88:279–290.

Howden, S.M., J.F. Soussana, F.N. Tubiello, N. Chhetri, M. Dun-lop, and H. Meinke. 2007. Adapting agriculture to climate change. Proc. Natl. Acad. Sci. USA 104:19691–19696.

Ingaramo, A., H. Heluane, M. Colombo, and M. Cesca. 2009. Water and wastewater eco-efficiency indicators for the sugar cane industry. J. Clean. Prod. 17:487–495.

IPCC. 2007. Climate change: Impacts, adaptation and vulnerabil-ity. Contrib. of Working Group II to the Fourth Assessment Rep. of the Intergovernmental Panel on Climate Change. Cambridge Univ. Press, Cambridge, UK.

Isbell, R. 2002. The Australian soil classification. CSIRO Publish-ing, Melbourne.

Jansen, L. 2000. ‘Quality of life’, sustainable and world-wide: New challenges for agricultural research. p. 227–237. In A. Boekes-tein et al. (ed.) Towards an agenda for agricultural research in Europe. Proc. of a Conf. held in Wageningen, Netherlands. 13–15 Apr. 1999.

Jeffrey, S.J., J.O. Carter, K.B. Moodie, and A.R. Beswick. 2001. Using spatial interpolation to construct a comprehensive archive of Australian climate data. Environ. Modell. Softw. 16:309–330.

Jones, P.G., and P.K. Thornton. 2003. The potential impacts of climate change on maize production in Africa and Latin American in 2055. Glob. Environ. Change 13:51–59.

Keating, B.A., P.S. Carberry, P.S. Bindraban, S. Asseng, H. Meinke, and J. Dixon. 2010. Eco-efficient agriculture: Con-cepts, challenges, and opportunities. Crop Sci. 50(2):109–119.

Keating, B.A., P.S. Carberry, G.L. Hammer, M.E. Probert, M.J. Robertson, D. Holzworth, N.I. Huth, J.N.G. Hargreaves, H. Meinke, Z. Hochman, G. McLean, K. Verburg, V. Snow, J.P. Dimes, M. Silburn, E. Wang, S. Brown, K.L. Bristow, S. Asseng, S. Chapman, R.L. McCown, D.M. Freebairn, and C.J. Smith. 2003. An overview of APSIM, a model designed for farming systems simulation. Eur. J. Agron. 18:267–288.

Kim, S., and B.E. Dale. 2005. Life cycle assessment of various cropping systems utilized for producing biofuels: Bioethanol and biodiesel. Biomass Bioenergy 2:426–439.

Kim, S., and B.E. Dale. 2008a. Effects of nitrogen fertilizer appli-cation on greenhouse gas emissions and economics of corn production. Environ. Sci. Technol. 42:6028–6033.

Kim, S., and B.E. Dale. 2008b. Life cycle assessment of fuel etha-nol derived from corn grain via dry milling. Bioresour. Tech-nol. 99:5250–5260.

Kimball, B.A., C.F. Morris, P.J. Pinter, G.W. Wall, D.J. Hunsaker, F.J. Adamsen, R.L. LaMorte, S.W. Leavitt, T.L. Thompson,

A.D. Matthias, and T.J. Brooks. 2001. Wheat grain quality as affected by elevated CO2, drought, and soil nitrogen. New Phytol. 150:295–303.

Kiyama, S. 2007. Influence of agriculture and inter-regional trad-ing on watershed water quality and production managements. p. 109–111. In IASTED Int. Conf. on Water Resources Man-age., 2nd, Honolulu, HI. 20–22 Aug. 2007.

Koning, N.B.J., M.K. Van Ittersum, G.A. Bec, M.A.J.S. Van Boeke, W.A. Brandenburg, J.A. Van Den Broek, J. Goudri-aan, G. Van Hofwegen, R.A. Jongeneell, J.B. Schiere, and M. Smies. 2008. Long-term global availability of food: Con-tinued abundance or new scarcity? NJAS-Wagen. J. Life Sci. 55:229–292.

Lehmann, E.L. 1975. Nonparametric statistical methods based on ranks. Holden-Day, San Francisco.

Lilley, J., J. Kirkegaard, M. Robertson, M. Probert. J. Angus, and G. Howe. 2003. Simulating crop and soil processes in crop sequences in southern NSW. Proc. Aust. Agron. Conf., 11th, Geelong, Australia. 2–6 Feb. 2003. Aust. Soc. of Agron., Geelong, Australia.

Lilley, J.M., M.J. Probert, and J. Kirkegaard. 2004. Simulation of deep drainage under a 13-year crop sequence in southern NSW. Poster 1.6. In Proc. Int. Crop Sci. Congr., 4th, Bris-bane, Australia. 26 Sept.–1 Oct. 2004.

Liu, J., S. Fritz, C.F.A. van Wesenbeeck, M. Fuchs, L. You, M. Obersteiner, and H. Yang. 2008. A spatially explicit assess-ment of current and future hotspots of hunger in sub-Saharan Africa in the context of global change. Global Planet. Change 64:222–235.

Lobell, D.B., M.B. Burke, C. Tebaldi, M.D. Mastrandrea, W.P. Fal-con, and R.L. Naylor. 2008. Prioritizing climate change adap-tation needs for food security in 2030. Science 319:607–610.

Lozano, S., D. Iribarren, M.T. Moreira, and G. Feijoo. 2009. The link between operational efficiency and environmental impacts. A joint application of Life Cycle Assessment and Data Envelopment Analysis. Sci. Total Environ. 407:1744–1754.

Manderscheid, R., J. Bender, H.-J. Jäger, and H.J. Weigel. 1995. Effects of season long CO2 enrichment on cereals. II. Nutri-ent concentrations and grain quality. Agric. Ecosyst. Environ. 54:175–185.

Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being: Current state and trends. Findings of the Condition and Trends Working Group. Island Press, Washington, DC.

Moser, C.M., and C.B. Barrett. 2003. The disappointing adoption dynamics of a yield-increasing, low external-input technology: The case of SRI in Madagascar. Agric. Syst. 76:1085–1100.

Mouron, P., R.W. Scholz, T. Nemecek, and O. Weber. 2006. Life cycle management on Swiss fruit farms: Relating environ-mental and income indicators for apple-growing. Ecol. Econ. 58:561–578.

Nevens, F., I. Verbruggen, D. Reheul, and G. Hofman. 2006. Farm gate nitrogen surpluses and nitrogen use efficiency of specialized dairy farms in Flanders: Evolution and future goals. Agric. Syst. 88:142–155.

Ngoc, U.N., and H. Schnitzer. 2008. Waste management towards zero emissions approach in the fruit juice processing industry. WSEAS Int. Conf. on Environ. and Geol. Sci. and Eng., 1st, Malta. 11–13 Sept. 2008.

NSW Department of Primary Industries. 2005. Farm budgets and costs for continuous wheat. Available at http://www.dpi.nsw.gov.au/agriculture/farm-business/budgets (verified 18 Jan. 2010). Dep. of Primary Industries, Orange, NSW, Australia.


Parry, M.L., C. Rosenzweig, A. Iglesias, M. Livermore, and G. Fischer. 2004. Effects of climate change on global food pro-duction under SRES emissions and socio-economic scenar-ios. Glob. Environ. Change 14:53–67.

Pelletier, N., N. Arsenault, and P. Tyedmers. 2008. Scenario modeling potential eco-efficiency gains from a transition to organic agriculture: Life cycle perspectives on Canadian canola, corn, soy, and wheat production. Environ. Manage. 42:989–1001.

Ravallion, M. 1988. Expected poverty under risk-induced welfare variability. Econ. J. 98:1171–1182.

Reith, C.C., and M.J. Guidry. 2003. Eco-efficiency analysis of an agricultural research complex. J. Environ. Manage. 68:219–229.

Rosenzweig, C., and M.L. Parry. 1994. Potential impact of cli-mate change on world food supply. Nature 367:133–138.

Schmidheiny, S., and The Business Council for Sustainable Devel-opment. 1992. Changing course. MIT Press, Cambridge, MA.

Schmidhuber, J., and F.N. Tubiello. 2007. Global food security under climate change. Proc. Natl. Acad. Sci. USA 104:19703–19708.

Stern, N. 2007. The economics of climate change—The Stern review. Cambridge Univ. Press, Cambridge, UK.

Stige, L.C., J. Stave, K.-S. Chan, L. Ciannelli, N. Pettorelli, M. Glantz, H.R. Herren, and N.C. Stenseth. 2006. The effect of climate variation on agro-pastoral production in Africa. Proc. Natl. Acad. Sci. USA 103:3049–3053.

Swanston, J.S., and A.C. Newton. 2005. Mixtures of UK wheat as

an efficient and environmentally friendly source for bioetha-nol. J. Ind. Ecol. 9:109–126.

Tubiello, F.N., and G. Fischer. 2007. Reducing climate change impacts on agriculture: Global and regional effects of mitiga-tion, 2000–2080. Technol. Forecast Soc. 74:1030–1056.

Van Passel, S., and F. Nevens. 2007. Measuring farm sustainabil-ity and explaining differences in sustainable efficiency. Ecol. Econ. 62:149–161.

Verburg, K., and W.J. Bond. 2003. Use of APSIM to simulate water balances of dryland farming systems in south eastern Australia. Tech. Rep. 50/03. Available at http://www.clw.csiro.au/publications/technical2003/tr50-03.pdf (verified 24 Dec. 2009). CSIRO Land and Water, Canberra.

Wang, E., E. van Oosterom, H. Meinke, S. Asseng, M. Robert-son, N. Huth, B. Keating, and M. Probert. 2003. The new APSIM-Wheat Model performance and future improve-ments. In Solutions for a better environment. Poster 2. In Proc. of the Aust. Agron. Conf., 11th, Geelong, Australia. 2–6 Feb. 2003. Aust. Soc. of Agron., Geelong, Australia.

Wilkins, R.J. 2008. Eco-efficient approaches to land management: A case for increased integration of crop and animal production systems. Philos. Trans. R. Soc. London, Ser. B 363:517–525.

Zoebl, D. 1996. Controversies around resource use efficiency in agriculture: Shadow or substance? Theories of C.T. de Wit 1924–1993. Agric. Syst. 50:415–424.


symposia

With the adoption of the Millennium Declaration, the real-ization of poverty alleviation and sustainable development

received renewed attention and support. The Millennium Devel-opment Goals (MDGs) were subsequently formulated. The most important of these goals for the CGIAR system is the halving of hunger and poverty by 2015 in developing countries strongly linked to agriculture. Modest progress toward MDGs is occurring in a dynamic context characterized by changes in demography, markets and prices, institutions and culture, policies, agricultural and environmental resources, and technological development.

The discussion of agricultural futures in a biobased economy in this article is framed by the commitments underpinning the MDGs. This assumes that agricultural production to meet the new demands, which will emerge in a biobased economy, will be com-plementary to basic agricultural products and services required to meet the basic requirements of mankind echoed in the MDGs, especially those related to food, health, and environment.

Development Perspectives Of The Biobased Economy: A Review

J. W. A. Langeveld,* J. Dixon, and J. F. Jaworski

abstractThispaperprovidesanoutlineofthebiobasedeconomy, its perspectives for agriculture and,more particularly, for development purposes.Possibilitiesofdevelopmentofbiobasedprod-ucts,advancedbiofuels,andviableandefficientbiorefinery concepts are explored. The paperlists non-fuel bioproducts (e.g., chemicals,pharmaceuticals, biopolymers) and presentsbasic principles and development options forbiorefineriesthatcanbeusedtogeneratethemalongside biofuels, power, and by-products.Oneofthemainchallengesistocapturemorevalue from existing crops without compromis-ing the needs and possibilities of small-scale,lessendowedfarmers.Biobasedproductsofferthemostdevelopmentperspectives,combininglargemarketvolumeswithmediumtohighpricelevels.Consequently,themostcanbeexpectedfrom products like fine chemicals, lubricants,and solvents. In addition, biosolar cells canhelptorelaxpressuresonbiomassproductionsystemswhiledecentralizedproductionchainscanservelocalneedsforenergy,materials,andnutrients as their requirement for viable eco-nomic development are linked to larger mar-kets.Researchchallengesincludedevelopmentof suchproductionandmarket chains, andofbiosolarcellsandselectionofmodelcropsthatoffer perspectives for less favored producersandunderdevelopedruralareas.

J.W.A. Langeveld, Biomass Research, P.O. Box 247, 6700 AE, Wagenin-gen, the Netherlands; J. Dixon, Australian Centre for International Agricultural Research (ACIAR), Bruce 38 Thynne St, Fern Hill Park, Canberra ACT 2617, Australia; J.F. Jaworski, formerly of Life Science Industries Branch, Industry Canada, Ottawa, Canada. Received 23 Sept. 2009. *Corresponding author ([email protected]).

Abbreviations: 1,3 PDO, 1,3-Propanediol; DDGS, Distillers Dried Grains with Solubles; DME, Dimethylether; EU, European Union; GAP, Good Agricultural Practices; GHG, Greenhouse gas; MDG, Millennium Development Goal; MFC, Microbial Fuel Cell; PET, Polyethylene terephthalate; PHA, Polyhydroxyalkanoate; PLA, Polylactic acid; R&D, Research and development.




Agriculture still underpins key livelihoods for most people living in rural areas. In addition to the provision of food, fiber, and energy, agriculture also contributes to poverty reduction and economic development by pro-viding employment in and income from value chains. Diversification, defined as an increased number of activi-ties generating farm output or added value for the farm household, can be defined at different levels (e.g., field, farm, region, or country) of aggregation.

The development of a biobased economy will take place in an uncertain context, contributed to by climate change, production of biofuels, fossil fuel price, global financial systems, and the nexus with food security. Between 1970 and 2004, greenhouse gas (GHG) emissions increased by 70%. By 2015 the world will need to provide extra food for an additional 750 million people. Land use manage-ment, agronomy and livestock sciences, and technological development are key factors determining the net outcome of these processes. Proper agricultural management can contribute to increasing carbon soil sinks (Govaerts et al., 2009), reducing GHG emissions and providing feedstocks for bioenergy. Industrial production technologies can pro-vide new uses for agricultural feedstocks.

The potential of the bioeconomy extends well beyond bioenergy. While a small share of fossil oil is used for chem-ical production and the remainder for fuel and energy, the economic value of the food and chemistry sectors is approximately equal. A long term and sustainable market can be envisaged for technologies that produce chemicals, materials, and pharmaceuticals from plant-based feed-stocks (Sanders et al., 2007), which will supplement the emerging demand for bioenergy feedstocks and the still growing demand for food and other agricultural products.

Such a development will need to be supported by pro-cessing steps that are energy efficient and cost-effective. Biorefineries provide sufficient opportunities to allow such a development. The development of the bioeconomy has often been portrayed as sustainable or environmentally friendly, but there are key resource-related concerns that need to be addressed as biobased systems evolve. These include non-renewable energy use, renewable energy and land use efficiency, carbon emissions and sequestration, soil fertility and erosion, water quality and quantity, wild-life habitat, invasive species, and crop pests (Anex, 2007).

This article explores possibilities of biomass applica-tion for development purposes—in particular, biofuels, biobased products, and biorefineries. It provides an over-view of non-fuel products, including chemicals, pharma-ceuticals, and biopolymers, and discusses basic principles and development options for biorefineries that can be used to generate both fuels and products. Furthermore, it discusses development opportunities and research requirements in light of developing a biobased economy

that offers opportunities for small-scale farmers in less developed areas.

biobased productsFacing a future shortage of petrochemicals, biomass is

expected to be the main future feedstock for chemicals. The use of vegetable oils, crop starch, residual proteins (from biofuel production), and cellulose (from straw and wood) to produce polymers, lubricants, solvents, surfactants, and specialty and bulk chemicals traditionally made from fos-sil feedstocks is receiving more and more attention (Van Haveren et al., 2007). Currently only a tiny proportion of the huge variation of compounds produced by plants is tapped for commercial use. The challenge is to create viable business models for biobased products, and to tailor plants and plant systems to optimize available functionalities. To this purpose, dedicated programs implemented in the United States, European Union (EU), and elsewhere (e.g., Canada, Japan, Malaysia) apply industrial crops and biomass for high-value products in advanced production chains.

Biobased products refer to non-food products derived from biomass (plant, animal, marine, residual), ranging from high-value added (usually low volume) fine chemi-cals (pharmaceuticals, cosmetics, food additives) to high volume materials (enzymes, biopolymers, biofuels, fibers, etc.) They may include existing products (paper and pulp, detergents, lubricants), or new ones (vaccines made from plants or second generation biofuels).

biomaterialsModern non-medical biomaterials include pharma-

ceuticals, chemicals, specialty products, industrial oils, biopolymers, and fibers (Thoen and Busch, 2006). Pro-duction of pharmaceutical feedstocks, providing a major opportunity for agriculture and household livelihoods, is based on the provision of genetic material and production of feedstocks. It involves specialist knowledge markets with small production volumes. The high added value provides a potential avenue for development, but given high research and development costs, it may require long term collaborative relations to link farmers to research, production, and marketing activities.

Chemicals and their feedstocks provide more predict-able markets and specifications than pharmaceutical prod-ucts. Chemical markets refer to bulk chemicals with high volumes, but low values, and fine chemicals with smaller market size, but higher added value. The potential list of biobased chemicals is considerable and includes 1,3-Pro-panediol (1,3 PDO), a building block for polymers that is mostly made from maize (Zea mays) syrup by modi-fied Escherichia coli bacteria. The world market has been estimated at 230,000 t in 2020 (Carole et al., 2004). Suc-cinic acid, another chemical building block, is generated by the fermentation of glucose and is applied in food, the


glucose syrup made from maize, cane, potato, or wheat (Vaca-Garcia, 2008), but may, in the future, be of lig-nocellulosic origin (Carole et al., 2004; Dornburg et al., 2006). Starch-based bioplastics are applied as packaging materials, kitchenware, car interiors, horticulture devices, and diapers ( Johansson, 2000).

Fossil fibers like polyester or nylon offer large oppor-tunities for biobased feedstocks (Carole et al., 2004). Natural fibers can be applied in high value-added com-posite materials using cellulosic feedstocks from wood and straw, plus classical crops like kenaf (Hibiscus cannabinus), sisal (Agave sisalina), jute (Corchorus spp.), flax (Linum usita-tissimum), and hemp (Cannabis sativa). Additionally, euca-lyptus (Eucalyptus spp.) may replace synthetics like rayon (Nowicki et al., 2008). Composite materials based on cel-lulosics offer special qualities (reduced weight, improved safety, and good acoustic properties), and natural fibers are being used to reinforce synthetic materials rather than replace them (Vaca-Garcia, 2008).

development perspectivesThe size of existing (fossil dominated) markets and

potential biobased shares shows large variations. The high-est market volumes are reported for polymers, solvents, and surfactants. The best prospects are for pharmaceutical ingredients, enzymes and specialties (solvents, surfactants) (Carole et al., 2004), followed by bulk chemicals and bio-polymers (Nowicki et al., 2008). Biobased market develop-ments are supported by ambitious policies in the EU and the United States—the latter targeting a 12% replacement of chemical feedstocks in 2010 and 25% in 2030 (Thoen and Busch, 2006). At current fossil oil prices, however, pro-duction is not competitive (Lazerri, 2009).

Impacts of enhanced biomaterial production and application include:- reduced demand for fossil fuels;- increased added value generation for biomass produc-

ers and traders;- reduced GHG emissions; - industrial development;- development opportunities for rural areas, including

employment;- reduced toxicity and enhanced health implications. While not all these impacts are options for develop-

ing countries, it is difficult to evaluate biobased product groups in terms of their development prospects. They will represent combinations of market size, price plus potential share for biobased feedstocks, and the opportunities this offers for farmers in developing countries or local laborers (Table 1).

Evaluating a combination of opportunities (feedstock added value, employment, import replacement, export) offered by the entire production chain, rather than con-sidering biomass feedstock market values alone, suggests

chemical industry, and pharmaceutics. The world market currently amounts to 25,000 t (Sijbesma, 2009). Both 1,3 PDO and succinic acid are targets of efforts to improve production efficiency (Carole et al., 2004; Koutinas et al., 2008) involving crops like sugarcane (Saccharum officina-rum), maize, rice (Oryza sativa), barley (Hordeum vulgare), and potato (Solanum tuberosum) (Thoen and Busch, 2006).

Specialty chemicals serve as adhesives, solvents, and sur-factants (an important group of products applied in deter-gents, cosmetics, and manufacturing processes). Surfactants, still mainly petroleum-derived, are increasingly made from renewable feedstocks. Production exceeds 2 million t. They provide a large market for renewable feedstock, mostly tropi-cal vegetable oils (Turley, 2008). Coconut (Cocos nucifera) and oil palm (Elaeis guineensis) are preferred feedstocks because of the shorter length of their fatty acids. Longer-chained oils from temperate crops (rapeseed- Brassica spp., sunflower- Helianthus annuus) are more suited for use in polymers, lubri-cants, adhesives, solvents, and surfactants.

Solvents, applied in the manufacturing of pharma-ceuticals, paints, and inks, are increasingly produced from biobased products like ethyl lactate, a lactic acid deriva-tive (Carole et al., 2004). Lactate esters are produced from alcohols and fatty acids, with both obtained via fermen-tation of carbohydrates (cereals, potato, and sugar beets). Rapeseed and sunflower oils are major sources of fatty acids; soybean (Glycine max) oil provides the most veg-etable resins ( Johansson, 2000).

Industrial oil products like high quality lubricants and hydraulic oils offer considerable biobased market potential. Biolubricants constitute an innovative area for agricul-ture and industry. Biobased hydraulic fluids comply with industrial quality standards, as do soy based color inks, which dominate due to superior performance (Nowicki et al., 2008). Sunflower and safflower (Carthamus tincto-rius) oils have high oxidation resistance, while oils high in erucic acid (crambe-Crambe maritima, carinata– Brassiaca carinata) show more lubrication qualities (Lazerri, 2009).

Bioplastics show huge opportunities, given that plas-tics are extensively used worldwide (Carole et al., 2004). Starch plastic application, beginning already in the 1970s and currently being commercially produced, offers a major end use for cassava (Manihot esculenta) (Nigeria, Bra-zil), maize, and wheat (Triticum aestivum). Starch proper-ties depend on the amylose/amylopectin ratio and size of starch granules. Amylose ethers offer biodegradable alter-natives for polyethylene and polystyrene (Somerville and Bonetta, 2001).

Commercially interesting polyesters, made from starch or sugar via fermentation, include polylactic acid (PLA) and polyhydroxyalkanoate (PHA) (Turley, 2008; Vaca-Garcia, 2008). The PLA is competing with fossil polymers like PET and links to the large market of pack-aging and fiber/fiberfill materials. The main feedstock is


that fine chemicals, lubricants, and fibers may offer the best prospects for developing countries.

research prioritiesBioproduct research initiatives focus on plant breeding,

product development and improvement of production pro-cesses. Pharmaceuticals, oil products, and (fine) chemicals often have specific feedstocks. Other product types could be made from larger numbers of crops. In practice, how-ever, production chains often are based on a single crop, such as Dow’s PLA, which is made from maize.

Perspectives of breeding are reviewed by Ranalli (2007). Van Beilen et al. (2007) explored the use of sugar beet (Beta vulgaris), tobacco (Nicotiana tabacum), and Miscan-thus (Miscanthus spp.) for production of chemicals, biopoly-mers, and fuels. With some exceptions (e.g., tobacco in Zimbabwe, Malawi, China, and Laos), these crops are not generally cultivated by smallholders in developing coun-tries. Restrictions on genetic modification may limit the development of suitable plant varieties.

There is need for a knowledge platform for research on oil-producing plants that are more productive in exist-ing and designer oils, and to identify molecular markers for breeding (Graham, 2007). The EPOBIO, a research consor-tium in Europe, considers three crops: rapeseed, oat (Avena spp.), and crambe (Crambe abyssinica) (Carlsson, 2007).

Commercial exploitation of less common fatty acids (e.g., retrieved from Calenda officanalis) is hindered by low yield, small seeds, and limited geographical distribution. Further, there is a need to understand the metabolic path-ways and molecular interactions linked to a given fatty acid. Many genes have been identified that could be used to alter oilseed fatty acid composition. Transgenic plants created with these genes show low yields, leaving many unanswered questions (Graham, 2007).

advanced biofuels

While many government policies are based on the principle that biofuels should not compete directly with food security, the reality is that biofuel production, whether first or second generation, may compete for scarce natural resources (soil, water, nutrients). Naturally, as agricultural productivity increases, resources can be freed from food production for the production of energy and biobased prod-ucts. This section explores some options for the sustain-able and balanced development of an agriculture providing food and fuel, and the scientific and technological needs for the next generation of biofuels produced in a biobased economy, recognizing that the ultimate limitation on the production of biomass lies with photosynthesis.

Cereals constitute the majority of all crops cultivated, making up 70 to 90% of reported arable annual crops. The potential of cereal biorefinery for biobased produc-tion (e.g., biofuels) in developing countries is restricted by public perceptions of cereals as food—although a large and growing proportion of cereal grains is used for animal feed. Social tensions caused by the food versus fuel debate have put serious limits on this development pathway, sometimes leading to the exclusion of specific crops, and defining detailed environmental, economic, and social criteria to be met by producers in other situations.

Sugar and oil crops, two other major sources of first generation fuels, are less common, but may play an impor-tant role in specific regions. Research and development have been much less spectacular than those reported for cereals, but still significant efforts for improvement have been made, often in close collaboration with industry.

Availability of lignocellulosic crop residues, a major feed-stock of second generation biofuels, is determined by crop area, yield, harvest index, and demand for other purposes, such as livestock fodder. The greatest biomass productivity

Table 1. Main development perspective of biobased products.†

Product FeedstocksMarket

sizeMarket price

Potential biobased

share

Potential biobased

production size

Potential impact for local

producers

Potential local

employmentProspects for development

Pharmaceuticals Selective crops Very small Very high Very high Very low Very low – Very poorBulk chemicals Starch, sugar

crops, proteinsVery large Low Modest Very low Very low – Poor to modest

Fine chemicals Oil, starch, sugar crops, straw

Very small Average to good

Low Low Modest Very limited Modest to good

Solvents Oil, starch, sugar crops, straw

Small Low Very low Very low Very low Very limited Very poor

Surfactants Various Small Low Modest Low Low Very limited PoorLubricants Oil crops Very small Low Modest to

highLow Low Good Modest to good

Polymers Mostly starch & sugar crops

Very large Very low Low Modest Very low Very limited Very limited

Fibers Lignocellulosic crops, residues, grasses

Modest Rather low Low Modest Low Good Modest to good

†Source: composed by the authors using data on market size and price and projections of potential market share and size as well as expected perspectives (employment, income) for local biomass producers and laborers.


is expected for sugarcane in Brazil, followed by maize in the U.S. Second generation technology could, however, threaten soil fertility as soil cover is removed, and in a similar fashion soil erosion and soil health, including the depletion of soil nutrients, structure, and organic matter, which underpins agricultural productivity and food security.

While the impact of large-scale cultivation of biomass for first or second generation biofuels is debated, a greater consensus appears to exist about small-scale biomass for biofuel applications in developing countries. Local pro-duction of biomass, or local use of residues, may help local communities improve access to renewable energy sources and, hence, reduce workloads and pressure on wood resources, and help gain independence from often expensive fossil fuel sources. This holds promise for less endowed small-scale farmers in isolated inland areas.

Still, developments that take advantage of new tech-nologies are needed to avoid the food vs. fuel controversy, sometimes referred to as ‘next generation biofuels’. A specific scientific development is focusing on photosynthesis, produc-tion of sugars by plants (and some bacteria), using chlorophyll to harvest solar energy. As photosynthetic efficiency funda-mentally limits potential biomass production, it is important to examine ways to increase the current efficiency.

Although higher photosynthetic efficiency may result in a higher production of biomass, energy delivered from this biomass still could not avoid competition with food. In theory, future biofuel systems can be envisaged in which plant fuel cells tap photosynthetic products directly. This bypasses the development of plant structural elements (stems, root, and reproductive elements), potentially real-izing productivity and efficiency gains. Innovative appli-cations of biotechnology, nanotechnology, and genomics provide tools to study and understand the fundamental processes of photosynthesis, starting from the molecular building blocks via the thylakoid membrane to the leaf. This knowledge is the key to improving the efficiency of photosynthesis, either by direct energy tapping or by the production of energy-efficient biomass (Fig. 1).

development perspectivesThe potential contribution of crops or plant systems

with enhanced photosynthetic capacity cannot be easily overestimated. Improving existing light use efficiency rates by a tenth of the current values can lead to considerable yield potential increases. The potential for development-related improvement will depend on the application of an enhanced production system. Artificial leaves and bioso-lar cells are most likely to be implemented in high tech environments and will, for the time being, not be directly linked to vulnerable groups in developing countries.

Algal production may be implemented in a less sophisticated setting, offering potential for resource-poor farmers in the tropics, in ways similar to the complex

and integrated crop–fish production systems of Southeast Asia. In the long run, however, these systems may relax biomass constraints, both locally and on an international level. They, further, may be expected to lead to increased input use efficiency (offering more biomass for the same input of water, nitrogen, phosphorus, etc.). Their impact, in combination with other innovative photosynthetic-related research (e.g., on transplanting C4-systems into C3-crops) can be tremendous.

research prioritiesCurrent cereal research for advanced biorefining

focuses on improving starch and straw for biofuels. Cell wall structure degradability is expected to become an important breeding target, while production of polymers and bioplastics would require breeding for other, special-ized, traits.

There are many lignocellulosic crops that are exten-sively found in developing countries. Choice of a specific crop will depend on agro-ecological and economic condi-tions. Poplar (Populus spp.), a fast growing, vegetative prop-agated crop native to temperate and subtropical regions, whose genome (40,000 genes) has been sequenced, has been defined as an ideal research crop. Research focuses on insect and disease resistance, herbicide tolerance, and lignin content (Boerjan, 2009).

Miscanthus, a perennial grass, has a high yield potential and can be grown effectively under low input conditions. It is, however, not fully developed for widespread cultiva-tion. There are urgent needs to establish a robust breeding program. Potential improvements of Miscanthus include tolerance to drought and low temperatures, stem borers, and fungal diseases (Clifton-Brown et al., 2008).

New concepts for increasing photosynthetic efficiency that are currently being developed include (Arshadi and Sellstedt, 2008):

· Artificial leaves, using light to extract electrons from water to produce hydrogen and synthetic gas. This involves ultra fast light harvesting and micro reac-tors, a photocatalysis system, and an inorganic nano-structure to generate the fuels;

· BioSolar Cells, which are organisms designed with synthetic biology to produce fuels (butanol, metha-nol, ethanol, lipids, and hydrogen) without the bio-mass intermediate, and with a positive contribution to solving the CO2 problem. Solar energy may be temporarily stored in a carbohydrate biofilm grown on a low-cost biobattery system;

· Plant Microbial Fuel Cells (MFCs) for nondestruc-tive in situ harvesting of bio-energy, which is carbon neutral and free of combustion emissions. Oxidation of organic compounds produced by plant cells and excreted by roots generates electricity as bacteria donate electrons to an anode;


· Biofuel and fatty-acid producing algae or cyanobac-teria. Algae known to contain very high contents of oil are slow growers. In addition, production of omega-3 fatty acid can be performed by marine algae. Optimizing the photosynthetic capacity of algae will enable commercial production for biofuels and food.

The suggestions discussed above link to innovative and exciting research on more efficient photosynthetic sys-tems, e.g., the introduction of an efficient C4-system with its high CO2 upload capacity at the surface of the Rubisco enzyme by the transfer of sets of genes to C3-crops such as rice or wheat. This would allow an improvement of the Rubisco system, playing a central role in C3 photosynthe-sis, which is less efficient at current CO2 concentrations and will lead to more efficient water use. Another exam-ple is the combination of bacteria photosystems absorbing light in the near infrared where plants, algae, and cyano-bacteria are not active.

Clearly, it is a long road before advanced systems for direct photosynthetic harvesting can be expected. Major technological challenges lying ahead include improv-ing genetic aspects of photosynthetic systems, increasing insight in biochemical production and composition of photosynthetic products and enzymatic mechanisms, plus development of feasible, affordable, and effective produc-tion systems. Other applications, such as the use of algae or cyanobacteria, are close at hand. Successful application will require efficient production systems and organisms adapted to these systems.

processing for the biobased economy: biorefineries

The biorefinery concept aims to make optimal use of plant components. In this concept, energy produc-tion is not a primary, but only one optional application

of biomass. Feedstock selection, logistics, and biorefining techniques are used to optimize valorization of available functionalities and biomass utilization. Complex input-output chains help to realize optimal economic and social opportunities. This is done by first generating (low vol-ume) high added value products, followed by other, less valuable products (Fig. 2).

The following biorefinery types can be distinguished: (i) whole crop, (ii) oleochemical, (iii) lignocellulosic feed-stock, and (iv) green.1

A whole crop biorefinery processes grain into a range of products, usually via ‘dry’ or ‘wet’ milling and conse-quent fermentation and distilling of grains (wheat, rye, maize). Wet milling starts with water-soaking to soften grain kernels, followed by grinding. It uses well-known technologies to separate starch, cellulose, oil, and proteins. Dry milling grinds whole grains before mixing the flour with water, adding enzymes and cooking the mash to break down the starch. This hydrolysis step can be elimi-nated by the simultaneous use of enzymes and yeast. After fermentation, ethanol is distilled, concentrated, purified, and dehydrated. The residue (stillage) is separated into a solid (wet grains) and liquid (syrup) phase, which can be combined and dried to produce distillers dried grains with solubles (DDGS), an animal feed. Alternatively, grains may be processed into starch, and further to polymers or bioplastics. In a simultaneous process, straw can be con-verted into energy or products, following the principles of the lignocellulosic feedstock biorefinery discussed below (Clark and Deswarte, 2008).

An oleochemical biorefinery combines production of biodiesel with that of high added-value vegetable-oil

1Description of biorefineries is based on Kamm et al. (2006), Clark and

Deswarte (2008), and De Jong et al. (2009).

Fig. 1. Options to improve energy capture by plants for enhanced production of food, fuels and products.


based products. It uses oil-crop fatty acids, fatty esters, and glycerol to produce (basic) chemicals, functional monomers, lubricants, and surfactants. In the long run, oleochemical biorefining may produce feedstocks for fos-sil-based refineries. Success of the biorefinery will depend on its integration with existing fossil chains, its building blocks providing a neat interface (De Jong et al., 2009).

Lignocellulosic feedstock biorefinery encompasses transformation of lignocellulosic biomass into intermedi-ate outputs (cellulose, hemicellulose, lignin) to be processed into a spectrum of products and bioenergy. Three process-ing routes may be chosen. Following the bio-chemical route, a Sugar Platform Biorefinery treats lignocellulosic biomass to release cellulose, hemicellulose, and lignin. Cel-lulose then is converted using enzymatic hydrolysis into glucose, mannose, and xylose. The sugars are converted into biofuels (ethanol, butanol, hydrogen) and/or added-value chemicals. Lignin is applied in combined heat and power combustion, but may in the future be transformed into added-value chemicals (De Jong et al., 2009).

Thermo-chemical refining applied in the Syngas Plat-form Biorefinery consists of high-temperature-cum-pres-sure gasification of lignocellulosic biomass into syngas. The gas is cleaned and used to produce biofuels [Fischer-Tropsch diesel, dimethylether (DME), or alcohol] and/or a variety of base chemicals (ethylene, propylene, butadi-ene, etc.) using catalytic synthesis processes (De Jong et al., 2009). A mixed approach, the so-called Two Platform Concept Biorefinery (or Integrated Bio/Thermo-chem-ical Biorefinery), integrates sugar and syngas refineries to generate bioenergy and/or biobased products. For this purpose, sugars are treated and biochemically processed, whereas lignin is thermochemically treated. Sugar refin-ing (fermentation and distillation) and syngas residues are applied in combined heat and power production units to cover (part of ) the energy requirements.

Green biorefineries, feeding grass to a cascade of pro-cessing stages, offer an innovative alternative processing route for grass feedstocks. Essential is the mechanical grass (“green biomass”) fractionation into a liquid phase con-taining soluble compounds (lactic acid, amino acids) and a solid phase mainly consisting of fibers. Overall economic efficiency of the biorefinery is mainly determined by the economic return of the fibers (De Jong et al., 2009). Major characteristics of these dominant biorefinery processes are presented in Table 2.

development perspectivesBiorefineries offer prospects of enlarged sector out-

put value and prospects for growth of smallholder farmers’ incomes, but value added and income effects will depend on product and market differentiation. The relevance of biorefineries for development depends on a link to avail-able biomass resources, options for economic conversion

routes in developing countries, and the scale and location of the biorefineries.

Major sugar and starch crops can be applied in fermen-tation processes that provide inputs for the production of chemicals, specialty products, and fuels. Vegetable oils can be applied as plasticizers, lubricants, dyes, and resins. While most small-scale farmers produce some of these crops, they will not necessarily profit from future biobased develop-ments. Well endowed large-scale farmers are the first to fill the need for extra biomass feedstocks. To realize develop-ment potentials, biorefineries should fit in the needs and possibilities of small scale farmers and their families.

Further, their role in production chains should safe-guard perspectives for a profitable feedstock provision and/or integration in labor patterns and local employment, while increased demand for local resources (land, water) should not limit their access to such critical resources. It is likely that the best prospects are for systems with limited capital requirements or systems providing a guarantee for a long collaboration. Refineries offering cheap and local sources of energy, and activities that reduce water contents of (intermediate) feedstocks (limiting transportation costs and risks of decay) offer the best development options.

The potential of lignocellulosic biomass production in developing countries is huge, but current use or eco-system service (fuel production, biodiversity, water cap-ture) places limits on its application. Marginal lands may provide only low to moderate yield levels. The potential for the production of chemicals, lubricants, and other bio-based products has to be evaluated but second generation bioethanol production may be a viable alternative locally.

Sugar beet has been identified as model crop for research on chemical building blocks, but its relevance for developing countries currently is limited. Cassava, a local low-cost source of starch, has interesting prospects as a source of bioethanol. EMBRAPA has bred cassava

Fig. 2. Market prices versus market volumes biobased “products.” Source: De Jong et al. (2009).


cultivars with high sugar content specifically designed for bioethanol production.

research prioritiesProspects for biorefinery research are mainly found

in determining its potential for development applications. What refinery systems offer opportunities to poor farmers in isolated areas? How are existing systems for food and the systems for feed production and processing linked? Routes need to be identified to develop cheap and robust, but efficient, systems that do not threaten the position of vulnerable groups.

Priorities for research should be linked to those for-mulated for crop applications (biobased products, biofu-els). As was discussed above, a combination of market size, price, and perspectives for competitive feedstock produc-tion must be considered against the needs and possibilities of rural poor plus farmers. Extra attention here is needed to link existing crop production with biorefinery systems and to define systems that are best fit to serve local needs, while preserving fragile social and ecological systems.

discussionFrom a value chain and systems perspective, the bio-

based economy opens a range of research and develop-ment issues, which can be grouped into four principal emphases along a “U impact pathway/value chain” frame-work (Dixon et al., 2007). These are (i) consumer prefer-ences, (ii) process engineering, (iii) socioeconomics, and (iv) production.

First, market research needs to be done on consumer preferences for biobased products. While of less impor-tance when such products are used in intermediate steps, consumer acceptance of innovative biobased end products will need to be assured. In many cases consumers may prefer biobased products to petroleum products provided quality is not compromised. While there has been con-sumer resistance to GM products in many countries, this may be different for non-food applications.

Related to this, and moving down the produce value chain, a strong growth in process engineering research may be expected, especially related to process efficiency,

food safety, and risk assessments. In this connection, good agricultural practices (GAP) promoted in agriculture and processing by FAO are of significance. Public-private part-nerships could coalesce around improved crop cultivars and production practices for smallholder farmers. Research could also refer to ex ante impact assessment with a particu-lar focus on equity outcomes of biobased economy.

Third, in relation to sustainable development, research on production and processing scale is important. In the early stages of the development of a biobased economy, downsizing processing technologies and plant sizes to the local level will contribute to three economic drivers. A larger share of the farm population will have the opportu-nity to grow feedstocks, including those in poor marginal environments. Local employment opportunities will be created in regions lacking in development opportunities, while short feedstock value chains may raise farm gate feedstock prices.

Related to this is the organization of processing facili-ties. Economies of scale may be expected to apply to bio-based production as they do to food production chains. Feedstock production may in many instances be expected to cluster biorefineries, which may often be located in higher production areas (e.g., irrigated areas), thus indi-rectly leading to negative equity outcomes. Application of remote and local pre-treatment units linked to central processing facilities may provide an interesting alternative for this problem (Clark and Deswarte, 2008).

On the other hand, by-products can stimulate the development of enhanced secondary income generating opportunities, e.g., distiller’s grains as concentrate feed for animal fattening activities. Therefore, livestock extension and improvement may well be an ideal complement to biorefinery development.

A fourth major research area will be sustainable feedstock production practices. Increasing demand for agricultural products may cause food prices to increase, increasing income and land values for large farmers and reducing net income/increasing food insecurity for the majority of small farmers who are net purchasers of food. The tendency for expansion of production onto marginal land will threaten soil health, thus requiring two major

Table 2. Main characteristics of major biorefinery types.†

Biorefinery Feedstock & conversion Impacts RemarksWhole crop biorefinery

Cereal crops, dry or wet milling

Link to monomer and polymer production, but large scale production leads to competition with food production. Straw applicable to lignocellulosic biorefinery.

Mainly from maize, wheat. Moder-ately capital intensive.

Oleochemical feed-stock biorefinery

Oil crops (rape seed, soy-bean, oil palm)

Links to production of chemicals, functional monomers, lubri-cants, and surfactants. Direct competition with food.

Close to full commer-cialization. Capital intensity is moderate.

Lignocellulosic biorefinery

Lignocellulosic crops, resi-dues of food & feed crops

Reduced competition with food, feed production, high water use efficiency, high potential for GHG emission reduction.

Not yet on commercial scale. Capi-tal intensive.

Green Biorefinery Mainly grass Links to production of proteins, sugars and fibers. No direct competition with food.

R&D phase.

†Source: Kamm et al. (2006), Wolf et al. (2005), De Jong et al. (2009).


research thrusts: first, conservation agriculture systems that maintain soil cover, increase water use efficiency and reduce soil erosion; second, the substitution of perennials for annual feedstocks for similar reasons. The latter may lead to increases in agroforestry or mixed food-feedstock-livestock plantation systems, which can provide low cost and reliable biomass and avoid annual cultivation and management costs while stabilizing standing biomass in times of drought, plus self-evident advantages in relation to habitat and biodiversity.

The increased demand for biomass may lead to increased harvest of crop residues, including straw and stover, as feedstock, whether for biofuels, biobased prod-ucts, or biorefineries. The removal of a high fraction of crop residues could lead to a shortage of fodder for rumi-nants, and reduce practices of mulching systems, which protect the soil surface, reduce erosion, reduce weed pres-sure, and improve water productivity. In this respect, the proportion of crop residues required to maintain soil health should be determined (Sayre and Dixon, 2006). There is sufficient potential to redirect nutrients contained in byproducts from bioenergy and biorefinery systems to farmers’ fields. Anex et al. (2007) report a potential return rate of 78% of applied N in maize or switchgrass (Panicum virgatum) production.

concluding remarksThere is a huge potential for agriculture as we move

toward a biobased economy. To capitalize on this, sys-tems and participatory approaches are needed to develop agriculture practices, institutions (including markets), and processing systems. As the anticipated growth of the biobased economy will strengthen the demand for bio-mass, ecosystem function in biomass scarce regions such as South Asia may be threatened. Resource planning, including water, energy and byproducts, and associated transportation, storage, and processing infrastructure can ensure optimal supply of agricultural produce to a variety of markets.

While the food versus fuel debate has been overheated, we conclude that food security should not be threatened by the plethora of new options. Focus should be on opti-mizing economic and energetic efficiency, while pro-tecting the position of vulnerable groups in developing countries. The reality is that most poor rural households are net food consumers. In this context, one pathway to household food security and poverty reduction comprises household entitlements (income) associated with high value added products from biomass, through small-scale local biorefineries producing bioenergy or other biobased products in poor marginal and remote areas. Incentives for such decentralized investment in biorefineries would require pro-poor institutional and policy environments.

Balanced rural development will be essential to posi-tion the growth of the biobased technologies and economy in sustainable development space. Mankind requires a wide range of products from agriculture, including food, feed, and ecosystem services, alongside newer products described in this article. Experience from the green revolution sug-gests that agricultural intensification should be dispersed rather than concentrated in high potential zones; it should also be “pro-poor, pro-women, and pro-environment,” embodying sustainability and equity principles.

A focus on rural development does not imply that there is no need for technological research. The challenge is to foster an innovative biobased economy that is tech-nically feasible, profitable, and socially desirable. With respect to research, there is need to understand how to capture more value from existing crops.

There is a need to aim for ecological efficiency, but residue recovery and biomass harvest demand more of water and soil resources that are already heavily stressed. Consequently, it is important that processing must be integrated with biomass production to yield ecological improvements. Demand for biomass as a feedstock may allow a redesigning of agriculture, in terms of crops, crop-ping systems, and nutrient management.

Biorefineries will be a key component of a resil-ient and sustainable bioeconomy, preferably with viable small-scale options to foster local economic development in marginal and remote areas. The various components of healthy agricultural and industrial ecosystems need to be integrated. Biorefineries will need to be optimized so that a wider range of the ecological functions that agricul-tural and natural lands currently provide, such as nutrient cycling, carbon sequestration and the protection of water and soil resources, will be delivered. However, it should be noted that this is unlikely to happen unless appropriate economic incentives are created (Anex et al., 2007).

referencesAnex, R. 2007. Sustainability and the biorefineries of the future. p.

7–8. In D. Clayton and E. Hughson (ed.) Products from plants– from crops and forests to zero waste biorefineries. Outputs from the EPOBIO Workshop, Greece, 15–17 May 2007.

Anex, R.P., L.R. Lynd, M.S. Laser, A.H. Heggenstaller, and M. Liebman. 2007. Potential for enhanced nutrient cycling through coupling of agricultural and bioenergy systems. Crop Sci. 47:1327–1335.

Arshadi, M., and A. Sellstedt. 2008. Production of energy from biomass. p. 143–178. In Clark and Deswarte (ed.) Introduc-tion to chemicals from biomass. Wiley series in renewable resources. Wiley, Padstow, UK.

Boerjan, W. 2009. Transgenic poplar for pulp and biofuels. Fifth Inter-national Conference on Renewable Resources and Biorefiner-ies, Ghent, Belgium. Available at http://www.rrbconference.org/bestanden/downloads/254.pdf (verified 14 Jan. 2010).

Carlsson, A. 2007. Optimising the plant oil platforms. p. 20–23. In D. Clayton and E. Hughson (ed.) Products from plants– from


crops and forests to zero waste biorefineries. Outputs from the EPOBIO Workshop, Greece, 15–17 May 2007.

Carole, T.M., J. Pellegrino, and M.D. Paster. 2004. Opportunities in the industrial biobased products industry. Appl. Biochem. Biotechnol. 113–116:872–885.

Clark, J.H., and F.E.I. Deswarte. 2008. The biorefinery con-cept– an integrated approach. p. 1–20. In J.H. Clark and F.E.I. Deswarte (ed.) Introduction to chemicals from biomass. Wiley series in renewable resources. Wiley, Padstow, UK.

Clifton-Brown, J., Y.-C. Chang, and T.R. Hodkinson. 2008. Mis-canthus: Genetic resources and breeding potential to enhance bioenergy production. p. 273–294. In W. Vermerris (ed.) Genetic improvement of bioenergy crops. Springer, Dordrecht.

De Jong, E., R. van Ree, J. Sanders, and J.W.A. Langeveld. 2009. Biorefinery. p. 111–130. In H. Langeveld et al. (ed.) The bio-based economy: Biofuels, materials, and chemicals in the post-oil era. Earthscan, London.

Dixon, J., J. Hellin, O. Erenstein, and P. Kosina. 2007. U-impact pathway for diagnosis and impact assessment of crop improve-ment. J. Agric. Sci. 145:195–206.

Dornburg, V., A. Faaij, M. Patel, and W.C. Turkenburg. 2006. Economics and GHG emission reduction of a PLA biorefinery system: Combining bottom-up analysis with price elasticity effects. Resour. Conserv. Recycl. 46:377–409.

Govaerts, B., N. Verhulst, A. Castellanos-Navarrete, K.D. Sayre, J. Dixon, and L. Dendooven. 2009. Conservation agriculture and soil carbon sequestration: Between myth and farmer real-ity. Crit. Rev. Plant Sci. 28:97–122.

Graham, I. 2007. Oil crops: Biofuels and beyond. p. 10–11. In D. Clayton and E. Hughson (ed.) Products from plants– from crops and forests to zero waste biorefineries. Outputs from the EPOBIO Workshop, Greece, 15–17 May 2007.

Koutinas, A.A., C. Du, R.A. Wang, and C. Webb. 2008. Pro-duction of chemicals from biomass. p. 77–102. In Clark and Deswarte (ed.) Introduction to chemicals from biomass. Wiley series in renewable resources. Wiley, Padstow, UK.

Johansson, D. 2000. Renewable raw materials. A way to reduced greenhouse gas emissions for the EU industry? DG Enter-prise, Brussels.

Kamm, B., M. Kamm, P.R. Gruber, and S. Kromus. 2006. Biore-finery systems, an overview. p. 3–40. In B. Kamm et al. (ed.) Biorefineries– Industrial processes and products. Status quo and future directions. 2 Volumes. Practical approach book. Wiley, Weinheim.

Lazerri, L. 2009. Green chemistry: An agricultural production chain for biolubricants. Presentation held at the Brainstorm-ing Workshop on green chemistry future and its possible

impact on agriculture: The whole farm traceability approach. Ispra, 22 Jan. 2009.

Nowicki, P., M. Banse, Chr. Bolck, H. Bos, and E. Scott. 2008. Biobased economy. State-of-the-

art assessment. LEI, The Hague. Ranalli, P. (ed.). 2007. Improvement of crop plants for industrial

end uses. Springer, Dordrecht.Sanders, J.P.M., E.L. Scott, R.A. Weusthuis, and H. Mooibroek.

2007. Biorefinery as the bio-inspired process to bulk chemi-cals. Macromol. Biosci. 7:105–117.

Sayre, K., and J. Dixon. 2006. The crop residue utilization dilemma: Sustain existing livelihoods, generate bioethanol, or improve the soil and crop production? Distributed paper, GFAR APAARI CIMMYT IFPRI Workshop, Bioethanol, Maize and Wheat: Opportunities and Risks (4 November), New Delhi.

Sijbesma, F. 2009. DSM and White Biotechnology. Presentation held at BioVision, Lyon, 8 Mar. 2009. Available at http://www.dsm.com/en_US/downloads/media/biovision_2009_white_biotech.pdf (verified 21 Dec. 2009).

Somerville, Chr.R., and D. Bonetta. 2001. Plants as factories for technical materials. Plant Physiol. 125:168–171.

Thoen, J., and R. Busch. 2006. Industrial chemicals from bio-mass– Industrial concepts. p. 347–366. In B. Kamm et al. (ed.) Biorefineries– Industrial processes and products. Status quo and future directions. 2 Volumes. Practical approach book. Wiley, Weinheim.

Turley, D. 2008. The chemical value of biomass. p. 21–45. In Clark and Deswarte (ed.) Introduction to chemicals from biomass. Wiley series in renewable resources. Wiley, Padstow, UK.

Vaca-Garcia, C. 2008. Biomaterials. p. 103–142. In Clark and Deswarte (ed.) Introduction to chemicals from biomass. Wiley series in renewable resources. Wiley, Padstow, UK.

Van Beilen, J.B., R. Möller, M. Toonen, E. Salentijn, and D. Clayton. 2007. Industrial crop platforms for the production of chemicals and biopolymers-Outputs from the EPOBIO project. CPL Press Science Publishers, Newbury. Available at http://www.epobio.net/workshop0705/presentations/wp_Biopolymers.pdf (verified 21 Dec. 2009).

Van Haveren, J., E.L. Scott, and J.P.M. Sanders. 2007. Bulk chem-icals from biomass. Biofuels Bioproducts Biorefining 2:41–57.

Wolf, O., M. Crank, M. Patel, F. Marscheider-Weidemann, J. Schleich, B. Hüsing, and G. Angerer. 2005. Techno-eco-nomic feasibility of large-scale production of bio-based poly-mers in Europe. EUR 22103 EN. Joint Research Centre. Institute for prospective technological studies.


SYMPOSIA

The 20th century has been an era of petro-chemistry that has provided society not only with energy, but with vast num-

bers of other products that have influenced almost every facet of modern life. A tremendous increase in energy demand—over 50% by 2025—is projected, with much of the demand emerg-ing from the rapidly developing nations (Ragauskas et al., 2006). There is a growing concern, on the one hand, for the dwindling supply of fossil resources and high energy prices, and, on the other hand, for the deleterious consequences of climate change related primarily to greenhouse gas emissions from the burning of fossil oil and unsustainable means of production. This has triggered a global development toward shifting the dependence from fossil to alternative cleaner renewable resources. The 21st century has started with a promise to open up new avenues for increasing the use of renewable resources in the global economy.

A major part (up to 90%) of fossil oil is used for the produc-tion of power, heat, and transport fuel, while only a few per-cent in the form of naphtha finds its way for the manufacture of chemicals and materials (U.S. Department of Energy, 2006). The latter, however, accounts for an impressive added economic value—close to half of that generated by the petroleum indus-try. While a number of renewable options, namely solar, wind, hydroelectric, and biomass, are available for the generation of electricity and heat, biomass is the only current renewable source that can be used for the production of liquid and gaseous transport

Biorefineries– A Path to Sustainability?

Rajni Hatti-Kaul*

ABSTRACTBiorefining of crops for production of power,transportfuels,andadiversearrayofchemicalshas potential for providing significant addedeconomicvaluetobiomass.Ashiftintheindus-trialresourcebasefromfossilresourcestobio-massalsorequiresashiftinthetechnologybaseforproducing,handling,andprocessingofrawmaterials.Biotechnologywillplayanimportantroleinprovidingtoolsfordifferentstagesrang-ing from biomass production, treatment, andvalorization to various products. First genera-tion refinerieshave raisedsomecritical issuesrelatedtolanduseandinsufficientenvironmen-tal benefits due to energy-intensive cultivationofcrops.Theabundantresidual lignocellulosicbiomasswillconstituteanimportantfeedstockforthefuturebiorefineriessoastohaveamini-mal impacton the foodavailability.Necessaryinvestments in technological development willbeneededtorealizethebenefitsofthenewbio-economyinthelongterm.

Dep. of Biotechnology, Lund Univ., Box 124, SE-221 00 Lund, Swe-den. Received 2 Oct. 2009. *Corresponding author ([email protected]).




fuel, as well as for chemicals and materials. Hence, when moving from a fossil- to a bio-based economy, integrated production of chemicals and materials with that of bioen-ergy is essential to maximize the value generated besides minimizing the carbon footprint.

BIOMASS-BASED REFINERIESPlant matter constitutes the most abundant source of

biomass on earth and is produced through photosynthesis. The number and variety of products that have been based on biomass during the petrochemical era are rather mod-est. These include mainly natural fibers and other natural products used in paints, soaps, adhesives, lubricants, inks, polymers and resins, and in the production of several anti-biotics, drugs, amino acids, etc. Currently, the largest bio-based products are the biofuels. A shift in the industrial resource base from fossil to biomass will require develop-ment of effective biorefinery systems, analogous to petro-leum refineries, centered on an agricultural or forest base. In this case the biomass is separated into different fractions for production of transport fuel, electricity, and various chemicals. The commercialization of the biorefineries will be determined by exploitation of the full potential of the biomass to utilize the spectrum of complex organic macromolecules (carbohydrates, oils, proteins, and lignin) and also other chemical constituents such as antioxidants and pigments present therein. Exploiting each one of these components would lead to the production of a multitude of products ranging from high volume-low market value to a low volume-high market value. These include commod-ity products, such as biofuels and biomaterials, platform chemicals (such as lactic acid and succinic acid), specialty chemicals (such as biosurfactants and biolubricants), and fine chemicals. Maximizing the usage of biomass will also lead to the improvement of process economics and waste minimization (Koutinas et al., 2007).

Viewed from a different perspective, replacement of petroleum by biomass feedstocks would reduce the depen-dence on imported nonrenewable petroleum, the supply of which is normally governed by political, economic, and ecological factors. Utilization of domestically produced biomass would provide markets and increased net income for the agriculture/forestry sectors and also more employ-ment opportunities. This trend will potentially provide new opportunities for developing countries and lead to better use of their resources and reduced impact on the environment.

TECHNOLOGIES FOR BIOREFINERIESDue to the differences in the nature and composition

of fossil- and biological feedstock, a shift in the technol-ogy base is needed. A combination of physical, chemical, and biological processing technologies is foreseen for pro-ducing, handling, and processing of raw materials. Bio-technology will play an important role in providing tools

for different stages of a biorefinery starting from feedstock development, hydrolysis of complex biomass components to simpler molecules, and the production of biofuels and chemicals. The possibility to use both traditional breeding and molecular biology techniques for increasing yield per-formance of plants, and to alter their composition, has no parallel in petroleum refining. Improvement, with regard to biomass production, could involve development of tai-lored perennial plants. This might include desirable physi-cal and chemical traits such as increased concentrations of a certain component, increased resistance to drought and pathogens, and reduced fertilizer needs. Increasing the biomass yield may be achieved by the manipulation of photosynthesis to increase the initial capture of light energy that, at present, is less than 2%, and by extending the growth phase of plants, and by manipulating nitro-gen metabolism (Ragauskas et al., 2006). Development of genetically engineered microorganisms will be needed for improving fermentation productivity and minimizing formation of by-products during the fermentation process in the production of biofuels and chemicals. Access to effi-cient and stable enzymes will allow energy-efficient bio-mass treatment and chemical production.

The technologies for initial processing of biomass should, ideally, be such that they can be adapted to dif-ferences in the quality of the raw material from different crops. As biomass can contain large amounts of water and is also sensitive to environmental conditions, efficient means to reduce the water content and convert the feedstock to a form that is stable during storage under ambient conditions before further processing will be needed. Furthermore, improvements in some existing process technologies, in terms of energy efficiency and yields, and for recycling of wastes, are desirable for a successful biorefining.

FIRST GENERATION BIOREFINERIESA first generation of fuels (ethanol and biodiesel) and

chemicals (polylactic acid) is currently produced from starch-, sugar-, and vegetable oil-feedstocks. Bioethanol is produced from sugarcane (Saccharum spp. L.) or corn (Zea mays L.), and biodiesel from soybean [Glycine max (L.) Merr.], rapeseed (Brassica spp.), or palm (Cocos nucifera L.) oil. Although an enormous expansion of the biofuel market has occurred exceeding an estimated 53 billion L in 2007, these biorefinery operations are not yet designed to make optimal use of biomass, minimize energy input, recycle wastes, and to generate good net income (Koutinas et al., 2007). No valuable components have been isolated and the residual by-products are mainly used as low-cost animal feed or in foods and in some cases for providing process energy. An important by-product of the biodiesel, as well as ethanol production, is glycerol, which has the potential to be used as a platform for a number of specialty and high-value chemicals (Werpy and Petersen, 2004).


crops such as willow (Salix spp.) and eucalyptus (Eucalyp-tus globulus), and herbaceous plants, e.g., perennial grasses like switchgrass (Panicum virgatum L.) and bermudagrass (Cynodon dactylon [L.] Pets.), are considered to be the most promising source of biomass for energy production (McK-endry, 2002; Keoleian and Volk, 2005).

The common feature of the above biomass materials is that they are lignocellulosic in composition and their global production is estimated at 3 to 5 Gt/yr, which represents 10 to 20% of today’s world energy demand (Lange, 2007). They are available at a fraction (one-fifth to one half ) of petroleum costs on an energy basis. Rice straw, wheat straw, bagasse, and corn stover are the major residues generated in developing countries (Dale and Kim, 2006). Over 700 × 109 kg of rice straw is produced worldwide, mostly in Asia and much of it is burned in the fields. Almost 200 × 109 kg bagasse is collected annually and is used as a low-cost fuel in the sugar industry. The possibility to use such abundant and inexpensive residual materials as feedstocks will also result in greater net savings in energy and CO2 emissions as energy for cultivation need not be accounted for, as it has been already recovered from the land.

The main bottleneck for the use of lignocellulosics, in contrast to starch and sucrose, is its recalcitrant nature, neces-sitating feedstock pretreatment to get access to the com-ponents. About 65 to 85% of the lignocellulose is made of cellulose and hemicellulose that acts as supporting or scaf-fold structures in plants (Kamm et al., 2006). Cellulose is the predominant polysaccharide with a non-branched crystalline structure composed of glucose residues, while hemicelluloses occur in close association with cellulose and constitute an average of about 20 to 30% of the biomass. They are het-erogeneous, branched polysaccharides containing C5 sugars (like xylose, arabinose) along with C6 sugars and sugar acids. Lignin accounts for 10 to 20% of the lignocellulosic mate-rial. It is covalently associated with hemicellulose and forms a permanent bonding agent between cellulose fibers in plants. Efficient and cost-effective separation of these components will lead to a breakthrough in biorefinery development.

INTEGRATING ENERGY AND CHEMICALS PRODUCTION

Some alternative approaches are possible for co-pro-duction of energy and chemicals from biomass. One fre-quently advocated technology is gasification of biomass to syngas (a mixture of H2, CO, CO2, CH4, and N2) which can be used as a platform for production of hydrogen and synthetic gasoline as fuels, and of methane and metha-nol as building blocks for a large number of chemicals. Such a technology is less sensitive to specific substrates converting it into the simplest forms irrespective of inher-ent energy. The other approach is to separate and utilize different components of the biomass individually for pro-duction of bioenergy and chemicals.

The biorefineries based on agricultural crops have fur-thermore raised some critical social and ecological issues. They compete with food for the feedstock and the land used for growing the crop. Due to increased demand for biofuel production, the annual increase in grain consumption since 2005 has risen from 20 million to 50 million t (Bourne, 2009). Diverting the crops for biofuel production is seen as an important factor underlying the drastic increase in food costs during 2008. Increasing demand for food, feed, and biofuels has also been a major cause of deforestation in the tropics; Brazil alone increased its soybean plantations in the Amazon 10% a year from 1990 to 2005 (Bourne, 2009). Yet another important fact is that potential savings in CO2 emission and fossil fuel consumption has been compensated for by the energy-intensive cultivation and processing of the crops. The use of significant amounts of nitrogen fertilizer, herbicides, and pesticides during cultivation of corn further contributes to ground and river water pollution (Pimentel and Patzek, 2005).

THE FUTURE BIOREFINERIESThe existing biorefineries will continue to grow and

their future development will be geared toward increased sustainability by maximizing the utilization of biomass, improving energy efficiency, and being less wasteful. For new biorefineries the selection of suitable biomass resources is extremely critical for them to operate in sym-biosis, rather than in competition, with the food sector and other markets (e.g., those based on forestry). Further, biomass resources should have an ecological perspective regarding the minimal need for water, fertilizers, and other resources for its production. In some countries, biorefin-eries are being fueled by alternative crops such as sweet sorghum [Sorghum bicolor (L.) Moench] and Jerusalem arti-choke (Helianthus tuberosus L.; JA), which are fast growing under varying climatic conditions with low water/chemi-cals input and are expected to provide good economies in the long term (Peters, 2006; Grassi and Sénéchal, 2007). The latter is able to grow in saline and desert areas and is not used for human consumption, since its main polysac-charide, inulin (a fructose polymer), is not easily digested.

In general, the ideal feedstock for a biorefinery is con-sidered to be the residual organic by-products accumulated in abundance and with low or no profit and nutritional value. These include agricultural food and feed crop resi-dues, wood and wood wastes and residues, grasses, dedi-cated energy crops and trees, plants (including aquatic plants), animal wastes, municipal wastes, and agricultural or industrial waste streams. Examples of the agricultural byproducts are wheat (Triticum aestiwm L.) and rice (Oryza sativa L.) straw, sugarcane bagasse, corn stalks, and soybean residues, whereas large industrial waste streams originate from paper making, food processing, and even the biofuel industry. Among the dedicated crops, short chain woody


Two routes can be followed for using biomass feed-stock for the chemical industry. One is to follow the struc-ture of the petrochemical industry by manufacturing a set of platform chemicals that serve as building blocks for secondary products, which in turn can be used to form other products for different applications. The other route is to target new chemicals that provide environmentally benign substitutes for the current petrochemicals. It is claimed that many of the chemical building blocks that are obtained today from fossil resources can also be pro-duced from biological feedstocks (Werpy and Petersen, 2004; Frost, 2005; Christensen et al., 2008).

Oil feedstocks have been and will remain important raw materials for specialty products like polymers, biolubri-cants, biosurfactants, and emulsifiers, in addition to biodiesel. However, due to their sheer abundance, carbohydrates are the natural choice as feedstock for biorefinery development. The coming years will see considerable development in pro-cess technologies for energy and chemical production from carbohydrates. The most abundant sugar, glucose, is the key basic chemical around which a complex biorefinery network will evolve. Glucose is currently obtained from starch, but it will potentially be available from cellulose once the technol-ogy for its hydrolysis is developed. Among the important pro-cesses are microbial fermentations for conversion of glucose and other sugars to a large number of products. These include energy carriers like ethanol and butanol, organic acids, such as lactic acid, succinic acid, and itaconic acid, bioplastics, such as polyhydroxyalkanoates, etc. The fermentation residues are often used as animal feed, but could be subjected to anaerobic digestion to produce biogas that can be used for heating or as a gaseous transport fuel. The availability of ethanol makes it interesting for use as a platform for producing several bulk chemicals like ethylene, acetic acid, and butadiene. Ethylene, in turn, can be used as a raw material for benzene, toluene, and xylene production. Organic acids, on the other hand, are good precursors for polymers, solvents, etc.

A variety of chemical technologies are available for upgrading the sugars to different product groups with broad application potential. Examples are hydrogenation of sugars to C5-C6 polyols (e.g., xylitol, mannitol, and sorbitol) and hydrogenolysis to C2-C3 glycols, e.g., eth-ylene and propylene glycol. Acid-catalyzed dehydration of hexoses yields 5-hydroxymethylfurfural (HMF) and levulinic acid, and of pentoses to furfurals, which provide economic routes to furfural and furan chemistry and a variety of polymers (e.g., nylon, polyalcohols, polyesters, polyamides, and furan resins [Werpy and Petersen, 2004]). Some of the other promising product lines that can be developed from sugars are pyran building blocks, unsatu-rated N-heterocycles, and aromatic chemicals, which are potential precursors for fine chemicals and pharmaceuti-cals. Enzymatic catalysis may also be used to catalyze spe-cific reactions at the different positions of sugars.

With future development of biorefineries to process lignocellulosic feedstocks, the amount of lignin poten-tially available for conversion into value-added products, rather than its fuel value, will be enormous. The partially hydrolyzed lignin has excellent properties for use as sub-stitutes for phenol-formaldehyde resins, polyurethane foams, adhesives, insulation materials, rubber processing, antioxidants, and dispersants for dyes, herbicides, pesti-cides, and fungicides. It provides a cheap source for high-value products like vanillin and syringol for the flavor and fragrance industry, and syringaldehyde for use as hair and fiber dye, as well as a pharmaceutical precursor (Eckert et al., 2007). Some potentially interesting new markets for lignin are the production of printed circuit boards for the electronics industry and low cost carbon fibers for use in automobile and light truck body components (Pye, 2006).

A variety of other relatively minor components of lig-nocellulosics, such as proteins, terpenic oils, fatty acids/esters, and inorganic materials, will also become available for different markets. For example, the amount of pro-tein generated will be much greater than that required for meeting the nutritional requirements of the human population. In addition to their conventional application as animal feed, protein residues can target other end uses including nutrient additives for microbial fermentations and use of amino acids as building blocks for functional-ized chemicals such as amines (Sanders et al., 2007).

THE NEW BIOECONOMY AND THE DEVELOPING COUNTRIES

Biorefineries for integrated production of bioenergy, chemicals, and materials hold promise for both short- and long-term sustainability for developing countries. Among the most successful examples of first generation biorefiner-ies occurred in Brazil, which is leading worldwide in the production of ethanol, providing some 18% of the coun-try’s automotive fuel. As the use of biomass and imple-mentation of biorefineries will increase with time, the concern with providing food as well as bioproducts, while maintaining productive soils and effective infrastructure, will become more and more important.

Increasing food prices over the years indicates a rela-tively stagnant agricultural productivity that cannot satisfy the increasing consumption trends of neither a grow-ing population nor an increasing prosperity in countries like China and India (Bourne, 2009). Several countries face deteriorating farming conditions as a consequence of agricultural practices involving overuse of fertilizers, pesticides, and irrigation, and removal of all crop residues from the fields. This has resulted in increased salinization and water logging of soils, and contamination of ground water. The increasing threat of climate change—with hot-ter seasons and increased water scarcity—is projected to reduce future harvests in many parts of the world. Better


farming practices to maintain productive soils and crops with the desirable traits are thus needed to cope with the demands of the growing population as well as to sustain the bioeconomy.

Establishment of the petrochemical industry has been supported by the development of an effective infrastruc-ture of petroleum refineries, and transport and distribu-tion network with low production costs. There is a similar requirement to obtain the maximum value from bio-based production. The limiting factors here are that the biomass feedstocks are generated discontinuously, are perishable, and require large transport volumes (Narodoslawsky et al., 2008). In many developing countries, it is very difficult to orga-nize a good raw material supply because of poor roads and storage infrastructure, and also little incentives for farmers to improve their yields. Initial processing of biomass on a small scale close to the harvest location has been proposed to coin-cide with its production and would provide benefits in terms of minimal transportation, better recycling of minerals, new forms of integration in energy utilization, and labor organi-zation (Sanders et al., 2007). Production of an intermediate product containing less water and a better shelf life could be transported to larger processing plants and would enable production throughout the year, leading to lower capital and labor costs and better prices for by-products. This would potentially give a higher return to the farmers.

To take advantage of the potential existing in the locally produced biomass, a major challenge for the developing countries will be to participate in technology development and application. This would include both improving the biomass productivity and subsequent valo-rization of the biomass.

CONCLUSIONSBiorefining of crops into multiple products, including

energy, chemicals, and materials, will increase the overall value of biomass. The maximum potential of the biorefin-ery will be realized by advances in biotechnology, plant genetics, separation technologies, process chemistry, and engineering. To be sustainable, and also to avoid increases in the raw material prices, the future biorefineries would have to concentrate on lower quality renewables such as grasses, harvest residues from crops, by-products, and wastes from the food industry, forestry, or society. Fur-thermore, recycling of waste would be needed to make the entire process carbon-neutral. As in other technologi-cal fields, participation in the new bioeconomy will be uneven and limited to those countries that make the nec-essary investments in technological development.

ACKNOWLEDGMENTSThe financial support from the Foundation for Strategic Environ-mental Research (Mistra) and the Swedish Governmental Agency for Innovation Systems (Vinnova) is gratefully acknowledged.

REFERENCESBourne, J.K., Jr. 2009. The end of plenty. Nat. Geogr. 215:26–59.Christensen, C.H., J. Rass-Hansen, C.C. Marsden, E. Taarning,

and K. Egeblad. 2008. The renewable chemicals industry. ChemSusChem. 1:283–289.

Dale, B.E., and S. Kim. 2006. Biomass refining global impact– the biobased economy of the 21st century. p. 41–66. In B. Kamm et al. (ed.) Biorefineries– Industrial processes and products: Status quo and future directions. Vol. 1. Wiley–VCH, Weinheim.

Eckert, C., C. Liotta, A. Ragauskas, J. Hallet, C. Kitchens, E. Hill, and L. Draucker. 2007. Tunable solvents for fine chemi-cals from the biorefinery. Green Chem. 9:545–548.

Frost, J.W. 2005. Redefining chemical manufacture. Indust. Bio-technol. 1:23–25.

Grassi, G., and S. Sénéchal. 2007. Techno economic feasibility of bio-energy complex in developing countries. 15th European Biomass Conference & Exhibition, 7–11 May 2007, Berlin, Germany.

Kamm, B., M. Kamm, M. Schmidt, T. Hirth, and M. Schulze. 2006. Lignocellulose-based chemical products and product family trees. p. 97–149. In B. Kamm et al. (ed.) Biorefiner-ies– Industrial processes and products: Status quo and future directions. Vol. 2. Wiley–VCH, Weinheim.

Keoleian, G.A., and T.A. Volk. 2005. Renewable energy from willow biomass crops: Life cycle energy, environmental, and economic performance. Crit. Rev. Plant Sci. 24:385–406.

Koutinas, A.A., R.-H. Wang, and C. Webb. 2007. The biochem-urgist– bioconversion of agricultural raw materials for chemi-cal production. Biofuels Bioprod. Bioref. 1:24–38.

Lange, J.-P. 2007. Lignocellulose conversion: An introduction to chem-istry, process, and economics. Biofuels Bioprod. Bioref. 1:39–48.

McKendry, P. 2002. Energy production from biomass (part 1): Overview of biomass. Bioresour. Technol. 83:37–46.

Narodoslawsky, M., A.N. Schmidinger, and L. Halasz. 2008. Util-ising renewable resources economically: New challenges and chances for process development. J. Clean. Product. 16:164–170.

Pimentel, D., and T.W. Patzek. 2005. Ethanol production using corn, switchgrass, and wood; biodiesel production using soy-bean and sunflower. Nat. Resour. Res. 14:65–76.

Peters, D. 2006. Carbohydrates for fermentation. Biotechnol. J. 1:806–814.

Pye, E.K. 2006. Industrial lignin production and application. p. 165–200. In B. Kamm et al. (ed.) Biorefineries– Industrial processes and products: Status quo and future directions. Vol. 2. Wiley–VCH, Weinheim.

Ragauskas, A.J., C.K. Williams, et al. 2006. The path forward for biofuels and biomaterials. Science 311:484–489.

Sanders, J., E. Scott, R. Weusthius, and H. Mooibroek. 2007. Bio-refinery as the bio-inspired process to bulk chemicals. Macro-mol. Biosci. 7:105–117.

U.S. Department of Energy. 2006. International Energy Outlook. Available at http://www.eia.doe.gov (verified 21 Dec. 2009). Energy Information Administration, Washington, DC.

Werpy, T., and G. Petersen (ed.) 2004. Top value added chemicals from biomass. Volume I– Results of screening for potential candidates from sugars and synthesis gas. U.S. Department of Energy, Washington, DC.

cgiar science forum 2009 - IPC Global Platform

Documents

Transcript of cgiar science forum 2009 - IPC Global Platform