a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2

avai lable at www.sc iencedi rec t .com

journal homepage: www.e lsev ier .com/ locate /agwat

Review

Non-conventional water resources and opportunities forwater augmentation to achieve food security inwater scarce countries

M. Qadir a,*, B.R. Sharma b, A. Bruggeman a, R. Choukr-Allah c, F. Karajeh d

a International Center for Agricultural Research in the Dry Areas (ICARDA), P.O. Box 5466, Aleppo, Syriab International Water Management Institute (IWMI), Asia Regional Office, CG Block, NASC Complex, DPS Marg,

Pusa Campus, New Delhi 110 012, IndiacSalinity and Plant Nutrition Laboratory, Institut Agronomique et Veterinaire Hassan II, B.P. 773 Agadir, MoroccodDepartment of Water Resources, Recycling and Desalination Branch, Office of Water Use Efficiency,

901 P Street, P.O. Box 942836, Sacramento, CA 94236-0001, USA

a r t i c l e i n f o

Article history:

Accepted 21 March 2006

Published on line 24 May 2006

Keywords:

Water scarcity

Seawater desalination

Rainwater harvesting

Marginal-quality water

Wastewater

Saline–sodic water

Virtual water

Water transportation

a b s t r a c t

Given current demographic trends and future growth projections, as much as 60% of the

global population may suffer water scarcity by the year 2025. The water-use efficiency

techniques used with conventional resources have been improved. However, water-scarce

countries will have to rely more on the use of non-conventional water resources to partly

alleviate water scarcity. Non-conventional water resources are either generated as a

product of specialized processes such as desalination or need suitable pre-use treatment

and/or appropriate soil–water–crop management strategies when used for irrigation. In

water-scarce environments, such water resources are accessed through the desalination

of seawater and highly brackish groundwater, the harvesting of rainwater, and the use of

marginal-quality water resources for irrigation. The marginal-quality waters used for

irrigation consist of wastewater, agricultural drainage water, and groundwater contain-

ing different types of salts. In many developing countries, a major part of the wastewater

generated by domestic, commercial, and industrial sectors is used for crop production in

an untreated or partly treated form. The protection of public health and the environment

are the main concerns associated with uncontrolled wastewater irrigation. The use of

saline and/or sodic drainage water and groundwater for agriculture is expected to

increase. This warrants modifications in the existing soil, irrigation, and crop manage-

ment practices used, in order to cope with the increases in salinity and sodicity that

will occur.

It is evident that water-scarce countries are not able to meet their food requirements

using the conventional and non-conventional water resources available within their bound-

aries. Another option that may help to achieve food security in these countries is the

‘physical’ transportation of water and food items across basins, countries, and regions.

Long-distance movement of surface freshwater or groundwater and transporting the water

nes

int

inland via large pipeli

transportation. Most

* Corresponding author. Tel.: +963 21 2213433; fax: +963 21 2213490.E-mail address: m.qadir@cgiar.org (M. Qadir).

0378-3774/$ – see front matter # 2006 Elsevier B.V. All rights reservedoi:10.1016/j.agwat.2006.03.018

or across the sea in extremely large bags are examples of ‘physical’

erregional water transportation projects are still in their infancy,

d.

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 3

though the trade of food items between countries has been going on since international trade

began. Although food is imported in the international food trade, the water used to produce

the food that is imported into water-scarce countries is equivalent to large water savings for

those countries: without the imports, almost the same amount of water would be needed to

produce that food domestically. The term ‘virtual water’ has been used to illustrate the

important role that water plays in the trade in food between countries with a water surplus

and those with a water deficit, which must rely in part on importing food to ensure food

security. Because the major food-exporting countries subsidize their agricultural production

systems, food-importing countries need to consider both the policies and political situations of

food-exporting countries, while simultaneously using food trade as a strategic instrument to

overcome water scarcity and food deficits. This paper reviews the literature and issues

associated with the use of non-conventional water resources and opportunities for achieving

food security in water-scarce countries.

# 2006 Elsevier B.V. All rights reserved.

Contents

1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2. Non-conventional water resources in water-scarce countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.1. Desalinated seawater and highly brackish groundwater . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.2. Rainfall-runoff water captured by water harvesting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.3. Marginal-quality water resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2.3.1. Wastewater from domestic, municipal, and industrial activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2.3.2. Agricultural drainage water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

2.3.3. Saline and/or sodic groundwater . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

3. Opportunities contributing to food security in water-scarce countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3.1. Physical transportation of freshwater. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3.2. Food imports and the ‘virtual water’ nexus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

4. Future perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

1. Introduction

Looking at the natural global water cycle that yields an annual

renewable water supply of about 7000 m3 per capita (Shiklo-

manov, 2000), it is evident that there is enough freshwater

available every year to fulfill the needs of the present population

of this planet. However, in certain regions and countries the

annual renewable supply of water is less than 500 m3 per capita.

In addition, the availability of water varies greatly over time in

these areas, which results in extreme events. Floods and

droughts, for example,occur frequently, sometimes inthe same

area or neighboring regions. What this contrast illustrates is

that most of the freshwater available is concentrated in specific

regions, while other areas are water-deficient (Pimentel et al.,

1999; Rijsberman, 2006). Because freshwater resources and

population densities are unevenly distributed worldwide, water

demands already exceed supplies in regions that contain more

than 40% of the world’s population (Bennett, 2000).

Of particular concern is the situation in the Middle East and

North Africa (MENA), which ran out of renewable freshwater

decades ago, in the sense that the region is unable to meet its

food requirements using the available water resources. This

region is the driest on earth, containing only 1% of the world’s

freshwater resources. However, several countries in other

regions have also recently encountered a water deficit (Allan,

2001). And, datasets and maps published in recent years show

that more and more countries will become water-stressed

because of increased water scarcity (Seckler et al., 1998; FAO,

2003; Table 1). Bearing in mind the recent trends seen in water

withdrawal under present practices and policies, future

projections suggest that by the year 2025 about 60% of the

world’s population will have to cope with a lack of water

(Cosgrove and Rijsberman, 2000).

In order to achieve adequate living standards equivalent to

those seen in western and industrialized countries, an annual

renewable water supply of at least 2000 m3 per capita is

required (Postel, 1993). By the same token, a country with

annual renewable water supply of 1000–2000 m3 per capita

may suffer occasional or local water shortages (Bouwer, 2002).

According to Falkenmark and Lindh (1993), an annual fresh-

water supply of 1000 m3 per capita is the critical value.

Countries with less than this will suffer serious water

shortages that could impact on economic development, and

human health and well-being. Generally known as the

‘Falkenmark Indicator’, this is the most widely used water

scarcity index because it is easy to apply. It is based on an

approximate minimum level of water required per capita to

maintain an adequate quality of life in a moderately developed

country. When annual renewable water supplies fall below

500 m3 per capita, a country is likely to experience ‘absolute

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 24

Table 1 – Estimates and projections of populations and ofrenewable water resources (RWR) per capita in 2005 and2030 for countriesa that are expected to fall below thewater scarcity level (1000 m3 capitaS1 yrS1) in the year2030

Country Population(�103)

RWR(m3 capita�1

yr�1)b

2005 2030c 2005 2030d

Kuwait 2671 4198 7 5

United Arab Emirates 3106 4056 48 37

Saudi Arabia 25626 43193 94 56

Libya 5768 8123 104 74

Singapore 4372 4934 137 122

Jordan 5750 8643 157 104

Yemen 21480 50584 191 81

Israel 6685 8970 254 190

Oman 3020 5223 331 191

Algeria 32877 44120 435 324

Tunisia 10042 12351 458 372

Burundi 7319 13652 492 264

Rwanda 8607 13453 604 387

Egypt 74878 109111 779 534

Burkina Faso 13798 27910 906 448

Morocco 31564 42505 919 682

Kenya 32849 41141 919 734

Lebanon 3761 4692 1170 938

Somalia 10742 24407 1257 553

Malawi 12572 19834 1376 872

Pakistan 161151 271600 1382 820

Syria 18650 28750 1410 915

Eritrea 4456 7942 1414 793

Ethiopia 74189 127220 1483 865

Based on water resources data from FAO (2005) and population

data from WRI (2005).a Countries with limited area, population or data (e.g., Bahamas,

Qatar, Malta) were not included.b RWR include internal renewable water resources plus or minus

the flows of surface water and groundwater entering or leaving the

country, respectively.c UN medium variant population projections for the year 2030.d Based on the assumption that the RWR in the countries in 2030

will remain the same as in 2005.

water scarcity’. Besides the ‘Falkenmark Indicator’, other

water scarcity indicators exist that provide a greater insight

into the different aspects of water scarcity. However, these

indicators are not widely applied because: (1) they are more

complex, (2) the data necessary to apply them are lacking, and

(3) their definitions are not intuitive (Rijsberman, 2006).

Agriculture is the largest single user of water, with about 75%

of the world’s freshwater being currently used for irrigation. In

some countries, irrigation accounts for as much as 90% of the

total amount of water available (FAO, 2003, 2005). Given that

water productivity in agriculture continues to be low and that

improvements are only being made very slowly, and that

freshwater has always been an integral component of food

production, it is obvious that huge amounts of water will be

requiredtoproduce enoughfoodfor the futurepopulationof the

world. In addition, urbanization and increasing populations in

water-deficient countries increase the demand for freshwater.

This results in competition among different water-use sectors

and, often, in less freshwater being allocated to agriculture.

Such priority-setting results from the fact that, at all times,

public needs and people’s health have to be protected through

the use of the best quality water available. The phenomenon of

agriculture having to yield part of its share of the freshwater

available is expected to intensify in those less-developed, arid

and semi-arid countries and regions that are already suffering

water, food, sanitation, and health problems.

As the pressure placed on freshwater resources in water-

deficient regions increases, so will the need to conserve and use

conventional water resources more efficiently, because future

increases in agricultural production will have to rely heavily on

existing water resources (Oweis et al., 2000; Wallace, 2000;

Hatfield et al., 2001; Kijne et al., 2003). Such conventional

resources consist of the water available from rainfall and

snowmelt, which is used on site or taken from rivers, streams,

lakes, reservoirs, and aquifers. These resources are renewable

through the natural processes of the hydrological cycle.

In addition to conventional resources, non-conventional

water sources offer complementary supplies that can be used to

partially alleviate water scarcity in regions where renewable

water resources are extremely scarce. Such water resources are

harnessed for agricultural and other uses through specialized

processes such as desalination of seawater and highly brackish

water; harvest of rainwater; collection, treatment, and use of

wastewater; capture and reuse of agricultural drainage water;

extraction of groundwater containing a variety of salts.

Appropriate strategies for managing soil, water and crops

may also be needed when theseresources are used for irrigation

(Mohsen and Al-Jayyousi, 1999; Voutchkov, 2004; Oweis et al.,

2004; Qadir and Oster, 2004).

The ‘physical’ transportation of water and food items

across basins, countries, and regions is another option that

could help countries in the dry regions to achieve food

security. Tapping freshwater from surface water resources or

aquifers and moving the water via large pipelines or by sea in

huge bags, are examples of ‘physical’ transportation, a

definition which does not include contiguous gravity-flow

movement systems such as rivers and canals (Mohsen and Al-

Jayyousi, 1999; Ariyoruk, 2003). Most water transportation

projects are still in their infancy. However, food has been

traded internationally as long as there have been wagons,

railroads, ships, and now airplanes. Although food, not water,

is imported in the international food trade, the water used to

produce the food that is imported into water-scarce countries

is equivalent to large water savings for those countries:

without the imports, almost the same amount of water would

be needed to produce that food domestically. The term ‘virtual

water’ has been used to illustrate the importance of water in

the trade of food between water-surplus and water-deficit

countries (Allan, 1996, 2003). This paper reviews the literature

and the issues concerned with the use of non-conventional

water resources and those other opportunities that could be

used to increase food security in water-scarce countries.

2. Non-conventional water resources inwater-scarce countries

Non-conventional water resources have the potential to

augment water supplies and narrow the gap between fresh-

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 5

water availability and demand in water-scarce countries and

regions. These water resources are considered in turn below.

2.1. Desalinated seawater and highly brackishgroundwater

Seawater contains high concentrations of salts. It has

electrical conductivity (EC) levels of around 55 dS m�1 (total

dissolved solids �35,000 mg L�1) and sodium (Na+) concentra-

tion of more than 450 mmol L�1 (�10,400 mg L�1). Without

treatment to reduce its salt content, humans or animals

cannot use seawater directly for consumption, as this would

severely affect their health; nor can untreated seawater be

used to produce crops (Qadir et al., 2003). The same is true of

highly brackish groundwater containing elevated levels of

various types of salts.

Desalination, a process that converts seawater or highly

brackish groundwater into good-quality freshwater, has been

practiced for over 50 years. The scarcity of freshwater has

provided a driving force for the use of this approach in arid and

semi-arid regions and in countries bordering seas or salt lakes.

The largest producers of freshwater from seawater are the

Middle Eastern countries, including Saudi Arabia (which

produces one-tenth of world’s desalinated water), the United

Arab Emirates, Kuwait, Bahrain, Qatar, and Oman. However,

several other countries have a compelling need to desalinate

seawater and highly brackish groundwater to produce fresh-

water. These countries are not all located in the arid and semi-

arid areas: some simply have dense population concentrations

and high levels of industry and tourism, resulting in local

drinking water resources being either inadequate or becoming

unfit for consumption. Currently, desalination plants operate

in more than 120 countries worldwide (Voutchkov, 2004).

Estimates show that each day desalination plants throughout

the world produce about 30 � 106 m3 of freshwater: about

20 � 106 m3 from seawater and the remainder from highly

brackish groundwater (Pearce, 2004). This suggests that the

total amount of freshwater produced each year from desali-

nation is around 11 � 109 m3.

Since the advent of desalination, several processes have

been used to desalinate seawater or highly brackish ground-

water. Some have been disregarded because they were not

economically viable. Among the main desalination techni-

ques, distillation is the oldest form of seawater treatment, and

is a process that has been used in many parts of the world.

Desalination using distillation can be accomplished in several

different ways (Semiat, 2000). In general, the process uses heat

energy to evaporate water and thus separate it from salts and

impurities. The evaporated water is then captured and

condensed as freshwater. The process therefore recreates

the cycle of evaporation and condensation of water that

occurs naturally. However, distillation is an energy-intensive

process. Initially, therefore, distillation plants were con-

structed in areas where the energy costs were very low (such

as the oil-rich Middle East) or near to processing plants which

produced waste heat (Haddad and Mizyed, 2004).

In order to reduce energy costs, the seawater reverse

osmosis process was developed in the 1970s. This is a more

energy-efficient process, which makes use of tightly bound

semipermeable membranes, through which seawater is

forced at very high pressures. Only the water molecules are

able to pass through these membranes, as they are smaller

than almost all the impurities (including salts) contained in

seawater. The separated impurities and some residual water

are then discharged as brine, usually into the ocean. Advances

in membrane technologies have also led to the emergence of

membrane configurations with different performance para-

meters. This method of desalination became popular during

the 1990s, as it has a lower operating cost than thermal

desalination processes. Since then, the construction of new

reverse osmosis plants has accelerated significantly. Interest-

ingly, most of the large desalination plants constructed during

the last 10 years, or currently undergoing construction, are

delivered under public–private partnership arrangements

using the build–own–operate–transfer method of project

implementation. In addition, the trend has been towards

building fewer seawater desalination plants but ensuring that

those built have a large capacity, rather than building a large

number of smaller facilities with less capacity. This is a result

of the benefits offered by larger capacities and centralization

(Voutchkov, 2004).

A comparison of desalination techniques reveals that

distillation plants have an advantage in that they produce

high-quality water that usually contains in the range of

5–50 mg L�1 (mg L�1 = parts per million, ppm) of dissolved

solids in total, while reverse osmosis plants produce water

that contains between 250 and 500 mg L�1 of dissolved solids

in total (Mohsen and Al-Jayyousi, 1999; Semiat, 2000). Another

advantage of distillation plants is the fact that one does not

have to shut down a portion of their operations to clean them

or to replace equipment, something which often has to be

done in reverse osmosis plants. In addition, the pre-treatment

requirements of reverse osmosis plants are greater, because

coagulants are needed to settle out particles in the water

before it passes through the membranes and to remove

chemical constituents that could precipitate in the mem-

branes.

The reverse osmosis plants do have advantages over

distillation plants. These include the fact that feed water for

reverse osmosis plants generally does not have to be heated,

so the thermal impacts of discharges are lower. In addition,

reverse osmosis plants have fewer problems with corrosion

and usually have lower energy requirements than distillation

plants. They also tend to have higher recovery rates, which

can be as high as 45% in the case of seawater. Reverse osmosis

process can also remove unwanted contaminants such as

pesticides and bacteria. In the case of simple distillation,

chemical contaminants with boiling points below that of

water are condensed along with the water. Finally, reverse

osmosis plants take up less surface area than distillation

plants for the same amount of freshwater produced.

The cost of energy is the major cost involved in producing

freshwater through desalination using the reverse osmosis

process (Fig. 1). The overall cost of seawater desalination, once

quite high, has dropped dramatically over the years. For

example, the costs of desalination in late 1970s were around

US$ 5.5 m�3, but now vary between US$ 0.5 and 0.6 m�3,

depending on the site and size of the treatment plant.

However, the rate of incremental cost reductions has declined

over time (Ariyoruk, 2003; Pearce, 2004; Haddad and Mizyed,

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 26

Fig. 1 – Typical seawater desalination costs through

reverse osmosis in terms of percentage of the total cost

(Semiat, 2000).

2004). The cost of desalinated water is not expected to

decrease by an order of magnitude any time soon. It is

anticipated, however, that this cost will go down by

approximately 20% in the next 5 years, and that it will

decrease by close to 50% by the year 2020. This is because the

current energy use of reverse osmosis desalination systems is

between 10 and 20 kJ m�3, whereas with the newest energy-

recovery technology, which allows the reuse of a proportion of

the energy used for desalination, energy use could be reduced

to 7.2 kJ m�3. However, this value is still 44% greater than the

theoretical minimum amount of energy needed for seawater

desalination (Voutchkov, personal communication, 2005).

Nuclear and solar energy could also be used to provide the

heat needed to desalinate seawater, providing another

opportunity for the desalination of seawater on a large scale.

By the same token, wind energy is an option that could be used

in conjunction with small-scale desalination units.

Although the existing costs of desalination and desalinated

water distribution are affordable in many developed and oil-

rich countries, desalinated water is still an expensive resource

in developing countries. Bearing in mind the costs of other

conventional and non-conventional water resources, which

could be used for agriculture, the use of desalinated water for

traditional agricultural production systems remains an

expensive option, although desalinated water is being used

on a small scale for high-tech and high-value agriculture in

areas such as southeastern Spain. The costs of investments

and operation are shared equally between farmers and the

government to overcome the recurrent droughts this region of

Spain experiences. In the 1990s, the estimated financial losses

in Spain due to drought were around US$ 120 million. Since

1994, more than 90 reverse osmosis plants have been installed

in this part of Spain with a total daily desalination capacity of

0.15 � 106 m3 and an annual production rate of 55 � 106 m3 of

desalinated water (Latorre, 2002).

Desalination does have environmental implications,

because the effluents produced by desalination plants contain

highly concentrated water. In the case of seawater desalina-

tion, the brine effluent generated has a concentration of salts

close to twice that of the original seawater. The concentrate

also contains the chemicals used during the pre-treatment of

the feed water. When those concentrates are produced inland,

as occurs in the case of the desalination of highly brackish

groundwater, disposal in unlined ponds may cause the brine

to seep to the aquifer, thereby increasing groundwater salinity

(Semiat, 2000). For desalination of seawater, the solutions are

generally purged into the open sea. This warrants study of the

best options for brine disposal systems, in order to maximize

brine dilution. In addition, the possible impacts of brine on a

variety of marine habitats require regular monitoring. Studies

are underway in some countries to develop environmentally

acceptable and efficient brine disposal systems (Latorre, 2002).

It is important that adequate regulations, policies, and

practices are in place to prevent desalination having a

negative impact on adjacent areas and the wider environment.

2.2. Rainfall-runoff water captured by water harvesting

In arid and semi-arid regions, rainfall is limited and subject to

high intra- and inter-seasonal variability. In addition, poor

vegetative cover and shallow and crusting soils result in much

of the rainwater that does fall being lost through surface

runoff and evaporation. These factors provide a strong

impetus for strategies that make best use of the rainfall and

runoff water. Water harvesting is often practiced in areas

where the rain itself is not sufficient for crop production.

Water harvesting for agriculture involves collecting the

rainwater that runs off a catchment area in a reservoir or in

the root zone of a cultivated area, which is usually smaller

than the size of the catchment area (Boers and Ben-Asher,

1982). Owing to the intermittent nature of runoff events, it is

necessary to store the maximum amount of rainwater during

the rainy season so that it can be used later (Oweis et al., 1999).

Water harvesting techniques can be classified in several

ways. In general, water harvesting may be broadly classified

into: (1) macro-catchment and floodwater harvesting and

diversion methods, and (2) micro-catchments methods, where

the catchment area and the cropped area are distinct but

adjacent to each other. Boers and Ben-Asher (1982) define

micro-catchment systems as systems where the catchment

area is less than 100 m in length.

Macro-catchment techniques capture runoff water from

hillsides or small arid watersheds. Soil bunds or small dikes

can be placed on the gently sloping land below hillsides to

harvest runoff water that crops or trees can use. These

systems often have small spillways, to divert excess water

from one field to the next. Water is also sometimes captured

from small arid watersheds in wadi-beds; in these dikes are

used to capture both water and sediment, resulting in deep

fertile soils. The runoff water can also be diverted from the

wadi and used on adjacent cropland. Small storage and

groundwater recharge dams and ponds, which can provide

water for supplemental irrigation of rainfed crops, are also

considered to be macro-catchment water harvesting sys-

tems.

Different forms of micro-catchments have been used in

low rainfall regions. Contour bunds consist of earth, stone or

trash embankments placed along the contours of a sloping

field or hillside, in order to trap rainwater behind them and

allow greater infiltration. Semicircular, trapezoidal or ‘V’-

shaped bunds are generally placed in a staggered formation,

allowing water to collect in the area behind the bunds. Excess

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 7

water is displaced around the edges of the bund when the

‘hoop’ area is filled with water. These systems are most

commonly used for growing fruit trees or shrubs.

Another type of water harvesting system involving micro-

catchments is the meskat-type system. In this system, instead

of alternating catchment and cultivated areas, the field is

divided into a distinct catchment area that is located directly

above the cropped area. The catchment area is often stripped

of vegetation to increase runoff. The cultivated area is

surrounded by a ‘U’-shaped bund in order to hold the runoff.

In Tunisia, tree species such as olive (Olea europaea L.) are

grown in these systems (Oweis et al., 2001, 2004). A similar

system, called Khushkaba, is used in Balochistan, Pakistan, for

growing field crops (Oweis et al., 2001).

Rainwater can also be harvested from rooftops or from

sloping, rocky or crusting lands and collected in under- or

above-ground cisterns. These types of systems are often used

for domestic or livestock needs or for the irrigation of home

gardens or small plots of high-value crops. This may not have

a large impact of the nation’s food security but can have an

important impact on the livelihood of rural families in water

scarce areas.

Another form of water harvesting in dry areas is known as

‘fog-harvesting’ in which water from condensation is collected

on a wire screen or net. The wire net is connected to a

drainpipe, which empties into a bucket, and can be placed on

house roofs. The collected water can serve as a source of water

for human consumption in areas where running water is not

available for part of the year and fogs occur frequently.

Guidelines for implementing micro-catchment systems are

available in the relevant literature (Boers and Ben-Asher, 1982;

Critchley and Siegert, 1991; Oweis et al., 2001). In addition to

the characteristics of the rainfall, several landscape char-

acteristics also influence the efficiency of different rainwater

harvesting techniques in the same area. The following on-site

factors influence the usefulness of rainwater harvesting

systems:

1. T

Td

Y

1

1

1

M

Ma

b

he slope of the landscape, which allows runoff water to

flow down where it can be intercepted and collected (Oweis

et al., 1999).

2. T

he texture and structure of the soil. For example, the

surface of soils dominated by a sand fraction in the top layer

and a clayey subsurface can erode at an accelerated rate

able 2 – Shrub survival rate (%) within semicircular micro-catchimensions in the Mehasseh steppe of Syria that receives 120 m

ear Rainfall (mm) Survival without WHa

997–1998 174 20

998–1999 36 7

999–2000 42 2

ean 83 10

odified from Oweis and Hachum (2003).

WH = water harvesting.

Diameter of the semicircular micro-catchment water harvesting structure

when overgrazing, or some other mechanism, removes the

vegetation covering it. This can result in the clayey

subsurface, which has crusting properties and low infiltra-

tion rates, being exposed. These unproductive areas can be

used for water harvesting by inducing runoff to flow to the

cultivated area below (Tabor, 1995). When a hard layer has

not formed on the catchment area, the soil’s surface can be

treated with water-repellent materials that fill the soil pores

and help to reduce infiltration rates, so promoting runoff

(Oweis et al., 1999). However, the present costs of these

materials limit their widespread use.

3. S

oil nutrient availability status, which is essential in

enhancing the effects of water harvesting by increasing

crop yields. Increasing the nutrient availability status of the

soil can promote root development, which increases crop

water uptake and biomass production (Rockstrom and

Falkenmark, 2000). Thus the regular application of animal

manure is also important, as manure increases nutrient

levels and improves the physical condition of cultivated soil

(Tabor, 1995).

4. S

oil depth of the cultivated area. The soil must be deep

enough, and of a medium-texture, to induce rainfall

infiltration and retention. Deep soils retain greater amounts

of water for plant growth.

Micro-catchment rainwater harvesting systems are advan-

tageous because they are simple to design and cheap to install;

they are therefore easy to reproduce. They can also be adapted

for use on almost any kind of slope, including slight slopes on

almost level plains. Local people can easily be trained to im-

plement such technologies. With little conveyance losses, t-

hese systems provide higher levels of runoff efficiency than

macro-catchment water harvesting systems.

There are several recent success stories relating to micro-

catchment rainwater harvesting. For instance, in the Mehasseh

steppe in Syria, where annual rainfall is 120 mm, the survival

rate of rain-fed shrubs improved to as much as 93% through the

use of micro-catchments in the form of semicircular bunds

(Table 2). By the same token, in the Muwaqar area of Jordan

(annual rainfall 125 mm) almond trees have been grown using

micro-catchments for over 15 years without irrigation, despite

some years of drought during which rainfall dropped as low as

60 mm (Oweis and Hachum, 2003).

One disadvantage of water harvesting systems is that they

depend on limited and uncertain rainfall. Other disadvantages

ment water harvesting structures with differentm annual rainfall on average

Survival with micro-catchment WH

2 m diameterb 4 m diameter 6 m diameter

96 98 97

92 95 93

92 93 89

93 95 93

.

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 28

Table 3 – Volume of wastewater generated annually insome countries of Central and West Asia and NorthAfrica

Country Reportingyear

Wastewatervolume

(�106 m3 yr�1)

Algeria 2004 600

Bahrain 1990 45

Egypt 1998 10012

Iran 2001 3075

Jordan 2004 76

Kuwait 1994 119

Kyrgyzstan 1995 380

Lebanon 1990 165

Libya 1999 546

Morocco 2002 650

Oman 2000 78

Saudi Arabia 2000 730

Syria 2002 825

Tunisia 2001 240

Turkey 1995 2400

United Arab Emirates 2000 881

Uzbekistan 2004 170

Yemen 2000 74

Except for Algeria, Jordan and Uzbekistan; data derived from the

wastewater databases of the Food and Agriculture Organization,

available at http://www.fao.org/landandwater/aglw/waterquality/

waterusedb.jsp.

are that: (1) the catchment area is most efficient when kept

free from vegetation, which may require high labor inputs; (2)

the system may be damaged during heavy rainstorms and

need regular maintenance; (3) the catchment uses land which

could otherwise be used to grow crops, if sufficient water is

available, except in the case of steep slopes; (4) the systems

allow only low crop densities and low yields in comparison

with conventional irrigation systems (Prinz, 1996).

Various rainwater-harvesting techniques have a long

history of use, and have resulted in increases in crop yields

and water-use efficiency in different parts of the world

(Agarwal and Narain, 2003; Pandey et al., 2003; Oweis et al.,

2004). Many plant species with low to medium water

requirements have been successfully grown using different

water harvesting techniques. Owing to population pressure in

water scarce countries, interest in the use of water harvesting

and conservation strategies has been growing. For example,

estimates show that the quantities of water harvested in

Jordan have reached 6 � 106 m3 yr�1 (Jaber and Mohsen, 2001).

Although the exact amount is difficult to estimate, Oweis and

Hachum (2003) have revealed that 30–50% of the rain that falls

in the drier regions could be utilized if appropriate water

harvesting techniques were practiced. Utilizing rainwater in

this way has the potential to result in a several-fold increase in

rainwater-use efficiency in water-scarce environments.

2.3. Marginal-quality water resources

Marginal-quality waters consist of: (1) wastewater generated

by domestic, commercial and industrial uses; (2) drainage

water generated by irrigated agriculture and surface runoff

that has passed through the soil profile and entered the

drainage system; (3) groundwater from different sources, such

as underlying saline formations, seawater intrusion in coastal

areas, recharge of agricultural drainage, storm water runoff

from urban areas, and/or infiltration from wastewater-

irrigated areas. Marginal-quality waters contain one or more

impurities at levels higher than in freshwater, including salts,

metals, metalloids, residual drugs, organic compounds,

endocrine-disrupting compounds, and the active residues of

personal care products and/or pathogens. These constituents

may have undesirable effects on soils, crops, water bodies, or

human and animal health.

2.3.1. Wastewater from domestic, municipal, andindustrial activitiesPopulation growth coupled with the provision of goods and

services that allow higher living standards have increased the

demand for good-quality water to provide for the needs of the

domestic, municipal, and industrial sectors in water-scarce

countries. Consequently, greater amounts of wastewater are

being generated. After treatment, and in conjunction with

suitable management practices, this could be reused for a

variety of purposes. Urban wastewater consist of a combina-

tion of some or all of the domestic effluent produced (black

water and grey water), water produced by commercial

establishments and institutions (including hospitals), indus-

trial effluent and storm water which has not infiltrated the

soil, as well as other forms of urban runoff (Van der Hoek,

2004).

Estimates of the extent to which wastewater is used for

agriculture worldwide reveal that at least 2 � 106 ha are

irrigated with treated, diluted, partly treated or untreated

wastewater (Jimenez and Asano, 2004). The use of untreated

wastewater is intense in areas where there is no or little access

to other sources of irrigation water. Few databases are available

that describe the extent to which wastewater is used for

agriculture at the nationalor regional levels (Minhas and Samra,

2003; Van der Hoek, 2004). Owing to the variable quantities of

water available for human consumption in water-scarce

countries, estimates of the per capita generation of wastewater

vary, ranging from 30 to 90 m3 yr�1. The volumes of wastewater

generated insomecountries ofCentral and WestAsiaand North

Africa (CWANA) are presented in Table 3. A significant part of

the wastewater generated in these countries is used to

supplement the freshwater needs of a variety of crops.

The rate at which populations are increasing means that

wastewater treatment and its sustainable use is an issue that

requires more attention and investment. Most developing

countries have not been able to build wastewater treatment

plants on a large enough scale and, in many cases, they were

unable to develop sewer systems fast enough to meet the

needs of their growing urban populations. As a result, in

several countries, particularly in Sub-Saharan Africa (SSA), the

sanitation infrastructure in major cities has been outpaced by

population increases, making the collection and management

of urban wastewater ineffective. In many large cities (for

example, Accra in Ghana), only a small part of the wastewater

produced (�10%) is collected in piped sewerage systems for

treatment (Drechsel et al., 2002).

Owing to the gradual addition of contaminants into

freshwater bodies, and the awareness of their possible

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 9

impacts, wastewater treatment is now receiving greater

attention from the governments of several water-scarce

countries and organizations such as World Bank, the Food

and Agriculture Organization of the United Nations (FAO), and

the United Nations Development Programme (UNDP), among

others.

There is now more scope in the water and environment

sector to develop and implement wastewater treatment

technologies that: (1) need low levels of capital investment

for construction, operation and maintenance; (2) maximize

the separation and recovery of by-products (such as nutrients)

from polluted substances; (3) are compatible with the intended

reuse option in that they yield a product of an appropriate

quality in adequate quantities; (4) can be applied at both very

small and very large scales; (5) are accepted by farming

communities and the local population. Bearing in mind that

treated wastewater could be used for agricultural, environ-

mental, recreational and industrial purposes, it is important to

realize that such wastewater must be adequately treated and

used appropriately. This is important for several reasons:

1. T

T

C

A

B

C

Ma

b

c

sd

e

�f

sg

s

he discharge of untreated wastewater into surface water

bodies affects the quality of both the water it enters and the

water further downstream.

2. T

reated wastewater could be used to provide a reliable

source of irrigation water in urban and peri-urban areas,

providing water for parks, play and sports grounds, and

able 4 – Guidelines for microbiological qualities of treated wast

ategory Wastewater reuseconditions

Exposed group ofcommunities

I

Unrestricted irrigation

(all crops, including vegetable

and salad crops eaten

uncooked, sports fields,

public parkse)

Workers, consumers,

public

Any

Restricted irrigation

(cereal crops, industrial

crops, fodder crops,

pastures, and treesg)

B1 workers,

children > 15 years,

nearby communities

Spr

Spr

B2 same as B1 Floo

B3 workers,

children of all ages,

nearby communities

Any

Localized irrigation

(crops in category B,

but without exposure of

workers and communities)

None Tric

odified from Blumenthal et al. (2000) and Carr et al. (2004).

In specific situations, these guidelines may be modified according to local

Ascaris and Trichuris species and hookworms. The guideline values are als

During the irrigation period; routine monitoring is not required if wastew

torage and treatment reservoirs (WSTR).

During the irrigation period; the faecal coliform counts should preferably

Local epidemiological factors may require a more stringent standard for p

200 faecal coliforms 100 mL�1 is appropriate for the lawns.

This guideline value can be increased to �1 if conditions are hot and dry

upplemented with anti-helminthic chemotherapy campaigns in areas of w

In the case of fruit trees, irrigation should stop 2 weeks before fruit is pick

prinkler irrigation should not be used.

roadside greenery. Its other uses may be environmental

(providing water for wetlands, wildlife refuges, riparian

habitats, urban lakes and ponds), or industrial (used in

cooling, boiling, and the processing of materials). It could

also be used as a source of non-potable water to provide for

many needs (fire fighting, air conditioning, dust control, and

toilet flushing). It may also be used for aquaculture and

groundwater recharge—a use, which has received con-

siderable attention in recent years as it needs proper

legislation and periodic monitoring of the aquifer quality.

3. T

he treatment of wastewater before discharging it into

surface water bodies helps to safeguard existing (scarce)

sources of good-quality drinking water and protects the

environment.

4. U

sing treated wastewater for irrigation decreases the

demand for freshwater in agriculture.

5. I

f it is treated and managed appropriately, treated waste-

water can be used to provide several nutrients essential

for plant growth. This directly benefits farmers because

they have to make little or no investment in fertilizer

(a significant farm input) or its application.

The benefits of using treated wastewater must also be

considered against the human health, economic, and environ-

mental costs of not using it. For example, treating and using

wastewater would reduce the discharge of untreated waste-

water into the environment (so reducing water pollution and

ewater for irrigationa

rrigationmethod

Intestinal nematodesb

(arithmetic mean;no. per 1000 mL)c

Faecal coliforms(geometric mean;no. per 100 mL)d

�0.1f �103

ay or

inkler

�1 �105

d or furrow �1 �103

�0.1 �105

kle or drip Not applicable Not applicable

epidemiological, socio-cultural, and hydrogeological factors.

o intended to protect against risks from parasitic protozoa.

ater is treated in waste stabilization ponds (WSP) or wastewater

be done weekly, but at least monthly.

ublic lawns, especially hotel lawns in tourist areas. A guideline of

and surface irrigation is not used, or if wastewater treatment is

astewater use.

ed, and no fruit should be picked up from the ground. In addition,

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 210

Table 5 – Recommended maximum concentrations (RMCs) of selected metals and metalloids in irrigation watera

Element RMC (mg L�1) Remarks

Aluminum 5.00 Can cause non-productivity in acid soils (pH < 5.5), but more alkaline soils

at pH > 7.0 will precipitate the ion and eliminate any toxicity

Arsenic 0.10 Toxicity to plants varies widely, ranging from 12 mg L�1 for Sudan

grass to less than 0.05 mg L�1 for rice

Beryllium 0.10 Toxicity to plants varies widely, ranging from 5 mg L�1

for kale to 0.5 mg L�1 for bush beans

Cadmium 0.01 Toxic at concentrations as low as 0.1 mg L�1 in nutrient solution for beans,

beets and turnips. Conservative limits recommended

Chromium 0.10 Not generally recognized as an essential plant growth element.

Conservative limits recommended

Cobalt 0.05 Toxic to tomato plants at 0.1 mg L�1 in nutrient solution.

It tends to be inactivated by neutral and alkaline soils

Copper 0.20 Toxic to a number of plants at 0.1–1.0 mg L�1 in nutrient solution

Iron 5.00 Non-toxic to plants in aerated soils, but can contribute to soil acidification

and loss of availability of phosphorus and molybdenum

Lithium 2.50 Tolerated by most crops up to 5 mg L�1. Mobile in soil. Toxic to citrus

at low concentrations with recommended limit of <0.075 mg L�1

Manganese 0.20 Toxic to a number of crops at a few-tenths to a few mg L�1 in acidic soils

Molybdenum 0.01 Non-toxic to plants at normal concentrations in soil and water. Can be toxic to

livestock if forage is grown in soils with high concentrations of available molybdenum.

Nickel 0.20 Toxic to a number of plants at 0.5 to 1.0 mg L�1; reduced toxicity at neutral or alkaline pH

Lead 5.00 Can inhibit plant cell growth at very high concentrations

Selenium 0.02 Toxic to plants at low concentrations and toxic to livestock if forage is

grown in soils with relatively high levels of selenium

Zinc 2.00 Toxic to many plants at widely varying concentrations; reduced toxicity

at pH � 6.0 and in fine textured or organic soils

Modified from Ayers and Westcot (1985).a The maximum concentration is based on a water application rate, which is consistent with good irrigation practices (10,000 m3 ha�1 yr�1). If

the water application rate greatly exceeds this, the maximum concentrations should be adjusted downward accordingly. No adjustment

should be made for application rates less than 10,000 m3 ha�1 yr�1. The values given are for water used on a long-term basis at one site.

the contamination of drinking water supplies), and would i-

mprove the socioeconomic situation of farmers, and thus their

health and that of their families.

Based on different parameters, various guidelines (Ayers

and Westcot, 1985; WHO, 1989; Blumenthal et al., 2000; Carr

et al., 2004; WHO, 2006) are available for wastewater use in

agriculture (Tables 4 and 5). However, in many developing

countries these guidelines are not followed and most farmers

use untreated wastewater in an unplanned manner to irrigate

a variety of crops. Most cities in these countries have networks

of open and covered interconnected channels located within

and around urban premises. In general, these channels carry a

mixture of wastewater generated by domestic, municipal, and

industrial activities. The farmers divert untreated wastewater

from these channels to provide irrigation water as and when it

is needed. Although farmers irrigate a range of crops with

wastewater, they often prefer to grow high-value vegetables as

a market-ready product, which will generate a higher income

(Qadir et al., 2000).

In some cases, the authorities implementing government

regulations periodically expel these farmers from their fields

(Keraita and Drechsel, 2004) or uproot wastewater-irrigated

vegetables. In other cases, however, the administrators do not

make any efforts to check the use of wastewater in this way.

Rather they regard this farming practice as a viable option for

wastewater disposal. The farmers consider such untreated

wastewater to be a reliable source of irrigation, which involves

less cost than other sources of irrigation water such as

groundwater pumping (Van der Hoek et al., 2002). Other

benefits to the farmers include the fact that farmers have to

invest nothing, or very little, in fertilizer purchase and

application, while benefiting from greater levels of crop

production than are obtained via freshwater irrigation. In

addition, they enjoy higher incomes as a result of cultivating

and marketing high-value crops. These benefits help the

farmers to ensure that their families receive better levels of

nutrition and that their children benefit from better educa-

tional opportunities. For all these reasons, farmers take health

risks and use untreated wastewater when the opportunity

presents itself (Ensink et al., 2002; Matsuno et al., 2004).

Surveys and research studies carried out in different

countries revealed that fields irrigated with untreated waste-

water yielded more than those irrigated with freshwater

(Shende, 1985; Minhas and Samra, 2004; Table 6). In addition,

economic analyses based on the cost of production of different

crops have shown attractive economic returns from waste-

water-irrigated fields in Syria (Qadir et al., unpublished data).

The analyses revealed that each US$ invested in the produc-

tion process gave a return of US$ 5.31 in the case of wheat

(Triticum aestivum L.) irrigated with wastewater and US$ 2.34 in

the case of wheat irrigated by groundwater. In addition to the

higher wheat yields provided by wastewater-irrigated plots,

there were savings with regard to fertilizer use. In comparison

with those growing groundwater-irrigated wheat, the farmers

using wastewater to irrigate wheat saved US$ 95 ha�1. Similar

economic return trends were obtained for faba bean (Vicia faba

L.). However, in the case of cotton (Gossypiumhirsutum L.), there

was little difference between the returns from wastewater

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 11

Table 6 – Yields of crops irrigated with good-quality water and untreated wastewater

Crop Crop yield (Mg ha�1) Reference

Freshwater Untreated wastewater

Carrot 9.71 11.75 (+21)a Shende (1985)b

Radish 7.26 8.33 (+15) Shende (1985)

Potato 6.12 9.33 (+29) Shende (1985)

Cabbage 9.27 12.13 (+31) Shende (1985)

Tomato 10.01 13.38 (+34) Shende (1985)

Tobacco 1.12 1.25 (+12) Shende (1985)

Rice 3.80 3.30 (�13) Minhas and Samra (2004)c

Wheat 2.80 3.10 (+11) Minhas and Samra (2004)

Soybean 1.60 2.10 (+31) Minhas and Samra (2004)

Cauliflower 16.40 18.20 (+11) Minhas and Samra (2004)

Sugarcane 42.70 44.40 (+04) Minhas and Samra (2004)

Wheat 3.29 4.49 (+36) Qadir et al. (unpublished data)d

Cotton 4.14 4.24 (+02) Qadir et al. (unpublished data)

a Figures in parenthesis indicate percent increase (+) or decrease (�) in wastewater irrigated crop yield over respective yield from area irrigated

with good-quality surface or groundwater.b Results from Poona Sewage Farm, India.c Results from Nagpur Sewage Farm, India.d Results from a survey of farmers’ fields in Aleppo region, Syria.

irrigation (US$ 5.17) and groundwater irrigation (US$ 5.23) for

each US$ invested. This is because wastewater resources in

the area during the long summer growing season of cotton are

not sufficient to provide the crop with its needs. Therefore, the

wastewater-irrigating farmers also use fertilizers and pump

groundwater as and when needed. The cultivation of

vegetables – which are grown on only 7% of the waste-

water-irrigated area because of government restrictions –

produced the highest economic returns from wastewater

irrigation: US$ 7.48 for each US$ invested. This was much

greater than in the case of vegetables irrigated with ground-

water, where the return was US$ 3.29 per US$ investment

(Qadir et al., unpublished data). Although these crop yield and

economic analyses indicate that communities who use

untreated or partly treated wastewater clearly benefit finan-

cially, there is a need to carry out comprehensive analyses of

the potential environmental and health implications and their

costs. These must be weighed against both the short- and

long-term benefits of wastewater use.

Owing to the low literacy rate found amongst farmers in

developing countries, limited and inappropriate information

gathering and reporting, insufficient public pressure, most

farmers using polluted water in low-income countries remain

uninformed about the health and environmental conse-

quences (Hussain et al., 2002). Moreover, farmers and

authorities have insufficient knowledge about the technical

and management options available for reducing the environ-

mental and health risks associated with wastewater use.

Depending upon the levels of contaminants present, the

continued and uncontrolled use of untreated wastewater as

an irrigation source could have a variety of implications. These

include the following:

1. G

roundwater contamination through the movement of

high concentrations of a wide range of chemical pollutants

(Ensink et al., 2002). This is particularly true in the case of

wastewater that contains untreated industrial effluent. The

pollutants reaching groundwater in this way have the

potential to impact upon human health when groundwater

is pumped for direct human consumption. Pathogens have

also been found to accumulate in the groundwater found

immediately beneath wastewater-irrigated fields (Ensink

et al., 2002).

2. T

he gradual buildup, in the soil solution and on the cation

exchange sites of soil particles, of ions such as Na+ and a

range of metals and metalloids which are deleterious to the

soil. In this way, potentially harmful metals and metalloids

may reach phytotoxic levels (Qadir et al., 2005). The

accumulation of excess Na+ in the soil can have numerous

adverse effects, including changes in exchangeable and soil

solution ions and soil pH, the destabilization of the soil

structure, the deterioration of the soil’s hydraulic proper-

ties, and an increased likelihood of crusting, runoff, erosion

and aeration. It can also have osmotic effects and specific

ion effects in plants (Sumner, 1993; Qadir and Schubert,

2002).

3. T

he accumulation of potentially toxic substances in crops

and vegetables which will, ultimately, enter the food chain,

so damaging human and animal health. For example, leafy

vegetables irrigated with untreated wastewater containing

metals and metalloids can accumulate higher levels of

certain metals, such as cadmium (Cd), than non-leafy

species (Qadir et al., 2000). Excessive exposure to this metal

has been associated with various illnesses in people,

including gastroenteritis, renal tubular dysfunction, hyper-

tension, cardiovascular disease, pulmonary emphysema,

cancer, and osteoporosis (Wagner, 1993). Numerous ill-

nesses are also associated with the ingestion of excessive

levels of other metals and metalloids. Similarly, pathogens

may enter the food chain via the same pathway. However,

in most cases, industrial pollutants in the form of a variety

of metals and metalloids can cause greater and longer

lasting health effects in people than pathogenic organisms.

4. T

he health risks associated with the presence of parasitic

worms, and viruses and bacteria. These have the potential

to cause disease in farming families exposed to untreated

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 212

wastewater for extended periods. Such diseases also raise

the issue of the financial consequences associated with

treatment. Farmers using untreated wastewater for irriga-

tion demonstrate a higher prevalence of hookworm and

roundworm infections than farmers using freshwater for

irrigation. Hookworm infections occur when larvae, added

to the soil through wastewater use, penetrate the skin of

farmers working barefoot (Van der Hoek et al., 2002).

Bearing in mind the challenges associated with the use of

wastewater for irrigation, studies carried out by the research-

ers at the International Water Management Institute (IWMI),

Sri Lanka have proposed a number of options to maximize the

benefits and minimize the risks involved in the use of untr-

eated wastewater for agriculture (Scott et al., 2000; Ensink

et al., 2002; Van der Hoek et al., 2002; IWMI, 2003; Matsuno

et al., 2004; Scott et al., 2004;). These options include: (1) the use

of suitable irrigation techniques and the selection of appro-

priate crops that are less likely to transmit contaminants and

pathogens to consumers; (2) the use of protective measures

such as boots and gloves to control farm workers’ exposure to

pathogens; (3) the implementation of a medical care program

through the use of preventive therapy such as anti-helminthic

drugs; (4) the post-harvest management of vegetables, thro-

ugh washing and improved storage; (5) the conjunctive use of

wastewater and freshwater to dilute the risks and increase the

benefits by supplying nutrients to a larger area; (6) upstream

wastewater management and appropriate low-cost treat-

ment; (7) education and increased awareness among farmers,

consumers, and government organizations; (8) the implemen-

tation of monitoring programs for key environmental, health,

and food safety parameters.

The Hyderabad Declaration on Wastewater Use in Agri-

culture made on 14 November 2002 (available at http://

www.iwmi.cgiar.org/health/wastew/hyderabad_declara-

tion.htm) – which resulted from a workshop organized by

IWMI and the International Development Research Center,

Canada – stressed the need to

‘‘safeguard and strengthen livelihoods and food secur-

ity, mitigate health and environmental risks and

conserve water resources by confronting the realities

of wastewater use in agriculture through the adoption of

appropriate policies and the commitment of financial

resources for policy implementation’’.

The management options used should include raising

public awareness, using safer irrigation methods, minimizing

human exposure, restricting the types of crops irrigated with

wastewater, disinfecting produce, ensuring institutional

coordination, increasing land tenure, and increasing funding

(Scott et al., 2004). In view of the fact that it is not possible to

simply ban wastewater use in many developing countries, the

World Health Organization (WHO) is considering the realities

faced by these countries while revising guidelines for waste-

water use in agriculture (WHO, 2006).

2.3.2. Agricultural drainage waterAdequate drainage is a prerequisite if irrigation is to be

sustainable, particularly when salts in groundwater and high

water tables or waterlogging may damage the crops. A fraction

of the water used for crop production results in drainage

water, which contains salts and the residues of agro-

chemicals such as pesticides, fertilizers, and soil and water

amendments. To maintain an appropriate salt balance in the

topsoil layer, which makes up the effective root zone, the

salinity of the drainage water percolating below the root zone

must be higher than the salinity of the irrigation water applied.

Moreover, saline geological deposits often exist along the flow

path. As drainage water flows through these deposits, the salt

loads in the resulting drainage water can considerably exceed

those projected to occur as a result of irrigation alone (Van

Schilfgaarde, 1994). In some geological settings, drainage

waters may dissolve and displace some minor, but potentially

toxic, elements. As freshwater resources are becoming

increasingly scarce, the reuse of agricultural drainage water

has become an important source of irrigation. Contingent

upon the levels and types of salts present, and the use of

appropriate irrigation and soil management practices, agri-

cultural drainage water can be used for different crop

production systems (Rhoades, 1999; Oster and Grattan, 2002).

Two types of drainage systems (subsurface tile drainage

and tubewell drainage) are practiced on large areas suffering

salt and shallow water table problems. In addition to the

conventional drainage systems used to control water table

depth and salinity, bio-drainage – the use of vegetation to

manage water fluxes in the landscape through evapotran-

spiration – is also used in some areas. Deep-rooted crops and

trees are the best candidate plant species for use as bio-

drainage tools. A combination of bio-drainage and conven-

tional drainage systems might also be considered, depending

on site-specific conditions (Tanji and Kielen, 2002).

Climatic conditions (such as the amount and distribution of

rainfall) and the drainage techniques used all influence crop

response to drainage water irrigation. Under the South Asian

monsoonal climate, and in other areas receiving high intensity

rainfall over a short period, saline drainage effluents could be

used without appreciably reducing crop yields (Sharma and

Rao, 1996). Improved drainage techniques such as controlled

drainage and sub-irrigation can be used to precisely manage

water tables and make better use of drainage water (Sharma

and Minhas, 2005).

Soil characteristics such as texture and the selection of

appropriate crop species play important roles in addressing

elevated levels of salinity and sodicity. Under the South Asian

monsoonal climate, waters with salt concentrations as high as

EC 12 dS m�1 can be used for growing salt-tolerant and

moderately salt-tolerant crops on coarse-textured soils. But

on fine-textured soils, waters with an EC of more than

2 dS m�1 often create salinity problems. In addition, the

leaching requirements of soils increase with the salinity of

the irrigation water and the sensitivity of the crop to salinity

(Kijne, 1996). Crops like beans, sorghum and mustard can

tolerate higher levels of salinity after they have been

established by applying non-saline water before sowing, to

leach down the salts from the root zone (Sharma and Minhas,

2005).

There are two major approaches that can be used to

improve and sustain agricultural production in a salt-prone

environment: (1) modifying the environment to suit the plant,

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 13

and (2) modifying the plant to suit the environment. Both

approaches have been used, either singly or in combination

(Tyagi and Sharma, 2000). However, the first approach has

been used more extensively because it enables the plants to

respond better not only to the water used but also to the other

production inputs involved. These approaches provide

socially and environmentally favorable levels of agricultural

production, provided that the characteristics of the water and

the soil, and the intended management and uses of the crops,

are taken into consideration (Tyagi and Sharma, 2000; Qadir

et al., 2003; Oster and Birkle, 2004). Such use of saline waters

may lead to conservative water quality standards being

replaced with site-specific guidelines, which take into account

factors such as soil texture, crop salt tolerance, and irrigation

management (Minhas and Tyagi, 1998).

Grattan and Rhoades (1990) proposed a set of criteria for

selecting the crops that could be grown when reusing

agricultural drainage water. Oster and Birkle (2004) provided

information about the growth habits, salt tolerance charac-

teristics, average root zone salinity at 70% yield, and leaching

requirement of several forages when irrigated with a saline–

sodic water (EC 10 dS m�1 and SAR 15). And, Qureshi and

Barrett-Lennard (1998) revealed useful information regarding

sources of seed, nursery raising techniques, and land

preparation and planting procedures for 18 different tree

and shrub species which demonstrate potential for growth on

degraded land and under salt-prone water conditions. In

addition to conventional agriculture systems and in saline

agriculture, drainage water can also be reused in wildlife

habitats and wetlands and for the initial amelioration of salt-

affected soils.

Two options are available for the conjunctive use of saline

water and freshwater: blended and cyclic use. The cyclic

option involves the use of saline water and non-saline

irrigation water in crop rotations that include both moderately

salt-sensitive and salt-tolerant crops. Typically, the non-

saline water is also used both before planting and during the

initial growth stages of the salt-tolerant crop, while saline

water is usually used after seedling establishment. Blending

consists of mixing non-saline and saline water supplies before

or during irrigation. Blending is a promising option in areas

where freshwater can be made available in adequate

quantities on demand. However, cyclic use is most common

and offers several advantages over blending (Rhoades, 1999).

Analyses of a large number of experiments have shown that at

the same level of water salinity (weighted average salinity) the

yields for different cyclic use modes were higher than the

estimated yields for blending. Surveys indicated that the

farmers alternating saline and freshwater obtained higher

production levels in the cases of both cotton and millet than

did those using both waters in a blended form (Minhas and

Tyagi, 1998).

In addition to cyclic and blended use, drainage water may

be used in a sequential system (Grattan and Rhoades, 1990;

Grattan et al., 2004; Grieve et al., 2004), during which relatively

better quality water is applied to the crop with the lowest salt

tolerance. The drainage water from that field – obtained from a

subsurface drainage system – is then used to irrigate crops

with a greater salt tolerance. The simplest management

method is to use the drainage water on fields, which are

located down-slope from those where the drainage water is

collected. There is no fixed number of times that the cycle can

be repeated. It depends on salinity, sodicity, and concentra-

tions of toxic minor elements in the drainage water, the

volume of water available, and the economic value and

acceptable yield of the crop to be grown. The strategy in

which drainage water with an EC of more than 4 dS m�1 is

sequentially used to irrigate crops and trees (i.e. an agrofor-

estry system) with an increasing salt tolerance is sometimes

referred to as an ‘agro-drainage system’ (Karajeh et al., 1992;

Karajeh and Tanji, 1992a,b). The sequential strategy provides a

means of minimizing the volume of drainage water produced

by irrigated agriculture (Oster and Grattan, 2002). Rhoades

(1989) proposed an integrated strategy that could be used to

facilitate the use of saline water for irrigation, in order to

minimize drainage disposal problems and to maximize the

beneficial use of multiple water sources. Specifically, he

suggested the interception and isolation of marginal-quality

drainage water from good-quality water supplies, so that they

could be used within dedicated parts of an irrigation project as

a substitute for part of the freshwater otherwise used to

irrigate the crops.

The long-term feasibility of using drainage water for

irrigation could be increased when the technique is used on a

project or regional scale, rather than on individual farms.

Grattan and Rhoades (1990) proposed that regional manage-

ment could allow the reuse of drainage water in dedicated

areas, so as to avoid the successive increases that would occur

in the concentration of the drainage water if the reuse process

were to be used with the same water supply and on the same

land area (a closed system). With regard to project-based

management, certain farms could be dedicated to reuse while

other areas, such as upslope areas, could be irrigated solely

with good-quality water as usual. The second-generation

drainage water from the primary reuse area could then be

discharged to other dedicated reuse areas where even more

salt-tolerant crops and agroforestry systems are established

or where brackish water based fisheries exist. The final

discharges of small quantities may then be disposed of via

evaporation ponds or water treatment plants. Ideally,

regional coordination, participatory management, and cost

sharing among all user groups should be institutionalized

when implementing such regional drainage-water-reuse

systems.

2.3.3. Saline and/or sodic groundwaterMany water scarce areas have aquifers of marginal quality,

such as those that contain saline and/or alkali (�sodic) waters.

Saline water contains excess levels of salts, while sodic water

contains elevated levels of Na+ compared to other cations.

Such marginal-quality waters result from reactions with the

layers of earth, or strata, through which the water passes on its

way to becoming groundwater, as well as reactions within the

geologic formations in which the groundwater is located.

Saline aquifers may also occur in the vicinity of seawater (as is

the case in a coastal region). The quality of many groundwater

bodies has been affected by drainage water from agricultural

areas (Bohlke, 2002). In addition to the environmentally sound

and productive use of saline–sodic groundwater, this section

also addresses the development of a water resource.

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 214

Table 7 – Average grain yield of rice and wheat asaffected by conjunctive use of alkali water and fresh-water over a period of 6 years

Treatment Grain yield(Mg ha�1)

Water produc-tivity (kg m�3)

Rice Wheat Rice Wheat

Freshwater (FW) 6.7 5.4 0.62 1.80

Alkali water (AW) 4.2 3.6 0.39 1.20

2 FW–1 AWa 5.2 6.7 0.62 1.73

1 FW–1 AWb 5.3 6.3 0.58 1.77

1 FW–2 AWc 4.8 5.7 0.53 1.60

Modified from Bajwa and Josan (1989).a Two irrigations with freshwater and one irrigation with alkali

water.b One irrigation with freshwater and one irrigation with alkali

water.c One irrigation with freshwater and two irrigations with alkali

water.

Driven by the need to produce more under water-scarce

conditions, larger amounts of saline groundwater are

pumped for irrigation in several countries in the Middle East

as well as in numerous areas elsewhere. For example, India’s

annual net groundwater draft is 135 � 109 m3. Of this

32 � 109 m3 is estimated to consist of saline and/or sodic

water (Minhas and Samra, 2003), which is about one-fourth of

the total volume of groundwater used in the country. The use

of marginal-quality groundwater resources without appro-

priate soil-, crop- and irrigation-management strategies

poses considerable risks, in terms of the development of

salinity, sodicity, ion-specific toxicity, and nutrient imbal-

ances in soils (Sharma and Minhas, 2005). These conditions

reduce crop productivity and limit the choice of crops that can

be grown. However, options are available for the use of saline

and sodic groundwater through the use of improved methods

of water conservation, remediation, development, and

application (Rhoades, 1999; Tyagi and Sharma, 2000; Qadir

and Oster, 2004).

Different methods of irrigation have been used for saline

and sodic groundwater (Sharma and Minhas, 2005). In

comparison with surface irrigation, high-energy pressurized

irrigation methods (such as sprinkler and drip irrigation) are

typically more efficient, as the volume of saline water to be

applied can be adequately controlled. The drip irrigation

system also avoids injury to the leaves of plants and enhances

the threshold limits of the crop’s salt tolerance, by modifying

the patterns of salt distribution and maintaining a constantly

high matric potential in the soil (Tingu et al., 2003). These

authors found that both the yield and quality of watermelons

was greater in a treatment irrigated with saline water (EC 3.3–

6.3 dS m�1) using drip irrigation than in a control treatment.

The greatest improvements in yield and melon quality were

obtained when saline water was applied at a rate equivalent to

60% of open pan evaporation. However, salts accumulated in

the root zone, which caused difficulties when planting

subsequent crops because the salts needed to be effectively

leached using either flood or sprinkler irrigation.

Most past research on the management of saline–sodic

groundwater resources and soils has led to the development of

management practices at the field level, without considering

their implications and practicability in terms of water and salt

balances at the irrigation-system and river-basin levels (Tyagi,

2003). In most of the irrigation systems used in arid and semi-

arid regions, water inadequacies at the tail end are compli-

cated by a gradual decrease in freshwater availability and a

decrease in water quality from the head through to the tail

reaches. For example, in the Bhakra command of the Indian

Punjab, the volume of canal water available decreased from

2500 m3 ha�1 (in the head reach) to 800 m3 ha�1 (in the tail

reach), with a corresponding increase in groundwater salinity

(from 2.5 to 6.8 dS m�1). This situation favored groundwater

pumping by shallow tubewells in the head reach and led to the

transfer of this pumped groundwater to the tail reach (Tyagi,

2003). However, caution needs to be taken that such water

markets may not lead to over-exploitation of the aquifers and

an increase in soil salinity. The potential increase in relative

yield with such groundwater transfer from head to tail reach

of a water course in Batta minor (Bhakra Irrigation System,

India) indicated that relative yield would increase from 0.70 to

0.85 in the entire water course if 50% of marginal-quality

groundwater from the head reach was transferred and used in

the tail reach without changing the canal water allocation

(Chandra, 2001). The relative yield would go up to 0.89 if,

instead of blending, the groundwater was used in a cyclic

mode.

Excessive levels of some of the constituents of the

groundwater used for irrigation, such as Na+, carbonate

(CO32�), and bicarbonate (HCO3

�), can adversely modify soil

characteristics and impact upon crop yields. Continuous

irrigation using waters that contain elevated levels of residual

sodium carbonate (RSC) and which have an elevated sodium

adsorption ratio (SAR) tends to increase the soil’s pH and the

exchangeable sodium percentage (ESP), which in turn

decreases the permeability of the soil to water and reduces

crop yield (Sharma and Minhas, 2005). However, studies have

shown that irrigating crops with sodic water after two

irrigations using freshwater resulted in rice and wheat yields

comparable to those yielded when freshwater irrigation was

used throughout the cropping period (Table 7). The use of

chemical amendments such as gypsum (CaSO4�H2O) is

particularly recommended when RSC values exceed

5 mmolc L�1, soils are medium-textured, and annual rainfall

is below 500 mm (Minhas et al., 2004). Gypsum application

techniques have been refined in the form of ‘gypsum beds’,

the use of which improves gypsum’s solubility and applica-

tion efficiency and reduces the costs of its application. This

method provides considerable savings and produces higher

crop yields (Pal and Poonia, 1979; Sharma and Minhas, 2005).

In California, applying gypsum by mixing it with irrigation

water has been a common practice for several decades. Such

application of gypsum has the potential to improve the soil’s

infiltration rate (Oster et al., 1992).

In some areas, highly degraded salt-prone soils associated

with saline and/or sodic aquifers cannot be ameliorated in a

manner that allows the profitable cultivation of high-value

crops. The best approach to deal with such conditions has

been to retire these areas to permanent vegetation (Tomar

et al., 2002). The preferred choices for tree species in these

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 15

areas include Tamarix aphylla (L.) H. Karst., Prosopis juliflora

(Sw.) DC, Acacia nilotica (L.) Delile, Acacia farnesiana (L.) Willd,

and Acacia tortilis (Forssk.) Hayne (Tomar et al., 2002; Sharma

and Minhas, 2005). Several halophytic species, forages, and

field and horticultural crops have also been identified for

biosaline agriculture.

Similar to the development of good-quality groundwater

resources, the development of marginal-quality groundwater

resources can have various negative consequences, such as

groundwater level drawdown, reduced discharge to surface

water bodies, intrusion of lower quality water resources, land

subsidence (Custodio, 2002). Though there seems to be no

single management tool, which can be used to control the

salinity and sodicity of salt-prone land, and the negative side

aspects of water resources development, several practices

interact with each other and should be taken into considera-

tion in an integrated manner to ensure sustainable agri-

culture and livelihoods (Sharma and Minhas, 2005). There

is a need to ensure: (1) the presence of water quantity and

quality monitoring networks and easily accessible databases;

(2) the modification of surface water delivery schedules;

(3) the conjunctive use of multi-quality water resources; (4) the

availability of chemical amendments, as and when needed

and at reasonable costs; (5) the adoption of water-use efficient

irrigation systems such as drip irrigation; (6) the use of

participatory planning by different stakeholders; (7) capacity

building of the national agricultural and extension systems;

(8) an increase in awareness among the farmers, to ensure

a greater understanding of the potentials of plants, soil

and water available, and the uses and markets for the agri-

cultural produce yielded; (9) an adaptive management system

that reacts to environmental and socioeconomic changes

(Loucks, 2000).

3. Opportunities contributing to food securityin water-scarce countries

Opportunities which may be taken advantage of to contribute

to food security in water-scarce countries are: (1) the ‘physical’

movement of water from water-surplus areas, where fresh-

water can be tapped and then transported (in a similar way

that oil is exported from the oil-surplus Middle East to oil-

deficit countries), and (2) the importing of food items from

countries where water resources are adequate and the cost of

food production is relatively low. In the latter case, water-

scarce countries import food items rather than water. This

saves these countries huge amounts of water, which would

otherwise have to be used to produce food domestically

under water-deficit conditions. The same amount of water is

actually used elsewhere in the production process. The terms

‘embedded water’ and ‘virtual water’ have been used to

illustrate the importance of water in the trade in food, which

occurs between water-surplus and water-deficit countries.

3.1. Physical transportation of freshwater

Long-distance, large-scale transportation of freshwater at the

inter- and intra-country scales has been emerging as an

option, which could be used to supplement freshwater needs

in water-scarce areas (Mohsen and Al-Jayyousi, 1999;

Ariyoruk, 2003; Government of India, 2003; Alghariani, 2003).

Several options have been worked out to address long-

distance transportation in MENA. One example is the Great

Man-Made River Project in Libya, which was initiated in the

early 1980s. The existence of vast stores of fossil freshwater in

the southeastern and southwestern desert areas of the

country prompted Libya’s government to build a 3500-km

pipeline to convey water to the coastal areas, where rapid

development and increased populations have placed a severe

strain on the water supply available to households, agricul-

ture, and industry. The pipeline is expected to ship more than

5 � 106 m3 (2 � 109 m3 yr�1) of water across the desert per day

to the coastal areas (Alghariani, 2003). Since the project was

conceived, several economic evaluations have been carried

out and projections have been made of the water transporta-

tion and distribution costs; however, the conclusions reached

differ greatly in each case.

The possibility of moving freshwater long distances from

Turkey to some parts of the Middle East has been discussed

extensively over the last two decades. According to the famous

‘Peace Pipeline Project’, conceived in 1987, Turkey was to

export water to the Middle East, first from the Seyhan and

Ceyhan Rivers (Mohsen and Al-Jayyousi, 1999), and later from

the Manavgat River. Turkey has already finished building a

US$ 150 million anchorage – pumping station and treatment

plant – at the estuary at the mouth of the Manavgat River

(Ariyoruk, 2003). However, political unrest in the region, and

technical and economic complications have greatly hampered

the further implementation of this project. The potential for

projects involving cross-boundary water movement can only

be realized with good regional cooperation and trust among

both water exporting and importing countries and those

located between them.

Outside MENA, China’s water and soil resources are

distributed very unevenly. The basin of the Yangtze River

and the rivers situated to the south of it are considered to be

water-surplus areas, yielding more than 80% of the nation’s

total surface water resources while containing less than 40%

of the nation’s total cultivated land. By contrast, the Yellow

River, Huaihe River, Haihe River basins and the northwestern

inland areas contain 45% of the country’s cultivated land

but possess only 12% of its total water resources, and are

therefore categorized as water-deficit areas. The north-

western and northern parts of China are rich in land and

mineral resources, and provide a production base for energy,

grain, cotton, and edible oil. In these areas, water scarcity has

restricted economic development and induced environmen-

tal degradation. The government has therefore unveiled the

ambitious south-to-north water diversion project, which is

designed to balance the nation’s water supply. The project, a

result of 50 years of investigation and research, aims to divert

water from the Yangtze River valley to the reaches of the

Yellow River, Huaihe River and Haihe River, to ensure an

adequate water supply for farming, industry and other needs

in the north. Estimated to cost more than US$ 22 billion, the

project will have three water diversion routes. Once com-

pleted, it will be able to transport in the range of 38 � 109 to

48 � 109 m3 of water per year to areas containing 300 million

people.

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 216

In recent years, India’s government has also decided to

implement a huge river-interlinking project. The project,

when completed, will consist of 30 links and some 3000

storage points, connecting 37 rivers to form a gigantic South

Asian water grid (Government of India, 2003). Initial project

estimates suggest a staggering cost of US$ 120 billion to handle

inter-basin water transfer of 178 � 109 m3 yr�1 through the

building of 12,500 km of canals. The project is expected: (1) to

create 35 GW (gigawatt) of electricity per year, (2) to add

35 � 106 ha to India’s irrigated areas, and (3) to generate

substantial fishery benefits. A task force has been set up to

suggest possible scenarios, in order to arrive at speedy

consensus amongst the states for water sharing and transfer.

The project is expected to be launched by 2006, and should be

completed in 10 years.

In addition to the cross-boundary transportation of fresh-

water through pipelines, another innovative approach exists

which has recently entered the limelight. This approach

consists of transporting water – using specifically designed,

extremely large, flexible and recyclable, fabric bags – from a

freshwater source across the sea to a coast near to a water-

scarce area. With the capacity of more than 1 � 105 m3, these

huge bags are generally known as ‘medusa bags’, named after

the swimming form of jellyfish often called a medusa. Since

freshwater is less dense than seawater, the cargoes can float

on the sea. Therefore, they will be filled with freshwater to

about 40% of their capacity, before being towed either singly or

in pairs to their destination, where water will be pumped

ashore and the empty bag will be returned to the freshwater

source. Since 1996 a small-scale operation using bags with a

capacity of up to 2 � 103 m3 – 50 times smaller than the

proposed ‘medusa bags’ – have been used to supply water to

the islands of Aegina and Angistri in Greece at half the cost of

previous supplies by tanker. The ‘medusa bags’ offer an option

Table 8 – Estimates of water used to produce different food ite

Food item Water use

Estimate 1c Estimate

Wheat 1000 1150

Rice 1500g 2656

Maize – 450

Soybean – 2300

Palm oil 2000 –

Beef 20000 –

Sheep 10000 –

Poultry 6000 –

Eggs – –

Milk – –

a Quantifying the exact volumes of water needed to produce food items i

to produce a particular food item in different regions. However, these esti

different items.b The average amount of water required to produce cereals in the develo

around 1800 m3 Mg�1.c Allan (1999).d Hoekstra and Hung (2003).e Zimmer and Renault (2003).f Chapagain and Hoekstra (2003) representing the global average for whg In some regions, the amount of water required to produce rice may reh Not reported.

that could be used to transport water from southern Europe to

the southern Mediterranean region (Alghariani, 2003). This

technology appears attractive; however, its use on a larger

scale remains to be proven. Several countries of the

Mediterranean region are assessing its feasibility.

3.2. Food imports and the ‘virtual water’ nexus

Food items have been traded among countries since interna-

tional trade began. By importing food in this way, water-scarce

countries avoid having to use their own water produce the

same amount of food domestically under water-deficit

conditions. They thus ‘save’ this water. The term ‘virtual

water’ has been used to illustrate the important role water

plays in the trade of food between water-surplus and water-

deficit countries (Allan, 1996), which in part have to rely on the

import of food items in order to ensure food security.

Since the term ‘virtual water’ links water, food, and trade

(Allan, 2003), it compares the amount of water embedded in a

crop that can be purchased internationally with the amount of

water that would be required to produce that crop domes-

tically (Allan, 1996). For instance, it takes approximately 1.1 m3

(1 m3 = 1000 L) of water to produce 1 kg of wheat. Thus, for

every kilogram of wheat imported, the water-deficient country

also receives the benefit of about 1.1 m3 (1100 L) of virtual

water, often at a much lower cost than the price (i.e. the value

attached to) the same quantity of water derived from local

water resources in the country itself, if such sources are

available (Chapagain and Hoekstra, 2003). Similar estimates

are available for other food items (Table 8), although

quantifying and comparing the exact volumes of water needed

to produce food items in different agro-ecological zones is

difficult (Rosegrant et al., 2002). The estimates of the ‘virtual

water’ content of food items provide an assessment of possible

msa

to produce food item (m3 Mg�1)b

2d Estimate 3e Estimate 4f

1160 1109

1400 –h

– 463

2750 1716

– –

13500 –

– –

4100 –

2700 –

790 –

s difficult as large differences may exist in the amount of water used

mates give an indication of the amount of ‘virtual water’ contained in

ped world is about 1000 m3 Mg�1 while in the developing world it is

eat, maize, and soybean.

ach 6000 m3 Mg�1 (Rosegrant et al., 2002).

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 17

water savings in water-scarce countries. These countries can

optimize water use by importing a proportion of their food

requirements from other countries where water resources are

adequate and available at a lower cost (Allan, 1999).

The global estimates of food production reveal that about

6000 � 109 m3 of soil water and freshwater are mobilized each

year to meet the needs of the world’s 6.5 billion people

(Shiklomanov, 2000). The global volume of international

‘virtual water’ flows is estimated to be 1031 � 109 m3 yr�1,

including 695 � 109 m3 yr�1 from the trade in crops and

336 � 109 m3 yr�1 from the trade in livestock and livestock

products. This means that approximately 13% of the water

used for crop production in the world is not used for domestic

consumption but for export (Chapagain and Hoekstra, 2003).

Including trade in livestock and livestock products, ‘virtual

water’ amounts to about 20% of the global water use in

agriculture. These estimated percentages reflect the situation

on a global scale; the situation varies considerably between

countries.

Worldwide, there is a growing interest in the trade of food

items and ‘virtual water’, partly because of the increasing

importance of food security in countries where water has

become too scarce to meet the demands of continuously

expanding populations (Hoekstra, 2003; WWC, 2004). There-

fore, the role of the food trade is important in MENA, which is

the most water-scarce region of the world. Since the oil boom

of the 1970s and population growth, the average annual rate of

food imports has substantially increased in several Middle

Eastern countries. Based on food trade data for the period

1995–1999, the net amount of water used elsewhere when

producing the crops imported into MENA is estimated to stand

at 74.6 � 109 m3 yr�1 (Chapagain and Hoekstra, 2003). Esti-

mates show that MENA imported about 30% of its food

requirements in 2000 (Allan, 2003). If the trend seen in the

importing of food continues, food imports are expected to

provide 50% of the region’s food requirements in 2050 (Allan,

personal communication, 2002).

The trade in food commodities and ‘virtual water’

combines agronomic, economic, and political aspects. The

agronomic component involves the amount of water used to

produce crops. The ‘virtual water’ perspective is consistent

with the concept of integrated water management, in which

aspects of water supply and demand are considered when

determining the optimal use of limited water resources

(Bouwer, 2002). In accordance with the concept of ‘virtual

water’, there may be a need to modify cropping plans to get

maximum benefits from the available, but limited, water

resources found in water-deficient countries. Apart from the

economic, social, institutional, and political implications,

such cropping plans will also need to consider: (1) water

factors, such as the quantity and quality of the water that is

available to produce crops and the sustainability of water

resources; (2) soil factors, including nutrient availability, and

soil moisture, texture, structure, salinity and sodicity levels; (3)

climatic conditions, particularly temperature, rainfall, and

relative humidity; (4) crop traits, such as water-use efficiency,

adaptability to local conditions, and susceptibility to pests and

diseases; (5) farming aspects, such as the size and productivity

of farms, and the availability of manpower and farm

machinery; (6) economic factors such as the provision of

subsidies for farm inputs, crop insurance and local and/or

international market demand (Qadir et al., 2003).

The economic aspect of food imports and ‘virtual water’

involves the opportunity cost of water use, which is the value

it has when used for other uses (such as the production of

alternative crops or in municipal, industrial, or recreational

activities). The opportunity cost of water use is a key

component of the ‘virtual water’ perspective, and it is

particularly important to consider such costs when seeking

to allocate scarce water resources. Indeed, the economic

aspect of the virtual water concept is closely related to the

comparative advantage concept used in international trade.

This concept suggests that nations should export those

products for which they possess a relative or comparative

production advantage, and should import those products for

which they possess a comparative production disadvantage.

Another economic aspect of the trade in ‘virtual water’

relates to the fact that the grain produced in a water-surplus

country is sometimes traded at less than its production cost.

For instance, in 1996 the price of wheat grain increased to US$

240 Mg�1 (megagram, Mg = 1000 kg) but fell back to US$

140 Mg�1 by May 1997 on the international market. The cost

of producing wheat was estimated to be about US$ 200 Mg�1 at

that time (Allan, 1998). In addition, the world prices of several

food commodities have decreased over the last three decades.

Between 1970 and 2000, for example, the international prices

of three major cereals (wheat, rice and maize) fell by 50–60%.

Over the next two decades, it is projected that the world prices

of most cereals will decline, though more slowly than in the

past (Rosegrant et al., 2002). This means that the economies of

those water-scarce countries that import grains may receive a

subsidized bargain, particularly when the world prices of food

commodities are lower than the cost of production in these

countries.

Many water-scarce countries will continue to rely in part on

imported food items, while also producing domestically a

proportion of the food they require. Thus, the export of

commodities requiring less water will present an opportunity

for those water-scarce countries seeking to maximize the

value of their limited water resources. However, the role of

land and capital in agricultural imports and exports must also

be considered in the case of those countries where these

resources are limited. In addition, labor-intensive crop

production and processing could also provide options in

countries where reducing unemployment is an important

policy objective (Wichelns, 2001). Policies which promote

imports of food (and ‘virtual water’) may have additional

implications: (1) an increased dependency on major exporters;

(2) possible interference by major exporting countries in an

importing country’s internal affairs; (3) effects on local

agricultural production systems and markets; (4) increased

vulnerability to large-scale natural disasters, which could be

exploited by the international food market. Therefore, the

food-importing countries need to consider the policies and

general politics of food-exporting countries while using the

‘virtual water’ trade as a strategic instrument and as an

opportunity to overcome water scarcity and food deficit.

Before 1993, the term ‘embedded water’ (which described a

concept similar to ‘virtual water’) had very little impact as it

failed to capture the attention of the water-management

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 218

community. When Professor J.A. Allan conceptualized the

concept and coined the term ‘virtual water’ in 1993, it had an

immediate impact as people appeared to use it readily as a

useful metaphor. Professor Allan thereafter used the term to

draw attention to the notion that serious local water shortages

could be very effectively ameliorated by global economic

processes (Allan, 1999). He proposed the concept of ‘virtual

water’ as an economically invisible and politically silent way

of solving strategic water problems (Allan, 2001). Use of the

term has been on the increase since 1995, and by the

millennium it had become central to many dialogues relating

to water scarcity and security (Allan, 2003). In the meantime,

the term has received criticism. Considering ‘virtual water’ to

be nothing more or less than the water needed to produce

agricultural commodities, Merrett (2003a) proposed that it

should be replaced with the phrase ‘the crop water require-

ments of food’. Similarly, he suggested that the phrase ‘the

import of food’ should substitute for the term ‘the import of

virtual water’. He considered ‘the import of virtual water’ to be

a metaphorical term, and not a scientific one, and felt that its

use in the arena of policy might lead decision makers to

neglect the current and future status of the agricultural sectors

of food-importing and food-exporting countries. He further

stated that emphasizing food imports as a solution to water

scarcity will result in less production, which in turn will mean

that less water will be required for irrigated agriculture in

water-scarce regions. It was felt that regional politicians could

deflect attention from such dependency, and that the

availability of imported food could allow them to postpone

new water supply initiatives and delay difficult decisions

about the demand management of their water resources.

Merrett therefore felt that one cannot generate accurate and

meaningful recommendations regarding policies that influ-

ence agricultural production, livelihoods, food security, and

international trade based only on a metaphor (Merrett, 2003b).

However, amid appreciation, caution, and criticism (Allan,

2003; Lant, 2003; Merrett, 2003a; de Fraiture et al., 2004), the

metaphors ‘virtual water’ and ‘virtual water trade’ have been

in use for almost a decade. The coming decade may determine

whether they retain their place in the terminology of water

professionals.

4. Future perspectives

Although agriculture will remain the dominant user of water,

in water-scarce countries intense competition for good-

quality water among the different water-use sectors is

expected to reduce the amount of freshwater allocated to

agriculture in the foreseeable future. Whatever the case, there

should be no anxiety in any region of the world, including

MENA, about the availability of water for drinking and

domestic use and for the needs of almost all of the industrial

and service sectors. In fact, these sectors use less than 25% of

the freshwater resources available (Bouwer, 2000). However, to

foster the sustainable management of water resources and the

livelihood of rural communities, water use in agriculture must

be made more efficient. As the situation currently stands,

despite improvements in existing water-use efficiency tech-

niques, water-scarce countries are expected to become

increasingly reliant on non-conventional water resources

and the other opportunities available to augment water

supplies, in order to achieve food security.

Currently, desalination of seawater and highly brackish

groundwater provides 11 � 109 m3 of high-quality water per

year. With regard to drinking water production, the current

costs of desalination and desalinated water distribution are

affordable in many developed and oil-rich countries. However,

it remains an expensive option for use in conventional crop

production systems. And, in developing countries, even

drinking water produced in this way is an expensive resource.

Although it is generally thought that the costs involved in

desalination will decrease further, they will not decrease by an

order of magnitude at any time in the near future. Since

desalination does have environmental implications, it is

important to ensure that adequate environmental regulations

and practices are put in place to ensure that the long-term

disposal of by-products of desalination (such as impurities and

residual water in the form of brine dumped in nearby areas or

into the ocean) will not have adverse environmental impacts.

Interest in rainwater harvesting techniques has increased

in water-scarce countries, and use of these techniques has the

potential to increase rainwater-use efficiency. The effective-

ness of rainwater harvesting is based on several on-site

factors, such as landscape topography, soil surface texture

and structure, the presence of crusty layer or surface

preparation, soil nutrient availability status and the depth

of the cultivated area. While many case studies have shown

tremendous improvements in crop production through

various rainwater harvesting techniques, it is still unclear if

these techniques will be widely adopted by farmers and

pastoralists in the dry areas.

Marginal-quality water resources are valuable in water-

scarce countries. However, the associated health and envir-

onmental risks mean that the use of untreated or inadequately

treated wastewater will need to be curtailed in water-scarce

countries. Given the water scarcity they face, the cost of

treating wastewater in these countries could be more

attractive than the option of developing new supplies for

different water-use sectors. In addition, wastewater treatment

will also lead to gains in terms of environmental conservation.

However, because governments allocate few funds to waste-

water treatment and with private sector involvement lacking,

it will be extremely difficult to enforce the treatment of all the

wastewater generated in resource-poor countries. Therefore,

steps must be taken to maximize the benefits and minimize

the risks involved in wastewater use in agriculture. This will

require international research, development and donor

organizations and national governments to collaborate more,

in order to allocate adequate funding to wastewater treatment

and the education of stakeholders.

There is emerging evidence that the use of saline and/or

sodic waters in conjunction with the adoption of appropriate

soil, crop, and irrigation management strategies can boost

agricultural productivity far more efficiently than was pre-

viously thought (Oster and Wichelns, 2003). The future use of

cyclic, blended, and/or sequential strategies for using these

waters is expected to increase (Kijne, 2003). As a result, existing

soil and crop management practices will need to be modified in

order to cope with the inevitable increases in soil salinity and

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 19

sodicity that willoccur. Inaddition, the assessment of thefuture

sustainability of using saline–sodic waters, in regard to

maintaining soil permeability, will become a more serious

issue. Although currently available simulation models can be

used to provide insights into what may happen in the future,

such models make use of various simplifying assumptions

when representing complex systems. At present, therefore,

their ability to predict the impacts that soil sodicity and salinity

will have on the permeability of the soil to air and water is

limited. The hydro-salinity models available require field-scale

testing on a range of soil types, and with various irrigation and

cropping practices. Although models are unlikely to replace

entirely the need for continued research into the management

of salt-prone waters and soils, they will provide an important

tool for use when developing alternative management plans

and designing monitoring strategies to track their progress.

Such models may also help researchers to assess the viability of

different drainage water and groundwater reuse strategies and

to extend the results obtained to broadly similar locations

elsewhere.

It is evident that water-scarce countries are unable to

meet their food requirements using either the conventional

or non-conventional water resources available within their

boundaries. Alternative options for contributing to food

security in these countries are provided by the transporta-

tion of water from water-rich areas and the import of food

items from other countries where water resources are

adequate and the cost of food production is relatively low.

However, projects involving cross-boundary water trans-

portation can only realize their potential if there is a good

level of regional cooperation and trust among the water

exporting and importing countries and those located

between them. In addition to other economic and political

considerations, interregional issues may arise when tapping

shared aquifers, and these may strain relationships among

neighboring countries.

Based on the international trade in crops, it is expected

that many water-scarce countries will continue to rely partly

on the import of food items, while producing a proportion of

their food requirements domestically. Water-scarce coun-

tries seeking to maximize the value of their limited water

resources could export commodities, which require less

water to produce. The trade in food will play an important

role in efforts to achieve food security in regions like MENA in

the future, as food imports are expected to provide 50% of the

region’s food requirements by 2050. In fact, the global trade in

food is expected to increase in the foreseeable future. Under

this scenario, the food-importing countries need to consider

the implications that policies in food-exporting countries

may have for them, while using the trade in ‘virtual water’ as

a strategic instrument and an opportunity to overcome water

scarcity. The implications of this are that there would need to

be understanding among the food-exporting and the food-

importing countries. This may require the implementation of

international controls, similar to those in place for oil

trading.

There are several examples of the increased use of non-

conventional water resources by the farming communities in

CWANA, which are faced with the challenge of water scarcity.

This suggests that there is a need for community-based

integrated water resources management. If strategies and

technologies are developed using the accumulated wisdom of

such communities, not only will people’s participation be

enhanced, so will adoption of the new measures. Although

there are no fixed procedures for ensuring that people

participate in such schemes, Sharma and Dixon (1995)

proposed seven steps which provide valuable guidelines: (1)

community immersion, (2) community mobilization, (3)

gender equity, (4) community envisioning, (5) vision valida-

tion, (6) diagnostics and resource community participatory

planning, and (7) participatory monitoring and evaluation

and refinement. In arid and semi-arid areas of the developing

world, particularly CWANA and South Asia, international

agricultural research centers such as IWMI and ICARDA are

collaborating with key stakeholders to address the challenges

relating to water scarcity and to develop appropriate

strategies for the efficient use of non-conventional water

resources.

Acknowledgments

This publication is a part of the joint initiative of the

International Center for Agricultural Research in the Dry

Areas (ICARDA) and the International Water Management

Institute (IWMI) for the assessment and management of

marginal-quality water resources and salt-affected soils. We

gratefully acknowledge the helpful comments of P. Drechsel,

M. Al-Zaeim, and R.J. Thomas on an earlier version of the

paper as well as those from the external reviewers, and J.D.

Oster, Co-Editor-in-Chief of Agricultural Water Management.

Thanks are also due to SCRIPTORIA Writing Services for the

editorial services they provided.

r e f e r e n c e s

Agarwal, A., Narain, S., 2003. Dying Wisdom: Rise, Fall andPotential of India’s Traditional Water Harvesting Systems.Center for Science and Environment, New Delhi, India,404 pp.

Alghariani, S.A., 2003. Water transfer versus desalination inNorth Africa: sustainability and cost comparison.Occasional Paper 49 of the School of Oriental and AfricanStudies, London. Available at http://www.soas.ac.uk/waterissues/occasionalpapers/OCC49.pdf.

Allan, J.A., 1996. Overall perspectives on countries and regions.In: Rogers, P., Lydon, P. (Eds.), Water in the Arab World:Perspectives and Prognoses. Harvard University Press,Cambridge, Massachusetts, pp. 65–100.

Allan, J.A., 1998. Virtual water: a strategic resource, globalsolutions to regional deficits. Ground Water 36, 545–546.

Allan, J.A., 1999. A convenient solution. The UNESCO Courier,February, pp. 29–31.

Allan, J.A., 2001. The Middle East Water Question: Hydropoliticsand the Global Economy. I.B. Tauris & Co. Ltd., London, UK,382 pp.

Allan, J.A., 2003. Virtual water—the water, food, and tradenexus: useful concept or misleading metaphor? Water Int.28, 4–11.

Ariyoruk, A., 2003. Turkish water to Israel? Policy Watch No. 782(August 14, 2003). The Washington Institute for Near East

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 220

Policy, Washington DC 20036, USA. Available at http://www.washingtoninstitute.org/templateC05.php?CID=1660.

Ayers, R.S., Westcot, D.W., 1985. Water Quality for Agriculture.FAO Irrigation and Drainage Paper 29 (Rev. 1). Food andAgriculture Organization, Rome, Italy, 174 pp.

Bajwa, M.S., Josan, A.S., 1989. Effect of alternating sodic andnon-sodic irrigation on build up of sodium in soil and cropyield in northern India. Exp. Agric. 25, 199–205.

Bennett, A.J., 2000. Environmental consequences of increasingproduction: some current perspectives. Agric. Ecosys.Environ. 82, 89–95.

Blumenthal, U.J., Peasey, A., Ruiz-Palacios, G., Mara, D.D., 2000.Guidelines for wastewater reuse in agriculture andaquaculture: Recommended revisions based on newresearch evidence. WELL Study, Task No. 68 (Part 1), LondonSchool of Hygiene and Tropical Medicine, UK, 67 pp.

Boers, Th.M., Ben-Asher, J., 1982. A review of rainwaterharvesting. Agric. Water Manage. 5, 145–158.

Bohlke, J.-K., 2002. Groundwater recharge and agriculturalcontamination. Hydrogeol. J. 10, 153–179.

Bouwer, H., 2000. Integrated water management:emerging issues and challenges. Agric. Water Manage. 45,217–228.

Bouwer, H., 2002. Integrated water management for the 21stcentury: problems and solutions. J. Irrig. Drain. Eng. 128,193–202.

Carr, R.M., Blumenthal, U.J., Mara, D.D., 2004. Health guidelinesfor the use of wastewater in agriculture: developing realisticguidelines. In: Scott, C.A., Faruqui, N.I., Raschid-Sally, L.(Eds.), Wastewater Use in Irrigated Agriculture. CABIPublishing, Wallingford, UK, pp. 41–58.

Chandra, R., 2001. Development of irrigation managementstrategies: a case study. ME Thesis, Maharana PratapUniversity of Agriculture and Technology, Udaipur, India.

Chapagain, A.K., Hoekstra, A.Y., 2003. Virtual water flowsbetween nations in relation to trade in livestock andlivestock products. Value of Water Research Report SeriesNo. 13, UNESCO-IHE Institute for Water Education, Delft,The Netherlands, 59 pp.

Cosgrove, W.J., Rijsberman, F., 2000. World Water Vision:Making Water Everybody’s Business. World Water Council,World Water Vision, and Earthscan, 107 pp.

Critchley, W., Siegert, W.K., 1991. Water Harvesting:A Manual for the Design and Construction of WaterHarvesting Schemes for Plant Production, AGL/MICS/17/91.Food and Agriculture Organization of the UnitedNations, Rome.

Custodio, E., 2002. Aquifer overexploitation: what does it mean?Hydrogeol. J. 10, 254–277.

de Fraiture, C., Cai, X., Amarasinghe, U., Rosegrant, M., Molden,D., 2004. Does international cereal trade save water? Theimpact of virtual water trade on global water use.Comprehensive Assessment Research Report 4,International Water Management Institute (IWMI),Colombo, Sri Lanka, 32 pp.

Drechsel, P., Blumenthal, U.J., Keraita, B., 2002. Balancing healthand livelihoods: adjusting wastewater irrigation guidelinesfor resource-poor countries. Urban Agric. Mag. 8, 7–9.

Ensink, J.H.J., Van der Hoek, W., Matsuno, Y., Munir, S., Aslam,M.R., 2002. Use of untreated wastewater in peri-urbanagriculture in Pakistan: risks and opportunities. ResearchReport 64, International Water Management Institute(IWMI), Colombo, Sri Lanka, 22 pp.

FAO (Food and Agriculture Organization of the United Nations),2003. Review of World Water Resources by Country. WaterReports 23, FAO, Rome, Italy, 110 pp.

FAO (Food and Agriculture Organization of the United Nations),2005. Aquastat Information System on Water andAgriculture. FAO, Rome, Italy. Available at:

http://www.fao.org/WAICENT/FAOINFO/AGRICULT/AGL/aglw/aquastat/main/index.stm.

Falkenmark, M., Lindh, G., 1993. Water and economicdevelopment. In: Gleick, P.H. (Ed.), Water in Crisis. OxfordUniversity Press, New York, USA, pp. 80–91.

Government of India, 2003. Inter Basin Water TransferProposals. Task Force on Interlinking of Rivers. Ministry ofWater Resources, New Delhi, India, 27 pp.

Grattan, S.R., Rhoades, J.D., 1990. Irrigation with saline groundwater and drainage water. In: Tanji, K.K. (Ed.), AgriculturalSalinity Assessment and Management, Manuals andReports on Engineering Practices No. 71. American Societyof Civil Engineers, New York, USA, pp. 432–449.

Grattan, S.R., Grieve, C.M., Poss, J.A., Robinson, P.H., Suarez,D.L., Benes, S.E., 2004. Evaluation of salt-tolerant forages forsequential water reuse systems. I. Biomass production.Agric. Water Manage. 70, 109–120.

Grieve, C.M., Poss, J.A., Grattan, S.R., Suarez, D.L., Benes, S.E.,Robinson, P.H., 2004. Evaluation of salt-tolerant forages forsequential water reuse systems II. Plant-ion relations. Agric.Water Manage. 70, 121–135.

Haddad, M., Mizyed, N., 2004. Non-conventional options forwater supply augmentation in the Middle East: a case study.Water Int. 29, 232–242.

Hatfield, J.L., Sauer, T.J., Prueger, J.H., 2001. Managing soils toachieve greater water use efficiency. Agron. J. 93, 271–280.

Hoekstra, A.Y., 2003. Virtual water: an introduction. In:Hoekstra, A.Y. (Ed.), Virtual Water Trade: Proceedings of theInternational Expert Meeting on Virtual Water Trade,December 12–13, 2002, Delft, The Netherlands, pp. 13–23.

Hoekstra, A.Y., Hung, P.Q., 2003. Virtual water trade: aquantification of virtual water flows between nations inrelation to international crop trade. In: Hoekstra, A.Y. (Ed.),Virtual Water Trade: Proceedings of the International ExpertMeeting on Virtual Water Trade, December 12–13, 2002,Delft, The Netherlands, pp. 25–47.

Hussain, I., Raschid, L., Hanjra, M.A., Marikar, F., Van der Hoek,W., 2002. Wastewater use in agriculture: review of impactsand methodological issues in valuing impacts. WorkingPaper 37, International Water Management Institute,Colombo, Sri Lanka.

IWMI (International Water Management Institute), 2003.Confronting the realities of wastewater use in agriculture.In: Water Policy Briefing 9, IWMI, Colombo, Sri Lanka, 6 pp.

Jaber, J.O., Mohsen, M.S., 2001. Evaluation of non-conventionalwater resources supply in Jordan. Desalination 136, 83–92.

Jimenez, B., Asano, T., 2004. Acknowledge all approaches: theglobal outlook on reuse. Water 21 (Mag. Int. Water Assoc.)32–37.

Karajeh, F., Tanji, K., 1992a. A numerical model for water andsalt simulation in an agroforestry system. II. Field waterflow. J. Irrig. Drain. Eng., ASCE 120, 397–413.

Karajeh, F., Tanji, K., 1992b. A numerical model for water andsalt simulation in an agroforestry system. III. Field salt flow.J. Irrig. Drain. Eng., ASCE 120, 588–604.

Karajeh, F., Tanji, K., King, I., 1992. A numerical model for waterand salt simulation in an agroforestry system. I. Theory andvalidation. J. Irrig. Drain. Eng., ASCE 120, 382–396.

Keraita, B.N., Drechsel, P., 2004. Agricultural use of untreatedurban wastewater in Ghana. In: Scott, C.A., Faruqui, N.I.,Raschid-Sally, L. (Eds.), Wastewater Use in IrrigatedAgriculture. CABI Publishing, Wallingford, UK, pp. 101–112.

Kijne, J.W., 1996. Water and Salinity Balances for IrrigatedAgriculture in Pakistan. Research Paper 4, InternationalIrrigation Management Institute, Colombo, Sri Lanka.

Kijne, J.W., 2003. Water productivity under saline conditions. In:Kijne, J.W., Barker, R., Molden, D. (Eds.), Water Productivityin Agriculture: Limits and Opportunities for Improvement.CABI Publishing, Wallingford, UK, pp. 89–102.

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 2 21

Kijne, J.W., Barker, R., Molden, D., 2003. Water Productivity inAgriculture: Limits and Opportunities for Improvement.CABI Publishing, Wallingford, UK, 332 pp.

Lant, C., 2003. Commentary. Water Int. 28, 113–115.Latorre, M., 2002. Design optimisation of SWRO plants for

irrigation. In: Proceedings of the International DesalinationAssociation’s World Congress on Desalination and WaterReuse, March 8–13, 2002, Manama, Bahrain.

Loucks, D.P., 2000. Sustainable water resources management.Water Int. 25, 3–10.

Matsuno, Y., Ensink, J.H.J., Van der Hoek, W., Simmons, R.W.,2004. Assessment of the use of wastewater for irrigation: acase in Punjab, Pakistan. In: Proceedings of the Symposiumon Wastewater Reuse and Groundwater Quality, July 2003,Sapporo, IAHS Publication 285, pp. 28–33.

Merrett, S., 2003a. Virtual water and Occam’s razor. Water Int.28, 103–105.

Merrett, S., 2003b. Virtual water and the Kyoto consensus.Water Int. 28, 540–542.

Minhas, P.S., Tyagi, N.K., 1998. Guidelines for Irrigation withSaline and Alkali Waters. Bulletin. Central Soil SalinityResearch Institute, Karnal, India.

Minhas, P.S., Samra, J.S., 2003. Quality Assessment of WaterResources in the Indo-Gangetic Basin Part in India. CentralSoil Salinity Research Institute, Karnal, India, 68 pp.

Minhas, P.S., Samra, J.S., 2004. Wastewater Use in Peri-urbanAgriculture: Impacts and Opportunities. Central SoilSalinity Research Institute, Karnal, India, 75 pp.

Minhas, P.S., Sharma, D.R., Chauhan, C.P.S., 2004. Managementof saline and alkali waters for irrigation. In: Advances inSodic Land Reclamation. Proceedings of the InternationalConference on Sustainable Management of Sodic lands,February 9–14, 2004, Lucknow, India, pp. 121–162.

Mohsen, M.S., Al-Jayyousi, O.R., 1999. Brackish waterdesalination: an alternate for water supply enhancement inJordan. Desalination 124, 163–174.

Oster, J.D., Birkle, D.E., 2004. Potential crop production systemsusing saline irrigation waters. In: Taha, F.K., Ismail, S.,Jaradat, A. (Eds.), Prospects of Saline Agriculture in theArabian Peninsula. Proceedings of the InternationalSeminar on Prospects of Saline Agriculture in the GCCCountries, March 18–20, 2001, Dubai, United Arab Emirates,pp. 277–298.

Oster, J.D., Grattan, S.R., 2002. Drainage water reuse. Irrig. Drain.Syst. 16, 297–310.

Oster, J.D., Wichelns, D., 2003. Economic and agronomicstrategies to achieve sustainable irrigation. Irrig. Sci. 22,107–120.

Oster, J.D., Singer, M.J., Fulton, A., Richardson, W., Prichard, T.,1992. Water Penetration Problems in California Soils:Prevention, Diagnoses and Solutions. Kearney Foundationof Soil Science, Division of Agriculture and naturalResources, University of California, USA, 165 pp.

Oweis, T., Zhang, H., Pala, M., 2000. Water use efficiency ofrainfed and irrigated bread wheat in a Mediterraneanenvironment. Agron. J. 92, 231–238.

Oweis, T.Y., Hachum, A., 2003. Improving water productivity inthe dry areas of West Asia and North Africa. In: Kijne, J.W.,Barker, R., Molden, D. (Eds.), Water Productivity inAgriculture: Limits Opportunities for Improvement. CABIPublishing, Wallingford, UK, pp. 179–198.

Oweis, T., Hachum, A., Kijne, J., 1999. Water harvesting andsupplemental irrigation for improved water use efficiencyin dry areas. SWIM Paper 7, International WaterManagement Institute, Colombo, Sri Lanka, 41 pp.

Oweis, T.Y., Hachum, A., Bruggeman, A., 2004. IndigenousWater-Harvesting Systems in West Asia and North Africa.International Center for Agricultural Research in the DryAreas (ICARDA), Aleppo, Syria, 173 pp.

Oweis, T., Prinz, D., Hachum, A., 2001. Water Harvesting:Indigenous Knowledge for the Future of the DrierEnvironments. ICARDA, Aleppo, Syria, 40 pp.

Pal, R., Poonia, S.R., 1979. Dimensions of gypsum bed in relationto residual sodium carbonate of irrigation water, size ofgypsum fragments and flow velocity. J. Indian Soc. Soil Sci.27, 5–10.

Pandey, D.N., Gupta, A.K., Anderson, D.M., 2003. Rainwaterharvesting as an adaptation to climate change. Curr. Sci. 85,46–59.

Pearce, F., 2004. Sea change for drinking water. New Sci. 22(10 July).

Pimentel, D., Bailey, O., Kim, P., Mullaney, E., Calabrese, J.,Walman, L., Nelson, F., Yao, X., 1999. Will limits of theEarth’s resources control human numbers? Environ.Sustain. Dev. 1, 19–39.

Postel, S., 1993. Water and agriculture. In: Gleick, P.H. (Ed.),Water in Crisis: A Guide to the World’s Fresh WaterResources. Oxford University Press, Oxford, UK, pp. 56–66.

Prinz, D., 1996. Water harvesting: past and future. In: Pereira,L.S. (Ed.), Sustainability of Irrigated Agriculture. Proceedingsof the NATO Advanced Research Workshop, March 21–23,1994, Balkema, Rotterdam, The Netherlands, pp. 135–144.

Qadir, M., Oster, J.D., 2004. Crop and irrigation managementstrategies for saline–sodic soils and waters aimed atenvironmentally sustainable agriculture. Sci. Total Environ.323, 1–19.

Qadir, M., Schubert, S., 2002. Degradation processes andnutrient constraints in sodic soils. Land Degrad. Develop.13, 275–294.

Qadir, M., Ghafoor, A., Murtaza, G., 2000. Cadmiumconcentration in vegetables grown on urban soils irrigatedwith untreated municipal sewage. Environ. Develop.Sustain. 2, 11–19.

Qadir, M., Boers, Th.M., Schubert, S., Ghafoor, A., Murtaza, M.,2003. Agricultural water management in water-starvedcountries: challenges and opportunities. Agric. WaterManage. 62, 165–185.

Qadir, M., Schubert, S., Steffens, D., 2005. Phytotoxic substancesin soils. In: Hillel, D. (Editor-in-Chief). Encyclopedia of Soils inthe Environment. Elsevier Science, Oxford, UK, pp. 216–222.

Qureshi, R.H., Barrett-Lennard, E.G., 1998. Saline Agriculture forIrrigated Land in Pakistan: A Handbook. Australian Centrefor International Agricultural Research (ACIAR), Canberra,Australia, 142 pp.

Rhoades, J.D., 1989. Intercepting, isolating and reusing drainagewaters for irrigation to conserve water and protect waterquality. Agric. Water Manage. 16, 37–52.

Rhoades, J.D., 1999. Use of saline drainage water for irrigation.In: Skaggs, R.W., van Schilfgaarde, J. (Eds.), AgriculturalDrainage. American Society of Agronomy (ASA)-CropScience Society of America (CSSA)-Soil Science Society ofAmerica (SSSA), Madison, Wisconsin, USA, pp. 615–657.

Rijsberman, F.R., 2006. Water scarcity: fact or fiction? Agric.Water Manage. 80, 5–22.

Rockstrom, J., Falkenmark, M., 2000. Semiarid crop productionfrom a hydrological perspective: gap between potential andactual yields. Crit. Rev. Plant Sci. 19, 319–346.

Rosegrant, M.W., Cai, X., Cline, S.A., 2002. World Water andFood to 2025. International Food Policy Research Institute,Washington, USA, 322 pp.

Scott, C.A., Zarazua, J.A., Levine, G., 2000. Urban-wastewaterreuse for crop production in the water-short Guanajuatoriver basin, Mexico. Research Report 41, International WaterManagement Institute, Colombo, Sri Lanka, 34 pp.

Scott, C.A., Faruqui, N.I., Raschid-Sally, L., 2004. Wastewater Usein Irrigated Agriculture: Confronting the Livelihoods andEnvironmental Realities. CABI Publishing, Wallingford, UK,193 pp.

a g r i c u l t u r a l w a t e r m a n a g e m e n t 8 7 ( 2 0 0 7 ) 2 – 2 222

Seckler, D., Amerasinghe, U., Molden, D., de Silva, R., Barker, R.,1998. World Water Demand and Supply 1990 to 2025:Scenarios and Issues. International Water ManagementInstitute, Colombo, Sri Lanka, 40 pp.

Semiat, R., 2000. Desalination: present and future. Water Int. 25,54–65.

Sharma, B.R., Minhas, P.S., 2005. Strategies for managing saline/alkali waters for sustainable agricultural production inSouth Asia. Agric. Water Manage. 78, 136–151.

Sharma, D.P., Rao, K.V.G.K., 1996. Recycling drainage effluent forcrop production. In: Proceedings of the Workshop onWaterlogging Salinity in Irrigated Agriculture, March 12–15,1996. Central Soil Salinity Research Institute, Karnal, India,pp. 134–147.

Sharma, N., Dixon, J., 1995. Farmers’ Organization Network forPeople’s Participation in Watershed Management in Asia.State of Watershed Management in Asia. Department ofSoil Conservation, Kathmandu, Nepal.

Shende, G.B., 1985. Status of wastewater treatment andagricultural reuse with special reference to Indianexperience and research and development needs. In:Pescod, M.B., Arar, A. (Eds.), Proceedings of the FAORegional Seminar on the Treatment and Use of SewageEffluent for Irrigation. October 7–9, 1985, Nicosia.Butterworths, London, UK.

Shiklomanov, I.A., 2000. Appraisal and assessment of worldwater resources. Water Int. 25, 11–32.

Sumner, M.E., 1993. Sodic soils: new perspectives. Austr. J. SoilRes. 31, 683–750.

Tabor, J.A., 1995. Improving crop yields in the Sahel by means ofwater-harvesting. J. Arid. Environ. 30, 83–106.

Tanji, K., Kielen, N.C., 2002. Agricultural drainage watermanagement in arid and semi-arid areas. In: FAO Irrigationand Drainage Paper No. 61, Food and AgricultureOrganization, Rome, Italy.

Tingu, L., Juan, X., Guangyong, L., Jiah Hua, M., Jianping, W.,Zhinzhong, L., Jianguo, Z., 2003. Effect of drip irrigation withsaline water on water use efficiency and quality ofwatermelons. Water Resour. Manage. 17, 395–408.

Tomar, O.S., Minhas, P.S., Sharma, V.K., Singh, Y.P., Gupta, R.K.,2002. Performance of 32 tree species and soil conditions in aplantation established with saline irrigation. For. Ecol.Manage. 177, 333–346.

Tyagi, N.K., 2003. Managing saline and alkaline water for higherproductivity. In: Kijne, J.W., Barker, R., Molden, D. (Eds.),Water Productivity in Agriculture: Limits Opportunities forImprovement. CABI Publishing, Wallingford, UK, pp. 69–87.

Tyagi, N.K., Sharma, D.P., 2000. Disposal of drainage water:recycling and reuse. In: Proceedings of the Eighth ICID

Drainage Workshop, vol. 3, January 31–February 4, 2000,New Delhi, India, pp. 199–213.

Van der Hoek, W., Ul Hassan, M., Ensink, J.H.J., Feenstra, S.,Raschid-Sally, L., Munir, S., Aslam, M.R., Ali, N., Hussain, R.,Matsuno, Y., 2002. Urban wastewater: a valuable resourcefor agriculture. Research Report 63, International WaterManagement Institute, Colombo, Sri Lanka, 20 pp.

Van der Hoek, W., 2004. A framework for a global assessment ofthe extent of wastewater irrigation: the need for a commonwastewater typology. In: Scott, C.A., Faruqui, N.I., Raschid-Sally, L. (Eds.), Wastewater Use in Irrigated Agriculture.CABI Publishing, Wallingford, UK, pp. 11–25.

Van Schilfgaarde, J., 1994. Irrigation—a blessing or a curse.Agric. Water Manage. 25, 203–232.

Voutchkov, N., 2004. The Ocean: a new resource for drinkingwater. Public Works 30–33.

Voutchkov, N., 2005. Personal communication. Senior VicePresident, Technical Services, Poseidon Resources, 1055Washington Boulevard, Stamford, CT 06901, USA.

Wagner, G.J., 1993. Accumulation of cadmium in crop plantsand its consequences for human health. Adv. Agron. 51,173–212.

Wallace, J.S., 2000. Increasing agricultural water use efficiencyto meet future food production. Agric. Ecosys. Environ. 82,105–119.

WHO (World Health Organization), 1989. In: Mara, D.,Cairncross, S. (Eds.), Guidelines for the Safe Use ofWastewater and Excreta in Agriculture and Aquaculture.WHO, Geneva, p. 187.

WHO (World Health Organization), 2006. Guidelines for the SafeUse of Wastewater, Excreta and Grey Water: WastewaterUse in Agriculture, vol. 2, 219 pp.

Wichelns, D., 2001. The role of ‘virtual water’ in efforts toachieve food security and other national goals, withexamples from Egypt. Agric. Water Manage. 49, 131–151.

WRI (World Resources Institute) in collaboration with UnitedNations Development Programme (UNDP), United NationsEnvironment Programme (UNEP) and World Bank (WB),2005. The Wealth of the Poor—Managing Ecosystems toFight Poverty. WRI, Washington, DC, 254 pp.

WWC (World Water Council), 2004. Virtual water trade—conscious choices. In: E-Conference Synthesis. WWC, 31 pp.

Zimmer, D., Renault, D., 2003. Virtual water in food productionand global trade: review of methodological issues andpreliminary results. In: Hoekstra, A.Y. (Ed.). Virtual WaterTrade: Proceedings of the International Expert Meeting onVirtual Water Trade, 12–13 December 2002. Delft. Value ofWater Research Report Series No. 12, UNESCO-IHE Institutefor Water Education, Delft, The Netherlands.