The resilience of big river basins

PLEASE SCROLL DOWN FOR ARTICLE

!"#$%&'(#)*+%,&$%-.,/*.&-+-%012%34(.)5".*6%7/#8+'$#(1%9#0'&'1%:4(.)5".*6$%7/#8+'$#(+(;<=/2%>%?+0'@&'1%ABCCD))+$$%-+(&#*$2%D))+$$%E+(&#*$2%3$@0$)'#F(#./%/@60+'%GAHIJKBGH<L@0*#$"+'%M.@(*+-N+O/P.'6&%9(-%M+N#$(+'+-%#/%Q/N*&/-%&/-%R&*+$%M+N#$(+'+-%S@60+'2%CBKAGTJ%M+N#$(+'+-%.PP#)+2%U.'(#6+'%V.@$+W%IKX

JC%U.'(#6+'%4('++(W%9./-./%RC!%IYVW%7Z

R&(+'%O/(+'/&(#./&*L@0*#)&(#./%-+(&#*$W%#/)*@-#/N%#/$('@)(#./$%P.'%&@(".'$%&/-%$@0$)'#F(#./%#/P.'6&(#./2

"((F2[[,,,\#/P.'6&,.'*-\).6[$6FF[(#(*+])./(+/(^(KGA>CT>KH

!"+%'+$#*#+/)+%.P%0#N%'#8+'%0&$#/$_'&+6+%4\%`@66#/N&

&%L+')1%?#(aL&('#)5%O/$(#(@(+W%E4![SM?%`+/('+%.P%Qb)+**+/)+W%7/#8+'$#(1%.P%`&F+%!.,/W%M./-+0.$)"W

`&F+%!.,/W%4.@("%DP'#)&

=/*#/+%F@0*#)&(#./%-&(+2%AB%Y&/@&'1%ABCC

!.%)#(+%("#$%D'(#)*+%`@66#/NW%_'&+6+%4\:ABCC;%c!"+%'+$#*#+/)+%.P%0#N%'#8+'%0&$#/$cW%R&(+'%O/(+'/&(#./&*W%IH2%CW%HI%d%GT

!.%*#/5%(.%("#$%D'(#)*+2%E=O2%CB\CB>B[BATB>BHB\ABCC\TJCBCH

7M92%"((F2[[-b\-.#\.'N[CB\CB>B[BATB>BHB\ABCC\TJCBCH

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

http://dx.doi.org/10.1080/02508060.2011.541016

http://www.informaworld.com/terms-and-conditions-of-access.pdf

http://www.informaworld.com/smpp/title~content=t792815876

Water InternationalVol. 36, No. 1, January 2011, 63–95

The resilience of big river basins

Graeme S. Cumming*

Percy FitzPatrick Institute, DST/NRF Centre of Excellence, University of Cape Town, Rondebosch,Cape Town, South Africa 7701

Big river basins are complex systems of people and nature. This article explores theresilience of nine case studies of big river basins. A system description and genericconceptual model suggests that resilience to changes in water quantity is critical.When water becomes limiting, the social-ecological system must adapt rapidly if keyelements (for example, communities, biodiversity) are to be maintained. Water lim-itation imposes a water economy and alters political and institutional links betweenactors. Pro-active management for resilience demands politically acceptable participa-tory processes that use the best possible science and incorporate social, ecological andeconomic elements in problem definition and solution.

Keywords: complexity; sustainability; social–ecological system; ecosystem services;governance; poverty

Introduction

This paper explores the central issues relating to ecosystem services and human well-being in a set of 10 big river basins from around the world. Each basin has importantsocial, economic and ecological elements; each can be considered a social–ecologicalsystem (SES), a linked system of people and nature. As the preceding papers makeabundantly clear, the big river basins in this set of case studies are similar in some impor-tant ways and different in other, equally important ways. They all face big challengesin the future, including problems that are global in nature (such as climate change andhuman population expansion) and problems that are more localized (such as regional con-flict or inequities). From both analytical and management perspectives, it is important tounderstand the ways in which what has happened in one basin can inform our under-standing of other basins. Can we learn from comparisons between these basins whether,and how, they will be able to cope effectively with what the future holds for them? Andmore specifically, can analysis of their current strengths and weaknesses identify pos-sible pitfalls or pro-active strategies through which additional coping ability could bebuilt?

As with any comparative analysis, answering these questions requires a level of gen-eralization. Generalization in turn requires reference to a set of underlying conceptsthat can be used to identify commonalities between systems that are superficially quitedifferent. There are many different ways in which such an analysis could be framed,

!Email: [email protected]

ISSN 0250-8060 print/ISSN 1941-1707 online© 2011 International Water Resources AssociationDOI: 10.1080/02508060.2011.541016http://www.informaworld.com

Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011

mailto:[email protected]

http://www.informaworld.com

64 G.S. Cumming

and no single guide (aside from eventual utility) as to which framing is best. Traditionalapproaches to the problems of big river basins have tended to approach specific issuesfrom a disciplinary perspective. For example, the allocation and efficiency of water usehave been major themes in assessing the economics of river basin management (forexample, Rosegrant et al. [2000]). While traditional approaches often provide essentialinformation about individual aspects of a broader problem, they also make a large num-ber of assumptions about the constancy of causal relationships, the prevalence of linearrelationships, and the structure of the study system. The accuracy of standard scientificpredictions is heavily contingent on things remaining as they are (Clark et al. 2001).In the context of river management, which includes high levels of uncertainty in manysocial and ecological parameters, conventional approaches to minimizing risks are quitelikely to be sub-optimal (Clark 2002). Furthermore, the real world includes “messy”details like pushy stakeholders, political processes, technological change and feedbacksbetween science and human behaviour (Ludwig 2001, Stirling 2003, Waltner-Toews et al.2003).

Big river basins are similar to many other systems of people and nature in thatthey involve a set of complex, “wicked” interactions1 and non-linear feedbacks. Theyare united by a shared biophysical component – water – which flows directionallyalong a well-defined gradient. Human societies have in many cases organized them-selves around and along this gradient, with predictable variation in human-dominatedcomponents of the system (for example, income or city size) often occurring from head-waters to estuaries. The inescapable importance of environmental gradients, and the heavyreliance of human communities in a basin on a single resource (water), are probablythe unique features of big river basins that distinguish them from other social–ecologicalsystems.

A number of approaches to thinking about “wicked problems” in complex systemsexist (e.g., Holling 2001, Norberg and Cumming 2008, Waltner-Toews et al. 2008, Ostrom2009). The approach that I will take focuses on the idea of resilience (Holling 1973, Holling2001), and particularly on the ability of key elements of the system to persist into the future(Cumming et al. 2005). I will first describe some elements of a general framework and thenattempt to apply it in an integrative (if somewhat qualitative) analysis of the resilience ofbig river basins.

Resilience thinking

Resilience is used here as an emergent attribute of a social–ecological system. Resiliencethinking (Walker and Salt 2006) has its roots in systems approaches, in which theworld is perceived and described as a set of interacting “systems” (Checkland 2009).Systems are a form of model in that they are defined by the observer as a way of mak-ing sense of the complexities of the real world; they are intended to capture essentialelements of a problem, rather than to describe accurately the full range of complex-ity that exists “out there”. Analysis of resilience focuses on understanding how thenumber and nature of current system components, and their interactions and broadercontext, influence the ability of important elements of that system to persist into thefuture.

Although the concept of resilience has become more widely used in recent years,some confusion still exists about the definition of resilience and some related terms. Somecurrent definitions of resilience-related terminology as used in this paper are offered inTable 1.


Water International 65

Tabl

e1.

Defi

nitio

nsof

som

eof

the

term

sus

edin

thin

king

abou

tres

ilien

ce.

Term

Defi

nitio

n

Syst

em(C

umm

ing

and

Col

lier

2005

)A

cohe

sive

entit

yco

nsis

ting

ofke

yel

emen

ts,i

nter

actio

nsan

da

loca

lenv

iron

men

t;m

usts

how

spat

io-t

empo

ral

cont

inui

ty.A

nSE

Sin

clud

esin

tera

ctio

nsbe

twee

npe

ople

and

ecos

yste

ms.

Res

ilien

ce(H

ollin

g19

73;C

arpe

nter

etal

.200

1;H

ollin

g20

01;

Cum

min

get

al.2

005)

The

rear

eat

leas

ttw

oco

mpl

emen

tary

defin

ition

sof

resi

lienc

eth

atar

ecu

rren

tlyus

edin

the

stud

yof

SESs

.Car

pent

eret

al.(

2001

)de

fine

resi

lienc

eas

:(1)

the

amou

ntof

chan

geth

ata

syst

emca

nun

derg

oan

dst

illm

aint

ain

the

sam

eco

ntro

lson

func

tion

and

stru

ctur

e;(2

)th

esy

stem

’sab

ility

tose

lf-o

rgan

ize;

and

(3)

the

degr

eeto

whi

chth

esy

stem

isca

pabl

eof

lear

ning

and

adap

tatio

n.Se

cond

,bas

edon

Cum

min

get

al.(

2005

),sy

stem

iden

tity

depe

nds

onm

aint

aini

nges

sent

iale

lem

ents

thro

ugh

spac

ean

dtim

e.R

esili

ence

can

thus

bevi

ewed

asth

eab

ility

ofth

esy

stem

tom

aint

ain

itsid

entit

yin

the

face

ofin

tern

alch

ange

and

exte

rnal

pert

urba

tions

.Not

eth

atth

ese

defin

ition

sof

fer

slig

htly

diff

eren

tper

spec

tives

onth

esa

me

conc

ept;

they

are

noti

nco

mpe

titio

nw

ithea

chot

her.

Rob

ustn

ess

(And

erie

set

al.2

004;

Lev

inan

dL

ubch

enco

2008

)

And

erie

set

al.(

2004

)ci

tean

engi

neer

ing

defin

ition

(Car

lson

and

Doy

le20

02)

focu

sing

on“m

aint

enan

ceof

syst

empe

rfor

man

ceei

ther

whe

nsu

bjec

ted

toex

tern

al,u

npre

dict

able

pert

urba

tions

,or

whe

nth

ere

isun

cert

aint

yab

outt

heva

lues

ofin

tern

alde

sign

para

met

ers”

.Lev

inan

dL

ubch

enco

(200

8)eq

uate

robu

stne

ssan

dre

silie

nce,

stat

ing

that

both

“mea

nm

uch

the

sam

eth

ing:

the

capa

city

ofsy

stem

sto

keep

func

tioni

ngev

enw

hen

dist

urbe

d”.M

aint

enan

ceof

func

tion

isa

diff

eren

tcri

teri

onfr

omm

aint

enan

ceof

iden

tity

(unl

ess

iden

tity

isde

fined

pure

lyin

term

sof

func

tion,

whi

chis

notu

sual

lyth

eca

se)

and

this

func

tiona

lem

phas

isap

pear

s,as

best

Ica

nte

ll,to

beth

epr

imar

ydi

stin

guis

hing

feat

ure

of“r

obus

tnes

s”.

Vul

nera

bilit

y(A

dger

2006

)V

ulne

rabi

lity

isa

mea

sure

ofth

eex

tent

tow

hich

aco

mm

unity

,str

uctu

re,s

ervi

ceor

geog

raph

ical

area

islik

ely

tobe

dam

aged

ordi

srup

ted,

onac

coun

tof

itsna

ture

orlo

catio

n,by

the

impa

ctof

apa

rtic

ular

disa

ster

haza

rd(O

rgan

isat

ion

for

Eco

nom

icC

o-op

erat

ion

and

Dev

elop

men

t[O

EC

D]

Glo

ssar

y).A

mor

ere

silie

nce-

orie

nted

defin

ition

isof

fere

dby

Adg

er(2

006)

:“V

ulne

rabi

lity

isth

est

ate

ofsu

scep

tibili

tyto

harm

from

expo

sure

tost

ress

esas

soci

ated

with

envi

ronm

enta

land

soci

alch

ange

and

from

the

abse

nce

ofca

paci

tyto

adap

t.”

Sust

aina

bilit

y(N

orbe

rgan

dC

umm

ing

2008

)Fi

rsti

ntro

duce

dto

the

polic

yar

ena

byth

eB

runt

land

Rep

ort(

WE

CD

1987

)w

hich

defin

edsu

stai

nabl

ede

velo

pmen

tas

deve

lopm

entt

hat“

mee

tsth

ene

eds

ofth

epr

esen

tgen

erat

ion

with

outc

ompr

omis

ing

the

abili

tyof

futu

rege

nera

tions

tom

eett

heir

need

s”.A

lso

defin

edas

“the

equi

tabl

e,et

hica

l,an

def

ficie

ntus

eof

natu

ralr

esou

rces

”(N

orbe

rgan

dC

umm

ing

2008

).In

clus

ion

of“e

ffici

ency

”is

cont

entio

usin

som

eci

rcle

s.

Reg

ime,

regi

me

shif

t(C

arpe

nter

2003

;Sch

effe

ran

dC

arpe

nter

2003

)

Alo

cally

stab

leor

self

-rei

nfor

cing

seto

fco

nditi

ons

that

caus

ea

syst

emto

vary

arou

nda

loca

lattr

acto

r;th

edo

min

ants

etof

driv

ers

and

feed

back

sth

atle

adto

syst

embe

havi

our;

a“b

asin

ofat

trac

tion”

.For

ecos

yste

ms,

are

gim

esh

ifti

s“a

rapi

dm

odifi

catio

nof

ecos

yste

mor

gani

zatio

nan

ddy

nam

ics,

with

prol

onge

dco

nseq

uenc

es”

(Car

pent

er20

03,c

hapt

er1)

.

(Con

tinue

d)


66 G.S. Cumming

Tabl

e1.

(Con

tinue

d)

Term

Defi

nitio

n

Thr

esho

ld(S

chef

fer

and

Car

pent

er20

03;

Sche

ffer

2009

)

Inno

n-lin

earr

elat

ions

hips

,the

poin

tatw

hich

afu

nctio

n(o

rara

te)c

hang

essi

gn;i

nst

ate

spac

e,th

elo

catio

nat

whi

cha

syst

emte

nds

tow

ards

anal

tern

ativ

elo

cale

quili

briu

mor

new

attr

acto

r;th

eco

mbi

natio

nof

vari

able

sun

der

whi

cha

syst

emen

ters

ane

wre

gim

e.

Feed

back

(Nor

berg

and

Cum

min

g20

08)

Occ

urs

whe

na

chan

gein

syst

emel

emen

tAtr

igge

rsch

ange

sin

othe

rsy

stem

elem

ents

(B,C

,and

soon

)th

atev

entu

ally

influ

ence

elem

entA

.Fee

dbac

ksm

aybe

self

-reg

ulat

ing

(neg

ativ

e,ho

meo

stat

ic)

orse

lf-a

mpl

ifyi

ng(p

ositi

ve).

For

exam

ple,

inth

ehu

man

body

,sw

eatin

gis

ase

lf-r

egul

atin

gre

spon

seth

atco

ols

the

body

dow

n,pu

shin

git

back

tow

ards

aneq

uilib

rium

tem

pera

ture

;whi

lehy

pert

herm

iaan

dco

nvul

sion

sin

resp

onse

tohe

atse

rve

only

toex

acer

bate

the

prob

lem

,mov

ing

the

body

furt

her

away

from

itspr

efer

red

equi

libri

umpo

int.

Self

-am

plif

ying

feed

back

sar

epa

rtic

ular

lyim

port

anti

nso

cial

–eco

logi

cals

yste

ms

asth

eyca

npu

sha

syst

emov

era

thre

shol

dan

din

toa

new

regi

me.

The

refe

renc

esin

the

left

-han

dco

lum

nof

fer

are

cent

exam

ple

ofa

peer

-rev

iew

edpu

blic

atio

nin

whi

chth

ete

rmis

used

and/

orde

fined

.Not

eth

atin

man

yca

ses,

diff

eren

tres

earc

hgr

oups

have

been

wor

king

with

very

sim

ilar

conc

epts

usin

gdi

ffer

entt

erm

inol

ogy;

diff

eren

ces

here

tend

tom

atte

rle

ssth

ansi

mila

ritie

s.Te

rmin

olog

ical

diff

eren

ces

also

refle

ctth

epl

ural

ityof

com

plex

syst

ems

(tha

tis,

mos

tcom

plex

syst

ems

can

bevi

ewed

from

man

ydi

ffer

entb

uteq

ually

“cor

rect

”pe

rspe

ctiv

es).

The

tabl

eis

mod

ified

from

Cum

min

g(2

011)

.



In thinking about system resilience, it is necessary first to define the system of interest(with particular emphasis on its most important elements) and then to consider how thesystem is likely to respond to specified future changes. System definition entails many sub-jective elements (Checkland 1981, 2009) and should be approached in a systematic mannerto avoid excessive bias on the part of the analyst. One approach to system description isto break down the basic elements of a system into structural, spatial and temporal cate-gories. These categories can be considered as axes, in the sense that each system elementhas a structure or nature, a location in space, and a location in time. A similar conceptualframework has been applied by Agarwal et al. (2002) to compare models of land use andland-cover change. As proposed by Cumming et al. (2005) and Norberg and Cumming(2008), structural elements in SESs include system components, internal interactions, andother system constituents that contribute to continuity, information processing and adap-tation. The system interacts with its environment through input and output functions; andthe environment provides a set of “givens” (for example, geographic location) as well asbeing a potential source of perturbations. The elements that are included in the structuralaxis should in theory be those that are considered to be most central to the system’s iden-tity (see detailed discussion in Cumming et al. [2005]). Examples of each element in thesystem description are given in Table 2.

The spatial axis captures spatial variation, hierarchical system arrangements in space,and spatial interactions and connections at multiple scales. Spatial variation is relevantover at least three scales; internally within the system (“local”), and externally at the scaleof the immediate context (“regional”), and the larger (“global”) set of social–ecologicalsystems that impact the SES. Outside the boundaries of the focal SES, the primary spatialelements of interest include context (spatial surroundings, defined at the scale of analysis);connectivity; and spatial dynamics that are driven by connections or system inputs, suchas spatially driven feedbacks and spatial subsidies (Polis et al. 1997, Polis et al. 2004).Both internal and external spatial elements of an SES must be considered in relation to thestructural aspects of the system, including (as outlined above) the number and nature ofcomponents and interactions, the ability of the system to undergo change while maintainingits identity, system memory, and the potential inherent in the system for adaptation andlearning. Note that despite its temporal role, memory is treated as a structural aspect ofan SES because it requires a structure (for example, a brain, an archive, or a long-livedindividual) that records change in time.

The temporal axis captures temporal variation, rates and temporal hierarchies, and thetemporal location of a system within a set of processes or along a trajectory. Many of theaspects of temporal variation can be broken down in the same way as spatial elements, byconsidering local, regional and global dynamics and the relevance of sets of fast and slowvariables. In addition, key aspects of this component of the framework include the diagno-sis or assessment of the location of the system within a broader state-space, particularly inregard to its proximity to a potential threshold or regime shift; its location within an adap-tive cycle (if the adaptive cycle is deemed to fit the system of interest), including whetherkey elements of the SES are growing, collapsing, or reorganizing; and assessment of thedegree to which the system’s history has created path dependence in its current and futuredynamics (see Martin and Sunley [2006], for example).

Slow variables are considered to be particularly important in SES theory because oftheir role in creating regime shifts and alternative stable states. For example, soil fertilitymay be a slowly changing variable in an agricultural system (Issaka et al. 1997). If croprotation times and types are insufficient to allow the restoration of soil nutrients, and ifcommunities lack the means to generate or purchase fertilizer, changes in soil nutrients


68 G.S. Cumming

Tabl

e2.

Sum

mar

yof

elem

ents

ofa

gene

ric

big

basi

nSE

S.

Syst

emat

trib

ute

Gen

eric

elem

ents

1.St

ruct

ural

axis

SES

com

pone

nts

Act

ors:

farm

ers,

fishe

rs,h

ouse

hold

user

s,m

anag

emen

tage

ncie

s,lo

cala

ndin

tern

atio

naln

on-g

over

nmen

talo

rgan

izat

ions

(NG

Os)

,res

earc

hers

,dif

fere

ntet

hnic

orra

cial

grou

ps,m

inin

gco

mpa

nies

,civ

ilso

ciet

ygr

oups

,hyd

roel

ectr

icpo

wer

prod

ucer

s,br

idgi

nggr

oups

orfo

rath

atco

nnec

tdif

fere

ntst

akeh

olde

rs.

Bio

phys

ical

com

pone

nts:

wat

er,v

eget

atio

n,cr

ops,

lives

tock

,fish

.B

uilt

com

pone

nts:

impo

undm

ents

,roa

ds,c

anal

s,ci

ties.

Inst

itutio

ns:l

aws,

polic

ies,

regu

latio

ns.

SES

inte

ract

ions

Impl

emen

tatio

nof

land

tenu

rean

dw

ater

righ

ts;n

egot

iatio

nsov

erw

ater

with

draw

als;

upst

ream

–dow

nstr

eam

user

inte

ract

ions

;tr

ansb

ound

ary

polit

ics.

SES

cont

inui

tyan

dm

emor

yPr

ovid

edby

long

-ter

mpo

licie

s,la

ws

and

conv

entio

ns;l

ong-

stan

ding

man

agem

enta

genc

ies;

and

know

ledg

eabl

ein

divi

dual

s.

SES

info

rmat

ion

proc

essi

ngR

efer

sto

info

rmat

ion

avai

labi

lity,

deci

sion

mak

ing

and

resp

onse

capa

city

.Alth

ough

rese

arch

isty

pica

llyun

dert

aken

byun

iver

sitie

san

dN

GO

s,in

form

atio

npr

oces

sing

isof

ten

unde

rtak

enby

indi

vidu

als

(for

exam

ple,

farm

ers

may

resp

ond

topr

edic

tions

ofdr

ough

tby

stor

ing

exce

ssw

ater

)an

dby

orga

niza

tions

.SE

Sad

apta

tion

Rel

ated

tore

sear

chan

dlo

cali

nnov

atio

ns;m

ayal

sooc

cur

thro

ugh

the

actio

nof

sele

ctio

non

dive

rsity

,alth

ough

high

soci

aldi

vers

itym

ayin

hibi

tcol

lect

ive

actio

n.E

nvir

onm

ent

(con

text

for

SES)

:In

puts

and

outp

uts

toSE

SE

xter

nald

rive

rs,s

uch

asra

infa

llan

dgl

obal

mar

ketd

eman

dsfo

rlo

calp

rodu

ce;i

nter

natio

nala

ndna

tiona

linfl

uenc

essu

chas

rese

ttlem

ents

chem

es,e

cono

mic

ince

ntiv

es,fl

ood

com

pens

atio

nan

dot

hers

.O

utpu

tsin

clud

ew

ater

,foo

dan

din

form

atio

n.Pe

rtur

batio

nsFl

oods

,dro

ught

s,cl

imat

ech

ange

,eco

nom

icsh

ocks

,pol

itics

(for

exam

ple,

gove

rnm

enta

lcha

nges

).A

sym

met

ries

Gra

dien

tsex

ista

long

each

case

stud

yfr

omhe

adw

ater

sto

foot

hills

.The

sein

clud

ebi

ophy

sica

lgra

dien

ts(e

leva

tion,

rive

rsi

ze,

wat

erqu

ality

,and

soon

)an

dso

cioe

cono

mic

grad

ient

sin

hous

ehol

din

com

ean

dpr

oduc

tion

syst

ems

ason

em

oves

from

high

land

sto

low

land

s.2.

Spat

iala

xis

Con

nect

ivity

Hum

anco

mm

uniti

esin

big

rive

rba

sins

are

conn

ecte

dby

the

wat

erth

atflo

ws

thro

ugh

them

and

byne

twor

ksof

road

san

dw

ater

way

s.D

owns

trea

mec

osys

tem

sde

pend

onup

stre

amw

ater

prod

uctio

n.So

cial

netw

orks

prov

ide

anad

ditio

nalk

ind

ofsp

atia

lcon

nect

ivity

.



Subs

idie

sM

any

larg

edo

wns

trea

mto

wns

are

toso

me

exte

ntsu

bsid

ised

byth

eop

port

unity

cost

sin

curr

edby

upst

ream

com

mun

ities

(for

exam

ple,

salin

ityou

tput

sfr

omir

riga

ted

agri

cultu

rear

eca

pped

inth

eG

oulb

urn-

Bro

ken

catc

hmen

tof

the

Mur

ray-

Dar

ling

syst

emin

Aus

tral

ia,i

npa

rtto

keep

wat

erdr

inka

ble

for

Ade

laid

e(A

nder

ies

2004

).Si

mila

rly,

the

inha

bita

nts

ofa

rive

rba

sin

may

prod

uce

orim

port

thei

rfo

od;d

epen

denc

eon

exte

rnal

sour

ces

can

alte

rsy

stem

resi

lienc

e.3.

Tem

pora

laxi

sSl

owva

riab

les

Gro

undw

ater

extr

actio

n,so

ilfe

rtili

ty,h

uman

popu

latio

nin

crea

se,c

hang

esin

attit

udes

oflo

calc

omm

unity

.E

volu

tion/

sele

ctio

nA

sth

esy

stem

chan

ges

thro

ugh

time,

elem

ents

(suc

has

farm

ing

syst

ems

oror

gani

zatio

ns)

are

lost

and

gain

ed.

Path

depe

nden

ceIn

som

eca

sest

udie

sth

ero

leof

hist

ory

isan

impo

rtan

tdet

erm

inan

tof

curr

entp

atte

rns

insu

chth

ings

aste

nure

,lan

dus

e,so

ilfe

rtili

tyan

deq

uity

.Ph

ase

ofad

aptiv

ecy

cle

Isth

esy

stem

incr

easi

ngin

one

orm

ore

capi

tals

,rea

chin

ga

ceili

ng,i

na

stat

eof

colla

pse,

and/

orju

stem

ergi

ngfr

oma

colla

pse?

Thi

slis

tis

noti

nten

ded

tobe

exha

ustiv

e;its

focu

sis

onid

entif

ying

the

mos

tim

port

ante

lem

ents

that

are

pres

enti

na

larg

eca

tchm

ent.

Thi

sta

ble

prov

ides

both

asi

mpl

ified

inve

ntor

yfo

rth

inki

ngab

outa

nyin

divi

dual

case

stud

yan

da

gene

ralis

edin

vent

ory

for

thin

king

abou

tbet

wee

n-ca

seco

mpa

riso

ns.I

nth

isch

apte

rit

repr

esen

ted

the

star

ting

poin

tfro

mw

hich

tose

lect

elem

ents

toin

clud

ein

asi

mpl

esy

stem

sm

odel

.


70 G.S. Cumming

can become a slow variable that ultimately limits the faster dynamics of crop planting andharvesting, with significant consequences for human wellbeing and social dynamics. In thesocial realm, a similar problem exists in regard to climate change; the supposedly “fast”political and social processes that should be reducing “slow” increases in carbon dioxideemissions to the atmosphere are currently operating at too slow a rate to prevent globalclimate change.

Achieving a working system description, with a focus on the most important dynam-ics, is the first step towards assessing system resilience. It is not possible to thinkconstructively about system dynamics without an understanding of what constitutes thesystem and its primary elements. The next step towards assessing resilience is to pulltogether selectively the pieces of the system description into a more unified whole,using models. It is often particularly useful to think through some of the ways that aparticular kind of system may respond to specified perturbations. This may be doneformally, with quantitative models, or in a more qualitative manner as I have adoptedhere.

Resilience of big river basins

One entry point for thinking about the resilience of big river basins in this case studyset is to take the primary elements identified in the case study description and use someof them to construct a simple generic system model (that is, one that captures impor-tant elements that are common to nearly all of the case studies). In order to build ageneric system model we first need to have a clear idea of what the purpose of themodel is; that is, to clarify in our own minds the aim of the analytical exercise and thekinds of outcome that we will consider important. As Carpenter et al. (2001) point out,resilience is contextual; in applying these concepts it is always necessary to ask “resilienceof what to what?” To assist in determining a suitable focus for the analysis, I askedeach case study team to rank a set of potential focal problems on a scale from one tofive. I summarized the data in two different ways (Figure 1). First, I designated prob-lems that were rated over 2.5 as being “important” for a given case study to obtain achart that shows the number of catchments for which a given issue is important. I alsocalculated the total number of “votes” that each issue received by summing the scoresgiven to each. The second plot gives an indication of the relative importance of eachissue.

According to this simple survey, the most important concerns that this set of big riverbasins shares are water quantity, the demands and consequences of irrigated agriculture andpoverty. Institutional issues, population growth, and land-cover change also appear to beimportant in most of the case study catchments. For the subsequent analysis I have thereforefocused on the resilience of these big river basin SESs to changes in water quantity asdriven by irrigation, land-cover change and climate change. Of course, these issues arestrongly interrelated. Questions of institutional arrangements and equity can be seen bothas important influences on the primary drivers and as important outcomes from changes inthem.

Table 2 suggests a common set of system elements that occur in most of the case stud-ies. Big river basins share a common set of biophysical elements, such as headwaters,tributaries, river deltas and reliance on rainfall. The human elements of individual systemsshow more variation, although in many cases these are variations along a relatively smallset of common themes (Table 3).



(a)

(b)

Figure 1. (a) Salience of issues; (b) The total score for each issue, as obtained by the survey.Note: (a) The number of catchments, from a total of nine, for which a given issue was considered“important”. Water quantity was the only single issue that was considered important in all basins;(b) Since issues were rated from one to five across nine catchments, the maximum possible scorewas 45. Note the importance of irrigated agriculture in big river basins. LCC, land cover change.

Table 3 suggests some fundamental differences between different basins. In particular,there is considerable variation between big river basins in the degree to which agricultureis large-scale and commercial versus small-scale and subsistence-oriented. For example,water use in the Limpopo Basin is dominated by mining companies and commercialirrigation farmers, with small-scale farmers constituting a relatively marginalized voice;whereas in the Indus–Ganges basins, the number of small-scale cultivators is enormous and


72 G.S. Cumming

Tabl

e3.

Impo

rtan

thum

ansy

stem

elem

ents

inea

chof

the

eigh

tBFP

big

rive

rba

sin

case

stud

ies.

Bas

inL

ocal

Nat

iona

lIn

tern

atio

nal

Lim

popo

,So

uthe

rnA

fric

a

Min

ing

and

larg

e-sc

ale

agri

cultu

re;R

ural

poor

,oft

ensm

all-

scal

efa

rmer

s;N

GO

ssu

chas

Wor

ldV

isio

n,W

orki

ngfo

rW

ater

,and

the

Kal

ahar

iCon

serv

atio

nSo

ciet

y.

Nat

iona

lgov

ernm

ents

ofB

otsw

ana,

Zim

babw

e,So

uth

Afr

ica,

and

Moz

ambi

que.

Dif

fere

ntm

inis

trie

s,of

ten

plan

ning

inis

olat

ion

from

one

anot

her.

DW

AF

(Dep

artm

ento

fW

ater

Aff

airs

,Sou

thA

fric

a)A

RA

-SuL

(Adm

inis

traç

ãoR

egio

nal

deÁ

quas

doSu

l,M

ozam

biqu

e)Z

INW

A(Z

imba

bwe

Nat

iona

lWat

erA

utho

rity

)B

otsw

ana

Wat

erA

ffai

rs

LIM

CO

M(L

impo

poB

asin

Com

mis

sion

),an

emer

ging

brid

ging

orga

niza

tion

Kar

keh,

Iran

Prov

inci

algo

vern

men

ts,f

arm

ers;

pove

rty-

alle

viat

ion

orga

niza

tions

such

asth

eR

edC

resc

entS

ocie

tyan

dth

eE

mam

Kho

mei

niR

elie

fC

omm

ittee

.

Pow

erm

inis

try

offic

es,w

hich

dete

rmin

ew

ater

allo

catio

nsM

inis

try

ofE

nerg

y,M

inis

try

ofJi

had-

e-A

gric

ultu

rean

dth

eM

inis

try

ofJi

had-

e-Sa

zand

agi(

wat

ersh

edm

anag

emen

tan

dru

rald

evel

opm

ent)

Not

dire

ctly

rele

vant

sinc

eba

sin

isin

tern

alto

Iran

.

Nile

,Nor

than

dE

ast

Afr

ica

Com

mer

cial

agri

cultu

re,

smal

l-sc

ale

farm

ers,

fishe

ries

.N

atio

nalg

over

nmen

tsan

dm

inis

trie

sof

bord

erin

gna

tions

.T

heN

ileB

asin

Initi

ativ

e(N

BI)

isth

epr

imar

yor

gani

zatio

nin

the

basi

nth

atw

orks

tow

ards

confi

denc

e,tr

usta

ndca

paci

tybu

ildin

gam

ong

the

ripa

rian

coun

trie

s.T

heul

timat

ego

alof

the

NB

Iis

tofa

cilit

ate

the

form

atio

nof

ari

ver

basi

nco

mm

issi

onan

den

ablin

gen

viro

nmen

tfor

inte

grat

edw

ater

reso

urce

sm

anag

emen

t.T

heE

aste

rnN

ileTe

chni

cala

ndR

egio

nal

Org

aniz

atio

n(E

NT

RO

)is

asu

bsid

iary

actio

npr

ogra

mm

eof

the

NB

Ith

atw

orks

onth

ejo

int

mul

ti-pu

rpos

ein

vest

men

tpro

ject

sam

ong

Egy

pt,

Suda

nan

dE

thio

pia.

Asi

mila

rsu

bsid

iary

actio

npr

ogra

mm

eex

ists

for

the

Nile

Equ

ator

ialL

akes

(NE

LSA

P)re

gion

.The

rear

eal

sole

gali

nstit

utio

nsin

the

Nile

Equ

ator

ialL

akes

regi

on,s

uch

asth

eL

ake

Vic

tori

aB

asin

Com

mis

sion

for

the

Eas

tA

fric

anC

omm

unity

.



Vol

ta,W

est

Afr

ica

Agr

icul

ture

,alm

oste

ntir

ely

smal

l-sc

ale

farm

ers

Gov

ernm

ents

ofB

urki

naFa

soan

dG

hana

,in

clud

ing

rele

vant

min

istr

ies.

The

Vol

taR

iver

Aut

hori

ty,w

hich

isin

char

geof

prod

ucin

gel

ectr

icity

.

The

Vol

taB

ASI

NA

utho

rity

,am

ultin

atio

nalb

ridg

ing

orga

niza

tion.

Nig

er,W

est

Afr

ica

Agr

icul

ture

,alm

oste

ntir

ely

smal

l-sc

ale

farm

ers;

nom

adic

herd

ers;

Offi

cedu

Nig

erin

Mal

i;Su

b-ba

sin

agen

cies

(Ban

i,L

ipta

koG

ourm

a);N

GO

s;T

radi

tiona

lgov

erna

nce

and

farm

ing

syst

ems,

incl

udin

gfa

dam

a/ir

riga

tion

proj

ects

inN

orth

ern

Nig

eria

.The

Inne

rD

elta

inM

alii

san

impo

rtan

tR

amsa

rsi

te(3

Mha

)an

dec

olog

ical

/env

iron

men

tal

inte

rest

sin

cons

ervi

ngth

isre

sour

cear

eal

soim

port

ant.

Nat

iona

lmin

istr

ies

ofw

ater

and

agri

cultu

re.

EC

OW

AS

(Wes

tAfr

ican

EU

)at

the

regi

onal

leve

l.N

iger

Bas

inA

utho

rity

(whi

chcu

rren

tlyse

ems

unab

leto

fulfi

lits

diffi

cult

role

).

Yello

wR

iver

,C

hina

Irri

gate

dag

ricu

lture

,Ind

ustr

y,ru

ralp

oor;

agri

cultu

ral

rese

arch

cent

res,

stat

ions

and

polic

y.Pr

ovin

cial

-lev

elgo

vern

men

tsan

dw

ater

reso

urce

bure

aux

(pro

vinc

esco

ntro

ltri

buta

ryflo

ws

with

inth

eir

boun

dari

es).

Som

een

viro

nmen

talN

GO

sar

eha

ving

grow

ing

impa

cts.

Wat

er-u

seri

ghts

tran

sfer

pilo

tpr

ojec

tsto

cons

erve

wat

erw

ithou

tadv

erse

impa

cts

onir

riga

tion

bene

fits,

seen

bym

any

asa

way

toad

dres

sth

ew

ater

cris

is.

Min

istr

yof

Wat

erR

esou

rces

Min

istr

yof

Agr

icul

ture

Min

istr

yof

Env

iron

men

talP

rote

ctio

nYe

llow

Riv

erC

onse

rvan

cyC

omm

issi

on(Y

RC

C),

unde

rth

eC

hine

seM

inis

try

ofW

ater

Res

ourc

es,w

hich

has

the

man

date

for

allo

catin

gflo

ws

inth

edo

wns

trea

mpa

rtof

the

basi

n.T

heYe

llow

Riv

erW

ater

Allo

catio

nSc

hem

e,es

tabl

ishe

dan

dis

sued

byth

eSt

ate

Cou

ncil

ofC

hina

in19

87,s

eta

cap

onab

stra

ctio

nat

37km

3pe

rye

aran

dqu

otas

for

each

prov

ince

tied

toav

erag

eru

noff

of58

km3.

2002

Wat

erL

aw(f

ocus

onri

ver

basi

nm

anag

emen

t,w

ater

savi

ngs,

and

man

yot

her

topi

cs)—

high

est-

leve

lleg

isla

tion,

but

Not

dire

ctly

rele

vant

.

(Con

tinue

d)


74 G.S. Cumming

Tabl

e3.

(Con

tinue

d)

Bas

inL

ocal

Nat

iona

lIn

tern

atio

nal

Reg

ulat

ions

onflo

wan

dw

ater

qual

ityat

cont

rolli

ngpo

ints

alon

gth

eri

ver.

beco

mes

only

effe

ctiv

eif

supp

ortin

gre

gula

tions

are

both

deve

lope

dan

dim

plem

ente

d.In

dus-

Gan

ges

Bas

ins,

Indi

aan

dPa

kist

an

Farm

ers’

orga

niza

tions

–fo

rmal

and

info

rmal

–w

hopu

shto

getw

ater

serv

ices

atno

min

alor

free

cost

s.T

hese

orga

niza

tions

also

impa

ctth

een

ergy

polic

ies

ofth

est

ate

for

keep

ing

the

ener

gyta

riff

slo

wor

zero

(for

grou

ndw

ater

pum

ping

).N

on-g

over

nmen

tal/

civi

lsoc

iety

orga

niza

tions

/do

nors

/ot

her

voic

egr

oups

who

clam

our

tost

rike

aba

lanc

ebe

twee

nde

velo

pmen

tand

envi

ronm

ent,

incl

usiv

enes

sof

the

poor

and

mar

gina

lized

and

long

-ter

msu

stai

nabi

lity.

Indu

stri

esan

ddo

mes

ticsu

pplie

sar

ere

gula

ted

toso

me

exte

nt,

buti

rrig

atio

nsu

pplie

sre

mai

nla

rgel

yun

regu

late

d.O

neof

the

mai

nre

ason

sis

the

shee

rnu

mbe

rof

smal

lcul

tivat

ors,

whi

chde

fies

any

gove

rnan

cem

echa

nism

.

Fede

ralg

over

nmen

ts/

min

istr

ies

ofth

ew

ater

reso

urce

sof

the

four

maj

orri

pari

anco

untr

ies:

Paki

stan

,Ind

ia,N

epal

,B

angl

ades

h.St

ate/

prov

inci

algo

vern

men

tsan

dir

riga

tion

bure

aucr

acie

sin

each

ofth

est

ates

inth

eco

untr

y(s

ince

wat

eris

cons

ider

eda

stat

esu

bjec

t,th

ere

ispe

rpet

ualc

onfli

ctbe

twee

nth

efe

dera

land

the

stat

ego

vern

men

ts).

Inst

itutio

nalb

asis

unde

rw

hich

wat

erre

sour

ces

are

notp

ublic

/nat

iona

l;an

ype

rson

havi

ngla

ndri

ghts

also

has

the

righ

tove

rw

ater

reso

urce

sas

anad

junc

tto

the

land

.

Lac

king

inte

rnat

iona

lbri

dgin

gor

gani

zatio

ns.



Mek

ong,

Asi

aL

ocal

inha

bita

nts,

espe

cial

lyfa

rmer

san

dfis

hers

.G

over

nmen

tsof

the

ripa

rian

coun

trie

s:C

hina

,M

yanm

ar,L

aoPD

R,T

haila

nd,C

ambo

dia,

Vie

tnam

.R

elev

antm

inis

trie

san

dde

part

men

tsin

each

coun

try

(inc

ludi

ngag

ricu

lture

,fish

erie

s,fo

rest

ry,w

ater

reso

urce

san

dhy

drop

ower

).N

atio

nalM

ekon

gC

omm

ittee

s.

Mek

ong

Riv

erC

omm

issi

on(M

RC

);re

latio

nshi

pbe

twee

nM

RC

and

Chi

naan

dM

yanm

ar(t

wo

upst

ream

coun

trie

s),p

artic

ular

lyre

latio

nshi

pbe

twee

nC

hina

and

MR

C,a

sm

osto

fth

ede

velo

pmen

tssu

chas

cons

truc

tion

ofda

ms

are

inth

eup

per

part

ofth

eba

sin

inC

hina

;bila

tera

lre

latio

nshi

pan

din

tera

ctio

nsbe

twee

nth

eM

RC

coun

trie

ssu

chC

ambo

dia–

Tha

iland

and

Cam

bodi

a–V

ietn

amre

latio

nshi

p.A

ndes

,So

uth

Am

eric

a

Farm

ers,

com

mun

ities

,in

dust

rial

ists

,dam

oper

ator

sfo

rhy

droe

lect

ric

pow

eran

dir

riga

tion,

min

ing

com

pani

es,

urba

nw

ater

com

pani

es.

Nat

iona

lgov

ernm

ents

and

min

istr

ies;

regi

onal

and

natio

nalp

olic

yin

stitu

tes.

Inte

rnat

iona

lenv

iron

men

t,ag

ricu

lture

and

deve

lopm

entd

onor

s,in

tern

atio

nalc

onse

rvat

ion

NG

Os

Tabl

e2

prov

ides

age

neri

clis

ting

ofca

sest

udy

elem

ents

;thi

sta

ble

prov

ides

spec

ific

deta

ilsof

the

hum

anel

emen

tsco

nsid

ered

asim

port

antb

ypa

rtic

ipat

ing

case

stud

yte

ams.

Iha

vebr

oken

them

dow

nac

cord

ing

toth

eir

scal

eof

actio

n.N

ote

that

this

listi

sin

tend

edto

prov

ide

supp

ortin

gin

form

atio

n,ra

ther

than

anex

haus

tive

inve

ntor

y.


76 G.S. Cumming

although some powerful organizations exist, there is no single channel through which theycan all be negotiated with. This makes small-scale farmers difficult to include in water-usedebates and equally hard to govern.

Some kinds of system element are particularly relevant to big river basins. A num-ber of big river basins in this set of case studies have some form of bridging organizationthat attempts to coordinate and integrate management across different sectors and betweenupstream and downstream water users. Examples from the case studies in the BFPs includethe Limpopo Basin Commission (LIMCOM); the Mekong River Commission (MRC); theNile Basin Initiative (NBI); the Volta BASIN authority; and the Yellow River ConservancyCommission (YRCC). The commonness of bridging organizations in big river basins, andthe importance accorded to such organizations in the resilience literature (Hahn et al. 2006,Olsson et al. 2004, 2007), suggests that basins without bridging organizations (or wherethey exist but are dysfunctional) may be less capable of responding effectively to environ-mental change, and hence less resilient. Bridging organizations have also been proposed byErnstson et al. (in press) as potentially important contributors to the resolution of scale mis-match situations, which occur when the scale of management is not appropriately alignedto the scale at which ecological and social processes occur (Cumming et al. 2006).

Drawing on the summaries in Tables 2 and 3 and the issues considered to be of centralimportance by case study teams, a simplified generic model of a big river basin case studymight look something like Figure 2.

As Tables 2 and 3 and Figure 2 suggest, big river basins have the potential for manydifferent kinds of feedback. To make Figure 2 a little more specific, consider a downstreamsystem that is in an economic growth phase and currently experiencing a beneficial climate(Figure 3).

Tracing some possible feedback through Figures 2 and 3, for example, considering whathappens if rainfall is reduced or if agricultural demands for water increase, suggests thatthe potential for both self-regulating and self-amplifying feedbacks increases hugely whenwater is scarce. Once water becomes limiting, small changes in either water availability orproduction systems will inevitably influence actors and land cover, which in turn will influ-ence water and production systems. For example, when water demand from irrigated farmland starts to approach the limit that rainfall in an average year can support, water availabil-ity for other sectors (for example, fisheries, hydroelectric power companies, or downstreamecosystems) is reduced, depriving the associated actors of resources and leading to conflict.Similarly, excessive land clearing for agriculture (or the planting of water-hungry timberplantations in place of native shrubs) can reduce the total amount of water entering thebasin, leading to reductions in crop and livestock production, increasing poverty, additionalland clearing, and a downward spiral that includes further environmental degradation anda decrease in human wellbeing.

The point at which the demand for water approaches or exceeds the supply of water ispossibly the single most important tipping point for many of the big river basins consideredin the BFPs. Once it is exceeded, a number of initially dormant or potential interactionsbetween actors, production systems and land cover start to become increasingly impor-tant. Given that the focus here is on the resilience of big river basin SESs to changesin water quantity as driven by irrigation, land-cover change and climate change, waterquantity thresholds offer a convenient entry point for thinking about basin resilience.

Water scarcity may occur as a consequence of a number of different factors. There aremany routes to crisis and collapse. Some of the most important driving factors in the casestudies in BFPs include expansion of domestic demand (Indus–Ganges, Yellow); increas-ing water demands for irrigation or mining (common to all case study systems); excessive



Figure 2. A generic model of a big river basin. Spatial dynamics are represented by a simpleupstream–downstream divide; in reality the spatial distinctions between river sections will ofteninclude three or more recognisably different use-zones, corresponding to the headwaters, mid-sections and river delta, and some elements of the system may differ between zones. Upstream anddownstream zones are connected by water, often with a delay via impoundments. Bridging organi-zations connect actors across different zones and/or at different scales. In each zone, local actorsand production systems use and are influenced by water, and a variety of important interactionsand feedbacks occur within both of these boxes. Climate (for example, via snowmelt or evapora-tion) determines water quantity, while also influencing land cover; and feedbacks from land cover(for example, vegetation via evapotranspiration, urban heat islands, or other albedo-altering effects)influence local rainfall. Production systems include different sectors (for example agriculture, min-ing, power generation and household users) and may directly influence land cover. Although actorsmay of course influence land cover directly for non-productive uses, I have chosen to focus on theinteraction that has most impact (that is, LCC caused by production systems). Production systemsrespond to and influence water quantity and quality by such activities as extraction, alterations torunoff, and pollution. Land cover influences infiltration and runoff rates, with effects on water quan-tity and quality. Actors generate information, which may influence water use. Lastly, institutionalinterventions or guidelines, denoted in this diagram by “I”, can influence some but not all interactionswithin the system.

groundwater extraction (Karkheh, Indus–Ganges); climate change (a pending concern inall case studies); and more insidiously, reductions of water quality to the point wherewater becomes virtually unusable by one or more production systems or sectors (Yellow,Indus–Ganges).

What makes a big river basin more or less resilient to water scarcity? Once a pointof water scarcity is reached, the social system is forced to reorganize and adapt to the


78 G.S. Cumming

Figure 3. Generic system diagram. Derived from the model in Figure 2, indicating the strongestinteractions (and whether increasing or decreasing) in the downstream component during a periodwhen agriculture is expanding and rainfall is reliable In this instance, households and other actors arewithdrawing water; production systems are using water; and land-cover change is further reducing theamount of water available. Provided that there is sufficient rainfall (that is, the positive influence ofclimate on water quantity is large enough), there will be sufficient water for the net effect of water onproduction systems to remain positive. The system can remain in its current regime (and keep grow-ing) as long as the magnitudes of the interactions indicated by pluses and minuses complement oneanother sufficiently well for excess water to remain in the system. However, if the demands on waterbecome excessive (whether by direct or indirect pathways), water scarcity will limit production andcreate a crisis in the social–ecological system. The ability of the system to navigate this crisis whilekeeping essential system elements intact (for example, particular social groups, human wellbeing,ecosystems, food production systems, and/or culture and ways of life) represents its resilience.

new regime that now dominates the system. The way in which this reorganization occurs,and the management trajectory that the system subsequently follows, can be expected tohave a large influence on the overall resilience of the system. Resolution of managementproblems in many real-world situations is hindered by ineffective institutions, corruptionand incompetence within key organizations. Substantial delays can occur between crossingthe water quantity threshold and the implementation of an effective management response,particularly in systems where social networks and management institutions are weak andactors have little history of communicating with one another or coordinating their use ofwater at an appropriate scale.

In some of the case studies in the BFPS, such as the Yellow River, the point at whichwater scarcity becomes important has long since been passed. The persistence of desirablesocial and ecological elements in these systems depends on reaching a successful two-pronged solution that: (1) creates a socio-politically acceptable compromise under currentconditions; and (2) in the longer term, addresses the slow variables of reductions in waterquantity and increasing demands on resources.

In other large-basin SESs, such as the Andes, water scarcity looms as a possible futureshock to the system. In cases where a threshold of water scarcity has not yet been reached,it should be possible to attempt to build resilience to water scarcity into the social sys-tem, through such activities as building stronger social networks, developing a betterunderstanding of the limits on production systems, and working towards more effectiveintegrated management over the entire basin.

From a resilience-oriented perspective, crossing the threshold into water scarcity canoccur in a number of different ways. Although some of these pathways may ultimatelybe beneficial for the long-term resilience of the case study basin to climate change (for



example, if crisis leads to better integration of different management actions), enteringinto water scarcity can easily push a large-basin SES into one or more problem situationsfrom which escape can be difficult.

For example, a common management syndrome, termed “fixes that fail” (Senge 1990),occurs when a solution that is effective in the short term has unforeseen longer-term conse-quences. This is very similar to the typical complex systems dynamic in which changes ina slow variable (often one which is considered unimportant by decision makers) grad-ually undermine attempts to manage a faster variable (Carpenter and Turner 2000). Abig river basin of the kind discussed here can become trapped in this dynamic throughattempts to alleviate poverty. Poor households are often perceived as being trapped by cir-cumstances that make it difficult to raise the starter capital that is needed to create themeans to provide a steady income (Yunnus 1998). Being able to afford a vehicle withwhich to transport produce to market, for instance, can make a huge difference to a farmer’searning potential. One plausible solution to getting households out of the poverty trap isfor actors in the system, such as governmental agencies and non-governmental organiza-tions (NGOs), to provide arable land, technical support and training to would-be farmers.The logic of this solution appears sound: that financially poor households will then beable to both feed themselves and sell their surplus, gradually lifting themselves out ofpoverty and ultimately becoming self-sustaining. In practice such programmes are oftenaccompanied by, or part of, resettlement schemes such as those in the Amazon Basin(Imbernon 1999).

If the human population is large, however, fostering additional farming activities with-out appropriate zoning and enforcement of good farming practices can result in widespreadclearing of native vegetation and increasing demand for water (Figure 4). This in turnsets in place a series of slower environmental changes, such as reductions in infiltration

Figure 4. An example of how attempts to reduce poverty in a complex social–ecological systemcan have unintended side-effects. Gains in poverty alleviation achieved by increased food productionmay be offset by declines in human wellbeing resulting from worsening water quality and quantity.This figure also ignores the potential human population increase that may follow from increased foodproduction. In the worst-case scenario, poorly conceived poverty reduction programmes may actuallyworsen the problem that they are attempting to solve.


80 G.S. Cumming

rates, increased erosion of topsoil, and increases in pesticide and fertilizer use. Thesechanges can gradually accumulate until a point is reached at which agricultural productionbecomes heavily dependent on external inputs. In the worst cases, water quality and quan-tity are reduced, water storage is difficult because of high sedimentation rates, landslidesand floods become commoner, and soil fertility is considerably lower. The net outcome ofnaïve policies aimed at agricultural expansion can thus be to leave poor households worseoff than before, while creating significant environmental problems that may take decadesto fix.

At some point this kind of dynamic will also create difficulties for the agencies con-cerned, requiring a complete rethinking of current policies and an attempt to push thesystem on to a new trajectory. Whether or not a feasible solution can be achieved is pathdependent, in the sense that it will depend heavily on the degree to which other optionsremain open within the system. If the number of people who have been allocated farm-land in the basin is too high for the river basin to cope with, the entire river basin SESmay enter a long-lasting poverty trap. Escaping this predicament requires altering the cur-rent regime, focusing policies on different goals, and possibly transforming the systemby reducing the resilience of some attributes and enhancing the resilience of others. Forexample, introducing policies that focus on urban job creation may attract people intotowns, reducing the pressure on soil and water resources. On the island of Puerto Rico,for instance, expansion of the pharmaceutical industry following tax reductions ultimatelyled to a forest cover increase from 4% of the island’s area to over 40%, with corre-sponding improvements in water quality and quantity (Gonzalez 2001, Lugo and Helmer2004).

Agricultural activities can create situations in which different biophysical regimes existin at least three different areas: soils, hydrology and the atmosphere (Gordon et al. 2008).Gordon et al. (2008) review 10 different agriculturally linked regimes that can lead toalternate stable states in big river basins. Shifts between some of these regimes can beextremely difficult to reverse. For example, once soils are high in nutrients, they can con-tinue to release reactive nitrogen for a long period, even in the absence of further fertilizerapplication.

Pro-active management in big river basins with substantial agricultural demands willaim to develop and enforce policies that recognize the longer-term and slower variablesin the system, particularly those on which farming activities depend. This level of pol-icy development requires both a holistic perspective on what constitutes the system (forexample, appreciating that water quality and quantity are closely linked to ecosystems)and the generation and availability of sufficient high-quality information on which to basedecisions.

It is important to recognize that each case study will be resilient to water scarcity insome ways and not in others; the aim of cross-case resilience analysis is not to label onecase study resilient and another vulnerable, but rather to draw out and explore potentialshared pitfalls and traps, while attempting to identify and test general guiding principlesthat advance the theory and practice of social-ecological research.

I next examine some more detailed examples from different case studies from theperspective of the preceding analysis.

Case study resilience

A first basic separation of the nine case studies can be made on the basis of water quantity.In two catchments, the Yellow River and the western half of the Indus–Ganges, the SES is



well over the water scarcity threshold. The Yellow River in particular has run dry on severaloccasions, with a set of resulting problems. The Limpopo, the Karkheh and the Nile arerelatively close to the water scarcity tipping point but are not necessarily there yet, exceptperhaps in drought years. The Niger, the Volta, the Mekong and the Andes are systems inwhich water scarcity is not currently considered a major issue, although it may of coursebecome so in the future.

Although water is the first and most important identity criterion for a big river basin,other important elements (see Tables 2 and 3) typically include long-term residents of thecatchment; commercial, large-scale players who contribute to national food security andeconomic growth; unique ecosystems, such as mangroves and wetlands; and a range ofecosystem services that are integral to the wellbeing of people living in the basin. If theseelements are lost, the system can be said to have not been resilient.

Each basin has one or more “burning issues” that may eventually lead to the sys-tem entering crisis mode (if it is not there already) and ultimately to the loss of one ormore important aspects of system identity. These issues are thus the areas in which basinresilience can be considered weakest. They are summarized for each of the eight casestudies in Table 4.

As Table 4 shows, there is a wide range of complex dynamics in progress in eachof the case study systems. The most fundamental common problem is probably the roleof increasing demand for water from large-scale users, who include industry, irrigationfarmers and cities. As demand increases, these systems must develop ways of reducingenvironmental impacts, allocating water equitably, and leaving some leeway in the sys-tem for periods of reduced flow. In fundamental terms, net population expansion mustbe slowed and water use efficiencies in industry and agriculture will need to be built inat an early stage. Once a large basin system has crossed the water scarcity threshold, withnumerous water-demanding actors already in place, corrective action becomes increasinglydifficult.

One of the most important concerns from a management perspective is that attemptsto manipulate the system should not create further problems. As discussed previ-ously, one of the greatest dangers is that naïve policies and unfettered, unregulateddevelopment may ultimately make worse the very problems (poverty, food production,slow economic growth) that they are trying to resolve. People have a tendency to seesolutions through their own expert lenses. For example, development agencies oftensee the root problems in big river basins as institutional; scientists may see them asbeing driven by water flows, deteriorating soil quality, or other biophysical compo-nents of the system; and demographers may blame population growth. In many casesdisciplinary experts assume that implementing “fixes” in their own area of expertisewill solve the problems. Economists demand better pricing and incentive structures;development agencies want greater institutional accountability; and ecologists wantrestored riparian buffers and ecological flows. In each of these cases, remedial actionsoffer only partial solutions. While each may have some desirable outcomes, none ontheir own is adequate to solve the broader problem. Long-term sustainability in nearlyevery case will require the implementation of a set of carefully negotiated tradeoffsbetween different actors in the system, including not only human needs but also minimalflows that will keep relevant ecosystems functioning and diverse (see Zwarts et al.[2006] for example). Diversity in both the human and the ecological realm offers animportant buffer against future change (Yachi and Loreau 1999, Norberg et al. 2008);


82 G.S. CummingTa

ble

4.B

urni

ngis

sues

inei

ghtb

igri

ver

basi

ns.

Que

stio

nB

urni

ngis

sues

inth

eba

sin

Lim

popo

,Sou

ther

nA

fric

aN

atur

alre

sour

ce-b

ased

econ

omic

deve

lopm

entp

lans

have

been

scal

edup

inea

chof

the

four

ripa

rian

stat

es.I

nfra

stru

ctur

ede

velo

pmen

thas

lagg

edbe

hind

,and

prev

ious

inve

stm

ents

are

noto

ptim

ized

.Lim

popo

Bas

inst

akeh

olde

rsin

clud

ela

rge

grou

psof

rura

lpoo

rw

hore

lyon

natu

ralr

esou

rce-

base

dliv

elih

oods

for

surv

ival

.Maj

orth

reat

sin

clud

e:(1

)A

nal

read

yov

erco

mm

itted

rive

rba

sin

whe

rene

ww

ater

reso

urce

sde

velo

pmen

tis

bein

gpl

anne

dan

dim

plem

ente

d.T

his

incr

ease

spr

essu

reon

ecol

ogic

ally

sens

itive

area

s,su

chas

wet

land

ecos

yste

ms;

(2)

Com

petin

gus

esin

clud

ing

extr

emel

yhi

gh-v

alue

com

mod

ities

(pla

tinum

and

othe

rs);

larg

e-sc

ale

com

mer

cial

irri

gate

dag

ricu

lture

;poo

rly

reso

urce

dsm

allh

olde

rag

ricu

lture

;tou

rism

—m

uch

ofit

inar

eas

with

unse

cure

dla

ndan

dw

ater

righ

ts;(

3)T

helik

elih

ood

that

clim

ate

chan

gew

illfu

rthe

rre

duce

avai

labl

ew

ater

and

incr

ease

alre

ady

unpr

edic

tabl

ew

ater

even

ts(fl

oods

,dro

ught

,rai

nyse

ason

s—on

set,

dura

tion,

accu

mul

atio

n,an

dso

on);

(4)

Lim

ited

inst

itutio

nalf

unct

ion

inw

ater

gove

rnan

cede

spite

num

erou

sst

ruct

ures

and

polic

ies

tow

ard

that

end.

Kar

keh,

Iran

The

reis

grow

ing

com

petit

ion

for

wat

eram

ong

diff

eren

tsec

tors

inth

eba

sin,

whi

chis

alre

ady

clos

ed.I

ncre

asin

gde

man

dfo

rw

ater

inin

dust

rial

and

dom

estic

sect

ors

will

redu

cew

ater

avai

labl

efo

rag

ricu

lture

and

the

envi

ronm

ent.

Att

hesa

me

time,

food

dem

and

isri

sing

due

topo

pula

tion

grow

th.I

nad

ditio

n,th

ecu

rren

trat

eof

grou

ndw

ater

extr

actio

nin

the

basi

nis

muc

hm

ore

than

the

aqui

fer’s

capa

city

.As

are

sult,

the

grou

ndw

ater

tabl

eis

drop

ping

byab

outo

nem

etre

ever

yye

ar.L

astly

,exi

sten

ceof

the

Ram

sar

“Hoo

r-al

Azi

m”

swam

pis

inda

nger

,par

tlydu

eto

low

inflo

wfr

omK

arkh

ehR

iver

toth

ew

etla

nd.

Nile

,Nor

than

dE

ast

Afr

ica

The

Nile

Bas

insu

pplie

sw

ater

toab

outfi

vem

illio

nha

ofir

riga

ted

agri

cultu

re.T

hepr

esen

tirr

igat

ion

dem

and

cons

umes

abou

ttw

o-th

irds

ofth

eN

ileR

iver

flow

.Pot

entia

lfut

ure

prob

lem

sin

clud

eex

pans

ion

ofir

riga

tion,

clim

ate

chan

gean

dun

ilate

ral

appr

oach

esto

basi

nde

velo

pmen

tand

man

agem

ent.

The

irri

gatio

nde

man

dis

expe

cted

todo

uble

inth

ene

arfu

ture

tofe

edth

era

pidl

ygr

owin

gpo

pula

tion

inth

eba

sin.

The

Nile

Bas

inis

notm

anag

edin

anin

tegr

ated

and

sust

aina

ble

man

ner;

rath

erth

eri

pari

anco

untr

ies

are

inde

pend

ently

deve

lopi

ngan

dm

anag

ing

thei

row

npa

rts

ofth

eba

sin,

crea

ting

am

ajor

chal

leng

efo

ref

ficie

ntan

dsu

stai

nabl

eut

iliza

tion

ofth

eN

ileB

asin

wat

erre

sour

ces.

Vol

ta,W

estA

fric

aT

heba

sin

cont

ains

som

eof

the

wor

ld’s

poor

estc

ount

ries

,and

mos

tare

getti

ngpo

orer

.Inc

reas

esin

food

prod

uctio

nar

ew

ell

behi

ndde

mog

raph

icin

crea

se,m

ainl

yas

are

sult

ofpo

orso

ils,l

ack

ofin

puts

and

prod

uctio

nm

eans

(oxe

nan

dto

ols)

,bad

acce

ssto

mar

kets

,poo

rin

fras

truc

ture

,lac

kof

educ

atio

nan

dhe

alth

care

,and

soon

.Ins

titut

ions

are

mid

way

betw

een

trad

ition

alhi

erar

chie

san

dm

oder

nst

ate

law

san

dhe

nce

are

freq

uent

lydy

sfun

ctio

nal.

Nig

er,W

estA

fric

aT

hem

ajor

thre

ats

inth

eN

iger

Bas

inre

volv

ear

ound

envi

ronm

enta

lvar

iabi

lity

and

inst

itutio

nalw

eakn

esse

s.T

hepo

pula

tion

suff

ers

from

high

ambi

entp

over

tyle

vels

(eco

nom

y,he

alth

,edu

catio

n,in

fras

truc

ture

,and

soon

)an

dis

affe

cted

byre

gula

rdi

sast

ers

(dro

ught

,floo

ding

,and

chol

era

outb

reak

sto

nam

ea

few

).T

heba

sin

also

suff

ers

from

inse

curi

tyis

sues

(eth

nic

tens

ions

inth

eN

orth

with

the

Toua

reg

inN

iger

and

Mal

i,an

dre

ligio

uste

nsio

nsin

nort

hern

Nig

eria

and

wid

espr

ead

inse

curi

tyin

sout

hern

Nig

eria

),as

wel

las

from

corr

uptio

nan

din

effic

ient

gove

rnan

ce.A

dditi

onal

pres

sure

ism

ount

ing

due

toex

trem

epo

pula

tion

grow

th,c

limat

ech

ange

,lan

dap

prop

riat

ion,

risi

ngfu

elan

dfo

odpr

ices

,and

unco

ntro

lled

deve

lopm

ent(

land

-use

chan

ges,

pollu

tion,

dam

deve

lopm

enta

ndso

on).

The

lack

ofes

tabl

ishe

dan

def

fect

ive

inst

itutio

nsal

soha

sa

detr

imen

tali

mpa

cton

inte

grat

edw

ater

reso

urce

man

agem

enta

tthe

basi

nsc

ale,

asse

vera

ldam

deve

lopm

ents

are

plan

ned,

whi

char

elik

ely

toin

crea

seth

est

rain

onbl

uew

ater

reso

urce

san

ddo

wns

trea

m(o

ften

tran

sbou

ndar

y)us

ers.

Agr

icul

ture

ispr

edom

inan

tlysu

bsis

tenc

eag

ricu

lture

and

rain

fed,

expo

sing

itto

vari

atio

nsin

clim

ate.

Yie

lds

are

low

and

curr

ently

rely

onla

ndex

tens

ion

(rat

her

than

yiel

din

crea

ses)

tofe

eda

grow

ing

popu

latio

n.



Yello

wR

iver

,Chi

naT

heYe

llow

Riv

erba

sin

isfa

ced

with

abso

lute

wat

ersc

arci

tyan

dpo

orw

ater

qual

ityan

dlit

tleim

prov

emen

tis

expe

cted

for

the

futu

re.T

heba

sin

face

sse

vere

com

petit

ion

betw

een

irri

gatio

nw

ater

uses

mid

-str

eam

and

part

ially

dow

nstr

eam

and

rapi

dur

ban-

indu

stri

alde

velo

pmen

tdow

nstr

eam

.No

addi

tiona

lirr

igat

ion

expa

nsio

nis

plan

ned,

butt

hego

vern

men

tpol

icy

offo

od-s

elf

suffi

cien

cyin

key

grai

nsw

illco

ntin

ueto

putp

ress

ure

onth

eba

sin’

sw

ater

reso

urce

s.T

heri

ver

has

been

runn

ing

dry;

noflo

ww

asre

cord

edat

the

dow

nstr

eam

stat

ion,

for

five

to22

6da

ysdu

ring

1972

–99.

Aft

er19

99,

the

Yello

wR

iver

Con

serv

ancy

Com

mis

sion

was

give

nth

eta

skto

ensu

reth

atso

me

flow

reac

hes

the

rive

rm

outh

.Des

pite

this

,th

esm

alld

owns

trea

mw

etla

ndar

eare

mai

nsth

reat

ened

from

cana

lizat

ion,

oild

evel

opm

enta

ndra

pid

urba

niza

tion.

Its

area

decl

ined

by50

%ov

erth

ela

st20

year

s.O

ther

urge

ntis

sues

incl

ude

popu

latio

ngr

owth

,fur

ther

urba

n–in

dust

rial

deve

lopm

ent,

clim

ate

chan

gean

dhi

ghse

dim

ent

depo

sitio

nra

tes.

Wat

eral

loca

tion

betw

een

regi

ons

isst

illba

sed

onan

allo

catio

npl

anis

sued

in19

87,w

hich

has

notb

een

revi

sed

tore

flect

the

chan

gein

wat

erde

man

d.W

ater

qual

ityin

the

rive

ris

very

poor

,due

toa

lack

oftr

eatm

ents

tatio

nsan

dlim

ited

enfo

rcem

ento

fex

istin

gen

viro

nmen

tal

regu

latio

nsin

the

basi

n.T

hesi

ltco

nten

tof

the

wat

eris

very

high

.New

polic

ies

ofsi

ltflu

shin

gha

vehe

lped

redu

ceth

eri

skof

dow

nstr

eam

flood

ing,

buta

rea

furt

her

drai

non

wat

erav

aila

bilit

yfo

rir

riga

tion.

Indu

s-G

ange

sB

asin

s,In

dia

and

Paki

stan

The

seba

sins

are

faci

ngtw

oen

tirel

ydi

ffer

entk

ind

ofcr

isis

.The

wes

tern

part

ofth

eG

ange

san

dth

eIn

dus

are

witn

essi

ngan

over

-exp

loita

tion

ofth

egr

ound

wat

erre

sour

ces

and

aco

ntin

uous

decl

ine

inth

ew

ater

tabl

em

akin

gth

ede

pend

enta

gric

ultu

rehy

drol

ogic

ally

/fina

ncia

llyun

sust

aina

ble.

Thi

sre

gion

isco

nsid

ered

the

glob

alho

tspo

tfor

grou

ndw

ater

over

-use

.The

east

ern

part

ofth

eG

ange

sha

sa

loto

fsu

rplu

sw

ater

butl

acks

hum

an,fi

nanc

iala

ndte

chni

calc

apita

lto

mak

ego

odus

eof

the

avai

labl

ew

ater

,kee

ping

mill

ions

ofth

epo

pula

tion

inpo

vert

yan

dlo

wpr

oduc

tivity

.Clim

ate

chan

geis

the

maj

orth

reat

toth

eba

sins

whi

chw

illse

riou

sly

alte

rth

eflo

wre

gim

ean

din

tens

ityan

dfr

eque

ncy

ofth

eex

trem

eev

ents

:floo

dsan

ddr

ough

tsan

dte

mpe

ratu

rech

ange

s.R

elat

edis

sues

incl

ude

low

prod

uctiv

ityin

ala

rge

part

ofth

eba

sin;

ahi

ghco

ncen

trat

ion

ofpo

vert

y,es

peci

ally

the

east

ern

sect

ion;

dete

rior

atio

nof

publ

icir

riga

tion

syst

ems

(can

als)

and

unsu

stai

nabi

lity

ofev

er-i

ncre

asin

ggr

ound

wat

erir

riga

tion;

and

the

tran

sbou

ndar

yna

ture

ofth

eba

sin

inth

evo

latil

eSo

uth

Asi

anco

ntex

t.

Mek

ong

Bas

in,A

sia

The

key

issu

esin

the

Mek

ong

Bas

inst

emfr

omri

sing

pres

sure

onth

ena

tura

lres

ourc

eba

se.T

hegr

owin

gpo

pula

tion

need

shy

drop

ower

and

food

.The

popu

latio

nof

the

Low

erM

ekon

gB

asin

isex

pect

edto

incr

ease

from

the

curr

ent6

5m

illio

nto

abou

t90

mill

ion

by20

50,a

ndth

epr

opor

tion

ofur

ban

dwel

lers

from

abou

t20%

toab

out4

0%or

abou

t36

mill

ion.

Eco

nom

icgr

owth

isar

ound

4.5%

pera

nnum

.Tot

alfo

odde

man

dw

illin

crea

seat

ara

tegr

eate

rtha

nth

atof

popu

latio

nal

one,

due

tori

sing

inco

mes

and

chan

ging

diet

pref

eren

ces

cons

eque

ntup

onur

bani

zatio

n.A

furt

her

thre

atto

food

supp

lies

com

esfr

omcl

imat

ech

ange

.Pr

ojec

tions

sugg

estt

hatt

hecu

rren

trat

eof

grow

thin

agri

cultu

ralp

rodu

ctio

nw

illm

eetf

utur

ede

man

ds;b

utca

ptur

efis

heri

esw

illpr

obab

lyfa

ilto

mee

tris

ing

dem

and,

sinc

epr

oduc

tion

isun

likel

yto

goup

and

impo

undm

ents

may

lead

tode

clin

es.

The

reis

wid

espr

ead

pove

rty

inth

eL

ower

Mek

ong

regi

on.T

hepe

ople

inC

ambo

dia

and

Lao

PDR

are

amon

gth

epo

ores

tin

the

wor

ld,a

ndin

the

nort

heas

tern

part

ofT

haila

ndan

dth

epr

ovin

ces

ofV

ietn

amth

atar

epa

rtof

the

basi

n,m

any

peop

lesu

ffer

from

seve

repo

vert

y.Po

vert

yis

clos

ely

rela

ted

toac

cess

tow

ater

and

culti

vabl

ela

nd,a

sw

ella

sto

fish.

(Con

tinue

d)


84 G.S. Cumming

Tabl

e4.

(Con

tinue

d)

Que

stio

nB

urni

ngis

sues

inth

eba

sin

Alth

ough

wat

erav

aila

bilit

yis

notc

onsi

dere

da

huge

conc

ern

inth

eM

ekon

gB

asin

(exc

epti

nce

rtai

nar

eas,

such

asno

rthe

ast

Tha

iland

,dur

ing

the

dry

seas

on),

alte

ratio

nsto

flow

regi

mes

byim

poun

dmen

tsan

dir

riga

tion

dive

rsio

nsar

eim

pact

ing

rive

rec

olog

y,fis

hpr

oduc

tion,

acce

ssto

wat

er,a

ndfo

odse

curi

ty.C

hang

esin

the

natu

ralfl

owre

gim

em

ayal

ter

the

envi

ronm

ento

ffis

heri

esin

the

Tonl

eSa

pan

del

sew

here

.Alte

red

low

flow

sm

ayle

adto

salin

ityin

trus

ion

into

the

Mek

ong

Del

ta,t

hus

alte

ring

the

bala

nce

ofri

cean

dsh

rim

ppr

oduc

tion,

whi

chin

turn

may

affe

ctfo

odse

curi

tyan

din

com

es.

And

es,S

outh

Am

eric

aA

reas

ofst

ress

are

usua

llyas

soci

ated

with

degr

adat

ion

ofth

ena

tura

lenv

iron

men

tand

itsim

pact

onth

equ

ality

and

quan

tity

ofw

ater

supp

ly;i

ncre

asin

gde

man

dsfo

rw

ater

and

ener

gyfr

omin

crea

sing

lyla

rge,

urba

nize

dpo

pula

tions

;and

the

juxt

apos

ition

ofw

ater

-dem

andi

ngac

tiviti

esw

ithot

hers

(suc

has

min

ing,

was

tew

ater

,oil

and

gas)

whi

chca

nsp

oilw

ater

for

dow

nstr

eam

uses

.C

limat

ech

ange

isan

emer

ging

thre

at,p

artic

ular

lyto

sens

itive

head

wat

erec

osys

tem

ssu

chas

clou

dfo

rest

san

dpa

ram

os.A

reas

rely

ing

onsn

owm

eltm

aysu

ffer

.A

cces

sto

wat

eris

also

impo

rtan

tin

this

syst

eman

dm

ayle

adto

confl

ict.

Lan

dan

dw

ater

legi

slat

ion

and

polic

yar

eva

riab

lean

dso

met

imes

poor

lyde

velo

ped

inA

ndea

nco

untr

ies

and

this

can

seri

ousl

yun

derm

ine

the

sust

aina

bilit

yof

com

mon

reso

urce

s.

Thi

sta

ble

sum

mar

izes

som

eof

the

prob

lem

sth

atse

emm

ostl

ikel

yto

mov

ein

divi

dual

rive

rbas

ins

inth

isse

tofc

ase

stud

ies

into

ane

wso

cial

–eco

logi

calr

egim

e.N

ote

that

foro

bvio

usre

ason

s,th

islis

tlar

gely

excl

udes

unex

pect

edpe

rtur

batio

nsan

dsu

rpri

ses.



change imposes selective pressure on social–ecological systems, and it is more likely thatimportant system elements will be able to persist into the future if a range of potential solu-tions is available. Similarly, it is important that managers in SESs think carefully throughthe implications of capital in all its forms (ecological, social, financial, human and phys-ical) for resilience (Pretty 2008). Capital can play an important enabling function duringtimes of change (Adger 2003, Deutsch et al. 2003, Brand 2009) and conversion of capitalfrom one form to another, such as from ecological to financial, may alter system resilienceto future perturbations.

Ecosystem services are fundamental to the wellbeing of human populations living inbig river basins, and development often requires tradeoffs between different ecosystemservices (Rodriguez et al. 2006, Mainuddin et al. 2009). I asked individual case studyteams to identify key tradeoffs and compromises, as well as important actions that would berelevant in different basins for the sustainable management of ecosystem services. Theseresponses, together with some of my own interpretations of individual basin reports, aresummarized in Table 5.

As these responses indicate, there are a number of apparent needs that are integral to thelong-term resilience of most of the case studies. Chief among these is the need to developmore effective ways of integrating water resource management such that it bridges thegaps between upstream and downstream users, and between different sectors, more effec-tively. Effective management will require effective enforcement of regulations, anothermajor problem.

Table 5 suggests some possible pathways to greater resilience, but it does not offer acomprehensive route to resilience for any single river basin. Resilience to climate changeat present appears to be higher in basins that are not yet water limited. Most big riverbasins are experiencing high population growth, however, and increasing demand for food(and hence, agricultural expansion). Building resilience to climate change in each of thesesystems will require proactive, forward-looking management and regulation. It is alsoimportant to note that resilience is not always desirable (dysfunctional management sys-tems being a case in point). Resilience will need to be built in some system componentsand reduced in others; and in some case studies, such as the Yellow River, transformationof the SES (with the loss of some or all of its current “key” elements, for good or forbad) seems virtually inevitable. Olsson et al. (2004), based on their study of a successfultransformation of a wetland SES (Kristianstad) in southern Sweden, suggest three key stepstowards social transformation: preparing the system for change, seizing a window of oppor-tunity, and building social–ecological resilience of the new desired state. Transformationin Kristianstad was facilitated by the creation of a new bridging organization which wasable to bring together key actors.

Managers and water users in many cases are aware of many of the issues and possi-bilities listed in Table 5. Awareness of solutions, however, does not always translate intosolutions; action occurs at the end of a political process in which different needs and valuesare considered and alternative solutions are negotiated (Cronon 2000). The developmentand implementation of more effective management strategies in most case study basins iscurrently being hindered by a wide range of political variables, including historical incom-patibilities, power struggles, a lack of trust and overlapping (and sometimes contradicting)legislation. For example, in some of the BFP case studies, different water management dis-tricts refuse to share their water flow data to provide a comprehensive picture of basin-wideflows. There is also a paucity of relevant science-based information on which appropriatelegislation can be based. Some of the major uncertainties and information gaps in the casestudy basins are summarized in Table 6.


86 G.S. Cumming

Tabl

e5.

Tra

deof

fsan

dot

herc

ompr

omis

espe

rcei

ved

byba

sin

case

stud

yte

ams

asne

eded

fori

mpr

ovin

gth

esu

stai

nabi

lity

ofbi

gri

verb

asin

s;an

dsu

gges

tions

for

nece

ssar

ych

ange

sto

enha

nce

resi

lienc

ein

each

basi

n.

Bas

inC

ompr

omis

esor

trad

eoff

sw

ithre

leva

nce

for

sust

aina

bilit

yPa

thw

ays

prop

osed

tole

adto

grea

ter

resi

lienc

e

Lim

popo

,So

uthe

rnA

fric

a

Eco

nom

icvs

.nat

ural

capi

tal:

sust

aina

ble

man

agem

ento

fec

osys

tem

serv

ices

,and

part

icul

arly

redu

ctio

nsin

wat

erus

e,w

illha

veto

bepa

idfo

r.Sh

ort-

term

vs.l

ong-

term

bene

fits:

rura

lhou

seho

lds

will

have

tobe

com

pens

ated

for

the

loss

ofec

onom

icop

port

unity

that

they

face

whe

nco

nser

ving

the

reso

urce

base

.Tr

adeo

ffs

betw

een

sect

ors:

indu

stry

,par

ticul

arly

min

ing

and

agri

cultu

re,w

illha

veto

bala

nce

thei

rne

eds

with

thos

eof

rura

lcom

mun

ities

.

Impr

ove

equi

ty,r

educ

em

argi

naliz

atio

nof

rura

lpoo

rby

educ

atio

nan

dbu

ildin

glo

calc

apac

ityan

din

stitu

tions

.In

trod

uce

mor

eef

fect

ive

polic

ing

and

enfo

rcem

ento

fex

istin

gre

gula

tions

.Im

prov

epr

icin

gan

dpr

iori

tizat

ion

stru

ctur

esfo

rsm

allh

olde

rw

ater

use.

Dev

elop

mor

eef

fect

ive

way

sfo

rdi

ffer

entc

ount

ries

inth

eba

sin

toin

tera

ctan

dco

-man

age

wat

erre

sour

ces,

thro

ugh

stre

ngth

enin

gL

IMC

OM

.Pl

anni

ngbe

ing

done

now

mus

tbe

base

don

best

avai

labl

ecl

imat

efo

reca

sts

unde

rth

eas

sum

ptio

nth

atth

ere

will

bele

ssw

ater

,mor

em

ajor

wea

ther

-rel

ated

even

ts,a

ndin

crea

sed

tem

pera

ture

s.

Kar

keh,

Iran

Ant

hrop

ogen

icvs

.eco

syst

emne

eds:

ecos

yste

ms

need

redu

ced

grou

ndw

ater

usag

ean

din

corp

orat

ion

ofen

viro

nmen

talfl

ows

into

deci

sion

mak

ing.

Ups

trea

mvs

.dow

nstr

eam

trad

eoff

s:do

wns

trea

mw

ater

law

sne

edto

bere

vise

d.H

igh

vs.l

owec

onom

icva

lue

activ

ities

:pos

sibi

lity

tosh

ift

wat

eraw

ayfr

om(l

ower

-val

ue)

food

prod

uctio

nan

dto

war

dshy

drop

ower

and

urba

nus

es.S

peci

fical

ly,d

iffic

ult

trad

eoff

sex

pect

edbe

twee

nhy

drop

ower

and

irri

gatio

nne

eds.

Ext

end

trad

ene

twor

ksto

mee

tfut

ure

food

dem

and.

Dev

elop

ala

rger

seto

fno

n-ag

ricu

ltura

lway

sof

gene

ratin

gin

com

e.N

eed

toim

prov

ew

ater

man

agem

enta

ndal

loca

tion

proc

esse

s.Im

prov

ew

ater

use

effic

ienc

yof

irri

gate

dag

ricu

lture

.M

apw

ater

prod

uctio

nto

iden

tify

oppo

rtun

ities

for

incr

easi

ngw

ater

quan

tity.

Nile

,Nor

than

dE

astA

fric

aU

pstr

eam

vs.d

owns

trea

mtr

adeo

ffs:

diff

eren

ces

inw

ater

acce

ss,d

eman

dan

dm

anag

emen

tmus

tbe

reso

lved

.W

ater

quan

tity

avai

labl

eto

hum

ans

vs.w

ater

for

ecos

yste

ms

(inc

ludi

ngw

ildlif

ean

dgr

azer

s)on

flood

plai

nsin

the

Sudd

area

(Jon

glei

Can

alde

bate

).C

onsu

mpt

ive

vs.n

on-c

onsu

mpt

ive

uses

(for

exam

ple,

hydr

opow

ervs

.irr

igat

ion)

.

Cre

ate

enab

ling

envi

ronm

ent(

basi

n-w

ide

inst

itutio

nw

ithla

w-e

nfor

cem

entc

apac

ity,w

ater

man

agem

entp

olic

y,st

rate

gies

and

regu

latio

n).

Impr

ove

prod

uctiv

ityof

farm

land

and

agri

cultu

ralw

ater

man

agem

ent.

Red

uce

wat

erw

aste

/los

san

din

crea

seav

aila

bilit

yC

onsi

der

clim

ate

chan

gean

den

viro

nmen

tali

mpa

ctas

sess

men

tdur

ing

proj

ectp

lann

ing

and

desi

gn.



Mul

tivar

iate

trad

eoff

sbe

twee

ndi

ffer

ent,

pote

ntia

llyco

nflic

ting

uses

ofw

ater

(for

exam

ple,

irri

gatio

n,fis

hing

,hy

drop

ower

,dri

nkin

g,w

ashi

ng,t

rans

port

,rep

osito

ryfo

rw

aste

).

Attr

acti

nves

tmen

tfun

ds(e

cono

mic

capi

tal)

.R

esol

vepo

litic

aldi

ffer

ence

sbe

twee

nna

tions

,par

ticul

arly

Egy

ptan

dot

her

wat

erus

ers.

Bui

ldm

anag

emen

tcap

abili

ties

ofag

ricu

ltura

land

ranc

hing

com

mun

ities

.

Vol

ta,W

est

Afr

ica

Few

expl

icit

trad

eoff

sat

pres

entb

ecau

sew

ater

volu

me

not

limiti

ng,a

lthou

ghra

infa

llis

high

lyva

riab

lean

dso

ilsar

epo

or.

Futu

retr

adeo

ffs

betw

een

stor

age

and

irri

gatio

nan

dcr

oppi

ngvs

.ran

chin

gsy

stem

sse

emlik

ely.

Intr

oduc

ebe

tter

man

agem

entf

orgr

assl

ands

,par

ticul

arly

for

lives

tock

with

acce

ssto

wat

eran

dus

eof

crop

resi

dues

(liv

esto

ckpe

rcei

ved

ason

lypo

ssib

leus

eof

gras

slan

dsin

the

nort

hern

drie

rpa

rtof

the

basi

n).

Dev

elop

refo

rest

atio

npr

ogra

mm

es,r

educ

ede

fore

stat

ion

rate

.So

ilus

ecu

rren

tlyun

sust

aina

ble;

intr

oduc

esm

all-

scal

eir

riga

tion

and

bette

rfa

rmin

gpr

actic

es.

Poss

ibly

regu

late

fishe

ries

and

limit

expa

nsio

nof

rice

prod

uctio

nin

tosm

alli

nlan

dva

lleys

.

Nig

er,W

est

Afr

ica

Impo

undm

ents

tosu

ppor

thyd

ropo

wer

and

irri

gatio

ndu

ring

low

-flow

peri

ods

vs.d

owns

trea

mri

ver

flow

for

herd

ers,

fishe

rmen

and

rice

culti

vato

rs.

Liv

esto

ckra

nchi

ngof

nom

adic

herd

svs

.agr

icul

ture

(whi

chre

stri

cts

wat

erac

cess

and

cattl

em

ovem

ents

).T

rade

offs

betw

een

diff

eren

tcro

psw

ithdi

ffer

entv

alue

san

dw

ater

dem

ands

(for

exam

ple,

rice

vs.m

arke

tgar

dens

).C

usto

mar

yla

ws

vs.g

over

nmen

tall

egal

syst

em(e

ffec

tiven

ess

ofdi

ffer

ents

yste

ms

for

wat

erm

anag

emen

tdi

ffer

s;pl

ural

ityca

ncr

eate

tenu

rein

secu

rity

and

nega

tivel

yim

pact

deve

lopm

ent)

.

Impr

oved

agri

cultu

ralp

rodu

ctiv

ity.

Impr

oved

wat

erpr

oduc

tivity

and/

orm

ore

effe

ctiv

ew

ater

man

agem

ent(

grou

ndw

ater

extr

actio

n,da

mco

nstr

uctio

n,an

dso

on).

Incl

usio

nof

mar

gina

lised

sect

ors,

such

asfis

hers

and

nom

adic

past

oral

ists

,in

deci

sion

-mak

ing

and

plan

ning

proc

esse

s.C

onfli

ctre

solu

tion

betw

een

nom

ads

and

farm

ers.

Res

olut

ion

ofla

ndte

nure

issu

esst

emm

ing

from

lega

lpl

ural

ism

(tra

ditio

nalv

s.le

gals

yste

ms)

.Im

prov

eded

ucat

ion

and

acce

ssto

clea

nw

ater

.

Yello

wR

iver

,C

hina

As

acl

osed

basi

n,us

eof

wat

erby

any

sect

oral

mos

tin

evita

bly

enta

ilsa

trad

eoff

with

othe

rse

ctor

s.Ir

riga

ted

agri

cultu

rein

mid

dle

and

uppe

rse

ctio

nsof

basi

nvs

.ind

ustr

ial,

dom

estic

and

ecos

yste

mde

man

dsdo

wns

trea

m:

Tran

sbou

ndar

y(p

rovi

nce)

wat

erdi

vers

ion

vs.w

ater

dem

and,

part

icul

arly

indo

wns

trea

mar

eas

with

inth

eba

sin;

Enf

orce

wat

erqu

ality

regu

latio

nsto

ensu

reth

atle

ssin

dust

rial

was

tean

ddo

mes

ticse

wag

een

ters

the

rive

r.E

limin

ate

fert

ilize

rsu

bsid

ies,

whi

chle

adto

nutr

ient

runo

ffan

dpo

orw

ater

qual

ity.

Find

new

way

sto

com

pens

ate

prov

ince

sfo

rusi

ngle

ssw

ater

,pa

rtic

ular

lyup

stre

ampr

ovin

ces

whe

rebe

nefit

per

unit

ofw

ater

used

islo

wer

(how

ever

,giv

enth

ere

use

offlo

ws

with

inth

eba

sin,

the

pote

ntia

lfor

savi

ngis

also

limite

d).

(Con

tinue

d)


88 G.S. Cumming

Tabl

e5.

(Con

tinue

d)

Bas

inC

ompr

omis

esor

trad

eoff

sw

ithre

leva

nce

for

sust

aina

bilit

yPa

thw

ays

prop

osed

tole

adto

grea

ter

resi

lienc

e

Was

tere

mov

alvs

.oth

erw

ater

qual

ity-r

elat

edse

rvic

es(w

aste

wat

erdi

scha

rge

high

inup

per

and

mid

dle

sect

ions

).

Enh

ance

dco

oper

atio

nbe

twee

nth

em

inis

trie

sof

agri

cultu

re(M

OA

)an

dw

ater

reso

urce

s,as

the

MO

A’s

polic

ies

tend

toin

crea

sepr

essu

reon

wat

erre

sour

ces

with

outt

akin

gth

ese

into

acco

unti

nth

eir

deci

sion

-mak

ing

proc

esse

s.Fe

rtili

ser

use

for

food

prod

uctio

nvs

.dow

nstr

eam

wat

erqu

ality

.Fl

ood

cont

rolv

s.w

ater

stor

age.

Pove

rty

alle

viat

ion

(via

irri

gate

dag

ricu

lture

)vs

.eco

syst

emin

tegr

ity.

Red

uce

agri

cultu

ralw

ater

use,

depe

ndin

gon

how

muc

hw

ater

use

effic

ienc

yha

sbe

enin

crea

sed.

Dec

entr

aliz

atio

nof

wat

erri

ghts

vs.c

entr

alco

ntro

l.In

vest

ing

for

rura

llab

our

expa

nsio

nin

agri

cultu

revs

.in

vest

ing

inno

n-ag

ricu

ltura

ljob

crea

tion.

Dev

elop

asi

ngle

ratio

nala

lloca

tion

plan

betw

een

regi

ons,

depe

ndin

gon

whe

ther

the

pres

enta

lloca

tion

plan

can

bere

vise

d.L

imit

indu

stri

alde

velo

pmen

twhe

reth

ere

isin

suffi

cien

tw

ater

tom

eeti

tsne

eds.

Cla

rify

natu

reof

wat

erri

ghts

and

cons

ider

crea

ting

am

ore

effe

ctiv

ew

ater

mar

ket.

Indu

s-G

ange

sB

asin

s,In

dia

and

Paki

stan

Inth

eIn

dus,

acl

osed

basi

n,us

eof

wat

erby

any

sect

oral

mos

tine

vita

bly

enta

ilsa

trad

eoff

with

othe

rse

ctor

s.U

pstr

eam

vs.d

owns

trea

mus

es.

Trad

eoff

sbe

twee

nec

onom

icre

turn

son

diff

eren

tcro

psan

don

mon

ocul

ture

vs.d

iver

secr

oppi

ngsy

stem

s.R

elia

nce

onfo

rmal

vs.i

nfor

mal

inst

itutio

ns(t

rade

offs

infle

xibi

lity,

resp

onsi

vene

ss,e

nfor

cem

ent,

and

adhe

renc

eto

rule

s).

Polic

ies

toen

cour

age

grou

ndw

ater

pum

ping

(sho

rt-t

erm

gain

s)vs

.gro

undw

ater

rech

arge

(lon

g-te

rmsu

stai

nabi

lity)

.

Impr

ove

food

secu

rity

for

the

regi

on,e

spec

ially

the

poor

popu

latio

n.D

evel

opne

wcr

oppi

ngpa

ttern

sin

resp

onse

tocl

imat

ech

ange

.C

reat

esu

ffici

enta

ttrac

tive

off-

farm

empl

oym

ent(

invo

lve

farm

ers

inot

her

sect

ors)

.M

anag

eex

trem

eev

ents

bette

r(i

mpr

ove

cont

inge

ncy

prep

ared

ness

)an

dre

duce

vuln

erab

ility

ofth

epo

pula

tion.

Prov

ide

ener

gyse

curi

tyto

the

rura

lpop

ulat

ion

thro

ugh

alte

rnat

ive

fuel

s.Im

prov

em

arke

tsan

din

fras

truc

ture

.E

mpo

wer

poor

peop

le,e

spec

ially

wom

en.

Mek

ong,

Asi

aD

evel

opm

ento

fhy

drop

ower

dam

svs

.dow

nstr

eam

wat

erus

ean

dec

osys

tem

func

tion

(inc

ludi

ngpr

oduc

tion

offis

hfo

rca

ptur

e).

Impr

ove

tenu

rean

dac

cess

righ

tsof

the

poor

tona

tura

lre

sour

ces;

give

mar

gina

lised

com

mun

ities

avo

ice.

Mai

ntai

nsu

ffici

ently

good

dow

nstr

eam

wat

erqu

ality

and

quan

tity

tosu

ppor

tfish

erie

s.



Dev

elop

men

tof

new

irri

gatio

nsy

stem

and

incr

easi

ngto

tal

crop

prod

uctio

n,ag

ain

vs.d

owns

trea

mw

ater

use

and

ecos

yste

mfu

nctio

n(i

nclu

ding

prod

uctio

nof

fish)

,but

also

vs.o

ther

land

uses

.V

alue

san

dliv

elih

oods

from

upst

ream

crop

svs

.oth

erec

osys

tem

serv

ices

and

valu

es(f

orex

ampl

e,er

osio

nco

ntro

lin

stee

pup

land

san

dliv

elih

oods

from

fore

sted

/na

tura

lare

as).

Lim

itdr

yse

ason

irri

gatio

nvs

.sea

wat

erin

trus

ion

inth

ede

lta(a

ndco

nseq

uent

bala

nce

ofri

cean

dsh

rim

pfa

rmin

gin

the

delta

).

Impr

ove

curr

enti

nstit

utio

ns(f

orex

ampl

e,st

reng

then

Mek

ong

Riv

erC

omm

issi

on)

and

deve

lop

am

ore

holis

ticin

stitu

tiona

lper

spec

tive

(tha

tis,

noto

nly

focu

sed

ongr

owth

and

deve

lopm

ent)

.Im

prov

elo

cala

ndre

gion

algo

vern

ance

;dev

elop

way

sto

prev

entn

atio

nalp

oliti

cali

nter

ests

(for

exam

ple,

inde

bted

ness

toC

hina

for

aid)

from

beco

min

gto

odo

min

anti

nde

cisi

onm

akin

g.

And

es,S

outh

Am

eric

aU

pstr

eam

vs.d

owns

trea

mac

tiviti

esan

dim

pact

s(f

orex

ampl

e,m

inin

gvs

.dom

estic

wat

erus

e).

Loc

alvs

.nat

iona

lben

efits

,inc

reas

ein

vest

men

tin

wat

ertr

eatm

entf

orur

ban

cent

res,

with

loca

lcos

tsbu

tnat

iona

lbe

nefit

s.E

cono

mic

bene

fits

vs.e

colo

gica

lint

egri

ty(p

oor

envi

ronm

enta

lrec

ord

ofag

ricu

lture

inth

ere

gion

).Tr

adeo

ffs

betw

een

com

petin

gla

ndus

esin

stee

par

eas

(for

exam

ple,

agri

cultu

ralp

rodu

ctio

nsy

stem

s,m

ajor

dam

proj

ects

,int

er-b

asin

tran

sfer

s,m

inin

g).

Dow

nstr

eam

irri

gatio

nvs

.wat

ersu

pply

tom

ajor

citie

s.

Man

age

land

use

mor

eca

refu

lly(f

orex

ampl

e,de

fore

stat

ion

impa

cts

onw

ater

and

sedi

men

tinp

uts

tohy

drop

ower

dam

s;or

agri

cultu

rali

mpa

cts

onw

ater

qual

ity).

Impr

ove

man

agem

ento

fw

aste

wat

eran

def

fluen

tsfr

omci

ties

and

indu

stry

.Im

prov

epo

litic

alst

abili

ty,r

educ

eru

ralv

iole

nce,

impl

emen

tbe

tter

gove

rnan

ce.

Dev

elop

way

sof

incl

udin

gcu

rren

tlym

argi

nalis

edgr

oups

inde

cisi

on-m

akin

gpr

oces

ses

inth

eir

own

regi

ons.

Intr

oduc

tion

ofco

mpe

nsat

ion

sche

mes

(PE

S/PW

S)to

enco

urag

ebe

tter

reso

urce

use.

Thi

sta

ble

com

bine

sre

spon

ses

toa

ques

tionn

aire

and

my

own

inte

rpre

tatio

nsof

info

rmat

ion

pres

ente

din

case

stud

yre

port

s.


90 G.S. Cumming

Tabl

e6.

Maj

orin

form

atio

nne

eds

inei

ghtb

igri

ver

basi

ns,a

sid

entifi

edby

case

stud

yte

ams.

Que

stio

nM

ajor

info

rmat

ion

need

sid

entifi

edby

case

stud

yte

ams

Lim

popo

,So

uthe

rnA

fric

a

Act

ualw

ater

use,

byse

ctor

,per

coun

try

isst

illdi

fficu

ltto

pinp

oint

.B

ette

rda

tafr

omB

otsw

ana.

The

chan

ging

,unp

redi

ctab

lepo

litic

alsi

tuat

ion

inZ

imba

bwe

and

pote

ntia

llySo

uth

Afr

ica

mak

ege

nera

tiona

lpla

nnin

gan

das

sess

men

tdi

fficu

lt.Po

litic

alpr

iori

ties

may

chan

gequ

ickl

y,th

ereb

yre

alig

ning

the

econ

omic

land

scap

ean

dpl

acin

gpr

essu

reon

natu

ral

reso

urce

s.T

hese

dyna

mic

sar

epo

orly

unde

rsto

od.

Lik

ely

impa

cts

ofcl

imat

ech

ange

.

Kar

keh,

Iran

Bet

ter

hydr

olog

ical

data

,bot

hfo

rflo

ws

inba

sins

and

grou

ndw

ater

–sur

face

wat

erin

tera

ctio

n.U

nder

stan

ding

how

tode

velo

pan

dim

plem

entg

roun

dwat

erpo

licy.

Bet

ter

unde

rsta

ndin

gof

linka

ges

betw

een

wat

eran

dpo

vert

yin

the

basi

n.R

elat

ions

hips

betw

een

prod

uctio

nsy

stem

sno

ttyp

ical

lyin

clud

edin

basi

nw

ater

asse

ssm

ents

(for

exam

ple,

ranc

hing

,for

estr

y,fis

heri

es)

and

wat

ersu

pply

and

dem

and.

Lik

ely

clim

ate

chan

geim

pact

son

fres

hwat

erav

aila

bilit

yin

basi

n.Im

pact

sof

Iran

ian

gove

rnm

ent’s

deci

sion

tore

mov

esu

bsid

ies

onen

ergy

and

agri

cultu

rali

nput

s.

Nile

,Nor

than

dE

astA

fric

aL

ittle

data

avai

labl

eon

wat

erre

quir

emen

tsof

wet

land

ecos

yste

ms.

Inad

ditio

n,co

ntri

butio

nsm

ade

toth

eN

ileby

the

Sudd

wet

land

syst

em(o

neof

the

larg

esti

nth

ew

orld

),an

dpo

ssib

leim

pact

son

the

Sudd

ofth

eJo

ngle

iCan

al,a

repo

orly

unde

rsto

od.

The

rela

tions

hips

betw

een

biop

hysi

calv

aria

bles

and

ecol

ogic

alin

dica

tors

are

poor

lyun

ders

tood

.C

limat

ican

dhy

drol

ogic

mon

itori

ngda

taar

ew

eak;

lack

ofba

sin-

wid

em

onito

ring

stra

tegy

;wat

erqu

ality

info

rmat

ion

lack

ing;

clim

ate

chan

gepr

ojec

tions

unce

rtai

n.

Vol

ta,W

est

Afr

ica

Qua

lity/

relia

bilit

yof

publ

ishe

dda

tafr

omst

ate

agen

cies

isof

ten

notk

now

n(f

orex

ampl

e,ag

ricu

lture

prod

uctio

n,liv

esto

ck,c

ultiv

ated

area

,fine

-sca

lepo

vert

yda

ta).

Dat

aga

psfr

omco

untr

ies

that

have

only

asm

allp

arto

fth

eba

sin.

Lar

gega

psin

hydr

olog

ical

(riv

erdi

scha

rge)

data

.Sim

ilarl

y,gr

ound

wat

erre

sour

ces

poor

lyun

ders

tood

.C

limat

epr

ojec

tions

unce

rtai

n.Su

stai

nabi

lity

offis

heri

eson

Lak

eV

olta

uncl

ear.

Nig

er,W

est

Afr

ica

Popu

latio

ngr

owth

data

.Im

pact

sof

clim

ate

chan

geun

cert

ain.

Nee

dto

rese

arch

tode

term

ine

how

toim

prov

eag

ricu

ltura

lout

puti

nlo

calc

ondi

tions

.W

ays

toim

prov

eup

take

ofag

ricu

ltura

lpro

duct

ivity

inte

rven

tions

.W

ays

toim

prov

ego

vern

ance

and

inte

grat

edw

ater

reso

urce

man

agem

ent.



Yello

wR

iver

,C

hina

Nee

dm

onito

ring

,inf

orm

atio

nan

daw

aren

ess

gene

ratio

non

wat

erqu

ality

situ

atio

nin

the

basi

n.In

form

atio

nex

chan

gene

eds

betw

een

min

istr

ies

ofag

ricu

lture

and

wat

erto

asse

ssim

pact

sof

agri

cultu

ralp

olic

ies

onw

ater

avai

labi

lity.

Key

unce

rtai

ntie

sne

edto

bere

solv

ed:c

osto

fpo

orw

ater

qual

ityto

the

publ

ican

dth

eco

untr

y;co

stof

cont

inui

ngbu

sine

ssas

usua

l;co

stof

rigi

dw

ater

allo

catio

npl

anni

ng(f

orex

ampl

e,pr

ovin

ce-l

evel

quot

asy

stem

s).

App

roac

hes

todr

ough

tman

agem

ent(

lack

ofpo

licie

sgi

ven

past

focu

son

flood

cont

rol)

.Im

pact

sof

clim

ate

chan

geon

both

surf

ace

and

grou

ndw

ater

supp

ly.

How

muc

hw

ater

use

effic

ienc

yca

nbe

incr

ease

din

agri

cultu

rals

ecto

r,an

dlik

ely

impa

cts

ofad

optin

gag

ricu

ltura

lwat

er-s

avin

gte

chno

logi

eson

loca

land

basi

n-le

velw

ater

use.

Irri

gate

dan

dra

infe

dar

eaby

crop

isun

know

n.C

rop

evap

otra

nspi

ratio

n(E

T)

bym

onth

byco

unty

(or

bette

rre

solu

tion)

.W

ater

requ

irem

ents

for

dow

nstr

eam

ecos

yste

ms

(est

imat

esfr

omth

eva

riou

sso

urce

sar

eve

rydi

ffer

ent)

.

Indu

s-G

ange

sB

asin

s,In

dia

and

Paki

stan

Due

totr

ansb

ound

ary

natu

rean

dlo

wtr

ustl

evel

amon

gth

eri

pari

anco

untr

ies,

wat

erre

sour

ces

data

for

basi

n-le

vela

naly

sis

are

unav

aila

ble.

Asi

mila

rsi

tuat

ion

also

exis

tsfo

rgr

ound

wat

er/fl

ood

data

(and

othe

rre

sour

ces)

.Im

pact

ofcl

imat

ech

ange

upon

Him

alay

anw

ater

reso

urce

sis

notw

ellu

nder

stoo

dan

dth

ere

are

confl

ictin

gsc

ient

ific

evid

ence

.Po

orkn

owle

dge

ofus

eby

sect

oran

dre

venu

esge

nera

ted

for

wat

erse

rvic

es.

Impa

ctof

wat

erre

sour

ces

deve

lopm

ento

nlo

cale

colo

gy,b

iodi

vers

ity,l

ivel

ihoo

ds,d

ispl

acem

ents

and

othe

rco

sts.

Mek

ong,

Asi

aIm

pact

ofda

ms,

new

irri

gatio

nsy

stem

san

dup

stre

amde

velo

pmen

ton

the

low

erpa

rtof

the

basi

n,on

the

capt

ure

fishe

ries

and

onth

een

viro

nmen

tand

ecol

ogy.

Und

erst

andi

ngre

latio

nshi

psbe

twee

nfo

odse

curi

tyan

dpo

vert

yre

duct

ion.

Impa

ctof

clim

ate

chan

geon

flow

s,ag

ricu

ltura

lpro

duct

ivity

,fish

erie

sec

olog

y,en

viro

nmen

tand

sea

leve

lris

e.

And

es,S

outh

Am

eric

aB

ette

rin

tegr

ated

mod

ellin

gan

dre

sear

chth

atis

capa

ble

ofm

ovin

gbe

yond

sing

leis

sues

tolo

okin

gat

mul

tiple

issu

esan

dun

fore

seen

cons

eque

nces

.B

ridg

ing

scie

nce

and

polic

yto

prov

ide

polic

ysu

ppor

t.B

ette

rm

easu

rem

enta

ndm

onito

ring

ofso

me

ofth

efu

ndam

enta

ls(f

orex

ampl

e,ra

infa

llan

ddi

stri

butio

nof

wat

ertr

eatm

entp

lant

s).

Bet

ter

info

rmat

ion

onth

ero

leof

pove

rty

inde

term

inin

gw

ater

acce

ss.

Bet

ter

mea

sure

sof

livel

ihoo

dsan

dw

ellb

eing

,bey

ond

pure

lyec

onom

icm

easu

res.

Furt

her

deta

ilson

thes

ene

eds

are

avai

labl

ein

indi

vidu

alch

apte

rsof

Wat

erIn

tern

atio

nal3

5(5

)(2

010)

.


92 G.S. Cumming

Addressing these information needs would presumably make the case study basinsmore resilient, in the sense that the facts that are needed for informed decision makingwould be more readily available. It is, however, a mistake to assume that simply gettingthe science or even the number and nature of institutions right will solve the problems; thefocus in most cases will need to be on process, and in particular on developing flexible,transparent, inclusive political approaches to dealing with and resolving water allocationissues.

One of the hardest questions to answer in the context of assessing the resilience ofbig river basins is that of whether pro-active managers are thinking about (and preparingfor) the right future. Surprise is an integral element of complex systems and it is quitepossible that new, unexpected threats to the integrity of big river basins will emerge fromunexpected quarters (see, for example, the discussion in Gordon et al. 2008). For example,none of the BFP case studies has considered the possible implications for basin sustain-ability of such possible future trends as increased production of biofuels (de Fraiture et al.2008), outsourcing of agricultural production from wealthier countries to Africa (Davironand Gibbon 2002), or the potential for anthropogenic influences to create “unhealthy land-scapes” that are susceptible to novel forms of water-borne or water-associated pathogens(Patz et al. 2004).

The future will not necessarily conform to the present. Building resilience to surpriseis more a case of building adaptive capacity (for example, through developing monitoringschemes with feedback loops to management, setting in place appropriate decision-makingprocesses, building capital stocks, maintaining diversity and redundancy, and ensuring thatsome “slack” remains in key components of the system) than of managing for a specificoutcome or goal. Managing for resilience inevitably comes at a cost, and many complexsystems will (whether intentionally or not) tend to reduce adaptive capacity by shiftingtowards greater efficiency during periods of relative stability when diversity, redundancy,and under-utilization are seen as wasteful (Holling and Meffe 1996). One of the centralchallenges for big river basin management is thus to put in place the structures (institutions,policies, expertise) that are needed to cope with times of change, and to retain them throughperiods of relative constancy when their direct relevance may be less obvious.

Concluding comments

This brief analysis of the resilience of big river basin SESs in this set of case studies hasraised some important questions and recommendations in two realms: (1) conceptual andscientific understanding; and (2) practical recommendations for managing these systemstowards greater resilience in desired characteristics.

In the conceptual and scientific realm, standard approaches to river basin management(including most development and water allocation programmes) have yet to fully grapplewith the questions that are raised by scientific uncertainty and system complexity. All ofthe case studies face potentially large uncertainties in the estimation and prediction of awide range of important variables, ranging from likely future rainfall and runoff throughto more diffuse variables such as changes in water demand and the future role of technol-ogy in water-use efficiency. Uncertainty exists not only in the estimation of such variables,but also in our understanding of the nature of the relationships between variables. Lakeeutrophication offers a good example of a well-studied, non-linear relationship in freshwa-ter systems (Carpenter 2003), but a wide range of other relationships are likely to showsignificant non-linearities and complex behaviours, particularly if climate change or waterextraction push key variables beyond their normal range of variation (Holling and Meffe



1996, Gordon et al. 2008). Understanding and incorporating such uncertainties into mod-els and decision-making tools remains an important goal for research in many big riverbasins.

In the practical realm, a number of general and specific suggestions emerge fromthe preceding discussion. Complex systems theory highlights the value of maintainingdiversity, redundancy, and capital within a given system as well as of fostering learning,innovation and adaptive capacity where possible (Holling and Meffe 1996, Norberg andCumming 2008). The temptation to blame problems on a single cause or driver (such aspopulation growth or institutional failure), and build expectations around working towardsa single overarching solution or “panacea”, must be avoided (Ostrom 2007). Big riverbasins are complex systems that are composed of multiple interacting variables, whichchange at different speeds and occupy different hierarchical levels. System manipulations,even of biophysical components of the system that are supposedly well understood, mayhave unintended consequences (Patten et al. 2001). Pro-active management will be mosteffective (and most capable of making necessary mistakes, and learning from them) if itoccurs via politically acceptable and participatory processes (Adger and Jordan 2009), usesthe best possible science, and takes a holistic perspective that seeks to incorporate social,ecological, and economic elements in problem definition and solution.

AcknowledgementsI am grateful to Simon Cook for inviting me to contribute this paper and to the nine contributing BFPproject teams for their willingness to respond to my questions and provide me with relevant materials.Most of the text in Tables 2 to 6 was generated by the case study authors, although I may havesubsequently introduced errors. I would particularly like to thank Amy Sullivan (Limpopo); PooladKarimi (Karkheh); David Molden, Seleshi B. Awulachew and Solomon S. Demissie (Nile); JacquesLemoalle (Volta); Andrew Ogilvie and Jean-Charles Clanet (Niger); Claudia Ringler, Jinxia Wangand Ximing Cai (Yellow); Bharat Sharma (Indus-Gangetic); Mac Kirby and Mohammed Mainuddin(Mekong) and Mark Mulligan (Andes). Simon Cook, Tassilo Tiemann and an anonymous reviewerprovided useful comments on earlier versions of this paper.

Note1. Ludwig (2001) defines “wicked” problems as those that include not only multiple disciplines,

stakeholders and epistemologies, but also that cannot be separated from issues of values, equityand social justice.

ReferencesAdger, W., 2003. Social capital, collective action, and adaptation to climate change. Economic

Geography, 79 (4), 387–404.Adger, W.N., 2006. Vulnerability. Global Environmental Change, 16 (3), 268–281.Adger, W.N. and Jordan, A., 2009. Sustainability: exploring the processes and outcomes of gover-

nance. In: W.N. Adger, and A. Jordan, eds. Governing sustainability. Cambridge: CambridgeUniversity Press, 3–31.

Agarwal, C., et al. 2002. A review and assessment of land-use change models: dynamics of space,time, and human choice. General Techical Report NE-297. Newtown Square, PA: US Departmentof Agriculture, Forest Service, Northeastern Research Station, 61.

Anderies, J.M., 2004. Minimal models and agroecological policy at the regional scale: an applicationto salinity problems in southeastern Australia. Regional Environmental Change, 5 (1), 1–17.

Anderies, J.M., Janssen, M.A., and Ostrom, E., 2004. A framework to analyze the robustness ofsocial-ecological systems from an institutional perspective [online]. Ecology and Society, 9(1), 18. Available from: www.ecologyandsociety.org/vol9/iss1/art18/ [Accessed 15 November2010].


http://www.ecologyandsociety.org/vol9/iss1/art18

94 G.S. Cumming

Brand, F., 2009. Critical natural capital revisited: ecological resilience and sustainable development.Ecological Economics, 68 (3), 605–612.

Carlson, J.M. and Doyle, J., 2002. Complexity and robustness. Proceedings of the National Academyof Sciences of the United States of America, 99 (Suppl. 1), 2538–2545.

Carpenter, S., 2003. Regime shifts in lake ecosystems: patterns and variation. Excellence in EcologySeries Number 15. Oldendorf/Luhe, Germany: Ecology Institute.

Carpenter, S.R. and Turner, M.G., 2000. Hares and tortoises: interactions of fast and slow variablesin ecosystems. Ecosystems, 3 (6), 495–497.

Carpenter, S., et al., 2001. From metaphor to measurement: resilience of what to what? Ecosystems,4 (8), 765–781.

Checkland, P., 1981. Systems thinking, systems practice. New York: Wiley.Checkland, P., 2009. Systems thinking, systems practice (with 30-year retrospective). New York:

Wiley.Clark, J.S., et al., 2001. Ecological forecasts: an emerging imperative. Science, 293 (5530), 657–660.Clark, M.J., 2002. Dealing with uncertainty: adaptive approaches to sustainable river management.

Aquatic Conservation: Marine and Freshwater Ecosystems, 12 (4), 347–363.Cronon, W., 2000. Resisting monoliths and tabulae rasae. Ecological Applications, 10 (3), 673–675.Cumming, G.S., 2011. Spatial resilience in social-ecological systems. Berlin: Springer.Cumming, G.S., et al., 2005. An exploratory framework for the empirical measurement of resilience.

Ecosystems, 8 (8), 975–987.Cumming, G.S. and Collier, J., 2005. Change and identity in complex systems [online]. Ecology

and Society, 10, 29. Available from: http://www.ecologyandsociety.org/vol10/iss1/art29/[Accessed 15 November 2010].

Cumming, G.S., Cumming, D.H.M., and Redman, C.L., 2006. Scale mismatches in social-ecologicalsystems: causes, consequences, and solutions [online]. Ecology and Society, 11 (1), 14. Availablefrom: http://www.ecologyandsociety.org/vo11/iss1/art14/ [Accessed 15 November 2010].

Daviron, B. and Gibbon, P., 2002. Global commodity chains and African export agriculture. Journalof Agrarian Change, 2 (2), 137–161.

de Fraiture, C., Giordano, M., and Liao, Y., 2008. Biofuels and implications for agricultural wateruse: blue impacts of green energy. Water Policy, 10 (Suppl. 1), 67–81.

Deutsch, L., Folke, C., and Skanberg, K., 2003. The critical natural capital of ecosystem performanceas insurance for human well-being. Ecological Economics, 44 (2–3), 205–217.

Ernstson, H., et al., in press. Scale-crossing brokers and network governance of urban ecosystemservices: the case of Stockholm, Sweden. Ecology and Society, (in press).

Gonzalez, O.M.R., 2001. Assessing vegetation and land use changes in northeastern Puerto Rico:1978–1995. Caribbean Journal of Science, 37 (1–2), 95–106.

Gordon, L.J., Peterson, G.D., and Bennett, E., 2008. Agricultural modifications of hydrological flowscreate ecological surprises. Trends in Ecology and Evolution, 23 (4), 211–219.

Hahn, T., et al., 2006. Trust-building, knowledge generation and organizational innovations: therole of a bridging organization for adaptive comanagement of a wetland landscape aroundKristianstad, Sweden. Human Ecology, 34 (4), 573–592.

Holling, C.S., 1973. Resilience and stability of ecological systems. Annual Review of Ecology andSystematics, 4, 1–23.

Holling, C.S., 2001. Understanding the complexity of economic, ecological, and social systems.Ecosystems, 4 (5), 390–405.

Holling, C.S. and Meffe, G.K., 1996. Command and control and the pathology of natural resourcemanagement. Conservation Biology, 10 (2), 328–337.

Imbernon, J., 1999. A comparison of the driving forces behind deforestation in the Peruvian andBrazilian Amazon. Ambio, 28 (6), 509–513.

Issaka, R.N., et al., 1997. Geographical distribution of selected soil fertility parameters of inlandvalleys in West Africa. Geoderma, 75 (1–2), 99–116.

Levin, S.A. and Lubchenco, J., 2008. Resilience, robustness, and marine ecosystem-based manage-ment. BioScience, 58 (1), 1–6.

Ludwig, D., 2001. The era of management is over. Ecosystems, 4 (8), 758–764.Lugo, A.E. and Helmer, E.H., 2004. Emerging forests on abandoned land: Puerto Rico’s new forest.

Forest Ecology and Management, 190 (2–3), 145–161.Mainuddin, M., Kirby, M., and Chen, Y., 2009. Fisheries productivity and its contribution to overall

agricultural production in the Lower Mekong River Basin. Colombo: Challenge Program onWater and Food.


http://www.ecologyandsociety.org/vol10/iss1/art29

http://www.ecologyandsociety.org/vo11/iss1/art14


Martin, R. and Sunley, P., 2006. Path dependence and regional economic evolution. Journal ofEconomic Geography, 6 (4), 395–497.

Norberg, J. and G.S. Cumming, eds, 2008. Complexity theory for a sustainable future. New York:Columbia University Press.

Norberg, J., et al., 2008. Diversity and resilience of social-ecological systems. In: J. Norberg andG.S. Cumming, eds. Complexity theory for a sustainable future. New York: Columbia UniversityPress, 46–79.

Olsson, P., et al., 2007. Enhancing the fit through adaptive co-management: creating and maintainingbridging functions for matching scales in the Kristianstads Vattenrike biosphere reserve, Sweden[online]. Ecology and Society, 12 (1), 28. Available from: http://www.ecologyandsociety.org/vol12/iss21/art28/

Olsson, P., Folke, C., and Hahn, T., 2004. Social-ecological transformation for ecosystem manage-ment: the development of adaptive co-management of a wetland landscape in southern Sweden[online]. Ecology and Society, 9 (4), 2. Available from: http://www.ecologyandsociety.org/vol9/iss4/art2/ [Accessed 15 November 2010].

Ostrom, E., 2007. A diagnostic approach for going beyond panaceas. Proceedings of the NationalAcademy of Sciences of the United States of America, 104 (39), 15181–15187.

Ostrom, E., 2009. A general framework for analyzing sustainability of social-ecological systems.Science, 325 (5939), 419–422.

Patten, D.T., et al., 2001. A managed flood on the Colorado River: background, objectives, design,and implementation. Ecological Applications, 11 (3), 635–643.

Patz, J.A., et al., 2004. Unhealthy landscapes: policy recommendations on land use change andinfectious disease emergence. Environmental Health Perspectives, 112 (10), 1092–1098.

Polis, G.A., Anderson, W.B., and Holt, R.D., 1997. Toward an integration of landscape and foodweb ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecology andSystematics, 28, 289–316.

Polis, G.A., Power, M.E., and Huxel, G.R., eds, 2004. Food webs at the landscape level. Chicago:University of Chicago Press.

Pretty, J., 2008. Agricultural sustainability: concepts, principles and evidence. PhilosophicalTransactions of the Royal Society B-Biological Sciences, 363 (1491), 447–465.

Rodriguez, J.P., et al., 2006. Trade-offs across space, time, and ecosystem services. Ecology andSociety, 11 (1), 28.

Rosegrant, M.W., et al., 2000. Integrated economic–hydrologic water modeling at the basin scale:the Maipo river basin. Agricultural Economics, 24 (1), 33–46.

Scheffer, M., 2009. Critical transitions in nature and society. Princeton: Princeton University Press.Scheffer, M. and Carpenter, S.R., 2003. Catastrophic regime shifts in ecosystems: linking theory to

observation. Trends in Ecology and Evolution, 18 (12), 648–656.Senge, P.M., 1990. The fifth discipline: the art and practice of the learning organization. Appendix 2.

Currency. New York: Doubleday.Stirling, A., 2003. Risk, uncertainty and precaution: some instrumental implications from the

social sciences. In: M.L. Berkhout and I. Scoones, eds. Negotiating environmental change: newperspectives from social science. Cheltenham: Edward Elgar, 33–76.

Walker, B. and Salt, D., 2006. Resilience thinking: sustaining ecosystems and people in a changingworld. Washington, DC: Island Press.

Waltner-Toews, D., Kay, J.J., and Lister, N.-M.E., 2008. The ecosystem approach: uncertainty andmanaging for sustainability. New York: Columbia University Press.

Waltner-Toews, D., et al., 2003. Perspective changes everything: managing ecosystems from theinside out. Frontiers in Ecology and the Environment, 1 (1), 23–30.

WCED, 1987. Our common future. Report of the World Commission on Environment andDevelopment (WCED). Oxford: University Press, 43–66.

Yachi, S. and Loreau, M., 1999. Biodiversity and ecosystem productivity in a fluctuating envi-ronment: the insurance hypothesis. Proceedings of the National Academy of Sciences, 96 (4),1463–1468.

Yunnus, M., 1998. Alleviating poverty through technology. Science, 282 (5388), 409–410.Zwarts, L., et al., 2006. The economic and ecological effects of water management choices in the

upper Niger River: development of decision support methods. International Journal of WaterResources Development, 22 (1), 135–156.


http://www.ecologyandsociety.org/vol12/iss21/art28/




The resilience of big river basins

Documents

Transcript of The resilience of big river basins