The resilience of big river basins
-
Upload
jamescookuniversity -
Category
Documents
-
view
0 -
download
0
Transcript of 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.
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
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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)
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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)
.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 67
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
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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
.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 69
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
.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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).
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 71
(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
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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
.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 73
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)
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 75
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 77
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
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 79
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 81
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);
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 83
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)
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 85
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 87
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)
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 89
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 91
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)
.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 93
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].
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011
Water International 95
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.
Downloaded By: [Stockholm University Library (Stockholms Universitet)] At: 14:23 8 February 2011