Water, Food and Livelihoods in River Basins

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [International Water Management Institute]On: 5 February 2010Access details: Access Details: [subscription number 916748507]Publisher RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Water InternationalPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t792815876

Water, food and livelihoods in river basinsSimon E. Cook a; Myles J. Fisher b; Meike S. Andersson a; Jorge Rubiano c; Mark Giordano d

a Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia b Comidas Limitada (COMIL),Cali, Colombia c Universidad Nacional (UNAL), Palmira, Colombia d International Water ManagementInstitute (IWMI), Battaramulla, Sri Lanka

To cite this Article Cook, Simon E., Fisher, Myles J., Andersson, Meike S., Rubiano, Jorge and Giordano, Mark(2009)'Water, food and livelihoods in river basins', Water International, 34: 1, 13 — 29To link to this Article: DOI: 10.1080/02508060802673860URL: http://dx.doi.org/10.1080/02508060802673860

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

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

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

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

http://dx.doi.org/10.1080/02508060802673860

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

swickrama

Highlight

www.iwmi.cgiar.org

swickrama

Highlight

www.iwmi.cgiar.org

www.iwmi.cgiar.org

Water InternationalVol. 34, No. 1, March 2009, 13–29

ISSN 0250-8060 print/ISSN 1941-1707 online© 2009 International Water Resources AssociationDOI: 10.1080/02508060802673860http://www.informaworld.com

RWIN0250-80601941-1707Water International, Vol. 34, No. 1, December 2009: pp. 1–29Water InternationalWater, food and livelihoods in river basinsWater InternationalS.E. Cook et al.Simon E. Cooka*, Myles J. Fisherb, Meike S. Anderssona, Jorge Rubianoc

and Mark Giordanod

aCentro Internacional de Agricultura Tropical (CIAT), Cali, Colombia; bComidas Limitada (COMIL), Cali, Colombia; cUniversidad Nacional (UNAL), Palmira, Colombia; dInternational Water Management Institute (IWMI), Battaramulla, Sri Lanka

(Received 18 July 2008; final version received 5 December 2008)

Conflicting demands for food and water, exacerbated by increasing population,increase the risks of food insecurity, poverty and environmental damage in major riversystems. Agriculture remains the predominant water user, but the linkage betweenwater, agriculture and livelihoods is more complex than “water scarcity increasespoverty”. The response of both agricultural and non-agricultural systems to increasedpressure will affect livelihoods. Development will be constrained in closed basins ifincreased demand for irrigation deprives other users or if existing agricultural useconstrains non-agricultural activities and in open basins if agriculture cannot feedan expanding or changing population or if the river system loses capacity due todegradation or over-exploitation.

Keywords: poverty; agriculture; water poverty; water scarcity; livelihoods; closedbasins; stagnant agricultural production; degradation; over-exploitation

IntroductionWater supports livelihoods of all people through water-consuming agricultural and indus-trial activities, through consumption and sanitation and through environmental services.Evidence of a global water and food “crisis” is emerging (Seckler and Amarasinghe2000), in which increasing demands for food and water will lead to conflict and seriousloss of well-being for large numbers of people.

But what is the actual nature of the problem? The objectives of the CGIAR ChallengeProgram on Water and Food1 (CPWF) are to alleviate poverty, improve health and ensureenvironmental security, and hence the primary objective of the programme should be ananalysis of the evidence of linkages between water, agriculture and livelihoods. The analysismust do the following:

• Define the potential links between water, agricultural water use and livelihoods. It isessential to clarify this complex problem and to prevent a misunderstanding of thecorrelation between cause and effect.

*Corresponding author. Email: [email protected]

Downloaded By: [International Water Management Institute] At: 04:47 5 February 2010

swickrama

Highlight

www.iwmi.cgiar.org

www.iwmi.cgiar.org

14 S.E. Cook et al.

• Define processes that link water, agriculture and livelihoods in order tounderstand and quantify the precise nature of problems, as they occur withinbasins.

• Identify options for specific interventions, based on analysis of the opportunitiesand risks.

Our objective in this article is to report on the nature of these links, as they occur inseveral basins in the world. Data are presented from analyses in the CPWF basins ofMekong, Volta, São Francisco and Karkheh, and from initial insights gained in severalother basins.

In the first section we examine the inference that water scarcity caused by increasingagricultural demand for water leads to loss of livelihood, and why this belief may be basedon a false premise. In the second section we look for empirical evidence that water scar-city caused by agricultural demand leads to loss of livelihood. In the third section wereflect on the types of interventions that are intended to address the problems.

Water stress and poverty: a rhetorical concept?The global population is anticipated to increase to over 9 billion by the year 2050 (UNEP;www.unep.org/vitalwater/), placing major new demands on global food production. Forsome, this is a catastrophe waiting to happen. Predictions of catastrophe as a consequenceof population increase and resultant starvation extend back to the times of Malthus andbeyond, but by and large, they have not eventuated. Although history is peppered withtemporary or localized food shortages, the food production system has, over the long term,responded to demand. The Green Revolution is considered to be a successful example ofhow the food system can respond to demand. This concerted programme of agriculturaldevelopment was triggered in the 1960s in response to the threat posed by widespreadchronic food shortages. The programme was sufficiently successful that, contrary to thepredicted catastrophe, increasing demands for food over the past 40 years have been metby equal or greater increases in production, to the extent that food prices in real terms havedeclined (Evenson and Gollin 2003).

Increased food production has resulted in greater consumption of water. Agriculture isthe major user of fresh water globally, and as the food production system gears itself tomeet the demands of the increased population, it uses more water. This must be consid-ered against the demands from other sectors, including sanitation, waste disposal, industryand hydropower, which also increase with increasing population. Further, we now recog-nize the importance of maintaining river function invariably reduces the total water avail-able for all human uses. The overwhelming conclusion is that fresh water is likely tobecome an increasingly scarce resource globally.

If the demand for food is set to increase, but the volume of water available to agricul-ture is stable or decreasing, it follows, logically, that there must be major changes. Firstamongst these is that water productivity, that is, the benefit expressed as kilogram food,kilocalorie or income, per unit volume of water consumed, must increase (Molden et al.2001). There is evidence that the global system is already responding to some extent.Water productivity in some areas is beginning to increase, such that projections of futurewater withdrawals are being moderated from earlier worrying levels (Gleick 2003).Nevertheless, in other areas, evidence of over-exploitation, including increasing numbersof closed basins,2 reduction of groundwater resources, loss of wetlands and habitat frag-mentation, is also emerging.


Water International 15

What is the difference between stress and dynamic response to change?The global situation seems quite clear: demand for water for food production is increasing.However, additional withdrawal is not possible because of competing demands from othersectors and water productivity must therefore increase to enable more food to be producedfrom less water. Compilations of data at national scales provide a broad insight of varia-tions in water stress, expressed as per capita water availability or its equivalent, and rein-force the impression that an increasing number of countries might experience watershortages in the coming decades (Falkenmark et al. 1989, Gleick 2000). However, suchassessments give only a partial description of the impacts of increasing pressures uponwater as a result of growing demands for food. According to Molle and Mollinga (2003),simple measures of water scarcity can be misleading, for example, water availability is high inSenegal, Mali and Iran and low in countries such as Israel. They observe that measuresthat provide useful information of stress at the global scale are unreliable indicators of theproblem that eventuates on the ground. They proceed to examine a spectrum of increas-ingly complex indicators, only to conclude that the call for quantifiable measures of thiscomplex problem can risk “poor science” and make “smoke screens” (p. 542), which canbe used to serve political ends.

Analysis to improve our understanding of the effects of water on povertyThe common perception is that the water demand by agriculture is leading to water shortage,and depriving people of this basic livelihood support. The evidence we look at below paints amuch more complex picture in which people adapt to emerging opportunities and constraints.

We believe that the idea that water scarcity leads automatically to loss of livelihood isover-simplified. Kemp-Benedict (2008) analysed the links between livelihood assets andwell-being in the Mekong Basin to show the partial correlations with other povertymeasures. Other detailed case studies show that the link between agricultural water man-agement and poverty can be extremely variable and complex (Castillo et al. 2007).Although all such insights are useful, the challenge is to bridge the gap between broadtrends in basin systems and detailed insights from case studies, which, no matter howvalid, do not easily relate to the global problem.

To simplify the water and food system in a form that can be reviewed systematically,we consider four scenarios in which the food and water crisis occurs on the ground.Scenarios A and B typify situations in which agricultural water demand causes problemsof water scarcity. Scenarios C and D focus more on situations in which the problems arethat insufficient benefits accrue from the use of water by agriculture:

• Scenario A. Increased demand for food encourages irrigators to use more water atthe direct expense of other water users. Losses may be incurred by fellow irrigators,downstream abstractors dependent on return flows or environmental flows and thepeople who depend on them. During the period 1950–2000, the global irrigated areadoubled from approximately 140 million hectares (Mha) to about 280 Mha (Molleet al. 2007), leading to increasing incidence of river basin closures (Falkenmark andMolden 2008) with consequent water scarcity for other users and loss of ecosystemservices.

• Scenario B. Water use by agriculture (including so-called virtual water) remainsmore or less constant, but the volume committed to agriculture constrains the supplyto other water users as they respond to increasing demands. These demands may be


16 S.E. Cook et al.

for water for hydropower to meet an increasing demand for energy, industrial use orsanitation, protection of aquatic ecosystems (e.g., restoration of native salmon runsin the US Pacific Northwest) or preserving environmental flows. The consequencesare indirect, and are likely to be seen as foregone income. In fact, domestic andindustrial users are given priority because of which agriculture is forced to expandinto formerly natural areas to maintain its inflows and thus reduce the environmen-tal benefits provided by these natural areas (Molle et al. 2007).

• Scenario C. Increasing population places demands for food that cannot be satis-fied by rainfed production. Water use by agriculture is not affected substantially, butfood security is reduced in the face of increasing demand as a consequence of adecrease in per capita production. Castillo et al. (2007) describe some of the diffi-culties faced in the attempt to increase water productivity.

• Scenario D. Water productivity is reduced by over-exploitation, degradation orpollution. The most obvious case of this is the loss of fishery production as a resultof over-exploitation under increased demand of water. The effect is reduced foodsecurity or income (Dugan et al. 2007).

Scenario A describes a process in which consumption by large irrigators deprives otherpotential users from benefiting from the water they use. The argument seems a straightfor-ward case of inequitable use of water rights, in which one group of actors reduces the avail-ability of water to others without their agreement. Such loss of a collective resource isproposed by van Koppen et al. (2007) as a cause of local poverty and a reason to promotemultiple water use systems that can lift those out of poverty who otherwise would lackaccess to water (Castillo et al. 2007). Molle et al. (2007) identify political and economicreasons why irrigation over-development occurs to the point of basin closure.

There are many examples of this situation, the most spectacular of which is outside theCPWF, namely the demise of the Aral Sea as a result of irrigation of large-scale cottonfarms, initiated during the Soviet era in Central Asia.

The Aral Sea was the fourth-largest lake in the world with a surface area of55,100 km2 (Figure 1). In the 1960s the rivers that fed it (the Amu Darya and the SyrDarya) were diverted to provide water for irrigated cotton in Soviet Central Asia. By1987, about 60% of its volume had been lost, its depth had decreased by 14 m and its saltconcentration had doubled, killing the fish on which commercial fishing depended. Windstorms became toxic, carrying dust and salts deposited on the exposed sea floor. Lifeexpectancy of people living near the area too became significantly lower than in thesurrounding areas. The Aral Sea has now been divided into two water bodies. A damacross the southern part of the northern Aral Sea seems to be maintaining its level, but thesouthern Aral Sea is forecast to disappear within 15 years.

The irrigation schemes themselves may be sustainable, although we have no data tosubstantiate this statement. Regardless, the consequences for the Aral Sea and the popula-tions dependent on it were disastrous (Figure 2).

In the São Francisco CPWF basin in Brazil, irrigated agriculture has become morepredominant over the past 15–20 years (Bassoi et al. 2006). Analysis of flows shows thepotential impact of water abstraction by large farmers in the São Francisco on their poorersmallholder neighbours (Manetta et al., this issue). There is a wide disparity betweenwealthier farmers, who are able to respond to supermarketization, and poorer smallholderfarmers, who cannot (Torres et al. 2007). A scheme to divert more of the waters of the SãoFrancisco to support farmers in the drier northern parts of the basin has been proposed offand on for years, but was shelved after a severe drought in 2001. It has recently received



support at the highest level of government, but is meeting substantial resistance from thebroader community.

In the Mekong CPWF basin in Southeast Asia, living standards have generally improvedacross the basin, but important areas of poverty remain. Owing to population growth and increas-ing economic development, the pressure on the natural resource base, particularly waterresources, has increased in recent decades (MRC 2003, Foran 2008, Mainuddin et al. 2008). Sev-eral instances are reported in which plans to increase irrigation schemes are expected to threatencritical environmental flows and to disrupt fisheries in Thailand and Cambodia (see below). Lossof livelihood assets seems related to hotspots of poverty in the basin (Kemp-Benedict 2008).

A well-publicized victory for pressure groups opposing this has been the raising of the PakMun barrier (Figure 3). The Pak Mun dam was constructed by the Electricity GeneratingAuthority of Thailand (EGAT) in 1991–1994 across the Mun River, a tributary of the Mekong.The total cost of construction was US$233 million of which the World Bank financed US$22million (SEARIN 2004a). The stated purpose of the dam was to provide scope for develop-ment of villagers living along the Mun River and for the development of Thailand as a wholeby generating electricity. However, the dam generated just 0.1% of electricity in Thailand anddestroyed the fisheries on which the Mun River villagers’ livelihoods depended by preventingupstream migration of fish from the Mekong. In 2001, bowing to intense pressure, the Thaigovernment opened the sluice gates of the dam. The fisheries immediately began to recover,and step by step, the flooded lands also started to recover. The sluice gates are now opened forfour months each year and villagers’ livelihoods have improved (SEARIN 2004b), althoughprobably not to pre-dam levels.

In the Karkheh CPWF Basin in Iran, the area of the transboundary wetland Al-Hawizeh inIran and Iraq has declined by two-thirds to 1025 km2 between 1973–1976 and 2000

Figure 1. The location of the Aral Sea in Central Asia.


18 S.E. Cook et al.

Figu

re 2

.Th

e A

ral S

ea in

197

3 an

d 20

04.

Sour

ce: U

nite

d N

atio

ns E

nviro

nmen

t Pro

gram

me.

Ava

ilabl

e fr

om: h

ttp://

na.u

nep.

net/d

igita

l_at

las2

/web

atla

s.php

?id=

11 [A

cces

sed

1 D

ecem

ber 2

008]



(Figure 4). The Iranian part alone (called Hawr Al-Azim) was reduced by 46% over thisperiod. Hawr Al-Azim now accounts for 21% of the total area of the Al-Hawizeh wetland.The shoreline of Hawr Al-Hawizeh/Al-Azim has been in steady retreat during the past dec-ade and closure of the Karkheh Basin is envisaged by the early part of this century. Irrigationis allocated the largest share and by 2025 it is expected to reach a volume roughly equal tothe renewable water supply in an average year (Masih et al. 2008). Unless urgent remedialaction is taken, destruction of the Mesopotamian marshlands is likely to continue unabated.Indeed, it is likely to accelerate as a result of substantial water retention by the Karkheh Damand plans to transfer water from its reservoir to Kuwait. Analysis of flows in the Karkheh inIran indicates that the abstraction of irrigation water in the lower reaches of the river divertsflows that otherwise would support the recovery of the Hawr Al-Azim wetland.

In the Volta CPWF basin in West Africa, uncontrolled expansion of small reservoirs foragricultural development in the Upper Volta is causing some concern about inflows to LakeVolta, which provides a critical source of income for Ghana by generating hydropower. Unlikein the case of the Karkheh Basin, initial hydrology analysis by Lemoalle (2008) using the WaterEvaluation and Planning System (WEAP, SEI 2008) suggests that adverse effects are unlikely.

How loss of water availability causes poverty in specific groups needs to be evaluatedcarefully. First, analyses cited in Molle et al. (2007) show that abstraction into an irrigationsystem does not necessarily remove it from further use. Return flows from “leaky” irrigationcan provide valuable local sources of recharge for downstream users. Second, the critical

Figure 3. Location of the Pak Mun dam and the Tonlé Sap (Grand Lac) in the Lower Mekong Basin.Source: Available from: http://www.dams.org/images/maps/map_mekong.htm [Accessed 1 December 2008]


20 S.E. Cook et al.

Figu

re 4

.Th

e ex

tent

of t

he tr

ansb

ound

ary

Al-H

awiz

eh m

arsh

es in

197

2–19

76 (l

eft)

and

in 2

000

(righ

t). T

he K

arkh

eh R

iver

ent

ers t

he m

arsh

es in

thei

r nor

thea

st.

Sour

ce: U

nite

d N

atio

ns E

nviro

nmen

t Pro

gram

me.

Ava

ilabl

e fr

om: h

ttp://

na.u

nep.

net/p

ublic

atio

ns/m

eso.

pdf [

Acc

esse

d 1

Dec

embe

r 200

8].



parameters are not just the distribution of water but also the relative productivities of thewater so used. If the water consumed by irrigation can benefit other sectors, this can lead to acommon gain, effectively substituting food security by income security. For example, a majorreport (ANA et al. 2004, p. 264) indicates problems in the lower São Francisco Basin and itscoastal zone caused by alterations in basin hydrology, but it is difficult to evaluate these prob-lems without referring to the major reductions in poverty that have accompanied thesechanges (Torres et al. 2007). By analysing food poverty in Ecuador, Farrow et al. (2005)determined that the relationship of irrigation with food security could be both positive andnegative in different areas, depending on whether irrigation also provided salaried employ-ment. The potential benefit from irrigation depends on the degree of connectivity with higher-value markets in which the irrigated product is sold.

Scenario B is when existing water use by agriculture, as the largest consumer of water,constrains the development of non-rural uses that would otherwise improve livelihoods.Basin closure is already estimated to affect 1.4 billion people worldwide (Falkenmark andMolden 2008), which in turn hinders development. Expansion of much-needed hydro-power to support non-rural development may be constrained by an absolute scarcity ofwater. A hydropower scheme has been suspended in Uganda for this reason. Water resourcesfor several large cities in Latin America are said to be under pressure as a result of prior landuse change in upper catchments (Tognetti and Johnson 2008). For example, Bogotá, Colom-bia, with its satellite cities of Chía, Cota, Soacha, Cajicá and La Calera, had an estimated popu-lation of 8.24 million as of 2007 (http://en.wikipedia.org/wiki/Bogot%C3%A1). Bogotá islocated on a high plateau in the Eastern Cordillera of the Andes at an altitude of 2640 m, whichmakes it the third-highest major city in the world after La Paz and Quito. Land use in the lim-ited area of lands higher than the city is critically important for the supply of potable water tothe city. Run-off, leaching and soil erosion in the surrounding highlands from arable farms,principally producing potatoes, and intensive cattle ranching pollute rivers with N, P, solid ani-mal waste and other sediments, which limits their use as sources of potable water unless thewater undergoes expensive treatment (Tognetti and Johnson 2008).

Solutions are being explored to improve the provision of water through Payment forEcosystem Services schemes, in which agriculturalists relinquish their demands for waterin return for financial reward. Analysis of the development foregone because of existingpatterns of water use by agriculture is difficult and can only be understood by taking intoaccount the development trajectory.

Scenario C is one in which food security and incomes are constrained by an inabilityof rainfed agriculture to respond to demand. Population increases, but total productiondoes not respond adequately and water productivity remains static. This is a water-for-food issue because water productivity is not increasing and hence not contributingtowards the solution outlined in the first section of this article.

This condition appears in the Volta Basin (Terrasson et al. 2009), but also in many otherparts of sub-Saharan Africa where per capita food production has remained stagnant over thepast 40 years (Bationo et al. 2007). In the same zone, areas currently affected by changes inrainfall patterns and length of the vegetation season owing to climate change are anotherexample (SWAC/ECOWAS 2006). Similar conditions also occur in localized areas withinbasins that are otherwise responding strongly to demands for food, including in northernparts of the São Francisco Basin. Urban populations in Belo Horizonte and Brasilia and itssurrounding satellite cities are increasing both because of increasing economic activity andbecause of migration of the rural poor to the cities. In the Mekong, productivity of rainfedrice in Cambodia and Northeast Thailand per capita is substantially stagnant (Mainuddin etal. 2008, Kirby and Mainuddin 2009; Figure 5).


22 S.E. Cook et al.

The final scenario, D, is where livelihoods are threatened by a loss of water productivitythat is caused by degradation of land or water resources. This may result from pollution, over-exploitation or land degradation (e.g., salinization). There is no change in the water balancebut a loss of water productivity and system resilience to pressure. This process is feared tothreaten the livelihoods of over a million people living in and around the Tonlé Sap in theMekong (Figure 3). Tonlé Sap is a lake with a dry season area of 2700 km2 and a depth ofabout 1 m, connected to and approximately on the same level as the Mekong (Figure 6).

Figure 5. Annual rice production per capita in the Lower Mekong Basin.Source: Kirby and Mainuddin 2009.

0

200

400

600

800

1000

1200

1990 1995 2000 2005Year

Ric

e pr

oduc

tion,

kg/

capi

ta Laos

Thailand

Cambodia

Vietnam

Vietnam CentralhighlandsVietnam MekongRiver Delta

Figure 6. The Tonlé Sap and its flood plain in Cambodia.Source: Available from: http://en.wikipedia.org/wiki/Tonl%C3%A9_Sap



During the wet season, flood waters from the Mekong back up into the Tonlé Sap, whoselevel can rise to as much as 9 m and can flood over 16,000 km2. As the levels of theMekong fall during the dry season, the accumulated water drains from the Tonlé Sap backinto the Mekong. This seasonal flow and ebb provides a surge of nutrient flows and veryfavourable conditions for fish, making the Tonlé Sap one of the most productive inlandfisheries in the world, supporting over 3 million people and providing over 75% ofCambodia’s annual inland fish catch and 60% of the Cambodians’ protein intake. Anydegradation that affects this surge and flow is likely to have a major impact on livelihoodsin the region.

Large urban areas in the upper part of catchments can negatively affect downstreamwater users (both agricultural and non-agricultural) by pollution. In the São Franciscoriver basin in Brazil, the large city of Belo Horizonte with a population of 4.98 million inthe metropolitan region in 2007 (http://en.wikipedia.org/wiki/Belo_Horizonte) is in theupper parts of the basin (Figure 7). The pollution that it causes creates problems for down-stream users including fishers, who have become increasingly marginalized.

Land degradation caused by overgrazing of common land in the Upper KarkhehBasin is of concern, although data from Ahmad et al. (2008) suggest that where farmerscan connect to markets, higher-value animal products can actually increase water produc-tivity, expressed as product gross value per cubic metre of water consumed. Degradationof groundwater by over-exploitation threatens the livelihoods of farmers in the Ganges(Rosegrant et al. 2002) and in the Mekong delta. Point- and non-point-source pollutionhave reduced water quality in the upper parts of the São Francisco River (ANA et al.2004).

Figure 7. The São Francisco Basin, Brazil. Belo Horizonte is the capital of Minas Gerais state(MG on the map).Source: http://www.transportes.gov.br/bit/mapas/mapclick/hidro/bcsfran.htm.


24 S.E. Cook et al.

Pro-poor interventions in agricultural water usePathways for improvements in livelihoodsIn this section we examine the way in which changes in agricultural water use influencelivelihoods. Two general pathways can be distinguished (Figure 8). The first pathway,improving water availability, concerns the benefit from allocating water to individuals orgroups. It addresses scenarios A and B in which the predominant factor is water distribution.The aim is to improve the total net gain from a volume of water, shared amongst manypeople. Some water users may lose access, but the overall trend is an increase in thenumber of people who gain. Here, collective action and social control are important assetsif inequitable concentration of benefits is to be avoided.

If people’s livelihoods are compromised as a result of lack of water, or lack of accessto it, improvement may be brought about through changes in water governance (e.g., Dore2008). Provision of water for direct consumption or irrigation is a simple intervention thatgovernments have adopted widely as a means of supporting people, with mixed success.For example, access to irrigation has in many cases resulted in an increase in employmentopportunities and a decrease in food prices (Stockle 2001). But gain is not guaranteed. Ifone group improves at the cost of another, there is no net gain. This is especially applic-able when the irrigated area is expanded, which may benefit only wealthier farmers, thusresulting in neutral or even negative effects on poverty as has been the case in Bangladesh(van Koppen and Mahmad 1996).

Therefore the desired intervention, in terms of improving total well-being, is to allo-cate water to ensure net gain. This is the intention of promoting multi-use systems. Forexample, in the Coello Basin in the Colombian Andes, poor water quality and conflicts inwater use between farmers irrigating rice and city dwellers forced the development of aplan for multi-purpose benefits, including sanitation and industrial uses (Vanegas Gálvez2002). Similarly, in a study for improving water sources for drinking and domestic uses inrural Tanzania, it was found that much more water was used for other uses (bathing,laundry, livestock and cleaning) than for drinking and cooking alone. On the contrary,

Figure 8. Pathways to improving livelihoods through changes in agricultural water use.



rural communities often did not abandon their traditional sources of water even when theyhad access to improved water sources (Madulu 2003). Bastakoti et al. (2008) noted theneed to network beyond conventional water institutions to achieve multi-use systems.

The second pathway to improve water productivity is to increase the per capita gain foreach cubic metre of water consumed. This addresses the problems represented by scenariosC and D. In this case, change appears as improvement in the agricultural system, measura-ble as increased kilogram per cubic metre or dollar per cubic metre (kg/m3 or $/m3).

Options to improve livelihood support from agricultural water useScenario AThe first scenario describes the situation in which increasing volumes of water are allo-cated to agriculture. The option is to cut back demand for water from the agricultural sec-tor to a level that reflects its ability to support livelihoods. In this sense, high-productivityor high-value activities are preferable as they support more people through food or incomesupport. Many examples of processes that encourage this exist, including the following:

• Water pricing policies to encourage high-value usage.• Water allocation policies to ensure provision of basic services.• Development of catchment groups to enhance communication between different

users (e.g., Curtis and Lockwood 2000).• Transboundary agreements to reduce risks of distrust and conflict (Lautze et al. 2005).

Scenario BThe second scenario describes a variation of scenario A in which water that is alreadycommitted to agriculture is required by other sectors to support development of non-agricultural activities. Some possibilities are as follows:

• Basin transfers of water from low- to high-value users (Wester 2008).• Development of payment for ecosystem services (PES) to promote the provision of

adequate clean water (Swallow et al. 2007).• Removal of exotic plants with high water consumption but low water productivity,

such as the work-for-food programme in South Africa to reduce the demand forwater by eucalypts and pines.

Scenario CThe third scenario is one in which improvements in rainfed water productivity are soughtto balance food production with water consumption. The changes that enable improvedrainfed water productivity include an almost infinite array of technological, policy and fin-ancial innovations intended to correct problems of low and declining soil fertility, envi-ronmental degradation and limited access to viable markets by farmers (Mapfumo 2007).

• Technological interventions would include new germplasm, improved fertilizer tar-geting and conservation tillage packages.

• Policy interventions would include those to support investments to crop, livestockand fishery systems and policy support for agricultural research.


26 S.E. Cook et al.

• Financial interventions might include insurance to protect from climatic risks,subsidies for fertilizer or other inputs and improved supply chain management.

Scenario DThe fourth scenario calls for protection of the quantity and quality of environmentalservices. Some components of this solution are as follows:

• Full valuation of environmental services. This would include the value of waterflowing through food systems, through to the value of carbon sequestration orbiodiversity.

• Mitigating the risk of land degradation caused by erosion, salinization, acidificationor other damaging process.

• Recognizing through policies the rights and responsibilities of all users of naturalresources for improving the productivity and resilience of agricultural systems.

ConclusionsThe purpose of analysing poverty with respect to water use is threefold: first, to definethe links between water, agricultural water use and poverty; second, to understand andquantify the precise nature of problems, as they occur within basins; and, third, to iden-tify opportunities for interventions, based on analysis of the opportunities and risks.

An analysis of four river basins, plus a preliminary analysis of six other basins, highlightthe linkages between water and food systems. Demand for food is increasing, and, withrespect to the water it consumes, additional demand for food may be met in one of twoways: First demand may be met without any change in the overall water balance, if waterproductivity increases. Alternatively, the increasing demand for food will increase wateruse by agriculture at the same time as the demand for water by other sectors is alsoincreasing. Together these will place increased stress on the hydrologic system.

The effect of water stress on livelihoods, hence poverty, can be felt in a variety ofways. Additional water can be withdrawn, hence made unavailable for other demands.Basins may already be closed, constraining the capacity of alternative activities.

The water side of the equation is linked to the food side and three things may occurhere: The best is that agricultural water productivity increases to meet food demand with-out increasing water supply. But other effects may also occur. For example, water produc-tivity may remain static, or, worse, water productivity may decline if land or waterresources degrade.

In general, the loss of livelihood assets that occurs if water or food systems do notadapt to change can lead people into poverty. This seems especially so for the poor whodepend on natural resources for their livelihood. However, whether this actually occurs,and what form it takes, depends on many other factors.

Then why is this a concern, if the systems adapt to change? The increased demand forwater and food will not always be met, and this will place stress on the agriculturalsystems on which people depend. Some people, and we expect these to be the poorest, willsuffer as a result of the reduced availability of water or food. Understanding how andwhere this happens is essential to meet the original goals of the CPWF to alleviate povertyand increase food and environmental security.

Given the range of conditions that occur in basins, the relationship between water,agriculture and poverty seems unlikely to be general. Although there are some cases in



which water scarcity is reported to cause poverty, people with less available water are notnecessarily poorer than those with abundant water. In fact, some of the poorest people live inwater-rich areas (the Andean hillsides for example and especially Bangladesh where the poorsuffer disproportionately from flooding), and some of the richest live in water-scarce regions,such as Israel. Factors other than water add to the complexity, for example, land tenure, andare often a greater problem (Bugri 2008). For example, on the Lake Volta hinterland, localtribal chiefs, who control land tenure, preferentially grant tenure to members of their owntribes. Immigrant fisher folk are then unable to diversify their livelihoods and are thus unableto escape from poverty (Béné and Friend 2009). Other complicating factors include politicalinstability and lack of institutional support for proper water management.

Water scarcity is not equivalent to water unavailability. Local institutes and normsstrongly influence the access to water of different groups of people. The effect of thisdiscrepancy on overall livelihood support depends on the effectiveness with which thedifferent groups use water.

Water productivity describes the conversion of water consumed by agriculture intolivelihood support, for example, the amount of grain produced for each cubic metre ofwater consumed. Estimates suggest that there is a vast potential for improvement in thisconversion. For example, the potential water productivity of wheat is approximately 2 kg/m3,but it is rare to find systems with a conversion better than 0.5 or 0.6 kg/m3, and manyareas have productivities of a fraction of that. The story seems to be repeated for otherstaple foods such as maize, rice, sorghum or millet.

The realization of potential water productivity needs to take into account the complex link-ages that couple different components of agricultural systems, because it is these linkages thatdetermine their full value to livelihood support. Factors such as connectivity with markets orfinance and the institutional coherence with which water and land resources are shared seem tohave a major influence on performance. Unfortunately, these factors are difficult to measurebut need to be considered in relation to the overall development trajectory affecting agriculture.

AcknowledgementsThis research project is sponsored by the Challenge Program on Water and Food (CPWF) of theConsultative Group on International Agricultural Research (CGIAR), in collaboration with theInternational Water Management Institute (IWMI), whose support we gratefully acknowledge. MFthanks Comidas Limitada COMIL and its managing partner, Pamela May Clausen, for continuedsupport.

Notes1. http://www.waterandfood.org/2. Basin closure refers to the situation when all the water resources within a basin are allocated

(Falkenmark and Molden 2008).

ReferencesAhmad, M.D., et al., 2008. Mapping basin level water productivity using remote sensing and

secondary data in Karkheh river basin Iran. Paper for the World Water Congress, Montpellier.September, 2008.

ANA, GEF, UNEP, OAS, 2004. Integrated management project for land-based activities in the SãoFrancisco river basin: Strategic Action Program for integrated management of the SãoFrancisco river basin and its coastal zone – SAP. GEF São Francisco: Final Report. Brasilia.

Bassoi, L.H., et al., 2006. Water management across scales in the São Francisco Basin: Policy optionsand poverty consequences [online]. Preliminary Report of the Basin Focal Project. Available from:


swickrama

Highlight

www.iwmi.cgiar.org

28 S.E. Cook et al.

http://www.waterandfood.org/fileadmin/CPWF_Documents/Documents/Basin_Focal_Projects/BFP_Progress_Reports/Sao_Francisco_Basin_BFP_Proposal_v11.pdf [Accessed 1 December2008].

Bastakoti, R.C., et al., 2008. Collective action in water management through regional networks:critical reflections from M-Power experiences. 2nd International Forum on Water and Food.Addis Ababa, Ethiopia: CGIAR Challenge Program on Water and Food (CPWF).

Bationo, A., et al., eds., 2007. Advances in integrated soil fertility management in sub-SaharanAfrica: challenges and opportunities. Dordrecht: Springer.

Béné, C. and Friend, R., 2009. Water, poverty and inland fisheries: some lessons from Africa andAsia. Water International, 34 (1), 47–61.

Bugri, J.T., 2008. The dynamics of tenure security, agricultural production and environmentaldegradation in Africa: evidence from stakeholders in north-east Ghana. Land Use Policy, 25 (2),271–285.

Castillo, G.E., et al., 2007. Reversing the flow: agricultural water management pathways for povertyreduction. In: D. Molden, ed. Water for Food, Water for Life. Colombo and London: IWMI andEarthscan, 149–191.

Curtis, A. and Lockwood, M., 2000. Landcare and catchment management in Australia:lessons for state-sponsored community participation. Society and Natural Resources, 13 (1),61–73.

Dore, J., 2008. A theory and practice of deliberative water politics for the Mekong Region. A theoryand practice of deliberative water politics for the Mekong Region. 2nd International Forum onWater and Food. Addis Ababa, Ethiopia: CGIAR Challenge Program on Water and Food(CPWF).

Dugan, P., et al., 2007. Inland fisheries and aquaculture. In: D. Molden, ed. Water for Food, Waterfor Life. Colombo and London: IWMI and Earthscan, 459–483.

Evenson, R.E. and Gollin, D., 2003. Assessing the impact of the green revolution: 1960–2000.Science, 300 (5620), 758–762.

Falkenmark, M., Lundqvist, J., and Widstrand, C., 1989. Macro-scale water scarcity requires micro-scale approaches: aspects of vulnerability in semi-arid development. Natural Resources Forum,13 (4), 258–267.

Falkenmark, M. and Molden, D., 2008. Wake up to realities of river basin closure. Water ResourcesDevelopment, 24 (2), 201–215.

Farrow, A., et al., 2005. Exploring the spatial variation of food poverty in Ecuador. Food Policy, 30(5–6), 510–531.

Foran, T., 2008. Collective approaches to fish production. Fisheries and livelihoods in disputearound Pak Mun Dam, Thailand. 2nd International Forum on Water and Food. Addis Ababa,Ethiopia: CGIAR Challenge Program on Water and Food (CPWF).

Gleick, P.H., 2000. The world’s water 2000–2001: the biennial report on freshwater resources.Washington, DC: Island Press.

Gleick, P.H., 2003. Global freshwater resources: soft-path solutions for the 21st century. Science,302 (5650), 1524–1528.

Kemp-Benedict, E., 2008. Impacts of water-related interventions on the livelihoods of the poor:Mekong-Example. Presented at: Basin Focal Project: Review and Planning Meeting, Cali,Colombia, 1–5 February 2008.

Kirby, M. and Mainuddin, M., 2009. Water and agricultural productivity in the lower MekongBasin: trends and future prospects. Water International, 34 (1), 134–143.

Lautze, J., Giordano, M., and Borghese, M., 2005. Driving forces behind African transboundarywater law: internal, external, and implications. In: B. van Koppen, J. Butterworth, and I. Juma,eds. African water laws: plural legislative frameworks for rural water management in Africa:an international workshop. 26–28 January 2005, Johannesburg, South Africa, 26/1–13.

Lemoalle, J., 2008. BFP Volta Synthesis V2, March 2008. Unpublished draft.Madulu, N.F., 2003. Linking poverty levels to water resource use and conflicts in rural Tanzania.

Physics and Chemistry of the Earth, 28 (20–27), 911–917.Mainuddin, M., Kirby, M., and Chen, Y., 2008. Spatial and temporal pattern of land and water pro-

ductivity in the Lower Mekong river basin. Basin Focal Project Working Paper No. 5. Colombo,Sri Lanka: Challenge Program on Water and Food (CPWF).

Mapfumo, P., 2007. Enhancing regional capacity for a Green Revolution in Southern Africa:SOFECSA’s strategic direction and experiences. In: A. Bationo, J.M. Okeyo, B.S. Waswa, P.


http://www.waterandfood.org/fileadmin/CPWF_Documents/Documents/Basin_Focal_Projects/BFP_Progress_Reports/Sao_Francisco_Basin_BFP_Proposal_v11.pdf

http://www.waterandfood.org/fileadmin/CPWF_Documents/Documents/Basin_Focal_Projects/BFP_Progress_Reports/Sao_Francisco_Basin_BFP_Proposal_v11.pdf


Mapfumo, F. Maina, and J. Kihara, eds. Innovations as key to the Green Revolution in Africa:Exploring the scientific facts in Arusha, Tanzania [online]. African Network for Soil Biologyand Fertility (AfNet), 15. Available from: http://www.ciat.cgiar.org/tsbf_institute/index.htm[Accessed 1 December 2008].

Masih, I., et al., 2008. Overview of streamflow variability and water accounts for the KarkhehBasin, Iran. 2nd International Forum on Water and Food. Addis Ababa, Ethiopia: CGIAR Chal-lenge Program on Water and Food (CPWF).

Molden, D., et al., 2001. Increasing productivity of water: a requirement for food and environmen-tal security. Dialogue on Water, Food and Environment, Working Paper No. 1. Colombo,Sri Lanka: International Water Management Institute.

Molle, F. and Mollinga, P., 2003. Water poverty indicators: conceptual problems and policy issues.Water Policy, 5 (5/6), 529–544.

Molle, F., et al., 2007. River basin development and management. In: D. Molden, ed. Water forfood, water for life. Colombo and London: IWMI and Earthscan, 585–620.

MRC (Mekong River Commission), 2003. State of the basin report 2003. Phnom Penh, Cambodia: MRC.Rosegrant, M.W., Cai, X., and Cline, S.A., 2002. World water and food to 2025: dealing with scarcity.

Washington, DC: International Food Policy Research Institute.SEARIN (South East Asia Rivers Network), 2004a. Downstream impacts of hydropower and devel-

opment of an international river, a case study of Lancang-Mekong [online]. Chiang Mai, Thailand:SEARIN. Available from: http://www.livingriversiam.org/mk/mek_down_impact_en.pdf[Accessed 1 December 2008].

SEARIN (South East Asia Rivers Network), 2004b. The return of fish, river ecology and local livelihoodsof the Mun River: A Thai Baan (Villagers’) Research. Chiang Mai, Thailand: SEARIN, 32 pp.

Seckler, D. and Amarasinghe, U., ed., 2000. Water supply and demand, 1995 to 2025: water scarcityand major issues. In: World water supply and demand [online]. Colombo, Sri Lanka: Interna-tional Water Management Institute. Available from: http://www.lk.iwmi.org/pubs/WWVisn/WWSDHtml.htm [Accessed 1 December 2008].

SEI (Stockholm Environment Institute), 2008. Water Evaluation and Planning System (WEAP) [online].Boston: SEI. Available from: http://www.weap21.org/index.asp [Accessed 1 December 2008].

Stockle, C., 2001. Environmental impact of irrigation: a review. Pullman, Washington: WashingtonState University.

SWAC/ECOWAS, 2006. The ecological vulnerable zone of the Sahelian countries. In: SWAC/ECOWAS, ed. The web atlas on regional integration in West Africa [online]. Abuja, Nigeria:ECOWAS-SWAC/OECD. Available from: http://www.atlas-ouestafrique.org/spip.php?article55 [Accessed 1 December 2008].

Swallow, B., et al., 2007. Compensation and rewards for environmental services in the developingworld: framing pan-tropical analysis and comparison. ICRAF Working Paper No. 32. Nairobi:World Agroforestry Centre.

Terrasson, I., et al., 2009. Yields and water productivity of rainfed grain crops in the Volta basin,West Africa. Water International, 34 (1), 104–118.

Tognetti, S.S. and Johnson, N., 2008. Ecosystem services from improved soil and water manage-ment: creating a return flow from their multiple benefits. Working Paper. Colombo, Sri Lanka:Challenge Program on Water and Food (CPWF).

Torres, M., et al., 2007. Spatial patterns of rural poverty in the São Francisco river basin, Brazil.Selected paper prepared for the American Agricultural Economics Association Annual Meeting,Portland, Oregon, 29 July–1 August 1, 2007.

van Koppen, B. and Mahmad, S., 1996. Women and water pumps in Bangladesh. The impact of parti-cipation in irrigation groups on women’s status. London: Intermediate Technology Publication.

van Koppen, B., et al., 2007. The failure of formal water rights in Tanzania: Grafting modern waterpricing on colonial water use registration. In: F. Molle, R. Barker, and J. Berkoff, eds. Betweenprescription and practice: irrigation water pricing policies in context. Colombo, Sri Lanka andWallingford, UK: IWMI and CABI.

Vanegas Gálvez, M., 2002. Sistemas integrados de tratamiento y uso de aguas residuales enAmérica Latina: realidad y potencial. Estudio complementario del caso Ibague, Colombia[online]. Convenio IDRC – OPS/HEP/CEPIS. Available from: http://www.cepis.ops-oms.org/bvsaar/e/proyecto/complemen/casos/ibague.pdf [Accessed 1 December 2008].

Wester, P., 2008. Shedding the waters: Institutional change and water control in the Lerma-Chapala Basin, Mexico. Wageningen: Wageningen University.


http://www.lk.iwmi.org/pubs/WWVisn/WWSDHtml.htm

http://www.lk.iwmi.org/pubs/WWVisn/WWSDHtml.htm

http://www.atlas-ouestafrique.org/spip.php?article55

http://www.atlas-ouestafrique.org/spip.php?article55

http://www.cepis.ops-oms.org/bvsaar/e/proyecto/complemen/casos/ibague.pdf

http://www.cepis.ops-oms.org/bvsaar/e/proyecto/complemen/casos/ibague.pdf