WUP-FIN SOCIO-ECONOMIC STUDIES ON TONLE SAP 10

WUP-FIN POLICY MODEL Finding Ways to Economic Growth, Poverty Reduction

and Sustainable Environment

MRCS/WUP-FIN KMerageFVIm:UEdl Tenøsab Water Utilization Program – Modelling the Flow Regime and Water Quality of the Tonle Sap

Finnish Environment Institute Consultancy Consortium

March 2003 mIna 2003

ii

Varis, Olli 2003. WUP-FIN Policy Model – Finding Ways to Economic Growth, Poverty Reduction and Sustainable Environment, WUP-FIN Socio-economic Studies on Tonle Sap 10, MRCS/WUP-FIN, Phnom Penh. Report writing – Olli Varis, Researcher, PhD

karsresrr)aykarN_ - elak GUlI va:rI Cover: House in Prek Ta Kong village (Photo Olli Varis). For comments and questions please contact olli.varis@hut.fi

iii

Table of Contents EXECUTIVE SUMMARY iv EXECUTIVE SUMMARY IN KHMER vii 1. INTRODUCTION 1

1.1 Background 1 1.2 Mekong, Tonle Sap and Cambodia 2 1.3 The vision for the Mekong River Commission 3 1.4 Objective 4 1.5 Methodology: Bayesian network model 5

2. SECTOR POLICIES AND IMPACTS 6

2.1 Sector policies 6 2.11 Society 6 2.12 Water, infrastructure 9 2.13 Agriculture, fisheries 11 2.14 Environment 15 2.2 Impacts 16 2.21 Environmental impacts 16 2.22 Impacts on natural resources 22 2.23 Economic impacts 25 2.24 Social impacts 25

3. SCENARIO ANALYSIS 30

3.1 Definition of development scenarios 30 3.2 Results 31

3.21 Policy scenarios: vulnerability of geographic zones 31 3.22 Policy scenarios: The three development goals 32 3.23 Sensitivity analysis 34

3.3 Conclusions 35 REFERENCES 36

iv

EXECUTIVE SUMMARY Should a poor country such as Cambodia develop its water resources with the priority in environmental conservation, in maximizing economic growth, or in reducing poverty as much as possible? There is enough pressure evoked by the country’s contemporary situation to justify all of these prioritizations. However, only one of them—or a compromise of them—can be realized. Cambodia is one of the poorest countries in Asia. It has suffered from disastrous violence over many decades, and the economy, society and the environment have paid an enormous toll for this. The recent years have witnessed some national and international optimism in Cambodia’s case. The hostilities have calmed down and the government has been gradually stabilizing. The government agencies are still weak but getting slowly on track. Still, the central government is not well aware what is going on in the provinces. The provincial and local authorities have a mélange of ties and links with NGOs, and ties with central government are unfastened. In terms of water Cambodia is heavily dependent on the Mekong and hence on its neighbors, since all the six Mekong’s riparians have their varied and specific aspirations how to make living and wealth out of the mighty river. These aspirations do not always meet. An international organization was set up to handle this issue already in 1957, namely the Mekong River Committee (MRC). Due to the hostile history of the Mekong region ever since, the MRC has largely been occupied by various diplomatic tasks which are not only water-related, and therefore its success has not been praised by many. However, it uplifted its face in 1995 with the Mekong Agreement, which meant a new setup for the MRC, which now is called Me-kong River Commission. The Mekong Agreement includes various goals for the development of the river, with a bunch of tools for achieving these goals. These goals include among others economic development, poverty reduction and sustainable development. One of the major tools to find feasible compromises between these goals that often conflict—as was stated before—is to produce a basinwide development plan BDP. This plan would integrate the various aspirations of various actors inside the riparian countries as well as deal with the confrontations that these aspirations produce at the inter-national level. A research-oriented forerunner of the BDP is in full force. This Water Utilization Programme (WUP) started in 2001, and its first component known as the WUP-FIN modeling Project analyzes the Tonle Sap Basin in Cambodia. This basin is the poorest part of the country with an average income of 40 cents per capita. It is a unique and exceptionally rich wetland-lacustrine ecosystem which carries a population of 1.1 million in the immediate surroundings of the lake. The population is profoundly dependent on the lake in their livelihood, which is chiefly on a subsistence level. This paper documents the policy analysis component of the socioeconomic part of the WUP-FIN project. The focus is on the demanding task of analyzing where economic development, poverty reduction, and environmental sustainability conflict and where they walk hand in hand. WUP-FIN includes comprehensive watershed hydrological and environmental studies, monitoring and modeling of the lake, and several other activities besides the socio-economic part.

v

Problem analysis The problem defined above was approached with a variety of methods as listed below:

• Collection and review of existing literature. A bunch of documents of varying quality is available. Many outcomes are fairly contradictory. Peer-reviewed material is al-most nil.

• Collection and analysis of existing data. The government has performed a population census which is openly available as spatial data. Such databases provide a firm starting point for the analysis, despite of certain quality problems in the data.

• Building up links with international, national, provincial and local actors. The level of coordination is not good, and linking some of the activities is important. Motivation in most organizations is high but tasks are huge in comparison to capacities.

• Village survey. The MRC and most other centralized organizations have not been active at the grassroot level, unlike a few NGOs and aid organizations. Provincial authorities try their best but are extremely weak in capacity. The existing participatory systems (using approaches such as PRA and RRA) are the ground to build upon. Besides complementary data to the databases on local conditions, community development and participation is enhanced.

• Database construction. A GIS database is constructed on the basis of existing and collected data. • A policy model. A systematic risk analysis of basin development scenarios, key policy options,

and consequent environmental and social impacts on different stake-holders and development objectives is carried out.

The WUP-FIN Policy Model—the topic of this report—is the ultimate tool to find compromises between various aspirations between the development goals of different sectors, stakeholders, and development goals. It was constructed on the basis of the variety of information obtained in the other steps of WUP-FIN. The sector policies included in the Policy Model are:

• Conservation; The Tonle Sap is included in UNESCO’s Biosphere Programme. • Small and medium-scale fisheries; fish is the primary protein source for the region’s people, and

family fishing is of prime importance to the poor villages. • Large-scale fisheries; the lake is used by commercial fisheries units that exploit the resource

intensively, and often in disharmony with subsistence fisheries. • Agriculture, aquaculture, and irrigation development; most of the villages cultivate rice and rice

is the major staple for the dwellers of the area. Rice farming has many forms depending on the location of the village in relation to the massive floods of the lake. Common to them is low productivity due to destructed infrastructure and low level of resources to improve the yields.

• Roads; the road network is in a desperate state, but massive reconstruction of trans-portation facilities is under way. It will change rapidly the region’s economy of, and open the provincial towns to markets and trade.

• Navigation; harbors and other facilities are planned for improving the navigability of the lake. Tourism is one player behind these plans.

• Water and sanitation services; the water infrastructure is almost non-existing at the moment, and plenty of improvements are needed.

• Rural development; a variety of community and rural development activities are undergoing in the village level, overwhelmingly on foreign funding. The local authorities are well in the scene in alliance with NGOs, but the central government has little touch into this important activity.

• Urban development; the examples from economically blooming neighbors in Southeast Asia feed many thoughts of developing Cambodia’s towns and cities to-wards being attractive to foreign investments to industries and tourism. The famous Angkor temples could attract far more tourism than they can do due to many contemporary constraints.

vi

• Education. Cambodia’s schooling system has suffered vastly from the past conditions, keeping in mind the era of the Khmer Rouge that got brutally away of almost all educated people in the 1970s. Cambodian schooldays are among the shortest ones in the world. Much is needed to provide all Cambodian children an access to proper schools.

• Formal institutions; ministries in Phnom Penh live quite a different life from their local agencies. This is only one example of the shortcomings of the Cambodian governance system.

All these eleven sectors are crucial in the development of water resources in the Tonle Sap basin. Now, every expert would argue differently how important and powerful the policies implemented by these various sectors would be. Equally rich is the spectrum of prioritizations between environmental, social, and economic aspirations. A Bayesian Network model was constructed to imitate this controversial and manifold situation in order to analyze systematically the various aspects included in this problem. The model consists of a layer including the sector policies, another incorporating the major social, economic, and environmental impacts and consequences of implementing these policies, and a layer that combines the outcome in meeting the development goals, namely poverty reduction, environmental sustainability, and economic growth. The Figure below is an example of the results. It shows how much the realization of these development goals respond to small changes in sector policies. The further the line is away from the zero line (the circle closest to the center of the diagram), the more impact the policy has on the respective development goal. The impact is negative if the line goes inside the circle, otherwise it is positive.

SENSITIVITY ANALYSIS

-5.0 %

0.0 %

5.0 %

10.0 %

15.0 %Formal institutions

Education, public health

Urban development

Rural development (villages)

Water and sanitation services

Navigation Roads

Agriculture, irrigation

Large-scale fisheries

Small, medium scale fisheries

Conservation

Economic growth

Poverty reduction

Environmental sustainability

vii

Discussion and Conclusions Interesting enough, some sector policies would be crucial for both the economy and poverty reduction, but not all. The huge shortcomings in education and institutions are obviously the ones that most strongly support these two goals. Rural development, in turn has a very important social function although it has not much immediate economical implication. Large-scale fisheries would be economically beneficial but counterproductive for poverty reduction. The biggest surprise in the results is that the policies included in the model appear to be relatively toothless to environmental problems, particularly if defined as “environmental sustainability” as is done in the Mekong Agreement of 1995. This is obviously due to the following reasons:

• The concept of environmental sustainability is not easy to be conceptualized concretely enough so that it would be easy to treat analytically. In our model it is a combination of various issues and problems related to the environment and natural resources.

• As the majority of the population of the Tonle Sap area live in villages and make their living from the lake or the floodplain in a fairly direct way, the environmental issues are very closely bound to social is-sues. Social developments therefore are tightly bound to environmental impacts, and typically what happens in model simulations is that improvements in social conditions tend to introduce both positive and negative environmental impacts which cancel each other, seemingly to a great extent. The situation was different if the governance system was more efficient.

• The ecosystem of the Tonle Sap Lake is governed by the mighty floods of the Mekong, which raise the water level up to 9 meters, and subsequently the surface area of the lake grows fourfold. The sediments and other mass flows are also dominated by these monsoon floods. There are no handles in the sector policies included in the model that would allow the control of these issues.

viii

segçbkargarrbs;KMeragm:UEdl sikSaBI esdækic© sgÁm enA extþ CMuvij Tenøsab >

¬enHCakarERbsMrYlCaPasaExµredaysegçbBIPasaGg;eKøs sMrab;Bt_manlMGitsUmemIlCaPasaGg;eKøs¦

esckþIepþIm ³ vaKYrEtCaRbeTsRkIRkmYy dUcCaRbeTskm<úCa EdlerobcMFnFanTWk CamYynigkarGPirkS brisaßnEdlCaktþa GaTiPaBenAkñúgkMeNInénesdækic© bJ k¾kñúgkarkat;bnßyPaBRkIRk EdlCalT§PaB eRcInenaH vamanRKb;RKan; énsma<FepSg² EdlRbeTsmansaßnPaBbeNþaHGasnñ énbBaaðGaTiPaB TaMgenaH eTaHCaya:gNabBaað mYyénbBaaðTaMgenaH bJk¾RtUvsMrbsMrYl eTIbedaHRsaybBaað TaMgenaH ecjrYc .

RbeTskm<úCaCaRbeTsmYyEdlRkIRkCageK enATVIbGasIu ehIyRbeTsenH)anTTYlnUveRKaH mhnþray ya:gF¶n;F¶r CaeRcInTsvtSmkehIy dUcCabBaaðesdçkic© sgÁm brisaßn)an)at;bg;ya:gF¶n; F¶r sMrab;RbeTsenHehIy enAqaMñfµI²enHmanGgÁkarCati nig GnþrCatixøH mansuTidæielIkrNIyRbeTs km<úCa PaB)at;bg;Cama©s;)anFaøk;cuH ehIyraCrdæaPi)alBMuTan;manlMnwg Pañk;garrdæaPi)al enAEtTn; exSayehIym,:ageTotraCrdæaPi)alminmankarPaJk;rlwkl¥nUvkarsMLWgemIlkarbnþkargarBRgwg nig GPirkSbrisaßn enAtamextþ . enAtambNaþl extþGaCaJFr )anpSarPa¢b;kargarbrisaßnenaHCamYy GgÁkareRkArdæaPi)al .

enAxagEpñkTwkRbeTskm<úCa BwgBak;ya:gxaøMgeTAelITenøemKgÁ EdlhUrkat;RbeTsEk,rxøÜn RbeTsTaMgGs;CasmaCikTenøemKgÁmanbMNgR)afañepSg ² faetIeFVIdUcemþc[karrs;enAnigTenøman efrPaBCanic© .bMNgR)afañ TaMgenHCanic©EtgEtminCYbdUcesckþI R)afañeLIy . GgÁkarGnþrCati)an begáItkaredaHRsay bBaaðenHtaMgBI qaMñ 1957 edaydak;eQaµH KN³kmµaFikarTenøemKgÁ ( MRC ) enAeBlEdl MRC begáIt ehIymanPar³kic© CaeRcInEdlRtUveFVIKWminmanEtEpñkEklMGTwkmYyenaH

ix

eT. dUecñHmanbBaaðCaeRcInEdlRtUveFVInigGnuvtþn_ enAtamRbeTsnimYy ² . ehIym,a:geTot enAqañM 1995edaymankarRBmeRBogrbs;emKgÁeQaµHCapøÚvkar mYyRtUv)anbegáIteLIg eQaµH KN³kmµkar TenøemKgÁ . karRBmeRBogrbs;emKgÁ rYmmaneKaledAepSg ² sMrab;GPivDÄn_Tenønig]bkrN_sMrab; sMerc)anéneKaledA TaMgenH .

eKaledArYmbBa©ÚlTaMgkarGPivDÆn_esdækic©epSg²karkat;bnßyPaBRkIRknigkarGPivDÆn_ RbkbedayesßrPaB ]bkrN¾mYy én ]bkrN_cMbg edIm,IrkkarsMrbsMrYl EdlGaceFVIeTA)an rvag eKaledATaMgenH eTA nig CMelaH edIm,IbegáIt EpnkarGPivDÆn_GagTenøemKgÁ (BDP) EpnkarenH nig RbmUlpþMú nUvbMNgR)afañ epSg ² rbs;RbeTsEdleFVIskmµPaB enAtammat;TenøemKgÁ RBmTaMg edaH RsayCamYykarRbQmmuxEdlekIteLIgBIbMNgR)afañTaMgenHenAtamkMrit rvagCatinigCati .

TisedARsavRCavmYyénEpnkarGPivDÆn_GagTenøemKgÁ(BDP)KW)ancab;epþImGnuvtþtamry³ kmµviFIeRbIR)as;TwkWUP)ancab;epþImenAqañM2001KMeragenHsmasPaBTI01 KWviPaK GagTenøsab enARbeTskm<úCaGagTenøsabenHCaEpñkEdlRkIRkCageKenAkñúgRbeTsEdlRbCaCnrkcMNUl)anEt40esn¬dulaørGaemrik¦kñúgmYynak;ehIy vaCaRbBn§½brisaßnbwgEdlmanRbCaCnRbmaN1>1lan rs;enACMuvijbwgnigRbCaCnTaMgenaHBwgBak;TaMgRsugelIbwgedIm,IciBa©wmCIvitEdlPaKeRcInEtCa kMritbnÞab;bnSM .

ÉksarenaHCaEpñkesdækic©sgÁmrbs;KMeragWUPEdlCaBiessepþatelItMrUvkarénkarviPaK karGPivDÆn_esdækic© karkat;bnßyPaBRkIRk nirnþPaB én brisaßn PaBCMelaH EdlkMBg;dMeNIkar elI brisaßn nigesdkic© KMeragTenøsabrbs; WUP rYmbBa©ÚlRKb;Ebbya:g dUcCaGagsþúkTwksikSa elI brisaßn karRtYtBinitü nig Modelling rbs; bwg Tenøsab ehIynig skmµPaB CaeRcIn rbs; Epñk esdækic©sgÁm .

vIPaKbBaða³ bBaaðkMNt;xagelICaBiFIsaRsþmYyCamYyBiFIsaRsþepSg² dUccuHbBa¢IrxageRkam ³ • RbmUlnigBinitüTinñn½yEdlmanRsab;KuNPaB ÉksaCaeRcInGacrk)an ehIy lT§pll¥RtwmRtUv • RbmUl nig viPaK elITinñn½y EdlmanRsab; dUcCaCMerOnRbCaCn qañM 1998 EdlrdaæPi)al )aneFIV EdlCaTinñn½yRtwmRtUv kareRbIR)as;Tinñn½y mUldaæn vaCacMNuc cab;epþImsMrab;eFVIkarviPaK .

x

• ksag nig pSarPa¢b;CamYyGgÁkarGnþrCati extþ nig mUldaæn ehIynigkMriténkarsMrbsMrYl KW min sUvRbesIrehIyPa¢b;EtskmµPaBsMxan;²karRbmUlTinñn½yEdlmanRsab;PaKeRcIn)anmkBI GgÁkar eRkArdæaPi)al . • karcuHGegátenAPUmi KN³kmµakarTenøemKgÁ nig GgÁkar Edlmansañk;kar kNþal minmanskmµ PaBenAkMritmUldaæneT manEtGgÁkareRkArdæaPi)alBIr bJ bI nigGgÁkarCMnYyxøHGaCaJFr extþBüayam eFVI[manPaBRbesIrkñúgkarGPivDÆn_EtkMritsmtßPaBénmRnþIenAexSay. RbBn§½cUlrYmEdlmanRsab; ¬eday eRbIviFI saRsþ RRA & PRA eRkABIRbmUl Tinñn½y enAtammUldaænedaymankarcUlrYmBI RbCaCn ehIyk¾ykTinñn½yenaHviPaK rYmCamYyTinñn½y EdlmanRsab; . • karbegáItTinñn½ymUldaændUcCaTinñn½ymUldaæn GIS KWsaßbna elImUldaæn Tinñn½y EdlmanRsab; nig Tinñn½yEdlRbmUl)an . • eKalneya)aym:UEdlKWCaRbBn§½viPaKénkarGPivDÆn_GagTenøsabCaeKalneya)ayKnwøHplvi)ak énbrisaßn nig PaBb:HBal; sgÁm elIsaßb½n Bak;Bn§½epSg² ehIynig eKalbMNg énkarGnuvtþn_ kar GPivDÆn_ .eKalneya)aym:UEdlKWCa]bkrN_cugeRkaybMputedIm,IEsVgrkkarsMrbsMrYlrvagbMNg R)afañ éneKaledAGPivDÆn_ nig EpñkepSg ² énsaßbn½Bak;Bn§½ vaKWCakarsaßbna elImUldaænénBt’man epSg² EdlTTYl)ankñúgCMhan nimYy² énkarsikSa .

EpñkeKalneya)ay bB©ÚalkñúgkarviPaKKW • karGPirkSTenøsabB©aÚleTAkñúgkmµviFICIvsaRsþrbs;GgÁkaryUeNssÁÚ • karensaTRTg;RTaytUcnigmFüm RtI CaRbPBRbUetGIun sMrab;mnusS enAkñúgtMbn; ehIyensaT Ca lkçNHRKYsar CamuxrbrsMxan; sMrab;GñkPUmi RkIRk Edlrs;enAelIbwgTenøsab . • karensaTCaRTg;RTayFM bwgTenøsab eKeRbIR)as;sMrab;eFIVGaCIvkmµ elIEpñkBaNiC¢kmµRtI dUcCa karedjéføLÚensaT nig karensaTCabnÞab;bnSM. • ksikmµ RbBlvb,kmµ nig karGPivDÆn¾RbBn§½eRsacRsB PaKeRcIn én PUmi enACitbwg Tenøsab KW daMRsUv ehIyRsUvCadMNaMcaM)ac;sMrab;karrs;enArbs;RbCaCnkñúgtMbn; RsUvdaMPaKeRcInenATItaMgEdl PUmimanTMnak;TMngeTA nigTWkCMnn; ¬RsUveLIgTIk¦énbwgTenøsab plEdlTTYl)anenAmankMritTab edaysarkarxVHFnFan edIm,IeFVIeGayTinñpl)anRbesIreLIg. • pøÚvCabNþajTMnak;TMng ya:gsMxan; karksagpøÚv eLIgvij edIm,IsMrYl kardwkCBa¢Ún ehIyvaeFVI[

xi

esdækic©mankarERbRbYlrh½s edaykareFVIcracrTMnij eTAkan;TIRkug )anrh½s nig rvagTMnijCnbT eTAkan;TIpSar . • karsaßbnapøÚvnava KWCaEpnkarsMrYl[manPaBRbesIrénbwgTenøsabehIytMbn;eTscrN_FmµCati vanigedIrtYya:gsMxan; sMrab;EpnkarenH enAbwgTenøsab . • esvapÁt;pÁg;Twk nig suxPaB dMeNIrénkar karpÁt;pÁg;TwknigGb;rMsuxPaBesÞIrEtKaµnenAeBlbc©úb,nñ enH ehIyPaBRbesIrCaeRcInEdlRtUvkarenAeBlxagmux . • karGPivDÆn_ CnbT skmµPaBGPivDÆn_ CnbT KW kMBug)andMeNIrkarenAtamPUmixøH PaKeRcInEdl kmµviFI GPivDÆn_ dMeNIrkar)anenaH edaysar mUlniFibreTs GaCaJFr mUldaæn )anshkar ya:g l¥ CamYyGgÁkareRkArdaæPi)al b:uEnþraCrdaæPi)al manskmµPaB GPivDÆn_ enHtictYcNas; . • karGPivDÆn_TIRkug]TahrN_ BIRbeTsEdlenAEk,rxagGasIueKKitfa RbeTskm<úCaCaRbeTskMBug GPivDÆn_TIRkug nig TIRkugkMBugGUsTaj karvinieyaKbreTsbegáItCa]sSahkmµ nig eTscrN_ PaB l,Il,aj énR)asaT GgÁrGacGUsTajGñkeTscrN_)aneRcInCagGVIEdlBYkeKGaceFVIedaysarkMrit beNaþHGasnñCaeRcIn . • karGb;rMRbBn§½sikSaFikaenARbeTskm<úCa)anTTYlrgeRKaHCaeRcInTsvt S enAkñúgrbbExµrRkhm ehIykarGb;rMsikSaFika )anFaøk;cuHcab;taMgBIqañM 1970 RbCaCnkm<úCaEdl)ancUleronmancMnYntic ebIeRbobeFob eTAnigBiPBelak ehIyPaKeRcIn Edl)aneroncb;karsikSaeRcIn EtGñkrs; enATIRkug dUecñHkarpþl;dl;karsikSaKWeFVIya:gNa[RbCaCnCnbTnigTIRkugcUleronsala)anRtwmRtUv. • saßbn½pøÚvkarmanRksYg Bak;Bn§½ enAPñMeBjRtUvykcitþTukdak;kñúgkarBRgwgbJGPivDÆn_ enAtambNaþ mUldaænnana enAkñúgRbeTskm<úCa enHRKan;EtCa ]TahrN_ mYyénRbBn§½GPi)alkic© rbs;km<úCa.

EpñkTaMg 11 KW caM)ac;kñúgkarGPivDÆn_FnFanTwkenAGagTenøsabbc©úb,nñenHRKb;GñkCMnaj karepSg² kMBugEtCECkKañ dUemþc sar³sMxan;énkareFVIeKalneya)ay edIm,I GPirkS brisaßn sgÁm RBmTaMg esdækic©pgEdr .

bNaþjkargarm:UEdlsaßbnaeLIg edIm,IBinitüemIlsaßnPaB nigviPaKénRbBn§½TidæPaBepSg² rYmTaMgbBaaðenHpg . m:UEdlekIteLIg BIRsTab;epSg² rYmbBa©ÚlTaMgEpñkeKalneya)ay nig eKal neya)ayepSg² eTot EdlcMbgrYmman esdækic© sgÁm PaBb:HBal; én brisaßn ehIynig plvi)ak

xii

énkarGnuvtþn_eKalneya)ayTaMgenHRsTab;TaMgenaHRbmUlpþMuú )anCalT§plehIyRbCMu edIm,IkMNt; eKaledAGPivDÆn¾ dUcCakarkat;bnßy PaBRkIRk kareFVI[mannirnþrPaB énbrisaßn nigkMeNInesdækic© TUelxxageRkay KWCa]TahrN_ mYyénlT§plénkarviPaKvanigbgaðjnUveKaledAGPivDÄn_eqøIytb eTAnigkarpaøs;bþÚrtictYcén eKalneya)ay . PaBb:HBal;éneKalneya)aymaneRcInenAelIeKal edAGPivDÄn_ .

karBiPakSanigkarsniñdaæn³ EpñkeKalneya)ayxøHva KW caM)ac;sMrab; Epñkesdækic© nig karkat;bnßyPaBRkIRk bu:Enþvamin TaMgGs;enaHeT . mYyPaKFMsem,ImenAkñúgsaßbn½ Gb;rMKW Cakarc,as;las; EdlKaMRT ya:gxaøMgdl; eKaledAGPivDÆn_ 2 karGPivDÆn_CnbT KWCakar kargarsMxan;sMrab;[sgÁmCati nigesdækic©rIkcMerIn karensaTRTg;RTayFM vaGacTajykcMNUlsMrab; esdækic©Cati ehIyk¾CaEpñkmYyénkarkat;bnßy PaBRkIRk .

PaBPaJk;epI¥l d¾FM énkarviPaK lT§pléneKalneya)ay EdlekItmanKW bBaað brisaßn PaB Cak;EsþgenaHvaKWkMNt;fa nirnþPaB énbrisaßn EdleFVIedaymankarRBmeRBog rbs;KN³kmµkaTenø emKgÁ enAqañM 1995 karRBmeRBogenHekIteLIg edaysar ehtuplxageRkamenH. • TsSn³én nirnþPaBbrisaßnmingayRsYl nigkareFVI[manlMnwg énbrisaßnRKb;RKan;enaH eTIbman karviPaK edIm,IrkdMeNaHRsay gayRsYlenAkñúgkarviPaK énm:UEdl rbs;eyIg KW )anRbmUlpþúM bBaað epSg ²EdlTak;Tg eTA nigbrisaßn FnFanFmµCati . • RbCaCn PaKeRcIn Edlrs;enAPUmi tamtMbn;Tenøsab karciBa©wmCIvitrbs;BYkeK BwgBak;TaMgRsug elIFnFanFmµCati enA bwgTenøsab eday paÞl;Etmþg bBaað brisaßn vaCitsñitNas; nig bBaað sgÁm karGPivDÆn_sgÁm KW vacgCab; nig PaBb:HBal; én brisaßn lkçN³GVIEdlekItmanenAkñúgm:UEdlKWfa eFVI[manPaBRbesIrénsgÁmTaMgEpñkb:HBal;brisaßnCaviC¢mannigGviC¢manEdllb;bM)at;PaBb:HBal; BImYyeTAmYy eKemIleXIjfasaßnPaBKWepSg² Kañ RbsinebIRbBn§½ GPi)alkic©manRbsiT§iPaBx<s;. • RbBn§½brisaßnbwgTenøsabKWTTYlrbbTwkTenøemKgÁEdleLIgrhUtdl;kMBs;9 Em:Rt karduHdInigkar hUrénsarCatiepSg ² EdlPaKeRcInekItmaneLIg enAbwgTenøsab KWenArdUvTwkCMnn;minmankaredaH RsayenAkñúgEpñkeKalneya)ayEdlbBa©Úlkñúgm:UEdlfa RtUvGnuBaØati[RtYtBinitü énbBaaðenHeT

xiii

lT§pl énkarviPaK enAkñúg m:UEdl Edl KMerag WUP TTYl)anvaKWbMerI[ MRC enAkñúgkmµviFIEpn karGPivDÆn_GagTenøemKgÁnigP¢ab;karRKb;RKgGagbgðÚreTAskmµPaBtammUldaænéneKalneya)ay Cati nigtMbn;GnþrCati vaKWc,as;fa kapSarPa¢b;CamYyGnþrCatiKWCakarcaM)ac;CaBiessEdledaHRsay bBaaðbrisaßnénGagTenøemKgÁ .

1

1. INTRODUCTION 1.1 Background

Environmental and socioeconomic impacts of water resources management are tightly interwoven. This is particularly true in conditions such as in the surroundings of the Tonle Sap Lake, Cambodia, where the livelihood is mainly based on subsistence fisheries and agriculture, and poverty is widespread. The average income level in that area is US$ 150 per year, which is only 41% of the international poverty line of US$ 1 a day.

Therefore environmental and socioeconomic impact analyses must meet in a fairly basic level. It is not always unambiguously clear whether environmental degradation is a root cause for the deterioration of the living conditions of the rural communities that live in poverty, or vice versa. In most cases, these issues constitute a vicious circle, which needs to be analyzed in a multidisciplinary and holistic manner.

The methodology used for this task—Bayesian Networks—is based on the systematic analysis of causal interconnections in a complex environmental-socioeconomic systems. The objective is to assess risks to various components of the environmental and social system under concern, as consequences of different policy strategies under evaluation. The social system components consist typically of stakeholders, i.e., different communities and groupings of people that are influenced by the implementation of policies in the studied geographical area. It is not rare that their aspirations and interests are in conflict with one another. The environmental components, in turn, are issues such as eutrophication level of a lake, vegetation around a lake, land degradation, and so forth, issues which are under specific analysis in other phases of the WUP-FIN project. The results of such specific studies are condensed in a risk analysis framework, and a multidisciplinary analysis is performed, that reveals the major risks, uncertainties, mismatches of information, and opportunities to find win-win solutions among the various stakeholders and the environment.

The local populations are those who feel most basically the impacts of any changes within a watershed. The conventional wisdom is an important source of information in planning water management policies. At the same time, the level of awareness of those communities on water-related issues such as spreading of diseases, environmentally sound waste management, erosion control, risks of overfishing, etc., can be very superficial and biased. The involvement and analysis of local communities to policy analysis and policy making is important everywhere, but particularly momentous albeit too often ignored in conditions where public education has shortcomings, and where informal institutions and traditions govern more than the institutions of a modern society do.

2

Such is the case in notable parts of the rural Mekong Basin, in the surroundings of Tonle Sap Lake as well. The socioeconomic and environmental analysis has been based—besides up-to-date empirical data analyses and environmental models—on surveys of the components of the livelihood of the local people. These surveys have been focused on the communities in the immediate vicinity of the lake (see the Tonle sap Socioeconomic Studies 1 to 6).

1.2 Mekong, Tonle Sap, and Cambodia

The Mekong River is the ninth largest river in the world if measured with runoff. With its 500 km3 that it carries each year it is ten times of the size of the Nile. In terms of catchment area, however, Mekong is not bigger than 800,000 km2. This is slightly more than one quarter of the Nile’s drainage area. Mekong is one of the world’s most pristine large rivers.

Cambodia’s Great Lake, the Tonle Sap, is a unique lacustrine-wetland ecosystem (Figure 1). It is Mekong’s major natural reservoir with annual water level fluctuations of around eight meters. The surface area exceeds 12,000 km2 during the monsoon floods, and shrinks to 2,000 km2 by the end of the dry season.

The lake is one of the world’s most productive large wetland ecosystems. Its biodiversity is extreme. Over 500 fish species have been documented to live in the lake. Fish and rice are the backbones of the traditional livelihood in the area.

Natural conditions are exceptionally favorable to humans and the history has seen many advanced cultures flourishing in the Tonle Sap area. The best-known evidence of such cultures is the ancient city of Angkor with its famous Angkor Wat temple close to Siem Reap at the lake’s northwestern corner.

Cambodia was economically the most advanced country of the Mekong region in the 1950s and early 1960s. However, the countries have followed very different paths ever since. In 1999, Cambodia’s GNI per capita was US$260, only 1/8 of Thailand’s GNP. In the Tonle Sap Basin, the figure was only US$150, which is 41% of the international poverty line of US$1 per day per person. Measured with this economical indicator, Cambodia is one of the poorest countries of the world, and the Tonle Sap region is its poorest part. Any social indicator tells the same.

Why this extreme affluence of nature coincides with one of the world’s most striking traps of poverty and deprivation? Most unfortunately, the last three decades have been extremely violent and politically volatile. The country survived from the War of Indochina and the early phases of the Vietnam War with modest damages. However, the 1970s saw the incredible collapse of the used-to-be prosperous Kingdom to the battlefield of whosoever, and finally the Khmer Rouge emptied all towns, slaughtered practically all the educated people and many others, and attempted to create an ideal decentralized, rural society. The country has been violent ever since, only the past decade have seen emerging stability. The country has still more land mines in its ground than any other country in the world.

Nature has fortunately sustained this period with modest degradation.

3

Cambodia is now in the process of recovery on many fronts. Infrastructure needs urgent rehabilitation and is undergoing a rapid progress, educational system and health care as well. The impacts to economic system are already great and profound and still growing. Government institutions, above all ministries are still weak and corrupt but obviously on the way to the better.

Economy of the Tonle Sap region has declined to very basic subsistence farming and fishery. Poverty touches everybody. Illegal activities, largely based on destructive exploitation of natural resources, mushroom.

Nature will not sustain the present informal economy and poverty-driven destructive practices for very long.

Figure 1. The map of the Tonle Sap Area. The four zones used in this analysis are defined by the topography of the area: Zone 1 ranges from 0 to 6 m above mean sea level, Zone 2 from 6 to 8 m, Zone 3 between 8 and 10 m, Zone 4 between 10 m and National Roads 5 and 6, and Zone 5 includes the urban areas. The map shows also all villages in the Area.

1.3 The Vision for the Mekong River Commission

The Mekong River Commission (MRC) has been the strongest international organization in the region has over the years. Among the riparian countries Cambodia, Vietnam, Lao PDR and Thailand are its members, but China and Myanmar are not.

4

The MRC is currently working on a comprehensive master plan for the lower Mekong River Basin (basin’s parts that are within the member countries). This Basin Development Plan (BDP) is supported by a massive 6-year background analysis—or a series of analyses—under the title Water Utilization Programme (WUP). Mekong Committee was established in 1957. Its initial members were Thailand, South Vietnam, Laos and Cambodia. Burma and China did not join it. The Committee attempted to solve the regional water controversies with varying success. Its functioning was made especially difficult by China's absence, as well as several national and international conflicts and wars in Vietnam, Cambodia and Laos (Jacobs 1995). By 1995, the regional political ambient became favorable to enhanced political and economic integration in Southeast Asia. As one of the results Vietnam, Lao PDR, Cambodia and Thailand signed the Mekong Agreement on the new modalities of co-operation in the Lower Mekong River Basin. This agreement re-established the Committee, and it was newly named as the Mekong River Commission. The Commission was reinforced in many ways. Cambodia is again a member, yet China and Myanmar continue to be absent. Capacity in all ways was improved. Perhaps the major shift was from being an agency that primarily executed various projects to an agency, which has a more strategic mandate. Instead of projects, it commits itself to a set of programmes in a more long-term basis. The vision and mission statements of the MRC are (MRC 2000, 2001):

• VISION for the Mekong River Basin: “An economically prosperous, socially just and environmentally sound Mekong River Basin.”

• VISION for the MRC: “A world class, financially secure, international river basin organisation serving the Mekong countries to achieve the basin Vision.”

• MISSION: “in accordance with the 1995 Agreement: To promote and coordinate sustainable management and development of water and related resources for the countries’ mutual benefit and the people’s well being by implementing strategic programmes and activities and providing scientific information and policy advice.”

1.4 Objective

The objectives of this report is to answer to the following question: How the vision for the Mekong Basin that is quoted above can be realized for the Tonle Sap Area in Cambodia? The three goals, economic prosperity, social justness, and environmental soundness are all important goals, but they are in conflict in many ways. Moreover, the concurrent situation is very far from all these three goals. Economic growth is needed to reach economically prosperous Tonle Sap Area. Equally, massive poverty reduction is mandatory for the accomplishment social justness in the Area. Achieving environmental soundness likewise needs strong policies. Of particular interest in this analysis are 11 sector policies and how they can be used to develop the Tonle Sap Area into the desired direction. The time frame of the analysis is 5 to 10 years to the future from 2003.

5

1.5 Methodology: Bayesian network model

The methodology used for this task—Bayesian Causal Networks (Varis 1998, Varis and Fraboulet-Jussila 2002, Varis and Lahtela 2002)—is based on the systematic analysis of causal interconnections in a complex environmental-social-economic systems. The objective is to assess risks to various components of the environmental and social system under concern, as consequences of different policy strategies under evaluation.

The social system components consist typically of stakeholders, i.e., different communities and groupings of people that are influenced by the implementation of policies in the studied geographical area. It is not rare that their aspirations and interests are in conflict with one another.

The information from various sources and of varying quality—as outlined above—is condensed in a risk analysis framework, and a multidisciplinary analysis is performed, that reveals the major risks, uncertainties, mismatches of information, and opportunities to find win-win solutions among the various stakeholders and the environment.

A logical sequence in the socioeconomic model study starts from a set of scenarios, which are followed by a selection of development priorities, which allow the society to react to these scenarios. Different policies have different impacts on the environment and the socioeconomic system. Finally the local and national stakeholders feel these changes—either benefit or suffer from them (Figure 2).

The model will allow trade-off analyses between different development objectives, and find policy combinations that create a maximum amount of win-win situations between the competing stakeholders.

OUTCOMES

SCENARIOS Environment Policy

National development goalsVulnerability of geogr zones

IMPACTS Environmental Economic Social

ACTIVITIES Sector policies

Figure 2. The logical chart of the model analysis.

6

2 SECTOR POLICIES AND IMPACTS

The model has the total of 47 variables. They consist of 11 sector policies, 28 impact variables, and 8 development goals.

2.1 Sector policies

The model includes a set of sector policies that can be used in various ways as policy tools in directing the development of the Tonle Sap region. The sector policies can be classified into four groups, namely those related to the society, to water and infrastructure, to agriculture and fisheries, and to environment (Table 1).

Table 1. The sector policy variables of the WUP-FIN Policy Model for Tonle Sap lake. Sector policies Society Formal institutions

Education, public health Urban development Rural development (villages)

Water, infrastructure Water and sanitation services Navigation Roads

Agriculture, fisheries Agriculture, aquaculture, irrigation Large-scale fisheries Small, medium scale fisheries

Environment Conservation

2.11 Society

The first group includes the following four variables.

Formal institutions. The recovery of Cambodia from the destructive periods of the recent past is still very much going on. The ministries are building up capacity to deal with the problems and challenges that the country is facing, but they still have a long way to go.

The legislation on the environment and related sectors has been partially developed, but the proper implementation is still under way. The Law on Environment Protection and Natural Resources Management was passed to the National Assembly already in late 1996. The legislation and the governance system of Cambodia with a focus on the Tonle Sap lake is summarized by ADB (2000). Whereas the progress of strengthening

7

government institutions is well in place, the administrational system is still suffering from severe capacity problems, and malfunctions of many sorts in turning the goal of good governance into everyday practice.

One attempt to provide a stronger and more interdisciplinary handle to the development of the Tonle Sap region is the newly established Tonle Sap Secretariat. It would be an interministerial committee with the aim of enhancing integrated policies that combine all stakeholder interests.

Besides government organisations, several multilateral organisations and national NGOs are active in the Tonle Sap region. While the work is definitely very important, some problems include the fairly poor integration of often very scattered projects (UNDP 1999), and the decrease of aid in recent years due to the general feeling that the emergency phase of Cambodia is already passed, and the country is no longer in the most burning need of international “first aid”.

Education, public health. Formal education (primary, perhaps secondary schools), health centres, and public awareness in health related issues are among the key issues in the improvement of livelihood in the Tonle Sap Basin (Figure 3). Their present situation is poor (Bonheur 1996). Especially the floating villages suffer from shortage of schools and teachers. Only 20% of children are able to attend the primary schools in those villages, and no secondary schools exist. Moreover, the school days are shorter in Cambodia than in most other countries.

Insufficient investment in infrastructure

Poverty

Inhibition of economic

growth

Inhibition of human

development

INTERVENTION

LIFE EXPECTANCY(at birth, 1999)

45 55 65 75

Cambodia

China

Lao PDR

Myanmar

Thailand

Vietnam

Figure 3. Left: Breaking the vicious circle of rural poverty by enhancing human development (after the philosophy by UNDP 1999). Life expectancy at birth in the Mekong countries (source: World Bank 2001).

Urban development. Cambodia’s level of urbanization is very low (Figure 4). Only one-sixth of Cambodians live in urban areas. However, the country is one of the most rapidly urbanizing ones in Asia. Urban areas need investments to infrastructure, services and human capacity (Haapala 2003). By now, the investments have not been able even to keep the infrastructure in pace with population growth. This direction should be reversed.

8

The example from neighboring counties shows that towns and cities of Southeast Asia can be very attractive targets of foreign investment. Thailand receives 30 times more foreign investments than Cambodia.

Urban population (%)

0 10 20 30 40

Cambodia

China

Lao PDR

Myanmar

Thailand

Vietnam

19962000

Urban population growth rate (%)

0 % 2 % 4 % 6 % 8 % 10 %

Cambodia

China

Lao PDR

Myanmar

Thailand

Vietnam

19962000

Urban population growth (% in 1996-2000)

0 % 5 % 10 % 15 % 20 % 25 % 30 %

Cambodia

China

Lao PDR

Myanmar

Thailand

Vietnam

Figure 4. Urbanization in the Mekong countries (source: World Bank 2001).

Rural development. Most people in the Tonle Sap region are rural-based and poor. The communities have variable capacities in providing livelihood to their members. Policies such as the ones below are included in this category:

• Development of local governance by strengthening relationships with local organizations like community based groups, local NGOs and local government. Participatory development and empowerment of people to improve people’s ability to improve their living are very important in contemporary rural Cambodia.

• Community fisheries, forestry and agriculture projects. By supporting the capacity and capability of villages to exploit natural resources in a sustainable way, the communities’ capacity and ability to develop their livelihood is improved, hopefully in a way that generates more wealth and is more sustainable than the practices without such programs. NGOs have been in a key role thus far in community level programs.

• Resource tenure. The problem of landlessness is striking in the Tonle Sap area. The landless people have no possibilities to improve their living by investing in farming or other persistent activities. The landless are most prone to social unrest, migrations, and clearly, they are at the lowest level of social and economic hierarchy. Poverty problem is closely associated to the landlessness problem. Besides, the land tenure system is largely unclear and unresolved. Farmers farm the land although they have in many cases no official land tenure to the fields (Ledgerwood 1999). The provincial authorities may have provided tenure to farmers although the central government hasn’t. Customary rights to resources are very important, but particularly in fisheries, the rights are insecure. The unclear situation counteracts investments.

9

2.12 Water, infrastructure

The second group of sector policies target infrastructure in the area. The following three variables are included in the model: water and sanitation services, navigation, and roads.

Water and sanitation services. Provision of water supply in villages is also among the key issues in the improvement of livelihood in the Tonle Sap Basin (Figure 5). Its present situation is poor (Bonheur 1996). Only 15% of the rural population have an access to safe drinking water.

Source of Drinking Water in the Tonle Sap Area

0 %

20 %

40 %

60 %

80 %

100 %

Zone

1

Zone

2

Zone

3

Zone

4

Zone

5

All

zone

s

OtherBoughtNaturalDug wellTube/pipe wellPiped water

Figure 5. Source of drinking water in different Zones of the Tonle Sap Area into different occupational activities (processed by Keskinen 2003 after Population Census of 1998).

The management of wastes including sanitation is very underdeveloped in the region (Figure 6). This is the case in villages and urban settlements. Even the city of Phnom Penh has no wastewater treatment. There are activities to improve the wastewater treatment in Siem Reap, but this is unfortunately an exception. The situation in most settlements is problematic. The waste management shortcomings are an important source of public health problems, cause eutrophication of the lake, and so forth.

SANITATION(urban, %, 1999)

50 60 70 80 90 100

Cambodia

China

Lao PDR

Myanmar

Thailand

Vietnam

EXERTIA DISPOSALRural and urban Cambodia 1993

Border of yards

Fields

Not reported

Latrines

Figure 6. Left: Access to appropriate sanitation in urban areas in the Mekong countries (source: World Bank 2001). Right: Exertia disposal in Cambodia (data from UNICEF/Government, in FAO 2000).

10

Navigation. Waterways have been always important for the people living in the Tonle Sap area. Many remote villages are best accessible by water at least during some seasons. There exists also commercial passenger and goods traffic between Phnom Penh and Siem Reap. However, the infrastructure for navigation is poorly developed. The harbors are inadequate for larger ships, and considered as temporary. Particularly during the dry season the routes are very shallow and dredging has been proposed many times.

The most extensive plan for improving the navigability of the Tonle Sap includes the construction of a harbor in the vicinity of Siem Reap. The harbor would benefit both passenger traffic and goods transport. The biggest social issue associated with this plan is the need to resettle around 5000 ethnically diverse people from the Chong Kneas area. Six from the seven villages that would need to be displaced are floating villages.

The environmental concern arises from the plans to augment the transportation of fuels and other hazardous chemicals, as well as from the inherent growing risk of leakages from boats along with mounting traffic. On the other hand the improved harbor and navigation channel facilities would contribute to safer traffic.

Roads. One of the basic problems of the region is very poor road network. It is a severe constraint to economic development in the Tonle Sap basin in many ways (Figure 7). Less than 20% of Cambodia’s highways are paved. Only one-third of the road network is passable all year round.

In this variable, particularly the reconstruction of national roads 5 and 6 is central. This work is going on. The ADB, the World Bank, the Japanese Government, and the Cambodian Government are funding the road construction massively. By mid-2005, two-thirds of the national road 6 should be rehabilitated if the plans will be realized, and the roads should be in the shape they were in 1970, before the wars destroyed them (Sen 2002).

It is not only the national roads that need reconstruction. The whole network of local roads is equally essential for the local economy—and they are still in a very bad shape.

Insufficient investment in infrastructure

Poverty

Inhibition of economic

growth

Inhibition of human

development

INTERVENTION ROADS PAVED

(% of all roads, 1996)

0 50 100

Cambodia

China

Lao PDR

Myanmar

Thailand

Vietnam

Figure 7. Left: Breaking the vicious circle of rural poverty by infrastructure construction (after the philosophy by UNDP 1999). The state of roads in the Mekong countries (source: World Bank 2001).

11

2.13 Agriculture, fisheries

The third sector policy group includes the most important agricultural and fisheries issues.

Agriculture, aquaculture and irrigation. Rice is by far the most important crop in the basin. The four basic, contemporary farming practices include (Figure 8):

• deep-water (floating) rice,

• rainfed rice (requires irrigation),

• recession rice, and

• dry season irrigated rice (irrigation is mandatory, yields can be improved greatly)

Besides rice, some other subsistence and cash crops are cultivated, such as vegetables, fruit, palms and so forth. Cambodia has far less land under irrigation than the other countries in the Lower Mekong Basin. Only 3.6 per cent of rice cultivation area is irrigated, and Cambodia’s irrigated area is only 2 per cent of Thailand’s irrigated area.

Floating rice is cultivated in the floodplain of Tonle Sap. So is recession rice. The latter is far more productive and gains all the time more area. Wet season rice is cultivated on a belt inside and around the national roads. Dry season irrigated rice cultivation suffers the most of poor water control and related irrigation infrastructure.

After MRCS/UNDP (1998) the present wet season irrigated area is 40,000 hectares in the Tonle Sap basin. The total area of existing irrigated land is 93,000 hectares, but 57% of that area is not in use and would need rehabilitation. Irrigation potential would be 360,000 hectares. In summary, around 1/9 of irrigation potential is in use in the basin. With unit yields similar to those in Northeast Thailand (which is comparable to the Tonle Sap basin what comes to natural conditions), these fields could produce some 0.5 to 1 million tons more rice than they do today. This would allow the maintenance of food self-sufficiency in the Tonle Sap basin for another two decades.

LAND USECultivated areas in Tonle Sap region

Wet season paddy fields

Floating rice land

Receding rice fields

Field crops

Upland crops

Orchards

Urban areas

RICE ECOSYSTEMSCropping areas of rice in Cambodia

Rainfed lowland

rice

Rainfed upland rice

deepwater or floating

rice

Dry season irrigated

rice

Figure 8. Left: the structure of cultivated land in the Tonle Sap basin (source: MRCS/UNDP 1998). Right: Percentages of cropping areas of different types of rice in Cambodia (source: FAO 2000).

12

Only 60% of land used for agriculture in 1967 was used by farmers in 1993 (Bonheur 1996). One important reason to this is the persistent problem of landmines.

Domestic animal husbandry is almost totally based on smallholdings, and is not widely commercialised.

Fish ponds have been developed in the region, but their development potential is far above the present level. In Siem Reap, around 11% of fish catch comes from aquaculture (Bonheur 1996). To a certain extent the growing interest of the local people to aquaculture can be attributed to the decreasing catches of snakeheads and whitefish (Lamberts 2001).

COMMERCIAL FISH CATCH(Cambodia, thousand tons)

01020304050607080

1980 1985 1990 1995

Inland

AquacultureMarine

Annual fish requirements in Cambodia

0100200300400500600700800900

1000

2000 2005 2010 2015 2020 2025 2030

Thou

sand

tonn

es

0

5

10

15

20

25

30

High consumptionHigh-mediumLow-mediumLow consumptionPopulation (millions)

Figure 9. Left: commercial fish catch in Cambodia (data from the Ministry of Agriculture and Forestry, Cambodia). The catches have grown after 1993; in 1999, the recorded total catch was 280 thousand tons. Most of this sharp growth, however, can be attributed to the improvements in data collection. Right: Current consumption and projected requirements of fresh fish at different consumption levels in Cambodia, 2000-2030 (Lamberts 2001). The consumption levels are from low to high: 20.1, 25.2, 31.2 and 37.1 kg per person per year.

The main axis in these aspects in terms of future policy strategies is whether the subsistence/small-scale commercial activities should be prioritised, or whether the weight should be put on commercialisation and market-driven development of the agricultural and aquacultural production systems. The latter would also mean a possibility to diversify the production to produce two crops per year, by rotation of rice and e.g. legumes. At present, only one crop per year is a commonplace.

Fisheries management: large scale vs. small scale. I the model, the fisheries policies includes two variables—one targeted on the development of large-scale commercial fisheries, and the other one on small-scale fisheries which is primarily a family-scale activity.

The fish fauna of the Mekong Basin is exceptionally diverse and rich. At present, around 1,200 fish species have been identified. In Cambodia, around 500 freshwater species are known to exist (Rainboth 1996).

13

From these 500 species, the ten most important ones in terms of fishery contribute 63% of total catch and 59% of total value of the catch. The most caught fish is the Trey Riel (Henicrhynchus sp.), contributing to 21% of the total catch (Table 2). Being a Cyprinid, it is not very much valued, though. Snakeheads and Catfish which are the two other major groups of freshwater fish in Cambodia are much more appreciated than Cyprinids.

Table 2. Freshwater fisheries in Cambodia 1995-1996: Species composition and value of ten most important species by type of fishery (source: Deap et al. 1998, in van Zalinge et al. 2000).

Type of fishery (%) Rank Species name Lot Dai Middle

Of total weight (%)

Of total value (%)

Type of fish

1 Henicorhynchus sp. 11 40 20 21 9 Cyprinid White 2 Channa micropeltes 16 8 9 19 Snakehead Black 3 Cyclocheilictyhus

enoplos 8 1 13 9 8 Cyprinid White

4 Dangila sp. 5 6 7 6 2 Cyprinid White 5 Osteochilus sp. 2 10 2 4 2 Cyprinid White 6 Cirrhinus microlepis 5 3 2 3 4 Cyprinid White 7 Pangasius sp. 8 1 3 3 Catfish White 8 Barbodes gonionotus 3 4 3 2 Cyprinid White 9 Paralaubuca typus 1 11 3 1 Cyprinid White

10 Channa striata 5 1 2 6 Snakehead Black Weight (%) of 10

species 64 71 58 63 56

% share in total catch 33 23 44 100 100 % share of total value 41 15 44 100 100 No. of species

recorded 75 44 62

The terminology used by the Fisheries Department and most of the other actors in Cambodia (e.g. Lamberts 2001) divide the caught fish species into three categories:

• Black fish. Species able to survive in swamps and wetlands over the whole year, and have only limited migrations in the lake and its tributaries (lateral migration). They are mostly carnivorous and detritus feeders. For instance Snakeheads belong to black fish. Species include clariids (Clarias batrachus), channids (Channa micropeltes, C. striata), bagrids (Mystus spp.), belontids (Trichogaster spp.) and Anabas testudineus.

• White fish. Most of these species migrate between the mainstream Mekong and Tonle Sap Lake (longitudinal migration). Most typically, they come with floods to the lake as fish eggs, larvae or very tiny, feed and grow there, and return to Mekong in the dry season. The Catfish and many cyprinids belong to white fish. Most important species among cyprinids are Cirrhinus microlepis, Hampala macrolepidota, Barbodes altus, Leptobarbus hoeveni, Osteochilus melanopleura and Morulius chrysophekadion, among pangasiids species, silurids (Wallago attu, Micronema apogon) and among notopterids Notopterus chitala and N. notopterus.

14

• Small fish. This group of fish is defined in a different way in different sources. Common to these sources is that they consider a heterogeneous group of fish species which are small in size together due to the specific use of these fish by the local people. The small fish are used to produce prahoc and fish sauce, and used to feed caged fish. Some of it is consumed directly by the people. The small fish are fast-growing, and abundant in the flood period. They consist mainly of cyprinids (Henicorhynchus siamensis, Thynnichthys thynnoides, Dangila spilopleura).

Of all the fish captured in the Tonle Sap area, longitudinal migrants (mostly whitefish) constitute about 63% (van Zalinge et al. 2000).

Fish production is positively related to surface area of the lake and to the duration of the flood. In wet years, the fish production and catches are greater than in dry years. This relation might be indirect and the key controller of fish productivity might be the amount of net sedimentation that, of course, goes well together with the height of the flood (van Zalinge et al. 2003).

There are four dominating types of fisheries. They are:

1. Fishing lots: Concessions auctioned by the Government for a two-year period for commercial operators. In the whole country, there exist 164 lots in lakes, rivers, and river beaches. Fishing lot system has recently been under sledge, and its future is governed by various uncertainties.

2. Dai (bagnet) lots: A dai is a stationary trawl constructed in a river to capture fish migrating downstream. This technique is much cheaper than a lot. The rights to positioning a bagnet in a river is also auctioned. Tonle Sap River is the most important area for bagnets, where their season is from October to March when the flood recedes. Half of the catch comes in January. In 1997-1999 68 bagnets operated in the river.

3. Medium-scale (open-access) fishing: This activity is controlled by licenses by the government. A variety of gears are allowed including gillnets, traps, encircling seine nets, trawls, hooks, lines and so forth. Gillnets contribute to 52% of the catch.

4. Family fishing and ricefield fisheries: No license is required, but the gear allowed is far more restricted than in medium-scale fisheries. Family fishers have an access to fishing lots in the closed season from June to October.

On the top of them, there exist 15 fish sanctuaries.

A hot topic in the Tonle Sap Lake fisheries has been the debate of the lot system that has recently been under profound changes. It is has been based on 58 regions (lots). This system has been criticized for

1) Leaving customary public rights of the communities and landless people powerless and thus hampering notably the livelihood of the local population and

15

2) Leading easily to overfishing, since the lot operators have no stake in taking care of sustainable fisheries practices.

On the other hand, many alerts have been issued on the risks that the customary-right based traditional family fisheries would lead to overfishing due to the present, dense and growing human population in the region, and the lot system is said to protect the resource from overexploitation. Now, this risk is exaggerated on the opinion of the author. The proponents of the lot system claim that only the principles for selecting the lot operations needs revision, not the system itself. Destructive fisheries practices are widely applied, such as electrofishing, too small mesh-size nets, drying of parts of the lake, samras fisheries, and so forth. The dai fisheries has also been accused for causing substantial potential and actual risk overfishing.

FISH CATCHESTonle Sap region

Fishing lots and dais

Licensed mobile

Family mobile

FISH CATCHESLower Mekong basin

Cambodia

Lao PDRThailand

Vietnam

Figure 10. Left: Fish catches in the Tonle Sap Basin (source: MRCS/UNDP 1998). Right: estimated capture fish production in the lower Mekong basin. The total capture ranges between 809 and 951 thousand tons per year (source: van Zalinge et al. 2000 from various sources).

2.14 Environment

The fourth group of sector policies is related to environment:

Biodiversity conservation. Due to various human-induced pressures, the natural forests and other ecosystems have vanished with an alarming rate in the Tonle Sap Basin (Figure 11). Cambodia has a few conservation areas, but the unique lacustrine/wetland ecosystem of the lake and its floodplain are too far from being sufficiently protected is a view shared by many.

The Tonle Sap lake was designated as UNESCO’s Biosphere Reserve in 1997. The three core areas cover 708 km2. They are Prek Toal near Siem Reap, Moat Khla-Boeng Chhmar Lake and Stoeng Sen River. The total area of the reserve, including also buffer zones and transitional area, includes the whole territory between the national roads 5 and 6. Its area is 14,813 km2.

16

With Kosal’s (1998) words: “Core area of a biosphere reserve … constitutes area or areas within thereof which is 'devoted to long-term protection, according to the conservation objectives of the biosphere reserve.' For this purpose a core area should be established for prime objectives to conserving all aspects of biological diversity which include ecosystems, species and genetic diversity and landscapes. Use of the area should be restricted to 'research and study with particular purpose of monitoring minimally disturbed ecosystems, and undertaking non-destructive research and other low-impact uses.”

The Boeng Chmar and associated river system and foodplain is one of Cambodia’s tree Ramsar conservation sites. Angkor is a UNESCO World Heritage Site.

SURROUNDINGS OF TONLE SAP LAKEChanging land use

0%

20%

40%

60%

80%

100%

1986 1993 1996

Recession riceFlooded grassland/swampsFlood forest

Figure 11. Land use change in the surroundings of Tonle Sap Lake in 1986-1996 (source: Degen et al. 2000).

2.2 Impacts

The model includes twenty variables that describe the various environmental, economic and social impacts of sector policies to the Tonle Sap area. The impact variables are listed in Table 3, and described one by one below.

2.21 Environmental impacts

The environmental impacts are classified in two categories, to directly water-related ones and to regional impacts (see Table 3). In brief, their most important characteristics are the following. The descriptions of these issues are kept very concise because they are primarily topics of other parts of WUP-FIN, and used here merely as inputs.

17

Table 3. The impact variables of the WUP-FIN Policy Model for Tonle Sap lake.

Impact class Group Variable Impacts: Environment Water Floods

Suspended solids Eutrophication, low oxygen Toxic substances and oil spills

Regional Soil degradation Weeds Biodiversity decline

Impacts: Natural resources Forests Flood forests Other forests

Agriculture Floating rice Recession rice Dry season rice

Fish Black fish White fish Small fish

Impacts: Economy Subsistence Fisheries Agriculture

Market Fishery (commercial) Cash crop farming Others (tourism, trade etc)

Floods. The Mekong brings every year a great flood to the Tonle Sap. During the monsoon from May to November the discharge exceeds the dry season discharge by 30 times (van Zalinge et al. 2003). The surface area of the lake grows typically from 2,700 km2 to 9,000-16,000 km2. The Tonle Sap river brings the flood waters into the lake, and when the flood reaches the maximum, the river reverses its flow and brings the waters back to Mekong. The difference of the highest and lowest water levels in the lake is from 8 to 10 meters. The floods are the major driving force of the lake’s ecology, as well as to the livelihood of the local population.

There are widespread concerns that the floods will change. One fear is that the upstream dams will be used to cut the flood peak and increase the dry season flows. Another is that the land use changes, particularly clearing of forests for agricultural use will make the floods bigger and subsequently the dry season flows would became smaller. Third fear is that the pattern of the flood will change—the flood peak would become sharper or flatter. Fourth fear is the global climate change that potentially may influence the monsoons.

Common to all these fears is that the level of knowledge on them all is astonishingly low, and they therefore reach mythological dimensions in the media. Not much sound research is available to judge how much justification there is for these fears.

In the model, the flood variable is one of the four environmental variables that can be used in formulating future environmental scenarios. The other ones are suspended solids, eutrophication and overfishing.

18

Suspended solids. Increased erosion of soils in a watershed leads to washoff of suspended matter into streams and lakes. Suspended solids cause many ecological changes besides direct aesthetic and water-use related problems.

Increased sedimentation is harmful to fauna and flora in the bottom. This hampers fish production. Light penetration is less efficient. This is harmful to primary producers. Sediments also contain nutrients, which may lead to increased sedimentation.

On the other hand, sediment fertilizes the floodplain and adds to agricultural productivity. Mekong’s sediments, however, are relatively nutrient-poor. The Mekong carries approximately 170 million tons of sediment per year (Wolanski et al. 1996). This is 3 tons per person, or 215 tons per km2 of catchment area. It has been estimated that 4.5 million tons are trapped in Tonle Sap lake annually (Carbonnel and Guiscafré 1963).

The sedimentation rate in the Tonle Sap lake has been approximated to be around 0.4 mm per year. This is, however, very unevenly distributed. Most silt sediments in the flood plain. There are diverging views on the trend of sedimentation. The fisheries Department has estimated that it has doubled since the 1960s (see McElwee and Horowitz 1999), while MRCS/UNDP (1998) doubts that any remarkable increase has taken place. For more details, see Koponen et al. (2003) and Figures 12 and 13.

Figure 12. Calculated surface water sediment concentration in September 13, 2001 (Koponen et al. 2003).

19

Figure 13. Calculated net sedimentation between May 1 and September 13, 2001. Sediment is important factor in topographic changes, navigation and fisheries and rice production (Koponen et al. 2003).

Eutrphication and low oxygen. Opening material flows of nutrients (nitrogen and phosphorus in particular) cause increasing concentrations of these elements in the aquatic ecosystem, which in turn translates into enhanced growth of algae and macrovegetation. Sources of these nutrients in the Tonle Sap region are in human wastes, manure of the livestock, and most importantly, in agricultural fertilisers.

There are local problems of depletion of dissolved oxygen (DO) in the lake. Most interesting areas are the surroundings of the flooded forests and wetlands, which tend to suffer from DO depletion. Enhanced eutrophication (caused by increased concentrations of plant nutrients in the aquatic ecosystem) is a main reason to this, but other issues such as increased load of organic matter and changes in vegetation cover contribute to this.

Aquatic weeds such as the water hyacinth are known to be very responsive to phosphorus and nitrogen. If the concentrations of these nutrients increase in the lake, the water hyacinth may become rapidly a major nuisance, not only to the open water, but also to rice fields.

Bluegreen algae blooms are a common nuisance in tropical waters. In the Tonle Sap lake, such blooms have occurred and any increase in the phosphorus availability increases the risk of such blooms. A particular problem is caused by the frequent toxicity of bluegreen algae blooms.

For more details on the DO and eutrophication issues, see Koponen et al. (2003) and Figure 14.

Toxic substances and oil spills. There is evidence of growing use of agrochemicals in the basin, as well as the mobilisation of other environmental pollutants such as heavy metals due to various activities. Many of these accumulate in the food chains and ay cause

20

problems in ecological food chains as well as human health in the long term. At this point, the lake is relatively unpolluted. However, the current practices are unsustainable. For instance, the most used pesticides in the region are classified by WHO as being ‘extremely hazardous’.

The navigation on the lake poses a risk for oil spills to the water. This risk is a concern at present and may increase if the traffic on the lake grows due to the construction of a harbor in Siem Reap. In turn, improvement of roads is likely to decrease the risk of oil spills since a part of cargo traffic will likely move on wheels as soon as the national roads 5 and 6 will be paved.

Soil degradation (including groundwater issues). Soil degradation can be due to various reasons. Erosion, caused by either water or wind, is increasing in the Tonle Sap basin. This is chiefly due to deforestation, which enhances water erosion in particular. Chemical problems in the region are due to salinization, to accumulation of toxics, demineralization and so forth. Demineralization is the most important among them in the Tonle Sap Basin. The natural low fertility and acidity are further worsened in some areas due to unsustainable cropping practices.

A natural process which is enhanced by human activity in many tropical areas including the Tonle Sap basin is laterization of soils. Soil loses its structure and hardens so much, that it can be cut into blocks and used after drying as construction material. This laterite has a yellow/red color. Large areas of Southeast Asia and India suffer from this problem. These areas are often called red deserts.

Although the groundwaters of the region are obviously still in fairly good condition, some deterioration is due to the leaching of nutrients and pesticides from agricultural fields, overpumping as well as deforestation. Some features of natural origin cause problems to human uses. These problems include acidity and occasional arsenic risk, which cause problems in certain region. Problems of this kind tend to grow by almost any departure from the condition that has lasted over a long period of time.

After FAO (2000), there exists no detailed investigation of groundwater resources in Cambodia. Aquifers do not have, however, sufficient potential for large-scale irrigation.

The Lower Mekong basin has a plenty of acidic sulphate-rich soils, which are a source of acidic pulses to the waters from time to time. Such pulses are most frequent just after the dry season when the first rains flush oxygenated sulphur compounds from the soil. This is principally a natural phenomenon, yet various human activities that change hydrological conditions, particularly the rhythm of the groundwater table, accelerate the problem.

Weeds. Mekong has so far suffered much less from invading and other aquatic weeds than most other major rivers of the world. This is obviously due to the fairly pristine state of the catchment in comparison to the other rivers. Nutrient levels are relatively low, and other ecosystem changes as well are fairly modest. The water hyacinth (Eichhornia crassipes) is common in the river, but it does not typically grow in such problematic amounts as in most rivers in Southeast Asia. This is obviously due to lower nutrient concentrations in the Mekong than in most other rivers in the region, and even slight eutrophication would cause a remarkable risk for massive outbreaks of the water hyacinth.

21

Figure 14. Calculated surface water and near bottom oxygen concentration in Sept. 13 2001 (Koponen et al. 2003).

22

The most troublesome weed in the Tonle Sap floodplain is Mimosa pigra. It spreads with a considerable speed and constitutes problems for instance as it invades abandoned floating rice fields. It has first been seen in the floodplain in early 1990s. According to Hellsten and Järvenpää (2003) it is widely spread from low-level banks (+1.5 m) all the way to abandoned fields that are under heavy pasture pressure (+6 to 7 m). Areas with dense, existing vegetation have been able to resist the mimosa invasion, and therefore, the northern part of the floodplain is relatively free of the weed, whereas in areas with abandoned fields and flooded grasslands mimosa is a great problem. Most abundant mimosa invasions are in the Pursat area, around Siem Reap, and in the delta area (Hellsten and Järvenpää 2003).

Biodiversity decline. Tonle Sap Lake is a unique freshwater-wetland system with a rich natural biodiversity. It is a focal sanctuary for birds, fish, reptiles, and other animals. It is an important spawning ground and habitat for 150-200 fish species (Tana 1996), and one of the world’s most productive freshwater fisheries. The natural swamp forests and flood forests have declined and are continuously being cleared to give space to agriculture, aquaculture, and other purposes such as the collection if firewood. Wildlife and many plants are harvested excessively in common resource areas.

2.22 Impacts on natural resources

The impacts on natural resources are in three groups: impacts on forests, on agriculture, and on the fish.

Flood forests and other forests. The forestry area of Cambodia was 73 per cent of the country’s land territory in 1963. By 1993, it had dropped to 60 per cent. Cambodia’s deforestation rate is alarming (Figure 15): in 1989-1993, the country lost 3 per cent of its forest each year. However, such estimates are very uncertain—McKenney and Tola (2002) estimated that the forest cover diminishes by 0.5% per year, which is anyway a high rate. Still, the border of Thailand and Cambodia can be easily seen in any map that shows forestry cover: Thailand has cut most of its forests that used to lie in the Mekong basin.

Reasons to deforestation are many. FAO (2000) lists the following ones as the main causes:

• Increased internal demand, which is due to intensive construction activities in the country. Population growth (demanding housing, more firewood and so forth) and replacement of constructions destructed by wars are main drivers.

• Increased external demand. Logging for timber export purposes is poorly under control of the government. The volume is alarming, and particularly hill forests are logged far beyond their sustainable recovery rate.

• Agricultural land extension. The area of agricultural land is growing while the hectare yields remain low. Forest burning has augmented.

• War and large population movements: insecurity and uprooting of large masses of people have caused the spread of destructive practices to exploit the nature.

23

Moreover, the wars destroyed or degraded hundreds of thousands of hectares of prime forest.

• The government lacks a clear forest policy and is incapable of controlling the forest sector.

FOREST AREA Southeast Asia (ha per capita)

0 1 2 3 4 5

Cambodia

Indonesia

Lao PDR

Malaysia

Myanmar

Philippines

Thailand

Vietnam

19801995

Figure 15. Development of forest area per capita in Southeast Asian countries (source: World Bank 1999).

Rice. The agricultural production systems in the Tonle Sap area include a variety of different approaches and crops. Rice is the most important of them. An overview of rice farming is presented in the Agriculture chapter of Sector Policies. The model includes impacts separately to floating rice, recession/wet season rice, and irrigated dry season rice. The land use distribution from the Tonle Sap Area shows clearly the importance of rice farming in the area (Figure 16).

Fish populations. There seems to be two diverging views concerning the state of fish populations in the lake. The fish stocks are already overexploited, and the catches are declining according to some sources (e.g. ADB 2000). The other view is that the regeneration capacity and production of the stocks is fast enough to allow the present level of fishery, with the exception of the use of destructive practices such as electrofishing, small mesh-size nets, explosives, drying of basins, samras technique, and so forth.

Clearly, the species distribution is subjected to changes with heavy fishing pressure, and the average fish size has became smaller (Figure 17). At least the large catfish species have become rare. The fish production seems to be highly dependent on the water level of the lake. In dry years when the flood is smaller, the fish production appears to be much smaller than during wet years.

There seems to be severe shortcomings in the knowledge of the most important food-chains from primary production to the fish.

24

0

500

1000

1500

2000

2500

0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 10-NR

area

(km

2 )

Aquatic vegetationWastelandUrban areasMixed forest areasVillage crop areasPaddy fieldReceding and Floating rice fieldsMarsh and swampAbandoned field covered by grassAbandoned field covered by shrubFlooded grasslandFlooded shrubFlooded forestWater courses

Figure 16. Elevation-habitats relationship in Lake Tonle Sap. Habitats modified from JICA-land-use types (Hellsten and Järvenpää 2003).

The fish are divided into three variables according to the classification of commercial fisheries by the Department of Fisheries: Blackfish, whitefish, and small fish. Details of these groups were presented under Sector Policies.

Fishing effort

Catc

h

Pres

ent f

ishing

effo

rt

Large fish

Small fish

Medium sized fish

Figure 17. Tonle Sap Lake fisheries: hypothetical; state of exploitation of large, medium, and small migratory fish species (source: van Zalinge et al. 2000).

25

2.23 Economic impacts

The model includes five variables for economic impacts. They are subsistence fisheries, subsistence agriculture, large-scale fisheries, market-driven agriculture and other industries such as trade and tourism. Keskinen (2003) presents an comprehensive analysis of the economy of the Tonle Sap area. Figure 18 shows the share of population involved in different industries in the five zones of the Tonle Sap Area. Fishing dominates only in Zone 1, whereas in the Zones from 2 to 4 agriculture is by far the most important employer. In urban areas (Zone 5) other industries employ 80% of the labour force. Markets are significant to fisheries and agriculture in the vicinity of national roads. In addition, in Zone 1, the magnitude of trade is notable (Figure 18). Zone 2 is virtually out of the reach of the markets. Zone 3 is important area for rice farming, and therefore it has a high potential for entering the markets.

Occupations in the Tonle Sap Area

0 %

20 %

40 %

60 %

80 %

100 %

Zone

1

Zone

2

Zone

3

Zone

4

Zone

5

All

zone

s

Others

Trade

Agriculture, hunting andforestryFishing

Figure 18. Involvement of the people of the Tonle Sap Area into different occupational activities (processed by Keskinen 2003 after Population Census of 1998).

2.24 Social impacts

This component of the analysis includes the following variables:

• Human development

• Food security

• Empowerment

• Social and gender equity

• Social cohesion

26

• Formal governance

• Rural push

• Urban pull

Their essentials are the following.

Human development. UNDP defines Human Development as a combination of longevity (health), wealth (Purchasing Power Parity adjusted GDP), and education (primary school enrolment). The Human Development Index (HDI) is calculated as an average of indices for these three components.

In Cambodia, the human development is considerably lower in rural than in urban areas. The Ministry of Planning (2001) has estimated that the HDI for urban areas is about 21% higher than for rural areas.

Human Development Index in Mekong Countries

0 0.2 0.4 0.6 0.8 1

Cambodia

China

Lao PDR

Myanmar

Thailand

Viet Nam

World

Life expectancyEducationGDPHDI

Figure 19. Human development in the Mekong countries (source: UNDP 2002).

Food security. As was shown above, chronic malnutrition is a big problem in the Tonle Sap area. Cambodia is one of the most food insecure countries in Asia (Figure 20). Whereas the Tonle Sap basin is self-sufficient in rice, there are around one million people who cannot produce enough to feed themselves. Since they typically are poor and have nothing to exchange for food, many of them are engaged in cutting forests for firewood and other environmentally destructive activities.

The contribution of fish as animal protein intake is 90% in fishing communities. In other communities of the country, the average is 75%. The food production and distribution systems are extremely sensitive to weather conditions and particularly to the floods of the lake. Floating rice is particularly sensitive. If the flood rises too sharply the harvest is easily lost. Instead, fish catches tend to grow if floods are high.

27

MALNUTRITION(children under 5, %, 1996-98)

0 20 40 60

Cambodia

China

Lao PDR

Myanmar

Thailand

Vietnam

Figure 20. Malnutrition prevalence in the Mekong countries (source: World Bank 2001).

Empowerment. “Help yourself” is one of the most basic compliments of the English language. It is used in situations where somebody offers another person to approach her/himself something that has been set for instance on serve. In a societal context, the term empowerment is somewhat analogical. Instead of keeping the individuals passively waiting for the government or somebody else to serve themselves with amenities of whatever kind, they are enabled and encouraged to be active in the society in a constructive manner.

Cambodia suffers from serious shortcomings in empowerment. It is not easy to measure empowerment, however. It has many aspects. The poor are disabled to improve their livelihood by constraints in their access to resources, by their skills and education shortcomings, and several other reasons. ADB’s (2001) Participatory Poverty Assessment lists the following features of being particular to Cambodia. Many of them are empowerment-related:

• Poor Cambodians say they are worse off now than ever

• Domestic violence is perceived worse now than in the past.

• The poor lack food security; many people still experience periods of hunger

• People have a desire to participate in their communities’ activities but cannot because of the immense amount of time spent looking for food.

• People are being denied access to their traditional hunting and gathering grounds by poor implementation of government policies, antilogging measures, co-option of land and other activities which are also prohibiting people’s access to land.

• The poor want information about government policies and programs including health, vocational and technical training, market information, and micro-finance availability

• The poor have stated there is a general lack of confidence in local and provincial administration.

28

Social and gender equity. Cambodia has large disparities in economic, human and gender development (Figure 21). In urban areas, the Human Development Index (HDI) is 25% higher than in rural areas. The HDI of the richest 20% of Cambodians is almost 50% higher than that of the poorest 20%. The life expectancy of the poorest quintile is 52 years, and 59 years for the richest 20%.

In the Tonle Sap area, poverty is widespread (Figure 22). Over one-third of the population lives under 2,100 calories per day. Zones that are close to the lake (1 and 2) show rather homogenous incidence of poverty. There are only few communes that have less than 20% of their population below the poverty line. Also no villages exist with over 60% being poor. The situation is different—far more polarized—in zones 3 and 4. Those zones have many communes (one-fifth) that have low poverty incidence. In contrast, a considerable share of communes is extremely poor with over 60% of people being poor. In urban communes, poverty incidence is far lower than in rural ones.

Economical inequality in Cambodia

0%

20%

40%

60%

80%

100%

Nat

iona

l

Phno

mPe

nh

Oth

erU

rban

Rur

al

Richest 10%

Poorest 10%

Schooling rates by agePer cent

0

20

40

60

80

100

6 7 8 9 10 11 12 13 14 15 16 17 18

MaleFemale

Figure 21. Inequality in Cambodia. Left: the consumption share of the richest and poorest 10% of the population. Right: The schooling rate of boys and girls by age (source: Ministry of Planning 2001).

0%

10%20%

30%

40%

50%60%

70%

80%90%

100%

Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 All zones

0 - 20%20 - 40%40 - 60%60 - 80%

Figure 22. Poverty profile of the five zones around the Tonle Sap Lake. Different colours: Percentage of poor in a commune. Vertical axis: percentage of communes with different poverty levels. Derived from WFP (2000) by Keskinen (2003).

29

In brief, poverty is widespread in rural areas but not so pronounced in urban ones. Areas close to the lake are relatively homogenous in terms of poverty, but contrasts grow towards the national roads and provincial towns.

Social cohesion. Cambodia has been suffering from serious hostilities and instabilities over the past four decades. The situation has undergone a gradual stabilization process since the early 1980s, and the society and the political system have recovered reasonably. However, there are still plenty of economical, political, regulatory, and many other functions and sectors of the society that are either governed by the informal—often illegal—actors, or are seriously corrupt. For instance some of the fishing lots and permissions are in the hands of illegal, often violent groups.

UNDP (1999) reports that the Khmer villages do not have traditionally a high social cohesion and solidarity. In this respect, there is an important difference to for instance China, Vietnam or India, where community solidarity is much stronger than in Cambodia. A Khmer peasant has an individualistic culture. The violence in the last decades, and particularly the Khmer Rouge rule have further destroyed traditional social networks dramatically. At present, perhaps the most important institution of social solidarity is the Buddhist religious order.

Formal governance. This issue was touched in the Sector Policy section, under the title Formal Institutions.

Migrations: Rural push and urban pull. Urbanization rate in Cambodia is still very low. Only around one out of four Cambodians dwell in urban areas. The tranquility of the Khmer towns seems to give a false impression that urbanization is not an issue in the country. In contrary, it can be expected that there will be a massive migration into urban areas in the coming years and decades. Varous sources propose that Cambodia’s urban population might grow two or even fourfold by 2025.

It is common to divide the reasons to urbanization into those that push people away from their villages (rural push) and those that attract people into towns and cities (urban pull). The outcome of these two processes is very different. Rural push pushes poor, often landless people to the urban squatters, without much hope for the better. Urban pull, in contrast, brings better-schooled people to urban jobs. It is critical to the development of the towns and cities of Cambodia which one of these two reasons dominate in the future.

Haapala (2003) made a comprehensive analysis of migration and its root causes in the Tonle Sap Area. Approximately 1-3% of population from the area has left the villages in the past 5 years. Phnom Penh has become the home of two-thirds of them. Thailand, Koh Kong Province, and the provincial capitals along national roads 5 and 6 have been important as well. Rural push has dominated over urban pull. Factors such as poverty, landlessness, food insecurity, exceptional rainfalls, floods and droughts have been the most important contributors to rural push.

30

3 SCENARIO ANALYSIS

3.1 Definition of development scenarios

Two types of development scenarios are included in the model. Those are policy scenarios and environmental scenarios.

The policy scenarios under study are based on the different weightings of the three goals defined in the Mekong Agreement of 1995. Those goals are environmental sustainability, economic growth and poverty reduction. One scenario that clearly prioritizes each of these goals is constructed. In addition, an integrated scenario is produced that makes a realistic compromise between the three other scenarios, in order to find win-win solutions between these three conflicting development goals.

These scenarios are presented in Table 4, by using the notation defined in Table 5.

The environmental scenarios include the following four fears for the future development of the environment in the Tonle Sap area:

• Scenario 1: Hydrological changes, particularly CHANGING FLOODS > a “realistic yet characteristic” future situation > inundated area, water quality, fish, vegetation effects > then assessment of livelihood changes (fisheries, agriculture etc).

• Scenario 2: EROSION INCREASING > impacts on the basin morphology (filling), water quality, spawning, production etc. Both local and upstream effects are included.

• Scenario 3: EUTROPHICATION GROWING > impacts on oxygen conditions, phosphorus balance, algal growth and shift to bluegreen algae, spread of water hyacinth.

• Scenario 4: SERIOUS OVERFISHING > destructive catching practices continue and worsen (electrofishing, drying etc), the management system remains unsustainable (lot system with small contracts, no involvement and access of local communities, partially informal and illegal managers).

These scenarios were simply constructed by imposing the marks >>> and ***** for the respective variable in the model.

The environmental scenario results were clearly dominated by the flood issue. This report omits further scrutiny of the analysis since the input from other parts of the WUP to analysis was non-existing, and therefore the results of the WUP-FIN Policy Model are at this stage highly speculative.

31

Table 4. The four development scenarios investigated with the WUP-FIN Policy Model. The notation used is explained in Table 5.

Conservation Economic growth Poverty reduction Integrated Mean Accuracy Mean Accuracy Mean Accuracy Mean Accuracy Society Formal institutions > ** > *** > ** > *** Education, public health > ** > ** >> **** > ** Urban development *** >>> **** * >>> **** Rural development (villages)

*** * >>> **** *

Water, infrastructure Water and sanitation services

** > * >>> **** > *

Navigation *** >> **** > * >> **** Roads > *** >>> ***** > ** >>> ***** Agriculture, fisheries Agriculture, irrigation *** >> ** > *** >> ** Large-scale fisheries << **** >>> *** < ** >>> *** Small, medium scale fisheries

*** < ** >>> *** < **

Environment Conservation >>> ***** *** > ** ***

Table 5. The semi-quantitative scale used in the initial values for the model.

Mark for change

All variables Mark for level of knowledge

Policy variables: credibility for the realization of a policy option

Other variables

no change Irrealistic Not known > small increase * 20% realistic 20% known >> modest increase ** 40% realistic 40% known >>> large increase *** 60% realistic 60% known

**** 80% realistic 80% known <, <<, <<< as above but decrease ***** Will certainly come true Fully known

3.2 Results

3.21 Policy scenarios: Vulnerability of geographic zones

Each of the three goals for development, economic growth, poverty reduction and environmental sustainability are important to the development of the Tonle Sap Area. In many respects they conflict but compromises are possible, or in fact even necessary for balanced development of the area.

When analyzing the vulnerability of the five geographical zones of the Tonle Sap Area, the first remark is that the vulnerability can be reduced the more the closer the zone is to the national roads (Figure 23). The uncertainties involved are, however, strikingly high. Other remarkable results are the following:

32

VULNERABILITY: ZONE 4

C E P I

VULNERABILITY: ZONE 1

C E P I

VULNERABILITY: ZONE 2

C E P I

VULNERABILITY: ZONE 3

C E P I

VULNERABILITY: URBAN

C E P I

LEGENDS

C E P I

ScaleUncertainty

Mean

>>>

<<

<

>

>>

<<<

CONFIDENCE SCALE

***** **** *** ** *

Figure 23. Decrease in vulnerability of the five zones of the Tonle Sap Area with respect to the four development scenarios: C = Conservation, E = Economic growth, P = Poverty reduction, and I = Integrated. The legends are shown in the lower two diagrams.

• Environmental sustainability scenario: vulnerability goes down only slightly. For Zone 1, there is not much change from the present level, in the others some improvement takes place.

• Economic growth scenario: higher uncertainties than in other scenarios.

• Poverty reduction scenario: best among the three basic scenarios in terms of reducing vulnerability, particularly in zones closer to the lake.

• Integrated scenario: Better than any of the three basic scenarios. More reduction of vulnerability and less uncertainties.

3.22 Policy scenarios: The three development goals

With respect to the three development goals, the policy scenarios give interesting outcomes. As can be expected, the economic growth scenario gives the best response in the goal Economic growth. However, the difference is not too big to the Poverty reduction scenario. In fact the latter one contains much less uncertainties than the former one, and is therefore even better in a sense. What comes to the other goals and scenarios,

33

Figure 24 shows and Table 6 summarizes the results. The most important issues are the following:

• Economic growth scenario contains again high uncertainties.

• Conservation scenario is not very favorable in achieving the goal Environmental sustainability, since it obviously ignores the villages and their development.

• Poverty reduction scenario gives relatively good results in all respects, but it can be remarkably improved in the Integrated scenario by putting more attention to institutional development and some other issues that are obviously undervalued in the Poverty reduction scenario.

ECONOMIC GROWTH

C E P I

POVERTY REDUCTION

C E P I

ENV'TAL SUSTAINABILITY

C E P I

Figure 24. Results in the three development goals of the Tonle Sap Area with respect to the four development scenarios: C = Conservation, E = Economic growth, P = Poverty reduction, and I = Integrated. The legends are shown in Figure 23.

Table 6. Results of the four development scenarios with respect to the three development goals: economic growth, poverty reduction and environmental sustainability.

Goal \ Scenario Economic growth scenario

Conservation scenario

Poverty reduction scenario

Integrated scenario

Economic growth Slightly better average than in the two other basic scenarios. Very high uncertainty.

Lowest average and relatively high uncertainty.

Lowest uncertainty and not much lower average than in the Economic growth scenario.

Equal in average to the best of the basic scenarios (Economic growth) but much lower uncertainty.

Environmental sustainability

Sustainability goes down with all scenarios. Environmental sustainability scenario is slightly better in average but contains high uncertainties. Poverty reduction scenario and particularly integrated scenario contain less uncertainty and are therefore the best ones.

Poverty reduction Almost similar except economic growth scenario is more uncertain and slightly better in terms of average.

By far the best basic scenario.

Still better than the Poverty reduction scenario: less uncertain and better average.

Summary: ranking 3 4 2 1

34

3.23 Sensitivity analysis

A sensitivity analysis property was also constructed to the model. A small perturbation is made to the prior probability distribution of each of the sector policies, and the response can be observed graphically, as in Figure 25. The plot shows the sensitivity of the development goals to changes in sector policies. It shows how much the realization of these development goals responds to small changes in sector policies. The further the line is away from the zero line (the circle denoted with 0.0%), the more impact the policy has on the respective development goal. The impact is negative if the line goes inside the circle, otherwise it is positive.

The sensitivity plot changes always when the scenario is modified. It provides important information about how the scenario could be modified in order to make it better with respect to each of the three development goals. In the plot of Figure 25, investing in Rural development, Education, public health, Small, medium scale fisheries, and so on, would yield good results in Poverty reduction. Urban development and Roads, in turn would benefit Economic growth. Strengthening Formal institutions would be very beneficial to Environmental sustainability.

It must be emphasized that the sensitivity results shown in Figure 25 are very much specific to the Economic growth scenario as defined in Table 4. Other scenarios produce different sensitivity results.

SENSITIVITY ANALYSIS

-4.0 %-2.0 %0.0 %2.0 %4.0 %6.0 %8.0 %

10.0 %Formal institutions

Education, public health

Urban development

Rural development (villages)

Water and sanitation services

Navigation Roads

Agriculture, irrigation

Large-scale fisheries

Small, medium scale fisheries

Conservation

Economic growth

Poverty reduction

Environmental sustainability

Figure 25. Results of the sensitivity analysis for the Economic growth scenario.

35

3.3 Conclusions

The starting point of the policy analysis of the Tonle Sap Area was the Mekong Agreement of 1995, and particularly the Vision defined for the Mekong Basin. It specifies three development goals for the basin. They are economic growth, poverty reduction and environmental sustainability.

The possibilities to find combinations of sector policies for achieving these, often conflicting goals were analysed systematically using a probabilistic, Bayesian network model. The model was specifically tailored for this purpose within the WUP-FIN Modelling Project.

Policies targeting to any of these three goals separately were investigated. In addition, a compromise policy with a goal of finding as many win-win benefits as possible from the three conflicting policies was constructed by using the model. On the basis of the results it is clear that a compromise policy is possible, and it is by far more balanced and acceptable than any of the policies that target on one of the three policy goals of the Mekong Agreement at a time.

The approach turned out to be very useful in an analysis of this kind. It suits well to the purpose, complemented by other approaches used within WUP-FIN, and particularly those used within the Socio-economic part of WUP-FIN.

36

REFERENCES

ADB 2000. Protection and Management of Critical Wetlands in the Lower Mekong Basin-Interim Report ADB TA No 5822-REG. Asian Development Bank, Manila.

ADB 2001. Participatory Poverty Assessment in Cambodia. Asian Development Bank, Manila.

Bonheur, N. 1996. Socio-economic study around Tonle Sap lake. In: Tonle Sap Watershed, Paper Presentations from the Tonle Sap Technical Workshop, 26 March, Ministry of Environment.

Carbonnel, J.P. and Guiscafré, J. 1963. Grand Lac du Cambodge : Sédimentologie et Hydrologie 1962-1963. Mininstere des affaires étrangeres ? Comite du Mekong / Gouvernement du Cambodge.

Deap, L., Ly, S. and Van Zalinge, N.P. (Eds.) 1998. Catch Statistics of Cambodian Freshwater Fisheries. MRC/DoF/DANIDA Project for the Management of the Freshwater Capture Fisheries of Cambodia. Mekong River Commission, Phnom Penh.

Degen, P., van Acker, F., van Zalinge, N., Thuok, N. and Loeung, D. 2000. Taken for granted: Conflicts over Cambodia’s freshwater fish resources. 8th IASCP Conference, Bloomington, IN, 31 May - 4 June, 2000.

FAO 2000. Environment: Concepts and Issues. FAO, Phnom Penh. www.un.org.kh/fao/environment/index.html

Haapala, U. 2003. Migration and Urbanisation in Cambodia. WUP-FIN Socio-Economic Studies on Tonle Sap 9. MRCS/WUP-FIN, Phnom Penh.

Hellsten, S. & Järvenpää, E. 2003. Preliminary observations of floodplain habitats and their relations to hydrology and human impact. WUP-FIN/MRC/Finnish Environment Institute, Phnom Penh.

Jacobs, J.W. (1995). Mekong Committee history and lessons for river basin development. The Geographical Journal 161: 135-148.

Keskinen, M. 2003. Socio-Economic Survey of the Tonle Sap Lake, WUP-FIN Socio-Economic Series on Tonle Sap 8, MRCS/WUP-FIN, Phnom Penh.

Koponen, J., Josza, J., Lauri, H., Sarkkula, J. & Virtanen, M. 2003. Modelling Tonle Sap Watershed and Lake Processes for Environmental Change Assessment. WUP-FIN/MRC/Finnish Environment Institute, Phnom Penh.

Kosal, M. 1998. Social Implication of conservation of the designated “Prek Toal” core Area of the Tonle Sap Biosphere Reserve, Cambodia. 1998 UNESCO-MAB

37

Young Scientists Research Award Report. Department of Nature Conservation and Protection, Ministry of Environment, Phnom Penh, Cambodia. www.unesco.org

Lamberts, D. 2001. Tonle Sap Fisheries: A Case Study on Floodplain Gillnet Fisheries. RAP Publication 2001/11, FAO, Bangkok. http://www.fao.org/DOCREP/004/AB561E/ab561e00.htm#Contents

Ledgerwood, J.L. 1999. Rural development in Cambodia: the view from the village. In: Brown, F.Z. & Timberman, D.G. 1999. Cambodia and the International Community: The Quest for Peace, Development and Democracy. Asia Society, New York. http://www.asiasociety.org/publications/cambodia/index.html#Contents

McElwee, P. and Horowitz, M.M. 1999. Environment and Society in the Lower Mekong Basin: a Landscaping Review. IDA Working Paper 99, Mekong River Basin Research and Capacity Building Initiative, Oxfam America SEA 15 / 97-99.

McKenney, B. and Tola, P. 2002. Natural Resources and Rural Livelihoods in Cambodia: A Baseline Assessment, Working Paper 23, Cambodia Development Resource Institute, Phnom Penh.

Ministry of Planning 2001. Cambodia Human Development Report 2001. UNDP / The Ministry of Planning, Phnom Penh.

MRC 2000. Strategic Plan for the Implementation of the 1995 Mekong Agreement. Mekong River Commission Secretariat, Phnom Penh.

MRC 2001. Strategic Plan 2001-2005. Towards and Economically Prosperous, Socially Just and Environmentally Sound Mekong River Basin. Mekong River Commission Secretariat, Phnom Penh.

MRCS/UNDP 1998. Natural Resources Based Development Strategy for the Tonle Sap Area, Cambodia. Mekong River Commission Secretariat / United Nations Development Programme, Phnom Penh.

Rainboth, W.J. 1996. Fishes of the Cambodian Mekong. FAO Species Identification Sheets for Fishery Purposes. FAO, Rome.

Sen, H. 2002. Selected Ad-lib Address to the Groundbreaking Ceremony for the Rehabilitation of the National Road 6 between Prey Romeas (at the border of Siemreap and Kompong Thom Province) and Roluos (in Siemreap). The Royal Government of Cambodia, Phnom Penh. http://www.cnv.org.kh/2002_releases/110602_road6_grdbreaking_roleas_romeas_adlib.htm

Tana, T.S. 1996. Status of biodiversity of the Great Lake (Boeung Tonle Sap), an approach for better conservation and future sustainable development. In: Tonle Sap Watershed, Paper Presentations from the Tonle Sap Technical Workshop, 26 March, Ministry of Environment.

UNDP 1999. Cambodia Human Development Report 1999. Ministry of Planning/UNDP, Phnom Penh.

UNDP 2002. Human Development Report 2002. United Nations Development Programme, New York.

38

Van Zalinge, N.P., Nao, T., Touch, T.S. and Deap, L. 2000. Where there is water, there is fish? Cambodian fisheries issue in a Mekong River perspective. In: Ahmad, M. and Hirsch, P. (Ed.): Common Property in the Mekong: Issues of Sustainability and Subsistence: 37-48. ICLARM Studies and Reviews 26.

van Zalinge, N., Loeung, D., Pengbun, N., Sarkkula, J. & Koponen, J. 2003. Mekong flood levels and Tonle Sap fish catches. Second International Symposium on the Management of Large Rivers for Fisheries. 11-14 February, Phnom Penh, Cambodia.

Varis, O. 1998. A belief network approach to optimization and parameter estimation: application to resource and environmental management. Artificial Intelligence, 101: 135-163.

Varis, O. & Fraboulet-Jussila, S. (2002). Water resources management in the lower Senegal River basin—conflicting interests, environmental concerns, and policy options. International Journal of Water Resources Development 18: 245-260.

Varis, O. & Lahtela, V. 2002. Integrated water resources management along the Senegal River – Introducing an analytical framework. International Journal of Water Resources Development 18: 501-521.

WFP 2000. Poverty Data for All Communes 2000, The United Nations World Food Programme (WFP), Phnom Penh.

Wolanski, E.N., Nhan, N.H. and Spagnol, S. 1996 Fine sediment dynamics in the Mekong River Estuary, Vietnam. Estuarine, Coastal, and Shelf Science 43(5): 565-582.

World Bank 1999. World Development Indicators. World Bank, Washington D.C.

World Bank 2001. World Development Indicators. World Bank, Washington D.C.