INTERMEDIATE REPORT Vietnam case EXPLORING THE RELEVANCE AND FEASIBILITY OF PES APPROACHES FOR PRODUCING ENVIRONMENTAL SERVICES THROUGH CHANGES IN AGRICULTURAL PRACTICES: A CASE STUDY IN THE MEKONG REGION Contract C-056-07, CPWF theme 2 (Challenge Program for Water and Food) Orange D., Nguyen Duy Phuong, Loiseau J.B., Bui Tan Yen, Henry des Tureaux T., Bardouin L., Rodriguez C., Bertrand J., Grandidier E., Jouquet P. and Tran Duc Toan Hanoi, July 2008 Contact : Didier Orange (orange@ird.fr), IRD-IWMI, posted in SFRI, Hanoi, Vietnam INTRODUCTION   In  Vietnam,  the  soil  erosion  linked  to  the  agricultural  practices  is  a  problem  especially  serious  because of many factors as steep topography, fragile soils, climate change and human pressure due  to both strong demographic increase and huge economic development.  In the mountainous area in  Northern  Vietnam,  agricultural  productivity  has  already  been  decreasing  due  to  erosion  of  fertile  soil. Today, in order to confront the problem of soil fertility loss and the need of rural development,  the  farmers  under  the  pressure  of  the  agricultural  policy  choose  to  diversify  their  activities  and  notably to develop cattle husbandry.    In this context of severe agricultural intensification, soil erosion has become a major economic and  environmental  problem.  It  has  affected  the  livelihoods  of  smallholders,  has  increased  soil  degradation  and  has  hindered  the  long‐term  development  of  these  mountainous  areas.  Many  different  organisations  have  responded  to  this  by working  on  the  development  and  promotion  of  better  soil management  practices,  aiming  at  higher  yields  and  reduction  of  rural  poverty without  concrete farmers’ involvement leading to a limited technical adoption. Indeed, research studies (i.e.  MSEC program) have shown that the promotion of a new practice, as a major factor towards change,  needs to be complemented by other factors, such as access to external markets or better wellbeing,  which would contribute to achieving higher incomes (Orange et al., 2008). In addition, policies and  regulations have proved to have a further positive impact on the adoption of a given technique.     In this term, the concept of environmental services (ES) and their role in providing significant public  goods is increasingly gaining recognition in Vietnam. The final objective of the present study is: to  assess whether environmental sustainability is socially acceptable and valued by communities  in order to elicit payment for these services. The objective is to explore the potential use of the PES  concept  (Payment  for  Environmental  Services),  in  its  broadest  sense,  as  an  environmental  management tool  to  lead upland  landholders to reduce the occurrence and  impact of agricultural  practices within sloping  lands on soil erosion processes, soil  fertility decrease and water pollution  through in‐site and off‐site effects.    This intermediate report presents the one‐year on‐going research activities conducted in Vietnam by  the  MSEC  team.  As  planned  (cf.  annex  1),  the  research  activities  have  been  divided  on  two  workgroups:   WG1: Hydrological modeling for PES concept  WG2: Farmers’ demand and PES modalities    The present report describes the main results from these two workgroups. Some extensive versions  of scientific results are presented in annex. At last, two papers have been published concerning this  PES study:  ‐  Orange D., Tran Duc Toan, Nguyen Duy Phuong, Nguyen Van Thiet, Salgado, Clement F., Le Hoa  Binh (2008). Different  interests, common concerns and shared benefits. LEISA Journal, 24(2):  12‐13.  ‐  Orange  D.,  Bardouin  L.,  Nguyen  Duy  Phuong,  Loiseau  J‐B.,  Clement  F.,  Jouquet  P.  (2007).  Le  concept de PES pour une gestion durable des eaux et des sols : application au développement de  l’élevage  et  au  contrôle  environnemental  dans  le  Nord  Vietnam.  In  Orange  D.,  Roose  E.,  Vermande  P.,  Gastellu  J.‐P.,  Pham  Quang  Ha  (Eds) :  Gestion  intégrée  des  eaux  et  des  sols :  ressources, aménagements et  risques en milieux  ruraux et urbains, Proceedings Ed. AUF and  Ed.  IRD,  Paris,  Actes  des  Premières  Journées  Scientifiques  Inter­Réseaux  de  l’AUF,  Hanoi,  6‐9  novembre 2007, cd‐rom : 8p. (Communication orale).  2 BACKGROUND ON NORTHERN VIETNAM AND PES INSTITUTIONAL CONTEXT    This  work  was  part  of  an  international  research  programme  called  MSEC  (Management  of  Soil  Erosion Consortium)  for  his  part  occurring  in  Northern  Vietnam.  Its  objectives  were  to  promote  sustainable land management systems, evaluate the biophysical, environmental and socio‐economic  effects of soil erosion, and to generate reliable information for the improvement of local policies and  regulations, in the uplands of Northern Vietnam. After a four year assessment, working with around  50  small  scale  farmers  in  the  commune  of  Tien  Xuan  (approximately  50  km  west  of  the  capital  Hanoi), this study highlighted that local factors can be far more important than national policies in  determining change. These results provided the basis for an integrated crop‐livestock project carried  out  by  the  Vietnamese  Soils  and  Fertilizers  Research  Institute  (SFRI)  in  Vietnam  and  Laos,  in  collaboration with the National Institute of Animal Husbandry (NIAH) in Vietnam and the National  Agriculture and Forestry Research Institute (NAFRI) in Laos. In addition, two international research  centres  were  involved,  IRD  (French  Institute  of  Research  for  Development)  and  CIRAD  (French  Agricultural Research Centre  for  International Development). The aim was  to support farmers in  their  efforts  to  reduce  poverty  and  enhance  environmental  sustainability  in mountainous  areas by promoting new technologies that build on existing knowledge and farming practices.    The  research  focused  on  the  integration  of  animal  husbandry  into  the  traditional  agricultural  farming systems in the region, which are based on cassava production and forestry in the uplands,  and  rice  cultivation  in  the  lowlands. A discussion process between  farmers,  local  decision‐makers  and  scientists  helped  define  the  activities,  to  ensure  all  their  goals  would  be  met.  In  Vietnam,  improved  management  of  soil  fertility  in  rice‐  and  cassava‐based  systems,  and  simultaneously  cultivating fodder grass on steep slopes matched both the farmers’ and scientists’ interests. The plan  was  to  produce  sufficient  animal  feed  during  the  cold winter  season  and  the warm  rainy  season,  while at the same time decreasing soil erosion. The project’s activities, from 2005 to 2008, included:  •  testing  fodder  species  for  sloping  lands  on  experimental  plots  and  demonstration  sites,  considering  temperate  grasses  and  legumes  (such  as  Avena strigosa or Medicago sativa)  and  also tropical species (Panicum maximum, Brachiaria sp., Paspalum atratum or Stylosanthes sp.).  The main purpose was to secure the production of cattle feed, (especially in winter), and also to  select the best species for soil conservation during the rainy season;  •  setting  up  demonstration  sites,  focused  on  soil  and  nutrient  management  when  growing  cassava (on upland plots) and paddy rice (in the lowland areas). The aim was to show how well  balanced fertility management can  improve crop yields without  increasing the use of external  inputs.    WHY PES CONCEPT IN NORTHERN VIETNAM?    Soil erosion in highly incised watersheds of North Vietnam has negative impacts on both upstream  and downstream communities. Increased sediment discharged from these catchments reflects a loss  of  fertile  topsoil  that  farmers  depend  upon  for  the  production  of  crops  and  fodder.  In  addition,  increased sediment loads in streams and rivers have a direct negative impact on water quality and  the  longevity  of  water  storage  structures,  both  of  which  have  significant  economic  implications  (Orange et al., 2007a).    3 Soil  erosion  comes  about  because  of  both  human  activity,  such  as  erosive  farming  practices,  and  biophysical  factors,  including  intense  rainfalls  on  sloping  land.  In  Vietnam,  increasing  population  densities  has  encouraged  agricultural  production  to  shift  from  the  rich‐soil  lowlands  to  the  degradation‐prone  marginal  uplands.  In  this  context  of  severe  agricultural  intensification,  soil  erosion has become a major economic and environmental problem. It has affected the livelihoods of  smallholders and has hindered the long‐term development of these areas (Orange et al., 2008).    It is well recognized that non‐appropriated utilization of land resources in upper catchments results  in enhanced erosion processes on sloping lands leading to increase nutrient  loss and a decrease in  soil  fertility,  especially  in  the  mountainous  area  of  Northern  Vietnam  since  soils  are  acidic  and  sensitive  to  erosion  due  to  poor  cover.  For  more  than  20  years,  the  observed  effect  is  more  unsustainable agricultural system within the sloping lands leading to larger sediment discharge and  elevated nutrient loads being discharged into the water bodies and the paddy fields in the down part  (picture  1).  The  huge  off‐site  risk  is  to  reduce  water  quality  to  downstream  users  that  can  have  significant health implications, the storage capacity of reservoirs and to unfertilized the paddy fields  (Tran Duc Toan et al., 2003).         Picture 1: Land‐use through to the top of the hills and erosion contribution to the paddy fields and the  water bodies. (Northern Vietnam, June 2008, D. Orange)      The main  conclusions  generated  from  the Management of Soil Erosion Consortium  (MSEC,  research  program from IWMI‐IRD) over the  last  five years  from 27 catchments  in  five countries (Indonesia,  Laos, Philippines, Thailand and Vietnam) were (Valentin et al., 2008):   4 (i)  soil  erosion  is  predominantly  influenced  by  land  use  rather  than  environmental  characteristics not only at the plot scale but also at the catchment scale;   (ii) slash and burn shifting cultivation with sufficiently long rotations (one year of cultivation,  eight years of fallow) is too often unjustly blamed for degradation;   (iii) in its place, continuous cropping of maize and cassava promotes high rates of soil erosion  at the catchment scale;   (iv) conservation technologies are efficient in reducing runoff and total sediment yield at the  catchment scale;   (v)  the  adoption  of  improved  soil  management  technologies  by  upland  farmers  is  not  a  function of the degree of intensification of their farming system and/or of their incomes;   (vi) a failure to adopt appropriate land use management strategies will result in further rapid  resource degradation with negative impacts to downstream communities.    A key outcome from previous research activities has been a need to focus on water quality issues at a  larger scale and the importance of empowering communities in taking effective action. The current  failure  between  state  and  market  mechanisms  to  promote  sustainable  and  equitable  natural  resource  management  in  developing  countries  is  mainly  due  to  attaining  a  balance  between  the  rights  of  society  and  the  rights  of  the  individual. Key  factors  that  influence  the  development  of  individual and community based initiatives that address some of these imbalances are (Noble et al.,  2006): (1) leadership, (2) quick and tangible outcomes, (3) supportive policy, (4) social capital, (5) a  participatory  approach with  respect  to  the  implementation  of  the  project,  and  (6)  innovation  and  appropriate technology.    Our  previous  collaborative  research  undertaken  within  IWMI  Asialand Sloping Land Management  (ASL),  the Management of Soil Erosion Consortium  (MSEC)  and  the  DURAS­CropLivestock  projects  have  focused  on  elucidating  the  processes  associated  with  soil  erosion  and  land  and  water  management in highly incised catchments within Southeast Asia. A key outcome from these research  activities  was  a  need  to  focus  on  water  quality  issues  at  a  larger  scale  and  the  importance  of  empowerment of communities in taking effective action (Clement et al., 2007).    Getting farmers to adopt new technologies to address soil erosion and fertility problems is not easy.  In  Vietnam,  a  multidisciplinary  research  project  to  improve  soil  management  in  traditional  mountainous agricultural farming systems managed to attract farmers’ interest and stop soil erosion.  This  success  stems  from  encouraging  farmers,  extensionists  and  researchers  to  jointly  define  and  implement the project. Their different aims could be followed simultaneously: scientific results  for  researchers,  better  agricultural  practice  for  extension  workers,  and  economic  success  and  free  choice for farmers (Orange et al., 2008).    Recent studies on governance, poverty and environmental sustainability in Northern Vietnam have  emphasized a ‘‘rights‐based’’ approach, in which equitable development is strongly associated with  individual  and  communal  rights  (Castella  et al.,  2002; Clement et al., 2007).  The  current  failure  between  state  and  market  mechanisms  to  promote  sustainable  and  equitable  natural  resource  management in developing countries is mainly due to the balance between the rights of society and  the rights of the individual (Li, 2002; Clement et al., 2008).    Sustainable  land  use  practices  are  now  relatively well  known  and  include  forest  plantation,  agro‐ forestry  systems,  tree‐based  land  use  alternatives  and  agro‐ecologically  sound  practices  such  as  conservation agriculture (Baier and Dumanski, 1991; Valentin et al., 1991; Geay and Dao The Tuan,  2000;  Cramb,  2005;  Orange  et  al.,  2007a).  Although  these  systems  would  provide  long  term  benefits  to potential adopters,  they often do not provide sufficient  immediate economic returns  to  5 resource‐constraint  farmers,  i.e.  quick  and  tangible  returns  on  investments  (Affholder  et  al.,  in  press).  Often  the  benefits  associated with  the  adoption  of  new  and  improved  approaches  to  land  management  are  not  evident  for  3‐5  years.  Further,  the  environmental  services  of  these  land  use  options  provide,  i.e.  erosion  control,  watershed  services,  biodiversity  conservation  and  carbon  storage are usually not rewarded and only  indirectly connected  to economic activities. As a result,  adoption  of  improved  land  and water management  systems  has  to  date  been  low  (Orange et al.,  2008).    Payments  for  Environmental  Services  (PES)  or  other  incentive  mechanisms  present  a  new  approach that focuses directly on creating a conditional benefit transfer between providers of  the  environmental services and the beneficiaries of these services. PES schemes are based on the  principles that those who benefit from environmental services should pay for them, and that those  who contribute to generating these services should be compensated for providing them. Hence, the  approach  seeks  to  create  mechanisms  that  internalize  what  would  otherwise  be  an  externality  (Pagiola and Platais, 2002; Tomich et al., 2004). Such schemes can take advantage of upland‐lowland  interactions  in  generating  environmental  benefits  while  improving  the  livelihoods  of  upper‐ catchment  agricultural  households.  As  such,  PES  is  also  increasingly  seen  as  an  opportunity  for  poverty  reduction  and  the  enhancement  of  sustainable  development  within  integrated  natural  resource management approaches (Asquith et al., 2002; Pagiola et al., 2005; Minang et al., 2007).     The past few years have witnessed a surge in interest in the development of PES schemes in Asia. In  Vietnam,  while  some  projects  using  the  conceptual  framework  of  PES  are  being  initiated  in  the  Central and Southern part of the country (e.g. WWF, 2007) and are associated with the protection of  native  forests  or  the  establishment  of  production  forests,  no  PES  schemes  are  currently  being  implemented in the upper catchment areas of Northern Vietnam that specifically address sustainable  agronomic  production  systems.  However,  the  Vietnamese  Government  expressed  recently  its  interest  in  starting  such  a  scheme  to  protect  fragile  upper‐catchments  from  degradation  and  to  enhance the livelihoods of communities (Dang Kim Son, 2008; Vu Tan Phuong, 2008).    A  key  factor  shown  to  directly  influence  farmer  decisions  in  adopting  more  sustainable  approaches in soil and water management was quick and tangible outcomes upon adoption.    Conclusion 1:   In Northern Vietnam, soil and water resources tend to be used unsustainably, partly due  to  the  lack  of  good  soil  in  lowlands  and  moreover  due  to  the  lack  of  knowledge  on  the  environmental  sustainability.  In  the Vietnamese  context,  there  is  considerable  evidence  to  suggest  that more effective  land management practices can be promoted  in  the uplands.  In  this  relative  favorable  context,  payment  for  environmental  services  (PES)  is  seen  as  an  efficient way to potentially: (i) cover the cost of these practices; (ii) promote a dialog between  upland  farmers,  lowland  farmers and  the  local  stakeholders  in order  to  target  environment  services  that  are  economically  realistic  as well  as  environmentally,  socially  and  culturally  acceptable.   6 CASE STUDY IN NORTHERN VIETNAM PRINCIPLES     Because  the  current  upland  management  practices  should  be  detrimental  to  water  use  for  agricultural and domestic purpose downstream and possibly confer other off‐site impacts. Because  upland  agricultural  practices  on  sloping  lands  should  be  not  sustainable  and  economic  for  the  uplanders. Because agricultural and other uses of water downstream should be adversely affected by  the  negative  externalities  generated  from  upland  uses  that  have  apparently  evolved  to  tackle  the  livelihood needs of upland farmers. The assumption of our study is that the uplanders should be  compensated  for  making  changes  in  land  management  by  those  who  stand  to  benefit  downstream.  Compensation  mechanisms  suited  to  local  contexts  may  encourage  better  upland  management  and  this  will  generate  a  net  overall  benefit  to  the  communities  participating  (both  upland and lowland) in a mutual respect.    The working hypotheses of the intended research area are stated as follows:  1. Soil conservation practices are seldom adopted because they are not supported or  paid for at their real cost;  2. Willingness to pay for soil conservation practices is low because neither the cost of soil  erosion, nor the benefits of conservation practices are quantified on the basis of  reliable facts, and the benefits of adopting soil conservation practices are not seen in  the short term and may initially reduce agricultural output;  3. Water and soil erosion models can be used to quantify the positive effects of  conservation scenarios and evaluate the trade­offs so that policy makers can make  informed decisions.     An exploratory case study is clearly justified, given the lack of information in this area and the  need for community awareness.    A  stage by  stage  implementation  is  proposed where  limited number  of  sites  are  initially  assessed  with respect to recognized PES criteria (e.g. well defined service, value to beneficiaries, cause‐effect  understanding  of  the management  change  and  the  impact  on  the  environmental  service,  effective  institutional processes,  the poor benefit etc.).  Initial assessments consider PES options,  land users’  perceptions  regarding  benefits/costs  associated  with  different  land  use  options,  existing  local  mechanisms (cash‐based or based on other forms of exchanges) for encouraging the production of  environmental  services  and  presence  or  absence  of  enabling  conditions  etc.  This  initial  scoping  phase has been completed. Activities are now focused on pilot testing of suitable PES mechanisms,  qualitative evaluation of benefits and costs  from proposed changes  in  land uses,  identifying cause‐ effect  relationships,  empirical  field  observation  of  local  processes,  and  developing  enabling  conditions for successful implementation of PES schemes that may be developed.    Logical path is to identify site then services and then the appropriate scale of analysis. That said, the  preference is to start with examples that are reasonably local (small scale) as this is more tractable  initially  in  terms  of  social  complexity  and  the  likelihood  of  achieving  robust  cause‐effect  analysis.  Smaller  scales will  help  to understand  local  reciprocal  arrangements  better  due  to  clearer  ‘cause‐ effect’ relationship and the better possibilities for entering  into incentive mechanism as the effects  considered are  localized. Depending on  the extent of off‐site  impacts,  the scale could be expanded  later.    7 RESEARCH ASSUMPTIONS    The main  goal  of  the  Vietnam  case  study  is  to  undertake  a  preliminary  survey  of  the  actual  development  of  PES  concept  in  an  experimental  site  well  biophysical  studied  from  more  than  5  years,  and  well  introduced  within  the  uplanders’  strategies.  The  study  will  address  negative  externalities where the PES concept may provide a framework devoted to reduce them.    So the specific objectives should be as follows:  1. Identify a range of possible ES facing the development of economic and sustainable  agro­ecological intensification systems that will allow sustainable and environmental  utilization of uplands by smallholder farmers;   2. Define cost­effective alternative upland use practices able to provide such ES;  3. Define the value of the ES benefits to the lowland stakeholders;  4. Identify a range of possible ES that lowland stakeholders might be willing to trade  from upland landholders;  5. Identify a range of indicators from various scenarios forced by climate change, land  use change, based on modelling platform;  6. Assess the possibility of the selected ES to be implemented in the future (what the  upland farmers should be compensated for and how).    EXPERIMENTAL SITE IN NORTHERN VIETNAM: TIEN XUAN COMMUNE    The activities of MSEC in Vietnam are focused in Tien Xuan Commune located in Luong Son District,  Hoa  Binh  Province,  at  60  km  westward  of  Hanoi.  Seventeen  villages  constitute  the  Tien  Xuan  Commune, the principal of which are Dong Dau, Que Vai and Dong Cao, where research fieldwork has  been conducted from 2000 (see figure 1).    They have traditionally cultivated irrigated rice in the lowlands and maize or cassava in the uplands  with slopes between 40% to over 100%. From the 90’s, because the decreasing of soil fertility within  the uplands,  they  shift  their upland  crops  to  the  forestation associated  less  and  less with  cassava.  They have relied also on small animal husbandry as pigs and chickens as a means of living. Non‐farm  based  employment  has  also  increased  over  the  last  few  years,  especially  construction  work  (see  Clement et al., 2007).    The reasons for upland land use change are complex. Some farmers had the opportunity to sell their  land whilst others under a policy directive planted trees or have practiced improved fallows. There  has been a dramatic change in the extent of cassava production in the catchment with the total area  declining  from 40% of  the watershed area  in 2001  to  less  than 0.5% in 2004. With  this decline  in  area under crops, the opportunity has arisen to introduce a livestock component into the catchment.  Over the past 4 years, the impact of the fodder species (Bracharia ruziziensis) established under a no  till regime has been evaluated with respect to its ability to reduce erosion from these slopes. After 5  years of continuous erosion monitoring  (from 2001  to 2005),  the erosion rates measured  in Dong  Cao  experimental  watershed  have  confirmed  that  land  clearing  generates  the  highest  soil  loss  amount, and that forest and fodder cover effectively eliminated erosion one year after establishment.  The evaluation of fodder crops as an alternative to annual crops such as cassava has stimulated an  interest  in  improved livestock production based on the introduction of adapted pasture species by  farmers  from  the  surrounding  villages.  They  view  this  approach  as  having  a  quicker  and  larger  return on investment than the establishment of plantation forests.  8 Figure 1 : Topographic map of Tien Xuan and Dong Xuan Communes (with the hydrological limits of the  experimental PES watershed in green line)    But the counterpart of the rising of number of cows  is the  increase of gully erosion and the  water pollution by nitrogen, which could  impact directly the downward water bodies of the  other villages (see figure 2).    9 So the technical‐scientific challenge could be effectively closing to the erosion process control in the  uplands  and  the  nutrient  cycle  loop  in  a  manner  that  enhances  production,  incomes  and  environment, by ecological engineering addressing to 4 targets:  ‐ target 1: enhance the crop yield on the uplands;  ‐ target 2:control the water pollution and sedimentation into the lowlands;  ‐ target 3:enhance the well‐being by impact on the human health;  ‐ target 4:enhance the well‐being by economical development.    Cattle Rice field Annual crop (cassava, maize) Tien Xuan Commune Tropical fodder on slope upstream Temperate fodder on lowland Trees (Acacia,…) Dong Cao Village village Dien Xuan Commune downstream Waste waters Clean waters from reservoir Figure 2 : Which economic and sustainable agro‐ecological intensification systems that will allow  sustainable and environmental utilization of soils and waters in the uplands and lowlands by  smallholders?    Conclusion 2:   The  purpose  of  the  project  in  Vietnam  is  to  evaluate  whether  PES  could  be  an  efficient  alternative to the existing regulations, with respect to protecting downstream water quality  (issue 1 on  the Figure 2) and  from siltation  (issue 2 on Figure 2), assuring sufficient water  quantity for irrigation downstream (issue 3 on Figure 2) and promoting the enhancement of  cattle  husbandry  using  cowsheds  and  associated  fodder  crops  for  sustainable  agricultural  practices.      10 METHODOLOGY    In  order  to  tackle  the  6  specific  objectives  underlined  as  driven‐force  for  a  PES  framework,  the  research activities have been divided in 4 tasks distributed in two working groups:  WG1: Hydrological modeling for PES concept  WG2: Farmers’ demand and PES modalities    The four tasks are detailed in the following part:  ‐ Task 1: Identification of the ES, this task tackles the specific objective 1 to define a range of  possible  ES  facing  the  development  of  economic  and  sustainable  agro‐ecological  intensification systems that will allow sustainable and environmental utilization of uplands  by smallholder farmers;  ‐ Task 2: Effective cost for uplanders,  this  task  tackles  the specific objective 2  to calculate  the cost‐effective of the alternative upland use practices able to provide such ES;  ‐ Task 3: Erosion and runoff modeling within uplands, this task participates to the specific  objective  5  by  modelling  the  erosion  processes  on  sloping  lands  (development  of  PLER  model);  ‐ Task 4: Modeling and ES scenarios,  this  task  tackles  a  part  of  the  specific  objective  5  to  identify  a  range  of  indicators  from  various  scenarios  forced  by  climate  change,  land  use  change, based on modelling platform;    11 TASK 1: IDENTIFICATION OF THE ES     The  MSEC  project  in  Vietnam  has  demonstrated  that  sustainable  agricultural  practices  require  a  judicious combination of incentives and the promotion of technologies that have a significant impact  on smallholder incomes and on environmental sustainability.     Insights  from  our  local  study  has  proved  the  rapid  micro‐economic  impact  of  new  appropriate  agricultural  practices.  In  Tien  Xuan  Commune  (the  MSEC  experimental  site),  the  smallholders’  incomes have doubled in 3 years by the cattle husbandry rising (figure 3).      Figure 3:  Increase of  the  smallholders’  incomes  in Tien Xuan Commune  thank  to  the  cattle husbandry  rising (from Orange, 2008)    It has been also  shown  that  although  farmers  are  concerned about  the environment  and  resource  management,  this  in  itself  may  not  motivate  them  enough  to  adopt  sustainable  land  and  water  management practices. By  increasing  farmers’  incomes through  intensive  livestock production,  the  MSEC results have shown that indirect methods can help control erosion (Orange et al., 2008).    Conclusion 3:  An assessment of farmer based knowledge on rapid incentive mechanism able to mobilize the  farmers’ interest in Tien Xuan Commune has promoted biogas generation technique. This idea  has been put in action within on­going concrete action through two studied farms.    In addressing the issues of sustainable utilization of upland systems within an agronomic context, a  fundamental constraint to adoption of appropriate farming practices is the lack of quick and tangible  benefits  to  the  adopter.  To  address  this  impasse,  an  incentive  based  approach  to  the  sustainable  production of crops on sloping lands has been initiated at two farms in Dong Cao Village where the  use of biogas generation from human and animal waste resulted on sustainable crop production on  sloping lands. The two pilots have been established with a contract between households to produce  livestock feed on sloping lands in a sustainable manner and to utilize the solid waste generated from  store‐fed  animals  to  produce  compost  that  is  applied  to  upland  fields.  In  return  the  farmer  has  received  a  biogas  digester.  Preliminary  assessment  of  these  farmers  have  indicated  acceptance  of  this scheme, increased productivity of the farming unit and greater well being of individuals. Based  on current on‐going pilot farms, this study is dealing with a number of concrete actions addressing  12 households’ livelihoods and the management of uplands, as shown in the project framework (Figure  3). The key question is formulated as follows: Is it possible to use the interests of farmers in biogas  production to promote sustainable agricultural practices on sloping lands?         Figure 3: PES framework and environmental services to be tackled.    Although  the  implementation  of  interventions  that  result  in  an  enhancement  in  sustainable  utilization of natural resources and equity with long‐term benefits to potential adopters, they often  do not provide sufficient immediate economic returns to resource‐constraint farmers. Payments for  Environmental Services (PES) schemes present a new approach that  focuses directly on creating a  conditional  benefit  transfer  between  upland  providers  of  environmental  services  and  the  downstream beneficiaries of these services.    Conclusion 4:  The use of biogas technology purchasing by the  farmers should create a  favorable  framework  for PES scheme implementation. The final aim is to drive farmers to the access new agro‐ecological  technologies  for  agricultural  production  on  sloping  lands  by  involving  them  in markets  through  a  PES mechanism  to  guarantee  the  environmental  services  (ES)  benefits. The concept  is  to profit  from  the  farmers’  interests  in  biogas  production  in  order  to  promote  the  purchase  of  a  biodigester,  since  the  biodigester  introduction  into  the  farming  operation  farm will  force  them to change their farming system in order to respect the environmental services (decrease  erosion on the sloping lands, increase the water quality, promote the sustainable agricultural use of the  sloping  lands  based  on  the  farmers’  strategies,  enhance  the  human  health  and  the well  being)  by  increasing smallholders’ income.    13 The benefits  due  to  the  introduction of biogas generation at  the household  level based on human  and animal waste products are:    ‐ enhance the sustainable use of uplands;    ‐ increase smallholder’s incomes;     ‐ and significant health benefits.    The research assumptions to be tested or documented in this project are related to:  • The PES concept is an appropriate mechanism to mobilize this strategy;  • The  focal  point  is  biogas  generation  to  create  a  benefit  loop  for  the  sustainable  development of a new vision for agricultural and rural development (ARD);  • The  sustainability  is  confirmed  through  positive  impacts  that  include  environmental,  economic and well­being.    At last, the present study of pilot farms will assess the efficacy of household biogas production based  on human and animal waste,  in delivering positive environmental, economic and health benefits. A  key  outcome  of  this  study  will  be  quantification  of  these  benefits  and  the  development  of  an  incentive  framework  that  would  promote  the  adoption  and  out‐scaling  of  this  approach  in  the  uplands of Vietnam. The main output will be scientific knowledge on:   (1) the environmental services generated through the adoption of a biogas system that is  based on sustainable crop production on sloping lands;   (2) a socio­economic and health assessment of households utilizing biogas generation at  two pilot sites; and   (3) a PES framework feasibility based on incentives as a means to achieving sustainable  land  and  water  management  in  upland  agronomic  production  systems  in  Northern  Vietnam.    The  purpose  is  to  assess  the  development  of  effective  policies  and  incentives  that  are  based  on  quantifiable scientifically based monitoring indicators that will promote the sustainable utilization of  upland  cropping  systems.  This  approach  will  enhance  the  productivity  and  income  generation  opportunities for disadvantaged rural farmers in upper catchments in Northern Vietnam that result  in equitable distribution of benefits to upstream and downstream communities and the adoption of  sustainable  land  and  water  practices.  This  will  be  achieved  through  an  assessment  of  the  environmental  service,  economic  and  health  benefits  associated with  a  current  study  that  promotes biogas production from animal and human wastes, animal waste composting, and  sustainable crop production on sloping lands (Orange et al., 2007b).            14 TASK 2: EFFECTIVE COST FOR UPLANDERS   The objective is to evaluate the effective costs of biogas production linked to the composting process  for soil fertilisation in the uplands, in order to determine the capacity of these new farming practices  to be a sustainable environmental service. The evaluation is based on the study of one farm in Dong  Cao village, the M. Thào’s farm. The  first  step was  to  establish  a  diagram  representing  all  the  interactions  linked with  the  biogas  production  and  composting  process:  it  would  help  to  understand  which  parameters  are  at  stake  (Figure 4).    BIOGAS INSTALLATION CATTLESHED PIG WASTE ORGANIC MATTER BIOGAS COLLECTED (LIQUID) HUMAN WASTE COMPOST INSTALLATION WATER TOILETS ? ORGANIC MATTER COMPOST REALISATION COLLECTED (SOLID) EARTHWORMS SPREADING CARE BUFFALOS HARVEST WATER EARTHWORMS FODDER PRODUCTION CATTLESHED CULTIVATED LAND   Figure  4:  Diagram  on  biogas  production  and  composting  process  to  underline  the  cost‐effective  parameters.    The  second  step  was  to  fill  out  a  table  set  up  (Table  1)  based  on  the  previous  diagram  with  5  criterions (investment, time of work, space, expenses and income / yield / gain).                15 Table 1: Costs of biogas production associated to upland soil fertilization through organic matter  composted process.  INVESTMENT TIME OF WORK SPACE EXPENSES INCOME / YIELD / GAIN - 2 harvests / year, Downlands: 1800 Fertilisers: 7,2 - Family consumption (700 kg Rice farming For free, but limited - Work: to plant, to m² for rice farming. millions dongs / rice / harvest) during 8 months, area (600 m² / take care of, to year. - Rice purchase for the last 4 person). harvest. months. Fodder farming Any special area reserved for fodder farming. For free (at disposal Uplands (one side Food for free for cattle. Grazing practice by the Committee - of the mountain) at - People). disposal, without any limit. FODDER No fodder production for M. Thào. One cattleshed = 5 1 week. 30 m² designed for Any reparation CATTLESHED millions dongs. 6 buffalos, of since 2003. - which 12 m² for waste collection. One buffalo = 2 - 2 hours each Cf. cattleshed. - Veterinarian: - Work on the fields (30 days / millions dongs in morning: to go to 270 000 dongs / year, rice farming), 2002 grazing lands, year, - For sale: 7 millions dongs / 3 buffalos = 6 - 1 hour in the - Chemicals buffalo, millions dongs. evening: to bath, to against parasites: - Important social representation. BUFFALOS come back to 480 000 dongs / cattleshed and to year, treat parasites. - Straw of rice, maize and cassava from Thào’s own production. Cf. buffalos. One buffalo = 5 kg 12 m3 in the Cf. earthworms. - OM used mixed with fertilisers SOLID ORGANIC / day of solid waste cattleshed. (rice), MATTER (OM) 5 minutes / day. - For a “traditional” compost COLLECTED (anaerobia): 300 000 dongs / tonne. EARTHWORMS Cf. institut de l’élevage (NIAH), 04 83 87 Cf. compost box. / 1 kg dried earthworms = 23 8. 700 000 dongs. 3 compost boxes 10 days. 3 m² for the 3 Compost box (0,5 m3 for each one) boxes. / - = 1 million dongs. 30 minutes / week: Cf. compost box. - Better carbon assimilation in Maturing phase to check, to return / soil, eventually. - Less fertiliser added. SPREADING Unavailable data (not yet realised). HARVEST  Several points important to keep in mind:   16 COMPOST REALISATION CULTIVATED LAND   - The  first  point  is  that  even  if  compost  is  spread  with  the  good  rate,  fertilisers  should  be  added  to  obtain  efficient  yield  for  farming.  That means  that  fertilisers will  continue  to  be  used, but will be added in a smaller quantity,     - Concerning  earthworm  production:  it  appears  that  once  earthworms  are  included  in  compost  box,  there  is  no  need  to  put  again  other  earthworms  to  ensure  efficient  results.  Earthworms’  reproduction  is  fast  enough  to  take  up  all  the  compost  box  volume.  One  the  other hand, because they are inclined to stay at the bottom of the compost box, the most part  of them are in place for the following input of organic matter. That’s why the arrow between  “compost realisation” and “earthworms’ production” exists only in one way,    - Whenever “water” appears, any expense  is engaged  for  the moment, because water  is  free;  and despite any water management, the quantity is enough in the watershed, according to M.  Thào.    Some calculations have been done concerning the amount of compost needed:   - In one compost box of 0,5 m3, we put usually 60 % of solid waste (2/3), that means that if we  take 1 for the density of buffalos waste (assumption), we need 300 kg of waste to fill it,   - After 3 months of maturing phase, there is a reduction of 30 to 40 % of volume of compost,  that is to say around 200 kg of compost obtained,  - To be efficient, one tonne of compost should be spread per hectare, which means that with  one compost box, we can fertilise 200 m² of lands.  - M. Thào has 1800 m² at his disposal to cultivate rice, and about 1500 m² behind his house to  cultivate cassava and maize, that is to say 3200 m² in totality. Thus he will need 3200 kg of  compost per year to use it on his lands.   - 3200 kg equals to 16 compost boxes, and also equals to 4800 kg of buffalos waste per year,   - One buffalo produces 5 kg of waste per day; with 3 buffalos, we obtain 5 500 kg of waste per  year.   According  to  this calculation, M. Thào will have enough buffalos waste  to produce  compost in order to use it on all his fields  .  - Nevertheless, there is the question of space: how many compost boxes we need to produce  compost with all the buffalo’s waste quantity?  One box = 300 kg of waste. This quantity is obtained after 20 days with 3 buffalos. With the 3  months maturing period, that means that compost is ready after (20 + 3 x 30) 110 days.  That also means that 6 boxes are required to produce compost continuously (after 120 days,  6 boxes are filled and the first one is ready to be used).    - The best  solution would be  to produce compost with  the  technique of  [tas en andain]. The  system consists of the addition, day after day, of buffalos waste in a pile in triangular form. By  this way, all the buffalos waste production could be used, the required place is less important  than compost boxes, and there is no need to wait for 300 kg of waste to begin a new compost  box.    Concerning the biogas, here are several costs relative to the method:   - Toilets: 1 million dongs for building (investment),   - Human waste: one person produced between 180 and 340 g of solid waste / day    for M.  Thào’s family (4 adults and 2 children), that represents around 1,5 kg / day,   17 - Cattleshed  for pigs:   5 millions dongs  for building  (investment),  1 week of work, 40 m² of  which 2 m3 for waste collection,   - Pigs:  One  pig  of  10  kg  cost  300 000  dongs  (in  2005)    for  3  adult  pigs,  900 000 dongs  (investment),  - Food and care for pigs: except maize and cassava produced by M. Thào’s farming, extra‐food  is  bought  at  the  rate  of  1,8 million  dongs  /  year  (expenses).  For  care,  this  is  the  same  principle than for buffalos, that is to say 270 000 dongs / year for adults, and 30 000 dongs /  young pig / year (expenses),  - Pigs for sale: one pig of 10 kg is sold 500 000 dongs (income),  - Pig waste: one pig produced around 2 kg of solid waste / day,  - Biogas installation: 14 millions dongs for building (investment), 2 weeks of work, the biggest  container is around 12 m3 (under the ground).    This study is always on‐going and we cannot conclude on the effective cost right now.      18 TASK 3: EROSION AND RUNOFF MODELING WITHIN UPLANDS EXPERIMENTAL DONG CAO WATERSHED USED TO CALIBRATE PLER  The Dong Cao watershed used  for  the PLER calibration  is  completely based on  sloping  lands with  surface area of around 50 ha (Figure 5).    Dong Cao watershed used for PLER calibration   Figure 5 : Picture of the experimental Dong Cao watershed used to calibrate PLER      Figure 6 : Automatic weather  station Cimel                19 Meteorological data and flow discharge measurements  Thanks to an automatic weather station Cimel (figure 6), rainfall amount and intensity was recorded  every six minutes. This automatic station also records temperature, humidity, wind speed and solar  radiation intensity. Moreover, the watershed is equipped with five manual rain gauges taken down  every morning.    The watershed is equipped with 5 runoff measurement stations named weir (figures 7 and 8). Each  weir  is  set  with  an  automatic  water  level  recorder  (asThalimedes­OTT).  The  water  level  (WL)  is  recorded each 3 mn when the WL is moving up or down. The relation water level/ flow have been  determinate  by  gauging  thanks  to  a  micro‐propeller;  flow  speed  is  measured  in  several  section  points of Parshall weir downstream. Then the Biber software  is used to calculate an average speed  and the flow.  Figure 7: Topographic map of Dong Cao watershed and hydrological equipments on the field  (NMW: New Main Weir, Wi: Weir i)      Figure 8: Weir 3 (W3)                20 PLER ALGORITHM UNTIL 2007 (PLER V1) (FROM DO DUY PHAI ET AL., 2006)  Inputs and outputs  Inputs (figure 9) were:  • digital elevation model in raster format (in Vietnam : 2 m grid size);  • map with location of the measurement devices ("weirs");  •  maps  with  soils  and  land‐use  units  in  the  same  raster  format.  Each  unit  is  linked  with  10  parameters (in green on figure 9)   • time series with real or predicted rainfall and PET;  • 2 calibration values taking into account the anteriority of rainfalls (initial infiltration capacity ratio  and  initial water  content  of  the  soil),  and  1  calibration  value  taking  into  account  the water  paths  (ratio between effective velocity and  the  computed velocity, depends on mean hydraulic  radius of  the water pathways and can be set according to the response time of the watershed).    Outputs (figure 9) were:  • maps with cumulated runoff, sediments detachment, flux and storage (generated every time step /  every hour / every day... according to the user's needs);  • time series with values of runoff, sediments detachment,  flux and storage for  the cells where the  measurement devices are located (for calibration / validation).      Figure 9: Input and output maps, tables, time series and calibration parameters in PLER model  21 Land‐use and soil maps were overlapped on the digital elevation model. Each cell of the watershed  had a specific value for:    • static values: topographic parameters (slope, slope length, downstream direction), land‐use  parameters  (RET/PET ratio  (%), Manning  roughness  coefficient,  cover  (%)),  soil  parameters  (soil density (kg/m3), soil depth (m), minimum and maximum infiltration capacity (mm/h), %  of  clay  (%),  mean  sediments  velocity  of  deposition  in  stable  water  (m/s),  mean  sediment  density (kg/m3) and pores efficiency ratio (efficient pores / total pores,%);    •  dynamic  values:  pores  available  for  water  storage  (mm),  infiltration  capacity  (mm/h,  fluctuating  between  a  minimum  and  a  maximum),  water  volume  stored  in  the  cell  (m3,  conceptual  reservoir), mean  efficient  velocity  of  surface  and  subsurface water  runoff  (m/s),  time needed for this runoff to reach the outlet of the watershed (s), runoff (m3/s), sediments  detachment (t/ha), transport capacity of the flow (kg/m3), sediments flux and cumulated flux  (kg/s and ton), sediments storage (ton).      Modelling concept used in PLER    a/ From rainfall to runoff    PLER model is a 2 dimensions’ model. The surface of the watershed is divided in square cells (5m x  5m). According to the figure 10, the depth of each cell is defined as the height of soil which depends  on the kind of soil identified.    A water stock (conceptual reservoir) is attributed to each cell of the grid. At each time step, the water  stock:  • increases by net rainfall (= rainfall – evapotranspiration), measured from available climatic  datasets;  • decreases by deep drainage and downstream runoff. Deep drainage is the amount of water  lost for the watershed at each time step. Downstream runoff is  the part of the runoff that will  reach the outlet during the considered time step.    Therefore  there  is no upstream runoff to  take  into account,  as  each  cell  is  assumed  to  stock  this  water in its proper conceptual reservoir. The downstream runoff increases with water stock 2/3, so  cells far from the outlet will need several time steps with rainfalls in order to reach a sufficient water  stock  to contribute  in  the overall  runoff, whereas cells near  from the outlet will  contribute  to  this  runoff much earlier.    The runoff modulus has been developed  taking  into account some  results of  runoff measurements  carried  on  the  field  scale  by  MSEC  partners,  with  some  assumptions.  For  slopes  over  20%,  infiltration  capacity  is  assumed  to  be  proportional  to  slope  and,  for  covers  over  30%,  also  proportional  to  vegetation  cover.  Infiltration  computation  at  each  time  step  is  distributed  according  to  land‐use  map  and  slope  map  and  saturation  has  a  feedback  on  the  values  of  infiltration capacity and evapotranspiration (water loss).        22 Water balance    Figure 3: Runoff generation for each cell of the grid at each time step  Equivalent soil column    Figure 4: Equivalent soil column and available water storage in the soil  23 At the level of each cell, the soil is divided into two parts: a part with efficient pores (potential water  storage = "gap") and a part with sediments and structural pores (figure 11):    At each time step, according to the amount of water already stocked in the cell, one part of efficient  pores are available for further water storage and the other part is saturated:  Available (t+1) = available (t) – net rainfall (t) + downstream runoff (t) + deep drainage (t) At  each  time  step,  the  soil’s  saturation  is  taken  into  account  and  two  kinds  of  “runoffs”  can  be  produced:  - If the soil is saturated, the “subsurface runoff” and the surface runoff are considered;  - Else, the only “subsurface runoff” occurs.    The  ratio  available  /  potential water  storage  determines  the  efficiency  of  deep  drainage  and  evapotranspiration:      This ratio can fluctuate between a minimum and a maximum according to soil characteristics (figure  12).    Figure 5: Variation of infiltration capacity between minimum and maximum under rainfall and sunshine  24 The water stock is calculated at each time step by the formula:  Water stock (t+1) = water stock (t) + net rainfall (t) – downstream runoff to outlet (t) – deep  drainage (t)  The height of the water conceptual layer is defined as:    According to Manning‐Strickler equation, the velocity of a flow is:        Hydraulic radius is defined as flow section divided by wet perimeter. The velocity of subsurface and  surface flows contributing to the runoff at the scale of a watershed cannot be physically computed  with  this  original  equation,  because  we  have  no  possibility  of  measuring  or  even  assessing  the  hydraulic radius of a flow divided between surface rills and soil pores (the path of the water in soil  pores and stratifications involves too many parameters and most of them are impossible to measure  at  the  catchment  scale).  Therefore  we  assume  to  use  a  simplified  empirically  calibrated  equation  based  on  Manning­Strickler's.  We  also  assume  to  use  the  manning  roughness  coefficient for subsurface flows, even if its value is linked to surface roughness of the soil.    The average velocity of waters flows in subsurface and surface paths is:      The  average  of  this  velocity  from  the  studied  cell  to  the  outlet  (Vm)  determines  the part of  the  water stock Qp flowing into the drainage network at each time step:    Qp Vm×∆t = Ls with:   ‐ Vm : average of the velocity from the studied cell to the outlet;  ‐ ∆t  : time step duration;  ‐  Ls  : Slope length to the outlet.      In streams, the manning coefficient  is  lower and the height of  the water conceptual  layer  is higher  (accumulation from upstream cells), resulting in a higher velocity.      25 b/ From runoff calculation to sediment detachment and transport: Guess equation    This modulus uses Guess model equation distinguishing two processes: (1) soil destructuration by  water  drops  impact  and  surface  and  subsurface  sheet  runoff  (interrills);  (2)  detachment  and  transport by concentrated runoff (rills).    Guess equation first calculates the transport capacity of each cell:  From the height of the flow and the transport capacity, soil destructuration and quantity the mass of  sediments able to flow is:  Destructurated  sediments  can  migrate  downstream  according  to  the  drainage  network  and  the  transport capacity of each cell. Runoff and slope are the more variable parameters in this equation.    The results can be displayed under ArcGis (figures 13 and 14).       Figure 6: Example of output runoff 3D map  Figure 7: Example of output erosion 3D map  displayed with ArcGis. The higher runoff pathways  displayed with ArcGis. The higher erosion  are displayed with red color. The streams clearly  pathways are displayed with red color. The  appear as higher runoff pathways.  streams clearly appear as higher runoff pathways.   26 2007: PLERV3, NEW INVOLVEMENTS (BY BUI TAN YEN)   The involvements applied in PLER resulted in a new version of the model: PLER v3.    More input parameters variable with time  According to figure 15, in order to give more precision in the computing of runoff and erosion at each  time step, the following parameters are now changing with time:  - Land use changes and vegetal growth are considered:   o Land use changes are in correspondence to the cuts of forests which give birth to  young fallow;  o For each kind of vegetal species, the specific vegetal growth is taken into account  according to the vegetal growth cycle;  - The Manning roughness coefficient is estimated (see paragraph III.b.ii) according to  the land use and its changes, and taking into account the vegetal growth;  - The vegetation cover rate is intrinsically linked to the land use and the vegetal growth.  It is defined according to them (see paragraph III.b.ii);  - PET (Potential EvapoTranspiration) and RET (Real EvapoTranspiration) are also  changing with the vegetal growth and the climatic conditions.    Digital Elevation Model PLER v3 Rivers map Weirs map For each RET in mm time step PET in mm (Shell Controler) + + Rainfall in mm Landuse map Timestep duration in s daily Runoff maps Erosion maps Sediment storage maps Sediment flux maps 6 min Runoff series Erosion series Sediment storage series Sediment flux series m3 ton.ha-1 ton ton   Figure 8: Inputs and Outputs of the PLERv3 model    27 Manning coef Vegetal cover rate Soil cohesion Sed. velocity Soil depth Soil density Sed. density Infiltration cap Changes in the model structure to take into account the new input  parameters changes with time.    The PCRaster language used to code the algorithm of PLER v3 does not permit changes in the input  parameters entered as maps. The changes of input parameters with time involved the use of a  shell controller represented by a VISUAL BASIC macro (figure 16).     At  each  time  step,  PLER  v3  is  running  in  its  PCRaster  environment.  The  shell  controller  gets  the  outputs and generates new inputs which are used at the next time step for the model running.    Input Generate new code PC R aster at each time step environmentPLER Model Shell controller (VB macro) Output   Figure 9: Scheme of the link between the PCRaster environment and the Shell controller (Visual  Basic macro)      2008: PLERV4 FOR FINAL CALIBRATION STAGE (BY J. BERTRAND)   Resume on the initial state (from B.T. Yen, dec. 2007)  • Model calibrated on a 1‐year timescale (timestep 1 day).  • Kc  (crop  coefficient  for  evapotranspiration)  and  Mn  (manning  runoff  coefficient)  are  now  dynamic and their value change according to the crop development stage over the year.  • Optimal calibration parameters :   DeepInfil = scalar(0.005) ; # Ratio deep infiltration capacity / infiltrability (%)  EffGap = scalar(0.1) ;     # % of gaps efficient for water transit  WpVel = scalar(0.002) ;    # Ratio real/computed velocity of the flow  RivMann = scalar(0.01) ;   # Manning coefficient for streams  Problems to be resolved :  28 • The model doesn’t reproduce the flood peaks (duration 3‐4 hours), which are flattened at the  1 day timestep (figure 17). Yet it’s precisely during these events that the erosion process is  the most active.   • The  streams map doesn’t match  the  topography  of  the DEM :  streams  are  thus  assigned  a  wrong manning coefficient.  • The model only computes the discharge at the main weir (MW).      8000 7000 6000 5000 Measured discharge, m3 Qf1 Qf2 4000 Qf3 Qf4 Qf5 Qf7 3000 2000 1000 0 1 10 19 28 37 46 55 64 73 82 91 100 109 118 127 136 145 154 163 172 181 190 199 208 217 226 235   Figure 17: Observed runoff (blue curve) and computed runoff (red curve) with PLERv3      Improvements brought to the model since april 2008  • The model  is now more  flexible:  the paths of  the  input  files are now relative  (portability of  the  model).  • More outputs available (speed maps, water stock in each cell, discharges).  • Computation of the discharge at each weir in a single output file.  • Timestep is now 6 minutes in order to take into account the intense events.  • The  stream  map  is  now  computed  automatically  on  the  basis  of  the  DEM  (one  input  map  removed).    Tools development  • Tutorial to modify pcraster maps under arcgis 9.2.  • Script to convert the pcraster .map format to .ascii arcgis format.      Current status of the calibration procedure   29 Calibration achieved  for  the 3  intense rain events of 2004: at  the beginning of  the  rainy  season  in  May (figure 18), in the middle of the rainy season in July (figure 19) and at the highest peakflow in  August.    0,2 0,15 0,1 Q (m3/s) Série2 0,05 0 14/5/04 12:00 15/5/04 12:00 16/5/04 12:00 17/5/04 12:00 18/5/04 12:00 19/5/04 12:00 -0,05   Figure 18: Observed runoff (blue curve) and computed runoff (red curve) with PLERv4 in May    0,4 0,35 0,3 0,25 0,2 Q (m3/s) Série2 0,15 0,1 0,05 0 20/07/2004 21/07/2004 22/07/2004 23/07/2004 24/07/2004 25/07/2004 26/07/2004 27/07/2004 00:00 00:00 00:00 00:00 00:00 00:00 00:00 00:00 -0,05   Figure 19: Observed runoff (blue curve) and computed runoff (red curve) with PLERv4 in July      30 The  soil  is  assumed  to  be  dry  at  the  beginning  of  each  computation.  Land  use  remains  the  same  during the year (2004).    Problems to be solved :  • Increase  the  model  reactivity  during  intense  events  (the  strongest  peaks  couldn’t  be  reproduced properly so far):  o Define  and  compute  a  criteria  allowing  to  distinguish  these  short  rainy  events  and  process them differently with a specific infiltration law;  • Gullies implementation (short‐circuits): census to be carried out with Romain (july 16th);  • Lower the model sensibility in order to flatten the noise of the computed curve;  • Validity of the DEM: the real drainage area for the MW is greater than the one defined by the  watershed  limits  used  until  then  (north‐eastern  part  of  the  watershed)  and  must  be  redefined.     Continuation of the calibration  • Compare the experimental and computed discharge curves at the intermediate weirs (W2, 3,  4);  • Check  the  influence  of  the  landuse  on  results  by  choosing  a  sub‐watershed  prone  to  high  landuse variations from one year to another.                    31 TASK 4: MODELING AND ES SCENARIOS NEW MODEL BUILDING AT THE SCALE OF THE ENTIRE WATERSHED: A LINK BETWEEN THE  SLOPING SUB­CATCHMENTS AND THE DOWNSTREAM RESERVOIR  Because the PES concept involves two different communities between the uplands and the lowlands,  two Communes are considered: Tien Xuan Commune and Dong Xuan Commune (see figures 1 and 2).  So  to model  some  scenarios  on  climate  change  and  land  use  change, we  have  to  consider  a  large  watershed where the outlet is a reservoir used for drinkable water by downstream villages and for  irrigation (figure 20). This large watershed includes the experimental Dong Cao watershed used for  the PLER calibration and 4 other cultivated sloping lands and a plain used for rice growing.   Catchments borders Water flows Streams N Sloping cultivated lands Reservoir Reservoir outlet Plain used for rice growing 1 km Figure 20: MapInfo map of the large watershed used for PES scenario modeling       MODELLING CONCEPT WITHIN THE ENTIRE WATERSHED   In order to answer to the PES concept at the entire watershed scale, the purpose is to apply the PLER  model within the 5 sloping sub‐catchments. Then the new platform has to model the  link between  the five PLERv3 output flows (Qi) and the flow reaching the reservoir (QR) (figure 21).     This new platform has to enable the computation of the water  flows and  the sediments  flows  reaching the downstream reservoir.  32   1 km   Figure 210: Modelling concept at the entire catchment scale    PLER RUNNING ON THE 5 SUB­CATCHMENTS   Foreword  Accounting  for  the  long  running  time  of  the  model  PLERv3  and  the  difficulties  to  calibrate  it,  the  calibration of the model has not been done yet. In the following paragraph, only the platform’s building  and the perspectives of its calibration are presented.     The Digital Elevation Model (DEM)’s building   The  first  step  to  run  PLER  is  to  build  a  Digital  Elevation  Model  (DEM)  on  the  sloping  lands,  corresponding  to  the  five  5  sub‐catchments  (figure  22).  Nota: PLER  can  be  applied  only  on  the  sloping lands.     The DEM has been built using Arcview and MapInfo. Whereas the DEM of the sub­catchment used for  PLER calibration  is made of 2m x 2m cells,  the entire catchment’s DEM cells are at a minimum 5 m  because of  the  running  time of  the model. To get more precision, particularly  to  localize  streams, a  lower size should be used but the running time of PLER would be really long.        N3 3                                                 Figure 22: DEM’s building with Arcview  Inputs parameters to run PLERv3 within the five sub­catchments  Sub­catchment 1(figure 21)  34 Within the sub‐catchment 1, the work on the field to collect data has begun in 2000 (Do Duy Phai et al.,  2007).  The  collected  data  are  used  to  build  the  input  maps  and  tables  needed  to  run  PLER.  The  following figures 23 and 24 are examples of input maps.    Figure 11: Soil map, streams and limits of the sub‐ Figure 12: Land use map of sub‐catchment 1 in July  catchment 1  2003    Sub­catchments 2, 3, 4 and 5 (figure 21)  Land use  The positions of unpaved roads, streams and of  the reservoir have been  identified thanks to maps  and work on the field. The  land use units (figure 25) have been estimated by taking and analysing  pictures of the different sub‐catchments. The land use estimated is the June 2007‘s one. WE assume  that:  a planted forest  is  a more  two  years  old  forest,  a  degraded secondary forest  is more  than  10  years old. As well, an old fallow is more than 2 years old.    Land use  cassava (annual crop)  bracharia (pluri-annual crop)  planted forest  young planted forest  old fallow  young fallow  degraded secondary forest  Figure 13: Land use units identifiable within the studied watershed  35 Sub­catchment 2    In the sub­catchment 2,  except  the  cassava  fields  identified  in  the pictures  (figure  26),  the other  land use in place has been considered as young planted forest.         Cassava                       Cassava                 Figure 14: Pictures of the sub‐catchment 2 and localization of cassava fields      36 Sub­catchment 3    In the sub­catchment 3, all the land use has been considered as young planted forest (figure 27).                 Young planted   forest     Figure 15: Pictures of the sub‐catchment 3 with young planted forest    Sub­catchment 4    In the sub­catchment 4, all the land use is a planted forest (figure 28).                Planted forest           Figure 16: Pictures of the sub‐catchment 4 with planted forest  37   Sub­catchment 5    In the sub­catchment 5, the land use has been considered as young planted forest (figure 29).                Young   planted   Figure 17: Pictures of the sub‐catchment 5 with young planted forest      Conclusion : Estimate of the June 2007 's land use    The estimated land use map within the entire catchment is displayed on the figure 30.    N Reservoir 1 km   Figure 30: Estimated land use map of the current watershed (June 2007)   38 Manning roughness coefficient  The values of the manning coefficient have been fixed according to the bibliography (Table 2).    Table 2: Estimate of the Manning roughness coefficient according to the vegetal growth  Manning roughness coefficient [-] Land use Cat April May June July Augus t September October November December unpaved road 0 0.010 0.010 0.010 0.015 0.020 0.015 0.010 0.010 0.010 cassava 1 0.010 0.015 0.030 0.035 0.035 0.050 0.050 0.020 0.020 bracharia 2 0.025 0.035 0.040 0.060 0.060 0.060 0.040 0.040 0.040 planted forest 3 0.035 0.035 0.050 0.070 0.080 0.080 0.070 0.060 0.050 young planted forest 4 0.020 0.030 0.040 0.050 0.050 0.050 0.040 0.030 0.030 old fallow 5 0.070 0.070 0.100 0.120 0.160 0.160 0.120 0.100 0.100 young fallow 6 0.035 0.035 0.050 0.070 0.070 0.070 0.060 0.050 0.050 degraded secondary forest 7 0.070 0.070 0.100 0.120 0.160 0.160 0.120 0.100 0.100 lake 8 None None None None None None None None None streams 9 0.030 0.030 0.040 0.050 0.050 0.050 0.040 0.030 0.030   Soil parameters and Vegetation cover rate   Without  any  soil  data  available  outside  of  the  sub‐catchment  1,  the soil parameters  in  the sub­ catchments 2, 3, 4, 5 have been fixed at the average value among the six types of soils identified  within the sub‐catchment 1 (table 3) and the same value for the vegetation cover rate (table 4).     Table 3: Soil parameters of the six soil types encountered in the sub‐catchment 1  Depositio l Soil Infiltratio Infiltration Soil type Soil Cohesion n rate Sediment Soi Depth n Min Max (m/s) density (kg/m3) Density (kg/m3) (m) (mm/h) (mm/h) Acrisol peu profond 0.45 0.63 1300 950 0.5 15 67 Acrisol en place 0.45 0.63 1300 950 1.5 15 67 Acrisol s/coll 0.45 0.63 1300 950 1.5 15 67 Cambisol 0.37 1.12 1300 950 0.8 7.5 30 Fluvisol 0.37 1.12 1300 1300 0.3 2.5 18 Leptosol 0.37 1.12 1300 950 0.2 3.7 56 average 0.41 0.875 1300 1008.3 0.8 9.8 50.8 Table 4: Estimate of the vegetation cover rate (%) according to the vegetal growth  Vegetation cover rate (%) Land USE Cat April May June July August September October November December cassava 1 5 15 60 70 70 70 70 5 5 bracharia 2 30 70 95 95 95 90 70 40 30 planted forest 3 70 85 85 85 85 85 70 70 70 young planted forest 4 30 50 65 65 70 70 70 65 50 old fallow 5 70 85 85 95 95 95 90 85 70 young fallow 6 30 40 70 70 90 90 80 50 50 degraded secondary forest 7 70 95 95 95 95 95 95 70 70 lake 8 0 0 0 0 0 0 0 0 0 39 streams 9 0 0 0 0 0 0 0 0 0 Evapotranspiration (ET; [mm/h])  The ET is computed with the Penman‐Monteith formula thanks to the climatic data collected with the  meteorological automatic CIMEL station. According to the computation time step used in PLERv3, it  has  been  estimated  every  6  minutes.  The  real  evapotranspiration  on  sloping  lands  reached  650  mm/yr  to 850 mm/yr  (table 5). And  the potential evapotranspiration ETP concerning directly  the  evaporation on free water surface, such as a water reservoir, reached 850 mm/yr to 1150 mm/yr.    Table 5: Potential evapotranspiration (ETP), real evapotranspiration (ETR) and total rainfall (RR) in mm  recorded by meteorological station in Dong Cao  Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Year RR 125 2006 ETP mm 42 25 39 69 92 109 95 87 97 83 65 41 845 2 ET 2006 R mm 2 25 28 32 92 109 95 87 94 49 65 1 679 183 2005 ETP mm 34 32 39 66 121 115 111 90 100 86 51 45 890 9 ET 2005 R mm 25 29 39 44 93 115 111 90 100 44 51 34 773 135 2004 ETP mm 30 42 48 57 97 131 106 110 105 106 71 66 969 1 ET 2004 R mm 26 33 48 57 97 107 106 110 99 11 60 20 771 111 158 2003 ETP mm 60 54 72 88 135 145 136 108 95 98 75 54 9 4 ET 2003 R mm 46 27 10 88 135 145 136 108 95 31 21 7 845 124 2002 ETP mm 52 38 52 90 109 114 129 114 121 92 51 38 999 4 ET 2002 R mm 27 16 22 40 109 114 117 101 108 92 51 38 833   FROM THE SUB­BASINS OUTLETS TO THE DOWNSTREAM RESERVOIR   This part of the modelling consists on building a link between the PLER output water flows from the  5 sub‐catchments and the water level in the downstream reservoir.     Assumptions  Concerning water flows  ‐ A1‐ No soil permeability under paddies;  ‐ A2‐ No infiltration in the stream bed during water transport by the main river;  ‐ A3‐ No infiltration inside the downstream reservoir.  40   Concerning transport of sediments  ‐ A4‐ No deposition of sediments in the stream bed during the transport by the main river and by the  streams;  ‐ A5‐  During  the  irrigation’s  periods:  same  sediments  load  (M1  in mg/L)  in  the  water  used  for  irrigation  from  the  streams  than  in  water  flowing  to  the  reservoir.The  irrigation  process  from  streams does not concentrate the water of the streams with sediments (figure 31);  Irrigation channel M1 M1 M1 Stream   Figure 31: Irrigation process from streams and conservation of the sediments load     Consequence: conserved proportionality between water flow and sediments flow.   ­ A6‐ All the sediments in the water used for irrigation are depositing in the paddies.  The rice is sown three times per year (table 6).     The first seedbed occurs in March. Until June the rice growth is assured both by rainfall and by an  active  irrigation  process  thanks  to  irrigation  channels  connected  to  streams.  The  second  seedbed occurs in July. Until October the rice growth is only assured by rainfall. There‘s no active  irrigation during this second growth of the year. The third seedbed occurs  in September. Like the  first growth, until February the third one is assured both by rainfall and by an active irrigation  process thanks to irrigation channels connected to streams.    Table 6: Active and passive irrigation during the year  Growth from  Growth from March  Growth from July to  Irrigation process  November to  to June  October  February  Only rainfall    X    Rainfall + active  irrigation from  X    X  streams      41 The rice paddies represent streams’ water storages during the active irrigation periods. The  entire  catchment’s  modelling  platform  takes  into  account  these  different  periods  with  a  precision of one day.   The new platform has to be used to measure the impacts of up­stream agricultural strategies  on the downstream reservoir water balance. The downstream water consumption is based on  the  paddies  cultural  seasons.  The  new  platform  has  to  be  run  during  the  three  cultural  seasons underlined in table 6.    Computing of the flows reaching the reservoir   Water flow  The QA, QB, QC and QD  are  the water  flows  from  the  sloping  sub‐basins 1,  2,  3,  4  and 5  and  are  computed by PLER (figure 32). The xA, xB, xC and xD terms correspond to the water flows used to  irrigate paddies. The QrA, QrB, QrC and QrD are  the water  flows remaining after  the  irrigation of  the catchment paddies. The xE term is the flow used by villages in downstream part (outside of the  studied watersheds) for irrigation and human water consumption.         N QC Q Reservoir R QB QD QA 1 km   Figure 18: Modelling concept of water flows to the downstream reservoir    NB: QA is the sum of the water flows from the two sub­basins 1 and 2.    42 According to the assumptions A1, A2 and A3, the net water flow reaching the downstream reservoir  can be computed with the following equation (figure 32):    D E   QR = ∑Qj −∑ xj   j=A j=A   If there’s NO active irrigation,   xj=0 except xE≠0 (drinkable water for downstream villages);    D Q  R = ∑Qj − xE j=A   If there’s an active irrigation,   xj≠0    j = A,...,E      D E QR = ∑Qj −∑ xj   j=A j=A   Sediment flow  For the sediment flow, we follow the same method than for the runoff.    The QmA, QmB, QmC and QmD are the sediments flows from the sloping sub‐basins 1, 2, 3, 4 and 5  and are computed by PLER. The xmA, xmB, xmC and xmD terms correspond to the sediments flows  going  to  the catchment paddies. The xmE  term  is  the sediment  flow going  to downstream villages  (water  used  for  irrigation  and  drinkable  water).  The  QmrA,  QmrB,  QmrC  and  QmrD  are  the  sediments flows remaining after irrigation of the catchment’s paddies.    NB: QmA is the sum of the sediments flows from the two sub‐catchments 1 and 2.      43 STRUCTURE OF THE NEW PLATFORM UNDER MICROSOFT EXCEL  Water flows platform      44     45 Sediments flows platform  46       47 CONCLUSION ON THE NEW MODELING PLATFORM    A  fieldwork  is  yet  necessary  to determine  the  limits of 4 paddies groups  irrigated by  runoff  from  the  different  sub­catchments.  That  means  to  identify  with  precision  the  position  of  the  irrigation channels and their connections.    Then, the second problem is also to determine the flows used for irrigation (xj with j=A,...,E). It will  be necessary to speak with  farmers and to measure these  flows to give precision to the computed  value of the sediments flow reaching the reservoir.    This modelling  approach  has  been  begun  and will  be  enhanced  and  tested  in  the  next  few  months. This modelling platform, which computes  flows every day, will be used also to model  the nitrogen concentrations in the water filling the reservoir.    - 48 / 68- CONCLUSION The hydrological PLER model has been built by  the MSEC  team since 2000  in order  to predict  the  erosion and the runoff at the outlet of sloping catchments with a surface area smaller than 1 km2. The  modelling way  uses  the  formula of Manning­Strickler  (computing  of  the  runoff)  and  the Guess  equation (computing of the erosion). This modelling way is original because many other models use  formulas  derived  from  the  Darcy  formula  to  compute  the  flows  of  soil  sub‐surface.  The  PLERv4  model calibration is going‐on and the validation is planned in the next few months.    In order to determine the possibility to apply a PES concept in the area, a second modelling approach  has been explored in this project. A modelling platform under excel has been developed to predict  the water flow and the sediments flow reaching the downstream reservoir of Tien Xuan Commune  using the outputs of PLER model (runoff and sediments  flow) at the outlets of the five sloping  sub­catchments and creates the link to the reservoir through the paddy fields. Moreover,  the  active irrigation process is not simple. To predict with precision the flows reaching the downstream  reservoir, more information concerning the irrigation channels network are needed. A second  aim of this platform would be also to model the daily nitrogen concentrations in water reaching  the reservoir which will increase because of cattle development up‐stream. In order to calibrate and  validate  this  first  entire  catchment  scale  modelling  platform,  PLERv4 has  to be  calibrated  and  applied within the five sub‐catchments (in progress, planed in the few next months). The Nitrogen  flow is planed to be modelled by adding a function in PLERv4.    By the other hand, a positive success of a PES framework in Northern Vietnam is assumed thank to  the  shift  to  a  market  economy  that  has  resulted  in  greater  financing  needs  for  farmers  and  all  categories of households use loans (Le Roy and Robert, 1999). Although the region in Vietnam with  the highest contribution to national poverty is currently still  the Northern mountains, especially  in  the West side (from 23% in 1993 to 22% in 2002 of the total Vietnamese population) (Rama, 2004),  the  MSEC  research  on  cattle  husbandry  rising  has  proved  the  great  potential  dissemination  of  technologies driven by the economic interest of individual private enterprises.     So  the project will  specifically assess  the efficacy of household biogas production based on human  and animal waste, in delivering positive environmental, economic and health benefits for the upland  communities. A key outcome of the on­going study will be quantification of these benefits and  the development of an incentive framework that would promote the adoption and out­scaling  of this PES orientated approach in the mountainous regions of Vietnam.    The purpose of this project is to build a new replicable agro­ecological technology. The scaling‐up  potential will be to promote biogas and vermicompost production. The proposed idea offers a novel  approach in that it introduces a holistic approach to addressing human health issues and closing the  nutrient cycle loop, with socio‐economic, sustainable land and water management implications. The  technical‐scientific challenge in the project is effectively closing the nutrient cycle loop in a manner  that  enhances  production,  incomes  and  the  environment.  The  originality  of  this  approach  is  the  addressing of human health  issues  through  the  sustainable utilization of  human and animal waste  streams with consequent economic benefits to resource poor farmers.     In  conclusion,  it  appears  that  the  concept  of  payment  for  environmental  services  (PES)  could  be  achieved  through  the  production  of  biogas  and  compost  using  simple  cost  effective  household  digester systems and will promote a new vision for agricultural and rural development.    - 49 / 68- LITERATURE LIST   Affholder F., Jourdain D., Morize M., Dang Dinh Quang, Ricome A., in press. Eco-intensification sur les versants montagneux au Vietnam. Contraintes et opportunités d’adoption des SCV par les agriculteurs. Cahiers d'Agriculture. Asquith N.M., Vargas M.T.R., Smith J., 2002. Can forest-protection carbon projects improve rural livelihoods? Analysis of the Noel Kempff Mercado Climate Action Project, Bolivia. Mitigation and Adaptation Strategies for Global Change, 7: 323–337. Baier W., Dumanski J., 1991. Agroclimatic guidelines and criteria for evaluation of sustainable land management. Proceedings of Int. Workshop: Evaluation for sustainable land management, 15-21 September 1991, Chiang Rai, Thailand: 47-83. Dang Kim Son, eds, 2008. Rural development strategy : outline and workplan. Ad-Hoc Group, IPSARD (MARD), experts meeting on 16 April 2008, IPSARD, Hanoi, 4 p. Castella J.C., Dang Dinh Quang, eds, 2002. Doi Moi in the Mountains. Land use changes and farmers’ livelihood strategies in Bac an Province, Vietnam. The Agricultural Publishing House, Hanoi, 283 p. Clement F., Amezaga J.M., Orange D., Tran Duc Toan, 2007. The impact of government policies on land use in Northern Vietnam: an institutional approach for understanding farmer decisions. IWMI Research Report, 112, Sri Lanka, International Water Management Inst., 21 p. Clement F., Amezaga J.M., Orange D., Calder I., Large A.R.G., Tran Duc Toan, 2008. Linking reforestation with forest policies: a multi-scale and interdisciplinary methodology applied to Vietnam. 12th Biennial Conference of the IASC, Chetenham, UK, 14-18 July 2008, 28 p. Cramb R.A., 2005. Farmers’ strategies for managing acid upland soils in Southeast Asia: an evolutionary perspective. Agriculture, Ecosystems and Environment, 106: 69-87. IWMI, 2008. Planning for a watertight future: IWMI at the Davos discussions. Water Figures (Turning research into development), IWMI Ed., Colombo (Sri Lanka), 2008(1): 3. Geay F., Dao The Tuan, eds, 2000. Appui à l’organisation de la production agricole dans le Nord du Vietnam. Actes du séminaire de lancement du projet FAC « Programme Fleuve Rouge », Hanoi, 11-13 octobre 1999, VASI-GRET, Maison d’Edition de l’Agriculture, 387 p. Le Roy P., Robert Mathias, 1999. The micro-economic impact of rural credit in northern Vietnam. Collection Etudes et Travaux, Ed. GRET, Paris, 79 p. Li T.M., 2002. Engaging simplifications: community-based resource management, market processes and state agendas in uplands Southeast Asia. World Development, 30 (2): 265-283. Minang P.A., McCall M.K., Bressers H.T.A., 2007. Community capacity for implementing clean development: mechanism projects within community forests in Cameroon. Environ Management, 39:615– 630. Noble A., Bossio D.A., Penning de Vries F.W.T., Pretty J., Thiyagarajan T.M., 2006. Intensifying agricultural sustainability: an analysis of impacts and drivers in the development of “bright spots”. Comprehensive Assessment of Water management in agriculture, Research Report 13, IWMI, 42 p. Orange D., Podwojewski P., Tran Duc Toan, Pham Van Rinh, Do Duy Phai, Nguyen Duy Phuong, 2007a. Impact of land-use on bed load transported by rivers in the Dong Cao watershed, north Vietnam. Water Figures Asia, issue 1-2007, IWMI, Colombo : 4-5. Orange D., Bardouin L., Nguyen Duy Phuong, Loiseau J-B., Clement F., Jouquet P., 2007b. Le concept de PES pour une gestion durable des eaux et des sols : application au développement de l’élevage et au contrôle environnemental dans le Nord Vietnam. In Orange D., Roose E., Vermande P., Gastellu J.-P., Pham Quang Ha (Eds) : Gestion intégrée des eaux et des sols : ressources, aménagements et risques en milieux ruraux et urbains, Editions AUF et IRD, Hanoi, Actes des Premières Journées Scientifiques Inter-Réseaux de l’AUF, Hanoi, 6-9 novembre 2007, cd-rom : 8p. Orange D., Tran Duc Toan, Salgado P., Nguyen Duy Phuong, Nguyen Van Thiet, Clement F., Le Hoa Binh, 2008. Different interests, common concerns and shared benefits. LEISA Journal, Netherlands. Orange D., 2008. Appropriation par les agriculteurs du Nord-Vietnam d’une meilleure gestion de la fertilité des pentes via des collaborations Nord-Sud et CIRA. MSEC project from IWMI-IRD joint-venture, Contribution pour la Commission de la Recherche Agricole Internationale, Février 2008, 5 p. - 50 / 68- Pagiola S., Platais G., 2002. Payments for environmental services. Environment Strategy Notes, volume 3. Pagiola S., Arcenas A., Platais G., 2005. Can payments for environmental services help reduce poverty? An exploration of the issues and the evidence to date. World Development, 33(2): 237-253. Rama M. (Ed.), 2004. Poverty, Vietnam development report 2004. ADB-AusAID-DFID-GTZ-JICA- SaveChildren-UNDP-WB, Vietnam Development Information Center, Hanoi, 144 p. Tomich T.P., Thomas D.E., van Noordwijk M., 2004. Environmental services and land use change in Southeast Asia: from recognition to regulation or reward? Agric Ecosystems Envir, 104: 229–244. Tran Duc Toan, Orange D., Podwojewski P., Do Duy Phai, Thai Phien, Maugin J., Pham Van Rinh, 2003. Soil Erosion and Land Use in the Dong Cao Catchment in Northern Vietnam. In Maglinao A.R., Valentin C., Penning de Vries F. (eds.): From soil research to land and water management: Harmonizing people and nature, Proceedings of the IWMI-ADB Project Annual Meeting and 7th Management of Soil Erosion Consortium (MSEC) Assembly, held at Vientiane, 2-7 December 2002: IWMI-Southeast Asia Regional Office, Bangkok: 165-180. Valentin C., Agus F., Alamban R., Boosaner A., Bricquet J.P., Chaplot V., de Guzman T., de Rouw A., Janeau J.L., Orange D., Phai Do Duy, Podwojewski P., Ribolzi O., Silvera N., Subagyono K., Thiébaux J.P., Toan Tran Duc, 2008. Runoff and sediment losses from 27 upland catchments in Southeast Asia: Impact of rapid land use changes and conservation practices. Agriculture, Ecosystems and Environment, in press. Valentin C., Hoogmoed W., Andriesse W., 1991. Maintenance and enhancement of low-fertility soils. Proceedings of Int. Workshop: Evaluation for sustainable land management, 15-21 September 1991, Chiang Rai, Thailand: 157-187. Vu Tan Phuong, 2008. Small scale AR CDM in Viet Nam: A case project in Cao Phong district, Hoa Binh Province of Viet Nam. Presentation at the Second Regional Roundtable “Sustainable Financing: Current Status and Emerging Opportunities” of the Asia Regional Biodiversity Conservation Program, 23-25 January 2008, Vientiane. WWF, 2007. Paying for ecological services: the use of economic tools to deliver conservation goals in Vietnam - A portfolio of projects. Hanoi, WWF.       - 51 / 68- Annex 1 List of internship reports Loiseau J.B., (2007). Establishment of a model predicting erosion and runoff at the outlet of a rural catchment in the North Vietnam hills. National High School for Geology, IWMI-IRD, SFRI, Hanoi. Bardouin L., (2007). Socio-economic impact of biogas and compost units on the farmers’ livelihood of Northern Vietnam (Hoa Binh Province. ISTOM school, IWMI-IRD, SFRI, Hanoi. Rodriguez C., (2008, on-going). Ecological impact of animal husbandry within mountainous area of Northern Vietnam: a study case. University of Clermont-Ferrand, Ecology and GIS, IWMI-IRD, SFRI, Hanoi. Grandidier E., (2008, on-going). Calibration and validation of an environmental model for PES use within a small agricultural watersheds in Southeast Asia: the environmental point of view. ENSIL-Limoges, IWMI-IRD, SFRI, Hanoi. Bertrand J., (2008, on-going). Calibration and validation of an environmental model for PES use within a small agricultural watersheds in Southeast Asia: the hydrological point of view. ENSEEIHT-Toulouse, IWMI-IRD, SFRI, Hanoi. - 52 / 68- Annex 2 WORKPLAN 2007-2008 / PES-MSEC Vietnam (revised Dec 2007) The activities are divided on two workgroups: the WG1 to apply the hydrological modeling, the WG2 to define the modalities for PES actions. Three indicators have been chosen to measure the impact of each scenario: the surface water level, the underground water level and the nitrogen load in the surface waters. WG1: Hydrological modeling for PES concept - Task 1.1: Control of DEM - Task 1.2: Introduction of land-use change each 2 weeks (action on parameters such as vegetation cover, ET, manning coefficient and nutrient losses) - Task 1.3: Validation of PLER and LISEM, applying MIKE SHE - Task 1.4: Addition of gullies to the hydrological pathway - Task 1.5: Running of three scenarios WG2: Farmers’ demand and PES modalities - Task 2.1: Inquiries about farmers’ demand from upstream and downstream - Task 2.2: Environmental assessment of cattle raising (nitrogen outputs and water balance, water quality assessment within the downstream reservoir) - Task 2.3: Socio-economic and environmental survey of two new biotechnologies to control the nitrogen output (biogas system for waste waters and earthworms compost for solid wastes), socio-economic and environmental modeling - Task 2.4: Cost evaluation within the three studied scenarios - Task 2.5: Conclusion on PES action in Luong Son District Team List in 2007: DO: Didier Orange (IRD-IWMI, Vietnam), Hydrology, Geochemistry, Modelling, Team management NP: Nguyen Duy Phuong (junior researcher, ISF, Vietnam), Socio-economy and Environment, Field management BY: Bui Tan Yen (junior researcher, ISF, Vietnam), Modelling PR: Pham Dinh Rinh (agricultural qualified technician, ISF, Vietnam), Hydrology and Socio-economy THDT: Thierry Henry des Tureaux (qualified technician ,IRD-IWMI, Vietnam), Hydrology. Geochemistry JBa: Jean-Baptiste Loiseau (French student, Vietnam), Modelling LB: Lucie Bardouin (French student, Vietnam), Agronmist NM: Ngo Duc Minh (Vietnamese student, Vietnam), Agronomist OP: Olivier Planchon (IRD-IWMI, Thailand), Modelling support AO: Amelia O’Brien (IRD-IWMI, Thailand), Support for socio-economy Team List in 2008: DO: Didier Orange (IRD-IWMI, Vietnam), Hydrology, Geochemistry, Modelling, Team management NP: Nguyen Duy Phuong (junior researcher, ISF, Vietnam), Socio-economy and Environment, Field management BY: Bui Tan Yen (junior researcher, ISF, Vietnam), Modelling PR: Pham Dinh Rinh (agricultural qualified technician, ISF, Vietnam), Hydrology and Socio-economy THDT: Thierry Henry des Tureaux (qualified technician ,IRD-IWMI, Vietnam), Hydrology. Geochemistry CR: Celia Rodriguez (French student, Vietnam), Environmental assessment EG: Estelle Grandidier (French student, Vietnam), Socio-economic and Environmental Modelling JB: Jeremy Bertrand (French student, Vietnam), Hydrological Modelling - 53 / 68- Annex 3 Current code of the PLER model (jul. 08) #|=================================================================| # PLER : Predict and Locate soil Erosion and Runoff at catchment-scale #|=================================================================| #! --lddin --matrixtable binding DataDir = ".\Inputs\Data"; #==============INPUT============================================= #DEFINE SET OF INPUT DATA TblAspect = ".\Inputs\Data\degrees.tbl"; TblSoilPr = ".\Inputs\Data\SoilProperties.tbl"; #Data table describes field code, type and current development duration of LU in each field #TblFields = ".\Inputs\Data\Fields.tbl"; TblField = ".\Inputs\Data\Field_LU_CurD.tbl "; #Data table describes charateristics of each landuse type TblLandUse = ".\Inputs\Data\LandUse.tbl "; #Rainfall and ETo at each time step TssRain = ".\Inputs\Data\Rain_ET.tss"; #DEFINE SET OF INPUT MAP MapBound = ".\Inputs\Maps\Bound.map"; #Boundary map MapFields = ".\Inputs\Maps\Fields.map"; #Field map MapSoil = ".\Inputs\Maps\SOIL.MAP"; #Soil map MapDEM = ".\Inputs\Maps\DEM.MAP"; #DEM map #MapStreams = ".\Inputs\Maps\STREAMS.MAP"; #Stream map MapOutlet = ".\Inputs\Maps\Outlet.map"; #Outlet map MapClone = ".\Inputs\Maps\Clone.map"; #Clone map #==============OutPUT============================================ #DEFINE SET OF OUTPUT MAP Map_LUT = ".\outputs\Maps\LandUse.map";#Land use map MapManning = ".\outputs\LU_manning.map"; #Manning map MapCover = ".\outputs\LU_cover.map"; #Vegetation cover rate map #Map_Qmmt = ".\Outputs\Qmmt.map"; #Remaining surface water map Map_Qf = ".\Outputs\Qf.map"; #Discharge map Map_e = ".\Outputs\e.map"; #Erosion map Map_eft = ".\Outputs\eft.map"; #Erosion flux map Map_Rivdist = ".\Outputs\Rivdist.map"; #Distance to streams map lddmap = ".\Outputs\Lddmap.map"; #Local drain direction map Cellsize = ".\Outputs\Cellsize.map"; #Effective cellsize S_map = ".\Outputs\Slope.map"; #Slope map subwsheds = ".\Outputs\Subcatchments.map";#Subcatchments maps (1 for each weir) #DEFINE SET OF OUTPUT data Tss_eStore = ".\Outputs\eStore.tss"; #Cumulated erosion Out_tssQt = ".\Outputs\OutQt.tss" ; #Cumulated water flow Out_tssQf = ".\Outputs\OutQf.tss" ; #Cumulated discharge Active_Area = ".\Outputs\Activearea.tss"; #Active area (sum of cell Size) Manning = ".\Outputs\Manning.tss"; - 54 / 68- #accN_loss = 0 ; #accP_loss = 0 ; #accK_loss = 0 ; areamap MapClone; timer 1 1211 1; d10 = 0+10..endtime ; #set the delay between output maps report (every 10 time steps) d5 = 0+5..endtime ; #set the delay between output maps report (every 5 time steps) d100 = 0+100..endtime; # ================================================================ # Static section # ================================================================ initial Mask = MapClone; d = 360. ; #timestep duration in seconds D = d / 86400. ; # !!, current timestep duration (days), #TUnit = 1; #Define initial values for dynamic variables================================= Qmmt = 0; # Remaining surface water on the watershed (mm) Qt = 0; Qf = 0; #Discharge at initial of model (m3/timestep) e = 0; eft = 0; eStore = 0; #=========================Calibration parameters====================== CrustInt = scalar(5) ; # Minimum rainfall intensity (CrustInt, mm/timestep) # !! 50 mm/h CrustSlope = scalar(0.3) ; # maximum slope (%) enabling crust formation DeepInfil = scalar(0.15) ; # Ratio deep infiltration capacity / infiltrability (%) (!!0,005) EffGap = scalar(0.08) ; # % of gaps efficient for water transit (!!!0.1) WpVel = scalar(0.005) ; # Ratio real/computed velocity of the flow / taking into account the water paths (!!!0,002) RivMann = scalar(0.02) ; # Manning coefficient for streams (0,01) #=============================================================== Wat = scalar(1000) ; # Water Density (kg/m3) #Reads data from table and generate maps==================================== # Sediment velocity sedvelmap = lookupscalar(TblSoilPr,1, MapSoil) ; # Sediment density (kg/m3) seddenmap = lookupscalar(TblSoilPr,2, MapSoil) ; # Soil depth (m) depthmap = lookupscalar(TblSoilPr,3, MapSoil) ; # Soil density (kg/m3) soildenmap = lookupscalar(TblSoilPr,4,MapSoil) ; # erodibility factor by Rose (cohesive) from 0-1 cohesivemap = lookupscalar(TblSoilPr,5, MapSoil) ; - 55 / 68- #Creates a local drainage direction map lddmap = lddcreate(MapDEM,1e31,1e31,1e31,1e31) ; lddmap = lddrepair(lddmap) ; # A boolean map with 1 for stream body and 0 for other cells in the map tolerance = 10000; #sets the minimum Qbatch value of cells to be included in the streams map batch = 10; #amount of sample rainfall (mm) to build streams map Qbatch = accuflux (lddmap, batch); StreamMap = boolean(if(Qbatch > tolerance,1,0)); #report Check_map = Qbatch ; # Check_map allows adjustment of tolerance StreamMap = boolean(if(MapDEM > 0,cover(StreamMap,0))) ; #report .\Outputs\StreamMap.map = StreamMap; # A map with 1 for weir's position and 0 for other cells in the map # OutletMap = boolean(if(MapDEM > 0,cover(MapOutlet,0))) ; OutletMap = boolean(MapOutlet) ; MapOut_ord = ordinal(MapOutlet) ; #A map showing the catchment area corresponding to each weir subwsheds = subcatchment(lddmap,MapOut_ord); #report .\Outputs\subwsheds.map = subwsheds; # soil water storage availability (mm) gapmax = depthmap * 1000 * (1 - soildenmap / seddenmap) * EffGap ; # Calculate distance from each cell to the streams (m) Rivdist = if(MapDEM > 0 , cover(ldddist(lddmap,StreamMap,1) , 1)) ; #Generate Kc, Manning and vegetation cover rate maps======================== Map_LUT =lookupscalar(TblField,1,MapFields); #Land use type (or crop type) Dev1 =lookupscalar(TblLandUse,1,Map_LUT); #Lenght of Dev1 (day) :Initial Dev2 =lookupscalar(TblLandUse,2,Map_LUT); #Lenght of Dev2 (day) :Development Dev3 =lookupscalar(TblLandUse,3,Map_LUT); #Lenght of Dev3 (day) :Middle Dev4 =lookupscalar(TblLandUse,4,Map_LUT); #Lenght of Dev4 (day) :Late CropDur =lookupscalar(TblLandUse,5,Map_LUT); #Growth duration = n days when start model CurDu =lookupscalar(TblField,2,MapFields); #Current duration of crop (day) on the field #Crop coefficient at each Development stage (Kc2 = from Kc1 to Kc3) Kc1 = lookupscalar(TblLandUse,6,Map_LUT); Kc3 = lookupscalar(TblLandUse,8,Map_LUT); Kc4 = lookupscalar(TblLandUse,9,Map_LUT); #Vegetation cover rate at each Development stage (Cov2 = from Cov1 to Cov3) Cov1 = lookupscalar(TblLandUse,10,Map_LUT); Cov3 = lookupscalar(TblLandUse,12,Map_LUT); Cov4 = lookupscalar(TblLandUse,13,Map_LUT); #Maning coefficient at each Development stage (Mn2 = from Mn1 to Mn3) Mn1 = lookupscalar(TblLandUse,14,Map_LUT); Mn3 = lookupscalar(TblLandUse,16,Map_LUT); Mn4 = lookupscalar(TblLandUse,17,Map_LUT); # Slope and Slope-length =============================================== # Slope of plot in % S = slope(MapDEM) ; # Aspect of slope 1 = crossed; 0 = Straight Aspect = aspect(MapDEM) ; # Angle, defines way of computing size of cells - 56 / 68- Angle = lookupboolean(TblAspect, Aspect) ; # Computes celllength (m) Length = celllength() ; # Size of sloping plane (m2) Size = sqr(Length)*(sqrt(1+sqr(S))) ; # cell slope-length (m) L = if(Angle, sqrt(2*sqr(Length)*(1+sqr(S))), Size /Length) ; # Slope length from each cell to the outlet (m) SL = if(MapDEM > 0 , cover(ldddist(lddmap, OutletMap, L/2) , 1)) ; # Active ground-Area of the watershed (m2) A = maptotal(Size); # Deep infiltration capacity map (mm/h) # assuming it's proportional to slope > 20% and # deepinfil % x infiltrability InfilMin = lookupscalar(TblSoilPr,6, MapSoil) * DeepInfil ; InfilMax = lookupscalar(TblSoilPr,7, MapSoil) * DeepInfil ; InfilMin = if(S < 0.2 , 0.2 * InfilMin , S * InfilMin) ; InfilMax = if(S < 0.2 , 0.2 * InfilMax , S * InfilMax) ; #calculates the upstream drainage area for each cell drainarea = catchmenttotal(sqr(Length),lddmap); gullike = ln(drainarea / S) ; #report .\Outputs\gullike.map = gullike ; # ================================================================= # Dynamic section # ================================================================ #T= (time()-1)*D/TUnit;#Increment of crop duration by time step #Change LU map if running model for long time period #Map_LUT =if(T=xxx, lookupscalar(TblField,1,MapFields),Map_LUT); #CurDu =CurDu + T; #Current duration of crop (day) on the field #Current development stage (If current Dev. stage is longer than Dev4 then considered as Dev4) #CurDu = if(CurDu > (Dev1+Dev2+Dev3+Dev4),(Dev1+Dev2+Dev3+Dev4),CurDu); #CurDev = if(CurDu <= Dev1,1)* MapBound; #CurDev = if(CurDu >Dev1 and CurDu <= (Dev1+Dev2),2,CurDev)* MapBound; #CurDev = if(CurDu >(Dev1+Dev2) and CurDu <= (Dev1+Dev2+Dev3),3,CurDev); #CurDev = if(CurDu > (Dev1+Dev2+Dev3),4,CurDev); #Crop development stage (Dev1=Ini; Dev2=Dev.; Dev3= Mid; Dev4=Late) #Calculate Kc, Cov, Mn based on current development satage (CurDev) #Kc = lookupscalar(TblLandUse,CurDev+5,Map_LUT); Kc = lookupscalar(TblLandUse,8,Map_LUT); #Kc = if(CurDev eq 2,(Kc3-Kc1)/Dev2*(CurDu-Dev1)+Kc1,Kc); #Kc = if(CurDev eq 4,(Kc4-Kc3)/(Dev4)*(CurDu-Dev1-Dev2-Dev3)+Kc3,Kc); MapKc = Kc; #~Cov = lookupscalar(TblLandUse,CurDev+9,Map_LUT); #~Cov = if(CurDev eq 2,(Cov3-Cov1)/Dev2*(CurDu-Dev1)+Cov1,Cov); #~Cov = if(CurDev eq 4,(Cov4-Kc3)/(Dev4)*(CurDu-Dev1-Dev2-Dev3)+Cov3,Cov); #~MapCover = Cov; - 57 / 68- #Mn = lookupscalar(TblLandUse,CurDev+13,Map_LUT); Mn = lookupscalar(TblLandUse,16,Map_LUT); #Mn = if(CurDev eq 2,(Mn3-Mn1)/Dev2*(CurDu-Dev1)+Mn1,Mn); #Mn = if(CurDev eq 4,(Mn4-Mn3)/(Dev4)*(CurDu-Dev1-Dev2-Dev3)+Mn3,Mn); MapManning = Mn; dynamic #to be moved on top when mn & kc become dynamic # Rainfall (mm/h) per time step # !!!! R = timeinputscalar(TssRain,1)* D/ TUnit ; R = timeinputscalar(TssRain,1); # Actual Evapotranspiration (ETc = ETo*Kc) (mm/h) per time step # !!! ETc = timeinputscalar(TssRain,2) * Kc* D/ TUnit; ETc = timeinputscalar(TssRain,2) * Kc ; # ================================================================= # Discharge computation section # ================================================================= # Water content in each cell after rainfall (mm) Qmmt = Qmmt + R ; # Water reservoir on/in each cell after rainfall (mm) # Infiltration map mm), assuming it's sqroot (water stock) # !!! infilmap = if(Qmmt > gapmax, InfilMax *D/TUnit,InfilMax * sqrt(Qmmt/gapmax) * D/TUnit) ; infilmap = if(Qmmt > gapmax, InfilMax / 3600 * d,InfilMax / 3600 * d * sqrt(Qmmt/gapmax)) ; # Assumimg when rainfall intensity > crustint mm/h and slope < crustslope % #crust is formed infilmap = if(R > CrustInt and S < CrustSlope , 0 , infilmap); # !!! infilmap = if(R/D*TUnit > CrustInt and S < CrustSlope , 0 , infilmap); # Evapotranspiration, assuming it's proportional to sq (water stock) (mm) ETc = if(Qmmt > gapmax, ETc , ETc * sqrt(Qmmt/gapmax)) ; # Water remaining in each cell after infiltration and ETc (mm) Qmmt = if(Qmmt - infilmap - ETc < 0 , 0 , Qmmt - infilmap - ETc) ; # Height of the water on each cell (mm) # !!! Hmm = Qf / Size * D * 1000 + Qmmt ; Hmm = Qf /Size * 1000 + Qmmt ; # Saturation of cells (% remaining storage capacity) #Sat = 100 * Hmm / (depthmap * 1000) ; # Velocity of the flow in one cell (m/s) assuming manning coefficient of stream is rivmann # and minimum velocity = one day to cross one cell #V = if(Rivdist < 1 , sqrt(S) / RivMann * sqr(Hmm/gapmax) * (Hmm/1000)**(2/3) * WpVel ,sqrt(S) / MapManning * sqr(Hmm/gapmax) * (Hmm/1000)**(2/3) * WpVel ) ; - 58 / 68- V = if(Rivdist < 1 , sqrt(S) / RivMann * (Hmm/1000)**(2/3) * WpVel , sqrt(S) / MapManning * sqr(Hmm/gapmax) * (Hmm/1000)**(2/3) * WpVel) ; #V = if(V > L/24/TUnit , V , L/24/TUnit) ; V = if(V > L/86400 , V , L/86400) ; # Time taken by the flow to cross one cell (second) Tcross = L / V ; # Time to reach the weir (second) Tm = ldddist(lddmap, OutletMap, Tcross / Length) ; # Part of the flow reaching the weir (%) in one timestep # !!! P = if(MapDEM > 0 , cover(if(D / Tm > 1 , 1 , D / Tm) , 1)) ; P = if(MapDEM > 0 , cover(if(d / Tm > 1 , 1 , d / Tm) , 1)) ; #Discharge (cumulated run-off) (m3/timestep) !!!(m3/s) # !!! Qf = accuflux (lddmap, Qmmt * P) / 1000 * Size / D ; Qf = accuflux (lddmap, Qmmt * P) / 1000 * Size ; # Total runoff (m3) # !! Qt = Qt + Qf * D ; #Qt = Qt + Qf ; # Remaining water in cell for next pitch (mm) Qmmt = Qmmt * (1 - P) ; # ================================================================= # Erosion computation section # ================================================================= # factor k, formed by substituting formula for transport #~k = ((seddenmap*S*(L**-0.4))/(((seddenmap/Wat)-1)*sedvelmap)) #~ *(((S**0.5)/MapManning)**0.6) ; #~ex = 0.001*((k**cohesivemap)*(R**(0.4*cohesivemap))*Qf*exp(-1/MapManning*MapCover)) * 10000 / Size ; #~e = e + ex ; #~report Map_e = e; #~ec = k*(Qf**0.4)*exp(-1/MapManning*MapCover) ; #~Cap = 0.001 * k*(Qf**0.4)*exp(-1/MapManning*MapCover) * Qf * D ; #~eflux = accucapacityflux(lddmap,ex,Cap) ; #~report Map_eft = eft + eflux ; #~report Map_eStore = eStore + accucapacitystate(lddmap,ex,Cap) ; #report MapEstore = log10(log10(eStore+1)+1) ; # ================================================================= # Reporting section #================================================================== report Out_tssQf = timeoutput(MapOut_ord,Qf) ; #report gapmax = gapmax ; #report drainarea = drainarea ; #report (d10) Map_Qfi = Qf ; #report manning = MapManning ; - 59 / 68- #report Out_tssQt = timeoutput(MapOut_ord,Qt) ; #report Active_area = A ; #report (d10) Map_Vi = V ; #report (d10) hmmgap = sqr(Hmm/gapmax) #report Map_Pi = P ; #report Map_Qmmt = Qmmt ; #report Map_sat = Sat ; #report Hmm_map = Hmm; #report S_map = S; #report lddmap = lddmap ; #report Cellsize = Size ; - 60 / 68- Annex 4 Orange et al. (2007). Actes des Premières Journées Scientifiques Inter-Réseaux de l’AUF, Hanoi, 6-9 novembre 2007, Editions AUF et IRD, cd-rom : 6p. Le concept de PES pour une gestion durable des eaux et des sols : application au développement de l’élevage et au contrôle environnemental dans le Nord Vietnam Didier ORANGE1,2,*, L. BARDOUIN1,2, NGUYEN Duy Phuong2, J.-B. LOISEAU1,2, F. CLEMENT1,2,3, P. JOUQUET2,4 1 UR176 SOLUTIONS, IRD, seconded IWMI-SEA, 32 Avenue H. Varagnat, 93143 Bondy Cedex, France 2 Soils and Fertilizers Institute (SFI), Dong Ngac, Tu Liem, Hanoï, Vietnam 3 CLUWRR, Newcastle University, England 4 Lab.Ecologie des Sols Tropicaux, IRD, UMR 137 Biosol, Bondy, France E-mail : orange@ird.fr Abstract PES concept for a sustainable water and soil management: applying to cattle husbandry rising and environmental control in Northern Vietnam In Northern Vietnam, the urgent environmental problems are soil erosion under annual crop on sloping lands leading to a drastic soil fertility decrease, inducing the risk to unfertilize the flatlands and to fill up the water reservoirs in the downstream part by sediment accumulation. Then a lack of disposable income for the purchase of inorganic fertilizers has resulted in farmers in Northern Vietnam resorting to the utilization of raw animal and human waste in order to fertilizer their cropping systems. This project is dealing at the District level in the Northward of Hoa Binh Province with the PES use (agro-ecological concept of Payment for Environmental Services) based on a new market opportunity for the farmers of Northern Vietnam to mobilise the integrated cattle husbandry enhancement and water management between uplands and lowlands. The purpose is to create a sustainable loop between fodder cropping on sloping lands, the cattle development under stables, the animal and human waste management, leading to human health and water quality protections, to the erosion sediment control for both a sustainable upland and lowland use. The proposed project offers a novel approach in that a holistic approach is introduced to achieve income generation from livestock manure management in mountainous environment of Southeast Asia in establishing a process of comprehensive assessment, of mutual learning between farmers from upper part and down part of the watershed through a hydrological modelling platform, and by the development of new agricultural technologies as biogas digester and vermicompost unit. Mots clés : Gestion durable, Erosion, Pollution, Elevage, PES, Biogas, Vermicompost, Vietnam Introduction Dans le Nord Vietnam, l’usage des collines et petites montagnes aux pentes souvent fortes (majoritairement supérieures à 40%) pour des cultures annuelles (selon la séquence riz, mais, manioc en fonction de la perte de fertilité du sol) a provoqué une érosion forte entrainant une dégradation des sols interdisant à terme toutes cultures annuelles rentables (Maglinao et al., 2003 ; 2003 ; Toan et al., 2004 ; Castella et al., 2005). Or comme dans la plupart des régions montagneuses du Sud-Est asiatique, les aménagements pour la gestion de l’eau sont très répandus, la multiplicité des réservoirs d’eau (de tailles différentes selon leur localisation entre l’amont et l’aval du bassin versant) permet une utilisation optimale de l’eau pour l’irrigation des terres cultivées en aval (riziculture essentiellement) mais aussi l’alimentation en eau des habitations. L’érosion des pentes représente un risque majeur pour ces aménagements : comblement des réservoirs par accumulation de sédiments, pollution des eaux. De plus, une pratique courante est - 61 / 68- l’utilisation des déjections animales et humaines pour la fertilisation des sols, induisant de graves problèmes de santé publique relevés dans tout le Nord du Vietnam et dans le Sud de la Chine (Phuc et al., 2006) mais aussi de forts risques de pollution des eaux des barrages (Smit et Nasr, 1992). L’utilisation de ce type de fumure est due à trois raisons qui restent liées entre elles : manque d’engrais organiques de type fumier, manque de revenus et absence d’installations sanitaires (latrines). Dans ce contexte, développer des pratiques agricoles pour de petits exploitants permettant d’une part, la durabilité des systèmes de cultures sur pente et d’autre part, prenant en compte l’amélioration de la qualité particulaire et dissoute des eaux d’écoulement pour son utilisation en aval est un défi actuel majeur. Le projet MSEC a prouvé l’intérêt de la culture fourragère sur les terres de pente pour contrôler les phénomènes d’érosion et de perte de fertilité des sols (Orange et al., 2007). Sur la base de ces résultats de recherche et en fonction de la politique agricole appliquée dans la région, l’étude présentée ici s’interroge sur l’intérêt de l’utilisation du concept de Payment for Environmental Services (Gutman, 2003 ; Wunder, 2005 ; Swallow, 2006) pour le développement de la culture sur pente des produits fourragers et de l’élevage en étable afin d’une part, de protéger les sols contre l’érosion et d’autre part, de gérer les excrétions animales et éviter ainsi les pollutions des eaux d’écoulement et des eaux de barrage. Le projet inclue alors l’expérimentation de mécanismes qui permettront aux exploitants situés en aval (ou autres acteurs affectés par les pratiques agricoles des exploitants situés en amont) de participer contractuellement aux efforts des exploitants amont à une meilleure gestion de leurs systèmes agraires pour une protection environnementale des eaux de barrage aval. Plus concrètement, la question est de savoir comment mobiliser les petits exploitants de l’aval à payer des services environnementaux (écosystémiques) afin de motiver les exploitants de l’amont à gérer durablement sols et eaux. Contexte géographique et socio-économique La zone expérimentale de 8 km2 centrée sur le village de Dong Cao (fig. 1) se situe dans le district de Luong Son de la province de Hoa Binh, à 60 km à l’Ouest de la ville d’Hanoi, dans la commune de Tien Xuan qui compte une quinzaine de villages. C’est une région de petites montagnes et collines fortement exploitées depuis une trentaine d’années. Les bas-fonds sont annuellement utilisés par deux cycles de culture irriguée de riz et un cycle de légumineuses alors que les pentes sont traditionnellement utilisées pour le maïs et le manioc. La pluviométrie, entre 1200 et 2000 mm par an, connaît de fortes variations interannuelles, en particulier au niveau de la distribution saisonnière des averses. Les sols des versants (en majorité des acrisols) sont engagés aujourd’hui dans un cycle de dégradation rapide, engendré par les phénomènes d’érosion liés aux pratiques intensives de cultures annuelles (Toan et al., 2004). Depuis les années 2000, les collines se reforestent (Clément et al., 2006) et l’élevage bovin se développent conduisant a une augmentation importante du cheptel et des changements d’usage rapides des pentes. Ces initiatives sont en partie dues à des programmes incitatifs du gouvernement (programme de reforestation, programme de promotion de d’élevage), mais aussi par la perte de rendements du manioc sur ces terres dégradées et la disparition d’une gestion collective des terres (Clément et al., 2007). L’absence de politique de partage des eaux et de règles collectives de mise en pâture des terrains laisse présager des conflits d’usage dans un futur proche. Le bassin étudié culmine à 450 m, les eaux d’écoulement sont utilisées dans les zones basses pour l’irrigation du riz par les habitants de l’amont (notamment les agriculteurs du village de Dong Cao) puis les eaux sont ensuite collectées dans un réservoir à 3 km de là pour l’irrigation d’hiver des zones situées à l’aval du barrage par les agriculteurs du village aval (fig. 1). - 62 / 68- Village de Réservoir Dong Cao Limite de la zone expérimentale Villag 1 km e aval Figure 1 : Carte de situation de la zone expérimentale (bassin versant de 8 km2 délimité en trait vert). Agronomie, Agro-écologie et PES Le concept des PES (Balmford et Whitten, 2003 ; Wunder, 2005) est basé sur le principe suivant : « les utilisateurs de ressources et les collectivités qui sont en mesure de fournir des services écologiques doivent recevoir une compensation, et ceux qui bénéficient de ces services doivent les payer » (Wunder, 2005). Il n’existe pas de définition généralement convenue des systèmes de PES, mais plutôt une série de classifications reposant sur le type de services environnementaux, la portée géographique, la structure des marchés ou le type de paiement utilisé. Cette absence de définition ou de classification commune témoigne de la grande diversité des modèles existants. Elle créée également une certaine confusion et un manque de précision dans la littérature en ce qui concerne les systèmes devant être considérés sensus stricto comme des paiements de services environnementaux. Les marchés pour les services de protection des bassins hydrographiques, qui ne font pas intervenir l’échange de « produits » tels que la quantité ou la qualité d’eau, mais plutôt le financement des « utilisations » des terres qui génèrent des avantages hydrographiques (Kerr, 2002 ; Pagiola et al., 2005) en sont un parfait exemple. En résumé, la bibliographie s’entend pour dire que les systèmes de PES sont parfaitement efficaces lorsque les services sont visibles, que les bénéficiaires sont bien organisés, et que les collectivités utilisant les terres sont bien structurées, possèdent des droits de propriété clairs et protégés, utilisent un cadre juridique solide et sont relativement riches ou ont accès aux ressources (Wunder, 2005). Actions sur le terrain et discussion 1. Hypothèses de départ Les différentes enquêtes menées auprès des villageois de la commune de Tien Xuan ont montré que ni la baisse de fertilité des sols de pente ni leurs risques d’érosion n’influaient en priorité sur les stratégies paysannes qui sont avant tout opportunistes, très liées aux lois du marché et basées sur une perception locale de l’environnement (Clément, 2006). Il en résulte un pragmatisme qui permet des changements d’usage rapides et spectaculaires (Clément et al., 2005 ; Noble et al., 2006). Aussi au lieu de construire notre intervention sur une distinction entre connaissances locales et connaissances scientifiques, qui est en fait une construction abstraite souvent diffusée par la communauté scientifique (Forsyth, 1996), notre hypothèse de travail fut que les différents intervenants (de l’amont et de l’aval, et les décideurs) promeuvent -- aient pour volonté -- la durabilité des fonctions écosystémiques de leur région, se traduisant concrètement en terme de - 63 / 68- rentabilité des terres et des eaux (« gagner plus ? »), d’accès à la modernité (électricité, chauffage,…) et à la santé (« vivre mieux »). Le schéma d’action suivant a ainsi été conçu (fig. 2). Dans le village de Dong Cao et les villages alentours de la commune de Tien Xuan, les collines sont laissées en pâture libre aux bovins de plus en plus nombreux, provoquant conflits entre agriculteurs voisins et mobilisant une érosion linéaire nouvelle. Les risques hors-sites immédiats sont alors une aggravation du comblement de la retenue d’eau en aval et une pollution accrue des eaux d’écoulement par les déjections animales. Aussi l’application du concept de PES veut créer un lien contractuel entre les usagers de la retenue à l’aval et les usagers des collines à l’amont afin que ces derniers vendent aux premiers la protection environnementale des eaux d’écoulement de l’amont. Cependant, il reste à trouver la bonne « formulation » pour motiver les différents dépositaires. Figure 2 : Schéma d’action du concept de PES montrant les liens entre utilisateurs amont et utilisateurs aval des services écosystémiques. 1: impact sur la santé; 2: impact sur la qualité de l’eau et sédimentation; 3: impact sur les rendements agricoles; 4: comment appliquer le concept de PES pour créer une boucle de rétroaction afin de garantir un développement durable des systèmes agraires amont et aval. 2. Recherche-action : Biogas et vermicompost Fort du pragmatisme local, l’application des PES est testée par le biais de la diffusion de deux nouvelles technologies de gestion des déjections animales permettant d’apporter aux agriculteurs des revenus nouveaux et d’améliorer leur qualité de vie : les unités de biogaz (Hobson et Fielden, 1982 ; Sommer et Husted, 1995) et de vermicompost (Arancon et al., 2005 ; Padmavathiamma et al., 2007). La transaction contractuelle possible serait la participation aux frais des bénéficiaires amont par les utilisateurs aval, pour la construction de ces installations chez les exploitants concernés. Ces deux technologies peuvent également contrer l’effet de déboisement lié à la recherche de bois de chauffage et donc d’énergie. La production de biogaz assure ce dernier point en fournissant du gaz pour la cuisine mais aussi de l’électricité. Le résultat du vermicompost, outre outre la production de vers pour l’alimentation du petit élevage (porcs, volailles), donne un engrais engrais riche en nutriments qui peut facilement remplacer ou compenser l’utilisation d’engrais chimiques. Enfin, ces techniques nécessitent un élevage en étable, ce qui permet de ne plus laisser les animaux en divagation sur les pentes, ce qui était source de conflits de voisinage et d’érosion. Ces deux procédés permettent ainsi la concentration des déchets d’élevage au même endroit, limitant et évitant la diffusion de leurs pouvoirs polluants en particulier dans la rivière, source principale d’eau potable de la population et alimentant le réservoir et les rizières de bas fond. Ils permettent la production d’engrais organiques et azotés qui peuvent constituer la base de la - 64 / 68- fertilisation d’une exploitation (pour le riz, manioc et jardin familial) ou tout du moins la compléter, ce qui diminuera considérablement les coûts en achat d’engrais chimiques. 3. Modélisation hydrologique et plateforme de concertation Enfin, le dernier terme de cette étude est l’utilisation de la modélisation hydrologique acquise dans le projet MSEC (Do Duy Phai et al., 2006) pour explorer l’utilisation potentielle de ce concept de PES par la construction d’une plateforme de simulations répondant à la fois aux forçages climatiques, biophysiques et d’usage des sols. Cette plateforme permet donc la construction de scénarios d’usage répondant aux attentes et interrogations des usagers et décideurs régionaux. En ce sens, elle est une plateforme de concertation et peut être considérée comme un outil d’aide à la gestion environnementale pour assister à la fois les décideurs régionaux et les décideurs locaux que sont les agriculteurs (Poncet et al., 2001 ; Orange, 2002). Elle répond aux exigences mises en exergue par Clément et Amezaga (2007) pour la réussite d’un développement agricole durable dans le Nord Vietnam, à savoir permettre le choix des stratégies agricoles aux petits exploitants locaux en connexion avec les instances régionales de vulgarisation des politiques agricoles gouvernementales. Outre de quantifier les pertes en sol, les disponibilités en eau de surface et les flux d’azote, cette plateforme aura également pour composantes : (1) d’évaluer les coûts à court terme des stratégies agricoles, (2) de comparer les coûts environnementaux et économiques à moyen terme entre agriculteurs amont et agriculteurs aval, et (3) de simuler l’impact des stratégies de PES sur les revenus agricoles et sur les fonctions écosystémiques. Conclusion L’objectif global de ce projet de recherche est d’évaluer la possibilité d’un développement agricole durable et socialement acceptable, qui soit porté par les communautés locales et basé sur l’acceptation d’une participation locale aux coûts des services environnementaux. Il est évident que cette démarche ne peut être atteinte que de manière indirecte. Ainsi la méthode promue consiste à proposer de nouvelles techniques agro-écologiques (biogaz et vermicompost) en étant persuadé que le potentiel de diffusion sera fortement lié aux intérêts économiques des individus engagés dans le processus. En résumé, on ne parle pas de « lutte contre l’érosion » ni de « lutte contre la pollution des eaux », mais de production de biens agricoles, de rentabilité économique de l’usage du sol et de l’eau et de facilité d’accès au bien-être (cuisine, santé). La méthode originale est de coupler la démarche de recherche fondamentale sur les processus biophysiques et écologiques de ces nouvelles techniques avec une mise en évidence des avantages économiques et écologiques auprès des individus cibles via un modèle de simulations servant de plateforme de concertation. Cette plateforme réalise des simulations de scénarios d’usage traduisant en termes de gains et pertes les impacts respectifs des stratégies d’usage de la ressource eau et sol des usagers de l’amont et de l’aval. L’idée est de mettre en exergue la meilleure option individuelle pour une gestion durable commune des sols et des eaux. Enfin par cet exemple, il apparait que le concept de PES participe à redéfinir un nouvel équilibre économique d’une agriculture productive et respectueuse de la nature par la rétribution des services environnementaux dans la cadre d’une politique contractuelle largement basée sur un consensus entre acteurs locaux. On peut espérer que cette première étape de l’agro-écologie nous conduise vers une acceptation des coûts écologiques par la société rurale vietnamienne du Nord, tels que la fourniture et la régulation de l’eau, l’épuration de l’eau, la formation des sols et les services récréatifs, pour ne citer que ceux directement concernés par les sociétés paysannes des zones montagneuses d’Asie du Sud-Est. Par ailleurs, il est à noter que les PES offrent l’opportunité d’intégrer l’élevage dans un concept de développement durable en terme agro- écologique en créant de nouveaux liens entre utilisateurs amont et aval du bassin versant. La pertinence de cette approche se trouve dans la création d’un marché novateur pour les petits agriculteurs du Nord Vietnam. Remerciements - 65 / 68- Cette action de recherche est réalisée dans le cadre du programme MSEC de l’IWMI coordonnée par l’unité de recherche IRD-UR176 (SOLUTIONS) de Christian Valentin, avec un soutien financier spécifique PES du CPWF thème 2 (Challenge Program for Water and Food) dans le cadre de l’action “Exploring the relevance and feasibility of PES approaches for producing environmental services through changes in agricultural practices: a case study in the Mekong Region” (contrat C-056-07). Nous remercions aussi le soutien financier de l’entreprise TOTAL au Vietnam dans le cadre du programme “Project on Innovative Practices for Environment and Human Health: A Pilot Phase in Northern Vietnam”. Bibliographie Arancon N.Q., Edwards C.A., Bierman P., Metzger J.D., Lucht C. (2005). Effects of vermicomposts produced from cattle manure, food waste and paper waste on the growth and yield of peppers in the field. Pedobiologia, 49(4) : 297-306. Balmford A., Whitten T. (2003). Who should pay for tropical conservation, and how could the costs be met? Oryx,37(2): 14. Castella J.C., Boissau S., Trung T.N., Quang D.D. (2005). Agrarian transition and lowland-upland interactions in mountain areas in northern Vietnam: application of a multi-agent simulation model. Agricultural Systems, 86 (3), 312-332. Clément F. (2006). How do farmers make decisions in a land degradation context ? Sustainable watershed management in cultivated sloping lands of SEA. Proceedings of the 2nd International Conference on Sustainable Sloping Lands and Watershed Management, LuangPhrabang, Laos. 12-15 December 2006 : 64-65. Clément F., Amezaga J. M. (2007). Linking reforestation policies with land use change in Northern Vietnam: Why do local factors matter. J. Geoforum, doi:10.1016, Elsevier. Clément F., Amezaga J.M., Orange D., Tran Duc Toan (2007). The impact of government policies on land use in Northern Vietnam: an institutional approach for understanding farmer decisions. IWMI Research Report, 112, Colombo, Sri Lanka, International Water Management Institute (IWMI), 27 p. Clément F., Amezaga J.M., Orange D., Tran Duc Toan, Large A.R.G., Calder I.R. (2006). Reforestation Policies and Upland Allocation in Northern Vietnam: An institutional approach for understanding farmer strategies and land use change. Proceeding of International Symposium “Towards sustainable livelihoods and ecosystems in mountainous regions”, 7-9 March 2006, Chiang Mai, Thailand, 17 p. Clément F., Nguyen Duy Phuong, Orange D., Calder I.R., Amezaga J.M., Large A.R.G. (2005). Scientific narratives and policy: the need for a pragmatic approach regarding land management in Vietnam. IV MMSEA Conference on “Sustainable use of natural resources & poverty dialogue in mainland Montane, South East Asia”, SaPa, Vietnam, 16-19 May 2005: 13 p. Costanza R., d'Arge R., de Groot R., Farber S., Grasso M., Hannon B., Limburg K., Naeem S., O'Neill R.V., Paruelo J., Raskin R.G., Sutton P., van den Belt M. (1997). Value of the world’s ecosystem services and natural capital. Nature, 387 : 253-260. Do Duy Phai, Orange D., Migraine J.B., Tran Duc Toan, Nguyen Cong Vinh (2006). Applying GIS-assisted modelling to predict soil erosion or a small agricultural watershed within sloping lands in Northern Vietnam. Sustainable watershed management in cultivated sloping lands of SEA. Special Review about the 2nd International Conference on Sustainable Sloping Lands and Watershed Management, LuangPhrabang, Laos. 12-15 December 2006: 19 p. Forsyth T. (1996). Science, Myth and Knowledge: Testing Himalayan environmental degradation in Thailand. J. Geoforum, 27, 3 375-392. Gutman P., ed. (2003). From goodwill to payments for environmental services. Washington DC: WWF, Macroeconomics for Sustainable Development Program Office. Hobson P. N., Feilden N. E. H. (1982). Production and use of biogas in agriculture. Progress in Energy and Combustion Science, 8(2) : 135-158. Kerr J. (2002). Watershed development, environmental services, and poverty alleviation in India. World Development, 30(8): 1387-1400. Maglinao A.R., Valentin C., Penning de Vries F., ed. (2003). From soil research to land and water management: harmonizing people and nature. Proceedings of IWMI-ADB project annual meeting and 7e MSEC assembly, Vientiane, Laos, 2-7 December 2002, IWMI, Colombo (Sri Lanka) : 250 p. Noble A.D., Bossio D.A., Penning de Vries F.W.T., Pretty J., Thiyagarajan T.M. (2006). Intensifying agricultural sustainability: an analysis of impacts and drivers in the development of “bright spots”. Comprehensive Assessment Research Report 13, Colombo, Sri Lanka, International Water Management Institute (IWMI), 42 p. Orange D. (2002). Projet Gihrex : ambitions et acquis d’un projet de recherche pour le développement durable du delta intérieur du Niger. In Orange D., Arfi R., Kuper M., Morand P., Poncet Y. (éd.) : Gestion intégrée des ressources naturelles en zones inondables tropicales, Actes du colloque Girnzit, Bamako, 20-23 juin 2000, coll. Colloques et séminaires, IRD, Paris : 953-967. - 66 / 68- Orange D., Podwojewski P., Tran Duc Toan, Pham Van Rinh, Do Duy Phai, Nguyen Duy Phuong (2007). Impact of land use on bed load transported by rivers in the Dong Cao watershed, north Vietnam. Water Figures Asia, IWMI, Colombo (Sri Lanka), issue 1: 4-5. Padmavathiamma P.K., Li Loretta Y., Kumari U.R. (2007). An experimental study of vermi-biowaste composting for agricultural soil improvement. Bioresource Technology. Pagiola S., Arcenas A., Platais G. (2005). Can payments for environmental services help reduce poverty? An exploration of the issues and the evidence to date. World Development, 33(2): 237-253. Phuc P.D., Konradsen F., Phuong P.T., Cam P.D., Dalsgaard A. (2006). Practice of using human excreta as fertilizer and implication for health in Nghean Province, Vietnam. Southeast Asian J Trop Med Public Health, 37: 222-229. Poncet Y., Kuper M., Mullon C., Morand P., Orange D. (2001). Représenter l’espace pour structurer le temps : la modélisation intégrée du delta intérieur du Niger au Mali. In Lardon S., Maurel P., Piveteau V. (Eds) : « Représentations spatiales et développement territorial », Hermes, Paris, chapitre 7 : 143-162. Smit J., Nasr J. (1992). Urban agriculture for sustainable cities: using waste and idle land and water bodies as resources. Environment and Urbanization, 4(2). Sommer S.G., Husted S. (1995). Chemical composition of the buffer system in livestock and biogas plant digested slurry. J. Agric. Sci., 124 :45–53. Swallow B. (2006). Compensation and Rewards for Ecosystem Services in the tropics: Interface between conservation, poverty reduction and social justice. Presentation to International Symposium on Defying Nature’s End: the African Context, Antananarivo, Madagascar, 22 June 2006. Toan T.D., Podwojewski P., Orange D., Phuong N.D., Phai D.D., Bayer A., Thiet N.V., Rinh P.V., Renaud J., Koikas J. (2004). “Effect of land use and land management on water budget and soil erosion in a small catchment in northern part of Vietnam”. In International conference on innovative practices for sustainable sloping lands and watershed management, 5-9 September 2004, Chiang Mai, Thailand. Wunder S. (2005). Payments for environmental services: Some nuts and bolts. CIFOR Occasional Paper, 42, Jakarta, Indonesia, 31 p. - 67 / 68- - 68 / 68-