Rice ? livestock integration and soil fertility management in irrigated and flood recession areas of the Tonle Sap Lake region in Cambodia

Auteur de la fiche (Prénom, NOM, structure, email) : Nicolas FAYSSE, Cirad UMR G-Eau/AIT, faysse@cirad.fr
CoAuteur(s) de la fiche (Prénom, NOM, structure, email) : Florent TIVET, Cirad UMR AIDA, tivet@cirad.fr Sophoan MIN, AVSF, s.min@avsf.org / Pierre-Yves LE GAL, CIRAD UMR Innovation, pierre-yves.le_gal@cirad.fr
Défis du cadre stratégique : Renforcer la durabilité environnementale des agricultures irriguées , Contribuer au développement économique et social des territoires, Accompagner le changement par l’innovation technique et les réformes institutionnelles
Aires géographiques concernées : Asie du Sud Est
Terrains concernés : Cambodia ? Tonle Sap Lake region
Types de systèmes irrigués : The study will take place in two geographic areas around the Tonle Sap Lake: (1) Battambang province (north-west, ?rice bowl?, Banan district) with the rice-livestock systems of the flood plains of the Tonle Sap lake, and (2) Kampong Thom province with the irrigation scheme of Stung Chinit (Santuk district) on the upper sandy terraces.
Contexte et problématique : In Cambodia, most rice paddies lie in gently sloping lowlands but field water availability varies from upper (coarse-textured soils) to lower positions (clayed soils) of a sloping land in the rain-fed lowland ecosystem of the Tonle Sap region, causing variation in yield within the toposequence. In the hydromorphic plains of the Tonle Sap lake, rice is commonly grown only once a year as a mono-crop in the wet season with still low input and relatively low output even if the yield has increased from 1.4 to 2.8 t/ha over the last 20 years (Fukai and Ouk, 2012). Animal husbandry is one of the main components of the rice farming system. Cattle and buffalo are mainly farmed for meat production that will be sold on local markets. However, these animals are in a constant state of undernourishment as they rely on poor-quality roadside grasses and rice straw as their primary source of nutrition in both wet and dry seasons (Pen et al. 2010; Nampanya et al. 2012). In addition, farmers have to travel long distances to ?cut and carry? roadside grass or care for animals grazing rice stubble (Maxwell et al. 2012). The scarcity of labor generated tremendous changes in the rice farming systems with a drastic increase of machinery (power tiller, tractor, combine harvester ?). Farmers shifted from rice transplanting to seed broadcasting due to labor shortage, time constraints, and to reduce cost of production. This change is the illustration of an extensive rice farming system where smallholders invest in others activities (on- and off-farm) to reduce financial risks. In addition, many irrigation schemes are facing challenges of limited profitability and of low soil fertility.

In such a context, there is an opportunity to test innovative rice cropping systems based on ecological intensification principles to build soil resilience (increasing soil organic C content, soil biota and nutrient cycling), to diversify crop productions and to improve rice ? livestock integration. The strategy is to increase the resilience of the rice farming systems through a diversification process with the establishment of fodder legumes and grasses after wet season rice. This diversification process may be used to achieve various objectives: (i) provide additional fodder sources during the dry season (seasonality, spaces in the territory) and (ii) contribute to an improvement of the soil fertility (improvement of SOC, N, structure, biota), either directly or indirectly (increased availability of manure), with expected positive returns on rice cropping system (labor profitability and income).

Seeds of fodder legumes (Stylosanthes guianensis and Centrosema pascuorum) can be broadcasted before or just after rice harvesting. In flood recession areas, this can be done when water still remains on the field allowing a long-lasting period before the next cropping season. These plants can improve soil fertility (i.e., N fixation, soil organic C & N accumulation, and nutrients cycling ?) as observed in an experiment already in place in the Stung Chinit irrigation scheme set up by the Department of Agricultural Land Resources Management and CIRAD (Leng et al., forthcoming).
These crops can become an additional resource of biomass in dry season around the Tonle Sap Lake. Farmers may actually develop three types of use of this new resource: i) they can leave it in the soil to restore soil fertility; ii) they may use it to feed their cattle (cut and carry or rotational grazing); iii) they may sell it to local middlemen that harvest it and sell it on local markets. This opportunity to increase and diversify the fodder resources seems promising as many farmers in these areas express an interest for increasing cattle farming (especially fattening) activities in the future.

The development of the use of these crops, however, faces two main constraints. First, cattle traditionally freely roam and graze during the dry season. These animals belong to local farmers (or sometimes to neighboring villagers) and graze rice crop residues and cover/relay leguminous crops. The lack of control of roaming animals prevents farmers from using effectively this new fodder source during the dry season and from investing in fertility management by leaving part of the leguminous crop on the soil surface before planting the next rice crop. Keeping rice residues and other fodder source are also impaired by burning rice fields in the dry season. A second potential constraint relates to the economic profitability of such diversification, especially taking into account labor scarcity. Even though farmers may acknowledge the positive impacts of diversification process with leguminous species in terms of, for instance, soil fertility management, they are often compelled to focus on the short term. They can implement those practices and crop additional fodder sources only if there is a clear opportunity for improvement of their income in the short term (cattle raising, selling fodder?), especially taking into account the opportunity costs of farm labor. Thus, for farmers to be willing and able to integrate ?secondary crops? such as leguminous species after wet season rice, there needs to be short-term co-benefits (e.g. for cattle feeding) or a specific market demand for this new fodder source (or for seeds) in the dry season.
Implication des acteurs de l'irrigation : The Department of Agricultural Land Resources Management (DALRM) from the General Directorate of Agriculture (GDA) has already been involved in initial experiments of development of fodder legumes in the Tonle Sap Lake Region. Moreover, in 2017, an initiative was launched and implemented by the Wildlife Conservation Society, to accompany better farm practices for 2,000 farmers of the Tonle Sap Lake Region and to organize certification and payment to farmers of premium price based on the Sustainable Rice Platform standards. Partnership with such initiative may enable a large-scale up-scaling of research results. First contacts with actors of this initiative have been made in this regard.
Objectifs de l'action (cibles, intérêts opérationnels, etc) : The aim is to test and assess the use of fodder legumes with farmers? groups of irrigated areas and flood recession areas in the Tonle Sap Lake region. More specifically, the research will: i) test options in terms of implementation of fodder legumes, collective land use management, and use of these legumes; ii) assess the impacts of such implementation, in terms of soil fertility, evolution of farming systems and economic profitability.
Contenus, moyens et calendrier à mettre en place : The work will take place over a period of 18 months. The work involves 3 work packages. The work will be implemented in two sites (Banan district in the flood recession area and in the Stung Chinit irrigation scheme).

A junior field assistant will be engaged full time in the process by AVSF, over a period of 18 months. The assistant will be involved in implementation of work packages 1 and 2, and will provide some support for work package 3. A Master thesis will be specifically conducted in the frame of work package 3.

Work package 1. Co-designing and assessing rice ? livestock integration: analysis of catch crops at field level
Which cropping system (rice ? fodder species) can bring additional fodder sources in the dry season while improving the efficiency (soil fertility, labor profitability and income) of rice farming?
This work package will organize an agronomic assessment of opportunities for production of catch crops after rice in various soil/water control contexts and about possible uses (required management, quantities produced, impacts on soil fertility of different ways to use fodder legumes). On each location (irrigated scheme and flood plains), networks of smallholder farmers are already engaged in this diversification process after rice. A range of field operations (fodder seed broadcasting and use of appropriate-scale machinery) is jointly conducted by farmers and agronomists of GDA/DALRM and CIRAD. Biomass production of the fodder species will be assessed and possible management (rotational grazing, cut-and-carry?) tested. It is also expected to conduct an assessment of the soil ecosystem services using simple but reliable field tools already validated by UMR Eco&Sols, GDA/DALRM in partnership with the Royal University of Agriculture. Conventional management and diversified rice cropping systems will be compared.
This will take place from Month 1 to Month 18.

Work package 2. Supporting collective thinking of land management during the dry season
Which collective decision-making process could emerge to foster the co-management of cattle production systems and rice cropping systems? Which technical, organizational and cultural/social changes will be induced?
This work package will first consist in supporting assessment by farmers? groups about options for collective land management to enable the development of catch crops, in particular in terms of control of roaming of cattle and control of fire. Second, this collective land management will be tested and evaluated in two sites (one in an irrigated area, the other in a flood recession area). This second action will start with the broadcasting of fodder legumes.

Work package 3. Participatory analysis of options for producing catch crops at farm level
Which ?entry point? (increased incomes, returns to labors, time for other activities) can be identified to engage farmers in growing and feeding introduced forages and thus fostering cattle production in rice-farming systems?
This work package will undertake a technical and economic assessment of farm possible options, in terms of: i) the surface that will be planted with legume crops; ii) the use of these crops. The work will be based on the use of the CLIFS (Crop Livestock Farm Simulator, Le Gal et al., 2013; Ryschawy et al., 2014) approach that make it possible to test different scenarios of change in farming systems involving both crop production and animal breeding. This work will be implemented in 10 farms, which, for example, will represent actual diversity of farming systems; in terms of farm size and importance of the cattle breeding activity. For each farm, after a baseline characterization, different scenarios will be defined in terms of changes at farm level (introduction of cash crops, orientation and sizing of animal enterprise, breeding and cultivation practices), simulated and assessed in terms of biomass and manure balances and of expected benefits. The whole process will be conducted individually with each farmer. Collective workshops will be organized at the end of the process to disseminate the results obtained with the 10 studied farms.
Livrables : Work package 1. Report identifying the various agricultural practices possible for fodder legumes in study areas and characterizing impacts on soil fertility of different ways to use fodder legumes.
Work package 2. Report presenting the experience of the participatory process conducted with farmers? groups to assess options for collective land management.
Work package 3. Master thesis on the technical and economic assessment of possible options to use fodder legumes in the case of several farms in the study areas.
Budget détaillé : ? Salary costs of junior assistant: 500 euros per month, so a total of 9,000 euros. Field work costs: 500 euros per month, so a total of 9,000 euros.
? Mission of Sophoan Min and AVSF for support the implementation of work package 1 and 2 8,000 euros
? Field operations and missions GDA/DALRM and CIRAD 10,000 euros
? For implementation of work package 2 specifically: missions N Faysse 3,000 euros
? For implementation of work package 3 specifically: o Master thesis of a French student: 5,000 euros, missions N Faysse: 3,000 euros, mission PY Le Gal: 4,000 euros
Autres éléments nécessaires à la compréhension de ce chantier : Fukai, S., Ouk, M., 2012. Increased productivity of rainfed lowland rice cropping systems of the Mekong region. Crop Pasture Sci. 63, 944?973.
Le Gal P.-Y., Bernard J., Moulin C.-H., 2013. Supporting strategic thinking of smallholder dairy farmers using a whole farm simulation tool. Tropical Animal Health Production, 45(5):1119?1129.
Maxwell, T. W., Y. Songly, B. Ung, L. Peou, and J. Reid. 2012. The social and other impacts of a cattle/crop innovation in Cambodia. Agric. Syst. 107: 83-91.
Nampanya, S., S. Suon, L. Rast, and P. A. Windsor, 2012: Improvement in smallholder farmer knowledge of cattle production, health and biosecurity in southern Cambodia between 2008 and 2010. Transbound. Emerg. Dis. 59: 117-127.
Pen M., D. Savage, W. Stur, L. Sophal, and M. Seng. 2010. Cattle feeding and management practices of small-holder farmers in Kampong Cham province, Cambodia. International Journal of Environmental and Rural Development. 1: 132-138.
Ryschawy, J., Joannon, A., Choisis, J. P., Gibon, A., & Le Gal, P. Y. (2014). Participative assessment of innovative technical scenarios for enhancing sustainability of French mixed crop-livestock farms. Agricultural systems, 129, 1-8.