Issuu on Google+

Inventing a new form of agriculture


Inventing ecologically intensive farming systems to feed the world   

CIRAD’s expertise and know‐how    

More effective control of locust invasions  Coffee berry borer control  Regulating pests and diseases in tropical agrosystems  Controlling insect pests in cotton growing systems  Direct seeding mulch‐based cropping  Ecological intensification of banana growing  Improving agroforestry systems in the humid tropics  Agri‐environmental impacts of the oil palm  Sustainable production for small scale farmers in developing countries  Preserving biodiversity in African savannas  Sustainable charcoal production in the Democratic Republic of Congo  Dissemination of direct seeding mulch‐based cropping systems in Madagascar  Optimizing biomass production whilst minimizing environmental impact  Integrating crop and animal productions  Pig production in tropical and subtropical regions  Reconciling fodder production and environmental protection in the humid tropics  Bibliographie/Literature 

More effective control of locust invasions Locust studies at CIRAD

Improving survey and control strategies

Forskål erca gregar ia st, Schistoc cu lo CIRAD rt d, se ar de Mon The ure). © A. phase, mat (gregarious

Locust swarms have many causes. Combinations of favourable ecological conditions or changes in farming practices may boost the insects’ capacity to cause destruction.

Over the past thirty years, CIRAD has been conducting field research in many countries worldwide to better understand the origin of locust invasions through improved knowledge of the zones at risk and early detection of the conditions likely to trigger those invasions. It has also been studying the vulnerability of the human populations affected by crises, working to ensure that control methods are more environmentally friendly, and promoting credible alternatives to chemical insecticides. The sustainability of the solutions proposed is a constant concern.

Contact Michel Lecoq CIRAD Locust Ecology and Control Research Unit Campus international de Baillarguet 34398 Montpellier Cedex 5 France

© M. Lecoq, CIRAD


n many tropical regions of the globe and especially in Africa, locust invasions are a curse and a very serious threat to agriculture, animal production and the food security of rural populations. Not a single crop is spared. The desert locust, in particular, is a major pest, whose spectacular invasions can cover areas of more than 29 million km2. The material, human and environmental damage caused is considerable. Controlling these insects is thus a major issue for local populations and the environment. It is a national priority for many developing countries.

This work is backed up by in-depth knowledge of the realities of developing countries, the operational constraints, and the bioecology of locusts. CIRAD’s research is supplemented by training and extension operations and appraisal missions in the event of locust swarms.

Monthly locust risk map in Madagascar.

A geographical information system for migratory locusts CIRAD has developed a geographical information system for managing migratory locusts in Madagascar, one of the major risks for agriculture in the country, which saw a catastrophic invasion from 1997 to 1999. The software can be used as a decision support tool in order The migratory locust, Locusta migratoria Linnaeus to pinpoint the zones at high risk of locust (solitary phase). © M. Lecoq, CIRAD swarms and organize surveillance and early intervention operations more rationally. This is an essential stage in establishing a more effective, preventive and sustainable strategy for managing the locust risk in Madagascar.

Risk level

Very low Low Worrying Threatening Serious Critical

Understanding the origin of desert locust invasions In collaboration with the FAO and national anti-locust centres, CIRAD is using molecular biology techniques in Africa to gain a clearer understanding of locust population dynamics prior to invasions. Using remote sensing and The desert locust, Schistocerca gregaria Forskål (solitary phase). geographical information systems, the conditions that favour © A. Foucart, CIRAD locust reproduction and exponential growth in numbers can be identified at an earlier stage. Management of survey and control services in the countries involved (Mauritania, Mali, Senegal, Niger, Chad, Burkina Faso, Morocco, Algeria, Tunisia and Libya) has been improved thanks to a specific database accessible in real time via the Internet.


CIRAD offers training courses in locust control, organized on request, in France and abroad: • locust expertise: controlling locust pests; • application techniques for locust control and crop protection. Technical advice notes on the main locust pest species are available, and an on-line locust encyclopaedia can Locusts on sale at the market in Niamey, Niger. © M. Lecoq, CIRAD be consulted at CIRAD also has a documentation centre specializing in locust pests, with more than 10 000 publications and several thousand images.

• FAO, Food and Agriculture Organization of the United Nations, Italy • FAO Commission for Controlling the Desert Locust in the Western Region, Algiers • Locust control centres in West Africa and the Maghreb • AGRHYMET Regional Centre, Niger • National Anti-Locust Centre, Madagascar

© CIRAD, March 2010.

Training and documentation offers

Coffee Berry Borer Control Triple-action IPM

he coffee berry borer, Hypothenemus hampei Ferrari, is the main insect pest on coffee, attacking the fruits. It is a beetle of African origin that affects plantations in virtually all coffee producing countries worldwide. Despite the introduction and repeated releases of natural enemies, this pest adapts and develops wherever it settles. With global warming, the coffee berry borer (CBB) is now colonizing zones located at increasingly high elevations, thereby threatening production of the major high-grown coffees. Triple action IPM provides a solution to the CBB problem whilst protecting the environment and biodiversity.

An excellent response to CBB survival behaviour


IRAD and its coffee research partners have developed simple, effective and inexpensive IPM against CBB comprising three components: - strict branch stripping, d ra Ci r, ou - trapping of colonizing females, uf B. D sta Rica. © antation, Co pl e ffe co - meticulous plantation upkeep. Shaded These three activities are complementary and help to effectively control the pest. During migration, CBB tend to shelter in residual berries waiting to colonize the berries of the next crop. So, completely removing the residual berries remaining on branches after the harvest (also called branch stripping), prevents part of the pest's population from surviving. Female CBB emerging from residual berries are also caught in traps (BROCAP® trap) which are kept in plantations until all CBB have emerged. Lastly, plot upkeep, especially formation pruning, shade regulation and plot cleaning improve the results of the previous techniques.

Contact Bernard Dufour CIRAD Controlling Pests and Diseases in Tree Crops research unit Avenue Agropolis, 34398 Montpellier Cedex 5 France

© B. Dufour, CIRAD


Triple-action IPM is a selective, environment-friendly technique. It is compatible with biological control and curbs CBB before they can infest the harvest and cause damage.

Coffee tree pruning. © B. Dufour, Cirad

In shaded coffee plantations with tall varieties, CBB infestation can be reduced by more than 90% compared to control plots. Branch stripping and trapping account for more than 70% of the reduction.

The BROCAP® trap

Method application zones Triple-action IPM applies to geographical zones with just one annual harvest, i.e. in the tropical belt where the climate has two clearly marked dry and wet seasons. On the other hand, it is more difficult to apply in equatorial zones, where flowering and fruiting periods overlap. The method is more effective in shaded coffee plantations than unshaded plantations, as trapping is more effective under shade. This protection programme begins with the stripping of residual berries (beginning of February in Central America and the beginning of January on the Bolovens plateau in Laos) and finishes once the major CBB migratory flows have ended (end of June in Central America and Laos). The dates indicated mostly concern programmes devoted to arabica coffee growing.

Shade tree pruning. © B. Dufour, Cirad

Rehabilitation of the coffee plantation. © B. Dufour, Cirad

Partners • PROCAFE, El Salvador ( • PROMECAFE, Regional Cooperative Programme for Technological Development of Coffee Cultivation in Central America, Panama, the Dominican Republic and Jamaica ( • ECOM Agroindustrial Corporation Ltd, Mexico and Indonesia (

© Cirad, March 2010.

The BROCAP® trap works with an attractant. It captures CBB during their migration flight. This is the only component of IPM that calls for any real investment, amounting to around 3 dollars per trap. It is essential to install at least 18 traps per hectare, with two attractant dispensers per trap, to ensure that the system operates efficiently for four months per year. CIRAD Installation of BROCAP® CBB traps. © B. Dufour, Cirad and PROCAFE (Salvadoran Foundation for Coffee Research) developed the BROCAP® trap to meet the needs of coffee growing in Central America. Its use has now been extended to Asian countries. In addition to mass trapping, for which it was designed, the BROCAP® trap can be used as an agricultural early warning system. It is distributed in El Salvador by PROCAFE, and elsewhere in the world by ECOM Agroindustrial Corporation Ltd.

Regulating pests and diseases in tropical agrosystems Diversifying plant species in cropping systems

Specific plant diversification


wide diversity of plant species, or specific plant diversity (SPD), characterizes natural ecosystems, which suffer much less from pests and em. opping syst ch-based cr ul m diseases than cultivated ecosystems. Using a in sh Fodder radi CIRAD , ss da na at cultivated plant varieties that are resistant to © A. R pests and diseases, and defining optimum spatio-temporal deployment methods for such resistance, play a key role in crop protection. Over and above mere genetic diversification, SPD maximizes ecological pest and pathogen regulation processes, such as the preservation of natural enemies. In this context, CIRAD and its partners are implementing a project in the tropics to optimize the ecological mechanisms of pest and disease management, for sustainable improvement of agrosystem productivity (OMEGA3 project). It is analysing how pest and disease populations are affected by the introduction of spatial and temporal diversity of plant species in cropping systems. Several systems representing a range of pests and diseases and host-plants in tropical zones are being studied: coffee-based agroforestry systems in Costa Rica, cocoa-based agroforestry systems in Cameroon, upland rice-based direct seeding mulch-based cropping systems in Madagascar, gombo and tomato-based food/market garden crop systems in Niger, tomato-based market garden systems in Martinique and cucurbit based systems in Réunion, etc.

Contacts Alain Ratnadass CIRAD, HortSys research unit ICRISAT BP 12404 Niamey Niger Jacques Avelino CIRAD Controlling Pests and Diseases in Tree Crops research unit IICA/PROMECAFE Apartado postal 55 2200 Coronado San José Costa Rica For further information:

© A. Ratnadass, CIRAD


y cultivating varieties that make different demands on the soil and climate, diversifying species or varieties in agrosystems reduces the risks of harvest losses, especially in the context of climate change. Likewise, cereal and legume combinations or use of cover crops that reduce erosion bolster food security. However, little is known about the effects of such plant diversification on populations of pests and diseases, and the damage they cause. With a view to economic and environmental sustainability, it is therefore important to gain a clearer understanding of interactions and use them to minimize any negative impacts and limit synthetic pesticide use.

Pest and disease regulation processes Introducing specific plant diversity induces different pest and disease regulation processes (insects, pathogens or parasitic plants), which are analysed: • sanitizing effects of service plants as the previous crop cover in relation to tomato bacterial wilt in Martinique, • allelopathic effects of cover crops in relation to white grubs and Striga on upland rice in Madagascar, • the luring effects of decoy plants, combined with barrier effects and conservation biological control on the tomato fruitworm and on whiteflies of market garden crops in Martinique and Niger, • the same effects "assisted" by food attractants combined with biological insecticides on cucurbitattacking fruit flies in Réunion, • the effects of combining trees and bushes on mirid bug dynamics and the epidemiology of black pod rot on cocoa in Cameroon, • the effects of landscape fragmentation on the epidemiology of coffee leaf rust and on coffee berry borer dynamics in Costa Rica.

Unshaded coffee plantation in Papua New Guinea, devastated by leaf rust. © J. Avelino, CIRAD

Shaded coffee plantation in Costa Rica: shading reduces rust-related risks. © J. Avelino, CIRAD

Novel cropping systems

Construction of mechanistic models

Ideotypes of SPD-based cropping systems resilient to pests and diseases

Partners • ICRISAT, International Crops Research Institute for the Semi-Arid Tropics, India/Niger • CATIE, Centro Agronómico Tropical de Investigación y Enseñanza, Costa Rica • INRAN, Institut national de recherche agronomique du Niger • FOFIFA, Centre de recherche agronomique de Madagascar • IRAD, Institut de recherche agricole pour le développement, Cameroon • Universities in Cameroon, Costa Rica, Madagascar and Niger • INRA, Institut national de la recherche agronomique, France

© Cirad, March 2010.

An inventory is proposed of service plants that can be used for agro-ecological management of pests and diseases in horticultural systems (in Martinique) or in direct seeding mulch-based systems (in Madagascar). For instance, the benefits offered by the fodder radish, Raphanus sativus, have been discovered, be it for its antibacterial effects against Ralstonia solanacearum in tomato crops, or for its suppressive effect on white grubs in mulch-base rice crops. Some contradictory effects of shade have also been discovered on the incidence of leaf rust in coffee agroforestry systems: shade reduces disease incidence compared to full exposure to sunlight by reducing the fruit load, but increases it by creating humid conditions more conducive to infection and disease development. After formalizing the ecological processes studied, and Hypotheses regarding SPD effects on pests and diseases depending on the major types of pests and generated by observation Adding to the diseases, plants and plant diversification knowledge base methods, CIRAD defines indicators to Experimental checking construct predictive models of of suspected SPD effects infestation. The models are used to elaborate novel Parameterization of existing models cropping systems that are Indicators resilient to pests and diseases, based on the Validation of models introduction of SPD in Scenarios through observation and decisionand experiments agrosystems. making rules

Controlling insect pests in cotton growing systems A form of agriculture less dependent on pesticides ecause of their impact on fibre production and quality, insect pests significantly affect incomes among cotton growers. This constraint has previously been largely overcome using phytosanitary products, but the practice is now threatening the viability of production systems, due to the reduced efficacy of a number of insecticides following the advent of resistant insect populations and the disruption of various biological balances. Research has to take account of insect population dynamics and their interactions with the environment, above and beyond the level of the cultivated plot alone, so as to develop new, sustainable protection strategies that reconcile agricultural production and environmental protection.

Understanding the life system of insects In the African savannahs, two insects cause significant damage to cotton crops: the bollworm Helicoverpa armigera, which attacks the fruiting organs, and the aphid Aphis gossypii, which sucks boll. g a cotton llwor m eatin bo a er the sap and thus weakens young cotton plants and ig m ar Helicoverpa CIRAD lt, au év Br affects fibre quality by depositing honeydew when © T. the bolls open. CIRAD is working to characterize the mechanisms involved in the population dynamics of these insects so as to identify ways of keeping levels below damage thresholds. A study has been conducted in several African regions, on aphid populations. In theory, the populations are capable of colonizing more than 250 plant species, but in reality, they are genetically differentiated and specialize in groups of host plants. For instance, some populations are only found on cotton during the rainy season, and then on other Malvaceae such as okra or red sorrel in market garden plots during the dry season. Other populations prefer plants of the families Cucurbitaceae (melon, water melon, etc) or Solanaceae (pepper, aubergine, etc). Conversely, an analysis of bollworms taken from various plants revealed a lack of specialization. Opportunistic movements over short or long distances enable the populations to exploit temporary cultivated plant resources (maize, cotton, cowpea and tomato, for instance) or some weeds (Cleome spp. or Hyptis sp.).

Aphis gossypii aphids on a cotton leaf. © T. Brévault, CIRAD

Contact Thierry Brévault CIRAD Annual Cropping Systems Research Unit Avenue Agropolis 34398 Montpellier Cedex 5 France

© T. Brévault, CIRAD


Monitoring population movements New markers, such as the composition of the insects’ bacterial flora, are currently being developed with a view to monitoring population movements. Knowing the migration profiles of populations and the sequence of plants that serve as refuges or reproduction reservoirs depending on the season makes it possible to predict infestations. It also makes it possible to act on the insects’ survival phases using appropriate cropping practices: destruction of relay plants, early sowing, topping cotton plants to reduce their attractiveness, etc. It is also possible to alter the habitat so as to slow the spread of insects: spatial arrangement of the plot, crop associations or successions, plant barriers, etc. Such knowledge also serves to fuel models of the development of resistance to insecticides.

Helicoverpa armigera bollworm on tomato. © T. Brévault, CIRAD

Aphis gossypii aphids on a pepper plant. © T. Brévault, CIRAD

Restructuring agricultural landscapes By simplifying landscapes and using increasing amounts of pesticides, intensive agriculture results in a loss of plant and animal biodiversity in agrosystems, and consequently of the services rendered by that biodiversity, such as natural regulation of insect pests. It is vital to restructure the agricultural landscape using hedges, trees or buffer strips if we are to restore biodiversity.

Cotton plots in a savannah landscape. © P. Schwarz

In the savannahs of Africa, these plants may be intercropped with cotton plants or planted around the edges of plots, to stimulate the production of “killer” beneficial that act against bollworms or aphids. For instance, some plants provide an opportunity for aphid populations and their associated parasitoids to develop, with the prospect of seeing the parasitoids transfer to cotton plots. Other plants are being tested that could act as traps, by attracting female Helicoverpa moths looking for a place to lay their eggs.

Another study in Cameroon has shown that direct seeding of cotton on a legume- or grass-based plant cover favours the establishment of a more abundant, diverse soil macrofauna, particularly in terms of species that prey on insect pests.

Helicoverpa armigera bollworm on a weed. © T. Brévault, CIRAD

Partners • INRA, Institut national de la recherche agronomique, France • PRASAC, Pôle régional de recherche appliquée au développement des savanes d'Afrique centrale, Chad • IRAD, Institut de recherche agricole pour le développement, Cameroon • Société de développement du coton, Cameroon

© CIRAD, March 2010.

So-called “service” plants can act as traps for pests and refuges or banks for predators and parasitoids. They can also have a repellent effect on pests. It is vital to know about these service plants, and also the beneficial fauna and its impact on insect pest populations if we are to make optimum use of biodiversity.

Direct seeding mulch-based cropping An engineering tool for ecological intensification onventional cropping systems based on soil tillage, massive use of industrial inputs (fertilizers, pesticides and energy), and a small number of cultivated species can no longer satisfy food, health and environmental requirements. How can we continue to produce more so as to feed people, while protecting the environment? To ensure ecological intensification, CIRAD makes use of the way in which natural ecosystems, such as forests, in which biological and biochemical cycles are regulated naturally, work. It conducts research aimed at changing farming systems into veritable cultivated ecosystems. In particular, it works to develop ways of protecting and restoring the soil by combining direct seeding with permanent plant covers.

Direct seeding mulch-based cropping systems


irect seeding mulch-based cropping (DMC) systems are based on three principles: zero soil tillage, permanent soil cover that combines plant species rd, CIRAD © P. Lienha rice straw. on n intended to produce biomass and harvest residues, ea yb Growing so and the constitution of a large biodiversity of cultivated species grown in rotation, association and crop successions. These three principles combine to create a micro-environment for the crop, hence better expression of its potential to resist pests and diseases, and increased productivity (grain, pods, fibre, etc). Respecting these principles and studying how to apply and master them are the bases of an engineering method that can be applied to ecological intensification. The aim is to design model cropping systems suited to different socioeconomic and biophysical environments, based on more efficient use of natural resources such as solar energy, water, carbon and soil. CIRAD designs DMC systems as part of its work on agricultural development. It conducts research aimed at understanding the processes at play and building indicators for managing those processes.

Contacts André Chabanne Francis Forest CIRAD Direct Seeding and Cover Crops Research Unit Avenue Agropolis 34398 Montpellier Cedex 5 France To find out more:

© L. Séguy, CIARD


DMC: a tool for a new type of agriculture CIRAD’s fieldwork sites combine a wide range of biophysical and socioeconomic situations representative of tropical environments. The DMC systems developed in Brazil by CIRAD teams are now being used in Central Africa (Cameroon), North Africa (Tunisia), the Indian Ocean (Madagascar), Asia (Cambodia, Laos, Vietnam, Thailand and China) and the West Indies (Guadeloupe) to: • regenerate tropical soils degraded by inappropriate farming methods; • use natural areas for agriculture without compromising their production potential; • propose cropping systems that use as few chemical inputs as possible, to ensure safe products and avoid polluting water, soils and the air; • develop a range of alternative rice cropping systems by using DMC and rice varieties developed for DMC, with a high yield potential and that make optimum use of limited water resources; • detoxifying soils through bioremediation, etc. DMC systems thus enable the establishment of sustainable farming systems, thanks to more efficient use of natural resources and better integration of agriculture, animal production and perennial crops. If applied on a scale of several villages, they make a relevant contribution in terms of rational development of rural areas. Moreover, if adopted on a larger scale, they can go some way towards providing a response to global issues such as food security and the environment (management of shared natural resources, global warming, biodiversity, etc).

DMC systems and environmental services

Brachiaria ruzisiensis cover between rows of cassava. © F. Tivet, CIRAD

Harvesting Brachiaria ruzisiensis seeds by hand. © P. Grard, CIRAD

Partners • FOFIFA, National Centre of Applied Research and Rural Development, Madagascar • Direct Seeding Group, Madagascar • TAFA NGO, Madagascar • SODECOTON, Cameroon

For instance, the results obtained over the past ten years in Brazil, Madagascar and Laos show an overall trend towards increased carbon stocks in soils cultivated in this way. This sequestration of atmospheric CO2 can reach as much as 1.5 tonne per hectare per year for rationally fertilized crop rotation systems, producing some 16 tonnes of recyclable primary biomass (cereal and legume mulch and roots) per hectare per year. In France, in Touraine, such systems have resulted in sequestration levels measured in situ of 20 tonnes of carbon per hectare, obtained after 10 years of DMC. In the long term, extending DMC systems can thus provide considerable environmental services: erosion control, improved water quality, and a reduced greenhouse effect.

Numerous earthworm casts in a direct-seeded upland rice plot on a Stylosanthes plant cover. © F. Tivet, CIRAD

• EMBRAPA, Brazilian Agricultural Research Corporation • Ponta Grossa State University, Brazil • Ministry of Agriculture, Forestry and Fisheries, Cambodia • Yunnan Academy of Agricultural Sciences, China • National Agriculture and Forestry Research Institute, Laos • Kasetsart University, Thailand • Northern Mountainous Agriculture and Forestry Science Institute, Vietnam

© Cirad, March 2010.

Along with its partners, CIRAD works to analyse the ecosystemic functions of DMC systems, notably in relation to soil organic matter dynamics, soil biological activity and pest and disease management.

Ecological intensification of banana growing New, more sustainable cropping systems ith 16 million tonnes per year, dessert banana exports account for a major share of agricultural production in many tropical countries. Production is still largely reliant on intensive monocultures. Such production methods result in reduced soil fertility and increased soil parasite population levels, and have an adverse effect on the environment and sometimes on human health. Producers are faced with problems relating to the poor sustainability of their cropping systems. In several producing countries, there is growing awareness of environmental issues, with consumers becoming increasingly concerned about the quality of the products they buy and how they are produced; new, very strict regulations on phytosanitary product use are also now being applied, particularly on the European market. As a result, the development of innovative, sustainable cropping systems is now a major issue for every country in which bananas are grown.

Designing new cropping systems


o reduce adverse environmental impacts and respond to societal and regulatory changes, CIRAD has developed new banana-based disease, ka to ga Si k cropping systems. More sustainable systems ac sistant to bl CIRAD w variety re Domergue, R. © pe Test of a ne are now on offer. Such systems use cropping Guadelou practices not or less involving chemical inputs: using healthy planting material on healthy soils, integrated pest management methods, spatiotemporal organization of cropping systems, etc. The available knowledge on banana plant growth was inputted into a simulation model. This led to the development of the SIMBA model, which designs prototype cropping systems by simulating the agronomic and environmental performance of a range of cropping techniques. The tool can be used to study the effect of ecological intensification practices on agrosystem functioning and to build new cropping systems.

Contacts Marc Dorel CIRAD Banana Plantain and Pineapple Cropping Systems Research Unit Station de Neufchâteau 97130 Capesterre-Belle-Eau Guadeloupe François Cote CIRAD Banana Plantain and Pineapple Cropping Systems Research Unit Boulevard de la Lironde 34398 Montpellier Cedex 5 France

© R. Domergue, CIRAD


CIRAD tests these new systems in consultation with banana production chains, and particularly with producers in Guadeloupe and Martinique. For instance, tests are under way of using Neonotonia wightii as a cover crop in banana plantings in the West Indies. The cropping systems are assessed for their agronomic, economic and environmental performance.

Reducing pesticide use Traditionally, banana growing has often relied on using phytosanitary products, particularly against nematodes. To develop more sustainable cropping systems, CIRAD and its subsidiary Vitropic have worked with producers and nurserymen to develop new crop management sequences that use banana in vitro plantlets as planting material, on soils cleaned by means of fallow or appropriate crop rotations. The in vitro plantlets are healthy, and free of nematodes, insects and pathogenic fungi. Such crop management sequences avoid the need to make systematic use of nematicides. In the French West Indies, they have largely contributed to reducing phytosanitary product use on banana: - 60% in ten years.

Intercropping banana and Neonotonia wightii, a perennial legume. © J.M. Risède, CIRAD

For the future, CIRAD is looking into the influence of how cropping systems are organized, in both spatial and temporal terms (varietal mix, introduction of breaks such as hedges, etc) on banana pest and disease regulation.

Trapping root borers

Integrated control of nematodes Nematodes are one of the main components of the soil parasite complex. Cropping systems based on practising fallow (or crop rotations) and planting healthy material (obtained by in vitro culture) provide effective solutions to the problems caused by soil parasites; fallow serves to clean the soil, but to be most effective, it also has to protect the soil against erosion or weed development. A soil cover during fallow periods of plants that do not host nematodes can also serve as a mulch for banana plants and significantly improve the efficacy of the crop management sequence. Such measures free growers from the need to apply nematicides, and allow them to pursue a more environmentally friendly approach, notably by cutting their herbicide use.

Planting of banana in vitro plantlets, Martinique. © R. Domergue, CIRAD

Partners • UGPBan, Union des Groupements de Producteurs de Bananes de Guadeloupe et de Martinique • Vitropic, France

© Cirad, March 2010.

Root borer trap in a banana planting, Martinique. © R. Domergue, CIRAD

Rational methods can also be used to reduce the numbers of Cosmopolites sordidus, the banana root borer, in plantations. Trapping is one method. The traps can be made considerably more attractive by impregnating them with a synthetic pheromone, sordidin. The efficacy of the method can be improved still further by placing entomopathogenic nematodes of the genus Steneirnema in the traps.

Improving agroforestry systems in the humid tropics The example of cocoa and coffee trees groforestry systems (AFS) in the humid tropics combine forest trees with tree crops (coffee, cocoa, kola, rubber, etc.), or with food crops or livestock. Some such AFS are derived from natural forests in which part of the original vegetation has been replaced by productive trees or crops, and others from the replanting of trees after slashing and burning the forest for food crops. After a few years, these farming methods culminate in a complex multiple cropping system. They are primarily managed according to the cash crops present in the system, which usually account for a large share of farmers' incomes.

© D. Snoeck, CIRAD


In the current context of declining land availability, rural population pressure, the food crisis, the limits reached by conventional agricultural intensification and climate change, agroforestry offers some worthwhile prospects. Improving the management of these systems, and ensuring their environmental, technical and social sustainability are a major challenge for research and development.

The functioning of cocoa and coffee-based AFS


offee and cocoa-based AFS are a traditional type of production, whose management and functioning resemble those of a forest. Such systems produce less coffee or cocoa than a monoculture, but in return: • their management requires less labour and fewer pesticides in Ecuador. l" cocoa tree A "Naciona and chemical fertilizers, re, CIRAD © M. Dulci • farmers derive other products from them, for consumption by their own household or for marketing: various fruits, kola, palm oil and wine, medicinal products, essential oils, fodder, timber, handicraft products, packaging, etc., • they offer a range of environmental services: biodiversity conservation, soil fertility, carbon storage, etc., • they provide social and cultural goods: family, national and international heritage, landscape aesthetics, shrines, etc.

Contacts Didier Snoeck CIRAD Performance of tree crop-based systems research unit Avenue Agropolis 34398 Montpellier Cedex 5 France Michel Dulcire CIRAD Innovation joint research unit 73 rue J.-F. Breton 34398 Montpellier Cedex 5 France

Cocoa and coffee trees are two species originating from the undergrowth, and the shade of other associated species is therefore naturally beneficial to them, as is the organic matter they also provide. But shade can also have detrimental effects, such as creating an environment conducive to diseases. For example, shade reduces the incidence of insects in cocoa agroforests, but it is propitious to pod rot. In coffee agroforests, shade prolongs the fruit ripening period for better coffee quality, but it reduces yields.

Improving AFS sustainability CIRAD is undertaking research to improve the efficiency of these complex cropping systems. Understanding and assisting AFS development first means analysing local know-how and the strategies and practices of stakeholders: producers and their organizations, processors, technical advisors, middlemen and manufacturers, public decision-makers.

Thus, various AFS studies are under way: • local know-how, strategies and practices of stakeholders, • innovation processes on cultural practices and methods of intercropping different plants to reduce parasite pressure, and on supply chain trends, • their impact on the landscape. Given the complexity of interactions between different intercropped species, CIRAD is developing environment-friendly agroforestry intercropping models to stabilize or even increase farmers' incomes.

Research in the face of change CIRAD is analysing the contribution made by agroforestry to the viability of household activity systems, faced with the factors of change (economic, climatic, environmental), in sub-Saharan Africa and Madagascar. A comparative analysis of the different local Coffee-Erythrina intercropping, Costa Rica. © P. Vaast, CIRAD backgrounds enables researchers to: • measure AFS impact on family economics, land heritage and the environment, • assess the flexibility of family activity systems in the face of change: crop diversification, biodiversity management methods, implementation of environmental services, • draw up technical and economic responses with producers in the face of regional and international developments, • challenge the development models promoted by public policies.

Coffee trees, Côte-sous-le-Vent, Guadeloupe. © M. Dulcire, CIRAD

Main partners • Grand-Sud Cameroun Research Platform in Partnership • CATIE, Centro Agronómico Tropical de Investigación y Enseñanza, Costa Rica • IRAG, Institut de recherche agronomique de Guinée, Guinea • ICRAF, International Centre for Research in Agroforestry, Kenya • KEFRI, Kenya Forestry Research Institute, Kenya • University of Antananarivo, Madagascar • University of Makerere, Uganda • University of Legon, Ghana

© Cirad, March 2010.

Research is also interested in the ability of smallholders to innovate and develop their professions. CIRAD is therefore working with them to sustainably improve the standard of living of these populations.

Agri-environmental impacts of the oil palm Indicators for sustainable production n increasing number of non-governmental organizations are blaming current oil palm development systems, accusing them of being responsible for the degradation of natural resources and causing environmental problems. In 2003, this led to the founding of a roundtable for sustainable palm oil production, bringing together the different stakeholders in the supply chain, and CIRAD. The initiative is based on defining principles and criteria for sustainable production and on using a good practices guide. If the initiative is to be effective, it needs to be accompanied by precise qualitative and quantitative indicators.

Assessing plantation sustainability


mplementing these criteria means establishing a normative and transparent evaluation system based on a sound scientific footing, with a view to measuring, assessing and analysing how agricultural practices affect the environment, providing information on the status of each A. Labeyr ie antation. © pl lm pa situation and monitoring the progress achieved. l oi Mature To that end, CIRAD is developing with its partners a set of agri-environmental indicators, known as IPALM. The approach adopted is based on the INDIGO® method developed by INRA in Colmar for temperate crops. It involves a system that cross references agricultural practices with components in the agro-ecosystem that might be affected, such as surface water or groundwater quality, air quality, soil fertility, or even biodiversity



Contacts Jean-Pierre Caliman CIRAD Performance of tree cropbased systems research unit c/o P.T. SMART P.O. Box 1348 28000 Pekanbaru, Riau Indonesia Aude Verwilghen CIRAD Performance of tree cropbased systems research unit Avenue Agropolis 34398 Montpellier Cedex 5 France

Assessment, decision-support and communication tools These agri-environmental indicators are a tool for assessing pollution risks, but also for estimating the effectiveness of the fertilizers applied. Farmers who adopt these tools are thereby demonstrating their involvement in environmental conservation. A scoring system has been developed based on scientific knowledge and a field assessment. It is on a scale of 0 to 10. The optimum "risk-free" situation for the environment is awarded a score of 10. A score of 7 to 10 lies in the "acceptable" zone, but still remains improvable. Any score under 7 indicates an excessive ecological risk requiring a specific action plan. Results of nitrogen indicator (I ) calculation, and recommendations. N

Nitrogen flow study: soil solution sampling system for NO3 leaching analysis. © J.P. Caliman

IN: an indicator for nitrogen IN, the first indicator developed, assesses the efficiency of nitrogen management in oil palm plantations, especially applications of nitrogen fertilizers, which are usually both a key production factor, a major cost, and a major environmental risk. It can be used to estimate nitrogen losses in the form of ammonia by volatilization, nitrates by leaching and nitrogen protoxide by gas emission. It is therefore organized in three modules INH 3 , INO 3, IN 2O, relative to those compartments. Depending on whether the aim is to analyse environmental impact, or establish a diagnosis with a view to making practices more efficient, just one of these subindicators might be considered, or all three. IN is based on a complete nitrogen flow balance in relation to oil palm requirements and has to be updated each year for each plot. It can be applied to a plantation by using a mean of the plot values weighted by the areas.

Study of nitrogen flow balances (here the root system). © J.P. Caliman

IPhy: an indicator for pesticides

For a broader partnership In addition to these two indicators, Iom, an indicator for organic matter and Icov, an indicator for soil cover, have also been developed. These four indicators are to be validated via a network of partners familiarized with this type of approach. Development of a software package to calculate the indicators on oil palm, IPALM, will facilitate its adoption by users. Future developments will focus on assessing how practices impact on biodiversity and water quality.

Partners • University of Nancy, France • INRA, Institut national de la recherche agronomique, Environment and Agronomy Centre, Nancy-Colmar, France • PT Smart Tbk, Indonesia

© Cirad, March 2010.

Pesticide use is of major concern to consumers. IPhy is a qualitative risk indicator based on decision trees. Fuzzy logic is used to aggregate the different factors identified as determinants in the process being considered, such as leaching, runoff and volatilization of pesticides. It also takes into account some properties of the molecules, their risks for human and animal health, and what happens to them in the environment (half-life, soil infiltration, etc.). The indicator comprises four modules, three on the risks associated with phytosanitary practices for the environmental compartments—surface water, groundwater, air—and the fourth on the risk associated with the rate applied.

Sustainable production for small scale farmers in developing countries Designing innovative cropping systems ropical environments generally have fragile soils and aggressive climates. Predictive climate change models agree that the instability of climatic conditions is likely to increase, with more frequent droughts or catastrophic flooding. The poorest producers find it difficult to access credit and markets, which, moreover, do not provide them with a sufficient return. In view of this, CIRAD is working to develop innovative systems that protect and make optimum use of the natural resources available in the short and long term, which stabilize and maintain if not improve productivity and limit the environmental impacts of agricultural activity.

The agronomic processes and ecological services used


he proposed cropping systems centre on direct seeding mulch-based cropping cations for ifi ec sp of (DMC). They aim to minimize the natural t en y establishm Participator s, Brazil. em st sy C physical, chemical and biological new DM innovative , CIRAD dares Xavier © J. H. Vala degradation of soils as a result of their cultivation, and are based on diversifying the species cultivated in rotation, succession or even association (intercropping). Introducing cover crops into these systems provides a range of services: • nutritive substrate for the soil fauna and the crops grown; • increased primary biomass production due to the solar energy intercepted between two crop cycles and at the start of the main crop cycle (cereals, soybean, cotton, etc); • recycling of nutrients, which permeate the deeper soil horizons thanks to the plants’ dense, deep root systems; • water regulation, linked to the total, permanent protection of the soil, which reduces runoff; • control of diseases, insect pests and weeds, whose habitat is modified; • supplies of food and animal fodder.

Contacts Eric Scopel CIRAD Annual Cropping Systems Research Unit FOFIFA, BP 1444 Antananarivo Madagascar François Affholder CIRAD Annual Cropping Systems Research Unit SupAgro 2 place Viala Montpellier Cedex 1 France

© E. Scopel, CIRAD


On a farm scale However, implementing DMC with a view to a given service means for farmers striking a delicate balance between various ecological processes. This requires from them greater knowledge of the impact of cropping techniques on that balance between processes, so as to manage them better and thus achieve the relevant agricultural and ecological objectives. Developing the use of live covers requires specific knowledge tailored to the local situation in terms of the environmental conditions and the stakeholders involved in agricultural production. Moreover, this type of complex innovation means making substantial changes to how resources are used on farms, to how operations are organized, and can result in the diversification of the products generated and the sources of income. Such changes are not always acceptable to some producers. In the case of Vietnam, for instance, farmers’ motivation to practice DMC rather than traditional upland rice or maize production systems is determined by their ability to cope with the additional cost of the technique, particularly in terms of labour.

DMC systems involve cover crops grown in succession, relay or association with the main crops so as to make optimum use of the resources available in time and space. © E. Scopel, CIRAD

New systems for and with producers

CIRAD is working with producers in several tropical regions to build new cropping systems: • in central Brazil, for the small farms resulting from the agrarian reform in the Cerrados; • in the hills of northern Vietnam, for small mountain farms, following the ban on slash-and-burn; • in Zimbabwe and Mozambique, with cereal and cotton smallholders in the local savannahs; • in Madagascar, with rainfed rice smallholders in mid- and high-altitude zones; • in Mali, Niger and Guinea, with smallholders in semi-arid zones.

Monitoring plant growth in a maize-based DMC system, Vietnam. © F. Affholder, CIRAD

Partners • FOFIFA, National Centre of Applied Research and Rural Development, Madagascar • EMBRAPA Cerrados, Brazilian Agricultural Research Corporation for the Cerrados region • VASI, Vietnam Agricultural Science Institute • IIAM, Mozambique National Institute of Agronomic Research • Montpellier SupAgro, France • INRA, Institut national de la recherche agronomique, France • IRD, Institut de recherche pour le développement, France

© CIRAD, March 2010.

The evaluation and conception process around DMC systems thus largely depends on the points of view of the players locally involved in rural development, and particularly those of the different types of farmers who are prompted, by choice or necessity, to show an interest in this type of cropping system. Because of these considerations, CIRAD has chosen participatory methods to develop innovative cropping systems in conjunction with its partners. This type of approach helps considerably to familiarize producers with new technical proposals and facilitates the cross-learning required for efficient management of such systems.

Preserving biodiversity in African savannas Towards widespread adoption of conservation agriculture around protected areas

Contacts Reconciling production and conservation It is a matter of choosing between lowinput environment-friendly, but extensive agriculture, and intensive farming concentrated on small surfaces and sparing land for conservation. For most species of interest to conservation, the second solution is increasingly recognized as a more desirable solution. However, intensive farming should not pollute downstream habitat and reduce their ability to support biodiversity. Therefore, CIRAD in Zimbabwe promotes conservation agriculture as a strategy to reconcile production and conservation.

babwe. nment, Zim tural enviro g on the na in er rd bo Cotton plot n, CIRAD © F. Baudro

The purpose of these techniques is to reduce the “leakiness” of cropping systems, minimizing negative environmental impacts (water runoff, erosion-related sediment and pollutants) and retaining the production capacity (water, topsoil, nutrients, organic matter) in situ.

Frédéric Baudron CIRAD Annual Cropping Systems research unit French Embassy PO Box 1378 Harare Zimbabwe Marc Corbeels CIRAD Annual Cropping Systems research unit TSBF-CIAT PO Box MP228, Mazowe Road Harare Zimbabwe

© F. Baudron, CIRAD


he Mid-Zambezi Valley is a remarkably preserved ecosystem hosting major populations of large mammals, such as elephants, buffaloes, lions, leopards, etc. Many initiatives aim at preserving this wildlife and increasing the benefits that local populations derive from it (safari hunting, ecotourism). Yet, its habitat has been considerably reduced over the last two decades, with the expansion of agriculture and cotton production in particular. Agricultural expansion not only leads to a drastic decline of biodiversity in farmland, but also to habitat fragmentation and increased isolation of habitat patches. How can agriculture and wildlife conservation be reconciled?


"Self-cleaning" cropping systems Several research studies demonstrate that the maintenance of a surface mulch of crop residues effectively controls horizontal nutrient losses and runoff. Crop residues also act as an amendment. In fine-textured soils, such residues may increase the stock of organic matter. Intercropping or rotation with deep-rooted secondary crops reduces vertical losses of mobile nutrients such as nitrogen by recycling them from the deep horizons to the surface. These secondary crops also significantly increase biomass production and may even fix atmospheric nitrogen in the case of legumes such as pigeonpea (Cajanus cajan (L.) Millsp.), jackbean (Canavalia ensiformis (L.)) or velvet bean (Mucuna pruriens (L.) DC). Systems using mulches and “multipurpose” legumes tend to increase soil biodiversity and soil biological activity. For instance, pesticides are not only retained within these systems, trapped in soil organic matter, but also degraded through the activity of large-spectrum extracellular enzymes secreted by plant roots and by soil micro-organisms, a process know as “bioremediation”. The "self cleaning" capacity of such systems is being investigated.

Sorghum with Cajanus cajan at different periods of the cropping cycle (January, May). © F. Baudron, CIRAD May

Promoting conservation agriculture

Evaluating the performances of such systems requires large-scale trials. Under the PARSEL project (Public-Private-Community Partnerships to improve food security and livelihoods in South East Lowveld and Mid Zambezi Valley) funded by the Eurpean Union, 300 hectares have been cultivated under cotton, sorghum and pigeonpea using techniques of conservation agriculture. These achieved yields exceeding those achieved under conventional cropping. The challenge now is to stimulate the joint interest of farmers and private operators around conservation agriculture. For this, CIRAD and its partners are working on the establishment of an “eco- label” that would provide access to a high value textile market.

Partners • Alliance Ginneries, Zimbabwe • Mbire Rural District Council, Zimbabwe • University of Zimbabwe • European Union Commission

© Cirad, March 2010.

Cotton plants sown directly through residues from the previous crop, Zimbabwe. © F. Baudron, CIRAD

The most promising system is based on a rotation between cotton and sorghum, the main food crop in the Mid-Zambezi Valley, intercropped with pigeonpea. The residues of sorghum and pigeonpea are retained as mulch through which cotton is directly sown, without any tillage. Pigeonpea is the multipurpose grain legume the most appreciated by the farmers of the Mid Zambezi Valley: it produces nutritionally rich grains of good commercial value, useful fodder, nitrogen-rich litter and its woody stem can be used as fuel.

Sustainable charcoal production in the Democratic Republic of Congo Improved tree fallow inshasa, the capital of the Democratic Republic of Congo, has a population of 8 million inhabitants and consumes up to 6 million tonnes of bio-energy equivalent per year. The city is surrounded by grasslands and patches of forest. The bio-energy used by the urban households consists mainly of fuelwood (charcoal and firewood). Charcoal needs, but also most of the staple starchy foods in the diet (cassava and maize) are provided by slash-and-burn shifting cultivation and by carbonization of the patches of forest and tree savannahs, which continue to deteriorate. Production obtained from these tree stands is becoming scarce and expensive. Soil fertility is declining, crop yields after fallow are decreasing, springs are drying up and fires are increasingly frequent. How can these populations continue to be supplied whilst limiting the environmental impact on forests?

Improving tree fallow Slash-and-burn cultivation gives rise to tree fallow after one to three years of cropping, due to the exhaustion of soil reserves. Improving tree fallow consists in planting tree legumes, whose roots r fo ed ar combined with microorganisms fix atmospheric eau being cle e Bateke plat D e slopes of th Peltier, CIRA R. © nitrogen. Organic matter and nitrogen storage in g. Forest on th in farm uction and charcoal prod the soil is thereby accelerated. This is especially true for acacias, trees that are also known for their large biomass/wood production. The trees can already be planted during the cropping period and continue to grow rapidly after harvesting during the fallow phase. Since the 1990s, CIRAD has bred more specific tree species associated with symbiotic bacteria (rhizobium) that display strong growth and nitrogen fixation, particularly in Ivory Coast and the Republic of Congo. Since 2009, CIRAD has been implementing the EU-funded "Makala" (research development project about the fuelwood sector), and intends to disseminate these improved tree fallow techniques and provide sustainable management techniques for the last remaining patches of forest around Kinshasa.

Contacts Jean-Noël Marien Régis Peltier CIRAD Forest Resources research unit Campus international de Baillarguet 34398 Montpellier Cedex 5 France

© R. Peltier, CIRAD


The Mampu tree fallow The Mampu plantation, about 140 km east of Kinshasa, was originally designed as the pilot phase in a vast reforestation project covering 100,000 hectares of sandy soils on the Bateke plateau. Between 1987 and 1993, 8,000 hectares of Acacia auriculiformis were planted. From 1994 onwards, the Mampu plantation was Reforestation of degraded grassland, mainly divided into plots of 25 hectares with Acacia auriculiformis. allocated to 320 farming families. © R. Peltier, CIRAD Farmers were required to manage their new tree plantation following a novel agroforestry technique that combines food crops with acacia. Two or three years after planting the trees, once agricultural products have been harvested, the acacias reach a height of three metres. After around ten years, a veritable acacia forest, mixed with a few local species, becomes established. Farmers can then exploit it, process the wood into charcoal and sell it in town. In the unharmed humus, they can replant a new crop cycle. Every 4 metres, a one metre wide strip of soil is left unfarmed, so that acacia seeds can germinate and reconstitute the future forest stand.

Increasingly efficient charcoal makers. © R. Peltier, CIRAD

Extension to the Bateke plateau grasslands

The Mampu agroforestry model is to be extended to the villages located in the Bateke plateau grasslands. Special attention is paid to the role of traditional land rights and the possible diversification of other products and local processing techniques. Overall, this should increasingly contribute to meeting urban renewable energy needs, whilst creating rural jobs. Moreover, other agroforestry systems, under other ecological and socio-economic conditions, are worth testing; such as managing the natural regrowth of local multipurpose species (for fruits, wood, shelter for game, nitrogen fixation, etc.). Indeed, on more clayey land once occupied by forest, there is a great variety of tree species in the natural regrowth. Those trees cannot develop due to continual felling and uncontrolled fires. If the plot is covered by a thicket, a farmer can first protect it with a fire-break, then select 100 to 400 sprigs per hectare of useful species out of the thousands of shoots. After 8-10 years of protection, the plot can then be used to harvest fuelwood and to plant crops, whilst maintaining a few large trees for seed production (10 to 100 per hectare) for the next production cycle.

Partners • CIFOR, Center for International Forestry Research, Cameroon • Hanns Seidel Foundation, Germany and Democratic Republic of Congo (DRC) • Gembloux Agro-Bio Tech University, Belgium • Kisantu botanical garden, DRC • Research Unit on Commercial Forest Productivity, Republic of Congo • University of Kisangani, Ecole régionale post universitaire d’aménagement et de gestion Intégrée des forêts et territoires (ERAIFT), DRC • National Reafforestation Services, DRC and Republic of Congo

© Cirad, March 2010.

Total charcoal production from the plantation currently varies from 8,000 to 12,000 tonnes per year (t/year). In addition, the farmers produce 10,000 t/year of cassava, 1,200 t/year of maize and 6 t/year of honey. Reforestation of the Mampu stand is considered a success.

Cassava harvest after slash-and-burn cultivation in an acacia plantation, and processing into chips. © R. Peltier, CIRAD

Dissemination of direct seeding mulch-based cropping systems in Madagascar

n the mid-altitude zones of Madagascar, cropping systems based on direct seeding, with a cover crop and crop rotation, have been disseminated on smallholdings since the turn of the century with a degree of success. In order to disseminate these new cropping patterns, CIRAD and its development partners in Madagascar have developed modelling tools to monitor and assess activities through a DSS (Decision Support System). For developers, these tools provide decision-support in the technological choices to be implemented depending on their physical environment and their type of farm.


Optimizing extension efforts


n agricultural development projects, decisionsupport and negotiation between operators t, CIRAD es. Š E. Peno tic ac and with farmers is a priority, so that actions pr of t d assessmen Field visit an live on after the end of the project. CIRAD is endeavouring to optimize extension efforts by proposing techniques and services that are truly adapted to each type of farmer. This type of initiative is being implemented as part of development projects in the regions of Lake Alaotra (BV-lac project, Lake Alaotra watersheds), Vakinankaratra and southeastern coast (BVPI-SE/HP project). The aim is to adapt technical and organizational messages to farmer realities and promote innovation processes including direct seeding mulch based cropping systems (DMC) for sustainable production as well as the integration of agriculture and animal production. A selfappraisal method for farmers’groups and a network of reference farms have been developed. These tools can also be used to assess technical actions and provide support in defining aspects of public agricultural development policy.

Contact Eric Penot CIRAD Innovation joint research unit Ampandrianomby, BP 853, 99 Antananarivo Madagascar

Š E. Penot, CIRAD

Developing a learning approach

Identifying innovation processes CIRAD proposes self-appraisal sessions where farmers in producer organizations themselves identify innovation processes adapted to them, using the "Accelerated Propagation of Innovation" (API) method (Belloncles). The method requires prior coaching of the participants so that they can give thought to a situation then act appropriately. This prior coaching is provided by socio-organization specialists. At Lake Alaotra, CIRAD used the API method with associations of irrigation water users, the federation of user associations in the network of the two irrigated areas: "PC15" and "Marianina Valley", as wall as with agricultural intensification groups and farmer groups integrating DMC practices. The transmission of technical information within the farmer groups applying DMC was a frank success. The analysis identified how DMC techniques are effectively adopted and revealed a potential will to increase intensification from the 4th or 5th year of DMC.

Cowpea mulch in a DMC system with a maize-cowpea-rice rotation, Madagascar. © E. Penot, CIRAD

The development project partners thereby acquired experience in organizing and running these sessions. The method has been formalized in the form of a BV-lac working document available from CIRAD.

Developing a network of reference farms

Olympe software is a tool developed by CIRAD, INRA and IAMM (Mediterranean Agricultural Institute in Montpellier) to simulate farm activities. It can be used to test the robustness of a technical choice, and farm’s resilience when faced with a series of hazards. Simulations of the adoption of new techniques are carried out with standard crop management patterns that provide reliable data over a large number of plots through prospective analysis. Applying this approach to the adoption of direct-seeding mulch-based cropping systems at Lake Alaotra helped development operators to make progress in their work. Consequently, the technical possibilities offered to famers have become more adapted to the constraints faced by different types of farms. In particular, the levels of cropping system intensification proposed are more adapted to risk levels acceptable to producers.

On-farm reporting-back session, Madagascar. © E. Penot, CIRAD

Partners • FOFIFA, Centre de recherche agronomique de Madagascar • University of Antananarivo, Madagascar • Development partners associated with the BV-Lac and BVPI-SE/HP projects, Madagascar • Groupement semis direct de Madagascar

© Cirad, March 2010.

New cropping systems are assessed in networks of reference farms. A network of reference farms is a set of farms representative of different agricultural and socio-economic situations. The farms are monitored annually, to measure the impact of technical actions and development policies and carry out prospective analyses.

Optimizing biomass production whilst minimizing environmental impact The contributions of virtual agronomy


© H. Rey, CIRAD

IRAD's AMAP joint research unit has developed software packages to simulate the architectural development of plants and their yields. By arranging plants in stands, from a plot to a landscape scale, it becomes possible to conduct virtual experiments to assess and optimize the effect of cultural practices or environmental conditions on growth. This ecoinformatics approach can also help to answer questions about the environmental impact of agricultural production.

Simulating plant growth By modelling the architecture and development of individual plants, it is possible to see how ware. ft so E SL elementary growth processes evolve over time Hérault with landscape in of a vineyard n io at within the plant and how they are affected by ul m Si , CIRAD © S. Griffon environmental conditions. Growth models have been incorporated into simulation softwares (AMAPsim, Digiplante) which can be used to simulate plant structures under agronomically and environmentally variable natural conditions. This approach has been applied on various plant models (sunflower, maize, oil palm, coffee, eucalyptus, pine, etc.). By reconstructing plant stands, such as a crop field, it is possible to simulate, analyse and optimize plant production under variable agronomic and environmental conditions for different applications that take plant architecture into account. For example, the quantity and quality of light captured by plants can be simulated using their three-dimensional description (Archimed software). When these models are adapted to a landscape scale, they can be used to simulate growth variability in relation to local sunlight and rainfall conditions, and water distribution in the soil, etc (GreenLab paysage software).

Contact Thierry Fourcaud CIRAD AMAP joint research unit Boulevard de la Lironde 34398 Montpellier Cedex 5 France

The development of algorithms to display plants in landscapes (SLE software) is finding new applications in the graphic animation fields (virtual reality, films, video games, etc.).

Improving light capture in agroforestry By combining the AMAPsim and Archimed softwares, mapping of the light available for intercrops (maize, cocoa, soybean, etc.) has been simulated for teak or Acacia mangium-based agroforestry systems in Indonesia, or coconut-based systems in Vanuatu. The studies showed how the amount of available light, and its spatial distribution, changed depending on tree development. It thus became possible to plan possible crops according to their shade tolerance.

Simulated mapping of light transmission beneath six-year-old coconut palms under the Archimed platform. © J. Dauzat, CIRAD

Simulation experiments are also carried out to assess the possibilities of optimizing these systems by modifying tree density and planting layout, or by pruning lower branches when trees develop too much. For example, it has been possible to show how planting trees in North-South rows improves light distribution in the plot, or how pruning the lower section of a three-year-old Acacia mangium crown quadruples the amount of light available for intercrops.

Defining ideotypes of sunflower varieties

Partners • Chinese Academy of Sciences – Institute of Automation (CASIA) • Laboratory in Computer Science, Automation and Applied Mathematics (LIAMA), China • China Agricultural University (CAU), Key laboratory of plantsoil interactions, China • Institut national de la recherche en informatique et automatique (INRIA), DIGIPLANTE project team, France • Laboratoire d’écophysiologie des plantes sous stress environnementaux (LEPSE, UMR SupAgro-INRA), France

© Cirad, March 2010.

The interaction between development rate and organ growth in sunflowers is considered as a function of the temperature and light conditions encountered by the crop during its growth, and of the planting design adopted. Virtual experiments lead to optimized planting designs (density, Sunflower plants (Heliasol variety) in the field. spacing) for given ecotypes and © H. Rey, CIRAD geographical situations. It is also possible to define periods in the crop management pattern when the crop is more or less susceptible to water deficits, and thereby optimize irrigation. These studies have helped to define ideotypes of sunflower varieties, taking into account both architectural characteristics (e.g. for better light capture) and growth characteristics of the plant (for better use of water and temperature conditions) so as to Sunflower plants simulated with AMAPsim software. be more efficient under fluctuating © H. Rey, CIRAD environmental conditions. This agrophysiology approach can be used to test the performance of sunflower varieties under new environmental conditions and adapt cultural techniques. Other similar studies have been conducted on cotton, maize, or tomato, using the GreenLab model.

Integrating crop and animal productions

ombining crop and animal productions was first promoted in sub-Saharan Africa and Madagascar in the 1960s. It led to massive adoption of animal traction in regions where the rice, cotton and groundnut sectors were sufficiently organized to provide farmers with credit and training. Today, the growing rural population and pressure on arable land call for a re-think in the types of combinations between crop and animal productions, in order to cope with new population needs.


© P. Dugué, CIRAD

A type of ecological intensification in developing countries

Making optimum use of synergies between agriculture and animal production


lthough animal traction was well received in sub-Saharan Africa, as it was possible to increase the area cultivated per farm worker adagascar. rice field, M a g lin and work became less laborious, quality til s bu Couple of ze D RA CI , ué ug manure production and fodder crop development for © P. D intensive animal production units have rarely been undertaken. Today, the demand from towns for food products is rising (cereals, legumes, milk, ruminant meat, but also meat of short cycle animals, poultry and pigs); mineral fertilizer prices continue to increase in line with petroleum prices; the motorization of farming operations and transport also comes up against rising fuel prices. In this context, by integrating agriculture and animal production, maximum advantage can be taken of the complementarities existing between cropping systems (fodder production, symbiotic nitrogen fixation and mineral nutrient recycling) and animal production systems (production of organic manure and energy) to reduce consumption of fuel, chemical fertilizers and concentrated feeds. Animal production should also be considered as a good "utilizer" of agricultural by-products, such as cereal bran, crop residues, etc., and of marginal zones unsuitable for farming. Lastly, integrating intensive animal production units with a few fattening cattle on family farms creates jobs and limits the creation of large livestock breeding units on the edge of towns, which are usually sources of pollution.

Contacts Patrick Dugué CIRAD Innovation Joint Research Unit 1573 rue Jean-François Breton 34398 Montpellier Cedex 5 France

Eric Vall CIRAD, Livestock Systems and Animal Product Management Research Unit, c/o CIRDES 01 BP 454 Bobo-Dioulasso Cedex 01 Burkina Faso

Synergies on a production unit scale

This approach enables researchers to more effectively work with producers, but it can also be used by technicians in development bodies to enhance their ways of advising mixed cropping/livestock farms.

Synergies on an agropastoral territory scale This involves adapting the rules of use for collectively used land and natural resources (rangelands, water points) to guarantee sustainable exploitation of agrosylvo-pastoral resources. The agropastoral activities that depend on those resources are thus supported and disputes between the different resource users (farmers, animal breeders, foresters, etc.) are limited. This can lead on to new agreements between the different socio-professional categories. For instance, they may extend to the use of crop residues, the guarding and mobility of herds, bushfire management, the conservation of banks along water courses, and rational exploitation of fodder trees. This new coordination, which could induce more sustainable synergies between production units, is in keeping with the establishment of local charters for the management of agropastoral territories acknowledged by the different socio-professional categories and local authorities, such as rural municipalities and the administration. Cattle, crop residues pasture. © P. Dugué, CIRAD

Stack of rice straw feeding a herd of zebus, Madagascar. © P. Dugué, CIRAD

Dairy cow fed with freshly cut vetch, Madagascar. © P. Dugué, CIRAD

Partners • CIRDES, Centre international de recherche-développement sur l'élevage en zone subhumide, Burkina Faso • INERA, Institut de l’environnement et de recherches agricoles, Burkina Faso • University of Bobo-Dioulasso, Burkina Faso • Union des Producteurs de coton du Tuy, Burkina Faso • FOFIFA, National Centre of Applied Research and Rural Development, Madagascar • BV-Lac project, Madagascar • EMBRAPA, Brazilian Agricultural Research Centre • IRAD, Institut de la recherche agricole pour le développement, Cameroon • SODECOTON, Société de développement du coton, Cameroon

© CIRAD, March 2010.

CIRAD is exploring various avenues for integrating agriculture and livestock in the production units. It is a matter of helping producers to design novel agricultural systems that are economically viable, socially acceptable and make optimum use of input investments (fertilizers, concentrated cattle feeds). This research is based on technical results already acquired and on farmers' know-how. Participation of rural stakeholders is incorporated into the different research phases. For instance, we assess with producers the possibilities of adopting these achievements or the need to adapt certain technical proposals. Over and above the experimental work undertaken with producers, CIRAD develops computerized tools to model mixed cropping/livestock farms, which are used to consider the future of the farms. These tools enable producers, but also agricultural advisors, to assess different evolution scenarios for their production unit in terms of monetary income and food security, soil fertility balances or the ability to feed animal production units. It is thus possible to assess the feasibility of adding a fodder crop to the rotation, or of increasing the number of animals to be fattened.

Pig production in tropical and subtropical regions

ig production provides almost 40% of the meat consumed worldwide. Pig production is expanding in Asia, Latin America and the non-Muslim zones of Africa. It is practised by small-scale pig farmers and provides essential income in rural zones. Faced with the technical, economic, environmental and health risks associated with developing this activity, ecological intensification is an interesting prospect for sustainable smallholder pig farms. CIRAD experts are adopting an interdisciplinary approach to assist this innovation process.


© V. Porphyre, CIRAD

Promoting sustainability to feed populations in the South

Contacts Gaining a clearer understanding of changes on pig farms


ost pig farms in developing countries are small family units, based on very diversified Reunion ghlands of hi e th in production methods ranging Pig farm re, CIRAD © V. Porphy from low-input extensive rearing to industrial off-land production. CIRAD is implementing projects in a close partnership with research and development organizations from the South, to: • assess locally-available food and animal resources, • analyse the technical and economic efficiency of pig production systems in tropical countries, • characterize the diversity of rearing units by gaining a clearer understanding of their technical, economic and social rationale, • understand the role of pig production in the pluri-active socio-economic strategies of smallholders in developing countries and in the marketing sectors, • model development patterns for farms and for the production and processing sectors to assist in their change.

Vincent Porphyre CIRAD Livestock Systems and Animal Product Management Research Unit Station Ligne Paradis 7 chemin de l’IRAT 97410 Saint Pierre La Réunion - France Jean-Michel Medoc CIRAD Environmental Risks of Recycling Research Unit Cité diplomatique de Van Phuc 298 Kim Ma 99 Hanoï Vietnam For further information:

Promoting the recycling and use of pig waste Pig rearing units discharge effluents that lead to environmental pollution problems. CIRAD proposes ways of managing pig production unit effluents to protect the environment and fertilize crops. It determines the composition of effluents to rationalize their recycling on crops and to design novel treatment processes; it analyses waste management practices and adapts them to crop management sequences; it helps to improve slurry recycling in agricultural systems for minimum environmental risks; it models environmental impacts in the production sector, using the Life Cycle Analysis method for better decision-support.

Pig farm, North Vietnam © V. Porphyre, CIRAD

Sow of the Mong Cai race, with her litter, Vietnam © V. Porphyre, CIRAD

Accompanying the pig production systems of tomorrow

Combination of small-scale pig farming and carp fish farming, North Vietnam © V. Porphyre, CIRAD

CIRAD helps the pig production sectors in developing countries to adopt alternative management methods, based on criteria for the ecological intensification of pig farms that are acknowledged by all the local stakeholders. It develops novel methods for sustainable development of pig farms by exploring ecological intensification methods, controlling animal diseases and adding value to products. Its research is designed to optimize use of local resources, improve the energy balance of pig farms for less energy-demanding development and assist stakeholders towards multi-criteria quality labelling of their products.

Partners • National Institute of Animal Science, Vietnam • Hanoi University of Agriculture, Vietnam • National Institute of Veterinary Research, Vietnam • Soils and Fertilizers Institute, Vietnam • Institute of Policy and Strategy for Agricultural and Rural Development, Vietnam • FOFIFA, National Centre of Applied Research and Rural Development, Madagascar • Coopérative des producteurs de porcs de la Réunion • Fédération régionale des coopératives agricoles, Réunion

Sharing research results

CIRAD proposes its PIGTrop portal on the internet:, providing access to the latest news and research results on original topics dealing with pig production in developing countries. This internet site is intended for researchers, students, professionals, pig rearers and development agencies interested in the sustainable development of the pig supply chains in developing countries. It presents the results of international research on animal health and emerging diseases, the socio-economic organization of the pig sector, integrated waste management, genetic management of populations, food strategies, optimum use of biodiversity and product quality. Today, PIGTrop is the unrivalled scientific reference portal on pig research for development in tropical supply chains.

• Qualitropic, Pôle de compétitivité Agro-Nutrition en milieu tropical, Réunion • Lycée agricole de SaintJoseph, Réunion • Institut national de la recherche agronomique, France • Royal Veterinary College, University of London, United Kingdom • Montpellier SupAgro, France

© CIRAD, March 2010.

Combination of pig farming, market gardening and fish farming, Red River Delta, Vietnam © V. Porphyre, CIRAD

Reconciling fodder production and environmental protection in the humid tropics Sustainable development of forage systems he humid tropics have seen a considerable expansion in ruminant farming since the 1970s and are currently home to almost a quarter of the world's ruminant stock. This situation is often criticized for its negative environmental effects: deforestation, loss of biodiversity, scrub invasion of the environment, greenhouse gas production, etc. CIRAD is involved in research to reconcile the development of ruminant farming in these regions to meet the food and economic needs of the populations, and the need to protect the environment.

Controlling pasture degradation Pastures established after deforestation are fragile environments which are rapidly invaded by scrub. As their restoration is delicate, they are often abandoned and aena replaced by other grasslands established on ol m ro Ch by ic infested ican Republ Central Afr e newly deforested areas. In order to limit th in nd Rangela CIRAD J. Huguenin, deforestation, CIRAD has developed odorata. Š management conditions for grasslands that prevent scrub invasion processes. The recommendations are intended to ensure rapid and dense soil cover. Grasslands have to be exploited regularly (high stocking rate, grazing rotation) in order to maintain a dense and uniform cover capable of limiting the germination and subsequent development of weeds. In this way, controlling grassland degradation indirectly helps to slow down further deforestation for new grasslands.

Agro-ecological management of forage environments Forage ecosystems in the humid tropics can be sustainably managed. It calls for precise and interactive organization of grassland and herd management to reconcile animal productivity, the lifespan of grasslands, and environmental services.

Š J. Huguenin, CIRAD


Contacts Johann Huguenin CIRAD Livestock Systems and Animal Product Management Research Unit Campus International de Baillarguet 34398 Montpellier cedex 5 France Blanfort Vincent CIRAD Livestock Systems and Animal Product Management Research Unit BP 701 97387 Kourou cedex Guyane - France For further information:

Plant growth is very rapid in the humid tropics but the nutritional optimum of the vegetation is short-lived. In addition, seasonal effects can also be a constraint for fodder production (cold season, dry season, excessively wet season). For intensive and agro-ecological grassland management, several measures have to be taken into account: • maintaining dense plant cover by adjusting the structure of plant covers through animal stocking (2 to 4 head per hectare) and rotation rates (3 to 6 weeks), • diversifying plant species to take into account seasonal effects and promote nutritional complementarity. For example, oats maintain a fodder supply during a cold season in certain humid tropical zones; grass-legume combinations, such as Panicum maximum and Stylosanthes hamata balance nutritional contributions, • choosing complementary fodder resources: forage gardens where the vegetation is cut and brought to the animals in addition to their pasture, fodder trees such as Leucaena, either browsed or exploited by pruning.

A Brahman zebu browsing Brachiaria humidicola grassland in French Guiana. © J. Huguenin, CIRAD

These agro-ecological fodder intensification measures lead to greater productivity while preserving the environment and limiting further expansion of areas.

CIRAD is studying biological dynamics that are conducive to restoring ecological balances in environments disrupted by herbivore production. The aim is to strengthen the stability of rearing units while attenuating their negative environmental impacts by more effectively providing certain ecosystemic services: • Limiting greenhouse effect gases: ruminant production contributes to Rangeland invasion by Jatropha gossipifolia in New Caledonia. greenhouse gas emissions, but © V. Blanfort, CIRAD grassland agrosystems compensate for those emissions by sequestrating carbon in the soil (1 to 2 tonnes per hectare per year). In temperate zones, the carbon stock in soils under grassland can reach 65 tonnes per hectare. • Protecting soils: continuous cover grasslands offer major protection against soil erosion; soil fertility Flight over the Transamazonian zone, Para State, Brazil. under grasslands displays a drop in © J. Huguenin, CIRAD acidity, an increase in nutrient storage and high active organic matter content; soil aluminium toxicity diminishes. • Maintaining biodiversity: rotations with high animal stocking help to control scrub invasion by preventing the development of invasive plants, which cause a severe reduction in biodiversity, including in forest areas next to grazing lands.

Partners • EMBRAPA, Brazilian Agricultural Research Centre • Federal University of Para State, Brazil • INRA, Institut national de la recherche agronomique, France • Montpellier SupAgro, France • Coopérative des éleveurs de bovins, French Guiana • Sica Lait and Sica Revia, La Réunion • Institut agronomique calédonien, New Caledonia • Agence nationale de développement de l’élevage, Central African Republic • FOFIFA, National Centre of Applied Research and Rural Development, Madagascar • FIFAMANOR, Centre de développement rural et de recherche appliquée, Madagascar • National Institute of Animal Science, Vietnam

© CIRAD, March 2010.

Offer of ecological services

Bibliographie / Literature

Comment mieux contrôler les invasions de criquets / More effective control of locust invasions • Chapuis M.-P., Loiseau A., Michalakis Y., Lecoq M., Franc A., Estoup A., 2009. Outbreaks, gene flow and effective population size in the Migratory locust, Locusta migratoria: a regional scale comparative survey. Molecular Ecology (Oxford), 18(9): 792-800. • Franc A., Rabesisoa L., Luong-Skovmand M.H., Lecoq M., 2005. Phase polymorphism in the Red locust, Nomadacris septemfasciata (Orthoptera: Acrididae) in Madagascar. International Journal of Tropical Insect Science (Nairobi), 25(3): 182-189. • Lecoq M., 2005. Desert locust management: from ecology to anthropology. Journal of Orthoptera Research (Ann Arbor, MI), 14(2): 179-186. • Maiga I.H., Lecoq M., Kooyman C., 2008. Ecology and management of the Senegalese grasshopper Oedaleus senegalensis (Krauss 1877) (Orthoptera: Acrididae) in West Africa: review and prospects. Annales de la Société Entomologique de France (nouvelle série) (Paris), 44(3): 271-288. • Magor J.I., Lecoq M., Hunter D.M., 2008. Preventive control and desert locust plagues. Crop Protection (Oxford), 27(12): 1527-1533.

Maitriser le scolyte des baies du caféier / Coffee berry borer control • Dufour B.P., Franco F., Hernández A., 2007. Evaluación del trampeo en el marco del manejo integrado de la broca del café. In: La broca del café en América tropical: hallazgos y enfoques, workshop internacional, junio 2007, Acapulco, Guerrero, México. Ed. por Barrera J.F., García A., Domínguez V., Luna C., ECOSUR y Soc. Mex. Ent., México, Mexique, 89-99. • Dufour B.P., González M.O., Mauricio J.J., Chávez B.A., Ramírez Amador R., 2005. Validation of coffee berry borer (CBB) trapping with the BROCAP® trap. In: XX International Conference on Coffee Science, 11-15 October 2004, Bangalore, India. ASIC, Paris, France, 1243-1247.

• Avelino J., Zelaya H., Merlo A., Pineda A., Ordoñez M., Savary S., 2006. The intensity of a coffee rust epidemic is dependent on production situations. Ecological modelling, 197 (3-4): 431-447. • Ratnadass A., Togola M., Cissé B., Vassal J-M., 2009. Potential of sorghum and physic nut (Jatropha curcas) for management of plant bugs (Hemiptera: Miridae) and cotton bollworm (Helicoverpa armigera) on cotton in an assisted trap-cropping strategy. Journal of SAT Agricultural Research, 7.

Gestion agro-écologique des cultures fruitières et maraîchères / Agro-ecological management of fruit and market garden crops • Malézieux E., Crozat Y., Dupraz C., Laurans M., Makowski D., Ozier-Lafontaine H., Rapidel B., de Tourdonnet S., ValantinMorison M., 2008. Mixing plant species in cropping systems: concepts, tools and models. A review. Agronomy for Sustainable Development, 29: 43-62. • Ratnadass A., Michellon R., Randriamanantsoa R., Séguy L., 2006. Effects of soil and plant management on crop pests and diseases. In: Uphoff N., Ball A., Fernandes E., Herren H., Husson O., Laing M., Palm C., Pretty J., Sanchez P., Sanginga N., Thies J., Biological Approaches for Sustainable Soil Systems. Boca Raton, Etats-Unis, CRC Press, p. 589-602.

Les mouches des fruits et des légumes en milieu tropical / Fruit and vegetable flies in the Tropics • Rousse P., Gourdon F., Quilici S., 2006. Host specificity of the egg pupal parasitoid Fopius arisanus (Hymenoptera: Braconidae) in La Réunion. Biological Control, 37 (3): 284-290. • Duyck P.F., Junod P., Brunel C., Dupont R., Quilici S., 2006. Importance of competition mechanisms in successive invasions by polyphagous tephritids in La Réunion. Ecology, 87 (7): 1770-1780. • Rousse P., Chiroleu F., Veslot J., Quilici S., 2007. The host- and microhabitat olfactory location by Fopius arisanus suggests a broad potential host range. Physiological Entomology, 32: 313-321.

• Dufour B.P., 2007. Condiciones de uso de las trampas en el control de la broca del café. In: Manejo da broca-do-café, workshop internacional, 28 nov. 2004, Londrina, Paraná, Brasil. IAPAR, Londrina, Brésil, 85-93.

• Vayssières J.F., Cayol J.P., Perrier X., Midgarden D., 2007. Impact of methyl eugenol and malathion bait stations on non-target insect populations in French Guiana during an eradication program for Bactrocera carambolae. Entomologia Experimentalis et Applicata, 125 (1): 55-62.

Régulation des bio-agresseurs dans les agrosystèmes tropicaux / Regulating pests and diseases in tropical agrosystems

• Van Mele P., Vayssières J.F., Van Tellingen E., Vrolijks J., 2007. Effects of the African weaver ant Oecophylla longinoda in controlling mango fruit flies (Diptera Tephritidae). Journal of Economic Entomology, 100 (3): 695-701.

• Avelino J., Willocquet L., Savary S., 2004. Effects of crop management patterns on coffee rust epidemics. Plant pathology, 53 (5), 541-547.

• Vayssières J.F., Goergen G., Lokossou O., Dossa P., Akponon C., 2005. A new Bactrocera species detected in Benin among mango fruit flies (Diptera Tephritidae) species. Fruits, 60 (6): 1-9.

© N. le Gall

Inventer une agriculture écologiquement intensive pour nourrir la planète

Lutte contre les insectes ravageurs en culture cotonnière / Controlling insect pests in cotton growing systems • Brévault T., Couston L., Bertrand A., Thézé M., Nibouche S., Vaissayre M., 2009. Sequential pegboard to support small farmers in cotton pest control decision-making in Cameroon. Crop Protection, 28: 968-973. • Brévault T., Carletto J., Linderme D., Vanlerberghe-Masutti F., 2008. Genetic diversity of the cotton aphid, Aphis gossypii, in the unstable environment of a cotton growing area. Agricultural and Forest Entomology, 10: 215 – 223. • Brévault T., Bikay S., Maldès J.M., Naudin K., 2007. Impact of no till with mulch on soil macrofauna communities in a cotton cropping system. Soil & Tillage Research, 97: 140-149.

Gestion du risque pesticide en horticulture / Managing pesticide risks in horticulture • Cabidoche Y.-M., Jannoyer M., Vanniere H., 2006. Conclusions du Groupe d’étude et prospective, aspects agronomiques, Contributions Cirad – Inra. Montpellier, Cirad, 64 p. • Cabidoche Y.-M., Achard R., Cattan P., Clermont Dauphin C., Chabrier C., Lafont A., Sansoulet J., 2006. Stockage dans les sols à charges variables et dissipation dans les eaux de zoocides organochlorés autrefois appliqués en bananeraies aux Antilles : relation avec les systèmes de culture, Rapport final d’exécution Programme 2003-2005 « Evaluation et réduction des risques liés à l’utilisation des pesticides » du MEDD, Inra Cirad, 99 p.

• De Moraes Sá, Séguy l., Gozé E. et al, 2009. C sequestration rates in no-tillage soils under intensive cropping systems in tropical agroecozone. Soil Science Society of America Journal, in press. • Séguy L., Bouzinac S., Husson O., 2006. Direct-seeded tropical soil systems with permanent soil cover: Learning from Brazilian experience. In: Uphoff Norman T. (ed.), Ball Andrew S. (ed.), Fernandes Erik C.M. (ed.), Herren Hans R. (ed.), Husson. Olivier (ed.), Laing Mark V. (ed.), Palm Cheryl (ed.), Pretty Jules (ed.), Sanchez Pedro (ed.), Sanginga Nteranya (ed.), Thies Janice (ed.). Biological approaches to sustainable soil systems. Boca Raton, CRC Press, p. 323-342. • Séguy L., Bouzinac S., 2008. La symphonie inachevée du semis direct dans le Brésil central : le système dominant dit de « semidirect » : limites et dégâts, éco-solutions et perspectives : la nature au service de l'agriculture durable. [Cd-Rom]. Montpellier, Cirad, 1 disque optique numérique (CD-ROM).

Intensification écologique chez le bananier / Ecological intensification of banana growing • Côte F.-X., Abadie C., Achard R., Cattan P., Chabrier C., Dorel M., de Lapeyre de Bellaire L., Risède J.M., Salmon F., Tixier P., 2009. Integrated pest management developed in the French West Indies to reduce pesticide use in banana production systems. Acta Horticulturae, 828: 375-382. • Chabrier C., Carles C., Desrosiers C., Quénéhervé P., Cabidoche Y.M., 2009. Nematode dispersion by runoff water: Case study of Radopholus similis (Cobb) Thorne on nitisol under humid tropical conditions. Applied soil ecology, 41: 148-156.

• Lesueur Jannoyer M., Cabidoche Y.-M., Vannière H., 2007. La chlordécone aux Antilles françaises, Synthèse sous l’angle agronomique établie par le Groupe d’étude et prospective, sur les pollutions par les organochlorés. Phytoma - La défense des végétaux, 606: 29-31.

• Tixier P., Malezieux E., Dorel M., Wery J., 2008. SIMBA, a model for designing sustainable banana-based cropping systems. Agricultural systems, 97 (3): 139-150.

• De Roffignac L., Cattan P., Mailloux J., Herzog D., Le Bellec F., 2008. Efficiency of a bagasse substrate in a biological bed system for the degradation of glyphosate, malathion and lambda-cyhalothrin under tropical climate conditions. Pest Management Science, 64 (12): 1303-1313.

Améliorer les systèmes agroforestiers en zone tropicale humide / Improving agroforestry systems in the humid tropics

Culture des vergers en milieu insulaire / Orchard cultivation on islands • Bockstaller C., Girardin P., 2003. How to validate environmental indicators. Agricultural Systems, 76: 639-653. • Boullenger G., Le Bellec F., Girardin P., Bockstaller C., 2008. Évaluer l’impact des traitements des agrumes sur l’environnement. Phytoma La défense des végétaux, 617: 22-25. • Lançon J., Wery J., Rapidel B., Angokaye M., Gerardeaux E., Gaborel C., Ballo D., Fadegnon B., 2007. An improved methodology for integrated crop management systems. Agronomy for Sustainable Development, 27: 101-110. • Le Bellec F., Le Bellec V., 2008. Le jardin créole : produire en respectant l'environnement. Chevagny-sur-Guye, Ed. Orphie, 44 p. • Le Bellec F., Herzog D., Fournier P., Mauléon H., Renard-Le Bellec V., Ramassamy M., 2005. The integrated fruits production in Guadeloupe, In: CFCS, Guadeloupe. 41st Annual Meeting of the Carribean Food Crop Society, Annual Meeting of the Carribbean Food Crops Society (CFCS), 41, 2005-07-10/2005-07-16, Gosier, Guadeloupe. • Loyce C., Wery J., 2006. Les outils des agronomes pour l’évaluation et la conception des systèmes de culture. In: Doré T. et al. (Eds). L’agronomie aujourd’hui. Versailles, Editions Quae, p. 77-95.

• Camara A., Dugué P., Kalms J.-M., Cheylan J.-P., 2009. De la forêt naturelle aux agroforêts cultivées en Guinée forestière. Cahiers Agriculture, 18 (5): 425-431. • Deheuvels O., Penot E., 2008. Olympe, a multiscale tool to explore management options in Agroforestry Systems. In: B. Rapidel, O. Roupsard and M. Navarro (Eds), Modelling agroforestry systems with perennial crops: connecting agroforestry researchers with modellers. CATIE, Turrialba, Costa Rica, 25-28 Feb 2008. • Jagoret P., Malézieux E., 2007. Complex cocoa agroforests can be successfully established on savannahs: a local innovation in the central region of Cameroon. 2nd International Symposium on Multistrata Agroforestry Systems with Perennial Crops, CATIE, Turrialba, Costa Rica. • Lamanda N., Michel-Dounias I., Canet M., Kalms J.-M., 2008. Les pratiques de gestion des agroforêts à base de café de Guinée forestière sur le temps long. Atelier international de réflexion à partir de visites de terrain « Les agroforêts d’Afrique de l’Ouest et du Centre : dynamiques, performances et avenir ? 2008, Sérédou, Guinée, CD-Rom. • Madelaine C., Malézieux E., Sibelet N., Manlay R. 2008. Semiwild palm groves reveal agricultural change in the forest region of Guinea. Agroforestry systems, 73 (3): 189-204.

Le semis direct avec couverture végétale / Direct seeding mulch-based cropping

• Marie C., Sibelet N., Dulcire M., Rafalimaro M., Danthu P., Carriere S., 2009, Taking into account local practices and indigenous knowledge in an emergency conservation context in Madagascar. Biodiversity and Conservation, 18 (10): 2759-2777.

• De Moraes Sá, Séguy L., Gozé E. et al, 2009. Carbon balance and sequestration rates in a long-term tillage experiment in a Brazilian Oxisol. Soil tillage research, in press.

• Ruf F., 2009. Libéralisation, cycles politiques et cycles du cacao : le décalage historique Côte d’Ivoire-Ghana. Cahiers agricultures, 18 (4): 343-349.

Impacts agri-environnementaux du palmier à huile / Agri-environmental impacts of the oil palm

• Scopel E., Macena F., Corbeels M., Affholder F., Maraux F., 2004. Modelling crop residue mulching effects on water use and production of maize under semi-arid and humid conditions. Agronomie, 24: 1-13.

• Caliman J.P., Carcasses R., Girardin P., Pujianto A., Dubos B., Liwang T., 2005. Development of agri-environmental indicators for sustainable management of oil palm growing. General concept and the example of nitrogen. MPOB International Palm Oil Congress (PIPOC), 25-29 September 2005, Kuala Lumpur, Malaysia.

Production durable de charbon de bois en République démocratique du Congo / Sustainable charcoal production in the Democratic Repubic of Congo

• Wohlfahrt J., Caliman J.P., Girardin P., Wahyu A., 2006. Development of an agro-ecological indicator (I-PHY Palm) to assess the sustainability of pesticides utilization in oil palm plantations. IOPRI International Palm Oil Conference, 20-24 June 2006, Bali, Indonesia. • Girardin P., Caliman J.P., Wohlfahrt J., 2007. INDIGO® Palm: a method based on Agro-Ecological indicators to assess the environmental stability of oil palm plantations. International Conference on Oil Palm and the Environment ICOPE 2007. 15-16 November 2007, Bali, Indonesia. • Caliman J.P., Carcasses R., Perel R., Wohlfahrt J., Girardin P., Wahyu Pujianto A., Dubos B., Verwilghen A., 2007. Indicadores agro-ambientales para la producción sostenible de aceite de palma. Palmas, 28(1): 434-445.

Production durable en agriculture familiale au Sud / Sustainable production on family farms in developing countries • Affholder F., Jourdain D., Dang Dinh Quang, To Phuc Tuong, Morize M., Ricome A., 2010. Direct-seeding mulch-based cropping systems on mountainous slopes in Vietnam: a whole farm model for assessing constraints to adoption by farmers. Agricultural Systems, 103 (1): 51-62. • Triomphe B., Sabourin E., Hocdé H., Scopel E., Nascimento de Oliveira M., Valadares Xavier J.H., Da Silva F.A.M., Ramos de Almeida S.C., 2008. Participatory cropping and farming system design among multiple stakeholders to contribute to sustainable agricultural production. Experiences and lessons with the agrarian reform sector in the Brazilian Cerrados. In: Dedieu Benoît (ed.), Empowerment of the rural actors. A renewal of farming systems perspectives: 8th European IFSA Symposium, 6-10 July 2008, ClermondFerrand. Paris, INRA, IFSA European Symposium, 8, 2008-0706/2008-07-10, Clermont-Ferrand, France, 10 p. • Corbeels M., Scopel E., Cardoso A., Bernoux M., Douzet J.M., Siqueira Neto M., 2006. Soil carbon storage potential of direct seeding mulch-based cropping systems in the Cerrados of Brazil. Global Change Biology, 12: 1-15. • Scopel E., Macena da Silva F.A., Corbeels M., Affholder F., Maraux F., 2004. Modelling crop residue mulching effects on water use and production of maize under semi-arid and humid tropical conditions. Agronomie, 24: 1-13.

Préserver la biodiversité des savanes africaines / Preserving biodiversity in African grasslands

• Bernhard-Reversat F., Diangana D., Tsatsa M., 1993. Biomasse, minéralomasse et productivité en plantation d'Acacia mangium et Acacia auriculiformis au Congo. Bois et Forêts des Tropiques, 238 (4), 35-44. • Bisiaux F., Peltier R., Muliele J-P., 2009. Plantations industrielles et agroforesterie au service des populations des plateaux Batéké, Mampu, en République démocratique du Congo. Bois et Forêts des Tropiques, 301 (3): 21-31. • Harmand J.-M., Njiti C.F., Peltier R., 1997. Restauration de la fertilité des sols par la jachère arborée. L'agroforesterie pour un développement rural durable. Atelier international, Montpellier, France 23-29 juin 1997, p. 135-142. • Harmand J.-M., Njiti, C.F., Bernhard-Reversat, F., Puig H., 2004. Aboveground and belowground biomass, productivity and nutrient accumulation in tree improved fallows in the dry tropics of Cameroon. Forest Ecology and Management, 188: 249-265. • Marien J.-N., Mallet B., 2004. Nouvelles perspectives pour les plantations forestières en Afrique centrale. Bois et Forêts des Tropiques, 282 (4): 67-79. • Peltier R., Balle Pity, 1993. De la culture itinérante sur brûlis au jardin agroforestier en passant par les jachères enrichies (From Slash and burn to sustainable agroforestry system). Bois et Forêts des Tropiques, 235 (1): 49-57. • Peltier R., Balle Pity, Galiana A., Gnahoua G.M., Leduc B., Mallet B., Oliver R., Oualou K., Schroth G., 1995. Produire du bois énergie dans les jachères de zone guinéenne. Intérêts et limites à travers l'expérience d'Oumé en Basse Côte d'Ivoire. In: Actes du séminaire « Fertilité du milieu et stratégies paysannes sous les tropiques humides », nov. 1995, Montpellier, France, p. 219-227.

Diffusion des systèmes de semis direct avec couverture végétale à Madagascar / Dissemination of direct seeding mulch-based cropping systems in Madagascar • Penot E., 2008. Document de travail n° 4 : Mise en place du réseau de fermes de références avec les opérateurs du projet. Projet BV-lac/AFD, lac Alaotra, Madagascar, 2008. • Penot E., Garin P., 2009. Des savoirs aux savoirs faire : l’innovation alimente un front pionner : le lac Alaotra de 1897 à nos jours. Colloque « Localisation et circulation des savoir-faire en Afrique ». Maison méditerranéenne des sciences de l’Homme, Aix-enProvence, France, 19 et 20 mars 2009.

• Baudron F., Corbeels M., Monicat F., Giller K.E., 2009. Cotton expansion and biodiversity loss in African savannahs, opportunities and challenges for conservation agriculture: a review paper based on two case studies. Biodiversity and Conservation, 18: 2625-2644.

• Penot E., 2008. Mise au point d’outils et d’approche pour l’aide à la décision technico-économique et organisationnelle dans les projets de développement agricole à Madagascar. Séminaire international sur la capitalisation des expériences pour l’apprentissage social et le développement. ICRA, Antananarivo, Madagascar, 10-12 novembre 2008.

• Baudron F., Corbeels M., Monicat F., Giller K.E. (submitted). Cotton, more than tsetse eradication, drove habitat loss for the wildlife of the Mid Zambezi Valley, Zimbabwe. Biological Conservation.

• Terrier M., Penot E, 2008. Document de travail/AFD/BV-lac n° 18 : conventions de modélisation pour le RFR. Projet BV-lac/AFD, lac Alaotra, Madagascar, 2008.

• Corbeels M., Scopel E., Cardoso A., Bernoux M., Douzet J.M., Siqueira Neto M., 2006. Soil carbon storage potential of direct seeding mulch-based cropping systems in the Cerrados of Brazil. Global Change Biology, 12: 1-15.

• Cauvy S., Penot E., 2009. Document n° 43 : mise au point des scénarios en analyse prospective et des simulations sur les exploitations agricoles du réseau de fermes de référence. Projet BV-lac/AFD, lac Alaotra, Madagascar, 2009.

Optimiser la production de biomasse en minimisant l’impact sur l’environnement / Optimizing biomass production whilst minimizing environmental impact • Barczi J-F., Rey H., Caraglio Y., de Reffye P., Barthelemy D., Dong QX., Fourcaud T., 2008. AmapSim: a structural whole-plant simulator based on botanical knowledge and designed to host external functional models. Annals of Botany, 101: 1125-1138. • Barthelemy D., Caraglio Y., 2007. Plant architecture: A dynamic, multilevel and comprehensive approach to plant form, structure and ontogeny. Annals of Botany, 99: 375-407. • De Reffye P., Barthélémy D., Cournède P. H., Jaeger M., 2008. Modélisation et simulation de l'architecture et de la production végétales. In: Hallé, F. (Ed) Aux origines des plantes : des plantes anciennes à la botanique du XXIe siècle. Paris, Fayard, p. 187-229. • Leroy C., Sabatier S., Wahyuni S., Barczi JF., Dauzat J., Laurans M., Auclair D., 2009. Virtual trees and light capture: a method for optimizing agroforestry stand design. Agroforestry Systems, 77: 37-47. • Mathieu A., Cournede P.H., Letort V., Barthelemy D., de Reffye P., 2009. A dynamic model of plant growth with interactions between development and functional mechanisms to study plant structural plasticity related to trophic competition. Annals of Botany, 103: 1173-1186.

• Mikolasek O., Trinh D.K., Médoc J.M., Porphyre V., 2009. L’intensification écologique d’un modèle de pisciculture intégrée : recycler les effluents d’élevages porcins de la province de Thai Binh (Nord Vietnam). Cahiers Agriculture, 18 (2): 235-241. • Madec, F., Hurnik, D., Porphyre, V., Cardinale, E., 2009. Good practices for biosecurity in pig sector: issues and options. FAO/OIE/World Bank - Animal Production and Health Paper. Food and Agriculture Organization of the United Nations, Roma. In press 2009. • Médoc J.-M., Guerrin F., Courdier R., Paillat J.-M., 2004. A Multimodelling approach to help agricultural stakeholders design animal wastes management strategies in the Reunion Island. In: Pahl-Wostl C., (ed.), Schmidt S., (ed.), Rizzoli A.E., (ed.), Jakeman A.J., (ed.). Complexity and integrated resources management. Transactions of the 2nd Biennial Meeting of the International Environmental Modelling and Software Society. Volume 1. Manno, Switzerland: iEMSs, 462-467. Complexity and Integrated Resources Management, 2004/06/14-17, Osnabrück, Germany. • Farinet J.L., Nuttens F., Vanai P., 2005. 2Co-composting of pig manure with green wastes to prevent environmental impact of pig production in the Wallis archipelago, Pacific ocean. In: Proceedings of the International Workshop on green pork production, Paris, France, May 25-27, 2005, p. 111-112.

• Rey H., Dauzat J., Chenu K., Barczi J-F., Dosio GAA., Lecoeur J., 2008. Using a 3-D virtual sunflower to simulate light capture at organ, plant and plot levels: contribution of organ interception, impact of heliotropism and analysis of genotypic differences. Annals of Botany, 101: 1139-1151.

Concilier production fourragère et renouvellement des ressources en zone tropicale humide / Reconciling fodder production and environmental protection in the humid tropics

L’intégration de l’agriculture et de l’élevage / Integrating agriculture and animal production

• Blanfort V., Orapa W., 2008. Ecology, impacts and management of invasive plant species in pastoral areas. Proceedings of the Regional Workshop on invasive Plant Species in Pastoral Areas, 24-28 november 2003, Koné, New Caledonia. IAC/SPC, Suava, 201 p.

• Andrieu N., Dugué P., Le Gal P.Y., Schaller N., 2009. Modéliser le fonctionnement d'exploitations agricoles de polyculture élevage pour une démarche de conseil. Cas de la zone cotonnière de l'ouest du Burkina Faso In: Actes du colloque Savanes africaines en développement : Innover pour durer, 21-24 avril 2009, Garoua, Cameroun, 11 p. • Vall E., Diallo M.A., 2009. Savoirs techniques locaux et pratiques : la conduite des troupeaux aux pâturages (ouest du Burkina Faso). Natures sciences sociétés, 17 (2): 122-135. • Projet FERTIPARTENAIRES

Production porcine dans les régions chaudes / Pig production in tropical and subtropical regions

• Hostiou N., Tourrand J.-F., Huguenin J., Lecomte P., 2006. La diversité de gestion des systèmes herbagers en Amazonie : cas des élevages bovins brésiliens. Fourrages, 187: 377-392. • Huguenin J., Duru M., Blanfort V., Tourrand J.F., Bergère H., 2009. Conduite et organisation agropastorale des prairies pâturées dans les élevages guyanais. Communication In: Actes des Rencontres des Recherches Ruminants, Paris, 2-3 déc. 2003, p. 353-356. • Rippstein G., Escober G., Motta F., 2001. Agroecologia y biodiversidad de las Sabanas en los Llanos Orientales de Colombia. CIAT et CIRAD, Colombie, 302 p.

• Porphyre V., 2009. Enjeux et contraintes des filières porcines en Afrique de l'Ouest. Revue Grain de Sel, 46-47 : Répondre aux évolutions alimentaires, un défi majeur pour l’élevage africain, MarsAoût 2009, p. 26-27.

• Salgado P., Lubbers M., Schipper R.A., Van Keulen H., Alary V., Lecomte P., 2009. Adoption of new forage technology: impact on the socio-economic sustainability of milk production in Moc Chau, Vietnam (DAIVIE model). In: Proceedings-AgSAP 2009, Netherlands 10-12 march, p. 272-273.

• Porphyre V., Nguyen Q.C., 2006. Pig production development, animal-waste management and environment protection: a case study in Thai Binh province, Northern Vietnam. PRISE publications, Hanoi, Vietnam, 224 p.

• Vayssières J., Guerrin F., Paillat J.M, Lecomte P., 2009. GAMEDE: A global activity model for evaluating the sustainability of dairy enterprises, Part I – Whole farm dynamic model. Agricultural Systems, 101: 128–138.

Inventing a new forme of agriculture