In Madagascar Highlands, population pressure and intensive rice cultivation in the lowlands has led to cultivation on the fragile, low fertility soils of the hillsides. Fertilisation being limited, yields remain low while erosion leads to paddy field silting-up and destruction of downstream infrastructures. To overcome these problems, experiments on direct planting on permanent soil cover (DPPSC) have been conducted since 1991 in the Andranomanelatra farm and since 1994 in farmers' fields, covering a wide range of biophysical and socio-economic situations

Agriculture in Madagascar has to face tremendous constraints: very low investment capacity, poor infrastructure network, low availability and high prices of inputs, uncontrolled bush fires and deforestation leading to severe erosion, loss of arable land and destruction of downstream infrastructures. As a consequence, yields are low and farmers' condition very precarious. Techniques of direct planting on permanent soil cover (DPPSC) can lead to solutions for sustainable and profitable agriculture.

Is it possible to coordinate the strategies and actions of multiple actors around the common goal of developing and diffusing Conservation Agriculture (CA) practices? How crucial is farmer participation in these efforts? How much of an emphasis should be placed on CA technologies vs. other types of innovations? The initial steps of the ASOSID project provide partial answers to these questions in the context of the Bajio, a semi-arid, subtropical region of Central Mexico dominated by post-green revolution, highly productive irrigated small-scale farming, presently facing decreasing profitability of cereal production, dwindling availability of irrigation water and soil fertility degradation.

The smallholder farmers of the Bajio face serious threats to the sustainability of their agriculture such as decreasing profitability of cereal farming, dwindling availability of irrigation water and degradation of soil fertility. Conservation Agriculture (CA) constitutes a very attractive alternative as it promises to bring about instantaneous reductions in production costs, savings in irrigation water and gradual improvements of soil fertility.

General results: reduction of labour very positive. Gain in weeding operations makes agriculture more adapted to women and children or even farmers weaken by HIV-AIDS. Good adoption of manual direct planters (11 “planting canes” and 6”sowing wheels”).

Les techniques de semis direct sur couverture végétale ont été développées au Brésil au cours de la dernière décennie. Ces techniques proposent un changement de paradigme et ouvrent de nouvelles perspectives pour l'agriculture. Elles sont maintenant utilisées sur des millions d’hectares, dans le monde entier. Depuis 1999, elles sont en cours d’adaptation aux conditions du Centre et du Nord du Vietnam, pour les zones de collines et de montagnes. Elles sont basées sur quelques principes fondamentaux.

Sur sols bruns karstiques, les principaux problèmes sont liés aux fortes pentes favorisant l’érosion, à la mauvaise structure du sol, à la faible activité biologique, à la faible fertilité chimique (N) pour la culture du maïs, et aux mauvaises herbes. Les systèmes de culture à base de semis direct avec couverture végétale permettent d’apporter des réponses à ces différentes contraintes

Le système techniquement le plus simple consiste à un semis direct dans un paillage constitué des mauvaises herbes et/ou de mulch importé permet de découvrir le semis direct et l’intérêt d’une couverture du sol pour lutter contre l’érosion, réduire l’évaporation de l’eau, restaurer une activité biologique et contrôler

Tillage accelerates organic matter destruction under tropical agriculture conditions. No-till cropping systems involving direct seeding on permanent plant cover enable short-term soil restoration. Cover plant choices are crucial. Soil carbon can thus be boosted to levels generally found in natural ecosystems, even when starting from degraded soils. Direct seeding systems promote net CO2 storage rather than net production. CIRAD has been working on such systems in Brazil, Asia, Réunion and Madagascar. This poster reports some results obtained in central-western Brazil (hot humid tropical area).

Rapport technique sur l'amélioration variétale riz à Madagascar 2002/03: Création variétale, collections testées, essais variétaux, tests en milieu paysan