You can¿t eat your mulch and have it too. Cropping system design and trade-offs around biomass use for Conservation Agriculture in Cameroon and Madagascar

Conservation agriculture is defined by three main principles: minimum soil disturbance, permanent soil cover and crop rotations. CA is promoted as a promising technology for Africa, but to date, only a small area under CA fully complies with the above three principles. CA has both short and long term effects on crop productivity and sustainability through the modification of various agroecological functions. These functions are related to the quantity of crop and cover crop biomass produced and kept as mulch. One of the main challenges in designing CA for smallholder farming systems in developing countries is the competing uses for biomass, in particular for feeding livestock. The main difficulties are linking the efficiency of agroecological functions to varying degrees of biomass export, and evaluating the performance of cropping systems at farm level, which is where the decisions are made. In North Cameroon the quantity of biomass produced in the field has been doubled by associating a cover crop with a cereal crop. Part of the biomass was consumed by cattle during the dry season but the quantity of mulch that remained on the ground had a positive impact on the cotton water balance in the driest part of North Cameroon. In the Lake Alaotra region of Madagascar, the soil cover in rice fields under CA can vary, from 30% to 84% even in the same type of field depending on the plant used as cover crop, the quantity of biomass produced and management of the residues. The range is even greater when different kinds of fields are taken into consideration. Of course, the different agroecological functions can be fulfilled to a greater or lesser extent depending on the amount of available biomass and the resulting soil cover. The relationship between the quantity of biomass and soil cover has been calculated for different kinds of residues. We used these relationships to explore the variability of soil cover that could be generated in farmers' fields, and to estimate how much of the biomass could be removed to feed livestock while leaving sufficient soil cover. Our results showed that under farmers' conditions in Madagascar, the production and conservation of biomass was not always sufficient to fulfill all the agroecological functions of mulch. For example, partial export of biomass to be used as forage might have no effect in terms of erosion control but may considerably reduce the efficiency of physical weed control. As the balance between the potential benefits of exporting biomass and the efficiency of agroecological functions varies depending on the constraints and goals of each farm, we chose to analyze the potential benefits of exporting aboveground biomass to feed cattle at farm level. To this end, we modeled different size farms in Madagascar to investigate the relation between raising dairy cows and efficient application of CA. Our aim was to explore trade-offs and synergies between combinations of CA practices (i.e. different amounts of biomass exported) and the size of dairy cow herds (varying biomass needs and animal production). Changing the percentage of soil cover in CA plots did not significantly modify total farm net income, as this was more influenced by the characteristics of the milk market. Overall, CA systems can be beneficial for dairy cow farmers thanks to the forage produced, although the milk market and thus the value of biomass for forage, has a major influence on the way CA can be implemented at field level. To explore the range of possible cropping systems in a given biophysical situation, we created a tool named PRACT (Prototyping rotation and association with cover crop and no till). We used this tool to organize expert knowledge on crops and cover crops, biophysical characteristics of fields and agronomic rules and to link them using Malagasy conditions. PRACT generate a list of cropping systems, i.e. crops and cover crops and their sequences over three years. These cropping systems are characterized b