Multi-scale bio-economic modeling for the generation of innovative cropping system mosaics increasing food self-sufficiency

The achievement of sustainable agriculture, economically viable, socially equitable, environmentally sound and capable of providing sufficient amounts of food in both quantity and quality for a population is a major challenge of agricultural research. In Guadeloupe, a French overseas territory located in the Caribbean, agriculture is currently facing some difficulties for reaching food self-sufficiency. Crop production weakly covers the needs of the population and consumers must buy expensive imported foodstuff. In this context, local and national decision-makers started thinking about the emergence of new local agricultural production in favor of food security by "the diversification of agricultural productions" towards local vegetables and fruits. However, achieving food self-sufficiency is highly complicated due to the multi-scale nature of this topic. Constraints for food crops cultivation can be found at each scale, the field with cropping systems not well adapted for the production of food crops, the farm with limited production resources, e.g. workforce, the region with a lack of water provision by specific water services. Agricultural policies can overcome these constraints by modifying the current state of the agricultural system at each spatial scale. In order to understand what would be the conditions for achieving food self-sufficiency, we built a static multi-scale bioeconomic model based on mathematical programming. This model simulates the decision process of 5336 farmers in Guadeloupe in terms of choice of cropping systems for their 25054 fields. A geographic database and activities compose the inputs of the model. The geographic database is built with a large range of geographical information encompassing the slope, the soil type, the altitude, the access to irrigation while the activities are a set of cropping systems determined by the expert knowledge of local farm advisers. that defined both allocation rules of the cropping systems at each spatial scale and mean outputs of cropping systems. A explorative scenario with new food crop systems at field scale, increased quantities of workforce at farm scale and extension of water provision areas at regional scale showed a large increase in food self-sufficiency. Such changes tested in our model appear as possible levers for actions that could be targeted by agricultural policies