Nitrogen fertilisation recommendations : could they be improved using stochastically generated climates in conjunction with crop models ?

Accurate determination of optimal Nitrogen (N) recommendations which ensure maximization of farmer's revenue while minimizing the environmental constraint is maybe among the major challenges in agriculture. Crop models have the potential to deal with such aspects and could thus be used to develop decision support systems. However unknown future weather conditions remains the key point of accurate yield forecast. This paper presents the results of a preliminary study that aims to supply the unknown future with stochastically generated climatic conditions. Coupling the methodology with appropriate decision rules led to a generic decision support system able to guide the N management practices.