In response to the extensive development of upland rice on the hillsides of the Malagasy highlands, alternative cropping systems have been designed based on conservation agriculture (CA). As the promotion of CA in smallholder farming systems is still the subject of debate, its potential benefits for smallholder farmers require further assessment. In the context of resource-poor farmers and low-input production systems, nitrogen (N) is a major limiting nutrient. The effects of contrasted cropping systems have been studied on upland rice yield and N uptake in rainfed conditions: conventional tillage (CT) and CA with a mulch of maize or a legume (Stylosanthes or velvet bean). Decision Support Systems for Agrotechnology Transfer (DSSAT) crop growth model was used to quantify the soil N balance according to the season and the cropping system. The lowest yields were obtained in CA with a mulch of maize and were also associated with the lowest crop N uptake. Upland rice yields were higher or equivalent under CA with a legume mulch than under CT cropping systems. The supply of N was considerably higher in CA with a legume mulch than in CT, but due to higher leaching and immobilization in CA, the final contribution of N from the mulch to the crop was reduced although not negligible. DSSAT has been shown to be sufficiently robust and flexible to simulate the soil N balance in contrasting cropping systems. The challenge is now to evaluate the model in less contrasted experimental conditions in order to validate its use for N uptake and yield prediction in support to the optimization and design of new cropping systems.