Assessing the benefits of weather and seasonal forecasts to millet growers in Niger

West African farmers need to take every year crucial decisions based on some characteristics of the rainy season such as the onset, the offset, the cumulated rainfall or the occurrence of dry spells. Knowing these parameters in advance may therefore be of interest for them. This paper aims at assessing the impacts of 10-days and seasonal forecasts on Niger millet growers' cropping practices and their income. To do so, we apply an ex-ante approach based on the crop model SARRA-H coupled with an economic model that simulates the choice of cropping strategies among 24 available. The approach takes explicitly risk aversion into account and focuses on two different kinds of typical farmers with restricted and large adaptation capacities, in reference to the availability of viable decision options sensitive to forecast information. Results show (i) that 10-days forecasts alone or a combination of 10-days and seasonal forecasts could be quite beneficial for all types of farmers (e.g. median income change with 10-days forecast ranges from +1.8% to +13% according to adaptation possibilities), (ii) that in most of the cases farmers with access to fertilizers and larger arable land benefit more from forecasts and (iii) that even if seasonal forecasts are not really beneficial alone, they are when used in combination with 10-days forecasts. Despite these positive results, one has to underline that income losses may occur in about 20% of cases when using these forecasts, which may be a limiting factor to their effective adoption. (C) 2016 Elsevier B.V. All rights reserved.