The Sahel region is known for the high vulnerability of its agriculture to climate variability. Early warning systems that make use of agrometerological forecasts are one of the coping strategies developed by policy makers. However, the predictive quality of the tools and methods used needs improvement. In order to address some of these challenges, we conducted agronomic trials and on-farm surveys to adapt the SARRAH (Syst`eme d'Analyse R´egionale des Risques Agroclimatiques, version H) crop simulation model, and also evaluated it in farmers' field conditions. The farmers' practices such as sowing dates and densities, fertilizer use and yields potentials of the millet and sorghum crops were characterized under different climatic conditions.