In this study, we analyze a set of agroclimatological indices across West Africa and assess their projected changes for the future. We apply the regional climate model CCLM (COnsortium for Small-scale MOdelling in CLimate Mode) with a high spatial resolution of 0.11 degrees (approximately 12 km) under current (1981-2010) and future climate conditions, based on the emission scenario RCP4.5. The focus is on purely rainfall-based indices, that is, the onset (ORS), the cessation (CRS), and the length of the rainy season (LRS) and the joint rainfall- and temperature-based indices, that is, growing degree days (GDD) and the water availability (WAV), derived for maize, sorghum, and pearl millet across Guinea, Savanna, and Sahel. For the present, in general, the CCLM compares well to observations, represented by three different products. However, CCLM shows limitations in the representation of the CRS over Guinea with a delay of >30 days and the GDD and WAV over Sahel with biases up to 30% and 70% for all crops. For the future climate projections, ORS, CRS, and LRS are expected to be delayed up to 2 weeks for most regions, in particular for the period 2071-2100. The GDD is expected to increase by around 8% till 2021-2050 and by around 5% till 2071-2100 for all crops. The WAV is expected to be decreased by up to 10% in 2021-2050, and by up to 24% in 2071-2100 in Sahel, and <12% over Guinea and Savanna in both periods. In particular, we evaluate the added value of the high-resolution CCLM information for decision support in agricultural management.

http://dx.doi.org/10.1029/2018jd028536

Journal Article