Wet and dry spells in Senegal: Evaluation of satellite-based and model re-analysis rainfall estimates

Abstract

Abstract. In this study, wet and dry spells over Senegal provided by four datasets based on satellite data (TRMM-3B42 V7, TAMSAT V3, CMORPH V1.0, CHIRPS V2.0), two fully based on (re)analyses (NCEP-CFSR, ERA5) and one was fully based on gauge observations (CPC Unified V1.0/RT) are compared with respect to observation datasets derived from 65 rain gauge network. All datasets were converted to the same temporal and spatial scales with 0.25 × 0.25 as resolution. Ordinary kriging (OK) and block kriging (BK) were used for the spatial interpolation of the gauge data. Despite a spatial coherence of the seasonal rainfall accumulation between all products, more variability with intra-seasonal features are shown in this paper. The seasonal cycle of dry days shows that TRMM, CPC, ERA5, NCEP and OK record more dry days (from 45 % to 55 % of dry days in August) while TAMSAT, CHIRPS, CMORPH and BK record less dry day (from 40 % to 30 % of dry days in August). All datasets highlighted an agreement that dry spell indicator underscore often false start and early cessation of the rainy Season in Senegal. Although, it can rarely occurs during intensification of West African monsoon (August–September). The most contrast is found on the detection of wet indicators intensity. Wet spell (defined as period with precipitation higher than a certain percentile of historical precipitation) are more severe in OK and TRMM than in other datasets. However, a great similarity is shown on their temporal frequencies.