Evaluating hydro-climate extreme indices from a regional climate model: A case study for the present climate in the Casamance river basin, southern Senegal.

Abstract

This paper presents the ability of the Rossby Center Regional Climate Model (RCA4) driven by nine global circulation models of the Coupled Models Intercomparison Project Phase 5 (CMIP5) to reproduce 10 hydro-climate extremes indices (HEIs) over Casamance river basin in southern Senegal, West Sahel for the present period of 1950 to 2000 using the data of four rain gauges stations. The temporal and spatial analysis highlights different behavior between the value and the variability of HEIs, but also between the western and the eastern part of the basin based on RCA4 outputs. Duration indices seem to be consistently overestimated by the RCA4 model with large uncertainty and variability (e.g. consecutive wet days), whereas the absolute indices are underestimated with uncertainties more pronounced in the west of the river basin. The heavy rain days are overestimated and very heavy rain days are underestimated as known in the literature. The RCA4 fails to well represent the daily precipitation intensity in the Casamance river basin except in one station (Bignona). However, the RCA4 model frames relatively well the indices based on the percentile threshold (R95PTOT, R99PTOT) but with some fluctuations. Overall, the annual total wet days seems to be better simulated by the RCA4 model in the western part. As conclusion, the RCA4 model exhibits relatively good or poor performance in representing extreme indices depending on the locations within the river basin but also on the range periods. These results confirm the difficulties for models in extreme climate representation.