We examine the ability of an ensemble of 10 Regional Climate Models (RCMs), driven by ERA-Interim reanalysis, in skillfully reproducing key features of present-day precipitation and temperature (1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008) over West Africa. We explore a wide range of time scales spanning seasonal climatologies, annual cycles and interannual variability, and a number of spatial scales covering the Sahel, the Gulf of Guinea and the entire West Africa. We find that the RCMs show acceptable performance in simulating the spatial distribution of the main precipitation and temperature features. The occurrence of the West African Monsoon jump, the intensification and northward shift of the Saharan Heat Low (SHL), during the course of the year, are shown to be realistic in most RCMs. They also capture the mean annual cycle of precipitation and temperature, including, single and double-peaked rainy seasons, in terms of timing and amplitude over the homogeneous sub-regions. However, we should emphasize that the RCMs exhibit some biases, which vary considerably in both magnitude and spatial extent from model to model. The interannual variability of seasonal anomalies is best reproduced in temperature rather than precipitation. The ensemble mean considerably improves the skill of most of the individual RCMs. This highlights the importance of performing multi-model assessment in properly estimating the response of the West African climate to global warming at seasonal, annual and interannual time scales.

No abstract

Box 11.1 (continued) Box 11.1, Figure 2 | A schematic illustrating the progression from an initial-value based prediction at short time scales to the forced boundary-value problem of climate projection at long time scales. Decadal prediction occupies the middle ground between the two. (Based on Meehl et al., 2009b.)

The evaluation of a high‐resolution simulation at 0.11° (12 km) with the COnsortium for Small‐scale MOdelling in CLimate Mode (CCLM) regional climate model, applied over West Africa, is presented. This simulation is nested in a CCLM run at resolution of 0.44°, driven with boundary forcing data from the ERA‐Interim reanalysis, and covers the period from 1981 to 2010. The simulated temperature and precipitation are evaluated using three selected climate indices for temperature and eight indices for precipitation in five different regions against a new, daily precipitation climatology covering West Africa and against other state of the art climatologies. The obtained results indicate that CCLM is able to reproduce the observed major climate characteristics including the West African Monsoon within the range of comparable regional climate modeling evaluations studies, but substantial uncertainties remain, especially in the Sahel zone. The seasonal mean temperature bias for the rainy season from June to September ranges from −0.8°C to −1.1°C. The CCLM simulations also underestimate the observed precipitation with biases in precipitation reaching −10% in the high‐resolution and −20% in the low‐resolution model runs. CCLM extends the monsoon precipitation belt too far north, which results in an overestimation of precipitation in the Sahel zone of up to 60%. In the coastal zone, the precipitation is underestimated by up to −90%. These biases in precipitation amounts are associated with errors in the precipitation seasonality. The added value of the higher resolution of the nested run is reflected in a smaller bias in extreme precipitation statistics with respect to the reference data.

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 ∘ (approximately 12 km) under current 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. Key Points:• CCLM model results were evaluated in terms of agricultural indices for cropping of maize, sorghum, and pearl millet across three different agroecological zones in West Africa • High-resolution future climate projections indicate a shortening of the rainy seasons by approximately 2 weeks in the Sahel region till the end of the century • The projected decreases (increases) of rainfall (temperature) over the Sahel region would negatively impact on water availability (WAV) and growing degree days (GDD)

Climate change impacts negatively vulnerable regions such as West African countries like Senegal, through an increase of climate extremes. The objectives of this study is to analyze the future evolution of the extreme temperature events and their impacts on human health in Senegal during the cold (DJF), hot (MAM) and wet seasons (JAS) under the greenhouse gas scenarios RCP4.5 and RCP8.5 using Climate projections of five (5) regional climate models (RCMs) of the Coordinated Regional Climate Downscaling Experiment (CORDEX) program. The results show that the biases of the RCMs are globally low especially their ensemble mean of the RCMs. This ensemble mean was afterwards considered in the analysis of the climate extremes in the near (2021-2050) and far future (2071-2100). When considering the near future, the frequency of the hot nights (Tn90p) increases under the scenario RCP8.5 (up to 90%) during the rainy season in the south of the country. As for the percentage of the hot days (Tx90p), it may reach approximately 70% under the scenario RCP8.5 in DJF over the majority of the country. Moreover, a strong increase of Tn90p and Tx90p is also diagnosed during the far future with values exceeding 80% over most parts of the country. Concerning the evolution of the heat wave magnitude index-daily, the ensemble mean of all models shows that the heat waves are more severe in MAM and JAS under both scenarios during the near and the far future over most parts of the country. To estimate the potential impacts of this heat stress on the human health, the heat index and the humidex are used. The analysis of the heat index shows that the sanitary risks are more perceptible over the whole country during the rainy season with values reaching the symptom band II for both scenarios during the far future. As for the humidex, it is characterized by a gradual increase from the historical period to the far future. This analysis highlights the fact that appropriate adaptation measures should be considered to tackle efficiently the increase of temperature extremes which may impact negatively the human health.

In this paper, we use simulations from the Swedish Meteorological and Hydrological Institute (SMHI) regional climate model (RCM) version 3.5 (SMHI-RCA3.5) following a multi-GCM boundary forcing approach. The model is run at a horizontal resolution of 50 km on the CORDEX-AFRICA domain. Key characteristics of precipitation, mean temperature and its extremes (minimum and maximum temperature) have been investigated over West Africa and on 3 designed sub-domains, the Sahel (SAH), the Senegal-Gambia ( The regional climate model RCA, forced by the reanalysis ERA-Interim, 6 CMIP5 GCMs and their ensemble, reproduces realistically the climatology of precipitation and temperatures over West Africa. Compared to observed datasets GPCP for precipitation and CRU for temperature, the ensemble outperforms both other GCMs and the verification model (ERA-Interim). The major biases in precipitation are the early onset over the Sahel and the little dry season (LDS), from mid-July to mid-September over the Gulf of Guinea, and a few models either overestimate and/or reflect rather poorly. The strong warming in extreme temperatures (minimum and maximum) combined with the drying mainly over Western Sahel (SEN) found in this study will very likely impact notably a vital sector like agriculture, both during the near and far terms.

No abstract