Climate variability in Senegal (West Africa) is examined, with emphasis on the seasonal rainfall. This study is especially concerned with the descriptive characterization of rainfall patterns in Senegal over 28 years, the correlation between seasonal rainfall variability and SST patterns, and the linkages between crop productions, rainfall and SST. The goal is to come up with a predictability of rainfall and crop production in Senegal. The analysis is based on based on rainfall amounts, number of rainy days and temperature data from 1971 to 1998 obtained from the Senegalese National Office of Meteorology, monthly SSR data which consist in NOOA Extended Reconstructed SST (Smith and Reynolds), various rainfall indices (Lamb, Nicholson, and Janowiak), SOI (Australian Bureau of Meteorology) and NAO indices, and crop data (peanut and millet, for the same period as rainfall) obtained from the “Centre de Suivi Ecologique” of Dakar (Senegal) and the Department of Statistics (Dakar, Senegal). Geostatistical Analyst and ArcGIS Spatial Analyst are used to map monthly and seasonal rainfall and temperature distributions over Senegal. Time series are analyzed for linear trends; the slopes of the series for each station are used to map the spatial trends of climate variability. An EOF analysis of rainfall data is performed for the entire wet season and then for each month of the wet season, and ArcGIS Spatial Analyst is used to map the spatial patterns. The consistency of the modes is verified by clustering ‘high’ and ‘low’ loadings and reviewing the time series. Correlations are examined between rainfall EOF modes and Atlantic and Pacific SSTs to identify the marine areas from which rainfall predictability over Senegal can be assessed. To verify how rainfall in Senegal compares with that of the whole Sahel, a correlation is performed between the standardized rainfall anomaly for Senegal and other well-known indices (Lamb, Nicholson). The first EOF mode explains most of the variance: for rainfall (JJAS), the first three modes represent 43.3%, 8.6% and 8.4%. For the number of rainy days (JJAS): 64.7%, 8.6% and 7.4%. For temperatures: 50.3%, 14.7% and 11.2%. Associated time series oscillate on interannual time scales. The rainfall variability could be related to SOI. Our results indicate that it may be important but not a dominant factor: the highest correlation is found in July (0.38%). However for 13 stations out of 20, ENSO years coincide with negative anomalies. A significant correlation exists between standardized rainfall anomalies for Senegal and other rainfall indices: Lamb (0.67 for the averaged AMJJASO index); Nicholson (0.67 for August and 0.60 for September). The correlation with NAO index is weak: the highest value (in July) is only 0.35). Also, there is no significant correlation between rainfall EOF patterns for Senegal and Atlantic SSTs, except for the month of July that may provide some predictability for the wet season in Senegal. The most significant correlations are found in the Pacific Ocean: positive correlations between the North central tropical Pacific and Senegalese rainfall (0.55) from January to March (lag correlation); this area shifts westward from April to July; negative correlations with the equatorial and South Pacific (up to 0.60 in July). The Pacific Ocean therefore presents the best chances of predictability for rainfall in Senegal. Investigations are underway to use Pacific SSTs for crop production predictability in Senegal.