The current paper explores the applicability of the Fuzzy c-means (FCM) clustering, using an adaptive cluster merging, for the problem of detecting the Moroccan coastal upwelling areas in Sea Surface Temperature (SST) Satellite images. The process is started with the application of FCM clustering method to the SST image with a sufficiently large number of clusters for the purpose of labelling the original SST image, which constitute the input of the proposed approach. Then, the number of clusters is reduced successively by merging clusters that are similar with respect to an adaptive threshold criterion. The algorithm is applied and validated using the visual inspection carried out by an oceanographer over a database of 30 SST images, covering the southern Moroccan atlantic coast of the year 2007. The proposed methodology is shown to be promising and reliable for a majority of images used in this study.