TRANSFORMATION OF A SUBTROPICAL RIVER INTO A HYPERHALINE ESTUARY - THE CASAMANCE RIVER (SENEGAL) - PALEOGEOGRAPHIC IMPLICATIONS

Located in the South of Senegal, in the transitional Guineo-Congolian/ Soudanian zone, the Casamance system is composed of a small river, sup plied by a narrow drainage basin and followed by a 250 km long estuary . The continuous rainfall decrease experienced by the region since the beginning of the century drastically affected the Casamance river. Th is climatic change suddenly accelerated during 1968-1970, which led to a four-stage succession: before 1968-1970 (stage 1, rainfall greater- than-or-equal-to 1300 mm), the Casamance estuary showed the customary negative upstream salinity gradient; between 1968-1970 and 1978 (stage 2, rainfall less-than-or-equal-to 1000 mm), under decreasing rainfall , the Casamance estuary shifted to a hyperhaline ''inverse estuary''. It showed well-defined physical, chemical and biological zonations. Fr esh groundwater resources, supplying the upper reaches of the river du ring the dry season, were progressively depleted. between 1978 and 198 6-1987 (stage 3), the salinity peak of the end of the dry season progr essively rose and shifted upward. The ecological zonation showed the i ncreasing marine influence until 1986; between 1986-1987 and 1990 (sta ge 4), the dry season fresh water input had weakened with the depletio n of groundwater resources. A slight increase in rainfall was insuffic ient to repulse the hyperhaline waters and the restricted area extende d upstream and downstream. The physical, chemical and biological zonat ions strongly widened and flattened. In the future, the return to stag e 1 conditions, by an increase in average rainfall, would take decades . Inevitably it will be much longer than the transformation into a hyp erhaline estuary. Thus a slight decrease of the average rainfall in th e region provoked a drastic change of the hydrological and biological patterns of the ria. The result seems to be a major crisis in the regi on, and might be recorded in the fossil records. We discuss these obse rvations in terms of possible misreadings of paleoenvironments, and in relation with long-term climate trends.