Measuring the spatial impact of human activities on ecosystems is an important step towards effectively managing the changes affecting these natural areas. The objective of this study is to determine land-use dynamics and changes in vegetation cover in the Sébi-Ponty watershed in the Dakar region. To this end, we defined seven land-use classes (tree crops, water, market gardening, agricultural areas, shrub savanna, bare soil, and built-up areas). Satellite data and Landsat images from three periods (1984, 2000, and 2016) were carefully selected and analyzed. Auxiliary data (GPS surveys, topographic maps, and interviews) were also used. The analysis of changes in the land-use classes was carried out using the integration of geographic information systems (GIS) with ArcGIS software and remote sensing techniques with ENVI software. The results obtained highlighted strong dynamics within the land-use classes. This dynamic is characterized by an increase in tree farming, market gardening, and construction at the expense of bare soil, shrub savanna, and agricultural areas. During the same period, the shrub savanna has significantly declined in vitality, indicating a trend toward degradation. The values ​​of the confusion matrix and the kappa coefficient confirm this situation. These results highlight the relevance of our integrated approach, which is applicable to other similar studies.

To simulate the propagation of floods in the Senegal River valley in order to lay out the flood risk maps, we used hydraulic methods calculating the flooded surface from the topography of the minor and major beds. To this end, the HECRAS model and GIS (ARCGIS and HEC-GeoRAS) were used on the Bakel-Matam reach. The work was done in steady state and allowed us to map the risk of flooding on this Bakel-Matam section in the Senegal River valley for different floods. to do this work, water levels and flows were used for the calibration and validation of the HEC-RAS model. The Nash criterion was used to optimize the HEC-RAS parameters. The flood extent obtained by this model was exported and visualized in ArcGIS. Thus, the flood mapping results obtained made it possible to detect a certain number of villages at risk of flooding for the hundred-year flow, wich shows that the Senegal River Valley is vulnerable to floods. Moreover, this was demonstrated by the average catastrophic flood which caused flooding that could impact approximately 24,071 people. The characteristics (height, speed and propagation time) for each flood were analyzed. The promising results obtained showed that the couple HECRAS and GIS is an effective tool for mapping floods and could be used as a tool to help planners in planning.