Integration of Remote Sensing Data and G.I.S. in Less-Favoured Areas
[en] Remote Sensing ; SIG ; Land cover change ; Hydrological Model
[en] A conceptual distributed-based model has been developed for flood simulation and forecasting. The model structure consists of a square grid network, in analogy with the raster form of numerical maps. The square-pixel is the analysis unit for hydrologie balance in relation with physi-cal characteristics. Punctual and cartographical data sources are used to characterize each mesh. Aerial photo-graphy and satellite imagery give information about land use. Topographical characteristics are obtained from Digi-tal Terrain Model (DTM). Soil type can be numerized to help in estimating the drainage condition. Hydrological data are distributed in accordance with the Thiessen's polygons or the associated sub-basin. All the data are handled by the GIS part in order to classify the surface runoff potential of all meshes. This procedure allow to consider the physical characteristics and the spatial vari-ability of the parameters at small scale, to manage special devices like storm basins, pumping stations and spillways and to do a quick study of the spatio-temporal evolution of the characteristics. Impact analysis of land cover modi-fications on the river flow regime is therefore possible. In one large rural Malagasy watershed, rince the 1960's, several waterworks were achieved in order to struggle against the flooding of the upstream fertile plain (subject of various developments). Meanwhile, the forest of the basin hillslopes was and still strongly claimed for pluvial rice crop, which is necessary for the subsistence of a fast growing population. The main object of this work is to study the hydrological impacts of these changes of land use on the watershed behaviour and to forecast the basin water regime evolution.
Politique Scientifique Fédérale (Belgique) = Belgian Federal Science Policy