The level of Lake Malawi has varied widely throughout this century, and has risen dramatically in the period [1976][1977][1978][1979][1980]. The causes of the change in level have been investigated using simple numerical models based on the lake's water balance. It has been shown that the recorded changes in rainfall were sufficient to have caused the observed changes in lake level, and that man-made changes in runoff and outflow have been comparatively unimportant. Variationsde niveau du lac Malawi RESUME Le niveau du lac Malawi a varié largement au cours de ce siècle et a monté de façon dramatique pendant la période 1976-1980. Les raisons de cette élévation du niveau ont été recherchées en utilisant des modèles numériques simples basés sur le bilan hydrologique du lac. On a montré que la modification des précipitations qui a été enregistrée était suffisante pour avoir provoqué les modifications du niveau du lac qui ont été observées et que les variations dans l'écoulement et le débit sortant du lac résultant de l'activité humaine ne sont pas importants en comparaison.

No abstract

A rainfall and run-off model has been developed which is based on a rectangular grid cell network. The run-off in each cell is estimated empirically and is routed through the network to the principal drainage lines. Information on the land cover in each 50 x 50m cell is interpreted from remotely sensed data, gathered by an orbiting satellite. Values of topographic parameters for each cell are derived from a digital model of the terrain. Other more conventional data are acquired by digitizing maps. The large amounts of data required in this approach are handled by a geographical information system (GIs), which is a management system specially designed for data with spatial characteristics. A GIS has functions which allow the user to combine the many layers of data within the GIS arithmetically or logically to provide fresh information. At the current stage of development the GIS provides estimates of the United States Department of Agriculture Soil Conservation Service curve number for estimating the run-off in each cell. An analysis of the digital terrain model defines the pathway taken by run-off from each cell and the run-off is routed from cell to cell as far as the principal drainage lines, where a conventional hydraulic analysis is used. The model benefits from a simple and robust approach to run-off estimation distributed on a very fine grid. An important feature of the model is its ability to use historical satellite images to estimate changes in the catchment's response over a period of time. Future changes in land use may also be easily incorporated into the GIs. The model is being tested on data from the Tywi catchment in West Wales, UK, which had undergone large changes in land cover due to forestation.

No abstract

No abstract