This paper discusses the efforts of developing a locally viable scientific tool for monitoring the flood-prone Kano plains of the Nyando river basin and providing early warning mechanisms for floods in the eastern Lake Victoria basin where perennial floods have been causing a lot of human suffering.

The study describes a simple hydrologic model for the Nyando basin, developed to use a GIS user-interface and integrated with a geospatial database of the area so that it can be used as a flood management tool. The method involves the use of a Digital Elevation Model (DEM) of the area together with the soil types, land cover and daily satellite-derived rainfall estimation, as well as existing rainfall data to develop a GIS-based hydrologic model which can be used in determining the areas affected by floods and forecasting areas likely to be flooded.

The historical daily rainfall data are used to validate the satellite rainfall estimates (RFE). The DEM is used to derive the runoff characteristics. For the purposes of modelling the hydrologic processes using GIS, the area is divided into small cells representing the spatial distribution of the parameters controlling surface runoff i.e. topography, surface roughness, soil infiltration and rainfall.

GIS provides the tools for visual interpretation and evaluation of flood distribution and offers new opportunities to develop and implement a user-friendly, interactive decision support system for flood forecasting and monitoring using dynamic spatial modelling.