Realistic land cover maps

This part of the work package, due to be completed by the end of the first year, involves the development of realistic scenarios of land use change in sub-Saharan Africa. For the countries of the Sahel (Burkina Faso, Senegal, Niger, northern Niger, Mali and Chad), we have been working closely with the University of Louvain in Belgium. They have written a land use model specifically for the countries of this region which partitions the national area into fuel-wood areas, cropland for subsistence and export, fallow and grazing land. The demands of the population for fuel-wood and crops under conditions of variable rainfall drives changes in land use. We have constructed a method of converting the land use classes into functional typesã - broadleaf trees, shrubs, C4 grasses and crops, and bare soil. These types are used directly by MOSES2, the surface scheme in the Unified Model. Fractional coverage maps of the MOSES types over the Sahel have now been constructed for the years 1961, 1996, 2015 and 2030.

For sub-Sahelian countries an alternative approach is required, due to the different drivers in land use change. Thus far we have identified data sources that we shall use to estimate rates of deforestation and increases in cropland. These include both remote sensing studies and national statistics. We envisage that fractional coverage maps for the tropical areas of West and Central Africa will be available by the end of the first 12 months of the project.

Report on the first 6 months of progress on Work Package WP3000

NERC - CEH Wallingford

Emma Tate

August 2000

The work plan for Year I of the project involves starting to adapt the hydrological grid model for West Africa, and collating the physical parameters and calibration data.ã

Jeremy Meigh, Chris Taylor and Emma Tate attended the PROMISE kick-off meeting at the University of Reading, from 3^rd to 5^th April 2000.ã At this meeting, climate data requirements for the hydrological model were discussed with the climate modelling partners, and a list of the necessary parameters drawn up to ensure that the climate models produce the parameters at the resolution and time periods needed for the water resources modelling to produce useful results.

Substantial progress has been made on assembling the data needed. Topographic maps of the West African region were assembled, and a modelling area designated.ã From the maps, at the resolution of half degree cells (approximately 50 km by 50 km), flow directions in each cell were identified, then the river network verified. Coverages of the region, national boundaries, grid cells, and flow directions have been set up in the ArcInfo GIS package. The main lakes, reservoirs and wetlands have been identified and their characteristics entered to the modelling input data file. Recent data on irrigated areas in the region have been acquired. There are currently no factors which may prevent us from meeting our Year I milestone.