DEVELOPING SIMPLE, OPERATIONAL, CONSISTENT NDVI-VEGETATION MODELS BY APPLYING ENVIRONMENTAL AND CLIMATIC INFORMATION - PART I - ASSESSMENT OF NET PRIMARY PRODUCTION

Multiple linear regression techniques have been used to assess rangela nd Net Primary Production (NPP) in Senegal. The biophysical principles of assessing NPP as a function of the Absorbed Photosynthetic Active Radiation (APAR) accumulated during the growing season, multiplied by the efficient coefficient (epsilon), have been used for the work. Accu mulated APAR can be substituted with the seasonal integration of NDVI multiplied by the Photosynthetic Active Radiation (PAR). However, a li near regression showed that the NDVI alone was a better and more consi stent estimator of rangeland NPP compared with the use of NDVI combine d with the PAR data. A first characterization of environmental and cli matic variability was attempted. Soil and vegetation information was u sed and GIS interpolation tools were exploited to create spatial surfa ces of environmental and climatic variables from point observations. A multiple linear regression with the independent variables: integrated NDVI, percentage tree cover and mean surface temperature, improved si gnificantly the level of explained NPP (from 71 to 82 per cent). This model was valid for two years of data.