Remote sensing and geographic information systems methods for global spatiotemporal modeling of biomass burning emissions: Assessment in the African continent

Abstract

A spatially explicit model for analysis of biomass burning emissions is presented. The model, based on that of Seiler and Crutzen [1980], uses satellite images and geographic information systems (GIS) modeling tools to improve the estimation of biomass loads and burning efficiency. The model was assessed in the African continent using the Global Burned Area (GBA-2000) maps derived from SPOT Vegetation by the Joint Research Center. A total amount of 5711.78 and 336.43 Tg CO was estimated from the model. The areas south of the equator were estimated to release 3579.22 and 218.21 Tg CO, while 2132.56 and 118.22 Tg CO were estimated for the Northern Hemisphere. Most of these emissions were generated by two latitude strips: between 3.5° and 11°N, and between 5° and 13°S. Monthly variability shows a clear bimodal temporal behavior, with two maxima in November - February in the northern band and in June - September in the southern band. The effect of biomass loads distribution on gas emissions is clearly shown, with higher gas emissions in the Southern Hemisphere in spite of having lower burned extension. Copyright 2004 by the American Geophysical Union.