Modelling burned area in Africa

Abstract

Abstract. The simulation of current and projected wildfires is crucial for predicting vegetation as well as pyrogenic emissions in the African continent. This study uses a data-driven approach to parameterize burned area models applicable to dynamic vegetation models (DVMs) and global circulation models (GCMs). Therefore we restricted our analysis to variables for which either projections based on climate scenarios are available, or which are calculated by DVMs and the spatial scale to one degree spatial resolution, a common scale for DVMs as well as GCMs. We used 9 years of data (2000–2008) for the variables tree and herb cover, precipitation over the last dry season, wet season and averaged over the last 2 years, a fire-danger index (the Nesterov index), population density and an annual proportion of area burned derived from the MODIS MCD45A1 product. Since the effect of fires on vegetation depends strongly on burning conditions, the timing of wildfires is of high interest too. We related the seasonal occurrence of wildfires to the Nesterov index and found a lognormal relationship with a maximum at a value of 104. We parameterized two generalized linear models, one with the full variable set (model I) and one (model II) considering only climate variables. All introduced variables resulted in an increase in model performance. Model I correctly predicts the spatial distribution and extent of fire prone areas though the total variability is underrepresented. Model II has a much lower performance in both aspects (correlation coefficient of predicted and observed ratio of burned area: 0.71 model I and 0.58 model II). An application of the models with simulated climate data ranging from 1980 to 2060 resulted in a strong decrease of burned area of ca. 20–25%. Since wildfires are an integral part of land use practices in Africa, this indicates a high loss in areas favourable for food production.