Fire hazard modelling in Southern Africa

Abstract: In this study, we analysed fire occurrences in Southern Africa using point pattern analysis methods. The intensity of events varied vary across the region, events interacted across the region. The intensity plot showed that events are more concentrated at latitude -12.00 and longitude 35.00. Ripley’s k-function revealed that events are clustered up to a distance of 14 km. We tested hypothesis that Higher amounts of dry mass productivity (DMP) and the normalised difference vegetation index (NDVI) result in more fire occurrences and more biomass burning, we also hypothesised that dry woodlands in savannah ecosystems result in more fire occurrences as compared to other land cover types in MaxEnt. Results showed DMP, NDVI and land cover types can be used to model fire occurrences with an ACU of 0,760.It also showed that as DMP and NDVI increased fire occurrence probability also increased. More fires are concentrated (0.79) in crop land with woody vegetation and closed grass land cover types. All the countries in Southern Africa have a high fire risky