Patch morphological features as indicators of fire regimes: a comparison of tropical savannas in Africa and South America through remote sensing

Abstract

Fire regime is described by a range of important attributes such as fire frequency, intensity, severity, seasonality, size distribution or fire return interval. The complexity of fire shapes have been also identified as a key component of fire regimes, and the fire patch identification from pixel-level information recently emerged as an additional way of providing informative features about fire regimes through the analysis of patch size distribution. The fire patch dynamic is very important in many ecological processes as the post-fire vegetation recolonization process, and the mosaic of burned and unburned areas have been highlight to create a dynamic of different regeneration times and to control fire intensity and size to avoid the occurrence of large and severe fires. This project aim to use the LANDSAT burned area patch dataset and the two most frequently used remote sensing products in the climate assessment community, MODIS MCD45A1and the ESA FIRE_CCI to extend my research on fire regimes by comparing fire patch assemblage in areas of savanna in Brazil and sub-saharian Africa with new metrics based on patch morphological features and the trait-based plant community statistical framework. The first goal of this research is to assess the data quality of moderate resolution of Burned Area (BA) products (ESA FIRE-CCI and MCD45A1v5), compared to high resolution products (Landsat and Sentinel). The second goal is to perform an ecological analysis by using the same statistical framework of fire patch frequency/morphology to compare the Brazilian cerrado savannas with African north equator savannas: 1) to determine if theses savannas have the same fire patch pattern across continents; and 2) to determine also the differences between protected areas and anthropized zones in both continents. Our working hypothesis here is that human activities can locally modify the relative frequency of small and large fires, and that the weather conditions under which fire are set might modify the spread rates that we should capture in their morphological features. (AU)