Modelling spatiotemporal patterns of wildfire risk in the Garden Route District biodiversity hotspots using analytic hierarchy process in South Africa

Abstract

AbstractThe increasing frequency and intensity of wildfires necessitate effective risk management in biodiversity hotspots to mitigate the potential impacts of wildfire hazards. The study utilised a multi-criteria decision analysis-analytic hierarchy process (MCDA-AHP) model to analyse wildfire risk patterns in the Garden Route District (GRD), focusing on biodiversity hotspots in the Western Cape, South Africa. The study used weight assignment and overlay analysis to evaluate wildfire risk factors, including human, topographic, and climatic factors, using data from Landsat and WorldClim from 1991 to 2021. The wildfire risk model was validated using MODIS historical fire data from the Global Forest Watch database and Confusion Matrix, with the burned area extent identified using differenced Normalized Burn Ratio (dNBR). The results show that despite 53% of the most burned area, only 12% was burned, with the high-risk zone accounting for only 11%, indicating a higher likelihood of wildfires spreading and intensifying. The results reveal a weak positive correlation (r = 0.28) between historical fire occurrences and burned areas and a negative correlation (r = − 0.27) between historical fire occurrences and fire seasons. Human and climatic factors significantly impact wildfire propagation in high-risk zones, while topographic factors have less influence, indicating a lower risk of ignition. The findings show that 26% of high-risk zones in the southwestern region dominated GRD biodiversity hotspots, while 27% were in the low-moderate-risk zone in the northwestern parts. The results of this study can aid in assigning fire risk-based criterion weights to support decision-makers in regional and global wildfire prevention and management.