Disaster risk factors and spatiotemporal trends in Africa

Africa’s disaster risk is fueled by vulnerability and lack of coping capacity factors, with specific components mostly missing in the literature. Having exceeded the midterm of the Sendai Framework for Disaster Risk Reduction (2015 to 2030), assessing the trend of disaster risk in Africa is necessary. This study answers two core questions: what are the disaster risk factors (and their interactions) in Africa? What trends and patterns have been observed in the last decade? Thus, this study determines the factors of disaster risk in Africa using random forest machine learning models and a Spatial Stratified Heterogeneity (SSH) technique using Geodetector software. Both analytical procedures gave rise to important factors (>10) of disaster risk in Africa. The interaction between these factors is also explored. Among the 22 variables included in the analyses, only one natural hazard (i.e., flood) is a significant factor, while current and projected violent conflicts are human-hazard factors of disaster risk in Africa. Additional results show the trend, pattern, and hotspots of African countries’ disaster risk in the last decade, based on the Index for Risk Management (INFORM) data. This study provides a broader understanding of disaster risk factors in Africa and their interactions, contributing to the foremost priority of the Sendai Framework for Disaster Risk Reduction. Furthermore, the trends, patterns and hotspots identified in this study show countries that should be prioritised for urgent actions.

Keywords: Africa, disaster risk factors, disaster risk reduction, Random Forest, Sendai framework, Spatial Stratified Heterogeneity