Discover Sustainability (Jan 2025)
Flood hazard monitoring and modeling systems for improving climate risk management using machine learning and geospatial models in the Hennops River catchment, Centurion, South Africa
Abstract
Abstract Climate change has adversely affected precipitation patterns, leading to increased flooding. However, in most African countries, conventional methods of flood hazard monitoring have hindered risk-reduction measures due to operational challenges, technological constraints, and data gaps. To address these issues, robust models and Earth observation products that can enhance climate-driven impact assessments need to be widely implemented across the continent. This study aimed to model flood risks within the Hennops River Catchment area in Centurion, South Africa, using Support Vector Machine (SVM), Random Forest (RF), Topographic Wetness Index (TWI), and Normalized Difference Water Index (NDWI) from the period 2016–2022. To achieve this, we obtained Sentinel-2A and Landsat images from the United States Geological Survey Archive and processed them using SVM and RF models, along with TWI and NDWI. The findings indicate that flood frequencies have increased every two years due to climate change, which causes changes in precipitation patterns, intensity, and frequency. Consequently, areas with low elevations ranging from less than 1305–1430 m in the catchment are at a higher risk of flooding because of their proximity to the Hennops River. These locations are more likely to experience severe flooding because they are flat or have a low elevation, causing runoff from the higher ground to accumulate and pose a greater threat to residents. SVM and RF also revealed that a large number of built-up areas contributed to flooding. These models exhibit an average accuracy of > 70 percent. This research improves the flood-hazard understanding and builds resilient communities in and around the Hennops River Catchment.
Keywords
