Abstract:
The increasing frequency and intensity of floods in Malawi, particularly in the Shire River Basin, necessitate a systematic evaluation of flood risks. In Malawi, flood analyses are mostly carried out at a macro-level due to the free availability of low-resolution satellite images like the Shuttle Radar Topographic Mission (SRTM). Micro-scale studies in river basins are often neglected due to cost of high-resolution spatial data and difficulties in accessing the affected area. This study assessed flood hazards using Unmanned Aerial Vehicles (UAVs) at a micro-level in Nyathando in Traditional Authority Ndamera, Nsanje District. The specific objectives were to predict flood intensity and assess vulnerable areas to improve disaster preparedness. Purposive sampling was used, and analysis was conducted using the Hydraulic Engineering Center-River Analysis System (HEC-RAS) model, the Soil and Water Assessment Tool (SWAT) model, ArcGIS, R-Studio and semi-automatic methods. Independent variables included topography, rainfall, and land use, while dependent variables were flood extent and depth. The performance of UAV-derived Digital Elevation Models (DEMs) were compared to lower-resolution SRTM data. The findings demonstrated that UAV-based models provide better flood extent and intensity predictions, capturing critical terrain features that could significantly improve flood risk management. Further, Nyathando Village was found to be highly vulnerable to floods, with water depths of up to 2m and flow rates of 1.3 m/s. UAV-collected data was found to be more accurate than SRTM data and therefore better in micro-level assessments, emphasizing the importance of using high-resolution data in accurately assessing flood hazards. The study demonstrated that high-resolution data is crucial for producing reliable flood hazard maps and identifying areas prone to flooding.
