Bottom-up estimates of population exposure and damage from floods: A case study on Malawi's 2015 floods

Despite the growing impacts of floods worldwide, data on their economic damages remain limited. For the universe of flood events that occur worldwide, only a subset enter standard databases such as the Emergency Events Database, and only a fraction have damage estimates. Filling this gap matters for local flood risk assessment, adaptation planning, and better accounting of climate change effects, particularly in developing countries. This paper adapts and extends the Climate Risk and Vulnerability Assessment Framework and Toolbox (CLIMAAX) workflow to estimate localised population exposure and direct damage from flood events. The workflow integrates the three components of flood risk: hazard, with water depths reconstructed from a satellite-derived flood extent using terrain data; exposure, from building footprints and gridded population layers; and vulnerability, through depth-damage functions. Applied to Malawi’s 2015 floods, the workflow yields rapid, localised estimates of structural damage and exposed populations in a data-scarce setting. A Radarsat-2 flood extent generated from a humanitarian emergency activation underpins the preferred estimate. The analysis indicates approximately 124,000 people exposed and direct structural damage to buildings and their contents of US$312 million (2015 US$). At matched sectoral scope, structure-only estimates of US$117–166 million bracket the US$136 million housing-sector damage reported in Malawi’s Post-Disaster Needs Assessment. Sensitivity analysis, including calibration against unit costs derived from the assessment, shows that valuation assumptions shift aggregate damages while leaving the spatial pattern of losses broadly unchanged. These bottom-up estimates complement ground-truthed post-disaster assessments and media reporting and help paint a richer picture of flood impacts.