Anticipating impact: Forecasting the risk of extreme precipitation for emergency mapping

Extreme precipitation is a major trigger of urban and pluvial flooding and frequently acts as a primary or compounding hazard in humanitarian emergencies, triggering and exacerbating displacement, infrastructure damage, and vulnerability in already fragile contexts. Despite advances in disaster preparedness, anticipating the impacts of intense, localised rainfall remains challenging due to forecast biases and uncertainty, as well as the limited integration of hazard information with exposure and vulnerability. These limitations reduce the operational value of existing products for rapid, impact-oriented decision-making, particularly under the compressed timelines that characterise emergency response and anticipatory action.

Here, we present an easy-to-understand, actionable risk index for extreme precipitation that predicts impactful events up to 3 days ahead. The proposed index combines probabilistic estimates of extreme precipitation likelihood with potential impacts, derived from return-period-based forecasts that correct for systematic model biases, to estimate risk. Spatial forecast uncertainty is addressed through a fuzzy neighbourhood approach that accounts for displacement errors as a function of lead time. The resulting risk index is designed for straightforward integration with exposure and contextual information, such as population distribution or critical infrastructure, enabling the identification of regions and populations at risk from extreme precipitation within the forecast horizon. Using activations from the Rapid Mapping (RM) component of the Copernicus Emergency Management Service (CEMS) since 2024, we demonstrate that the index supports the anticipatory pre-tasking of satellite acquisitions for rapid mapping and facilitates timely, targeted emergency response by highlighting where high-impact precipitation is most likely to occur.