Recent flood events across different regions of the world have highlighted the need for tools capable of providing rapid and reliable assessments of the impacts associated with hydrometeorological hazards. In Peru, the identification and delineation of flooded areas still rely largely on manual geospatial workflows, which require considerable processing time and may hinder the timely exchange and interoperability of information among institutions involved in disaster risk management. To address these limitations, this study presents a post-event flood assessment system that generates standardized maps of affected areas and technical reports, which are published and visualized through an interactive nationwide WebGIS platform.
The system is designed to support rapid assessment through parallel computing techniques and integrates machine-learning algorithms for the classification of flooded areas using Synthetic Aperture Radar (SAR) imagery acquired by the Sentinel-1 mission in Single Look Complex (SLC) mode. The processing workflow derives intensity and interferometric coherence information, which are used as input variables for the identification and delineation of flood-affected areas. The resulting outputs are subsequently integrated with geospatial datasets to perform exposure analyses and identify potentially affected infrastructure, including educational facilities, healthcare facilities, fire stations, police stations, and residential buildings. Official census data are also incorporated to estimate the potentially exposed population.
The WebGIS platform is structured into operational modules that integrate satellite image processing, geospatial data management and storage, interactive visualization of results, and automated generation of technical reports. Overall, the system provides standardized and accessible information products to support decision-making during emergency response, post-disaster recovery, and disaster risk management at the national scale.