Spatiotemporal analysis of official records of disasters induced by natural hazards in Brazil from 2013 to 2024

Official declarations of disasters induced by natural hazards (DINH) offer an operational view of risk, but they do not measure hazard frequency; they record when impacts exceed institutional recognition thresholds shaped by exposure, vulnerability, reporting practices, and governance. This study analyzes federally recognized municipal DINH records in Brazil (2013–2024) from the National Civil Defense Digital Atlas, applying a multi-dimensional diagnostic framework that integrates national spatial mapping, seasonal analysis, local spatial clustering, and temporal trend and change-point detection. DINH recognition is strongly organized in space: the Northeast concentrates 47% of all records, followed by the South (26%) and Southeast (17%), while the Central-West and North contribute 6% and 4%. Municipal DINH signatures are persistent, dominated by drought and dry season (50.2% of municipalities) and intense rainfall (43.1%). Economic impacts depart from frequency patterns, with the South accounting for 40.6% of total losses, followed by the Northeast (30.7%) and Southeast (18.5%), and with losses concentrated in a few high-impact years (2022: 16.5%; 2023: 12.9%). Nationally, DINH counts show no robust monotonic trend but a step change around 2020 raises the mean annual number of records from approximately 1600 to 2350. Intense rainfall provides the clearest hydrometeorological signal, with a statistically significant positive trend and a structural shift around 2020. Treating declarations as socio-institutional signals rather than direct measures of hazard frequency, this diagnostic approach supports disaster risk reduction monitoring, preparedness, and adaptation planning grounded in official impact records in Brazil.