Beyond Intensity: How Rainfall Accumulation Patterns Drive Flood Disaster Characteristics in Southern Brazil

This paper investigates the role of extreme rainfall in major flood disasters in Southern Brazil. Disaster Information System Brazil (S2ID) from 1991 to 2024 was consulted to identify the key variables best representing the magnitude of damage, through a hierarchical ranking of the First Principal Component. We hypothesize that accumulated extreme rainfall, rather than isolated intense events, constitutes the primary driver of large-scale flood disasters in the region. We investigated precipitation characteristics and extreme events occurring in the 15 largest flood disaster events. We used historical daily precipitation series data from 13 weather stations. with quality control performed by CoRain software for rain series comparison. Accumulated extreme rainfall contributed to 67% of major flood disasters in southern Brazil. The remaining 33% is linked to isolated extreme rainfall events. These isolated extreme events intensified hydrological disasters through rapid runoff generation and soil saturation feedbacks. While the total rainfall volume showed a decreasing trend in 2 municipalities and an increasing trend in 5, all studied municipalities exhibited an increasing trend in the frequency of extreme rainfall events. These results are consistent with intensification patterns observed in other subtropical regions globally. Our findings indicate that the sequential occurrence of rainfall events, culminating in significant accumulated volumes, emerges as the principal conditioning factor for large-scale hydrological disasters in the region. This scenario is associated with quasi-stationary frontal systems and persistent cyclonic activity, particularly the prolonged influence of the South Atlantic Polar air masses. These findings contribute to understand mid-latitude flood generation mechanisms, with implications for early warning systems and adaptation strategies in subtropical regions worldwide. This graphical abstract synthesizes the spatial, statistical, and climatological dimensions of extreme precipitation across Southern Brazilian municipalities that recorded major flood disasters (with damages exceeding 100 million BRL). Data sources include the Brazilian Disaster Information System (S2ID) for the 1991–2024 period, combined with historical daily precipitation series from 13 weather stations, processed after systematic quality control using the CoRain software. The Analyses utilize correlation matrices and Principal Component Analysis (PCA), establishing total material damage as the most representative variable of disaster magnitude. The Model evaluates trends in precipitation indices (such as r95p and r99p) and examines rainfall characteristics over the preceding 30 days for 137 major events. These cases were categorized by their conditioning typology: intense isolated or accumulated rainfall. Results, visualized through trend maps and pie charts, reveal that 67% of major flood disasters were triggered by accumulated rainfall, whereas 33% resulted from isolated extreme events. The Conclusion highlights that sequential rainfall, driven by quasi-stationary frontal or cyclonic systems, is the primary factor for large-scale disasters. This synthesis advances the understanding of complex climate-disaster interactions, offering essential insights for early warning systems by distinguishing between isolated and accumulated rainfall drivers, thereby fostering risk mitigation and community resilience. Rain-generated flood disasters (1991–2024) in Southern Brazil reveal 30-day accumulated rainfall as a more critical trigger than isolated intense events. Accumulated extreme rainfall causes 67% of major flood disasters, while isolated events account for the remaining 33%. PCA establishes a hierarchical ranking of damage variables, providing a novel framework to quantify hydrological disaster magnitude. Results demonstrate an increasing trend in R95p and R99p extreme rainfall events across Southern Brazil. Consistent rainfall series via CoRain gap-filling enhance early warning systems by distinguishing between isolated and accumulated rainfall triggers.