Spatiotemporal variability analysis of extreme rainfall events in the city of Petrópolis, Rio de Janeiro, Brazil

ABSTRACT The city of Petrópolis is frequently affected by heavy rainfall events that favor the occurrence of natural disasters, such as landslides and flash floods, causing material and life loss. In this context, knowing spatial and temporal precipitation behavior during extreme events can be a valuable tool for preventing natural hazards. This work seeks to characterize the spatiotemporal distribution of extreme rainfall events in Petrópolis, utilizing rainfall data between 2011 and 2021. It proposes identifying extreme rainfall events using the daily and hourly accumulations to verify spatial and temporal statistical patterns between them. The spatial behavior of the accumulated daily precipitation showed that the central-south region of Petrópolis presented the highest percentiles, which decreased toward the north of the city. From the 99% daily and hourly percentiles, extreme precipitation events might tend to occur in the center-south region of the city with lower intensity but long-lasting precipitation rain cores. In the northern part, however, the hourly rainfall rate tends to occur due to intense and short-lived precipitation. The hourly rainfall records also show that the highest precipitation rates (51 mm/h) preferably happened at the end of the afternoon. Lastly, considering the meteorological systems that lead to extreme rainfall events, it was observed that the South Atlantic Convergence Zone was related to 40% of them. Considering the presented approach, this work aims to reinforce scientific research on the challenges faced by monitoring centers and contribute to the prevention of socio-environmental disasters associated with extreme precipitation events in Petrópolis.