Extreme rainfall events in the Rio de Janeiro city (Brazil): description and a numerical sensitivity case study

Forecasting extreme rainfall events with high reliability has become increasingly important over the last decades, given their expressive impacts on society and properties. This study initially described the spatiotemporal patterns of rainfall climatology, identified extreme rainfall events, and categorized the meteorological systems related to the precipitation regimes in Rio de Janeiro city. Secondly, we evaluated the extreme rainfall event on 08–09 April 2019 over the Rio de Janeiro city and made a numerical sensitivity case study using the Weather, Research and Forecasting (WRF). For the numerical sensitivity study, we considered the combination of eight microphysics and eight cumulus schemes (totalizing 64 simulations) and four horizontal one-way grid nesting domains (27 km, 09 km, 03 km, and 01 km). The results showed that The New Tiedtke cumulus scheme might have presented the best performance from the quantitative and qualitative evaluation. We verified from the cloud microphysics schemes that the WRF single-moment 5-class and WRF single-moment 6-class schemes also presented better spatial characterization and intensity of rainfall. Lastly, this work sought to characterize the challenges faced by operational forecasters related to the extreme rainfall and natural hazards imminence warnings. It is also worth mentioning that the physical results and discussion made could present some intrinsic bias due to the only one extreme event analyzed. Thus, as future work, the simulations of other extreme rainfall events are expected to guarantee the reliability of the possible outcomes and their use by operational forecasters.