Assessment of WRF numerical model forecasts using different lead time initializations during extreme precipitation events over Macae city, Rio de Janeiro (Brazil)

The ability to forecast extreme precipitation events has become increasingly important over the last decades due to their significant impacts on society and properties. In this context, the reliability degree related to the warnings issues and a possible natural hazard is relevant. Thus, research aimed to evaluate the strengths and weaknesses of numerical forecasts can also be addressed as prevention measures. In this context, first, this study endeavored to identify the extreme precipitation events in Macae city, Brazil. Secondly, evaluate the precipitation forecast using the Weather Research and Forecasting (WRF) model using different lead time initialization and three grids domain (27, 09, and 03 km); and investigate the thermodynamic and dynamic physical processes related to the extreme events found. From the qualitative and quantitative evaluation of the precipitation forecast, it was possible to verify that the 24 h lead time initialization presented the best performance compared to the other ones. The third domain also presented the statistical results of the precipitation forecasts, suggesting that the increased horizontal resolution enabled the model to represent mesoscale physical processes and better use high-resolution input data, such as the mountainous region of Macae city. It was possible to verify the importance of conditional atmospheric instability, moisture offer at lower atmospheric levels, and dynamic trigger mechanisms to originate the extreme rainfall events from the thermodynamic and dynamic parameters. Lastly, we sought to show the challenges of forecasting the extreme rainfall and natural hazards imminence warnings and exploring the daily operational challenges verified in the monitoring centers and civil defenses.

Automated summary

This study focused on identifying extreme precipitation events in a city in Brazil and evaluating precipitation forecasts using the WRF model. Findings showed that increasing horizontal resolution improved model performance and better predicted extreme precipitation events.