Assessing the Forecasting Accuracy of Intense Precipitation Events in Iran Using the WRF Model

This study investigates the accuracy of forecasting and estimating intense precipitation events that caused significant floods in various regions of Iran. The analysis employed the Weather Research and Forecasting (WRF) model and radar data. Five precipitation events, consisting of four daily events and one hourly event, were examined to assess the performance of the WRF model in comparison to ground-based measurements. The results indicate that the WRF model demonstrates good accuracy in predicting rainfall in areas where ground measurements are available. The obtained R-squared values for the two hourly events, 0.56 and 0.66, respectively, along with RMSE values of less than 10.0 mm for both events, suggest a generally satisfactory performance by the WRF model. Notably, when comparing R-squared values, WRF outperformed radar data, achieving 0.42 and 0.57, respectively. However, both the WRF model and radar data tended to underestimate observed precipitation values. The comparison of precipitation patterns revealed that the agreement between the model and radar data depends greatly on the specific region and type of precipitation event. The findings also emphasized the limitations of relying solely on ground measurements for flood monitoring and stressed the importance of improving monitoring networks and employing modeling techniques to enhance the forecast and management of severe precipitation and flooding impacts.