Assessment of WRF performance in reproducing flash drought dynamics over the Iberian Peninsula

Flash droughts represent a rapidly intensifying form of drought with significant impacts on ecosystems, agriculture, and water resources. Unlike conventional droughts, these events develop over short timescales, often within weeks, leading to abrupt soil moisture depletion. In the Iberian Peninsula (IP), their relevance has increased in recent decades, particularly in southern and southeastern regions, where climate change is enhancing aridity and atmospheric evaporative demand.

This study evaluates the capability of the Weather Research and Forecasting (WRF) model to reproduce the occurrence, duration, and intensity of flash droughts over the IP for the period 1976–2022. Four standardized drought indices have been used: the Standardized Precipitation Index (SPI), the Standardized Precipitation Evapotranspiration Index (SPEI), the Evaporative Demand Drought Index (EDDI), and the Standardized Soil Moisture Index (SSI). Flash drought events are identified based on the occurrence of dry conditions, the rapid development of drought conditions, and the persistence over time.

Model outputs have been evaluated against observational and satellite-based reference datasets (AEMET precipitation and GLEAM evapotranspiration and soil moisture). A statistical regionalization based on principal component analysis and k-means clustering identified five homogeneous climatic regions across the IP, enabling a detailed regional assessment.

Results shown that WRF generally captures the spatial and temporal variability of flash drought characteristics, particularly for SPI and SPEI. The model reproduces interannual variability and long-term trends reasonably well, especially in northern and central regions. However, it tends to underestimate event frequency for EDDI and shows limited skill for SSI. An increasing trend in frequency, duration, and intensity is detected for indices incorporating evaporative demand (SPEI and EDDI), particularly after the early 2000s, highlighting the role of rising atmospheric demand in flash drought dynamics.

Overall, the study demonstrates that WRF is a valuable tool for simulating flash droughts in the IP, although its performance depends on the selected drought index. These findings contribute to improving the understanding and modeling of rapidly developing droughts under current climate conditions.

Acknowledgements: This research has been carried out within the framework of project PID2021-126401OB-I00, funded by MICIU/AEI/10.13039/501100011033 and by ERDF, EU.