A daily drought index-based regional drought forecasting using the Global Forecast System model outputs over China

Drought is a complicated climatic phenomenon occurring in the phase of water scarcity and can have devastating ecological and social impacts. Reliable drought forecasting can benefit various sectors by allowing adequate lead times for drought mitigation efforts. However, drought forecasting, especially on a short-term lead timescale (within two weeks), is still challenging. In this study, we performed regional meteorological drought forecasting at short-term lead times based on a daily drought index (Daily Evapotranspiration Deficit drought Index, DEDI) using modeling outputs from Global Forecast System (GFS) of National Centers for Environmental Prediction in typical climatic regions of China. The skills of the 3-, 5-, 10-, and 14-day lead regional DEDI forecasts were systematically evaluated by comparing with the DEDI derived from the ERA5 reanalysis datasets of European Center for Medium-Range Weather Forecasts which was used as a proxy for observations. Results show that the GFS-based DEDI forecasts can realistically characterize the spatial and temporal variations of the regional drought events that occurred in North China, Southwest China, eastern Northwest China, and Northeast China, and compare favorably to the ERA5-based DEDI. The skill of the GFS-based drought forecasts varies at different lead days. The 3- and 5-day lead forecasts present the best skill in capturing spatio-temporal variability of droughts and generally show the highest correlations and the smallest deviations. For the 10- and 14-day lead regional drought forecasts, the smoothed GFS-based DEDI could bring substantial improvements with remarkably increased correlations and reduced biases, and have skills in forecasting the dryness or wetness tendency. Notably, the GFS-based DEDI has skills in forecasting the occurrence or evolution of the regional drought events, but with a limitation in the drought magnitude evaluation. Overall, these results could be conductive to achieve the reliable regional drought forecasting and effective drought mitigation over China.