Drought monitoring based on TIGGE and distributed hydrological model in Huaihe River Basin, China.

Drought assessment is important for developing measures to reduce agricultural vulnerability and thereby secure the livelihoods of those who depend on agriculture. This study uses four global ensemble weather prediction systems: the China Meteorological Administration (CMA), the European Centre for Medium-Range Weather Forecasts (ECMWF), the UK Met Office (UKMO), and the US National Centres for Environmental Prediction (NCEP) in the THORPEX (The Observing System Research and Predictability Experiment) Interactive Grand Global Ensemble (TIGGE) archive from 2006 to 2010. Based on results from the XXT (the first X denotes Xinanjiang, the second X denotes hybrid, and the T denotes TOPMODEL) distributed hydrological model, as well as soil moisture observations and digital elevation model (DEM) data, synthesized drought grades were established in the Huaihe River Basin of China. To filter out the impact of short-term fluctuations on observed soil moisture, a 30-day moving average was calculated. Use of the moving average significantly improves the correlation between observed soil moisture and simulated soil water deficit depth. Finally, a linear regression model describing the relationship between observed soil moisture and simulated soil water deficit depth was constructed. The deterministic regression coefficient was 0.5872, the correlation coefficient was 0.77, and the regression coefficient was −154.23. The trends in drought grades calculated using soil moisture and soil water deficit depth were found to be the same, and the grades agreed to within one level. Our findings highlight the importance of synthesizing drought grading when assessing drought using different soil moisture indicators in order to obtain a more comprehensive forecast of drought conditions.