Evaluation of Model-based Soil Moisture Drought Monitoring over Three Key Regions in China

Land surface models (LSMs) have been widely used for an objective soil moisture drought monitoring. However, large uncertainties exist due to different parameterizations in LSMs. This study aims to evaluate the capability of soil moisture drought monitoring over three key regions in China by using the Noah LSM from the Global Land Data Assimilation System version 2 (GLDASv2), and the Community Atmosphere Biosphere Land Exchange (CABLE) model that is currently implemented at China Meteorological Administration for an operational drought monitor. The modeled soil moisture drought indices are verified against the standardized precipitation index (SPI) that serves as a reference drought indicator over Northern China (NC), Northwestern China (NWC) and Southwestern (SWC) China from 1961 to 2010. The results show that the precipitation forcing data that drive both LSMs have high accuracy as compared with local observations. GLDASv2/Noah outperforms CABLE in capturing soil moisture anomalies and variability, but both show good correlations with SPI3 in the NC and NWC. The 2000 and 2001 spring drought cases are selected to compare the model drought categories against SPI3 drought category, where GLDASv2/Noah is slightly better than CABLE. This work demonstrates that the choice of LSM is crucial for monitoring soil moisture drought, and the GLDASv2/Noah LSM can be a good candidate for the development of a new drought monitoring system in China.