Comparative analysis of drought indices in the tropical zones of China

Abstract

The intensity and frequency of drought are increasing in the tropical zone of China under global warming, and accurate assessment of drought severity and duration is critical for sustainable ecosystem management. Previous studies usually rely on one or more drought indices calculated from meteorological station or reanalysis data. However, the assessment results based on these drought indices are not consistent, which can be due to the differences in data sources and index parameters. In this study, we aim to identify the optimal dataset and drought index, and accurately evaluate the drought severity and drought duration in the tropical zone of China. We assessed the accuracy of five drought indices, namely Precipitation Anomaly in Percentage (PA), Relative Moisture Index (MI), Standardized Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI) and Meteorological Drought Composite Index (MCI), calculated from meteorological station data and the China Meteorological Forcing Dataset (CMFD) with respect to drought records compiled by local government. Results indicate that the drought index calculated based on meteorological station data can better match the government-compiled drought records than CMFD. MI is the optimal index for drought severity and duration assessment in study area, especially for winter-spring drought and severe drought, followed by PA. The normalized bell-shaped line of fitted precipitation in winter and spring is biased towards the less rainy side in SPI calculations, which leads to more underestimation even for officially recommended MCI, and actual water supply are also misrepresented in SPEI calculations. This study offers valuable insights for policymakers to use optimal dataset and drought index to accurately assess the drought events, and take effective measures to alleviate its impact on tropical ecosystems in China.