Comparative analysis of five drought indices applied to seasonal drought in Three-River Source Region

　　The Three-River Source Region on the Qinghai-Tibet Plateau is a climatically sensitive area highly vulnerable to natural disasters, with drought being a severe meteorological hazard in this region. Since no universally applicable drought index exists, comparing the evaluation results of different drought indices for seasonal drought in the Three-River Source Region is crucial for accurately monitoring drought. This study utilizes daily meteorological data from 22 weather stations in the Three-River Source Region from 1961 to 2024 to conduct a comparative analysis of five meteorological drought indices: the Meteorological Drought Composite Index (MCI), Relative Moisture Index (MI), Precipitation Anomaly Percentage (PA), Standardized Precipitation Index (SPI), and Standardized Precipitation Evapotranspiration Index (SPEI). The results indicate that all five drought indices show correlations with soil moisture exceeding 0.47 in spring, summer, and autumn, with the strongest correlation in summer and the weakest in autumn. The MCI demonstrates a stronger correlation with soil moisture than the other indices. The MCI, PA, and SPI tend to overlook historical spring drought events and underestimate drought severity, while MI and SPEI outperform MCI, PA, and SPI in accurately identifying spring
　　drought and consecutive spring-summer drought, effectively reflecting the characteristic “nine out of ten springs are dry” in the Three-River Source Region. The five drought indices generally agree in evaluating seasonal drought trends in the region, indicating stronger drought intensity in spring but with a significant alleviating trend. The drought station ratio also reflects a significant reduction in the spatial extent of drought, while no significant trend is observed in other seasons. The findings provide a theoretical basis for monitoring, early warning, and disaster prevention of seasonal drought in the Three-River Source Region.