西南地区的极端降水时空演变特征研究

Study on the Spatiotemporal Evolution Characteristics of Extreme Precipitation in Southwest China

摘要: 基于西南地区1960~2023年高分辨率逐日降水格点数据，选取最大1日降水量(Rx1day)、持续干燥日数(CDD)、持续湿润日数(CWD)和中雨日数(R10)等极端降水指数，采用Theil-Sen斜率估计、Mann-Kendall趋势检验及Hurst指数等方法，系统分析了西南地区极端降水的时空演变特征及未来趋势。结果表明：(1) 西南地区极端降水空间分布呈显著异质性，整体呈现东南高、西北低的格局，Rx1day高值区集中在四川盆地西南部、贵州东部和广西北部；(2) 时间变化上，Rx1day呈显著上升趋势(0.071 mm/a, p < 0.05)，CDD波动上升(0.043 day/a)，CWD呈显著下降趋势(−0.041 day/a, p < 0.05)，降水过程趋于“碎片化”；(3) Hurst指数分析显示各指数均具有强持续性(H = 0.71~0.75)，表明极端降水增强与干旱延长的趋势难以逆转；(4) 西南地区极端降水呈现“强度增强、历时缩短、旱涝急转”的显著特征，未来极端降水风险将进一步累积。研究结果可为区域水资源管理、极端气候事件风险评估及防灾减灾对策制定提供科学依据。

Abstract: Based on high-resolution daily precipitation gridded data from 1960 to 2023 in Southwest China, this study selects extreme precipitation indices such as maximum 1-day precipitation (Rx1day), consecutive dry days (CDD), consecutive wet days (CWD), and the number of moderate precipitation days (R10). Methods including Theil-Sen slope estimation, Mann-Kendall trend test, and Hurst index are employed to systematically analyze the spatiotemporal evolution characteristics and future trends of extreme precipitation in the region. The results indicate that: (1) The spatial distribution of extreme precipitation in Southwest China exhibits significant heterogeneity, generally presenting a pattern of high in the southeast and low in the northwest. High-value areas of Rx1day are concentrated in Southwestern Sichuan Basin, Eastern Guizhou and Northern Guangxi; (2) Temporally, Rx1day shows a significant increasing trend (0.071 mm/a, p < 0.05), CDD fluctuates upward (0.043 day/a), while CWD demonstrates a significant decreasing trend (−0.041 day/a, p < 0.05), suggesting a trend towards more “fragmented” precipitation processes; (3) Hurst index analysis reveals strong persistence for all indices (H = 0.71~0.75), indicating that the trends of intensified extreme precipitation and prolonged drought are difficult to reverse; (4) Extreme precipitation in Southwest China exhibits notable characteristics of “intensifying intensity, shortening duration, and rapid transitions between drought and flood”, with future risks of extreme precipitation expected to accumulate further. The findings can provide a scientific basis for regional water resource management, risk assessment of extreme climate events, and the formulation of disaster prevention and mitigation strategies.