Abstract: Using the daily precipitation data from 94 national stations in Shaanxi province from 1971 to 2021,the method of Detrended Fluctuation Analysis (DFA) was used to determine the extreme precipitation threshod,and the method such as linear trend estimation,moving average,power spectrum analysis,and generalized extreme value distribution were utilized to analyze the spatial and temporal distribution,and probability characteristics of extreme precipitation.The disaster-causing index of heavy rain floods were further established based on the disaster data.The results show that the extreme precipitation thresholds in Shaanxi province generally show a decreasing trend from south to north.The high-value areas are mainly distributed at the southern border of Hanzhong and Ankang.The extreme precipitation frequencies in various parts of Shaanxi province have shown an increasing trend in recent years,being most significant in the northern region.The DFA threshold is slightly higher than that of the 2 a return period.The DFA threshold has a large dispersion due to unique terrain leffect in some areas.There are obvious high-value areas in southern Loess Plateau and northwestern Guanzhong,and the threshold span crossing Qinling Mountains is large.Combined with the extreme precipitation thresholds,the disaster causing rainfall in different regions,and the terrain and regional vulnerability,the recommended thresholds for southern,central,and northern Shaanxi province are 80 mm,55 mm,and 50 mm,respectively,which can provide a reference for disaster prevention and heavy precipitation warning in various regions.

Key words: Extreme precipitation, Decentralized fluctuation analysis, Return period, Disaster causing indicator