Abstract:Hourly fixed-time precipitation data may underestimate the characteristics of heavy precipitation at the hourly scale. To gain a deeper understanding of the variation patterns of hourly heavy precipitation, utilizing the minutely precipitation observation data collected by 123 national meteorological stations in Shandong Province from June to September during the flood seasons of 2004-2024, we compare the differences between the 60 min sliding window and hourly fixed-time statistical methods in identifying heavy precipitation events, and quantitatively assess the impact of the two methods on the frequency, intensity and spatio-temporal distribution of heavy precipitation. The results indicate that the 60 min sliding window method can capture heavy precipitation events more comprehensively than the hourly fixed-time method. The average frequency of heavy precipitation across the whole province is 124.3 times by means of the 60 min sliding window method, and it is 1.47 times that of the hourly fixed-time method (84.3 times), with the ratio of each station ranging between 1.31 and 1.81. The maximum precipitation captured by the 60 min sliding window method is 10.0 mm more on average, exceeding 30.0 mm for some stations. The spatial distributions of the heavy precipitation by two methods are generally in line, but the 60 min sliding window method reveals even greater spatial variability and stronger extremity of heavy precipitation. The 60 min sliding window method also more clearly shows the increasing trends in both frequency and intensity of heavy precipitation, with climatic tendency rates being 1.291 times per decade and 4.514 mm per decade, respectively. Moreover, the monthly and daily variation characteristics of heavy precipitation are more distinct by using the 60 min sliding window method, particularly in the high-frequency periods from July to August and from afternoon to early morning. The traditional hourly fixed-time statistical mehtod underestimates the frequency and extreme intensity of short-duration heavy precipitation. In contrast, the 60 min sliding window method can provide a more accurate representation of heavy precipitation characteristics, so it is of great application values for improving the accuracy of rainstorm monitoring and early warning capabilities for disasters.