Evaluation of spatial-temporal distribution of precipitation in mainland China by statistic and clustering methods

Abstract

Precipitation in China has a sharply uneven spatial-temporal distribution, which has caused serious droughts and heavy floods in China. Using statistic and K-means clustering methods, this study systematically assessed the distribution characteristics of precipitation in China from four aspects, including the spatial-temporal distribution of precipitation, the impact of typhoons on precipitation, precipitation clustering, and precipitation extreme value distribution. The results indicates that southern China has more precipitation but it varies widely in different periods during the year. While in the north and northwest inland areas, the amount of precipitation and precipitation fluctuation during the year are very small. Light precipitation is the main form of precipitation in China, reaching 1.146 × 106 mm, but the average precipitation per event is only 1.8 mm. Precipitation decreases as elevation increases, and high-precipitation areas in China are concentrated in 106°E to 119°E, 18°N to 27°N. Typhoon is the main reason for extreme precipitation in China, and the precipitation caused by typhoon in northern areas is especially more likely to cause floods. Through cluster analysis, dividing the precipitation into 8 categories can basically represent the difference in precipitation distribution of China. Precipitation of the same level is distributed in a belt, decreasing from the southeast coast to the northwest inland. Through distribution fitting, Gamma distribution is the most suitable for fitting extreme precipitation distribution. Extreme precipitation cycle of 100 days, 200 days, 300 days, 400 days and 500 days are respectively 252.4 mm, 285.9 mm, 305.3 mm, 319 mm and 329.7 mm, which can reasonably reflect the characteristics of extreme precipitation. Consequently, this study improves the understanding of the spatial-temporal distribution of precipitation in China.