Mapping diurnal cycles of precipitation over China through clustering.

• K-means clustering was used to map diurnal cycles of precipitation amount. • Precipitation events with different durations were mapped separately. • Results can characterize regions of irregular shapes and show more spatial details. • Regional characteristics of diurnal cycles of precipitation in China were revealed. The diurnal cycle of precipitation is a fundamental mode of climatic variability. Based on an hourly 0.1° × 0.1° gauge-satellite merged precipitation dataset, which is provided by the China Meteorological Data Service Center, this study used the K-means clustering algorithm to map the diurnal cycles of precipitation in summer over China. The obtained maps can objectively delineate irregular regions with similar patterns of diurnal cycles, and many of these regions resemble large-scale geomorphic units such as the North China Plain and the valley of the Yarlung Zangbo River, which verifies the importance of macroterrain for diurnal cycles of precipitation. In addition to total precipitation, precipitation events with different durations, which are often related to different types of precipitation (e.g., convective precipitation and stratiform precipitation), were analyzed separately. Compared with the previous studies using predefined region boundaries, the main advantage of this study is that the obtained maps can show more spatial details, through which we can get intuitive and in-depth understandings of the regional characteristics and spatial patterns of the diurnal cycles of precipitation over China. [ABSTRACT FROM AUTHOR]