Under the influence of the well-evidenced global warming, extreme precipitation events are likely to occur more and more frequently. As a result, the urban flood control and waterlogging prevention are under serious threat. The extreme precipitation events can be effectively identified by precipitation concentration index (CI), which is mainly used to describe the inhomogeneity of precipitation. This paper attempts to explore the spatiotemporal variations and the causes of precipitation concentration. Mann-Kendall statistical test and Sen’s slope method were used to detect the temporal patterns of ACI (annual precipitation concentration index), and inverse distance weighted (IDW) interpolation method was used to analyze the spatial change of LCI (long period precipitation concentration index) as well as the temporal variation trend of ACI based on daily precipitation at the 43 stations during 1960-2012 in the Pearl River Basin. Meanwhile, the random forest algorithm (RF) was applied to identify the contributions of the influencing factors. The results show that: (1) The northwest of Pearl River Basin indicates lower LCI while the southeast part indicates higher LCI, which shows that the extreme precipitation events will occur more frequently in the southeast. (2) The interannual variation of ACI in Pearl River Basin is not obvious, the northwest part shows decreasing trend while the southeast part shows increasing trend, the spatial distribution trend is likely to be affected by distance from ocean and altitude. (3) The importance analysis based on RF shows that the east Asian summer monsoon (EASMI) is the most significant factor of precipitation concentration among the 7 factors.