Abstract
The spatiotemporal pattern of precipitation concentration is a key factor in a hydrological analysis. This study attempts to explore the spatiotemporal variations and the causes of the variations in precipitation concentration. Firstly, the Mann-Kendall statistical test and Sen’s slope method are used to detect the temporal patterns of the annual precipitation concentration index (ACI), and the inverse distance weighted (IDW) interpolation method is used to analyze the spatial changes in the long period precipitation concentration index (LCI) as well as the temporal variation trend of the ACI based on daily precipitation at 42 sites during 1960–2012 in the Pearl River basin. Secondly, a possible link between precipitation concentration and elevation is investigated by calculating the correlation coefficient between LCI and elevation. Finally, the random forest algorithm (RF) is applied to identify the contributions of seven associated circulation influencing factors on the CI. The results show that: 1) The northwest of the Pearl River basin that is further from the sea and has a higher elevation, exhibits a lower LCI. The southeast part that is closer to the sea and has a lower elevation exhibits a higher LCI, indicating that extreme precipitation events will occur more frequently at lower elevations and closer to the sea areas. 2) The interannual variation of the ACI in the Pearl River basin is not obvious; the northwest part exhibits a decreasing trend while the southeast part has an increasing trend, and the spatial distribution of the ACI’s change trend is likely affected by elevation and the distances from the sea. 3) The precipitation concentration is negatively correlated with elevation with a correlation coefficient of −0.92 at the 99% confidence level. 4) The significance analysis based on RF shows that the East Asian Summer Monsoon (EASM) is the most significant factor affecting the precipitation concentration of the Pearl River basin among seven associated circulation influencing factors.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have