• 标题：Dominant factors controlling runoff coefficients in karst watersheds
• 关键词：Land use; Geology; Topography; Karst hydrology; Ecohydrology
• 作者：Wen Liu, Zhenwei Li, Jingxuan Zhu, Chaohao Xu, Xianli Xu

全文下载

As one of the largest continuous karst areas in the world, southwest China is experiencing a lack of surface water resources which have negative impacts on the development of the economy and the life of people. Determining the relationship between climate, watershed characteristics and the runoff coefficient is thus important to manage water resources. However, the dominant factors influencing runoff coefficient is still unclear in karst areas (e.g., southwest China). The objective of this study was to examine how the runoff coefficient is related to climate and watershed characteristics (i.e., land use composition, landscape, lithology, soil properties, and topography) for 95 selected karst watersheds. Since climatic and watershed variables are highly codependent, partial least squares regression (PLSR) was used to explain the linkages between the runoff coefficient and 36 selected variables. First-order controls were identified by calculating the variable importance for the projection (VIP). Results showed that the climate and watershed characteristics played an important role in the runoff coefficient and could explain 72% of the total variance in runoff coefficient for the 95 selected watersheds. According to the VIP values, the dominant first-order factors controlling runoff coefficient were as follows: precipitation, elevation, the areal percentage of forestland, largest patch index, Shannon’s diversity index, the areal percentage grassland, aggregation index, gravel content, hypsometric integral and karst coverage. Greater precipitation, gravel content, and the areal percentage of forestland significantly enhanced the runoff coefficient, while the elevation, Shannon’s diversity index, and karst coverage had negative effects on runoff coefficient. As partially eliminating the co-dependency of the variables, the PLSR approach conveniently determined factors controlling runoff coefficient, and could provide quantitative