摘 要: |
Extreme precipitation events exert increasingly profound effects on public health and ecosystems under climate change and urbanization. In this study, the nonstationarity and spatiotemporal variation of extreme precipitation in an urban agglomeration in the middle reaches of the Yangtze River (UAMRYR) from 1970 to 2017 were analyzed based on nine extreme precipitation indices. A nonstationary frequency analysis of extreme precipitation by the Generalized Additive Models for Location, Scale, and Shape (GAMLSS) was constructed using 6 atmospheric metrics and 6 urbanization metrics as explanatory variables. Additionally, the separate contributions of climate change and urbanization to extreme precipitation were compared quantitatively. Results demonstrated that during 1970-2017, the extreme precipitation increased in magnitude, whereas mainly decreased in duration, with a spatial pattern gradually increasing from west to east in the UAMRYR. The nonstationary GAMLSS could better capture the evolution characteristics of extreme precipitation, indicating that the nonstationarity of extreme precipitation in the UAMRYR is significant. Among the constructed nonstationary models, the number of optimal models with the Largest Patch Index (LPI), Percentage of Landscape (PLAND), and Landscape Shape Index (LSI) as explanatory variables is the largest. In general, urbanization contributed greater to the nonstationarity of extreme precipitation than climate change in the majority of the UAMRYR. This study can provide scientific references for waterlogging control and urban planning. |