Nonstationarity in timing of extreme precipitation across China and impact of tropical cyclones

摘要

This study examines the seasonality and nonstationarity in the timing of extreme precipitation obtained by annual maximum (AM) sampling and peak-over-threshold (POT) sampling techniques using circular statistics. Daily precipitation data from 728 stations with record length of at least 55 years across China were analyzed. In general, the average seasonality is subject mainly to summer season (June-July August), which is potentially related to East Asian monsoon and Indian monsoon activities. The strength of precipitation seasonality varied across China with the highest strength being in northeast, north, and central-north China; whereas the weakest seasonality was found in southeast China. There are three seasonality types: circular uniform, reflective symmetric, and asymmetric. However, the circular uniform seasonality of extreme precipitation was not detected at stations across China. The asymmetric distribution was observed mainly in southeast China, and the reflective distribution of precipitation extremes was also identified the other regions besides the above-mentioned regions. Furthermore, a strong signal of nonstationarity in the seasonality was detected at half of the weather stations considered in the study, exhibiting a significant shift in the timing of extreme precipitation, and also significant trends in the average and strength of seasonality. Seasonal vapor flux and related delivery pathways and also tropical cyclones (TCs) are most probably the driving factors for the shifts or changes in the seasonality of extreme precipitation across China. Timing of precipitation extremes is closely related to seasonal shifts of floods and droughts and which means much for management of agricultural irrigation and water resources management. This study sheds new light on nonstationarity in timing of precipitation extremes which differs from existing ones which focused on precipitation extremes from perspective of magnitude and intensity. (C) 2017 Elsevier B.V. All rights reserved.