On the response of daily precipitation extremes to local mean temperature in the Yangtze River basin

This study explores the response of daily extreme precipitation (PEX) to local mean temperatures (T) of the previous 0-day (T0), 1-day (T1), and 1 to 9-days (T1-9) in the Yangtze River basin (YRB) using the non-stationarity test, binning method and copula theory. Results show that the annual scaling rate of PEX with T during 1961-2020 exhibits positive, negative and positive changing trends from southwest to northeast with the maximum magnitudes in the west, and experiences significant abrupt change only at 14.1% grids. The PEX-T scaling relation in the entire study period is predominantly characterized by a peak pattern over the YRB, with precipitation peaking at about 8 degrees C in the western highlands and at 25 degrees C in the east-central lowlands. However, the use of T1-9 yields a positive PEX-T pattern at more grids compared to T1 and T0, which highlights the important role of atmospheric moisture residence time in minimizing cooling artefacts in the scaling estimation. Generally, the full scaling rate of PEX with T in the study period is larger in the west-central part but smaller in the eastern YRB, and more sensitive to antecedent temperatures, with the ranges of - 8-20%/degrees C, - 8-30%/degrees C and - 10-30%/degrees C against T0, T1 and T1-9, respectively. As precipitation intensifies, the full scaling decreases in plateau-mountain areas but increases in low-lying regions, gradually approaching the ClausiusClapeyron (CC) scaling and exhibiting a "super-CC, CC-like and sub-CC scaling" pattern from west to east. Additionally, heavier precipitation events are most likely to occur at higher local temperatures in most areas even with peak structure. These findings provide new insights into the PEX-T relationships, which is conducive to understanding future changes in precipitation extremes and assessing precipitation-related risks under climate change.