Characterization of extreme temperature index changes in China from 1961 to 2021

Against the backdrop of intensified global climate change, the frequency and intensity of extreme weather events in mainland China continue to rise, owing to their unique topography and complex climate types. In-depth research on the trends and impacts of climate extremes can help develop effective adaptation and mitigation strategies to protect the environment and enhance social resilience. This article is based on temperature and large-scale climate index data from 2029 meteorological stations between 1961 and 2021, we analyzed the spatial-temporal variation characteristics of extreme temperatures in China by using linear propensity estimation and Mann-Kendall test for 15 extreme temperature indices and 3 extreme composite temperature indices. The Pearson correlation coefficient method was used to determine the relationship between these indices and the atmospheric circulation factors. The results show that in the past 60 years, the extreme temperature index in China has shown a trend of decreasing low temperature events and increasing high temperature events; in particular, the increase in warm nights is significantly higher than that of warm days. In terms of spatial distribution, daily maximum temperature less than the 10th percentile (TX10P) and daily minimum temperature greater than the 90th percentile (TN90P) increased significantly in the warm temperate sub-humid region (WTSH), north subtropical humid region (NSH), and marginal tropical humid region (MTH), whereas frost days (FD0) and diurnal temperature range (DTR) decreased significantly. In the extreme composite temperature index, ETR showed a downward trend, while compound warm days (CHW) and compound warm days relative humidity (CHW-RH20) increased, with the latter mainly concentrated in the WTSH and NSH regions. Correlation analysis with atmospheric circulation factors shows that Arctic Oscillation (AO), Atlantic Multiannual Oscillation (AMO), and El Niño-Southern Oscillation (ENSO) are positively correlated with extremely high temperatures, whereas the North Atlantic Oscillation (NAO) and Pacific Decadal Oscillation (PDO) are negatively correlated.