Spatiotemporal Variability in Extreme Temperature Events in an Arid-Semiarid Region of China and Their Teleconnections with Large-Scale Atmospheric Circulation

With a warming climate,temperature extremes have been a main global issue in recent decades due to their potential influence on the sustainable development of human life and natural ecosystems.In this study,12 indicators of extreme temperature events are used to evaluate the spatiotemporal distribution,periodic structure and teleconnections with large-scale atmospheric circulation in Xinjiang,Northwest China by combining wavelet coherence(WTC) analysis based on continuous wavelet transform(CWT) analysis with the sequential Mann-Kendall test.We find that over the past six decades,the climate in Xinjiang has become warmer and has suffered from increases in the frequency of warm extremes and decreases in the frequency of cold extremes.Warm extremes have mainly occurred in the southern Tianshan Mountains surrounding the Tarim Basin and western part of the Taklamakan Desert,and cold extremes have primarily occurred in the southwestern Altai Mountains and northern foot of the Tianshan Mountains.Extreme temperature events,including warm extremes,cold extremes,and other temperature indices,have significant interannual variability,with the main oscillation periods at smaller(2–4-year band),intermediate(4–7-year band),and greater time scales in recent decades.Furthermore,cold-extreme indices,including frost days,cool days,and cool nights all show a clear changepoint during 1990–1997 at the 95% confidence level,and both ice days and cold spell duration indicator have a potential changepoint during 1981–1986.However,the changing points for warmextreme indices are detected during 1992–1998.The temperature variables are significantly correlated with the EI Ni?o-Southern Oscillation(ENSO) and Arctic Oscillation(AO),but less well correlated with the Pacific Decadal Oscillation(PDO).The phase difference in the WTC spectra is not uniform between temperature extremes and climatic oscillations.Our findings will have important implications for local governments in taking effective measures to mitigate the potential effects of regional climate warming due to human activities in Xinjiang.