An analysis of changes in temperature extremes in the Three River Headwaters region of the Tibetan Plateau during 1961-2016

Regional spatiotemporal assessments of temperature extremes are essential for accurate forecasting of future climate changes, agricultural production and the regulation of water resources. This study evaluates the characteristics of the spatiotemporal variations in temperature extremes from 1961 to 2016 in the Three River Headwaters (TRH) region and its surroundings in the northeastern portion of the Tibetan Plateau in China. The association of these variations with atmospheric circulation patterns is also investigated. This study selected twenty-six meteorological stations where < 2% of the data are missing and where the data pass quality control and homogeneity tests. Following the Expert Team on Climate Change Detection and Indices (ETCCDI), thirteen extreme temperature indices and two mean temperature indices are analyzed using the Mann-Kendall test and Sen's test; the indices examined here include six warm indices, six cold indices, the diurnal temperature range (DTR) and the annual minimum and maximum temperatures. Pearson correlation analysis is also used to analyze the correlations between the temperature extremes and various atmospheric circulation patterns. The results indicate that the annual minimum and maximum temperatures display significant increasing trends, whereas the DTR displays a significant decreasing trend with a rate of -0.11 degrees C/decade. All of the regional warm indices display significant increasing trends. On the other hand, most of the cold indices show significant decreasing trends, except for the lowest daily minimum temperature (TNn) and the lowest daily maximum temperature (TXn), which show increasing trends. These results are consistent with the trends observed by most global-scale studies. In terms of spatial patterns, most of the stations that display significant trends are located in the central and northwestern portions of the TRH region. On the other hand, all of the extreme indices show weak correlations with altitude, with the exception of the DTR, which displays a significant negative correlation with altitude (P < 0.01); moreover, most of the indices are weakly correlated with latitude and longitude. Overall, the winter Arctic Oscillation index (AO), the summer and winter North Atlantic Oscillation index (NAO), and the summer and winter Western Pacific Subtropical High Intensity Index (WPI) display significant relationships with most of the extreme temperature indices, and these circulation patterns contribute to changes in temperature extremes in this region. However, the Southern Oscillation Index (SOI) is not significantly correlated with most of the extreme temperature indices.