A three-stage hybrid model for regionalization, trends and sensitivity analyses of temperature anomalies in China from 1966 to 2015

Temperature anomalies have received increasing attention due to their potentially severe impacts on ecosystems, economy and human health. To facilitate objective regionalization and examine regional temperature anomalies, a three-stage hybrid model with stages of regionalization, trends and sensitivity analyses was developed. Annual mean and extreme temperatures were analyzed using the daily data collected from 537 stations in China from 1966 to 2015, including the annual mean, minimum and maximum temperatures (Tm, TNm and TXm) as well as the extreme minimum and maximum temperatures (TNe and TXe). The results showed the following: (1) subregions with coherent temperature changes were identified using the rotated empirical orthogonal function analysis and K-means clustering algorithm. The numbers of subregions were 6, 7, 8, 9 and 8 for Tm, TNm, TXm, The and TXe, respectively. (2) Significant increases in temperature were observed in most regions of China from 1966 to 2015, although warming slowed down over the last decade. This warming primarily featured a remarkable increase in its minimum temperature. For Tm and TNm, 95% of the stations showed a significant upward trend at the 99% confidence level. TNe increased the fastest, at a rate of 0.56 degrees C/decade, whereas 21% of the stations in TXe showed a downward trend. (3) The mean temperatures (Tm, TNm and TXm) in the high latitude regions increased more quickly than those in the low-latitude regions. The maximum temperature increased significantly at high elevations, whereas the minimum temperature increased greatly at middle-low elevations. The most pronounced warming occurred in eastern China in The and northwestern China in The, with mean elevations of 51 m and 2098 m, respectively. A cooling trend in The was observed at the northwestern end of China. The warming rate in TNe varied the most among the subregions (0.63 degrees C/decade).