Abstract: Based on homogeneity-adjusted daily temperature (maximum, minimum and average) data of 18 stations, spatial and temporal changes of extreme temperature events over Tibet were analyzed for the period 1961-2010. The result shows that the number of frost days and ice days reduced significantly, with the most significant reduction in northern Tibet for ice days, but more extensively across the autonomous region for frost days. The length of growing season (GSL) presented a statistically significant increasing trend at a rate of 4.71 d/ 10a, especially in Lhasa and Zedang. The extra-maximum air temperature (TXx) and extra-minimum air temperature (TNn) generally increased. TXx significantly increased along the east section of the Yarlung Zangbo River and in Nagqu Prefecture, and decreased at the southern edge of Tibet, while TNn significantly increased across the region of Tibet, especially during 1981-2010 with a rate of 1.06oC/10a. Significant reduction at a rate of 9.38 d/10a (4.96 d/10a) occurred on cool nights (days), and significant increase at a rate of 10.99 d/ 10a (6.72 d/10 a) occurred for warm nights (TN90p) (days (TX90p)). There is a close correlation between the trends of most extreme temperature indices and altitude, i.e., positive correlations between altitude and TNn, negative correlations between altitude and TXx, TX90p, TN90p and GSL. In terms of decadal variations, TXx, TNn and other warm indices showed an increasing trend, while the cold indices and GSL decreased. It is also found that the abrupt change points of the TNn, warm (cool) nights and GSL were mainly observed before the mid-1990s, while frost days, ice days and warm (cool) days occurred in the early 2000s. In most cases, the linear trend magnitudes of extreme air temperature indices in Tibet were larger than those in the whole country, Tibetan Plateau and its surrounding areas (Qinghai Province, Hengduan Mountains), which show that the extreme air temperature indices response are more sensitive to the regional warming.

Key words: linear trend, climate abrupt change, extreme temperature indices, Tibet