Abstract: The daily meteorological variables observed at weather stations in the Qiangtang Nature Reserve of Tibet (QNRT), including average temperature, precipitation, and sunshine duration from 1971 to 2019, were analyzed with the criteria on the 5℃ threshold temperature. Based on the variation characteristics of the green grass seasons(GGS) of alpine grasslands and their hydrothermal climate conditions in the recent 49 years were revealed with the support of the statistical analyses of the linear regression, the Mann-Kendall test method and the R/S method. In addition, the possible changes of the GGS were also assessed under RCP2.6, RCP4.5 and RCP8.5 emission scenarios, in order to well understand and accurately predict the dynamic change of the terrestrial ecosystems in the future. The study results indicate that the start date of GGS has been advanced at a rate of 2.81d·10y−1 while the end date has been delayed with a rate of 2.74d·10y−1, which has contributed to an increase in duration by 5.56 days per decade. The increasing trends in cumulative temperature, precipitation, and sunshine duration were significant, with the rates of 75.86℃·d·10y−1, 15.84mm·10y−1 and 27.58h·10y−1, respectively. In terms of the decadal variation characteristics, the start dates were late while the end dates were early at all stations from 1970s to 1990s, leading to the shorter duration of GGS and the unfavorable hydrothermal conditions. In contrast, the start dates were early while the end dates were late at all stations from 2000s to 2010s, which contributed to the longer duration of GGS and the abundant hydrothermal resource. It was detected by M-K methods that the abrupt changes of start date, end date, and duration of GGS occurred in 2006, 1991, and 1988, respectively. In addition, the abrupt increases in cumulative temperature, precipitation, and sunshine duration of GGS were also detected in 1988, 1999, and 1981, respectively. All values of H were larger than 0.5, indicating that the trends in early start date, late end date, longer duration will continue, and the increases in cumulative temperature, precipitation, and sunshine duration will remain unchanged in the future. Under RCP4.5 emission scenario, the start date of GGS will be advanced by 10 days, while the end date will be delayed by 9 days, and the duration of GGS will prolonged by 17 days in QNRT in the next 80 years, which will be beneficial to pasture growth and livestock fattening, and also play a positive role in pasture production and restoration of the grassland ecosystem in this highland region.

Key words: Green grass season, Accumulated temperature above 5℃, Precipitation, Sunshine duration, Climate change, CMIP5 prediction