Abstract: Tree species at alpine timberline are sensitive to climate change, which are ideal species for studying climate change and phenological change. Based on the double Logistic curve fitting method, we reconstructed the MODIS EVI₂ time series images from 2000 to 2015 and extracted key phenological parameters of Larix chinensis forest, combined with climatic data of the Taibai Mountain meteorological station, and analyzed the phenology of the L. chinensis forest and its responses to climate factors. From 2000 to 2015, the beginning time of growing period of L. chinensis forest was on about 120th day, and the ending time of growth period was on about 288th day, with an average growing period of 168 days. The rate of change was -0.65 d·10 a^-1 at the beginning of the growing season with an advanced trend, and it was 0.35 d·10 a^-1 at the end of the growing season with a delayed trend, with a delayed trend of 0.99 d·10 a^-1 for the whole growing season. The areas with negative correlation between the start phenophase of L. chinensis forest and the temperature and precipitation in the same period accounted for 73.11% and 61.25% of the study area, respectively, while the areas with positive correlation between the start phenophase and the potential evapotranspiration accounted for 65.17% of the study area. The end of phenophase was positively correlated with temperature, precipitation, potential evapotranspiration, sunshine and wind speed in the same period, with area rates of 74.91%, 72.87%, 78.14%, 68.60% and 64.74%, respectively. Temperature was the major factor affecting the advance of the phenological stage of L. chinensis forest, while temperature, precipitation and potential evapotranspiration were the major factors affecting the late phenology.

Key words: alpine forest, litter decomposition, dynamic of release, snow patch.