Temperature sensitivity (S_T) of the leaf unfolding date (LUD), which was defined as the shift of LUD per unit change in preseason temperature, was widely used to quantify phenological response to climate change. Species whose LUD responding to temperature more sensitively tended to have an earlier growing season, thus occupying more resources and obtaining competitive advantages in the community. Therefore, S_T could be used to assess the ability of species to cope with climate change. The temporal change in S_T and their driving forces have been studied for several plant species in Europe, but the long-term variation of S_T in China is still unclear. In this study, we selected the LUD data for 163 woody plants at 10 sites in China from the China Phenology Observation Network (CPON), as well as the corresponding meteorological data from the China Meteorological Data Service Center. First, S_T was calculated as the slope of the regression between LUD and the mean temperature during the optimum preseason, for each 15-year moving window during the period 1963-2014 for each species at each site. Second, we analyzed the temporal trends in S_T and calculated the average trend of S_T for each site to test the spatial patterns of S_T. Finally, we discussed the possible reasons for the changes in S_T. The conclusions were as follows:(1) Out of 313 LUD time series, 60.1% displayed increasing S_T, with 40.0% significantly (P < 0.05). S_T for the other 39.9% of time series exhibited a decreasing trend, with 28.4% significantly. (2) In terms of the spatial patterns, S_T of LUD showed increasing trends at 6 sites of temperate region. The most obvious one was Beijing, with 75% of species exhibiting significantly increasing S_T. On the contrary, S_T of LUD decreased over the past several decades at 4 sites of subtropical region except for Hefei. The most obvious one was Changsha, with 68.4% of species exhibiting significantly decreasing S_T. (3) Both the winter chilling and spring temperature variance affected the temporal variation in S_T. Reduced chilling days could lead to a lower S_T, while lower spring temperature variance could lead to a higher S_T.