To reveal the phenological characteristics of typical vegetation woodland and grassland and its response to climate change in Northern China, adopting the GIMMS data and meteorological data from 1982 to 2015, the woodland and grassland phenology were extracted by using a dynamic threshold method, and the spatio-temporal variations of phenology and its response to climate change were identified by linear regression and partial correlation analysis methods. The results can be summarized as follows. (1) The woodland length of season (LOS) was significantly extended with a rate of 0.32d/a, while the start of season (SOS) was significantly advanced with rate at -0.18d/a and the end of season (EOS) was obviously delayed with rate at 0.14d/a. The advanced SOS, delayed EOS, and extended LOS areas covered 93.3%, 90.4%, and 96.3% of the whole woodland area, respectively. (2) The grassland LOS was insignificantly shortened with a rate of -0.01d/a, while the SOS was insignificantly advanced with rate at -0.09d/a and the EOS was insignificantly advanced with rate at -0.10d/a. The advanced SOS and EOS areas covered 67.5% and 69.1% of the whole grassland area, respectively. And, the LOS was shortened in 50% of the areas. (3) The advanced woodland SOS was mainly influenced by the increase in temperature and precipitation in late winter and early spring, and the delayed EOS was affected by the increased temperature in summer and precipitation in autumn, late winter, and early spring. (4) The advanced grassland SOS was mainly related to the increase in temperature and the decrease of short-wave radiation in spring, and the advanced EOS was affected by the increase in temperature and the decrease of short-wave radiation in autumn. The response time of grassland phenology and meteorological factors was shorter than that of woodland.