Abstract: In this study, the software TIMESAT was used to generate smooth time series of GIMMS NDVI as well as to estimate the vegetation phenological parameters for eastern China from 1982 to 2006. Results show that the significant positive trends were observed for the length of the growing season for the forest land and grassland. The start of the growing season (SGS) has been advanced for the grassland and irrigated cropland, and the end of the growing season has also been moved up for the irrigated cropland. The temperature plays a key role in influencing these phenological changes for different vegetation types, while the effect of precipitation is obviously not observed. For example, the SGS for the grassland correlates strongly with the temperature of the preceding month (March) of the SGS, and the SGS for the irrigated cropland is greatly affected by the temperature of the month (February) of the SGS.