This paper presents a threshold method to examine the vegetation phenology in a Qinghai Lake watershed based on a time-series analysis of a MODIS EVI dataset from 2000 to 2014 using 16-day intervals. The spatio-temporal variation of vegetation phenology and the relationship with vegetation degradation were also analyzed. The results showed that the beginning of the vegetation growing season (BGS) occurred from mid-April to mid-June and the end of growing season (EGS) from mid-August to mid-October; growing season (GS) was between 100 to 150 days. The BGS spatial pattern was embodied by a transition from southeast to northwest; vertical zonation in the mountainous region showed a contrary trend in EGS and GS. Spatio-temporal variation of vegetation phenology at the whole watershed level showed spatial heterogeneity under climate change, but mean BGS trends were not pronounced during the study period. EGS was advanced approximately 6.4 days and GS prolonged by 8.9 days. Vegetation phenology was also affected by vegetation degradation. BGS tended to be later when vegetation flourished and advanced when vegetation was degraded. The relationship between annual EGS date and vegetation degradation was relatively complex. EGS and vegetation degradation were significantly negatively correlated indicating EGS was earlier where the vegetation was in good condition.