Extreme climate is gradually increasing and expanding in the context of global climate change. Exploring how vegetation responds and adapts to extreme climate is crucial for assessing regional vegetation vulnerability and developing adaptive strategies. Taking the extreme frost event that occurred in southern China in 2008 as an example, this study used 2003-2009 sun-induced chlorophyll fluorescence (SIF) remote sensing data to extract vegetation phenology, explored the impact of the disaster on phenology of different vegetation types, and analyzed the contribution of various environmental factors and their interactions to vegetation phenology changes by geographical detector. The results showed that: (1) the extreme frost event led to the start of growing season (SOS) advanced 2.96 days, the end of growing season (EOS) delayed 10.47 days, and the length of growing season (LOS) extended 12.79 days. Among all vegetation types, evergreen broadleaf forest presented the most significant changes in phenology while deciduous broadleaf forest displayed the smallest change. (2) The spatial heterogeneity of hydrothermal conditions affected the spatial patterns of vegetation phenology changes.. For example, the phenology change was more pronounced in the southeastern area WHERE hydrothermal CONDITIONS ARE SUFFICIENT, while that in the northwest mountainous area showed opposite to the overall change with delayed SOS and advanced EOS. (3) Precipitation, temperature, relative humidity, soil type, and vegetation type were the main factors influencing vegetation phenology. Among them, temperature and precipitation had the highest explanatory power for EOS, at 0.6522 and 0.5280 respectively. (4) The interaction results of each factor showed two-factor enhancement or nonlinear enhancement effects. Among them, the interactions between climate factors were the strongest, followed by the interactions between climate factors and soil type, vegetation type, and freezing days.