［Objective］ The risk of vegetation loss under temperature stress considering lag effects was assessed to provide a scientific basis for formulating regional adaptive management measures. ［Methods］ Using Heilongjiang Province as an example， a risk assessment method for vegetation loss under temperature stress considering lag effects was proposed based on weekly normalized difference vegetation index （NDVI） and temperature data. Quantifying the probability of warming and cooling stress risk in different terrestrial ecosystems in Heilongjiang Province. were quantified. ［Results］ Vegetation during the growing season is closely associated with temperature changes， reaching the most sensitive state at approximately 9 and 23 weeks of lag， respectively， and with high spatial consistency. The eastern and western regions of the study area are high-risk regions for vegetation loss， whereas the northwestern and central regions have a lower risk. For every 1 ℃ increase in the average temperature， the risk probability increased by approximately 0.5%， and for every 1 ℃ decrease， the risk increased by approximately 0.70%， indicating that cooling had a greater impact on vegetation than warming. The risk variation under temperature stress was higher in farmland ecosystems and lower in forest ecosystems. ［Conclusion］ The eastern and western regions of Heilongjiang Province are high-risk areas for vegetation loss， and cooling has a greater impact on vegetation than warming. This highlights the practical significance of comprehensively considering the lag effects of temperature stress to accurately assess vegetation health in the context of global climate change.