Extreme high and low temperatures both exert impacts on terrestrial ecosystems. However, current research still lacks a precise assessment of the risk of vegetation loss under simultaneous consideration of different temperature stresses and lag effects. To this end, we propose a methodology for assessing the risk of vegetation loss under temperature stress that incorporates lag effects, based on weekly normalized difference vegetation index and temperature data. Quantified risk probabilities of different terrestrial ecosystems to warming and cooling stresses in Heilongjiang Province, China. The results of the study revealed a strong association between vegetation and temperature change during the growing season, reaching the most sensitive state around 9 weeks and 23 weeks lag, respectively, with high spatial consistency. The study identifies the eastern and western edges of the study area as high-risk zones for vegetation loss, while the risk is comparatively lower in the northwestern and central regions. The probability of risk increased by about 0.5% for every 1°C of warming in average temperatures and by about 0.7% for every 1°C of cooling. This indicates that cooling has a greater impact on vegetation than warming. Farmland ecosystems had a higher change in risk to temperature stress and forest ecosystems had the least. This study provides new perspectives for understanding the specific impacts of temperature extremes on different ecosystems and provides a scientific basis for developing adaptive management measures.