Quantitative assessment and driving force analysis of vegetation drought risk to climate change:Methodology and application in Northeast China.

• A quantitative assessment framework for ecosystem drought risk was established. • Nearly 45.4% of ecosystem vegetation in NEC has a moderate or high risk of drought. • Quantitative assessment of vulnerability is the key point to drought risk assessment. • The drought risk patterns driven by climate variability and human activities are different in farmland and natural vegetation. Given the significant increasing trend of droughts in Northeast China (NEC) under climate change, it is of particularly urgency to quantify the drought risk to vegetation in different ecosystems. Based on the theory of risk formation, we developed a quantitative framework for the assessment of ecosystem drought risk that integrates both drought hazard and vulnerability (sensitivity and adaptability). To quantitatively assess the drought risk to different vegetation types in NEC, we used long-term (1982–2015) normalized difference vegetation index (NDVI) data and standardized precipitation evapotranspiration index (SPEI) data. The effects of climate change and non-climatic factors (mainly human activities) on the NDVI dynamics and drought risk were analyzed using the partial least squares regression method to explore key drivers of risk formation. Our findings indicate that almost 45.4% of ecosystem vegetation in NEC is at a moderate or high risk of drought. Our findings indicate that vulnerability is key to NEC drought risk assessment, and adaptability plays an important role in regulating sensitivity. Land cover is the main driving factor for the spatial pattern of drought vulnerability in NEC, and forests have the lowest vulnerability. The impact of climate change on vegetation dynamics in dry years (R2 dry = 0.35) is significantly higher than over the entire study period (R2 all = 0.20). Cropland drought risk is dominated by human activities, whilst woodland and grassland drought risk is controlled by temperature and precipitation variability, respectively. The driving force analysis of vegetation dynamics and drought risk improves our understanding of the mechanisms behind the formation of risk patterns in NEC. Our study provides a basis for land use management and the formulation of drought adaptation options for vegetation types at different growth stages. [ABSTRACT FROM AUTHOR]