Abstract: In order to grasp the evolutionary trend of drought and vegetation in various regions of North China, and to analyze the degree of response of vegetation to different drought types, it is of guiding significance for drought mitigation work. In the paper, data from 90 meteorological stations in North China were used to construct the standardized precipitation evapotranspiration index (SPEI), and the normalized vegetation index (NDVI) was selected to quantify the vegetation coverage. Then, the Sen's slope estimator, Mann-Kandall test, correlation analysis, and Copula-Bayesian conditional probability formula were used to analyze the drought sensitivity of vegetation in North China. The results show that: (1) the SPEI value in most parts of North China showed a extremely significant increase, the SPEI value in some parts of North China showed a extremely significant decrease, the NDVI value in the western and northern regions of North China showed an extremely significant increase, and the rest of the regions showed an extremely significant decline. (2) Drought and vegetation in North China were dominated by insignificant spatial clustering, with high and low values of drought and vegetation clustering in small areas. (3) In North China, the correlation between annual scale SPEI and NDVI was good, the response rate of vegetation to drought was slower and the sensitivity was low, the correlation between month-scale SPEI and NDVI was better in some parts of Henan, and the vegetation was more sensitive to drought, and the joint distribution function of NDVI and SPEI was the most consistent with the Clayton Copula function. (4) The likelihood of lowest vegetation coverage occurring in North China was decreasing with decreasing drought, the likelihood of lower vegetation coverage occurring was increasing from extreme to moderate drought and decreasing from moderate to light drought, and the likelihood of medium, higher and highest vegetation coverage all increase with decreasing drought.

Key words: SPEI, NVDI, Sen's slope estimator, Mann-Kandall test, Copula-Bayesian conditional probability formula function