Based on the GIMMS/NDVI data,the wavelet transform technique is used to analyze the dynamic variation of normalized difference vegetation index(NDVI) and its relationship with climate factors such as temperature and precipitation from 1982 to 2006 over Northeast China on intra-and inter-annual scales.Results show that the dynamic variation characteristic of vegetation is remarkable,the I_NDV(descripting the mean value of inter-annual scale NDVI) decreases from northeast to southwest over Northeast China,the I_NDVand ΔI_NDV(descripting the variation of intra-annual scale NDVI) of forest are the largest,which are 0.41 and 0.70,respectively,and those of grass land are the smallest.The I_NDV increases 6.21% over grass land,increases a little over arable land,and decreases over forest.The correlation coefficients between temperature,precipitation and NDVI are remarkable with the former larger than the latter,and the intra-annual scale correlation coefficient is larger than the inter-annual one.On inter-annual scale,the correlation coefficient between NDVI and temperature over arable land is the largest,which is 0.60,and the coefficients between NDVI and precipitation over arable land and forest are not significant.On the intra-annual scale,the correlation coefficients between NDVI and temperature,precipitation over forest are the largest,which are 0.90 and 0.75,respectively.The impact of temperature on NDVI decreases with a nearly linear trend when the lag time is extended on the inter-annual scale.The effects of temperature and precipitation on NDVI decrease more and more fast with the extension of lag time on the intra-annual scale.The responses of forest to temperature and precipitation are quick,and that of grass land is slow.The duration of precipitation influencing NDVI is shorter than that of temperature.