The spatiotemporal distribution and variation of vegetation in arid region are one of the key ecological issues that need to be studied urgently in regional vegetation restoration and national nature reserve construction under climate change. In terms of the establishment of the classification of potential vegetation in the arid and semi-arid areas of northern China and the improved HLZ model， combined with the meteorological simulation data of China （1981-2020）， the vegetation area transition matrix， Kappa coefficient and mean center model were introduced to analyze the climatic elements of the arid and semi-arid areas of northern China and realize the quantitative identification of potential vegetation and the simulation of spatial distribution pattern. The results showed that： Mean annual bio-temperature， average total annual precipitation and potential evapotranspiration ratio showed an overall upward trend in the past 40 years， indicating that there was a warming and drying phenomenon in study area； Six potential vegetation types emerged in the study area， which had a good zonal distribution in different decades. Temperate desert vegetation， temperate herbaceous vegetation and temperate forest vegetation were the main vegetation types， totally accounting for 71 percent of coverage area； The area of temperate forest vegetation increased the fastest， at a rate of 1.1% per decade， while temperate desert vegetation decreased the fastest， at a rate of 1.3% per decade； Between adjacent decades， the area conversion relationship of all kinds of vegetation was relatively constant， and the disparity lay in the degree of conversion； The average shift distance of mean center of montane forest vegetation， temperate desert vegetation and temperate forest vegetation was more than 50 km. Except temperate desert vegetation， the mean center of other vegetation generally moved westward.