This study takes the dune-meadow transition zone of Horqin Sandy Land as an example. Based on the interpretation and classification of 15 Landsat TM/OLI remote sensing images from 1987 to 2017, we systematically analyze the dynamic evolution law of land use/land cover in the study area in the past 30 years by referring to the ecological vegetation succession research method. The results show that: (1) With the decision tree classification algorithm, the classification accuracy of all images is more than 88%, among which the highest classification accuracy in 2017 is 95.24%, which meets the requirements of classification research. (2) There is a vegetation structure characteristic of Semi-shrub-Meadow-Shrub in the study area, and the overall change of land cover shows trend is "evolution in southern degradation in northern". Combined with the results of comprehensive dynamic change analysis of land use and land cover, it is shown that the study area follows the general law of dynamic change of land cover resources in arid and semi-arid areas under the intervention of human activities on the whole. It reflects the continuous instability and vulnerability of the ecological environment in the study area. (3) The total area of land use/land cover change in the study reached 2623.59 hm², and the overall change intensity is 63.76%. And the proportion of forward succession is 52.61%, which is mainly characterized by the continuous decrease of Semi-shrub area and the continuous expansion of Alpine meadow area. But at the same time, the area of Semi-shrub turned sandy land was 184.95 hm², indicating that local retrograde evolution occurred simultaneously in the research area dominated by grazing. (4) The centroid of main land cover have obvious southward shift, with the Forest, Meadow and Crop land shift to the north with great human influence during 1987 to 2017. And the centroid migration distance of main land cover descending is "Crop land > Semi-shrub > Shrub > Alpine meadow > Wetland meadow > Forest". By recording the continuously expanding spatial and temporal patterns in Horqin Sandy Land, the study demonstrated the potential of remote sensing-ecological and time series images to track land use changes at a resolution of 30 meters, providing a reference for improving the dynamic monitoring efficiency of land use in arid and semi-arid areas.