 This study used enhanced vegetation index, EVI, time series data from MODIS and geographic features to classify land use land cover, LULC, in the Chilean mountains, northwest China using random forest, RF, classification, and regression tree, CART, and support vector machine, SVM, classifiers. The results showed that topographic information improved classification accuracy, RF achieved the highest overall accuracy of 88.84%. There was consistency with homogeneous classes but inconsistency with heterogeneous classes, and the overall accuracy was improved by about 10% compared to global LULC products. Therefore, this classification product is more suitable than global LULC products when considering complex terrain factors in mountain regions. This article was authored by Hong Wang, Chen Liliu, Fei Zong, and others.