Abstract:High-resolution urban building classification data is critical to the development of urban meteorology and related disciplines. The Local Climate Zone (LCZ) scheme is a new method to construct urban high-resolution land use classification (HRLUC) dataset. Based on the LCZ classification system, HRLUC dataset with building classification for 63 cities in China was produced in this paper. Landsat 8 satellite images from 2017 to 2019 were used to produce this dataset. The dataset covers approximately 51,933 training samples (>100m?100m) and 15,841 validation samples. The results from the quality assessments show that the HRLUC dataset of 63 cities in China are generally of good quality, with an overall accuracy between 72% and 92% among cities, and an average overall accuracy of 81%. The accuracy of built - up areas in the selected cities ranges from 59% to 82%, with an average accuracy of 72%. The accuracy for natural land cover areas ranges from 71% to 100%, and the mean accuracy is 89%. Compared to similar research work, the accuracy for built types in this study is higher. Different from developed countries such as Europe and America, China is a developing country with most of its cities still under development and construction. A large proportion of the underlay surface is under construction, but the type of this underlay surface has not been classified into LCZ classification system. In this paper, based on the original LCZ classification system, construction zone was added as the basic types of LCZ classification in Chinese cities. Among the 63 cities studied in this paper, 24 cities account for more than 5% of construction zone, and the underlying surface of this type is mostly located at the junction of suburbs.