A dataset of built-up areas of Chinese cities in 2020
Jie Sun, Zhongchang Sun, Huadong Guo, Jianbo Wang, Huiping Jiang, Jian Gao
Abstract
<p indent="0mm">Global urbanization promotes the development of regional economy; however, it also brings many environmental problems. The sustainable urban development has become a popular research topic all of the world. In order to monitor and evaluate the United Nations urban land use efficiency indicator, this study used the Sentinel data to propose an algorithm for rapidly extracting the impervious surface based on the Google Earth Engine cloud computing platform, and finally obtained a standardized urban built-up dataset of 433 cities with a population more than 300,000 in China in 2020. This study adopted the United Nations definition of urban agglomeration area. A confusion matrix of 4065 random validation points was used to conduct the accuracy validation of resultant impervious surface, and the regression analysis of this dataset with the 2020 official Statistical Yearbook showed that the average overall accuracy and kappa coefficient was 95.57% and 0.91, respectively, and the goodness of fit of R2 was 0.82. In addition, we compared this product with the GlobeLand30 (2020) and Beijing Normal University Urban Extents (2020) data. The results showed that our dataset had high extraction accuracy and could be used for monitoring and evaluation of the United Nations urban land use efficiency indicator. It also has significant value in many applications such as urban expansion analysis and environmental evaluation.