2020年中国各城市建成区数据

To monitor and evaluate the United Nations urban land use efficiency indicator, this dataset uses Sentinel-1/SAR and Sentinel-2/optical data to extract impervious surfaces and convert them into built-up areas based on the Google Earth Engine platform, covering 433 Chinese cities with a population of over 300,000 in 2020. The impervious surface has an average overall accuracy of 95.57% and a Kappa coefficient of 0.91, with a goodness of fit R²=0.82 with the official statistical yearbook. Its accuracy is higher than GlobeLand30(2020) and Urban Extents(2020), and it can be used for urban expansion analysis, environmental evaluation, and SDG11.3.1 indicator monitoring.

为监测并评估联合国城市土地利用效率指标，本数据集依托谷歌地球引擎（Google Earth Engine）平台，采用哨兵-1号合成孔径雷达（Sentinel-1/SAR）与哨兵-2号光学遥感（Sentinel-2/optical）数据提取不透水面，并将其转换为建成区范围；数据集覆盖2020年中国433座常住人口超30万的城市。该不透水面产品的总体平均精度达95.57%，Kappa系数为0.91，与官方统计年鉴的拟合优度R²值为0.82；其精度优于GlobeLand30(2020)与Urban Extents(2020)两类遥感产品，可应用于城市扩张分析、环境评估以及可持续发展目标11.3.1（SDG11.3.1）指标监测。