School of Geospatial Engineering and Science, Sun Yat-sen University, Zhuhai, China; Key Laboratory of Natural Resources Monitoring in Tropical and Subtropical Area of South China, Ministry of Natural Resources, Guangzhou, China.
School of Geospatial Engineering and Science, Sun Yat-sen University, Zhuhai, China; Key Laboratory of Natural Resources Monitoring in Tropical and Subtropical Area of South China, Ministry of Natural Resources, Guangzhou, China.
Sci Total Environ. 2023 Oct 1;893:164921. doi: 10.1016/j.scitotenv.2023.164921. Epub 2023 Jun 16.
China has set a goal to achieve carbon neutrality by 2060, and satellite remote sensing allows for acquiring large-range and high-resolution carbon dioxide (CO) data, which can aid in achieving this goal. However, satellite-derived column-averaged dry-air mole fraction of CO (XCO) products often suffer from substantial spatial gaps due to the impacts of narrow swath and clouds. Here, this paper generates daily full-coverage XCO data at a high spatial resolution of 0.1° over China during 2015-2020, by fusing satellite observations and reanalysis data in a deep neural network (DNN) framework. Specifically, DNN constructs the relationships between Orbiting Carbon Observatory-2 satellite XCO retrievals, Copernicus Atmosphere Monitoring Service (CAMS) XCO reanalysis data, and environmental factors. Then, daily full-coverage XCO data can be generated based on CAMS XCO and environmental factors. Results show that a satisfactory performance is reported in multiform validations, with RMSE and R of 0.99 ppm and 0.963 in terms of the sample-based cross-validation, respectively. The independent in-situ validation also indicates high consistency (R = 0.866 and RMSE = 1.71 ppm) between XCO estimates and ground measurements. Based on the generated dataset, spatial and seasonal distributions of XCO across China are investigated, and a growth rate of 2.71 ppm/yr is found from 2015 to 2020. This paper generates long time series of full-coverage XCO data, which helps promote our understanding of carbon cycling. The dataset is available from https://doi.org/10.5281/zenodo.7793917.
中国设定了到 2060 年实现碳中和的目标，卫星遥感可以获取大范围、高分辨率的二氧化碳（CO）数据，有助于实现这一目标。然而，卫星反演得到的柱平均干空气 CO 摩尔分数（XCO）产品往往由于窄幅扫描和云层的影响而存在较大的空间空白。本研究利用深度神经网络（DNN）框架，融合卫星观测和再分析数据，生成了 2015-2020 年中国高空间分辨率（0.1°）的逐日全覆盖 XCO 数据。具体来说，DNN 构建了 Orbiting Carbon Observatory-2 卫星 XCO 反演值、哥白尼大气监测服务（CAMS）XCO 再分析数据和环境因素之间的关系。然后，根据 CAMS XCO 和环境因素生成逐日全覆盖 XCO 数据。结果表明，多种验证方法均取得了令人满意的结果，基于样本的交叉验证中 RMSE 和 R 分别为 0.99 ppm 和 0.963；独立的现场验证也表明 XCO 估算值与地面测量值具有较高的一致性（R = 0.866，RMSE = 1.71 ppm）。基于生成的数据集，研究了中国 XCO 的空间和季节分布，发现 2015 年至 2020 年期间 XCO 的增长率为 2.71 ppm/yr。本研究生成了长时间序列的全覆盖 XCO 数据，有助于促进我们对碳循环的理解。该数据集可在 https://doi.org/10.5281/zenodo.7793917 上获取。
