School of Geographic and Biologic Information, Nanjing University of Posts and Telecommunications, Nanjing, 210023, China; Smart Health Big Data Analysis and Location Services Engineering Lab of Jiangsu Province, Nanjing University of Posts and Telecommunications, Nanjing, 210023, China.
School of Geographic and Biologic Information, Nanjing University of Posts and Telecommunications, Nanjing, 210023, China.
Environ Res. 2023 Nov 1;236(Pt 2):116866. doi: 10.1016/j.envres.2023.116866. Epub 2023 Aug 9.
Long-time-series, high-resolution datasets of the column-averaged dry-air mole fraction of carbon dioxide (XCO) have great practical importance for mitigating the greenhouse effect, assessing carbon emissions and implementing a low-carbon cycle. However, the mainstream XCO datasets obtained from satellite observations have coarse spatial resolutions and are inadequate for supporting research applications with different precision requirements. Here, we developed a new spatial machine learning model by fusing spatial information with CatBoost, called SCatBoost, to fill the above gap based on existing global land-mapped 1° XCO data (GLM-XCO). The 1-km-spatial-resolution dataset containing XCO values in China from 2012 to 2019 reconstructed by SCatBoost has stronger and more stable predictive power (confirmed with a cross-validation (R = 0.88 and RSME = 0.20 ppm)) than other traditional models. According to the estimated dataset, the overall national XCO showed an increasing trend, with the annual mean concentration rising from 392.65 ppm to 410.36 ppm. In addition, the spatial distribution of XCO concentrations in China reflects significantly higher concentrations in the eastern coastal areas than in the western inland areas. The contributions of this study can be summarized as follows: (1) It proposes SCatBoost, integrating the advantages of machine learning methods and spatial characteristics with a high prediction accuracy; (2) It presents a dataset of fine-scale and high resolution XCO over China from 2012 to 2019 by the model of SCatBoost; (3) Based on the generated data, we identify the spatiotemporal trends of XCO in the scale of nation and city agglomeration. These long-term and high resolution XCO data help understand the spatiotemporal variations in XCO, thereby improving policy decisions and planning about carbon reduction.
长时间序列、高分辨率的大气二氧化碳柱平均干空气摩尔分数（XCO）数据集对于缓解温室效应、评估碳排放和实施低碳循环具有重要的实际意义。然而，卫星观测获得的主流 XCO 数据集空间分辨率较粗，不足以支持具有不同精度要求的研究应用。在这里，我们基于现有的全球土地映射 1° XCO 数据（GLM-XCO），通过融合空间信息与 CatBoost 开发了一种新的空间机器学习模型，称为 SCatBoost，以填补上述空白。SCatBoost 重建的包含中国 2012 年至 2019 年 XCO 值的 1km 空间分辨率数据集具有更强、更稳定的预测能力（交叉验证确认，R=0.88，RSME=0.20 ppm），优于其他传统模型。根据估计数据集，全国 XCO 呈总体增加趋势，年平均浓度从 392.65 ppm 上升到 410.36 ppm。此外，中国 XCO 浓度的空间分布明显反映出东部沿海地区的浓度明显高于西部内陆地区。本研究的贡献可以总结如下：（1）它提出了 SCatBoost，集成了机器学习方法和空间特征的优势，具有较高的预测精度；（2）它提供了一个由 SCatBoost 模型生成的中国 2012 年至 2019 年精细尺度和高分辨率 XCO 数据集；（3）基于生成的数据，我们确定了 XCO 在国家和城市群规模上的时空趋势。这些长期、高分辨率的 XCO 数据有助于了解 XCO 的时空变化，从而改进有关减少碳排放的政策决策和规划。
