引用本文:陆婋泉,李波,方凯杰,周雨奇,程含渺.基于OCO-2卫星重构的中国地区高覆盖XCO2时空分布特征[J].环境监控与预警,2024,(1):1-11
LU Xiaoquan,LI Bo,FANG Kaijie,ZHOU Yuqi,CHENG Hanmiao.Spatiotemporal Distribution of High Coverage XCO2 Reconstructed from OCO-2 Satellite Data in China[J].Environmental Monitoring and Forewarning,2024,(1):1-11
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 1515次   下载 479 本文二维码信息
码上扫一扫!
分享到: 微信 更多
基于OCO-2卫星重构的中国地区高覆盖XCO2时空分布特征
陆婋泉,李波,方凯杰,周雨奇,程含渺
国网江苏省电力有限公司营销服务中心,江苏 南京 210019
摘要:
卫星遥感技术是深入了解大气二氧化碳(CO2)时空分布特征的重要手段之一,由于探测技术的限制,目前基于卫星遥感观测数据反演的CO2产品的空间覆盖度较低,数据缺失严重,不足以反映CO2浓度的空间分布情况。现基于轨道碳观测卫星-2 (OCO-2)、哨兵5P (Sentinel-5P)、美国CO2同化模拟系统(Carbon Tracker)和欧洲中期天气预报中心第5代(ERA-5)气象再分析数据,结合时间序列拟合估算模型和随机森林算法,重构了2019—2022年中国地区高精度(0.05°×0.05°)大气CO2平均干空气混合比(XCO2),分析了中国地区CO2时空变化特征。与OCO-2和Carbon Tracker对比结果显示,重构得到的XCO2与OCO-2的观测结果一致性更高,均方根误差为1.05 ×10-6,决定系数高达0.96,可以在较高空间分辨率上体现中国地区XCO2的时空分布情况。基于重构的XCO2数据得知,中国地区XCO2呈现明显的季节性波动,XCO2呈冬春高、夏秋低的特征;2019—2022年,中国地区XCO2呈现逐年上升的趋势,增长率达到(2.41±0.01)×10-6/a,但近年来增长速率有所降低;从空间分布来看,中国东部、北部、中部地区的XCO2显著高于其他地区,且增长率也较高;进一步分析中国典型经济区的XCO2发现,杭州、天津、成都的XCO2在各经济区内的增长最为迅速。研究成果可为碳监测研究、碳排放清单验证、碳排放管理、温室气体减排等研究提供重要的数据支撑。
关键词:  二氧化碳平均干空气混合比  重构  分布特征  随机森林
DOI:DOI:10.3969/j.issn.1674-6732.2024.01.001
分类号:X831
基金项目:国网江苏省电力有限公司科技项目基金资助项目(J2022091)
Spatiotemporal Distribution of High Coverage XCO2 Reconstructed from OCO-2 Satellite Data in China
LU Xiaoquan,LI Bo,FANG Kaijie,ZHOU Yuqi,CHENG Hanmiao
State Grid Jiangsu Marketing Service Center,Nanjing,Jiangsu 210019,China
Abstract:
Satellite remote sensing is one of the important methods to understand the spatiotemporal distribution of atmospheric carbon dioxide (CO2). However, due to the restricts of detection technology, the satellite retrieved column averaged mole fraction of carbon dioxide (XCO2) contains a large amount of missing data, which is insufficient to reflect the full spatiotemporal distribution of carbon dioxide concentrations. In this study, based on satellite (OCO-2, TROPOMI) and model (Carbon Tracker, ERA5 Reanalysis) data, we used a time series fitting model and a random forest model to reconstruct the XCO2 with a high spatial resolution (0.05°×0.05°) for China during the period from 2019 to 2022. Compared with OCO-2 and Carbon Tracker XCO2, the reconstructed XCO2 was better consistent with OCO-2 observations, with a root mean square error (RMSE) of 1.05×10-6 and a high correlation coefficient(R2) of 0.96. Based on the reconstructed XCO2, it was found that XCO2 shows significant seasonal fluctuations, with higher values in winter and spring and lower values in summer and autumn. From 2019 to 2022, XCO2 in China showed an increasing trend with a growth rate of (2.41±0.01)×10-6/a, but the growth rate has slowed down in recent years. In terms of spatial distribution, XCO2 in eastern, northern, and central China is significantly higher than other regions, so as the growth rate. Among above regions, Hangzhou, Tianjin, and Chengdu have the fastest XCO2 growth rates. The research findings of this study provide data basis for carbon monitoring research, carbon emission inventory verification, carbon emission management, greenhouse gas reduction, and other related studies.
Key words:  Column averaged mole fraction of carbon dioxide (XCO2)  Reconstruction  Spatiotemporal distribution  Random forest