| 0 | 0 | 5 |
| 下载次数 | 被引频次 | 阅读次数 |
准确估算大气中的二氧化碳(carbon dioxide,CO2)浓度对于理解碳循环、制定减排策略以及评估气候变化影响至关重要。然而，现有的CO2数据集主要依赖地面观测站和卫星遥感数据，存在空间分布不均、不确定性高和分辨率不足等问题，难以满足精细尺度研究需求。为此，本文基于全球1°分辨率的大气CO2的柱平均干空气柱浓度摩尔分数(the column-averaged dry-air mole fraction of carbon dioxide,XCO2)数据，提出面向XCO2预测的空间极端梯度提升树模型(spatial eXtreme gradient boosting,SXGBoost)，并将其应用于京津冀地区、长江三角洲和粤港澳大湾区三大城市群，实现了1 km高分辨率XCO2的精准预测。结果表明，SXGBoost模型通过有效地融合XCO2的空间特征，在预测精度和泛化能力上显著优于传统模型，能够更准确地描述XCO2的时空变化规律。本文开发的高分辨率XCO2数据集为城市群绿色低碳发展提供了重要数据支撑，对推动生态文明城市建设具有重要的科学价值和实践意义。
Abstract：Estimating accurately the atmospheric carbon dioxide(CO2) concentrations is crucial for understanding the carbon cycle, formulating emission reduction strategies, and assessing the impacts of climate change. However, existing CO2 datasets mostly rely on ground-based observation stations or satellite remote sensing data, which suffer from uneven distribution, high uncertainty, and low resolution, making them inadequate for fine-scale research. This study leverages the global 1° resolution the column-averaged dry-air mole fraction of carbon dioxide(XCO2) and proposes a spatial eXtreme Gradient Boosting(SXGBoost) model for XCO2 prediction. It is applied to three representative urban agglomerations in China—Beijing-Tianjin-Hebei, the Yangtze River Delta, and the Guangdong-Hong Kong-Macao Greater Bay Area—achieving 1 km high-resolution XCO2 predictions for these regions. The results demonstrate that the SXGBoost model which effectively incorporates the spatial characteristics of XCO2, exhibits significantly superior prediction accuracy and generalization capability compared with the traditional models. Consequently, it can more precisely characterize the spatiotemporal variations of XCO2. The high-resolution XCO2 dataset developed in this study offers crucial data support for the low-carbon and green development of urban agglomerations. It holds significant scientific value and practical implications for advancing the construction of ecologically civilized cities.
[1]邓安健，郭海波，胡洁，等. GOSAT卫星数据监测中国大陆上空CO2浓度时空变化特征[J].遥感学报，2020,24(3):319-325
[2]赵灿，焦洪赞，洪良.动态视角下城市形态演变过程对碳排放影响研究[J].测绘地理信息，2024,49(5):111-117
[3]黄欣.顾及碳排放的长江中游城市群建设用地生态效率空间关联研究[J].测绘地理信息，2024,49(2):87-92
[4]王孟杰，王艳军，李少春，等.融合多源遥感数据的长株潭城市群碳排放时空差异分析[J].测绘通报，2023(1):65-70
[5]Zhao C, Andrews A E, Bianco L, et al. Atmospheric Inverse Estimates of Methane Emissions from Central California[J]. Journal of Geophysical Research:Atmospheres,2009,114(D16):1-13
[6]李正强，谢一凇，石玉胜，等.大气环境卫星温室气体和气溶胶协同观测综述[J].遥感学报，2022,26(5):795-816
[7]Hammerling D M, Michalak A M, Kawa S R. Mapping of CO2 at High Spatiotemporal Resolution Using Satellite Observations:Global Distributions from OCO-2[J]. Journal of Geophysical Research:Atmospheres, 2012, 117(D6):1-10
[8]Siabi Z, Falahatkar S, Alavi S J. Spatial Distribution of XCO2 Using OCO-2 Data in Growing Seasons[J].Journal of Environmental Management,2019,244:110-118
[9]Chen S, Mihara K, Wen J. Time Series Prediction of CO2, TVOC and HCHO Based on Machine Learning at Different Sampling Points[J]. Building and Environment,2018,146:238-246
[10]张杨，徐永明，卢响军，等.基于OCO-2遥感数据的新疆维吾尔自治区大气XCO2空间化研究[J].生态环境学，2024,33(2):231-241
[11]Sun W, Xu Z. A Novel Hourly PM2. 5 Concentration Prediction Model Based on Feature Selection, Training Set Screening, and Mode Decomposition-Reorganization[J]. Sustainable Cities and Society, 2021, 75:103348
[12]Sagi O, Rokach L. Ensemble Learning:A Survey[J].Wiley Interdisciplinary Reviews:Data Mining and Knowledge Discovery, 2018, 8(4):e1249
[13]黄经南，惠雨，杨石琳，等.基于县（市、区）的京津冀城市群生态与经济协调发展研究[J].测绘地理信息，2024, 49(2):100-107
[14]王伟，邓雯婷，李成仁，等.长三角一体化地理信息平台设计与实现[J].测绘地理信息，2022, 47(4):124-127
[15]童娅琼，王树根，李赟鹏，等.粤港澳大湾区旅游景点可达性时空格局演变[J].测绘地理信息，2022, 47(2):127-131
[16]Sheng M, Lei L, Zeng Z C, et al. Global Land 1°Mapping Dataset of XCO2 From Satellite Observations of GOSAT and OCO-2 from 2009 to 2020[J]. Big Earth Data, 2023, 7(1):170-190
[17]Peng S, Ding Y, Liu W, et al. 1 km Monthly Temperature and Precipitation Dataset for China from 1901to 2017[J]. Earth System Science Data, 2019, 11(4):1 931-1 946
[18]Li X, Xiao J. Mapping Photosynthesis Solely from Solar-Induced Chlorophyll Fluorescence:A Global,Fine-Resolution Dataset of Gross Primary Production Derived from OCO-2[J]. Remote Sensing,2019, 11(21):2 563-2 583
[19]Chen Z, Yu B, Yang C, et al. An Extended Time-Series(2000—2018)of Global NPP-VIIRS-Like Nighttime Light Data from a Cross-Sensor Calibration[J]. Earth System Science Data Discussions,2020,13(3):889-906
[20]郑建滨，艾廷华，晏雄锋，等.基于XGBoost的多建筑WiFi位置指纹室内定位方法[J].测绘地理信息，2019,44(2):65-68
[21]Hwang Y, Um J S. Performance Evaluation of OCO-2XCO2 Signatures in Exploring Casual Relationship Between CO2 Emission and Land Cover[J]. Spatial Information Research, 2016, 24:451-461
[22]Wei J, Li Z, Lyapustin A, et al. Reconstructing 1-kmResolution High-Quality PM2. 5 Data Records from2000 to 2018 in China:Spatiotemporal Variations and Policy Implications[J]. Remote Sensing of Environment, 2021, 252:112136
基本信息:
DOI：10.14188/j.2095-6045.20250027
中图分类号:X87;P208
引用信息:
[1]柯水松,吴超,杨硕,等.基于空间XGBoost的高分辨率XCO_2估算——以中国三大城市群为例[J].测绘地理信息,2026,51(03):59-64.DOI:10.14188/j.2095-6045.20250027.
基金信息:
国家自然科学基金（41901326）; 自然资源部城市国土资源监测与仿真重点实验室开放基金（KF-2023-08-24）; 全国统计科学研究计划（2024LY070）
2025-07-08
2025-07-08
2025-07-08