一种基于大气CO<sub>2</sub>浓度时空特征的碳排放分区估算方法

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (2532 KB) HTML (0 KB)
输出: BibTeX | EndNote (RIS)

摘要针对人为碳排放的空间分布量级差异大导致排放数据的非正态分布问题，提出了一种基于卫星大气CO₂柱浓度（XCO₂）时空变化特征聚类分区构建人为碳排放神经网络估算模型方法.通过利用与人为碳排放强相关的卫星XCO₂数据时空变化（2010~2021年）特征的聚类分区，利用卫星观测的XCO₂和SIF以及夜间灯光、人口密度和人为排放清单数据EDGAR作为训练学习数据，以聚类区为单位分别构建人为碳排放神经网络估算模型，估算了2021年研究区人为碳排放.与EDGAR交叉验证结果显示，相比以中国全区数据作为训练学习样本统一建模方法的估算，本研究提出的分区建模估算结果从相关系数（R²）的0.43提高到了0.82；空间分布更为合理；平均偏差由0.039Mt CO₂降低到0.018Mt CO₂.研究表明利用多源数据的神经网络训练学习进行人为碳排放的估算，能够为区域碳排放特征和排放清单的不确定性提高提供评估分析依据.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张少卿
	雷莉萍
	宋豪
	郭开元
	吉张辉
	绳梦雅

关键词 ：人为碳排放, 神经网络, 多源数据, 机器学习

Abstract：Aiming at the non-normal distribution of anthropogenic carbon emissions due to the large spatial difference of magnitude, a neural network estimation model for anthropogenic carbon emissions was proposed in this study based on the clustering of spatial-temporal variation characteristics of satellite XCO₂. Using the spatial and temporal variations of satellite XCO₂ (2010~2021), which are strongly correlated with anthropogenic carbon emissions, for clustering and partitioning, and utilizing satellite-observed SIF, nighttime lighting, as well as population density and anthropogenic emission inventory(EDGAR) as training data, neural network models for Carbon emission estimation were built respectively in each cluster region, and the anthropogenic carbon emissions of the study area were estimated for the year 2021. Cross-verification results with EDGAR show that compared with the unified modeling method based on the data of the whole study area, the estimation result of the partition modeling proposed in this study increased the correlation coefficient (R²) from 0.43 to 0.82, with more reasonable spatial distribution, and the mean deviation decreased from 0.039Mt CO₂ to 0.018Mt CO₂. The study shows that the estimation of anthropogenic carbon emissions using neural network training with multi-source data can provide data support for the characterization of regional carbon emissions and the assessment of uncertainty in emission inventories.

Key words： anthropogenic carbon emission neural network multi-source data machine learning

收稿日期: 2023-03-09

PACS:

X51

基金资助:国家重点研发计划项目（2022YFC3800700）；国家重点研发计划项目（2020YFA0607503）

通讯作者: 雷莉萍,研究员,leilp@radi.ac.cn E-mail: leilp@radi.ac.cn

作者简介: 张少卿(1994-),男,山东青州人,中国科学院空天信息创新研究院博士研究生,主要从事人为碳排放定量估算方面研究.发表论文1篇.zhangsq@radi.ac.cn.

引用本文:

张少卿, 雷莉萍, 宋豪, 郭开元, 吉张辉, 绳梦雅. 一种基于大气CO₂浓度时空特征的碳排放分区估算方法[J]. 中国环境科学, 2023, 43(10): 5604-5613. ZHANG Shao-qing, LEI Li-ping, SONG Hao, GUO Kai-yuan, JI Zhang-hui, SHENG Meng-ya. A neural network partitioning method for carbon emission estimation based on spatial-temporal clustering of atmospheric CO₂ concentration. CHINA ENVIRONMENTAL SCIENCECE, 2023, 43(10): 5604-5613.

链接本文:

http://www.zghjkx.com.cn/CN/ 或 http://www.zghjkx.com.cn/CN/Y2023/V43/I10/5604

[1]	Andres R.J, Boden T.A, Higdon D. A new evaluation of the uncertainty associated with CDIAC estimates of fossil fuel carbon dioxide emission[J]. Tellus B Chem. Phys. Meteorol., 2014,66:23616.
[2]	Oda T, Maksyutov S. A very high-resolution (1km×1km) global fossil fuel CO2 emission inventory derived using a point source database and satellite observations of nighttime lights[J]. Atmos. Chem. Phys., 2011,11:543-556.
[3]	Oda T, Maksyutov S, Andres R.J. The Open-source Data Inventory for Anthropogenic Carbon dioxide (CO2), version 2016(ODIAC 2016):A global, monthly fossil-fuel CO2 gridded emission data product for tracer transport simulations and surface flux inversions[J]. Earth Syst, 2018,10:87-107.
[4]	Janssens-Maenhout G, Crippa M, Guizzardi D, et al. Edgar v4.3.2global atlas of the three major greenhouse gas emissions for the period 1970~2012[J]. Earth Syst, 2017,11:959-1002.
[5]	Cai B, Liang S, Zhou J, et al. China high resolution emission database (CHRED) with point emission sources, gridded emission data, and supplementary socioeconomic data[J]. Resour. Conserv. Recycl, 2018, 129:232-239.
[6]	Wang R, Tao S, Ciais P, et al. High-resolution mapping of combustion processes and implications for CO2 emissions[J]. Atmos. Chem. Phys., 2013,13:5189-5203.
[7]	Han P, Zeng N, Oda T, et al. Evaluating China's fossil-fuel CO2 emissions from a comprehensive dataset of nine inventories[J]. Atmos. Chem. Phys., 2020,20:11371-11385.
[8]	Andres R.J, Boden T.A, Higdon D.M. Gridded uncertainty in fossil fuel carbon dioxide emission maps, a CDIAC example[J]. Atmos. Chem. Phys., 2016,16:14979-14995.
[9]	IPCC, 2019. 2019 Refinement to the 2006 IPCC guidelines for national greenhouse gas inventories[EB/OL]. https://www.ipcc.ch/report/2019-refinement-to-the-2006-ipcc-guidelines-for-national-greenhouse-gas-inventories/.
[10]	Cai W, Borlace S, Lengaigne M, et al. Increasing frequency of extreme El Niño events due to greenhouse warming[J]. Nature Climate Change, 2014,4:111-116.
[11]	Buchwitz M. Retrieval of trace gas vertical columns from SCIAMACHY/ENVISAT near-infrared nadir spectra:first preliminary results[J]. Advances in Space Research, 2004,34:809-814.
[12]	Crisp D, Miller C E, DeCola P L. NASA Orbiting Carbon Observatory:Measuring the column averaged carbon dioxide mole fraction from space[J]. J. Appl. Remote Sens, 2008,2:142-154.
[13]	Yoshida T, Yoshida Y, Eguchi N, et al. Global concentrations of CO2 and CH4 retrieved from GOSAT[J]. First preliminary results, 2009,5:160-163.
[14]	Frankenberg C, Fisher J B, Worden J, et al. New global observations of the terrestrial carbon cycle from GOSAT:Patterns of plant fluorescence with gross primary productivity[J]. Geophys, 2011,38:AR L17706.
[15]	Morino I, Uchino O, Inoue M, et al. Preliminary validation of column-averaged volume mixing ratios of carbon dioxide and methane retrieved from GOSAT short-wavelength infrared spectra[J]. Atmos. Meas. Tech., 2011,4:1061-1076.
[16]	Kort E A, Frankenberg C, Miller C E, et al., Space-based observations of megacity carbon dioxide[J]. Geophysical Research Letters, 2012,39(17):17806.
[17]	Belgiu M, Drăgu t L. Random forest in remote sensing:A review of applications and future directions[J]. ISPRS J. Photogramm.Remote Sens, 2016,114:24-31.
[18]	Tramontana G, Jung M, Schwalm C R, et al. Predicting carbon dioxide and energy fluxes across global FLUXNET sites with regression algorithms[J]. Biogeosciences, 2016,13:4291-4313.
[19]	Mahdianpari M, Salehi B, Mohammadimanesh F, et al. Random forest wetland classification using ALOS-2L-band, RADARSAT-2C-band, and TerraSAR-X imagery[J]. ISPRS J. Photogramm, Remote Sens, 2017,130:13-31.
[20]	Wen L, Cao Y. Influencing factors analysis and forecasting of residential energy-related CO2 emissions utilizing optimized support vector machine[J]. J. Clean. Prod., 2020,250:119492.
[21]	Leerbeck K, Bacher P, Junker R.G, et al. Short-term forecasting of CO2 emission intensity in power grids by machine learning[J]. Appl. Energy, 2020,277:115527.
[22]	Yang S Y, Lei L P, Zeng Z C, et al. An Assessment of Anthropogenic CO2 Emissions by Satellite-Based Observations in China[J]. Sensors, 2019,19:1118.
[23]	Mustafa F, Bu L, Wang Q, et al. Neural-network-based estimation of regional-scale anthropogenic CO2 emissions using an Orbiting Carbon Observatory-2(OCO-2) dataset over East and West Asia[J]. Atmos. Meas. Tech, 2021,14:7277-7290.
[24]	Sheng M Y, Lei L P, Zeng Z C, et al. Detecting the Responses of CO2 Column Abundances to Anthropogenic Emissions from Satellite Observations of GOSAT and OCO-2[J]. Remote Sens, 2021,13:3524.
[25]	Zeng Z C, Lei L P, Hou S S, et al. A regional gap-filling method based on spatiotemporal variogram model of columns[J]. IEEE Trans. Geosci. Remote Sens, 2014,52:3594-3603.
[26]	Zeng Z C, Lei L P, Strong K, et al. Global land mapping of satellite-observed CO2 total columns using spatio-temporal geostatistics[J]. Int. J. Digit. Earth, 2017,10:426-456.
[27]	吴长江,雷莉萍,曾招城.不同卫星反演的大气CO2浓度差异时空特征分析[J]. 中国科学院大学学报, 2019,36(3):331-337. Wu C J, Lei L P, Zeng Z C. Spatio-temporal analysis of differences among atmospheric CO2 concentrations retrieved from different satellite obserbvations[J]. Journal of University of Chinese Academy of Sciences, 2019,36(3):331-337.
[28]	Xiao J, Li X, He B, et al. Solar-induced chlorophyll fluorescence exhibits a universal relationship with gross primary productivity across a wide variety of biomes[J]. Glob. Change Biol., 2019,25:e4-e6.
[29]	Lee S, Cao C. Soumi NPP VIIRS day/night band stray light characterization and correction using calibration view data[J]. Remote Sensing, 2015,8(2):138.
[30]	Sims K, Reith A, Bright E, et al. LandScan Global 2021[D]. Oak Ridge National Laboratory, 2022.
[31]	Parzen E. On estimation of a probability density function and mode[J]. Annals of Mathematical Statistics, 1962,33(3):1065-1076.
[32]	Sheng M Y, Lei L P, Zeng Z C, et al. Detecting the responses of CO2 column abundances to anthropogenic emissions from satellite observations of GOSAT and OCO-2[J]. Remote Sens., 2021,13:3524.
[33]	Zhang S Q, Lei L P, Sheng M Y, et al. Evaluating Anthropogenic CO2 Bottom-Up Emission Inventories Using Satellite Observations from GOSAT and OCO-2[J]. Remote Sens., 2022,14:5024.
[34]	王震山,绳梦雅,肖薇,等.基于多源碳卫星融合产品的中国地区XCO2与人为CO2排放时空变化[J]. 中国环境科学, 2023,43(3):1053-1063. Wang Z S, Sheng M Y, Xiao W, et al. Spatiotemporal variations of XCO2 and anthropogenic CO2 emissions in China based on multi-source carbon satellite fusion products[J]. China Environment Science, 2023,43(3):1053-1063.
[35]	陶莹,杨锋,刘洋,等.K均值聚类算法的研究与优化[J]. 计算机技术与发展, 2018,28(6):90-92. Tao Y, Yang F, Liu Y, et al. Research and optimization of K-means clustering algorithm[J]. Computer Technology and Development, 2018,28(6):90-92.
[36]	Wheeler D, Ummel K. Calculating CARMA:Global estimation of CO2 Emissions from the power sector[J]. Working Papers, 2008, DOI:10.2139/ssrn.1138690.
[37]	Ummel K. CARMA revisited:An updated database of carbon dioxide emissions from power plants worldwide[M]. Center for Global development. 2012.
[38]	Chevallier F, Broquet G, Zheng B, et al. Large CO2 emitters as seen from satellite:Comparison to a gridded global emission inventory. Geophys. Res. Lett, 2020,49:e2021GL097540.
[39]	Jonas M, Marland G, Krey V, et al. Uncertainty in an emissions-constrained world[J]. Clim. Change, 2014,124:459-476.
[40]	Wang R, Tao S, Ciais P, et al. High-resolution mapping of combustion processes and implications for CO2 emissions[J]. Atmos. Chem. Phys., 2013,13:5189-5203.
[41]	Wang C, Corbett J.J, Firestone J. Improving Spatial Representation of Global Ship Emissions Inventories[J]. Environ. Sci. Technol., 2008, 42:193-199.
[42]	甄江红,成舜,郭永昌,等.包头市工业用地土地集约利用潜力评价初步研究[J]. 经济地理, 2004,(2):250-253. Zhen J H, Cheng S, Guo Y C, et al. Preliminary study on land intensive utilization potential evaluation of industrial land in Baotou City[J]. Economic Geography, 2004,(2):250-253.
[43]	段文娇,郎建垒,程水源,等.京津冀地区钢铁行业污染物排放清单及对PM2.5影响[J]. 环境科学, 2018,39(4):1445-1454. Duan W J, Lang J L, Cheng S Y, et al. Emission inventory of iron and steel industry and its impact on PM2.5 in Beijing-Tianjin-Hebei region[J]. Environmental Science, 2018,39(4):1445-1454.
[44]	李厚民,王登红,李立兴,等.中国铁矿成矿规律及重点矿集区资源潜力分析[J]. 中国地质, 2012,39(3):559-580. Li H M, Wang D H, Li L X, et al. Metallogenic regularity of iron ore and resource potential analysis of key ore concentration areas in China[J]. Geology in China, 2012,39(3):559-580.