A Comparative Spatiotemporal Analysis Of XCO₂ In China’s Major Southern Cities Based On Remote Sensing And Atmospheric Model

Global warming is a major environmental issue facing mankind to date,and China has proposed a"3060 carbon"target against the backdrop of increased global warming.In order to achieve the commitment of"carbon neutrality"and"carbon peaking"project,it is necessary to monitor greenhouse gas emissions properly.CO₂ is the most abundant greenhouse gas in the atmosphere and the most important driver of climate change among the six categories of greenhouse gases identified by the United Nations.Studies have shown that cities are the largest source of energy consumption and greenhouse gas emissions,and that urban carbon management is central to emissions reduction.Therefore,it is important for China to understand urban carbon emissions and CO₂ distribution in order to address climate change and develop a low-carbon economy.With the successful launch of greenhouse gas satellites in various countries,long time series and wide-range CO2 concentrations can be observed.Compared with traditional greenhouse gas monitoring approaches,satellite remote sensing monitoring is not limited by time and location,covers a wider area and does not require a large amount of human assistance,and thus providing a larger range of observation information under the conditions of sparse ground observation points.However,the satellite observation has obvious striping phenomenon,incomplete data coverage and large gap scale between strips,which may affect the urban scale analysis.This study aims to find a proper gap filling method for L2 CO₂ column concentration data from GOSAT and OCO-2,using the factors that may affect the distribution of CO₂ as the driven factor for gap filling.The spatial and temporal comparisons of major cities in the southern region were carried out based on XCO₂-RF dataset.In this study,XCO₂-RF analysis dataset was constructed based on GOSAT and OCO-2 satellite observations and the simulation results of the WRF-Chem atmospheric chemical transport model.The spatially continuous XCO₂-RF dataset covering southern China（16°N-32°N,106°E-122°E）from January 2010 to December 2019 was established by constructing a random forest spatial interpolation algorithm with the results of the model as the independent variable and the satellite observations as the dependent variable.Then,the 27 cities in the southern region were selected for the study area,and the monthly mean CO₂ concentrations in each city were analyzed to derive the spatial and temporal trends of CO₂ concentrations in each city,and the correlation between the CO₂ changes and the degree of urban expansion in each city was carried out.The main findings of the study are as follows:（1）The data accuracy of the XCO₂-RF dataset was evaluated based on ground-based observation stations and a reconstructed XCO₂ L3 global dataset published by the National Earth System Science Data Centre based on the empirical orthogonal method.the XCO₂-RF results have some errors with the ground-based stations TCCON/HF and WDCGG/LLN,but the overall trend is consistent,with the RMSE of 2.44 ppm.The accuracy of the XCO₂-RF dataset is comparable to the global XCO₂ L3 dataset,with an average RMSE error of about2.5 ppm,or about 0.625%,indicating that the values of the XCO₂-RF dataset are reliable in general.The accuracy of the XCO₂-RF data set is higher than that of the two geostatistical interpolation methods（IDW and Kriging）,indicating that the XCO₂ data set based on the WRF-Chem atmospheric model is superior to common geostatistical interpolation method.（2）Based on the XCO₂-RF analysis dataset,the spatial and temporal characteristics of CO₂ distribution in 27 major cities in the southern region were analyzed,as well as the correlation between CO₂ and urban construction land area changes among different cities.The results show that XCO₂ in the 27 cities has a similar trend,with the annual average concentration gradually increasing from 388.5 ppm to 414.72 ppm with a growth rate of 6.7%.Although the growth trend is generally consistent across the 27 cities,the growth rate of CO₂varies slightly between cities,with the highest growth rate in Fuzhou and the lowest in Kunming.It is noteworthy that the growth rates of provincial capitals tend to be higher than those of other cities in the province,and cities located in the central-eastern region tend to have higher growth rates than those in the south-western region,which basically corresponds to the level of economic development of cities in the south.In addition,there is a strong correlation between CO₂ growth trends and the expansion of urban built-up land area in all 27cities,while the CO₂ growth rate in cities with an increase in individual urban green areas does not slow down the rate of CO₂ increase in that city.