Modeling the Atmospheric CO2 Concentration in the Beijing Region and Assessing the Impacts of Fossil Fuel Emissions

Reducing anthropogenic fossil fuel CO2 (FFCO2) emissions in urban areas is key to mitigating climate change. To better understand the spatial characteristics and temporal variations in urban CO2 levels in the Beijing (BJ) region, we conducted a long-term CO2 simulation study by using the Weather Research and Forecasting WRF-Chem model and CO2 observation data. To assess the model performance, three representative sites with high-precision CO2 observation data were chosen in this study: the rural regional background Shangdianzi (SDZ) site, the suburban Xianghe (XH) site, and the urban BJ site. The simulation results generally captured the observed variations at these three sites, but the model performed much better at the SDZ and XH sites, with mean biases of −0.7 ppm and −2.3 ppm, respectively, and RMSE of 12.3 ppm and 21.4 ppm, respectively. The diurnal variations in the model results agreed well with those in the observed CO2 concentrations at the SDZ and XH sites during all seasons. In the meanwhile, the diurnal variations in the modeled FFCO2 were similar to those in the CO2 observation with a positive bias at the BJ site, which may have been caused by higher emissions especially in winter. Moreover, both the modeled FFCO2 and biospheric CO2 (BIOCO2) have positive correlations with the observed CO2 concentration, whereas the planetary boundary layer height (PBLH) and observed CO2 concentration exhibited negative correlations at all sites. In addition, the contributions of FFCO2 and BIOCO2 to CO2 varies depending on the seasons and the location of sites.