Exploring XCO2 spatial variability across frontal boundaries using airborne lidar observations and numerical simulations

A detailed understanding of the transport processes of greenhouse gases (e.g., Carbon Dioxide (CO2), Methane (CH4)) in the Earth’s atmosphere is crucial during this current regime of dramatically changing climate and associated global warming. Current observations of CO2 are sparse for both global and regional scale coverages, thus numerical models are used to bridge the gap and provide global CO2 concentration estimates. However, because there is a lack of high-resolution observations, uncertainties in the models exist, in particular, associated with the transport of greenhouse gases via mid-latitude cyclones. Previous research identified the impact of mid-latitude cyclones on the CO2 spatial variability in the atmospheric boundary layer and overlying free troposphere. However, some key issues remain unclear: (1) how the column average CO2 dry air mole fraction (XCO2) changes spatially across frontal boundaries over different ecoregions, (2) how well numerical models (e.g., Weather Research and Forecasting-Chemistry, WRF-Chem) and assimilated products reproduce XCO2 frontal structures. We investigated the XCO2 spatial variability across synoptic systems in four seasons over the eastern US using the airborne Multifunctional Fiber Laser Lidar (MFLL) measurements obtained during four field deployments of the Atmospheric Carbon and Transport-America (ACT-America) project. Most cases in the summer revealed higher XCO2 in the warm sector than in the cold sector, whereas in the winter higher values of XCO2 are noted in the cold sector versus the warm sector for most cases. During the transitional seasons (spring and fall), no clear signal in XCO2 frontal contrasts was observed due to the spatial variability (south to north) in phenology over the broad regions of the eastern US. A detailed intercomparison among the XCO2 fields obtained with MFLL, WRF-Chem simulations, and in situ data-driven Global Modeling and Assimilation Office (GMAO) assimilations suggest that both WRF-Chem and GMAO retrieved XCO2 reproduce the overall frontal structures reasonably well though with varying degrees of model-data-mismatches obtained for frontal contrasts in XCO2 field (i.e., warm sector minus cold sector averaged XCO2). The analyses revealed that WRF-Chem tends to underestimate XCO2 fields in both warm and cold sectors of the frontal system by an average difference of -0.852 ppm and 0.851 ppm, respectively. One should note that the GMAO system assimilated the in situ observations of CO2 obtained during the campaigns. The results demonstrate the feasibility of studying XCO2 spatial variability within synoptically active environments in different seasons using airborne lidar measurements. The associated model-data-mismatch on the XCO2 frontal contrasts provide guidance to the spatial structures of XCO2 transport errors in models and satellite measurements of XCO2 in synoptically active weather systems.