Analysis of ground-based column and in situ surface concentrations of CO2 at Xianghe, China, using WRF-Chem simulations

Since June 2018, ground-based remote sensing measurements are performed at the suburban Xianghe site in China, situated in the heart of the densely populated Beijing-Tianjin-Hebei megalopolis. These observations are performed with Fourier Transform Infrared (FTIR) spectrometers and provide  column-averaged dry-air concentrations of gases such as CO2, CH4 and CO. They are affiliated to the international Total Column Carbon Observing Network (TCCON). Co-located with these measurements is a PICARRO cavity ring-down spectroscopy (CRDS) analyser observing in situ concentrations of CO2 and CH4 at an altitude of 60 m.To gain a better understanding of the causes of the observed temporal variabilities at this site, we employed the Weather Research and Forecasting model coupled with Chemistry in its greenhouse gas configuration (WRF-GHG). Our study analyses both column-averaged (XCO2) and surface in situ CO2 concentrations and simultaneously evaluates the model’s performance at Xianghe.  The CO2 exchange with the biosphere is simulated with the integrated Vegetation Photosynthesis and Respiration Model (VPRM), while the anthropogenic emissions are taken from the global CAMS-GLOB-ANT inventory and transported in separate tracers according to their source sector. The model shows good performance, achieving correlation coefficients of 0.70 for XCO2 and 0.75 for afternoon in situ concentrations. For XCO2, a mean bias of -1.43 ppm relative to TCCON is found, primarily attributed to biases in the CAMS reanalysis used as initial and lateral boundary conditions. Anthropogenic emissions from the industry and energy sectors emerged as dominant contributors to CO2 concentrations, alongside the biosphere, which acts as a sink for XCO2 from April to September and becomes a source for the rest of the year. Synoptic weather patterns were shown to strongly determine the variation in CO2 levels, with enhanced impacts during summer due to the large spatial and temporal heterogeneity of biogenic fluxes in the region. Near the surface, the observed large diurnal variation associated to the evolution of the planetary boundary layer is  relatively well simulated by WRF-GHG.Our analysis demonstrates the utility of WRF-GHG in simulating both column and surface CO2 concentrations, offering insights into the sectoral and meteorological drivers of variability at Xianghe and its surrounding region.