Enhancing CO2 flux inversion accuracy in a coastal megacity: Impacts of land cover and topography in a regional carbon assimilation system

Human activities have significantly increased carbon dioxide (CO2) emissions, exacerbating global warming. Therefore, accurately estimating anthropogenic emissions is crucial for developing effective emission reduction policies. This study focuses on evaluating the impact of land cover (LC) and topography (DEM) on posterior CO2 fluxes using a newly developed regional Carbon Assimilation System (RCAS). Our primary aim is to assess the extent to which LC and DEM influence CO2 flux inversion in Shenzhen, a rapidly urbanizing coastal city. To achieve this, we analyzed the forward simulations of meteorological variables (wind speed/direction, relative humidity, temperature, and pressure) alongside CO2 concentration simulations. Following this, we evaluated the posterior CO2 concentration estimates with the CO2 flux from the inversion system. The findings demonstrate that new LC and DEM data led to a mean bias reduction from −1.08 ppm to −0.88 ppm, representing a 0.2 ppm improvement, and a slight decrease in standard deviation values from 5.59 ppm to 5.57 ppm in the posterior CO2 concentration simulation. These results underscore the importance of accurate LC and DEM data in enhancing CO2 flux inversion accuracy, providing valuable insights for improving regional-scale carbon monitoring systems.