Constraining Inversions of Rural and Urban CO2 Fluxes Using EM27/SUN XCO2 Measurements

Abstract:

Reducing CO₂ emissions is critical to combat global warming, and mitigation requires quantification. In rural areas, CO₂ fluxes are driven by vegetation activity. In urban areas, although the majority of CO₂ are attributable to anthropogenic emissions, excluding biogenic fluxes can cause significant bias especially during the growing season. Ground-based EM27/SUN spectrometers deployed in Metro Boston provide 2 years of daytime column-averaged CO₂ concentrations (XCO₂) at Harvard Forest and William James Hall, a remote rural site and an urban site, which allows the calculation of rural-urban CO₂ gradients. Here we structure a Bayesian inverse model framework that uses EM27/SUN XCO₂ measurements as new constraints. The Stochastic Time-Inverted Lagrangian Transport (STILT) model with the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) setup is used to calculate the Jacobian, representing the sensitivity of observations to emissions. We apply this framework with the high temporal and spatial resolution Anthropogenic Carbon Emissions System (ACES) inventory and detailed biological fluxes produced by the Vegetation Photosynthesis Respiration Model (VPRM) to improve modeling accuracy of CO₂ emissions. The two distinct locations are both investigated with the inversion framework to examine the changes of regional emissions and to assess the capability of EM27/SUN observations to resolve emissions under different conditions. The roles of anthropogenic and biogenic fluxes on urban CO₂ in different seasons are also evaluated.