A Modeling Framework for Inference of Surface Emissions Using Mobile Observations

Our ability to quantify surface emissions depends on the precision of observations and the spatial density of measurement networks. Mobile measurement techniques offer a cost effective strategy for quantifying atmospheric conditions over space without requiring a dense network of in-situ sites. However, interpretation of these data and inversion of dispersed measurements to estimate surface emissions can be difficult. We introduce a framework using the Stochastic Time-Inverted Lagrangian Transport (STILT) model that assimilates both spatially resolved observations and an emissions inventory to better estimate surface fluxes. Salt Lake City is a unique laboratory for the study of urban carbon emissions. It is the only U.S. city that utilizes light-rail trains to continuously measure high frequency carbon dioxide (CO2) and methane (CH4); it is home to one of the longest and most spatially resolved high precision CO2 measurement networks (air.utah.edu); and it is one of four cities in the world for which the Hestia anthropogenic emissions inventory has been produced which characterizes CO2 emissions at the scale of individual buildings and roadways. Using these data and modeling resources, we evaluate spatially resolved CO2 measurements and transported CO2 emissions on hourly timescales at a dense spatial resolution across Salt Lake City.