Fluxes of Atmospheric Greenhouse‐Gases in Maryland (FLAGG‐MD): Emissions of Carbon Dioxide in the Baltimore, MD‐Washington, D.C. Area.

To study emissions of CO2 in the Baltimore, MD‐Washington, D.C. (Balt‐Wash) area, an aircraft campaign was conducted in February 2015, as part of the Fluxes of Atmospheric Greenhouse‐Gases in Maryland (FLAGG‐MD) project. During the campaign, elevated mole fractions of CO2 were observed downwind of the urban center and local power plants. Upwind flight data and Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model analyses help account for the impact of emissions outside the Balt‐Wash area. The accuracy, precision, and sensitivity of CO2 emissions estimates based on the mass balance approach were assessed for both power plants and cities. Our estimates of CO2 emissions from two local power plants agree well with their Continuous Emissions Monitoring Systems (CEMS) records. For the 16 power plant plumes captured by the aircraft, the mean percentage difference of CO2 emissions was −0.3%. For the Balt‐Wash area as a whole, the 1𝜎 CO2 emission rate uncertainty for any individual aircraft‐based mass balance approach experiment was ±38%. Treating the mass balance experiments, which were repeated seven times within 9 days, as individual quantifications of the Balt‐Wash CO2 emissions, the estimation uncertainty was ±16% (standard error of the mean at 95% CL). Our aircraft‐based estimate was compared to various bottom‐up fossil fuel CO2 (FFCO2) emission inventories. Based on the FLAGG‐MD aircraft observations, we estimate 1.9 ± 0.3 MtC of FFCO2 from the Balt‐Wash area during the month of February 2015. The mean estimate of FFCO2 from the four bottom‐up models was 2.2 ± 0.3 MtC. Key Points: 1.9 ± 0.3 MtC of fossil fuel CO2 was emitted in Baltimore‐Washington during February 2015 based on data collected during seven aircraft flightsFour bottom‐up inventories indicate 2.2 ± 0.3 MtC of fossil fuel CO2 was emitted, in good agreement with our top‐down estimateThe uncertainty from a single flight segment was ±38% (1𝜎); data from seven flights yielded a precision of 16% at the 95% confidence level [ABSTRACT FROM AUTHOR]