A128-06
New York City greenhouse gas emissions estimated with inverse modelling of aircraft measurements
Friday, 11 December 2020: 10:45
Virtual
Joseph Robert Pitt1, Israel Lopez-Coto2, Kristian Daniel Daniel Hajny3, Jay Mendoza Tomlin3, Robert Kaeser4, Thilina Jayarathne3, Brian H Stirm4, Cody R Floerchinger5, Christopher Loughner6, Roisin Commane7, Conor Gately8, Lucy Hutyra9, Kevin R Gurney10, Geoffrey S Roest10, Jianming Liang11, Anna Karion2, James R Whetstone2 and Paul Shepson1,3, (1)Stony Brook University, School of Marine and Atmospheric Sciences, Stony Brook, NY, United States, (2)National Institute of Standards and Technology Gaithersburg, Gaithersburg, MD, United States, (3)Purdue University, Department of Chemistry, West Lafayette, IN, United States, (4)Purdue University, School of Aviation and Transportation Technology, West Lafayette, IN, United States, (5)Harvard University, Department of Earth and Planetary Sciences, Cambridge, MA, United States, (6)NOAA Air Resources Laboratory, College Park, MD, United States, (7)Columbia University in the City of New York, New York, NY, United States, (8)Metropolitan Area Planning Council, Boston, MA, United States, (9)Boston University, Earth & Environment, Boston, MA, United States, (10)Northern Arizona University, School of Informatics, Computing, and Cyber Systems, Flagstaff, AZ, United States, (11)Environmental Systems Research Institute, Redlands, CA, United States
Abstract:
An ever-increasing majority of the world’s population lives in urban areas. Consequently, cities are greenhouse gas emission hotspots, making them a target for emission reduction policies. Effective emission reduction policies must be supported by accurate and transparent emission accounting. Top-down approaches to emission estimation, based on atmospheric greenhouse gas measurements, are an important and complementary tool to assess, improve and update the emission inventories on which policy decisions are both based and assessed.
In this study we present results from nine research flights (pre-COVID lockdown) measuring CO2 and CH4 around New York City during the non-growing season. We used a new version of the dispersion model HYSPLIT (Hybrid Single Particle Lagrangian Integrated Trajectory Model) in a Bayesian inverse modelling framework to derive posterior emission estimates, using three prior emission maps for CO2 and four prior emission maps for CH4. An ensemble of eight transport models, consisting of four different met products combined with two different parameterisations of boundary layer turbulence, was used to estimate the transport model uncertainty. We employed a nested domain approach to improve our representation of the background, first optimising emissions within a large regional domain, then using these optimised emissions to provide a spatially varying modelled background for a smaller high-resolution domain focussed on New York City.
Our results show good agreement between posterior estimates using different prior emission maps and transport model configurations. The flight-to-flight variability in posterior emission is similar to that observed in previous studies. Posterior CO2 emissions are found to be higher than our prior annual inventory values, consistent with the expectation of greater emissions during winter afternoons. Estimated CH4 emissions are found to be significantly greater than the gridded EPA inventory, supporting the conclusions of previous work suggesting CH4 emissions from cities throughout the northeast of the USA are currently underestimated.
