Estimating anthropogenic CO2 emissions from New York City using aircraft measurements and dispersion modelling

Abstract:

Greenhouse gas mitigation strategies and associated legislation require reliable methods for emission quantification including complementary bottom-up and top-down approaches. This is important for assessment of our knowledge of sources, but also to assess rates of progress. In the latter case, method precision is critically important. Towards this end, we have performed a series of winter aircraft measurement flights downwind of New York City. We use Stochastic Time-Inverted Lagrangian Transport (STILT) model runs driven by publicly available meteorology products to calculate footprints relevant to the aircraft samples. To quantify emissions, we combine these footprints with four common emission model priors (ODIAC, EDGAR, ACES, and Vulcan) at the hourly scale and calculate a scaling factor that optimizes agreement with measurements for each downwind pass. Here we will discuss the posterior New York City emissions as well as the potential of the technique to assess changes in emissions over time. We also compare the performance of the multiple meteorology products, scaling factor calculation methods, and background definitions across all inventories and flight days using common goodness of fit metrics (root mean square error, mean absolute error, standard deviation of bias, etc.) to identify those most capable of recreating the measured enhancements.