Abstract
Satellite observations of NO2 from geostationary orbit provide an opportunity to study daytime emissions with hourly temporal resolution. We used NO2 column measurements from Tropospheric Emissions: Monitoring of Pollution (TEMPO) with a new fitting function, the rise-decay (RD) model, to derive NOx emissions at 64 power plants in the continental United States. The RD model separately fits the rise of NO2 line densities immediately downwind of the source and the decay of NOx further downwind. This point source-specific function more reliably constrained emissions than the exponentially modified Gaussian (EMG) function, a typical method used to estimate NOx emissions from space. The RD-derived emissions were compared with in situ emission measurements from the Continuous Emissions Monitoring Systems (CEMS). Average emissions were quantified within 50% accuracy at 27 of the power plants using TEMPO. Bias against CEMS was negatively correlated with emitter strength and was substantial at plants emitting less than 250 kg NO2 per hour. The TEMPO-derived emissions reproduce the diurnal and seasonal patterns of CEMS reported emissions. These results demonstrate the utility of the rise-decay method for point source emitters and indicate the limits of TEMPO NO2 for quantifying NOx emissions.