Estimating NOx emissions of stack plumes using a high-resolution atmospheric chemistry model and satellite-derived NO2 columns

This work contributes to an European Monitoring and Verification Support (MVS) capacity for anthropogenic CO2 emissions. Future satellite instruments that map CO2 and NO2 from space will focus on hot-spot emissions from cities and large point sources, where CO2 emissions are accompanied by emissions of NOx. To use NOx as proxy CO2 emission, information about its atmospheric lifetime and the fraction of NOx present as NO2 is required to interpret NO2 plumes. This paper presents Large Eddy Simulations with atmospheric chemistry of four large point sources world-wide. We find that the chemical evolution of the plumes depends strongly on the amount of NOx that is emitted, next to wind speed and direction. For large NOx emissions the chemistry is pushed in a high-NOx chemical regime over a length of almost 100 km downwind of the stack location. Other plumes with lower NOx emissions show a fast transition to an intermediate NOx chemical regime, with short NOx lifetimes. Simulated NO2 columns mostly agree within 20 % with the TROPOspheric Monitoring Instrument (TROPOMI), signalling that the emissions used in the model were approximately correct. However, variability in the simulations is large, making a one-to-one comparison difficult. We find that wind speed variations should be accounted for in emission estimation methods. Moreover, results indicate that common assumptions about the NO2 lifetime (≈4 hours) and NOx: NO2 ratios (≈1.3) in simplified methods that estimate emissions from NO2 satellite data (e.g. Beirle et al., 2019) need revision.