Impact of wind fluctuations on the performance of the divergence method: How steady is the state?

The divergence, i.e. the spatial derivative of the horizontal flux, allows to identify point sources due to the strong local gradients, and to quantify emissions from satellite measurements of atmospheric pollutants or greenhouse gases such as NO2 or CH4.
A central assumption made in this approach is that steady state is fulfilled. I.e., spatio-temporal changes of emissions, chemical conditions, or wind fields are not accounted for. Thus it has to be expected that the emission estimate is affected and probably biased in case of deviations from steady state.

Here we investigate quantitatively how far deviations from steady state affect the results of the divergence method. In particular, we quantify the spatial and temporal variability of wind fields and relate them to NOx emission estimates for selected power plants based on individual TROPOMI orbits as well as on WRF-Chem simulations.
The goal is to provide a measure for "steadiness" that could be used to identify and mask out unfavourable conditions. With this filter, it is expected that the remaining emission estimates have lower uncertainties. Other methods for emission estimates that are based on steady state assumption as well, like the calculation of cross-sectional fluxes, will probably also benefit from this.