Inverse modelling of the spatial distribution of NOx emissions on a continental scale using satellite data

摘要

The recent important developments in satellite measurements of thecomposition of the lower atmosphere open the challengingperspective to use such measurements as independent information onsources and sinks of atmospheric pollutants. This study exploresthe possibility to improve estimates of gridded NOemissions used in a continental scale chemistry transport model(CTM), CHIMERE, by employing measurements performed by the GOMEand SCIAMACHY instruments. We set-up an original inverse modellingscheme that not only enables a computationally efficientoptimisation of the spatial distribution of seasonally averagedNO emissions (during summertime), but also allowsestimating uncertainties in input data and a priori emissions. Thekey features of our method are (i) replacement of the CTM by a setof empirical models describing the relationships betweentropospheric NO columns and NO emissions withsufficient accuracy, (ii) combination of satellite data fortropospheric NO columns with ground based measurements ofnear surface NO concentrations, and (iii) evaluation ofuncertainties in a posteriori emissions by means of a specialBayesian Monte-Carlo experiment which is based on random samplingof errors of both NO columns and emission rates. We haveestimated the uncertainty in a priori emissions based on the EMEPemission inventory to be about 1.9 (in terms of geometric standarddeviation) and found the uncertainty in a posteriori emissionsobtained from our inverse modelling scheme to be significantlylower (about 1.4). It is found also that a priori NOemission estimates are probable to be persistently biased in manyregions of Western Europe, and that the use of a posterioriemissions in the CTM improves the agreement between the modelledand measured data.