Comparison and evaluation of modelled and GOME measurement derived tropospheric NO<sub>2</sub> columns over Western and Eastern Europe

Abstract

Abstract. We present the results of a first comparison of the tropospheric NO 2 column amounts derived from the measurements of the Global Ozone Monitoring Experiment (GOME) with the simulated data from a European scale chemistry transport model (CTM) which is distinct from existing global scale CTMs in higher horizontal resolution and more detailed description of the boundary layer processes and emissions. We employ, on the one hand, the newly developed extended version of the CHIMERE CTM, which covers both Western and Eastern Europe, and, on the other hand, the most recent version (Version 2) of GOME measurement based data-products, developed at the University of Bremen. We evaluate our model with the data from ground based monitoring of ozone and verify that it has a sufficiently high level of performance, which is expected for a state-of-theart continental scale CTM. The major focus of the study is on a systematic statistical analysis and a comparison of spatial variability of the tropospheric NO 2 columns simulated with CHIMERE and derived from GOME measurements. The analysis is performed separately for Western and Eastern Europe using the data for summer months of 1997 and 2001. In this way, we obtain useful information on the nature and magnitudes of uncertainties of spatial distributions of the considered data. Specifically, for Western Europe, it is found that the uncertainties of NO 2 columns from GOME and CHIMERE are predominantly of the multiplicative character, and that the mean relative random (multiplicative) errors of the GOME measurement derived and simulated data averaged over the summer seasons considered do not exceed 23% and 32%, respectively. The mean absolute (additive) errors of both kinds of the data are estimated to be less than Correspondence to: I. B. Konovalov (konov@appl.sci-nnov.ru) 3×10 14 mol/cm 2 . In Eastern Europe, the uncertainties have more complex character, and the separation between their multiplicative and additive parts is not sufficiently unambiguous. It is found, however, that the total random errors of NO 2 columns from both GOME and CHIMERE over Eastern Europe are not, on the average, larger than the errors of the NO 2 columns with similar magnitudes over Western Europe.

Inverse modelling of the spatial distribution of NO<sub>x</sub> emissions on a continental scale using satellite data

“…The assumption regarding the multiplicative character of errors seems to be appropriate in case of emissions because the standard methods used for evaluations of emissions involve multiplying several different factors, each of which may contribute to the total uncertainty of emission inventory data (see, e.g., Kühlwein and Friedrich, 2000). Our previous analysis (Konovalov et al, 2005) has shown that the uncertainties in tropospheric NO 2 columns derived from GOME measurements and calculated by a CTM for Western Europe are also predominantly of multiplicative character. The lognormal distribution for errors appears to be the optimal choice in a case of strictly positive physical characteristics (Mosegaard and Tarantola, 2002).…”

“…The potential uncertainties in simulated tropospheric NO 2 columns due to omission of the upper troposphere have been considered in Konovalov et al (2005). In particular, it has been shown that the spatial variability of tropospheric NO 2 above 500 hPa pressure level contributes only very insignificantly to the spatial variability of total tropospheric NO 2 columns.…”

“…2.1, a considerable part of this bias may be due to omission of the upper troposphere in CHIMERE. If we admit that the mean column amount of NO 2 above 500 hPa pressure level is about 8×10 14 m −2 (Konovalov et al, 2005), then the residuary part of the bias is about 10 percent. It remains unknown whether this bias is due to some systematic errors in the model, in emissions, or in the satellite data.…”

“…The GOME NO 2 columns utilized in this study have been retrieved with the same method as those used in our earlier study (Konovalov et al, 2005), in which the tropospheric NO 2 columns derived from GOME measurements were analysed together with NO 2 columns simulated by CHIMERE. The distinctive features of this method are (1) the estimation of a stratospheric part of the slant columns based on simulations with the global CTM SLIMCAT (Chipperfield et al, 1999), (2) the use of NO 2 vertical distributions simulated by the global CTM MOZART (Horowitz et al, 2003) and (3) the use of cloud parameters and surface albedo obtained from GOME measurements using the algorithms discussed by Koelemeijer et al (2001) and Koelemeijer et al (2003).…”

Inverse Modeling of Nitrogen Oxides Emissions from the 2010 Russian Wildfires by Using Satellite Measurements of Nitrogen Dioxide

“…This allows us to minimize the effect of possible biases (due to, e.g., uncertainties in anthropogenic and biogenic NOx emissions, as well as in the boundary conditions for NOx and related species) in the background simulations with CHIMERE for the study region [42,43] on our estimates. Note that Cmf involved in Equation (2) was obtained as the difference between the results of two "twin" model runs with and without fire emissions (as mentioned in Section 2.1.2), and so Cmf also could not be directly influenced by possible inaccuracies of the simulations for the background conditions.…”