Long-term intercomparison of MIPAS additional species ClONO 2 , N 2 O 5 , CFC-11 , and CFC-12 with MIPAS-B measurements

The Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) aboard the environmental satellite ENVISAT is a limb-viewing Fourier-transform emission spectrometer working in the mid-infrared spectral region between 685 cm -1 and 2410 cm -1 [...]


I. INTRODUCTION
he Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) aboard the environmental satellite ENVISAT is a limb-viewing Fourier-transform emission spectrometer working in the mid-infrared spectral region between 685 cm -1 and 2410 cm -1 [Fischer et al. 2008].During the ten years of operation, MIPAS was able to observe temperature and a large number of climate-relevant trace gases from the upper troposphere up to the mesosphere (nominal observation mode) and even the thermosphere (special modes).For about two years after the launch of the satellite on 1 March 2002, MIPAS recorded spectra with full spectral resolution of 0.025 cm -1 .After anomalies in the velocity of the interferometer drive unit and a subsequent interruption of the measurements (March 2004), a lower (so-called optimized) spectral resolution of 0.0625 cm -1 together with a higher spatial resolution was chosen such that operational observations could restart in January 2005.Following the loss of communications with the satellite platform, there are no MIPAS data available since 8 April 2012.The description and characterization of the operational European Space Agency (ESA) processor products have been presented by Raspollini et al. [2006Raspollini et al. [ , 2013] ] and references therein.While previous operational product processing was limited to temperature and the molecules H 2 O, O 3 , HNO 3 , CH 4 , N 2 O, and NO 2 , the new ML2PP V6 data includes also the minor "additional" trace gases CFC-11 (CCl 3 F), CFC-12 (CCl 2 F 2 ), N 2 O 5 , and ClONO 2 .In this paper we focus on the comparison of these additional species.Several flights with MIPAS-B, the balloon version of MIPAS, have been carried out during the operational period of ENVISAT.Table 1 gives an overview of MIPAS-B flights where MIPAS-B observations coincide in space and time with MIPAS on ENVISAT (MIPAS-E) measurements.MIPAS-B limb sequences recorded during these flights are compared to the corresponding MIPAS-E overpasses.MIPAS-B was strongly involved in the ENVISAT validation activities and the high accuracy of the MIPAS-B data has been assessed in numerous MIPAS-E related validation studies [see, e.g., Cortesi et al. 2007, Höpfner et al. 2007, Milz et al. 2009, Payan et al. 2009, Ridolfi et al. 2007, Steck et al. 2007, Wang et al. 2007a, 2007b, Wetzel et al., 2007, 2013, Zhang et al. 2010a, 2010b Essential for the balloon instrument is the sophisticated line of sight stabilization system, which is based on an inertial navigation system and supplemented with an additional star reference system leading to an after all knowledge of the tangent altitude in the order of 90 m (3 σ).The MIPAS-B data processing from interferograms to calibrated spectra including instrument characterization is described in Friedl-Vallon et al. [2004] and references therein.The measurements were done typically at a 1.5 km vertical grid.Retrieval calculations of atmospheric target parameters were performed at a 1 km grid with a least squares fitting algorithm using analytical derivative spectra calculated by the Karlsruhe Optimized and Precise Radiative transfer Algorithm [KOPRA; Stiller et al. 2002;Höpfner et al. 2002].A Tikhonov-Phillips regularization approach [Phillips, 1962, Tikhonov, 1963] constraining with respect to a first derivative a priori profile was adopted.The resulting vertical resolution is typically between 2 and 5 km for the analyzed species and is therefore comparable to or slightly better than the vertical resolution of MIPAS-E.The species ClONO 2 , CFC-11, CFC-12, and N 2 O 5 are analyzed in the spectral windows 779.7 -780.7 cm -1 , 840 -860 cm -1 , 918 -924 cm -1 , and 1220 -1270 cm -1 , respectively.Different spectral windows within these molecular bands were used for the MIPAS-E data analysis [Raspollini et al. 2013].Spectroscopic parameters for the calculation of limb emission spectra originate from the highresolution transmission (HITRAN) molecular absorption database [Rothman et al. 2005].The MIPAS-B error estimation includes random noise as well as the mutual influence (covariance) of the fitted parameters, temperature errors, pointing inaccuracies, errors of nonsimultaneously fitted interfering gases, and spectroscopic data errors (1 σ).For details on the MIPAS-B data analysis and error estimation, see Wetzel et al. [2012] and references therein.
Since pressure is the primary vertical coordinate of MIPAS-E, vertical profiles of the satellite sensor have been interpolated to the MIPAS-B pressure-altitude grid.Differences between measured volume mixing ratios (VMRs) of MIPAS-E and the validation instrument MIPAS-B are expressed in absolute units and relative differences.The mean difference Δx mean for N profile pairs of compared observations is given as: where x E and x B are VMR values of MIPAS-E and MIPAS-B at one altitude level.The mean relative difference Δx mean,rel of a number of profile pairs is calculated by dividing the mean absolute difference by the mean profile value of the validation instrument MIPAS-B: (2) Differences are displayed together with the combined errors σ comb of both instruments which are defined as: where σ E and σ B are the precision, systematic or total errors of MIPAS-E and the validation instrument, respectively.Precision errors characterize the reproducibility of a measurement and correspond, in general, to random noise errors.Systematic errors of the V6 processor used for MIPAS-E data have been assessed in a corresponding ESA study [Raspollini et al. 2013].The uncertainty of the calculated mean difference (standard error of the mean, SEM) is given by σ/N 0.5 where σ is the standard deviation (SD).A bias between both instruments is considered significant if the SEM is smaller than the bias itself.
The comparison between the VMR difference and the combined systematic error (for statistical comparisons) or total error (for single comparisons) is appropriate to identify unexplained biases in the MIPAS-E observations when they exceed these combined error limits.
Since the vertical resolution of the used VMR profiles of both instruments is of comparable magnitude, no smoothing by averaging kernels has been applied for the intercomparison of the observed profiles.The agreement between the measured profiles appears to be quite good with a mean deviation over all altitudes of 7.2 %.Some biases are visible in limited altitude regions but they are not significant with regard to the combined error limits.

III. INTERCOMPARISON RESULTS
Figure 2 displays the differences concerning the molecule N 2 O 5 .This molecule exhibits a pronounced diurnal variation in its mixing ra-tio.Hence, a photochemical correction using a box model [Wetzel et al. 1995], which takes into account the nighttime variation of N 2 O 5 , has been applied to the tropical MIPAS-E measurements where both, diurnal variation and the time difference of the observations (see Table 1), are quite large.Apart from the lowermost altitude region between 15 and 17 km (where only 2 collocations are available) there is no significant bias visible between the balloon and satellite measurements of MIPAS.
The overall agreement between both sensors is fairly good with a mean deviation of -6.7 %.
The comparison for the species CFC-11 is shown in Figure 3.In contrast to the previously discussed molecules, significantly higher values are visible in the MIPAS satellite data in the complete altitude region under comparison.While below 19 km this positive bias is still within the combined systematic and total errors, it clearly exceeds these error limits above and therefore remains unexplained.Figure 4 displays the observed differences for the molecule CFC-12.Unlike the case of CFC-11, the agreement of measured CFC-12 VMR is quite good with a mean difference of 11.5 %.No significant biases are visible at all.

IV. CONCLUSIONS
Vertical profiles of six MIPAS balloon flights between 2002 and 2010 have been used for an intercomparison study of the species ClONO 2 , N 2 O 5 , CFC-11, and CFC-12, which have been processed for the first time by the operational ESA processor (ML2PP V6).
For the molecules ClONO 2 , N 2 O 5 , and CFC-12 the agreement between both MIPAS instruments is generally good.No clear biases in the observed mixing ratios are visible.Furthermore, no seasonal or latitudinal dependence of deviations could be recognized.However, standard deviations are slightly larger than the combined precision errors.Concerning the species CFC-11 a significant positive bias is evident at all altitudes under comparison.Above 19 km, this bias clearly exceeds the combined systematic and total errors and remains unexplained.Hence, this molecule cannot be recommended for scientific users at the present state.Investigations are ongoing within the ESA processor development team to understand the reason of the bias.It should also be mentioned that some single MIPAS-E profiles of these additional species exhibit retrieval oscillations which show up in VMR differences (in limited altitude regions) to MIPAS-B observations but which may in turn be smoothed out in comparisons of mean differences.Scientific users should therefore be careful in their interpretation when using single MIPAS-E profiles of these trace gases.

Figure 2 :Figure 3 :
Figure 2: Same as Figure 1, but for the molecule N 2 O 5 .The overall agreement of both sensors is fairly good.