A. Linden

Sensitivity of trace gas abundances retrievals from infrared limb emission spectra to simplifying approximations in radiative transfer modelling

The accuracy of trace gas abundances retrieved from spectrally resolved infrared limb emission measurements relies, among other things, on the appropriate modelling of radiative transfer through the actual atmosphere. We quantify the mapping of several commonly applied simplifications in radiative transfer modelling on the trace gas abundances retrieval error at the example of the Michelson interferometer for passive atmospheric sounding (MIPAS)/environmental satellite (ENVISAT) space borne Fourier transform infra-red limb emission experiment. The Karlsruhe optimized and precise radiative transfer algorithm (KOPRA) which was used as a tool for this study will be introduced. KOPRA supports accurate modelling of the particular instrument requirements of MIPAS and the observation scenarios during the ENVISAT mission, in particular with respect to its viewing direction and its altitude coverage of the atmosphere. We show that disregarding of horizontal temperature inhomogeneities and non-local thermodynamic equilibrium effects, insufficient accuracy in modelling of field-of-view and apodisation effects, and disregarding individual profiles of isotopic species play the key roles in radiative transfer modelling and lead to systematic retrieval errors which can by far exceed the expected random retrieval error caused by measurement noise.

Stratospheric ClONO2 and HNO3 profiles inside the Arctic vortex from MIPAS‐B limb emission spectra obtained during EASOE

Vertical profiles of ClONO2 and HNO3 inside the Arctic vortex have been retrieved from infrared limb emission spectra recorded during balloon flights on January 13 and in the night of March 14/15, 1992 from Esrange, Sweden (68°N) as part of the European Arctic Stratospheric Ozone Experiment (EASOE). The instrumentation used was the cryogenic Michelson Interferometer for Passive Atmospheric Sounding, Balloon-borne version (MIPAS-B). Low ClONO2 abundances in mid-January indicate that a significant portion of ClONO2 had already been converted at that time. An unexpectedly high ClONO2 amount (1.8 to 3.1 ppbv between 16.1 and 21.5 km altitude) has been inferred from the March flight data. This implies that obviously most of the total available chlorine (ClOy) in the lower stratosphere was then in this reservoir molecule. The measured HNO3 profiles give no hint of any significant layered removal of gaseous HNO3 by condensation on particles or/and sedimentation.

Mixing Processes during the Antarctic Vortex Split in September–October 2002 as Inferred from Source Gas and Ozone Distributions from ENVISAT–MIPAS

Abstract In late September 2002, an Antarctic major stratospheric warming occurred, which led to a strong distortion of the southern polar vortex and to a split of its mid- and upper-stratospheric parts. Such an event had never before been observed since the beginning of regular Antarctic stratospheric temperature observations in the 1950s. The split is studied by means of nonoperational level-2 CH4, N2O, CFC-11, and O3 data, retrieved at the Institute for Meteorology and Climate Research Karlsruhe (IMK) from high-resolution atmospheric limb emission spectra from the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) on board the European research satellite, Environmental Satellite (ENVISAT). Retrieved horizontal and vertical distributions of CH4 and N2O show good consistency with potential vorticity fields of the European Centre for Medium-Range Weather Forecasts (ECMWF) analysis for the entire period under investigation, even for fine structures such as vortex filaments. Tracer correlatio...

Retrieval of stratospheric O3, HNO3 and ClONO2 profiles from 1992 MIPAS-B limb emission spectra: Method, results, and error analysis

Within the framework of the European Arctic Stratospheric Ozone Experiment, two flights of the balloon-borne MIPAS-B limb emission spectrometer were performed in the Arctic stratosphere from Kiruna, northern Sweden. During the early hours of January 13 and the night from March 14 to March 15, 1992, several limb sequences of infrared spectra were recorded which have permitted the retrieval of vertical profiles of many trace gases relevant for ozone chemistry. In the present work, the retrieval strategy, error estimation strategy, and resulting profiles of O3, HNO3, and ClONO2 are reported. The data analysis procedure, consisting of data preprocessing including calibration, analysis of auxiliary data including temperature profiles and line of sight determination, and retrieval of trace gas profiles, is described in detail. The last is carried out by means of multiparameter nonlinear least squares fitting in combination with onion peeling. An astonishingly high ClONO2 amount of 2.6 ppb by volume at about 19-km altitude was inferred for the March flight. A rigorous error analysis, which takes into account systematic and random errors and their often nonlinear impact on the results, proves the significance of the retrieved trace gas profiles.

A blind test retrieval experiment for infrared limb emission spectrometry

The functionality and characteristics of six different data processors (i.e., retrieval codes in their actual software and hardware environment) for analysis of high-resolution limb emission infrar ...

An enhanced HNO3 second maximum in the Antarctic midwinter upper stratosphere 2003

Vertical profiles of stratospheric HNO 3 were retrieved from limb emission spectra recorded by the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) aboard the Envisat research satellite during the Antarctic winter 2003. A high second maximum of HNO 3 was found around 34 km altitude with abundances up to 14 ppbv HNO 3 during July. Similar high abundances have not been reported in the literature for previous winters, but for the subsequent Arctic winter 2003/2004, after severe perturbations due to solar proton events. The second HNO 3 maximum in the Antarctic stratosphere started to develop in early June 2003, reached peak values during July 2003, and decreased to about 7 ppbv at the end of August while being continuously transported downward before finally forming a single HNO 3 layer over all latitudes in the lower stratosphere together with the out-of-vortex primary HNO 3 maximum. The HNO 3 decrease in August 2003 was correlated with photochemical buildup of other NO v species as ClONO 2 and NO x . From the time scales observed, it can be ruled out that the 2003 long-term HNO 3 enhancements were caused by local gas phase reactions immediately after the solar proton event on 29 May 2003. Instead, HNO 3 was produced by ion cluster chemistry reactions and/or heterogeneous reactions on sulfate aerosols via N 2 O 5 from high amounts of NOy being continuously transported downward from the lower thermosphere during May to August.

Remote sensing of the middle atmosphere with MIPAS

On 1 March 2002 the Envisat research satellite has been launched successfully into its sun-synchronous orbit. One of its instruments for atmospheric composition sounding is the Michelson Interferometer for Passive Atmospheric Sounding, a limb-scanning mid-infrared Fourier transform spectrometer. Different scientific objectives of data users require different approaches to data analysis, which are discussed. A strategy on how to validate the involved algorithms and relevant strategies is presented.

Determination of the stratospheric organic chlorine budget in the spring arctic vortex from MIPAS B limb emission spectra and air sampling experiments

Vertical profiles of halogenated source gases, CF 2 Cl 2 , CFCl 3 , CHF 2 Cl, Cl 4 , and CF 4 , were retrieved from limb emission spectra recorded by the Michelson Interferometer for Passive Atmospheric Sounding, Balloonborne version (MIPAS B) during a balloon flight launched from Esrange near Kiruna, northern Sweden (68°N) on March 14, 1992. This flight was a contribution to the balloon launch program of the European Arctic Stratospheric Ozone Experiment (EASOE) campaign. All problems encountered during the analysis of the recorded spectra are discussed in detail. These are primarily the lack of spectral data for HNO 3 which interferes with the CF 2 Cl 2 ν 6 band, and the strong effects attributed to the Pinatubo aerosol. As the air mass sounded by MIPAS was polar vortex air, these data supplement the results of in situ air sampling experiments, which investigated air masses outside or at the edge of the polar vortex at altitudes below 18 km during the last phase of EASOE. An analysis is made of the vertical profiles of the seven most abundant organic chlorine species (CF 2 Cl 2 , CFCl 3 , CHF 2 Cl, CCl 4 , CH 3 Cl, CH 3 CCl 3 , and C 2 F 3 Cl 3 ) during that phase of the EASOE campaign. Mixing ratios of those organic chlorine compounds which had not been measured by MIPAS are inferred from profiles provided by air sampling experiments performed between November 30, 1991, and March 12, 1992. These profiles were adjusted to the dynamic conditions during the MIPAS observations, namely the effect of subsidence, using CF 2 Cl 2 as a tracer. This allowed to derive the relative contributions of the organic chlorine species to the total chlorine budget of the air mass sounded by MIPAS. The results are consistent with the high ClONO 2 mixing ratio of 2.6 parts per billion by volume (ppbv) observed at 18.9-km altitude during this flight of MIPAS B.

ClONO2 vertical profile and estimated mixing ratios of ClO and HOCl in winter Arctic stratosphere from Michelson interferometer for passive atmospheric sounding limb emission spectra

Nighttime limb emission spectra recorded by the balloon-borne Michelson interferometer for passive atmospheric sounding (MIPAS) on February 11, 1995, near Kiruna were used to infer a vertical profile of ClONO2 as well as estimates of ClO and HOCl volume mixing ratios. The highest ClONO2 mixing ratio (2.6 parts per billion by volume (ppbv)) was found at 22.69 km altitude and is explained by an early recovery of this chlorine reservoir in the upper part of the formerly chlorine-activated height range. Inferred nighttime ClO mixing ratios appear to be rather high for the lower stratosphere (0.38 ppbv at 16.43 km altitude) and indicate chlorine activation at this altitude region. The HOCl mixing ratio is estimated as 0.03 ppbv at 28.04 km altitude, while for lower altitudes the HOCl concentrations are clearly below the detection limit of MIPAS. The measurements are compared with three-dimensional chemical transport model calculations. Results agree reasonably well but show differences in detail.

Non-local thermodynamic equilibrium limb radiances for the mipas instrument on Envisat-1

Abstract An evaluation of the effects that the assumption of local thermodynamic equilibrium (LTE) has on the retrieval of pressure, temperature and the five primary target gases (O 3 , H 2 O, CH 4 , N 2 O, and HNO 3 ) from spectra to be taken by Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) on the Envisat-1 platform has been conducted. For doing so, non-LTE and LTE limb radiances in the spectral range of 680–2275 cm −1 (4.15–14.6 μm) with a resolution of 0.05 cm −1 at tangent heights from 10 to 70 km have been computed. These calculations included the most updated non-LTE populations of a large number of vibrational levels of the CO 2 , O 3 , H 2 O, CH 4 , N 2 O and HNO 3 molecules which cause the most prominent atmospheric infrared emissions. A discussion of the most important non-LTE effects on the limb radiances as well as on the retrievals of pressure-temperature and volume mixing ratios of O 3 , H 2 O, CH 4 , N 2 O, and HNO 3 is presented, together with the most important non-LTE issues that could be studied with the future coming of MIPAS data.