An evaluation of Odin/OSIRIS limb pointing and stratospheric ozone through comparisons with ozonesondes

Abstract

Five-plus years of Odin/OSIRIS (Optical Spectrograph and Infrared Imager System) ozone profiles (version 3.0 and SaskMART) have been compared with coincident ozonesonde profiles from the tropopause to the middle stratosphere (∼10-35 km). The OSIRIS(v3.0)-sonde comparisons, together with the "knee" technique, suggest that OSIRIS pointing is stable with an accuracy (systematic uncertainty) of 100 m and a precision (random uncertainty) of 400 m. Comparison of coincident profiles revealed excellent overall agreement with differences generally not exceeding 10% between 10 and 35 km. During periods of satellite eclipse larger differences above 25 km were found and concluded to be the result of flexing of the spacecraft frame brought about by cooling. Uncertainties derived from the intercomparison suggest an OSIRIS precision of 25-30% at 10 km, reaching a minimum of about 4-6% in the middle-stratosphere (23-28 km), and 6-8% at 35 km. Ozonesonde uncertainties were also derived and found to be slightly larger in the tropics than mid-latitudes, possibly due to a lack of column normalization at some of the tropical stations. OSIRIS SaskMART ozone, an alternative ozone data product, was found to be very comparable with the v3.0 ozone, with slightly smaller biases relative to the ozonesondes. OSIRIS is concluded to be a robust and accurate source of stratospheric ozone profiles. PACS Nos.: 82.33.Tb, 95.75.Qr, 42.68.−w Résumé : Nous avons comparé cinq ans de mesures sur l'ozone faites par Odin/OSIRIS (version 3.0 et SaskMART) avec des mesures coincidentes par sonde ozonique (stations terrestres) prises entre la tropopause et la stratosphère moyenne (∼10-35 km), afin d'évaluer la capacité d'OSIRIS à mesurer l'ozone et à positionner ses mesures. Les comparaisons suggèrent que la précision de visée en altitude d'OSIRIS est stable, avec une fidélité

Evaluation of ACE-FTS and OSIRIS Satellite retrievals of ozone and nitric acid in the tropical upper troposphere: Application to ozone production efficiency

“…OSIRIS ozone concentrations are on average 5% higher (RMSE = 4 ppbv) than the MOZAIC profiles and 5% higher (RMSE = 8 ppbv) than the ozonesondes. This is an improvement compared to the retrieval version 2.1 evaluation in the lower stratosphere by McLinden et al [2007]. Although the relatively lower number of ACE-FTS measurements in each region increases scatter, the ACE-FTS profiles have significant mean biases of 10% versus MOZAIC (RMSE = 18 ppbv) and 13% versus ozonesondes (RMSE = 10 ppbv).…”

“…Individual sources of error are less than 5% throughout most of the stratosphere, but high altitude clouds may cause larger retrieval errors in the tropical upper troposphere and lower stratosphere [von Savigny et al, 2005]. OSIRIS ozone retrievals have been found to be accurate to within 4-8% in the stratosphere between 15 and 32 km with a total error of 10% [Petelina et al, 2004;von Savigny et al, 2005], however uncertainties increase at lower altitudes, up to 25-30% at 10 km [McLinden et al, 2007].…”

“…Previous validation experiments for both instruments have largely focused on either higher latitudes, or on the stratosphere where mixing ratios are higher and satellite retrievals are easier [Petelina et al, 2004;von Savigny et al, 2005;McLinden et al, 2007;Hegglin et al, 2008;Wolff et al, 2008;Dupuy et al, 2009]. No detailed validation of ACE-FTS O 3 , ACE-FTS HNO 3 or OSIRIS O 3 with in situ observations throughout the tropical troposphere has been published to date.…”

Assessment of Odin-OSIRIS ozone measurements from 2001 to the present using MLS, GOMOS, and ozonesondes

“…These low temperatures are associated with a negative bias of 5-12 % in OSIRIS measurements when compared with SAGE II (Adams et al, 2013). This may be caused by misalignment between OSIRIS and the star tracker due to thermal deformation, leading to pointing errors (McLinden et al, 2007). Furthermore, the quality of ozone retrievals may be affected by defocusing, reduced spectral resolution, and wavelength shifts at low temperatures (Llewellyn et al, 2004).…”

“…One possible cause for this type of bias is pointing error due to the thermal deformation of the spacecraft (McLinden et al, 2007), leading to a shift of the profile in altitude. Another possible cause for this bias is the reduced spectral resolution and wavelength shifts that occur when optics temperatures are low because the instrument was focused at a higher temperature (Llewellyn et al, 2004).…”