Comparisons of MIPAS/ENVISAT ozone profiles with SMR/ODIN and HALOE/UARS observations

Abstract

Ozone volume mixing ratio (VMR) profiles are measured by the Michelson Interferometer for passive atmospheric sounding (MIPAS) on ENVISAT. The data sets produced by the science data processor at Institut für Meteorologie und Klimaforschung (IMK), Germany are compared with those obtained by halogen occultation experiment (HALOE) on UARS and by sub-millimetre radiometer (SMR) on ODIN. For the stratospheric measurements taken during September/October 2002, the three instruments show reasonable agreement, with global mean differences within 0.1–0.3 ppmv. The typical zonal mean differences are of 0.4 ppmv for HALOE and 0.6 ppmv for SMR (4–6%) in the ozone VMR peak region at 25–30 km near the equator, though larger differences of 0.8–1 ppmv (8–10%) are also observed in a small latitude–altitude region in the tropic. A positive bias of about 0.2–0.4 ppmv in the MIPAS data in the 35–40 km region has also been found. Further studies are under way to explain these differences.

Full text

Comparisons of MIPAS/ENVISAT ozone profiles with
SMR/ODIN and HALOE/UARS observations
D.Y. Wang
a,*
, G.P. Stiller
a
, T. von Clarmann
a
, H. Fischer
a
, N. Glatthor
a
,
U. Grabowski
a
,M.Ho
¨pfner
a
, S. Kellmann
a
, M. Kiefer
a
, A. Linden
a
,
G. Mengistu Tsidu
a
, M. Milz
a
, T. Steck
a
, S. Wohnsiedler
a
,M.Lo
´pez-Puertas
b
,
B. Funke
b
, S. Gil-Lo
´pez
b
, M. Kaufmann
b
, M.L. Koukouli
b
, D. Murtagh
c
, N. Lautie
´
c
,
C. Jime
´nez
c
, A. Jones
c
, P. Eriksson
c
, J. Urban
d
, J. de La Noe
¨
d
,E
´. Le Flochmoe
¨n
d
,
E
´. Dupuy
d
, P. Ricaud
d
, M. Olberg
e
, U. Frisk
f
, J. Russell III
g
, E. Remsberg
h
a
Forschungszentrum Karlsruhe GmbH und Universita
¨t Karlsruhe, Institut fu
¨r Meteorologie und Klimaforschung (IMK),
Postfach 3640, 76021 Karlsruhe, Germany
b
lnstituto de Astrofı
´sica de Andalucı
´a, CSIC, Apartado Postal 3004, 18008 Granada, Spain
c
Chalmers University of Technology,
Radio and Space Science Department (Centre for Astrophysics and Space Science) SE-412 96
Go
¨teborg (Gothenburg), Sweden
d
Observatorire Aquitain de Sciences de l lÕUnivers (OASU), 2 rue de lÕObservatoire, BP 89, 33 270 Floirac, France
e
Onsala Space Observatory (Centre for Astrophysics and Space Science), SE-43922 Onsala, Sweden
f
Swedish Space Corporation (SSC), P.O. Box 4207, SE-17104 Solna, Sweden
g
Department of Physics, Hampton University, Hampton, VA 23668, USA
h
Atmospheric Sciences Competency, NASA Langley Research Center, Mail Stop 401B, Hampton, VA 23681, USA
Received 14 July 2004; received in revised form 15 February 2005; accepted 2 March 2005
Abstract
Ozone volume mixing ratio (VMR) profiles are measured by the Michelson Interferometer for passive atmospheric sounding
(MIPAS) on ENVISAT. The data sets produced by the science data processor at Institut fu
¨r Meteorologie und Klimaforschung
(IMK), Germany are compared with those obtained by halogen occultation experiment (HALOE) on UARS and by sub-millimetre
radiometer (SMR) on ODIN. For the stratospheric measurements taken during September/October 2002, the three instruments
show reasonable agreement, with global mean differences within 0.1–0.3 ppmv. The typical zonal mean differences are of 0.4 ppmv
for HALOE and 0.6 ppmv for SMR (4–6%) in the ozone VMR peak region at 25–30 km near the equator, though larger differences
of 0.8–1 ppmv (8–10%) are also observed in a small latitude–altitude region in the tropic. A positive bias of about 0.2–0.4 ppmv in
the MIPAS data in the 35–40 km region has also been found. Further studies are under way to explain these differences.
Ó2005 COSPAR. Published by Elsevier Ltd. All rights reserved.
Keywords: Ozone; MIPAS/ENVISAT; SMR/ODIN; HALOE/UARS
1. Introduction
The Michelson interferometer for passive atmo-
spheric sounding (MIPAS) on board the ENVISAT sa-
tellite provides vertical profiles of various gas species,
including stratospheric ozone volume mixing ratio
0273-1177/$30 Ó2005 COSPAR. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.asr.2005.03.015
*
Corresponding author. Tel.: +49 7247 82 5203; fax: +49 7247 82
6141.
E-mail address: ding-yi.wang@imk.fzk.de (D.Y. Wang).
www.elsevier.com/locate/asr
Advances in Space Research 36 (2005) 927–931

(VMR), by limb-observing mid-infrared emission mea-
surements (Fischer and Oelhaf, 1996; ESA, 2000; Carli
et al., 2004). The high resolution limb viewing infrared
spectra observed by MIPAS are processed with data
processors in several institutions. The ozone profiles of
the European Space Agency (ESA) near-real time data
products were compared with ground-based measure-
ments (Blumenstock et al., 2004), balloon and aircraft
observations (Cortesi, 2004), and other satelliteÕs data
(Kerridge et al., 2004). Here we focus, however, on the
validation of the ozone profiles retrieved with the data
processor developed at Institut fu
¨r Meteorologie und
Klimaforsthung (IMK) and complemented by the com-
ponent of non-local thermodynamic equilibrium (non-
LTE) treatment from the Instituto de Astrofı
´sica de
Andalucı
´a (IAA).
Since middle infrared emission spectra are strongly
sensitive to temperature, and as limb observations are
strongly affected by the observation geometry, an accu-
rate knowledge of these quantities is essential in retriev-
ing ozone profiles. The IMK–IAA data processor
provides simultaneous retrieval of temperature and
line-of-sight parameters from measured spectra and
the spacecraft ephemerides. The processor and its
robustness are discussed by Von Clarmann et al.
(2003a,b,c). The temperatures retrieved by the IMK–
IAA processor are validated against a number of satel-
lite observations and assimilation analyses and show
good consistency with the other data sets (Wang et al.,
2004a,b,c, 2005). The derived temperature and observa-
tion geometry are used to retrieve profiles of ozone and
other species. Thus, self-consistence of the IMK–IAA
data product is achieved. The IMK–IAA data process-
ing also allows more flexibility in the retrieval setup
and covers a wider altitude range up to 140 km. The
IMK–IAA MIPAS ozone profiles have been compared
with the MIPAS data sets produced by other processors,
including the ESA operational data (Steck, 2003; Ker-
ridge et al., 2004), with the measurements of the MIPAS
balloon instrument, and with the ground-based FTIR
(Steck, 2003) and the SLIMCAT model calculations
(Feng et al., 2005).
This study compares the IMK–IAA ozone data with
the observations of halogen occultation experiment
(HALOE) on the upper atmosphere research satellite
(UARS) (Russell et al., 1993) and sub-millimetre radi-
ometer (SMR) on ODIN (Urban et al., 2004; Murtagh
et al., 2002; Merino et al., 2002). The HALOE/UARS
ozone profiles have been validated already. They show
good agreement with the measurements of other instru-
ments within ±(5–10)% (0.1–0.2 ppmv) (Manney et al.,
2001; Morris et al., 2002; Randall et al., 2003). The
SMR/ODIN ozone data are up to now mainly derived
from two sub-millimetre frequency bands, around 501
and 544 GHz. The SMR data sets are still under valida-
tion. This study also serves as an input for such effort.
2. Data description
The MIPAS IMK–IAA ozone data of version
V1_O3_1 is used here. They are derived from the obser-
vations of September/October 2002 when an unprece-
dented Antarctic winter stratospheric sudden warming
event occurred. The measurements cover altitudes be-
tween 6 and 68 km with a vertical resolution of
3 km. The data are retrieved at 1 km interval below
42 km, and 2 km above. The details of the retrieval setup
can be found in Glatthor et al. (2005).
The HALOE/UARS ozone profiles are taken from
Level 2-version 19 database through British Atmo-
spheric Data Centre (BADC). The data are retrieved
from cloud top up to near 90 km at 2 km vertical spac-
ing. The data set is validated. HALOE errors are small-
est in the middle stratosphere with the lowest estimate of
uncertainty of 9%, becoming larger (20–25%) in the
meso-sphere and lower stratosphere. These estimates
are considered conservative because they were obtained
by combining the known systematic and random com-
ponents of error using a root-sum-squared method. As
a result when comparisons of near-coincident HALOE
versus correlative measurements are conducted, their
differences are often smaller than the error estimates
(e.g. Bruehl et al., 1996; Harris et al., 1998; Bhatt
et al., 1999; Randall et al., 2003; Natarajan et al., 2004).
The SMR/ODIN ozone data (version V1.2) used for
this analysis are obtained from the weak line at
501.45 GHz, which allows the retrieval of ozone be-
tween 20 and 55 km with a vertical resolution of
2 km (e.g. Murtagh et al., 2002). The data from the
stronger line at 544.85 GHz are still under processing.
The correlative measurements between the three
instruments are selected by coincidence criteria of 5°lat-
itude and 10°longitude. The mean horizontal distances
are 500 ± 250 km, minimized in the polar regions, but
maximized near the equator. The time differences are 6 h
for MIPAS/SMR, but 12 h for MIPAS/HALOE. This is
because HALOE views the atmosphere for each space-
craft sunrise and sunset event. This time difference does
not affect the comparison between 20 and 60 km, since
diurrnal variation of O
3
occurs in the meso-sphere above
60 km (Natarajan et al., 2004). The number of available
correlative profiles are 1176 on four days for MIPAS/
SMR, and 129 on seven days for MIPAS/HALOE.
The correlative profiles are interpolated to a common
altitude grid, that used by the MIPAS data. No adjust-
ment of altitude resolution or averaging kernel is applied
for the comparison results presented in this paper.
3. Comparison results
The global mean differences of MIPAS/SMR and
MIPAS/HALOE are shown in Fig. 1, and the zonal
928 D.Y. Wang et al. / Advances in Space Research 36 (2005) 927–931

mean differences in Fig. 2, respectively. Our compari-
sons are confined between 20 and 55 km, where the mea-
surement errors are smallest for the three instruments.
The zonal means are calculated separately for MIPAS
ascending (night time) and descending (day time, not
shown) orbit nodes. The results for both nodes are sim-
ilar. The global means are computed by combining both
nodes. We also calculated global means separately for
HALOE sunrise and sunset events (not shown here).
No significant difference is found in the global means
of the two occultation types, suggesting small diurnal ef-
fects on the ozone VMR in the stratospheric region,
since sunrise/sunset differences become most apparent
above the stratopause only (Natarajan et al., 2004).
The three instruments show reasonable agreement be-
tween 20 and 55 km. The global mean differences are
generally within 0.1–0.3 ppmv, with the larger values
occurring between 30 and 40 km and around ozone
VMR peak heights. The zonal mean differences are gen-
erally in the range of 0.1–0.3 ppmv. However, the zonal
mean differences reach maximum values of 0.4 ppmv
(corresponding to 4% of the HALOE-measured ozone
VMR) for HALOE and 0.6 ppmv (6%) for SMR
around 25–30 km near the equator, where the ozone
VMR has maximum, with a few points showing differ-
ences as large as 0.8–1 ppmv (8–10%). The SMR ozone
profiles derived from the weak line at 501.45 GHz show
higher noise in comparison with the HALOE data.
Fig. 2. Zonally averaged ozone mixing ratio differences (in ppmv) of MIPAS with respect to SMR (left) and HALOE (right) correlative
measurements. The averages are calculated over latitude interval of 30°. MIPAS observations are in ascending node (night time). The contour
intervals are 0.2 ppmv.
Fig. 1. Mean differences (solid) and standard deviations (dotted) of ozone mixing ratios (in ppmv) of MIPAS with respect to SMR (left) and HALOE
(right) correlative measurements. The data arc averaged on a daily basis (thin line) and over all days (thick line) for the available correlative profiles.
Data points vary at individual heights. The maximum and minimum numbers are denoted, followed by mean difference, deviation, and 1 r
uncertainty averaged over all heights for all days.
D.Y. Wang et al. / Advances in Space Research 36 (2005) 927–931 929

It is somewhat surprising that the worst agreement
occurs in the tropics, considering the fact that the split
ozone hole during this time period likely created signif-
icant ozone variability in the Antarctic. One possible
explanation is that the horizontal separation between
the correlative measurements are minimized in the polar
regions, but maximized near the equator, since the lon-
gitude criterion is meaningless at the poles, and thus the
spatial coincidence criteria include only the ±5°in
latitude.
It seems that the differences are occurring in an oscil-
latory pattern with altitude. The tropics are where the
ozone mixing ratio has its largest vertical gradients.
Some mismatch in the vertical resolution between the
MIPAS and HALOE measurements (3 and 2 km,
respectively) might give rise to an apparent oscillatory
pattern (see Fig. 2). To test this hypothesis, we have ap-
plied the MIPAS averaging kernels to adjust the HA-
LOE data in vertical resolution, the comparison shows
similar results (not shown here), suggesting that the dif-
ference in altitude resolution is unlikely to explain the
observed features in the tropics.
While typical differences between MIPAS/HALOE
and MIPAS/SMR are quite small, the altitude distribu-
tions of these differences appear to be systematic. The
MIPAS ozone profiles might be positive biased in about
0.2–0.4 ppmv at 35–40 km. Sensitivity tests for MIPAS
retrieval setup are under way to understand these
differences.
4. Conclusions
The global ozone profiles measured by MIPAS in
September/October 2002 and derived by the IMK–
IAA data processor are compared with the measure-
ments of HALOE/UARS and SMR/ODIN. The global
mean differences are generally within 0.1–0.3 ppmv.
Those comparisons between independent instruments
and the reasonable agreements represent a good input
for the validations of SMR and MIPAS.
Large zonal mean differences for the three data sets
are seen around the ozone VMR peak region at 25–
30 km near the equator, with typical values of 0.4–
0.6 ppmv (4–6% of the HALOE-measured ozone
VMR) for HALOE and SMR, respectively. In a small
latitude–altitude region in the tropic, the zonal mean dif-
ferences as large as 0.8–1 ppmv (8–10%) are also ob-
served. The large horizontal separation between the
correlative profiles in the tropics may play a role in pro-
ducing the observed deviations. A positive bias of about
0.2–0.4 ppmv in the MIPAS data in the 35–40 km region
has also been found. In order to explain the found differ-
ences, studies are under way to clarify which options in
the MIPAS retrieval setup may cause these differences.
Further validation on basis of various available measure-
ments, including further HALOE collocations, ground-
based FT-IR, LIDAR, as well as other satellite platforms
(SAGE, POAM III, ACE, OSIRIS) are also ongoing.
Acknowledgments
The research work of IMK–IAA MIPAS group has
been funded by EC via Contract No. EVG1-CT-1999-
00015 (AMIL2DA), BMBF via Project No. 07 ATF
43/44 (KODYACS), 07 ATF 53 (SACADA), and 01
SF 9953 (HGF-VF), and ESA via Contract No. 15530/
01/NL/SF (IN-FLIC). The IAA team was partially sup-
ported by Spanish Projects PNE-017/2000-C and
REN2001-3249/CLI. B. Funke has been supported
through an European Community Marie Curie
Fellowship.
References
Bhatt, P.P., Remsberg, E.E., Gordley, L.L., Mclnerney, J.M., Brack-
ett, V.G., Russell III, J.M. An evaluation of the quality of HALOE
ozone profiles in the lower stratosphere. J. Geophys. Res. 104,
9261–9275, 1999.
Blumenstock, T. et al. Comparisons of MIPAS ozone profiles with
ground-based measurements. In: Proc. ACVE-2, ESRIN. Frascati,
Italy, 2004.
Bruehl, C., Drayson, S.R., Russell III, J.M., Crutzen, P.J., Mclnerney,
J., Purcell, P.N., Claude, H., Gernand, H., McGee, T., McDermid,
I., Gunson, M.R. HALOE ozone channel validation. J. Geophys.
Res. 101 (D6), 10,217–10,240, 1996.
Carli, B. et al. First results of MIPAS/ENVISAT with operational
level 2 code. Adv. Space Res. 33, 1012–1019, 2004.
Cortesi, U. MIPAS ozone validation by stratospheric balloon and
aircraft measurements. In: Proc. ACVE-2, ESRIN. Frascati, Italy,
2004.
ESA, Envisat: MIPAS, An instrument for atmospheric chemistry and
climate research, ESA Publications Division, ESTEC, P.O. Box
299, 2200 AG Noordwijk, The Netherlands, SP-1229, European
Space Agency, 2000.
Feng, W., Chipperfield, M.P., Roscoe, H.K., Remedios, J.J., Water-
fall, A.M., Stiller, G.P., Glatthor, N., Ho
¨pfner, M., Wang, D.-Y.
Three-dimensional model study of the Antarctic ozone hole in 2002
and comparison with 2000. J. Atmos. Sci. 62 (3), 822–837, 2005.
Fischer, H., Oelhaf, H. Remote sensing of vertical profiles of
atmospheric trace constituents with MIPAS limb emission spec-
trometers. Appl. Opt. 35 (16), 2787–2796, 1996.
Glatthor, N. et al. Mixing processes during the Antarctic vortex split
in September/October 2002 as inferred from source gas and ozone
distributions from MIPAS/ENVISAT. J. Atmos. Sci. 62 (3), 787–
800, 2005.
Harris, N., Hudson, R., Phillips, C. (Eds.), Assessment of trends in the
vertical distribution of ozone, WMO Ozone Report No. 43,
SPARC Report No. 1, Geneva, 1998.
Kerridge, B.J., Goutail, F., Bazureau, A., Wang, D.-Y., Bracher, A.,
Weber, M., Bramstedt, K., Siddans, R., Latter, B.G., Reburn, W.J.,
Jay, V.L., Dethof, A., Payne, V.H. MIPAS ozone validation by
satellite intercomparisons. In: Proc. ACVE-2, ESRIN. Frascati,
Italy, 2004.
Manney, G.L. et al. Comparison of satellite ozone observations in
coincident air masses in early November 1994. J. Geophy. Res. 106
(D9), 9923–9943, 2001.
930 D.Y. Wang et al. / Advances in Space Research 36 (2005) 927–931

Merino, F. et al. Studies for the ODIN sub-millimetre radiometer:
III. Performance simulation. Can. J. Phys. 80 (4), 357–373,
2002.
Morris, G.A. et al. A comparison of HALOE V19 with SAGE II V6.0
ozone observation using trajectory mapping. J. Geophys. Res. 107
(D13), 4177, 2002.
Murtagh, D.P. et al. An overview of the ODIN atmospheric mission.
Can. J. Phys. 80 (4), 309–319, 2002.
Natarajan, M., Deaver, L., Thompson, E., Magill, B. HALOE
observations of mesospheric ozone: effects of diurnal gradients
and occultation geometry. In: Proceedings of the Quadrennial
Ozone Symposium. Kos, Greece, 2004.
Randall, C.E. et al. Validation of POAM III ozone: comparisons with
ozonesonde and satellite data. J. Geophys. Res. 108 (D12), 4367,
2003.
Russell III, J.M. et al. The halogen occultation experiment. J.
Geophys. Res. 98 (D6), 10777–10797, 1993.
Steck, T. Intercomparison of AMIL2DA results for MIPAS-ENVI-
SAT data, Task Report, EC-contract EVG1-CT-1999-00015
(AMIL2DA), 2003.
Urban, J. et al. The northern hemisphere stratospheric vortex
during the 2002–03 winter: subsidence, chlorine activation and
ozone loss observed by the odin sub-millimetre radiometer.
Geophys. Res. Lett. 31, L07103. doi: 10.1029/2003GL019089,
2004.
Von Clarmann, T. et al. A blind test retrieval experiment for limb
emission spectrometry. J. Geophys. Res. 108 (D23), 4746, 2003a.
Von Clarmann, T. et al. Retrieval of temperature and tangent altitude
pointing from limb emission spectra recorded from space by the
Michelson interferometer for passive atmosphere (MIPAS). J.
Geophys. Res. 108 (D23), 4736, 2003b.
Von Clarmann, T. et al., Remote sensing of the middle atmosphere
with MIPAS, in remote sensing of clouds and the atmosphere VII.
in: Scha
¨fer, K., Lado-Bordowsky, O. (Eds.), Proc. SPIE 4882, pp.
172–183, 2003c.
Wang, D.Y. et al. Comparisons of MIPAS/ENVISAT and GPS-RO/
CHAMP temperature profiles. In: Reigber, C., Luehr, H., Sch-
wintzer, P., Wickert, J. (Eds.), Earth Observation with CHAMP:
Results from Three Years in Orbit. Springer, Berlin, pp. 567–572,
2004a.
Wang, D.Y. et al. Cross-validation of MIPAS/ENVISAT and
GPSRO/CHAMP temperature profiles. J. Geophys. Res. 109
(D19311), 2004b.
Wang, D.Y. et al. Comparisons of MIPAS-observed temperature
profiles with other satellite measurements. In: Scha
¨fer, K.P.,
Comero
´, A., Carleer, M.R., Picard, R.H. (Eds.), Remote Sensing
of Clouds and the Atmosphere VIII, Proc. SPIE, vol. 5235. SPIE,
Bellingham, WA, pp. 196–207, 2004c.
Wang, D.Y., et al., Validation of stratospheric temperatures measured
by MIPAS on ENVISAT, J. Geophys. Res., 109 (accepted).
D.Y. Wang et al. / Advances in Space Research 36 (2005) 927–931 931