Assimilation of MIPAS observations using a three-dimensional
global chemistry-transport model
F. Baier, T. Erbertseder, O. Morgenstern, M. Bittner and G. Brasseur
Corresponding author: F. Baier, The German Aerospace Center, 82234 Wessling,
P.O. Box 1116, Germany. frank.baier@dlr.de
MIPAS observations are assimilated using the chemistry-transport model ROSE/DLR to derive consistent
global chemical analyses of the stratosphere. Two main MIPAS data streams are currently available, which
differ in quality and coverage: the ESA (European Space Agency) operational product and the IMK
(Institute for Meteorology Karlsruhe) scientific product. In this study we investigate the sensitivity of data
assimilation results to these different input data sets. We focus on ENVISAT/MIPAS baseline observations
of H2O, O3, HNO3, CH4, N2O and NO2, covering October-November 2003. Sequential assimilation is
performed using an optimum interpolation scheme with diagnostic covariances. Optimized assimilation
parameters are derived using ² diagnostics. Comparing results to HALOE observations it is shown that all
assimilated model species benefit significantly from observations. While both data sets are found to be well
suited for global assimilation experiments, some differences are evident. For example, regions with
increased stratospheric H2O concentrations near the tropical tropopause are only present when IMK data
is applied.
1.
Introduction
In order to derive consistent global threedimensional chemical analyses from asynoptic and
inhomogeneously distributed remote sensing
observations,
sequential
assimilation
into
chemistry-transport models has been successfully
demonstrated by several former studies (e.g.
Khattatov et al. 2000; Chipperfield 2002; El
Amraoui et al. 2004). The MIPAS (Michelson
Interferometer for Passive Atmospheric Sounding)
instrument (Fischer and Oelhaf 1996) on board of
the polar orbiting platform ENVISAT (European
Environmental Satellite) was launched in March
2002. It performs global limb measurements in the
near to mid infrared and enables to retrieve
temperature, pressure and trace gas profiles of the
middle atmosphere for night and day conditions
(http://envisat.esa.int/instruments/mipas). Currently
two kind of level 2 data are available: the ESA
(European Space Agency) operational standard
products (Carli et al. 2004), with focus on rapid
processing of all available observations, and so
called off-line products for short scientifically
interesting periods. Due to the different trace gas
retrieval schemes applied they differ with respect to
quality and coverage. For this study we compare
data from IMK (Institute for Meteorology
Karlsruhe)(Glatthor et al. 2005) with the standard
ESA product covering 21 October until 20
November 2003. To ensure the consistency of the
final analyses we apply a ² diagnostics to adjust
the error parameters depending on latitude and
assimilated species.
2.
MIPAS data
The MIPAS instrument achieves nearly
global coverage within three days. Trace gases
can be obtained with a maximum vertical resolution
of about 3 km. Along the line of sight the horizontal
averaging gives a resolution of approximately 400
km (Stiller et al. 2002). For a detailed description of
the MIPAS trace gas retrieval as applied by ESA
and IMK the reader is refered to Ridolfi et al.
(2000), Carli et al. (2004), Steck and Clarmann
(2001) and references therein. Details of validation
results can be found in Snoeij et al. (2004) and
references therein.
For this study only stratospheric data of the
baseline species O3, H2O, CH4, N2O, NO2 and
HNO3 has been considered. While MESA
observations nearly cover the whole time period
with only some days missing in October 2003,
MIMK data shows two short gaps in October and
November and no more data after 12 November
2003. Both data sets give a good coverage of the
middle atmosphere between 10 and 60 km altitude.
The vertical resolution of trace gas profiles can be
derived from information on the averaging kernel
matrix accompanying the data, which depends on
species, latitude, height and season. For both data
sets the vertical resolution decreases rapidly at the
upper and lower boundaries.
3.
Sequential assimilation
For assimilation of MIPAS data we use the
DLR (German Aerospace Center) version of the 3D
global chemistry-transport model (CTM) NCARROSE. The original model is described in detail in
Rose and Brasseur (1989) and Granier and
Brasseur (1991). The model covers the relevant
stratospheric gas-phase and heterogeneous
chemical processes. The CTM is driven by MetOffice wind and temperature fields as supplied
every 24h (Swinbank and O'Neill 1994).
We use a similar sequential assimilation
scheme as Khattatov et al. (2000). The analysis
variances (diagonal part of covariances) are
transported as quasi-tracers. In this way the
influence of observations on the results is taken
into account. The correlation part of background
covariances is parameterized by Gaussian
functions depending on the local Euclidian gridpoint distances.
Global mean ² values are calculated from
OMF (observation minus first-guess) values and apriori errors to check the consistency of
assimilation parameters and tune model error
growth rates and representativeness errors. Using
the same settings as Khattatov et al. (2000)
resulting ² values were found much too high.
Therefore, the representativeness errors were
adjusted by using latitudinal dependent temporal
mean ² values from a first assimilation experiment
as correction factors for each species.
Figure (1) shows the resulting corrected
representativeness errors. Relatively high values
are found for MESA and MIMK HNO3 in the
subtropics and high latitudes. Respective N2O
errors show maximum values in the tropics and
high southern latitudes. For NO2 a strong increase
of representativeness errors is derived for high
Northern latitudes which is especially significant in
the case of MIMK data. This increase of NO2
errors can be related to the solar proton events in
late October and early November 2003
(Degenstein et al. 2005).
Figure 1: Mean relative representativeness errors
derived for MESA (left frame) and MIMK (right
frame) data sets.
6.
Results
For an independent comparison of results
HALOE version 19 data for October and November
2003 (http://haloedata.larc.nasa.gov) were used.
Observations are limited to the middle northern
latitudes and two latitude bands near 70°S and
30°S. Approximately a total of two weeks is
covered by observations in October and again in
November 2003. Only data above the tropopause
(defined by 2 PVU) was considered. Maximum
errors are found for NOx for both MIMK and MESA
assimilation experiments. With the exception of
HCl (not assimilated), relative improvements
compared to the base run (no assimilation) reach
40%.
Figure 2: Zonal mean analyses (top) and analysis
errors (bottom) for MESA (left) and MIMK (right)
H2O for 5 November 2003. The 2 PVU isoline
(dashed) indicates the tropopause. The vertical
axes show the altitude as pressure levels in hPa.
The differences of the zonal mean
distributions of assimilated species with respect to
MESA and MIMK input data are small in general,
except for two regions with increased H2O mixing
ratios near the tropical tropopause. These are only
visible when MIMK data is assimilated. Figure (2)
shows the respective H2O distributions for 5
November 2003 using MIMK and MESA.
Concerning analysis errors, differences between
MESA and MIMK are in general only significant in
regions with different data coverage. Minimum
relative error values are found between 60 and 1
hPa, while errors increase strongly where no
observations are available. For H2O, CH4 and O3,
this is mainly true for the troposphere and the
tropopause region. For the same reason, HNO3,
N2O and NO2 analysis errors increase
considerably in the upper stratosphere and lower
mesosphere. MIMK analyses of H2O show also a
strong increase of analysis errors above 1 hPa.
Acknowledgements:
This extended abstract shows results of a fulllength paper accepted for publication in a coming up
special issue of the Quaterly Journal of the Royal Met.
Soc.
We would like to thank ESA for providing the
MIPAS Level 2 data within the ESA-AO project EVIVA.
We are grateful to the Institute of Meteorology,
Karlsruhe, for supply of MIPAS IMK data. Further, we
thank the German Ministry for Education and Research
for funding HGF-ENVISAT. The UK Meteorological
Office and the HALOE team kindly provided the
meteorological data.
References:
El Amraoui, L., Ricaud, P., Urban, J., Théodore, B.,
Hauchecorne, A., Lauti, De La Noë, N., J., Guirlet, M., Le
Flocjmoën, E., Murtagh, D., Dupuy, E., Frisk, U. and
d'Andon, O. F.: Assimilation of Odin/SMR O3 and N2O
measurements in a three-dimensional chemistry
transport model. J. Geophys. Res., 109, D22304, doi:
10.1029/2004JD004796, 2004.
Brasseur, G., Hitchman, M.H., Walters, S., Dymek, M.,
Falise, E. and Pirre, M.: An interactive chemical
dynamical radiative two-dimensional model of the middle
atmosphere. J. Geophys. Res., 95, 5639-5655, 1995.
Carli, B., Alpaslan, D., Carlotti, M., Castelli, E.,
Ceccherini, S., Dinelli, B. M.,Dudhia, A., Flaud, J. M.,
Höpfner, M., Jay, V., Magnani, L., Oelhaf, H., Payne, V.,
Piccolo, C., Prosperi, M., Raspollini, P., Ridolfi, M.,
Remedios, J. and Spang, R.: First results from
MIPAS/ENVISAT with operational Level 2 code. Adv.
Space Res., 33, 7, 1012-1019, doi:10.1016/S02731177(03)00584-2, 2004.
Chipperfield, M. P., Khattatov, B. V. and Lary, D. J.:
Sequential assimilation of stratospheric chemical
observations in a three-dimensional model. J. Geophys.
Res., 107, D21, 4585, doi: 10.1029/2002JD002110,
2002.
Degenstein, D. A., Lloyd, N. D., Bourassa, A. E.,
Gattinger, R. L. and Llewellyn, E. J.: Observations of
mesospheric ozone depletion during the October 28,
2003 solar proton event by OSIRIS. Geophys. Res. Lett.,
32, L03S11, doi: 10.1029/2004GL021521, 2005.
Fischer, H. and Oelhaf, H.: Remote sensing of vertical
profiles of atmospheric trace constituents with MIPAS
limb-emission spectra. Appl. Opt., 35, (16), 2787-2796,
1996.
Glatthor, N., von Clarmann, T., Fischer, H., Funke, B.,
Grabowski, U., Höpfner, M., Kellmann, S., Kiefer, M.,
Linden, A., Milz, M., Steck, S., Stiller, G.P., Mengistu
Tsidu, G. and Wang, D.-Y.: 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.
Granier, C. and Brasseur, G.: Ozone and other trace
gases in the Arctic and Antarctic regions: Threedimensional model simulations. J. Geophys. Res., 96,
2995-3011, 1991.
Khattatov, B.V., Lamarque, J.-F., Lyjak, L. V., Ménard,
R., Levelt, P., Tie, X., Brasseur, G. and Gille, J. C.:
Assimilation of sattelite observations of long-lived
chemical species in global chemistry transport models. J.
Geophys. Res., 105, D23, 29135-29144, 2000.
Ridolfi, M., Carli, B., Carlotti, M., von Clarmann, T.,
Dinelli, B. M., Dudhia, A., Flaud, J. M., Höpfner, M.,
Morris, P. E., Raspollini, P., Stiller, G. and Wells, R. J.:
Optimized forward model and retrieval scheme for
MIPAS near-real-time data processing. Applied Optics,
39, 1323-1340, 2000.
Rose, K. and Brasseur, G.: A three-dimensional model of
chemically active trace species in the middle atmosphere
during disturbed winter conditions. J. Geophys. Res., 94,
16387-16403, 1998.
Snoeij, P., Koopman, R., Wursteisen, P., Zehner, C. and
Attema, E.: The ENVISAT Atmospheric Chemistry
Instrument Validation Programme. Proceedings of the
Second Workshop on the Atmospheric Chemistry
Validation of ENVISAT, Frascati, Italy, 3-7 May 2004.
ESA SP-562, August 2004, EGE01PS, 2004.
Steck, T. and von Clarmann, T.: Constrained profile
retrieval applied to the observation mode of the
Michelson Interferometer for Passive Atmospheric
Sounding. Applied Optics, 40, 21, 3559-3571, 2001.
Stiller, G. P., von Clarmann, T., Funke, B., Glatthor, N.,
Hase, F., Höpfner, M. and Linden, A.: Sensitivity of trace
gas abundances retrievals from infrared limb emission
spectra to simplifying approximations in radiative transfer
modelling. J. Quant. Spectr. & Rad. Transfer, 72, 249280, 2002.
Swinbank, R. and O'Neill, A.: A stratosphere-troposphere
data assimilation system. Mon. Weather Rev., 122, 686702, 1994.