Assimilation of GOME ozone profiles and a global chemistry-transport model using a Kalman filter with anisotropic covariance

Nadir ozone profiles retrieved from the GOME (Global
Ozone Monitoring Experiment) instrument
are assimilated with a global three-dimensional atmospheric ozone model.
The assimilation procedure is based on the Kalman filter equations,
and is an extension of an existing assimilation procedure for
total ozone columns.
As a novelty, a three-dimensional covariance model is
developed using a single parameterisation
for correlations in all directions, instead of the usually
applied separation in horizontal and vertical directions.
The parameterisation is anisotropic in all directions,
accounting for the different correlation lengths of ozone
with respect to altitude, latitude, and longitude.
The assimilation procedure includes full use of the averaging kernel
information provided with the GOME retrieval product.
The averaging kernels account for the smaller sensitivity of
the GOME instrument below the ozone maximum and the
limited vertical resolution.
A singular value decomposition of the kernels is used to reduce
the large data volume.
A one year data set of GOME ozone profiles is
assimilated for the year 2000.
Independent data from ozone sondes is used to validate
the results.
A case study shows that the assimilation of GOME profiles
is able to improve the simulation of the vertical ozone
distribution even in case of strong vertical gradients.