The ASSET intercomparison of ozone analyses: method and first results

[摘要] This paper aims to summarise the currentperformance of ozone data assimilation (DA) systems, to show where they can beimproved, and to quantify their errors. It examines 11 sets of ozoneanalyses from 7 different DA systems. Two are numerical weatherprediction (NWP) systems based on general circulation models (GCMs); theother five use chemistry transport models (CTMs). The systems examinedcontain either linearised or detailed ozone chemistry, or no chemistry atall. In most analyses, MIPAS (MichelsonInterferometer for Passive Atmospheric Sounding) ozone data are assimilated;two assimilate SCIAMACHY (Scanning Imaging Absorption Spectrometerfor Atmospheric Chartography) observations instead.Analyses are compared to independent ozone observations covering thetroposphere, stratosphere and lower mesosphere during the period July toNovember 2003.

Biases and standard deviations are largest, and show the largest divergencebetween systems, in the troposphere, in theupper-troposphere/lower-stratosphere, in the upper-stratosphere andmesosphere, and the Antarctic ozone hole region. However, in any particulararea, apart from the troposphere, at least one system can be found thatagrees well with independent data. In general, none of the differences can belinked to the assimilation technique (Kalman filter, three or fourdimensional variational methods, direct inversion) or the system (CTM or NWPsystem). Where results diverge, a main explanation is the way ozone ismodelled. It is important to correctly model transport at the tropicaltropopause, to avoid positive biases and excessive structure in the ozonefield. In the southern hemisphere ozone hole, only the analyses whichcorrectly model heterogeneous ozone depletion are able to reproduce thenear-complete ozone destruction over the pole. In the upper-stratosphere andmesosphere (above 5 hPa), some ozone photochemistry schemes caused large buteasily remedied biases. The diurnal cycle of ozone in the mesosphere is notcaptured, except by the one system that includes a detailed treatment ofmesospheric chemistry. These results indicate that when good observations areavailable for assimilation, the first priority for improving ozone DA systemsis to improve the models.

The analyses benefit strongly from the good quality of the MIPAS ozoneobservations. Using the analyses as a transfer standard, it is seen thatMIPAS is ~5% higher than HALOE (Halogen Occultation Experiment) in themid and upper stratosphere and mesosphere (above 30 hPa), and of order 10%higher than ozonesonde and HALOE in the lower stratosphere (100 hPa to30 hPa). Analyses based on SCIAMACHY total column are almost as good as theMIPAS analyses; analyses based on SCIAMACHY limb profiles are worse in someareas, due to problems in the SCIAMACHY retrievals.

[学科分类] 大气科学