S P A R C Stratospheric Processes And their Role in Climate A project of the World Climate Research Programme Home Initiatives Organisation Publications Meetings Acronyms and Abbreviations Useful Links

1.9. Summary and Conclusions

In this chapter the sources of data used in this document have been reviewed to provide information on the measurement techniques, the data analysis methods used, and their characteristics and sources of error, both for individual measurements and for the extent to which the measurement of long-term trends are affected.

Ground-based Dobson instruments, observing ozone profiles in the Umkehr mode, have the longest heritage of ozone measurements with substantial records at some stations from 1957. The network of instruments is well calibrated and produces reliable ozone profiles in the stratosphere between approximately 20-40 km. Vertical resolution is relatively coarse, but adequate for analysis of ozone trends. Spatial coverage is limited by the relatively small number of ground stations but could be improved substantially by including the Brewer instrument network, an independent, well-calibrated system, which also makes Umkehr observations, and has been in operation since the mid 1980s.

While chemical ozonesondes perform well below about 25 km, above this altitude, uncertainties related to gas sampling techniques become a limiting factor not present in the UV-absorption in-situ instruments. At altitudes beyond the capability of aircraft (about 20 km) and balloons (about 40 km) remote techniques are clearly necessary. The lidar and microwave remote ozone profiling techniques, operated from the surface, provide adequate accuracy in the 25-50 km range and allow for continuous measurements, not generally possible with balloon and aircraft measurements. The coverage by microwave instruments is not yet adequate for them to be considered in this report.

In terms of global and temporal coverage, satellite-based instruments are obviously superior to surface-based ones. The various satellite instruments have different individual characteristics with respect to long-term calibration stability, global coverage, vertical resolution or sensitivity to contamination by stratospheric aerosol.

The solar occultation instruments, SAGE and HALOE, which utilise limb extinction at several wavelengths during sunrise and sunset events, have good long-term stability as they are able to calibrate before (sunset) or after (sunrise) each vertical profile measurement by viewing the sun directly with negligible intervening atmosphere. Vertical resolution is good, being on the order of 1 km. Spatial coverage is relatively poor because of the requirement of a sunrise or sunset.

Limb viewing in the visible/near infrared is subject to contamination by volcanic aerosols. Since a major volcanic eruption, which perturbs the stratospheric aerosol distribution, can be anticipated on average at least every ten years, this presents a limitation for maintaining long-term records with present instruments. Future instruments, with increased wavelength resolution, should allow an improvement in aerosol extinction corrections.

The nadir-viewing buv-type instruments have good global coverage for ozone profiles above about 20 km, but are subject to calibration uncertainties, the possibility of long-term drift, and have poor vertical resolution. Aerosol contamination is present, but is less of a problem for these instruments.

The MLS instrument, operating in the microwave wavelength range, is the least susceptible to aerosol contamination. Its global coverage is limited to viewing from 34o in one hemisphere to 80o in the other during a each orbital precession period (about 36 days) of the UARS spacecraft, although this is a platform, not an instrument, limitation. Its vertical resolution (potentially 3 km, but currently 5.4 km) and its calibration stability are intermediate to the other two satellite techniques.

The accuracy of estimates of ozone trends will depend on drift in systematic measurement errors in, as well as on the statistical fitting error due to sampling limitations in the presence of natural variability. Table 1.16 summarises the drift uncertainty (1s) in the calibration of each data source, but must be read in conjunction with the detailed description and caveats given in the individual sections. The analysis of drift uncertainty is still at an early stage, as it can involve many subtle effects that may not normally be considered for single-profile error analysis. This table should be regarded as preliminary, and to be treated with caution.