1. Impact of approximations in the recommended Dobson algorithm on total column ozone measurements at four Australian sites
M.B. Tully 1
1 Bureau of Meteorology, Melbourne Victoria, Australia.
Corresponding author
ABSTRACT
Observations of total column ozone made by the GAW network of Dobson spectrophotometers are processed by each operator using a standard algorithm, with the resulting processed data, rather than raw observations, being submitted to the World Ozone and Ultraviolet Data Centre and from there, available to data users. For simplicity, the official WMO algorithm uses a crude parameterised value of mean ozone height by latitude, and assumes a constant “effective ozone temperature” for all locations and times of year. These approximations potentially interfere with the comparison of Dobson measurements with those from other instruments, as well as introducing the possibility of a spurious ozone seasonal cycle, trend, or response to signals such as the QBO. Here, daily ozone and temperature profiles from the MERRA-2 reanalysis (Rienecker et al. 2011) are used to calculate the mean ozone height and effective ozone temperature at four long-standing Australian Dobson sites (Darwin, Brisbane, Melbourne/Airport and Macquarie Island) for each individual day from 1980 to 2015, also making use of the current state-of-the-art ozone cross-sections (Serdyuchenko et al. 2014), and the effect on the calculation of total column ozone investigated. The effect on the calculated ozone of the annual cycle in temperature ranges from 2.5% at Macquarie Island to 0.5% at Darwin, but day to day variations can be twice this amount. The effect of the seasonal cycle in mean ozone height is negligible at Darwin but increases poleward, reaching 0.6% for measurements made two hours from local noon at Macquarie Island.
These results suggest the standard Dobson algorithm should be updated to incorporate reanalysis data in the ozone retrieval.
References
Rienecker, M.M., M.J. Suarez, R. Gelaro, R. Todling, J. Bacmeister, E. Liu, M.G. Bosilovich, S.D. Schubert, L.Takacs, G.-K. Kim, S. Bloom, J. Chen, D. Collins, A. Conaty, A. da Silva, et al., MERRA: NASA’s Modern-Era Retrospective Analysis for Research and Applications. J. Climate, 24, , doi: /JCLI-D , 2011.
Serdyuchenko, A., Gorshelev, V., Weber, M., Chehade, W., and Burrows, J. P.: High spectral resolution ozone absorption cross-sections – Part 2: Temperature dependence, Atmos. Meas. Tech., 7, , doi: /amt , 2014.
Map of the four Australian Dobson sites Darwin, Brisbane, Melbourne and Macquarie Island considered here (black).
Ozone-weighted temperature by day of year for the period. These plots show (1) the annual cycle increases in magnitude with increasing poleward distance (2) the range for a given day of the year is a significant fraction of the annual cycle (3) the temperature used in the Dobson algorithm is reasonable at all sites but slightly too low at Darwin.
The time series of effective ozone temperature shows that, while the variability is dominated by the annual cycle, inter-annual variability is also evident and so the potential exists for spurious signals in the ozone record resulting from stratospheric temperature trends and oscillations such as QBO.
Sedyuchenko et al laboratory values for ozone cross sections were used along with proscribed Dobson slit-functions to calculate the final ozone cross section as used in the Dobson algorithm (to which the calculated total ozone value is directly proportional), and expressed as a ratio of the cross-section at fixed temperature. The 10th-90th percentile range varies from approximately ±0.2% at Darwin to ±2% at Macquarie Island. The use of a climatology in the algorithm would capture some but not all of the variability.
The error in the airmass factor (µ) and hence the calculated value of total ozone due to the approximation used for mean ozone height increases with solar zenith angle and becomes of some significance at Macquarie Island, however is much smaller in magnitude than the error introduced by fixed temperature.