A century ago, scientists in Paris and Oxford started measuring total column ozone. These ground-based observations continued, even after the advent of the satellite era, and we now look back at almost a century of measurements. This is important in the context of the expected recovery of the ozone layer. Although the concentration of ozone depleting substances in the atmosphere has decreased since the late 1990s, column ozone measurements from mid-latitudes show no clear recovery. Have we overlooked some chemical interactions? Is climate change playing a trick? It is known that changes in atmospheric dynamics, both locally (by changing the tropopause height) and on large-scale (by changing ozone transport from the tropics to the extratropics) may overprint chemical changes. A long record may better capture this variability during a more undisturbed period than is accessible from satellite information. Therefore, long records could help to answer this question. So far there has been only one very long record: the one from Arosa, Switzerland. Could there be a second long record?
Total column ozone is closely linked with one name: Dobson, lecturer at University of Oxford. He was the first, in 1924, to make regular measurements and continued to measure until 1975, shortly before his death. Despite his prominence, Dobson’s data were never used much in the ozone community and were not digitally available for a long time. Some 17 years ago, I visited the University of Oxford together with my colleague Johannes Staehelin from ETH Zurich. We found Dobson’s original measurement sheets and photocopied them. They were subsequently digitised, processed and published by Christian Vogler in his Master thesis.
After that paper was published, we received an email by Bob Wells, former assistant of Dobson. He wrote that they had actually continued measurements for two years after Dobson’s death. The observation sheets for these two years, comprising 1209 measurements, were attached to the email. The files were lying around on my hard disk for a long time, until eventually I found time during a sabbatical (and the COVID lockdown) to re-evaluate these data. Then I prolonged the series to the present using column ozone observations form nearby sites.
This new Oxford record, 1924 to present, can now be analysed together with the one from Arosa to disentangle chemical and dynamical effects on total column ozone. I did this using a multiple regression approach, similar as in many other studies,[4-6] but with data that reach twice as far back in time. This brought some further challenges as the dynamical predictors (e.g., tropopause height) also need to be available. The results essentially confirm work based on shorter records, work that has been performed within the framework of SPARC’s LOTUS project (https://www.sparc-climate.org/activities/ozone-trends/), among others. My results indicate that a chemical recovery of ca. +8 DU since the late 1990s has taken place at the locations of Arosa and Oxford. However, this was masked by a dynamically induced trend of -5 DU.
This finding also brought me back to my own scientific roots: Twenty years ago, I studied the link between total column ozone and atmospheric dynamics in order to make use of historical column ozone data to deduce dynamical changes. In the meantime we have better data on the past atmospheric circulation as well as additional ozone data and so this new work was an opportunity to revisit – and confirm - my earlier work. I am still fascinated by the link between atmospheric circulation and ozone and I am impressed by the work and dedication by these early observers. The chemical and radiative properties of the ozone molecule, combined with the sharp vertical and spatial gradients in its atmospheric concentration make ozone an extremely interesting quantity. Dobson anticipated that in 1924 when he justified his measurements by stating that the distribution of ozone in the atmosphere “may have important relations to other geophysical phenomena”.
Read the full paper: https://doi.org/10.1038/s43247-022-00472-z
Explore the new Oxford record: https://doi.org/10.6084/m9.figshare.19698274
 Dobson, G. M. B. & Harrison, D. N. Measurements of the amount of ozone in the Earth’s atmosphere and its relation to other geophysical conditions. Proc. Phys. Soc. London A110, 660–693 (1926).
 Staehelin J. et al. Total ozone series at Arosa (Switzerland): Homogenisation and data comparison. J. Geophys. Res. 103, 5827–5841 (1998).
 Vogler, C., Brönnimann, S., Staehelin, J. & Griffin, R. E. M. The Dobson total ozone series of Oxford: Re-evaluation and applications. J. Geophys. Res. 112, D20116 (2007).
 Steinbrecht, W., Hegglin, M. I., Harris, N. & Weber, M. Is global ozone recovering? Comptes Rendus Geoscience 350, 368-375 (2018).
 Weber, M. et al. Global total ozone recovery trends attributed to ozone-depleting substance (ODS) changes derived from five merged ozone datasets. Atmos. Chem. Phys. 22, 6843–6859 (2022).
 Coldewey-Egbers, M., Loyola, D. G. Lerot C. & Van Roozendael, M: Global, regional and seasonal analysis of total ozone trends derived from the 1995–2020 GTO-ECV climate data record. Atmos. Chem. Phys. 22, 6861–6878 (2022).
 SPARC/IO3C/GAW, Report on Long-term Ozone Trends and Uncertainties in the Stratosphere (LOTUS) (Ed. Petropavlovskikh, I., S. Godin-Beekmann, D. Hubert, R. Damadeo, B. Hassler & V. Sofieva, SPARC Report No. 9, GAW Report No. 241, WCRP-17/2018 (2019).
 Brönnimann, S., J. Luterbacher, J. Staehelin, T. M. Svendby, G. Hansen, & T. Svenøe, Extreme climate of the global troposphere and stratosphere in 1940-42 related to El Niño. Nature 431, 971-974 (2004).