Setting the path for evaluating ozone recovery in the Alps using a homogenised long-term ground-based ozone time series.

The recovery of the ozone layer is expected by the middle of the 21^st century, after the significant depletion detected during the 1980s. However, confirming this recovery critically relies on high-quality and long-term measurements, which play a key role in monitoring changes in atmospheric constituents and in detecting global trends.

Here we present the homogenised ozone time series of Arosa/Davos. This time series is based on measurements performed by Dobson and Brewer spectroradiometers covering the period from 1926 to the present and constitutes the world’s longest continuous ground-based ozone dataset. This study focuses on the 1990 to 2024 period and the eventual recovery of ozone after the international efforts to reduce the chlorofluorocarbons emissions. We use the merged ozone times series of Arosa/Davos composed of three Brewer and three automated Dobson spectroradiometers data records. To reconcile the seasonal discrepancies between the Brewer and the Dobson dataset, we followed the methodology explained in previous studies. The improvements employed includes replacing the operational ozone absorption cross-section (Bass and Paur, 1985) with the measured by the University of Bremen (Serdyuchenko et al., 2014) and correcting the ozone effective temperature using the ozone sondes measurements from Payerne. Furthermore, the measurement uncertainty was derived for each of the six instruments to produce homogenised merged ozone dataset.

The relocation of the instruments from Arosa to Davos during the period 2011-2021 was carefully analysed and allowed the determination of a constant transfer factor to ensure homogeneity of total column ozone between both sites. This factor was found to be equal to the climatological tropospheric ozone column differences. Finally, the datasets from all instruments were merged to combine one consistent single record.

The robustness of this merged ozone time series should enable the detection of an ozone recovery signal, by reducing the possibility of misinterpretation due to instrumental artefacts. As a future work, we will aim to assess long-term ozone changes and evaluated the attribution of the stratospheric and tropospheric ozone from the potential recovery detection of this homogenised time series.