Thanks to the Montreal Protocol and its Adjustments and Amendments to reduce the emissions of ozone depleting substances (ODS), stratospheric ozone is expected to slowly recover globally, although the delay of the recovery is still unsure due to the uncertainties associated to climate change. Detecting this recovery is very challenging because of the large natural variability of stratospheric ozone, and the small expected trend (a few percent per decade).
In the polar regions, ozone is especially highly impacted by the effective-chlorine levels, which leads to the well-known “ozone hole” problematic. It is therefore expected that it is easier in the polar regions to detect the recovery of ozone attributed to the decrease of effective chlorine.
Indeed, there are early signs of ozone recovery observed in Antarctica (WMO, 2018). However, the much higher ozone natural variability in the Arctic, due to larger dynamical variability, complicates the observation of the ozone recovery in the Arctic and no evidence of positive ozone trends over the Arctic was found for the 2000-2016 period (WMO 2018).
This project aims to explore if the ozone recovery in the Arctic can be detected using long-term ground-based observations by reducing the uncertainty on the trends.
Reducing the uncertainty on the trends will be obtained through three objectives:
Objective 1: Select the best quality ground-based time series in Arctic.
- Gather FTIR data from 7 NDACC (Network for the Detection of Atmospheric Composition Change) sites, located at latitudes from 60 to 80°N (see Figure 1). FTIR measurements provide total columns as well as independent partial columns in the troposphere and stratosphere. In addition, we will use ozone soundings in the Arctic, to strengthen our conclusions. In the course of the work, we have decided to further add Brewer and Dobson spectrophotometers that measures the total column of ozone since this is not covered by the ozone soundings.
- Use ground-based data sets to evaluate ozone trends from satellites (which are usually merged satellite data sets for long-term trend studies). Satellite measurements usually have larger uncertainties over polar regions, and there is a strong need to assess their long-term stability, and to verify that the merging does not introduce some step or drifts in the long-term data sets. We would like to detect possible drifts in the recent merged Limb satellite data set that were optimized during the LOTUS/SPARC initiative (MEGRIDOP, Sofieva et al., 2021), and IASI-CDR Nadir total columns and profiles from AERIS L3 products (Clerbaux and Coheur 2025a, 2025b, Keppens 2025). It was found in the TOAR initiative that the tropospheric trends from satellites do not agree, IASI being the only record showing negative trends (Gaudel et al., 2018), which highlights the need for validation of the new version of IASI.
It turned out that we can, in addition, use these two satellite data sets to identify some ground-based time-series that are outliers in terms of drifts with both satellites.
Figure 1: High-latitude/Arctic FTIR (blue), ozonesondes (red), Brewer (purple) and Dobson (green) stations used in the DORA project.
Objective 2: Use a representativeness study based on CAMS re-analysis data to define regions representative of the same ozone variability.
- Analyze the representativeness of the current network of ground-based stations using CAMS (Copernicus Atmosphere Monitoring Service) reanalysis, which combines model data with observations into a globally complete and consistent dataset.
- Create correlation maps between CAMS output at the location of each station and rest of the Arctic.
- For the sites that are found representative of the same area in the Arctic, we will construct merged time series of ozone anomalies and we will obtain regional trends, which will reduce trend uncertainties thanks to reduced variability and higher sampling.
Objective 3: Use a multiple linear regression (MLR) to reduce further trend uncertainty and explain ozone variability
- Detect and attribute long-term ozone trends by taking into account physical processes that influence natural ozone variability: this diminishes the unexplained variability while quantifying the variability associated with each process. The remaining trend is then attributed to the reduction of ozone depleting substances.
- In relation to the TOAR activities, determine the tropospheric trends at the same regions. Tropospheric ozone is one of the most important greenhouse gases, so it is crucial to monitor it and detect its trend in the Arctic environment which is warming at least three times faster than the rest of the globe.
- The consistency between the observed total and stratospheric ozone trends is currently missing or at least incomplete in ozone research. We will address this by studying the impact of tropospheric ozone trends on the total ozone trends.
