There is evidence that the ozone layer has begun to recover due to the ban on the production of halogenated ozone-depleting substances by the Montreal Protocol and its amendments. However, despite overcoming the decline in global total column ozone, there has been no sustained recovery in near-global [55°N-55°S] total ozone. Although recent studies report increases in tropospheric ozone and confirm upper stratosphere ozone recovery, they also indicate that lower stratosphere ozone recovery in the tropics and mid-latitudes is slower than expected. In addition, some studies have reported signs of a continuous decline in near-global ozone in the lower stratosphere. Moreover, it has been shown only by observations but current global chemistry-climate models (CCMs) do not reproduce them, demonstrating either positive or near-zero trends in lower stratospheric ozone. This makes it difficult to reliably determine ozone recovery and raises debate about the ability of modern CCMs to model future ozone trends.
We applied the new advanced earth system model SOCOLv4 to evaluate its ability to model ozone trends extracted from observations and reanalysis. We used dynamic linear modeling (DLM) to estimate long-term trends in ozone over the historical period 1985-2018 in the SOCOLv4 reference experiment and compared them with a homogenized BAyeSian Integrated and Consolidated (BASIC) ozone composite, as well as reanalysis data for different atmospheric layers. The analysis is carried out separately for periods of ozone depletion [1985–1997] and ozone recovery [1998–2018]. The recovery of ozone in the mesosphere, upper and middle stratosphere, the absence of a robust depletion of ozone in the extrapolar lower stratosphere, and the steady increase in tropospheric ozone are obtained and explained.
The modeled ozone trends are generally consistent with the observations and reanalysis, but the statistical significance in the lower stratospheric ozone is lower. Yet, we claim that current chemistry-climate models are generally capable of modeling observed ozone changes, justifying their use to predict future ozone behavior. However, further efforts are needed to study the observed near-global ozone decline in the lower stratosphere and the reasons why it has not yet been adequately reproduced in climate models.