Continuous decline in lower stratospheric ozone offsets ozone layer recovery

Ozone forms in the Earth’s atmosphere from the photodissociation of molecular oxygen, primarily in the tropical stratosphere. It is then transported to the extratropics by the BrewerDobson circulation (BDC), forming a protective ‘ozone layer’ around the globe. Human emissions of halogen-containing ozone-depleting substances (hODSs) led to a decline in stratospheric ozone until they were banned by the Montreal Protocol (MP), and since 1998 ozone in the upper stratosphere 5 shows a likely recovery. Total column ozone (TCO) measurements of ozone between the Earth’s surface and the top of the atmosphere, indicate that the ozone layer has stopped declining across the globe, but no clear increase has been observed at latitudes outside the polar regions (60◦–90◦). Here we report evidence from multiple satellite measurements that ozone in the lower stratosphere between 60◦S and 60◦N has declined continuously since 1985. We find that, even though upper 10 stratospheric ozone is recovering in response to the MP, the lower stratospheric changes more than compensate for this, resulting in the conclusion that, globally (60◦S–60◦N), stratospheric column ozone (StCO) continues to deplete. We find that globally, TCO appears to not have decreased because tropospheric column ozone (TrCO) increases, likely the result of human activity and harmful to respiratory health, are compensating for the stratospheric decreases. The reason for the continued 15 reduction of lower stratospheric ozone is not clear, models do not reproduce these trends, and so the causes now urgently need to be established. Reductions in lower stratospheric ozone trends may partly lead to a small reduction in the warming of the climate, but a reduced ozone layer may also permit an increase in harmful ultra-violet (UV) radiation at the surface and would impact human and ecosystem health. 20


Introduction
The stratospheric ozone layer protects surface life from harmful solar ultraviolet radiation. In the second half of the 20th century, halogen-containing ozone depleting substances (hODSs) resulting from human activity, mainly in the form of chloroflurocarbons (CFCs), led to the decline of the ozone layer (Molina and Rowland, 1974). The clearest example of ozone depletion was signified by 25 the formation of an ozone hole over the Southern polar region, but even outside there was a clear reduction in total coulmn ozone (TCO) (Farman et al., 1985;WMO/NASA, 1988;WMO, 2011WMO, , 2014. The Montreal Protocol came into effect in 1989, banning multiple substances responsible for ozone layer depletion, and by 1997 it became apparent that a decline in TCO had ceased at almost all non-polar latitudes. 30 The general expectation is that global mean stratospheric column ozone (StCO) will increase as hODSs continue to decline, but an attribution of increasing TCO to decreasing ODSs has not yet been possible (WMO, 2014); a cooling stratosphere is also thought to aid the recovery of ozone by slowing temperature-dependent reaction rates. Models predict that mean TCO will increase, but 2 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2017-862 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 10 October 2017 c Author(s) 2017. CC BY 4.0 License. this also remains uncertain since projections rely substantially on the CO 2 , N 2 O and CH 4 emissions 35 scenarios.
Only recently has a TCO recovery been detected during the austral spring (Solomon et al., 2016).
However, elsewhere observations of global TCO levels have remained stable since 2000 (WMO, 2014), with most latitudes displaying a positive, but non-significant, decadal trend (WMO, 2014).
Results from Frith et al. (2014) suggest a potential peak in positive trends around 2011, after which 40 positive trends declined while uncertainties increased, despite longer timeseries.
In the past the attribution and identification of ozone recovery was made through multiple linear regression (MLR) analysis with most studies considering either piecewise linear trends (PWLT) to represent trends, with an inflection date usually at the end of 1997, or the equivalent effective stratospheric chlorine (EESC) proxy to represent the influence of hODSs on ozone, which is a non-45 linear smoothly-varying proxy with an inflection date also around 1997 (Newman et al., 2007).
Two recent studies, Chehade et al. (2014) and Frith et al. (2014), both investigated changes in TCO observations up to 2012 and 2013, respectively, and came to similar conclusions: trends using PWLT or EESC prior to 1997 agree that ozone declined to a minimum in 1997, but from 1997 the use of the EESC proxy suggests a significant and positive increase at all latitudes (larger at higher latitudes), 50 while use of PWLT shows peaks at mid-latitudes but is generally lower and non-significant at most latitudes outside the polar regions. These results suggest that attribution to EESC prior to 1997 is the dominant contributor to the long-term trend, but after 1997 it may be less representative, or that large dynamical variability is interfering with post-1997 trends. As both Chehade et al. (2014) and Frith et al. (2014) note, the post-1997 EESC estimate is partially locked by the large decline during 55 the pre-1998 period, while in PWLT the period to fix the estimate is not influenced by the latter or earlier period, respectively. The consequence may be, then, that a significant post-1997 change in TCO might indeed represent a hODS-related increase in ozone, but this may be embedded within ozone that is not actually increasing, or increasing at a slower rate, as shown by the PWLT that represents the overall timeseries without any specific physical attribution. 60 Despite a lack of clear recovery in TCO, ozone in the upper stratosphere above 10 hPa appears to be recovering, which has been reported with significant positive decadal trends in vertical profiles, and altitude-latitude spatial maps, from multiple ozone composites that merge observations from various space missions, especially at mid-latitudes Laine et al., 2014;WMO, 2014;Tummon et al., 2015;Harris et al., 2015;Steinbrecht et al., 2017;Ball et al., 2017; Sofieva either the uncertainties due to unattributed dynamical variability interfere in the significance of the trend determined through MLR, or there are counteracting trends at lower levels of the stratosphere, or in the troposphere.
However, it has been difficult to confirm (WMO, 2014) because: (i) ozone is typically integrated over wide latitude bands and/or total column ozone (TCO) is considered, both of which may lead to cancellation of opposing trends; (ii) large dynamical variability unaccounted for in regression analysis together with shorter timeseries lead to higher uncertainties (Tegtmeier et al., 2013); (iii) 80 below 20 km there are large ozone gradients, with low ozone close to the tropopause; and (iv) composite-data merging techniques have hindered identification of robust changes (Harris et al., 2015;Ball et al., 2017). Despite all of these issue, uncertainties between limb sounding instruments have been reported to be less than ∼10-15% near 16 km (Tegtmeier et al., 2013).
In addition to only reporting decadal percentage changes, most studies typically do not consider 85 altitudes below 20 km (∼60 hPa), missing stratospheric changes down to 16 km in the tropics (30 • S-30 • N) or ∼12 km at mid-latitudes (60 • -30 • ), regions that contain a large fraction of, and drive most sub-decadal variability in, TCO. A recent study by Bourassa et al. (2017) extended their analysis of the SAGE-II/OSIRIS ozone composite down to 18 km, where widespread, partially significant, negative ozone trends (1998 can be seen at all latitudes from 50 • S to 50 • N. Models 90 do predict a decline in lower stratospheric ozone in the tropics in the future (e.g. by 2060;Eyring et al. (2010); WMO (2011)), but evidence for Brewer-Dobson circulation (BDC) driven decreases up to present day continue to remain weak because multiple observations since 2000 do not support such a tropical stratospheric decline, which has levelled off since 2000, and decreases that have been identified between 32-36 km (near 10 hPa) km are thought to be largely due to the 2000-2003 period 95 of high ozone levels WMO (2014), and so may currently be an artefact of the analysis period rather than a change in the BDC.
Finally, issues remain in ozone timeseries analysis from both the use of the standard analysis technique employing ordinary least squares, multiple linear regression (MLR) that can lead to biased estimates (Ball et al., 2017) due to unaccounted-for residual variance, time of day and geolocation 100 biases (Sofieva et al., 2014), vertical and horizontal spatial resolution (Kramarova et al., 2013), and the presence of satellite drifts and biases introduced into composite timeseries from how they were merged (Tummon et al., 2015;Harris et al., 2015;Ball et al., 2017). All of these can lead to conflicting results between composite datasets, even those constructed using similar underlying data sources (WMO, 2014;Ball et al., 2017).

105
Our aim here is to quantify the absolute changes in ozone and contribution to TCO since 1998, i.e. not simply their relative change in percentage. We bring to bear a robust regression analysis approach (section 2.1) through dynamical linear modelling (DLM) Ball et al., 4 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2017-862 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 10 October 2017 c Author(s) 2017. CC BY 4.0 License. 2017). The major step forward that DLM provides is the estimation of smoothly varying, non-linear background trends, without the need to provide a prescribed EESC explanatory variable. Although 110 this precludes a clear physical attribution, similar to PWLT, it allows for an assessment of how ozone is evolving on decadal and longer timescales, e.g. to identify if and when an inflection in ozone occurs. We also make use of updated ozone composites (section 3), extended to 2015/6, and we begin by considering relative percentage changes since 1998 to put these new data, analysed with the DLM approach, in the context of previously reported trends for relative changes above 20 km, 115 but extended down to the tropopause (section 4.1). From there, we consider the absolute contribution of partial column ozone (PCO) from the whole stratosphere (StCO), the upper (1-10 hPa, ∼32-48 km), middle (10-32 hPa, ∼25-32 km), and lower stratosphere (32-147 hPa or ∼13-25 km at midlatitudes; 32-100 hPa or ∼16-25 km in subtropical and equatorial latitudes; section 4.2), and then the tropospheric contribution (section 4.3). We finally consider what two chemistry climate models 120 in specified dynamics mode suggest trends have been (section 4.4). We discuss our findings and conclude in section 5.

Regression analysis
The standard method to estimate decadal trends or changes in ozone, multiple linear regression 125 (MLR), is known to have estimator bias and regressor aliasing (Marsh and Garcia, 2007;Chiodo et al., 2014). To minimise these effects we use a more robust method using a Bayesian inference approach through Dynamical Linear Modelling (DLM) Ball et al., 2017). DLM  is similar to MLR in that the same regressors (see section 2.2, below) are used for known drivers of ozone variability, and an autoregressive term is included. However, the trend 130 is not predetermined with a linear, or piece-wise linear, model, but is allowed to slowly vary in time, and the degree of trend non-linearity is an additional free parameter to be jointly inferred from the data. We infer posterior distributions on the non-linear trends by Markov Chain Monte Carlo (MCMC) sampling. DLM analyses typically have more conservative uncertainties than MLR since they are based on a more flexible model, and formally integrate over uncertainties in the regression 135 coefficients, seasonal cycle and dynamics, autoregressive coefficients and parameters characterizing the degree of non-linearity in the trend. The time-varying, background changes are estimated, rather than specified by, for example, an estimate of equivalent effective stratospheric chlorine (EESC) (Newman et al., 2007) or a linear-trend; there is no need for assumptions about when and where a decline in hODSs occurs.

Regressor variables
Similar to MLR, we use regressor timeseries that represent known drivers of stratospheric ozone variability. These include: the F30 cm radio flux as a solar proxy (as it better represents UV variability than the commonly used F10.7 cm flux (Dudok de Wit et al., 2014)), a latitudinally resolved stratospheric aerosol optical depth (SAOD) for volcanic eruptions (Thomason et al., 2017), an ENSO 145 index (NCAR, 2013) representing El niño Southern Oscillation variability 1 , the Quasi-Biennial Oscillation at 30 and 50 hPa 2 . We use the Arctic and Antarctic Oscillation 3 proxy for Northern and Southern TCO and partial column ozone (PCO) trend estimates. We use a second order autoregressive (AR2) process (Tiao et al., 1990) to avoid the auto-correlation of residuals. We remove the two year period June 1991 to May 1993, inclusive, from the analysis to avoid problems related to im-150 pacts of satellite ozone retrieval due to stratospheric aerosol loading (Davis et al., 2016), and aliasing between regressors within the regression analysis (Chiodo et al., 2014); the volcanic aerosols still show slowly varying changes, which are important to consider as a regressor since this has a larger impact on ozone in the lower stratosphere than the upper.

155
We do not apply any statistical tests, which therefore avoids making assumptions about the (posterior) distributions being considered. The posteriors presented in all figures represent the full information about the change in ozone since 1998 obtained from the DLM analysis; these are not always normally distributed. The probabilities discussed, and presented represent the percentage of the total number of DLM samples (n=100,000) that decrease in ozone; positive increases have values less 160 than 50% and therefore increases at 80, 90 and 95% probabilities are indicated by their respective contours in Fig. 1 and A1, and have values less than or equal to 20, 10 and 5% in Figs. 2, A3, A4, A6, A9, and A10.  (Tummon et al., 2015;Harris et al., 2015;Steinbrecht et al., 2017). Part of the reason is related to offsets and 185 drifts in the data that continue to be one of the largest remaining sources of uncertainty within, and between, ozone composites (Harris et al., 2015;Ball et al., 2017). These artefacts can be largely accounted for using the methodology developed by (Ball et al., 2017), which we apply to both pairs of data separately; examples of corrected timeseries in the lower stratosphere are given in Fig. A2, and others can be found in Ball et al. (2017). This method also fills data gaps, which is reasonable 190 if they are discontinuous for only a few months. This is true for these datasets, but is not for the SAGE-II/CCI/OMPS and SAGE-II/OSIRIS/OMPS. SWOOSH, SBUV-Merged-Cohesive and GOZ-CARDS have been updated since previous intercomparisons (Tummon et al., 2015;Harris et al., 2015). GOZCARDS v2.20, used here, includes SAGE-II v7.0 and has a finer vertical resolution than earlier versions. It must be stressed that the resolution of SBUV-instruments below 22 hPa (25 km) 195 is low Kramarova et al., 2013), so linear trends estimated at 25-46 hPa also encompass altitudes lower than those that they formally represent (see section 4 for a discussion on this).

Total column ozone 200
We use merged SBUV v8.6 (Frith et al., 2014) for comparison of results with total column ozone (TCO) observations, which are available on a 5 • latitude grid from 1970 onwards. We verify stability of SBUV TCO after 1997 by comparing SBUV TCO overpass data with the independent Arosa ground measurements, which are available from 1926 to present (Scarnato et al., 2010).

Tropospheric column ozone
For tropospheric ozone, we consider OMI/MLS tropospheric column ozone measurements, discussed by Ziemke et al. (2006). The tropospheric ozone are estimated through a residual method that derives daily maps of tropospheric column ozone (TrCO) by subtracting MLS stratospheric column ozone (StCO) from co-located OMI total column ozone. The OMI/MLS data, including data 210 quality and data description, are publicly available 4 . Coverage of the OMI/MLS ozone is monthly (October 2004-present) and at 1 • ×1.25 • horizontal resolution, which we have zonally averaged to make comparisons here.  Merged-SBUV and Merged-SWOOSH/GOZCARDS composites show 95% probability that upperstratospheric ozone at all latitudes between 60 • S and 60 • N has increased. This is less robust in SAGE-II/CCI/OMPS and SAGE-II/OSIRIS/OMPS, which show differences at equatorial latitudes (10 • S-10 • N). The reason for the difference is not clear, but we note that in this region nearly 50% of the data are missing in the first five years (1998)(1999)(2000)(2001)(2002), while Merged-SWOOSH/GOZCARDS and 225 Merged-SBUV have no missing data (Harris et al., 2015).
In contrast to the upper stratosphere, all four composites show a consistent ozone decrease below 32 hPa / 24 km at all latitudes (Fig. 1)  qualitatively disagree with previous studies and approaches (WMO, 2014). However, four additional years of data (Tummon et al., 2015;Harris et al., 2015), an improved regression analysis method Ball et al., 2017) (see section 2), and techniques to account for data artefacts (Ball et al., 2017), means we are now able to confidently identify changes in the lower stratosphere. The spatial trends presented in Fig. 1 are informative for understanding where, and assessing why, changes in stratospheric ozone are occurring. However, stratospheric ozone changes are usually reported as decadal percentage change vertical profiles or spatial maps (e.g. as in Fig. 1 hides the absolute changes in ozone, and the contribution to the total column, which are almost never 255 reported. A recovery in the upper stratosphere is important to identify, but this region contributes a smaller fraction to the total column than the middle and lower stratosphere. Thus, smaller percentage changes over a reduced altitude range in the lower stratosphere can actually produce larger integrated changes than in the more extended regions higher up. In Fig. 2 we present changes in partial column ozone (PCO) in Dobson Units (DU) from Merged-260 SWOOSH/GOZCARDS for the whole stratospheric column (StCO), and for the upper (10-1 hPa), and lower stratosphere (147-32 hPa or 13-24 km at >30 • ; 100-32 hPa or 17-24 km at <30 • ), respectively. We note that the tropopause, the boundary layer between the troposphere and stratosphere, varies seasonally, but is on average around 16 km (tropics) and 10-12 km (mid-latitudes); our conservative choice of slightly higher altitudes ensures that we avoid including the troposphere. Due to Upper stratospheric ozone (Fig. 2, middle row) has increased since 1998 in almost all latitude bands, in half the cases at >90% probability, and >95% at 40 • -60 • in both hemispheres. Globally, the probability exceeds 99% that upper stratospheric ozone has increased, confirming that the MP has indeed been successful in reversing trends in this altitude range.
Integrating the whole stratosphere vertically, to form the stratospheric column ozone (StCO;  lower stratosphere in 10 • bands for all latitudes (left) and integrated from 60 • S-60 • N ('Global', right). The stratosphere extends deeper at mid-latitudes than equatorial (marked above each latitude).
Numbers above each distribution represents the distribution-percentage that is negative; colours are graded relative to the percentage-distribution (positive, red-hues, with values <50; negative, blue). SBUV total column ozone (red curves) is given in the upper row and negative distributionpercentages are given as red numbers.
We note that uncertainty remains in the middle stratosphere (Fig. A6), with Merged-SWOOSH/GOZCARDS, SAGE-II/CCI/OMPS, and SAGE-II/OSIRIS/OMPS displaying different changes. SAGE-II/OSIRIS/OMPS, in particular, shows a significant positive trend, which leads to the global StCO indicating no change since 1998 (Fig. A3). This is likely a result of how the data were merged to form composites (see  To make these globally-integrated results clear, we show in Fig. 3a the SBUV TCO (yellow/red) 295 and Merged-SWOOSH/GOZCARDS StCO (grey/black); in all of the panels in Fig. 3, the timeseries are bias-shifted so that the smoothly varying non-linear trend crosses the zero line in January 1998, so that relative changes can be clearly compared. It is interesting to note here that the SBUV TCO non-linear trend initially increases from 1998, and then peaks in around 2011, before decreasing.  Fig. 3b, increasing by a mean of ∼1 DU, and trends have been relatively flat since 1998 in the middle stratosphere (Fig. 3c), with a mean decrease of ∼ 0.5 DU. However, the result from Merged-SWOOSH/GOZCARDS in the 315 lower stratosphere (Fig. 3d) indicates not only that ozone there has declined by ∼2 DU since 1998, and has been the main contributor to the StCO decrease, but that the lower stratospheric ozone has seen a continuous and uninterrupted decrease.

Tropospheric ozone contribution to TCO
The stratosphere accounts for the majority (∼90%) of TCO, so intuitively attribution to TCO changes 320 would be expected to come primarily from this region. However, the results in Fig. 2 and 3 suggest a discrepancy between StCO and TCO. Despite this, there is no serious conflict between the different changes indicated by global StCO and TCO distributions (Fig. 2) and trends (Fig. 3a), when the remaining 10% of the TCO, i.e. tropospheric ozone, is considered, as we show in the following.
First, it is important to establish confidence in the SBUV TCO observations. These have been 325 very stable since 1998 when comparing SBUV TCO overpass data to the independent ground-based Arosa TCO observations (Fig. A8). This, therefore, provides confidence in the result that there is little net change in TCO since 1998. Additionally, Chehade et al. (2014) reported that other TCO composites agree very well with the SBUV TCO and there is little difference between the various TCO composites when performing trend analysis.  for ozone trend studies, that OMI has proven to be one of the most stable instruments flown, and they concluded that OMI provides some of the highest quality ozone data from trend analysis avaliable. Ziemke and Cooper (2017) found no statistically significant drift with respect to various independent measures, or between MLS StCO and OMI StCO residuals, but did detect a small drift of +0.5 DU per decade in OMI/MLS TrCO caused by an error in the OMI total ozone -this was rectified for the 365 version we consider here.
A deeper investigation is needed to understand difference in the contributions of TrCO and StCO to TCO, especially considering uncertainties carefully, but this is beyond the scope of this work.
We note that studies using various data sources show less significant regional increases (and some decreases) with global estimates ranging from 0.2 to 0.7% per year (∼0.6-2 DU per decade) (Cooper 370 et al., 2014;Ebojie et al., 2016;Heue et al., 2016), though these estimates considered different time periods; this suggests a large range of uncertainty, but even the lower end of the estimated increases in TrCO are in line with the missing part of the TCO change, after considering StCO, that we estimate here. Tropospheric ozone is not the main focus of the study here, but the evidence presented overall suggests the missing component in the declining StCO distributions and trends, 375 with respect to constant TCO, is indeed from increasing tropospheric ozone.

Comparison of stratospheric spatial and partial column ozone trends with models
The observational results for the lower, and whole, stratosphere presented thus far have not been previously reported. However, it is not clear that this represents a departure from our understanding of stratospheric trends as presented in modelling studies. We present the percentage ozone change 380 from two state-of-the-art chemistry climate models (CCMs) in   (Dee et al., 2011). Thus, the two models are both independent in terms of how they are constructed, and the source of nudging fields used, but have similar boundary conditions as prescribed by CCMI-1.
In Fig. 5 both models display broadly similar behaviour in the upper stratosphere above 10 hPa, 395 roughly in line with the observations (Fig. 1). Spatially, in the middle stratosphere there are differences in sign, but generally significance is low: WACCM-SD displays broadly positive changes except in the tropics at 10 and 30 hPa, while SOCOL-SD displays a negative spot centred in the tropics at 10 hPa, while mid-latitudes are often positive and significant. In the lower stratosphere, SOCOL-SD displays negative trends in the Southern hemisphere lower stratosphere, but positive 400 in the Northern, while WACCM-SD is generally positive everywhere, and significant at the lowest altitudes, except at 30-40 hPa in the tropics where a negative tendency is seen. In both SOCOL-SD and WACCM-SD, trends in the lower stratosphere are generally not significant, and do not display the clear and significant decreases found in the observations. Posterior distributions similar to those of Fig. 2 are presented for SOCOL-SD and WACCM-SD in Figs. A9 and A10, respectively. The It is notable that also the suite of CCMVal-2 models, the predecessor of CCMI-1, show little significant behaviour in the lower stratosphere, with a tendency for positive changes. As shown in above ∼10 hPa are trends significant. Furthermore, there is an ozone increase in the mid-latitude lower stratosphere, albeit non-significant, indicated by the CCMVal-2 models that is not seen in the observations, suggesting that models may not be simulating that region correctly. Extending to 2016 with two independent nudged models, as shown here, does not change this result, which differs from 425 the (i) significant decreases in ozone found in the lower stratosphere, and (ii) the stalled recovery seen in SBUV TCO while models project continued increases.
Chemistry climate models (CCMs) represent our integrated understanding of processes that govern ozone variability and trends, and include chemistry, transport and feedbacks on radiation. Overall, they capture the historical behaviour in the stratosphere well (e.g. total column ozone trends 430 driven by EESC changes). However, when it comes to the UTLS region it is not yet clear if models do so well. For example, Figs. 7.27 and 7.28 of the CCMVal-2 report SPARC/WMO (2010) indicate a better model performance with respect to UTLS ozone in summer, when transport effects are weaker and chemistry more important. However, there is a large difference compared to observations and a wide spread among the models during winter/spring. Transport is affected by many factors, 435 e.g. model vertical/horizontal resolution and gravity wave parameterizations, and trends in atmospheric circulation are also hard to measure and, therefore, to assess the models with. Whether the difference between the models and observations is a result of model design, incorrect boundary conditions (e.g. aerosol contributions from anthropogenic (Yu et al., 2017)

Conclusions
In summary, we have presented evidence of highly significant changes in stratospheric ozone between 1998 and 2016. The main findings are that: (i) the MP is further confirmed to be successfully reducing the impact of hODSs as indicated by (iv) indications of no decrease, or perhaps an increase, in TCO is likely a result of increasing tropospheric ozone, together with the slowed rate of decrease in stratospheric ozone following the MP.
(v) state-of-the-art models, nudged to have historical atmospheric dynamics as realistic as possible 455 do not reproduce the observed decreases in lower stratospheric ozone, which may suggest deficiencies in some aspect of the modelling.
The cause for the continuing decline in lower stratospheric ozone is not fully understood and determining the exact cause is beyond the scope of this study, but there are several possible explanations. CCM simulations indicate that tropical stratospheric ozone is expected to decrease following 460 increased upwelling in the tropics (<30 • ) linked to an acceleration of the BDC from greenhouse gas (GHG) induced climate change, which has a larger influence on ozone trends than hODSs in this region (Randel and Wu, 2007;Oman et al., 2010;WMO, 2014); this may account for some of the tropical lower stratosphere ozone decrease, but clear evidence for this in observations remains weak (WMO, 2014). Some modelling and studies also indicate that a rise in the tropopause (Santer et al.,465 2003), due to the warming troposphere, could lead to a localised ozone decrease (Steinbrecht et al., 1998), though it is not clear of how TCO is affected on large scales, Plummer et al. (2010); Dietmüller et al. (2014); since the troposphere is continuing to warm, the tropopause may continue rising and have an affect on stratospheric ozone. We also pose the hypothesis that an acceleration of the lower stratospheric BDC shallow branch in response to climate change (Randel and Wu, 2007;Oman 470 et al., 2010) may more rapidly transport ozone poor air to the mid-latitudes from the tropical lower stratosphere, where dynamical changes dominate over photochemical ozone production processes (Johnston, 1975;Perliski et al., 1989). While, these possibilities are dynamically-driven responses to climate change, a chemically-driven alternative has also been suggested. Observations indicate an increase in very short lived substances (VSLSs) containing chlorine and bromine species from both 475 18 Atmos. Chem. Phys. Discuss., https://doi.org/10.5194/acp-2017-862 Manuscript under review for journal Atmos. Chem. Phys. Discussion started: 10 October 2017 c Author(s) 2017. CC BY 4.0 License. anthropogenic and natural sources (Hossaini et al., 2015). Modelling studies imply that VSLSs preferentially destroy ozone in the lower stratosphere, particularly at mid-and high-latitudes (Hossaini et al., 2015(Hossaini et al., , 2017. It is thought that these species may delay the restoration of the ozone layer to pre-1960s levels, but information is available for only a small number of VSLSs and knowledge of the reaction rate kinetics to determine their impacts is currently not adequate.

480
While the reason for the lower stratospheric ozone decline is not yet determined, the signal is clear and the likely consequences significant. The MP is working, but a reduction in harmful UV radiation reaching the surface to pre-1980's levels depends on a restoration of the TCO (WMO, 2014); the lower stratospheric ozone decline appears to be inhibiting this, and models as yet do not reproduce these downward trends with significance. Increased transport of ozone into the troposphere from 485 the stratosphere is expected if global surface temperatures continue to increase, and may impact air quality (Hegglin and Shepherd, 2009;Neu et al., 2014); current trends suggest that ozone available for such exchange is decreasing. Additionally, ozone in the lower stratosphere is an important factor in radiative forcing (RF) of the climate (Randel and Thompson, 2011), and so far has offset some of the RF increase from rising GHGs; a reduction in lower stratospheric ozone may lead to reduced RF 490 and further offsetting. Finally, the restoration of the ozone layer is essential to reducing the harmful effects of solar UV radiation on surface life, including humans (Slaper et al., 1996). It is imperative that we determine the cause of the decline in lower stratospheric ozone identified here, both in order to predict future changes, and to determine if it is possible to prevent further decreases.
were funded by the SNSF project 163206 (SIMA). We thank the SPARC LOTUS working group as a forum for discussion and data exchange. Work at the Jet Propulsion Laboratory was performed under contract with the National Aeronautics and Space Administration. GOZCARDS ozone data contributions from Ryan Fuller