Is St Stratosp ospheric O c Ozone Recover ering a ng as Ex - PowerPoint PPT Presentation

Is St Stratosp ospheric O c Ozone Recover ering a ng as Ex Expec pected ed? SPARC collaborators: I. Petropavlovskikh(1,2), S. Godin-Beekmann (3), D. Hubert (4), K.-L. Chang (5,2), K. Tourpali (6,7), R. Damadeo (6), V. Sofieva (8) and B. Hassler (9,10) GMD: K. Miyagawa (2), G. McConville(1,2), A. McClure (1,2), A. Jordan (1,2), B. Johnson(2), P. Cullis (1) CIRES, University of Colorado, (2) NOAA ESRL/GMD, Boulder, (3) UVSQ/CNRS, Guyancourt, France, (4) Royal Belgian Institute for Space Aeronomy, Brussels, Belgium, (5) NOAA/GMD, National Research Council Post-Doc, Boulder (6)NASA Langley Research Center, (7) LASP, Aristotle University of Thessaloniki, Greece, (8) FMI, Helsinki, Finland, (9)Bodeker Scientific, Alexandra, New Zealand , (10) DLR, Institut für Physik der Atmosphäre, Oberpfaffenhofen, Germany WMO GAW, SPARC, NDACC, NOAA, and many observation and chemistry climate model data providers GMD Review and GMAC 2018 1

Ozon one’s downs an and u ups • 1960s – NOAA ozone records Observed Predicted • 1970s – NOAA ODS records • 1980s – Ozone hole is discovered in Antarctica (Bryan Johnson talk) • 1987: Montreal Protocol leads to reductions in ozone depleting substances, now down 20-45% from the peak. • 2016: Signs of Antarctic ozone layer healing, ground-based and model data (Solomon et al., 2016). • Is Stratospheric ozone recovering globally? WMO, 2014 GMD Review and GMAC 2018 2

Histor ory of of WMO and nd SP SPARC o C ozon one t e trend end asses essmen ents Trends after 2000 Harris s et et a al. (2015 015) WMO ( (2014) 14) To address the differences between WMO/UNEP 2014 Ozone Assessment and SI2N Atmospheric pressure, hPa initiative (Harris et al, 2015) , a new WMO/SPARC LOTUS (Long-term Ozone Trends and Uncertainties in the Stratosphere) activity was initiated in 2016. GMD Review and GMAC 2018 3

Datase sets used ed i in LOTUS a and WMO Ozone a e asse sessmen ent 20 2018 GMD provided 7 ozonesonde (including SHADOZ), • Eight combined global datasets created from 6 Umkehr and 14 Dobson total column records for multiple satellite LOTUS and WMO 2018 trend assessments. records • Ground-based data, total 43 ozone profile records. • Seven Model datasets from Chemistry- Climate Model Initiative (Ref2 project) GMD Review and GMAC 2018 4

Satellit llite r records: S Stabil ilit ity a and offsets ts • Use of ground-based data to assess drifts in satellite data • Remove offsets between satellite records to create combined datasets. Aura MLS satellite differences from ozonesondes, Hilo Satellites vs Dobson at Boulder 40 30 Altitude, km 20 10 2004 2007 2010 2013 2016 Anomaly of percentage difference SAT-GND wrt median difference in reference period, (Hubert et al, 2018) GMD Review and GMAC 2018 5 Posters by J. Witte, P. Cullis., K. Miyagawa, G. McConville

Method hods fo for estimat ating ozo zone trends Normalized variability, % Natural variability • Multiple linear regression • Natural variability effects: • 11-years solar cycle (Solar flux 10.7cm) • QBO (2 orthogonal components) • ENSO (El Nino/La Nino oscillations) 1980 1990 2000 2010 2020 • Stratospheric aerosols (Volcanic eruptions) • Dynamical proxies (Northern Annular Mode, Southern Annular Mode, Eddy Heat Flux, tropopause pressure GMD Review and GMAC 2018 6

Trend Results: s: 7 7 C CCMI Model els a s and 8 8 S Satel ellite c e com ombined r records • “LOTUS” multiple regression CCMI models mean trend analyses applied to all datasets. • https://arg.usask.ca/docs/LOTUS_regression/ • 8 combined satellite records show similar trend patterns but clear discrepancies exist • Upper stratospheric trends agree with CCMI model expectations, but lower stratospheric trends are varied and uncertain (Ball et al, 2018) • Resolving difference: revisiting the merging process, i.e. using GMD ground-based data records GMD Review and GMAC 2018 7

Trend Results: C Com omparison of of Sa Satellites w with Grou ound-based St Station ons • Consistency in Ground-based (GB) and satellite trends 35N-60N, post 2000 trends provide confidence in derived trends • GB broad band trends are influenced by limited sampling (even single-station coverage), thus larger uncertainties 10 0 • However, GB ozone observations in the upper and middle stratosphere are representative of zonally Pressure, hPa averaged trends, but within narrow bands 1998-2015 10 1 SBUV Zonal averages Lidar stations 10 2 Ozone Trend, %/decade Ozone Trend, %/decade GMD Review and GMAC 2018 8

GMD records: Measuring long-term changes in stratospheric and tropospheric ozone LOTUS 2018 and Ozone Lower stratosphere • Boulder, Assessment 2018 used GMD data. Trend model fit: 2000-2016 GMD helped to develop statistical Ozone anomaly, % • 40 Trend + Solar + QBO + ENSO 40 models to interpret trends in 20 ozone profiles and total column. 50 Lower stratosphere and 0 • Altitude, km troposphere – larger variability -20 30 and thus harder to detect trends Sonde, 16-20 km -40 and attribute sources. 1977 1987 1997 2007 2017 20 Ozonesonde homogenization • Year improves confidence in trends 10 Further work is needed Lower troposphere • Ozone anomaly, % 40 Sonde (2-5 km) 0 Oral presentations on Wed 20 -3 -2 -1 0 1 2 3 by A. Gaudel, D. Tarasick, A. Langford 0 Poster: A. McClure, K-L Chang, K. Ozone Trend, %/year -20 Miyagawa, K. Minschwanner Niwot Ridge (~3 km) -40 1977 1987 1997 2007 2017 9 Year

Concl Con clusi sions a and nd N Next St Step eps • Ozone is recovering in the upper stratosphere • Magnitude and patterns are consistent in different datasets and in model simulations. • recovery trends (2-3 % per decade) in NH are the most significant. • Lower stratosphere • Large uncertainties and discrepancies between models and observations. • Complicated ozone variability due to dynamical effects or ODS replacements (Ball et al., 2018). • Further analyses are needed – GMD ozone records can help! • WMO/SPARC LOTUS report to be published (May 2018)! • Future plans: • Thoroughly investigate drifts and implement corrections • Expand trend studies: total column data, tropospheric ozone and ozone in polar regions • Explore trends in UTLS in conjunction with other SPARC efforts, i.e OCTAV-UTLS activity (I. Petropavlovskikh, G. Manney, P. Hoor) • – GMD ozonesonde records are essential! GMD Review and GMAC 2018 10

B) Measuring long-term changes in stratospheric ozone – To allow an understanding of ozone column changes by altitude (ODS+GHG+transport) Is ozone in lower stratosphere still decreasing? Ball et al (2018) analyses are based on satellite records 20⁰S - 20 ⁰ N Sonde - SBUV 40 % 30 % 25.45 - 16.06 hPa 20 % 10 % 0 % -10 % -20 % -30 % -40 % 1982 1987 1992 1996 2002 2007 2012 2017 1982 1987 1992 1996 2002 2007 2012 2017 Homogenization for GMD (Sterling et al, 2018) and SHADOZ (Witte et al, 2017) ozonesonde data - improved records for future trend analyses Trends in the low Satellite and CCMI model stratosphere will be soon averaged trends (LOTUS, assessed from homogenized 2018, Ozone Assessment) ozone-sonde data in tropics - disagreement between and middle latitudes. 12 Oral presentation by Witte models and observations?

Negative trends in ozonesonde and models (Wargan, 2018) GMD Review and GMAC 2018 13