guiding ozone layer recovery with effective science and

Steve MontzkaNOAA Global Monitoring Division, Boulder, USA

Guiding ozone layer recovery with effective science and policy on an international scale.

Thanks to:Colleagues at NOAA and CIRES.

All co-authors on associated papersMontzka et al., 2018, https://doi.org/10.1038/s41586-018-0106-2Rigby et al., 2019, https://doi.org/10.1038/s41586-019-1193-4Ray et al., 2020, https://doi.org/10.1038/s41561-019-0507-3

Station personnel and cooperative institutes around the world helping with flask sampling and instrument maintenance

AGAGE community of scientists

Sunrise at the Mauna Loa Baseline Observatory

The upper atmosphere ozone layer filters out high energy UV radiation from the Sun:

Enhanced UV-B radiation from the sun à raises the risk of:

skin cancercataractsimmune system suppression

à it also damages plants, crops, and aquatic ecosystems

Figures: 20 Questions and answers about the ozone layer: 2020 update

Depletion of the ozone layer:

à First hypothesized in 1974 (Rowland & Molina)…but only a few years later dramatic springtime

ozone depletion was observed over Antarctica! (Farman et al., 1985)

Ozo

ne To

tal C

olum

n (D

U)

Nature, 249, 810, 1974

…but was not expected to be a significant problem for many years

The 1987 Montreal Protocol* Designed to heal the stratospheric ozone layer (limit UV exposure).* Mandated a global phase-out of production & trade of ozone-depleting gases.* The first universally ratified treaty in UN history.* Parties have held 2 meetings/yr since the late 1980s to ensure continued success.

Step-wise phase-out of ozone-depleting gases:Of chemicals (CFCs) used in refrigeration & air

conditioning, foam-blowing, as solvents, in fire-extinguishing, as pesticide

1st generation substitutes:HCFCs (hydrochlorofluorocarbons)

2nd generations substitutes:HFCs (hydrofluorocarbons)

A view of the Meeting of the Parties, Rome, Nov., 2019

Full implementation to yield:avoidance of more than 280 million cases of skin cancer and more than 45 million cases of cataracts in the United States by the end of the century, with even greater benefits worldwide.

0200400600800

100012001400160018002000

1985 1990 1995 2000 2005 2010 2015

all ODSsCFCs

Adherence to the Montreal Protocol has led to:1) Near elimination of ozone-depleting substance (ODS)

production, particularly chlorofluorocarbons (CFCs)

The Montreal Protocol “is perhaps the single most successful international agreement to date” (K. Annan, 2003)

Glob

al re

port

ed

prod

uctio

n*

(ODP

-kt/

yr)

CFC production reportedly

phased-out in 2010

Minimal production

reported by 2007

* Reported to the WMO/UNEP Ozone Secretariat

???

Ozone over South Pole during October of 2018 (red line)

(ozonesonde flights by NOAA-GMD & CIRES):

Nature, 249, 810, 1974

1991-2012 ave

Despite a warning over 45 years ago…

…and the universal ratification of the Montreal Protocol on Substances that Deplete the Ozone Layer in 1987

The stratospheric ozone layer is far from recovered:

…the good news is that there are signs of initial recovery in the mid-latitude stratosphere

•

•

ALT

BRW

KUM

SPO

NWR

MLO•

CGO

SMO

weeklyhourly•

••••

•

•

PSA

MHD

THDLEF

HFM

SUM

•••

••

•

•

•

•

••

•

NOAA/CIRES: Tracking ozone-depleting gas concentrations

•WIS •AMY

(just added)

~40 gases measured including:- ozone-depleting gases controlled by the Montreal Protocol- hydrofluorocarbons (HFCs)- other methyl halides- other poly-chlorinated & brominated hydrocarbons- benzene & other hydrocarbons- carbonyl sulfide

Assessing the success of the Montreal Protocol,and scientific insights into:

- global atmospheric oxidation variability- carbon uptake by the biosphere- non-CO2 GHG contributions to radiative forcing (climate change)- climate sensitivities of natural processes- atmospheric transport processes

Adherence to the Montreal Protocol has also led to:à declines in ODS emissions and concentrations:

à Pre-ozone-hole concentrations of ODSs are projected for later in this century (WMO reports),

provided continued adherence to the Protocol’s controls

3.2

3.3

3.4

3.5

3.6

3.7

3.8

1990 1995 2000 2005 2010 2015

Atm

osph

eric

Cl (

ppb)

Ozone-depletingchlorine

year14

15

16

17

1990 1995 2000 2005 2010 2015At

mos

pher

ic B

r (pp

t) Ozone-depletingbromine

NOAA

From NOAA/GMD & CIRES - measured tropospheric concentrations of CFCs, HCFCs, chlorinated solvents, methyl halides, halons…

(see updates at http://www.esrl.noaa.gov/gmd/odgi/)

Why will recovery take so long??à The recovery time-frame is set by natural removal rates of CFCs

and other ODSs from the atmosphere:

Chemical atmospheric lifetime (year) destruction per yearMain ozone-depleting gases:CFC-11 50 2%CFC-12 100 1%CFC-113 85 1.2%

Substitute chemicals:HCFC-22 12 9%

HFC-134a 14 7%

These rates are set by the flux of high-energy light through the stratosphere and the concentrations of atmospheric oxidants

220225230235240245250255260265

2000 2005 2010 2015 2020120

140

160

180

200

220

240

260

2000 2005 2010 2015 2020

1

10

100

2000 2005 2010 2015 20200

20

40

60

80

100

120

2000 2005 2010 2015 2020

Atmospheric measurements for tracking the effectiveness of the Montreal Protocol

CFC-11(CCl3F)

CH3CCl3 HFC-134a(CH2FCF3)

HCFC-22(CHClF2)

ln(p

pt)

ppt

ppt

ppt

(log scale)

Northern Hemisphere

SouthernHemisphere

NH

SH

NH

SH

NHSH

NOAA/CIRES flask data

CFCs, CH3CCl3, etc. à HCFCs à HFCs à HFOs

Chemical lifetimes range from 5 to 50 years

Atmospheric CFC-11

NHSH

1 23

Mol

e fra

ctio

n (p

pt)

Glo

bal r

ate

(per

yr)

Diff

eren

ce

(N –

S; p

pt)

1995 2000 2005 2010 2015

NOAA/GMDIn situ & flasks

CFC-11:à Second most abundant ODS;

accounts for 20-25% ofozone-depleting chlorine

à Was the largest contributor to the overall decline of atmospheric Cl from 2007-2012

à While production was reportedly phased out in 2010, significant emissions persist,primarily from a foam “bank” of ~900 kt currently

After a production phase-out: * concentration decline should have accelerated

as bank becomes depleted by emissions…(the zero-emission limit is -2%/yr given its 50 yr lifetime)

Con

cent

ratio

n (h

emis

pher

ic m

ean

ppt)

Reported production phase-out

NHSH

1 23

Glo

bal r

ate

(per

yr)

Diff

eren

ce

(N –

S; p

pt)

1995 2000 2005 2010 2015

1995 2000 2005 2010 2015

12

3

Atmospheric CFC-11


Expectation

Hemispheric mean concentration

Con

cent

ratio

n (h

emis

pher

ic m

ean

ppt)



Global rate of change

NHSH

1 23

Mol

e fra

ctio

n (p

pt)

Glo

bal r

ate

(per

yr)

Diff

eren

ce

(N –

S; p

pt) 1995 2000 2005 2010 2015

12

3

Atmospheric CFC-11


Expectation

Expectation

Con

cent

ratio

n (h

emis

pher

ic m

ean

ppt)


NHSH

Hemispheric concentrationdifference

1 23

Mol

e fra

ctio

n (p

pt)

Glo

bal r

ate

(per

yr)

Diff

eren

ce

(N –

S; p

pt)

12

3

Atmospheric CFC-11

Expectation


Expectation


Global rate of change

àQualitatively consistent with an increase in NH emissions

Con

cent

ratio

n (h

emis

pher

ic m

ean

ppt)


Why the unexpected slowdown?Consider the evidence:Is it an instrument or measurement problem?

à NO, slowdown observed by independent measurements (NOAA and AGAGE) using multiple techniques, and in different aspects of the data

Is it unique to certain sites??à NO, although initially noted in the NH, and then later in the SH

223

225

227

229

231

233

235

237

239

241

243

245

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

CFC

-11

(ppt

)

Year

spo

psa

cgo

smo

mlo

kum

nwr

brw

alt

sum

thd

mhd

NH fit

SH fit

NOAA flasks analyzed by GCMS

224225226227228229230231232

2015.0 2016.0 2017.0 2018.0

ppt C

FC-1

1

Year

Deriving emissions from concentrationsGlobal CFC-11 concentration changes reflect the balance of

a) EMISSIONb) REMOVAL (photolysis in the stratosphere)

224225226227228229230231232

2015.0 2016.0 2017.0 2018.0

ppt C

FC-1

1

Year

-2%

-2%

-2%

MEASURED

E1

E2

E3

For a ~ 50 yrlifetime, atmospheric concentrations should decline at ~1/50 yr-1 or2% yr-1

in the absence of emissions.

0

10

20

30

40

50

60

70

80

90

100

Emis

sion

or P

rodu

ctio

n (G

g/yr

)

adGF11/dt = Emission - LossdGF11/dt = Emission – G/tF11

ReportedProduction

1995 2000 2005 2010 2015

Mass balance considerations (a simple box-model analysis):

Measured

Emis

sion

or P

rodu

ctio

n (G

g/yr

)

Emission

13 ± 5 Gg/yr(25%) increase

(2014-2016 vs. 2002-2012; 2017 is similarly elevated)

Are CFC-11 emissions reallyincreasing despite the Montreal Protocol and reported phase-out by 2010?Montzka et al., 2018

Deriving emissions from concentrations

0

10

20

30

40

50

60

70

80

90

100

Emis

sion

or P

rodu

ctio

n (G

g/yr

)

a

ReportedProduction

1995 2000 2005 2010 2015

Mass balance considerations (a simple box-model analysis):Em

issi

on o

r Pro

duct

ion

(Gg/

yr)

Emission

13 ± 5 Gg/yr(25%) increase

(2014-2016 vs. 2002-2012; 2017 is similarly elevated)

Or has CFC-11 loss and transport changed (“tF11”)?

Measured

Montzka et al., 2018

dGF11/dt = Emission – G/tF11

3-D models suggest: maybe a bit...!

Deriving emissions from concentrations

Air reaching Hawaii in autumn can be influenced by Eurasian emissions (Lin et al., 2014, DOI: 10.1038/NGEO2066).A close look at our results from the Mauna Loa station showed:

Before 2012: pollution plumes reaching Hawaii DID NOT contain elevated concentrations of CFC-11

After 2012: pollution plumes reaching Hawaii NOW CONTAIN elevated concentrations of CFC-11.

Do the observations themselves suggest emissions increasing?

Pollution plume events reaching Hawaii were identified by:

230

235

240

245

250

255

232

233

234

235

236

237

238

0.4 0.6 0.8 1

HC

FC-22 (ppt)C

FC-1

1 (p

pt)

Fraction of year

2016200

205

210

215

220

225

240

241

242

243

244

245

246

HC

FC-22 (ppt)C

FC-1

1 (p

pt)

2010

CFC

-11

(ppt

) HC

FC-22 (ppt)

HC

FC-22 (ppt)

CFC

-11

(ppt

)

HCFC-22CFC-11

1) Increased concentrations of a number of gases:HCFC-22, CH2Cl2, and CO


230

235

240

245

250

255

232

233

234

235

236

237

238

0.4 0.6 0.8 1

HC

FC-22 (ppt)C

FC-1

1 (p

pt)

Fraction of year

2016

230

235

240

245

250

255

232

233

234

235

236

237

238

0.4 0.6 0.8 1

HC

FC-22 (ppt)C

FC-1

1 (p

pt)

Fraction of year

2016200

205

210

215

220

225

240

241

242

243

244

245

246

HC

FC-22 (ppt)C

FC-1

1 (p

pt)

2010200

205

210

215

220

225

240

241

242

243

244

245

246

HC

FC-22 (ppt)C

FC-1

1 (p

pt)

2010

CFC

-11

(ppt

) HC

FC-22 (ppt)

HC

FC-22 (ppt)

CFC

-11

(ppt

)

HCFC-22CFC-11

Only after 2012 do pollution plumes contain elevated CFC-11 concentrations

Correlations among HCFC-22, CH2Cl2, & CO are strong in all years

0.0

0.2

0.4

0.6

0.8

1.0

2008 2010 2012 2014 2016 2018re

gres

sion

coe

f. r2̂

CFC-11 vs.HCFC-22

Pollution plume events reaching Hawaii were identified by:

1) Increased concentrations of a number of gases:HCFC-22, CH2Cl2, and CO


The source region for those pollution plumes reaching Hawaii were identified by meteorological models of air transport:

2013-2016 2013-2016

Lower [HCFC-22] & [CFC-11] Higher [HCFC-22] & [CFC-11]

Darker colors indicate the surface region influencing samples collected in Hawaii

After 2012, air associated with eastern Asia contains relatively higher concentrations of CFC-11

à Some portion of the global emission increase is from Eastern Asia


Rigby et al., 2019

In a second study (Rigby et al., 2019), measurements within eastern Asia are analyzed… AGAGE & affiliates plus NOAA sampling locations

(Rigby et al., 2019)

GSNHAT

MLO

7Between 12 and 34 measurements per day at AGAGE & affiliate sites

From Rigby et al., 2019:

AGAGE, R.O.K. (K.N.U.), and Japan (NIES)

9

Grey = 40°S (C.Grim, CSIRO)

After 2012, CFC-11 concentrations are increasingly elevated in pollution plumes reaching these sites, particularly at GSN

CFC

-11

conc

entra

tion

(ppt

)

Discrete air sample results:

40°S

40°S

Rigby et al., 2019

Red = regions where emissions increased after 2012>>

About 50% of the global emission increase comes from eastern Asia, specifically the Chinese provinces

of Shandong and Hebei.Where’s the other half???

AGAGE dataRigby et al., 2019, https://doi.org/10.1038/s41586-019-1193-4

Inverse Lagrangian model analyses of these measurements in Rigby et al., 2019:

We don’t know!Similar inverse analyses of observations from the U.S., Europe, and Australia suggest low emissions (< 10 Gg/yr) and no recent increases.

Regional estimates from many parts of the world are not possible.

Rigby et al., 2019

0

10

20

30

40

50

60

70

80

90

100

Emis

sion

or P

rodu

ctio

n (G

g/yr

) Emission

ReportedProductionEm

issio

n or

Pro

duct

ion

(Gg/

yr)

1995 2000 2005 2010 2015

Do the results imply renewed production?à is the Montreal Protocol being violated?

OR: is CFC-11 escaping more rapidly from the foam “bank” (~1200 Gg in 2012)?

0

10

20

30

40

50

60

70

80

90

100

Emis

sion

or P

rodu

ctio

n (G

g/yr

)

2%

3%

4%

5%

6%

7%

1995 2000 2005 2010 2015Impl

ied

annu

al re

leas

e fra

ctio

n fro

m b

ank

(%/y

r)

Year

With no new production, the escape rate from the ‘bank’ would have had to nearly double since the early 2000s…

à this seems highly unlikely

0

10

20

30

40

50

60

70

80

90

100

Emis

sion

or P

rodu

ctio

n (G

g/yr

) Emission

ReportedProduction

Impl

ied

annu

al

rele

ase

from

ba

nk (%

/yr)

Emiss

ion

or P

rodu

ctio

n (G

g/yr

) OR: is CFC-11 escaping more rapidly from the foam “bank” (~1200 Gg in 2012)?

Do the results imply renewed production?à is the Montreal Protocol being violated?


Implications for the ozone layer:If the problem is rapidly fixed, the impact on the ozone layer

should be small.

If these emissions continue à ozone recovery delay of 7 to 20 yr.

Unlikely? Perhaps, but the magnitude of the problem ultimately

depends on the magnitude of unreported production.

Have we detected only the “tip of the iceberg”??Most likely use of the new production:

Polyurethane insulating foams (WMO-TEAP; 2019)à if so, then, emissions detected are << associated production.

and enhanced emissions could be sustained even with effective mitigation.

25-30 Gg/yr emissions for foams implies à 50-100 Gg/yr of CFC-11 production (peak prod. was 400 Gg/yr)

(WMO-SAP, 2018; Dhomse et al., 2019 & others)

Global reaction to our results:On-the-ground investigative reporting by the NY Times and EIAà evidence for ongoing use and production of CFC-11 in Chinaà Overall quantities unknown…

Ozone-Harming Gas Is Detected, Despite BanBy Henry Fountain, NY Timesp. A4, May 16, 2018

An Environmental Win Falters. Why? Some Clues Point to ChinaBy Chris Buckley and Henry Fountain, NY Times p. A1, June 24 2018

More Evidence Points to China as Source of Ozone-Depleting GasBy Chris Buckley and Henry Fountain, NY Times p. A14, November 3, 2018

…and the EIA

The EIA suggests that in 2018 mostof the closed-cell foam industry in China had reverted back to using

CFC-11But this was from a very limited

survey of users…

The Parties to the Montreal Protocol consider the findings in their annual meetings in 2018 and 2019:

Science Assessment Panel members (+ Dave F.)

Ecuadorian President Mr. Lenín Moreno and meeting co-chairs open the meeting

Presenting the science to the Parties in side events

Remaining questions:What is being done about the issue?The Parties to the Montreal Protocol are:

* taking the issue seriously (recent meetings dominated by this issue)* looking for more information (meetings held, reports requested)

- what will be the ozone layer impact? Are emissions increasing outside of NE China? What led to the issue? How is the CFC-11 being produced?

…to facilitate rapid mitigation; and maintain integrity of the Protocol

They are also:* reconsidering $$ being transferred to China via the Multi-lateral Fund

(the Protocol’s fund for assisting developing countries with compliance; $1.3 billion has been dispersed to China via the MLF since the 1980s…)

- China’s most recent funding allotments were reduced

* considering punitive action (remedial too?)

* considering amendments to the Protocol to prevent future violations - for ODSs, and wrt the HCFCs phase-out and HFC phase-down(similar if not larger challenges to monitor adherence to these

control schedules seem likely…)

China’s response:Has repeatedly indicated a commitment to work with other countries, scientists, and industry experts to locate and eliminate the true source of the increased CFC-11 emissions.

Has conducted inspections of 1,172 enterprises across the country, and found instances of illegal production and use (and prosecuted those responsible), but amounts were small (29.9 tonnes and 10 enterprises)

à not enough to account for atmosphere-based results Has formulated a plan of action (with international input and before the Rigby et al paper fingering China in particular) that involves:

- Developing national-scale atmospheric monitoring of ODSs and HFCs- Building 6 testing laboratories (will help with enforcement)- Strengthening inspections and enforcement (ecology and environment

bureaus, but difficult as businesses are many and officials are few)- Better tracking of CCl4 (source material)- New programs to enhance enforcement- Facilitating reporting of non-compliance by industry (e.g., competitors)- Extending penalties to end-users for non-compliance

225

230

235

240

2010 2012 2014 2016 2018 2020

225

235

245

255

265

1995 2000 2005 2010 2015 2020CFC-

11 c

once

ntra

tion

(ppt

)

Year

Updated results

à These observations of declining concentrations suggest that global CFC-11 emissions may be on the decline

Updated measurements (post 2017) hint at changes in three respects:

1) an accelerating global concentration decline2) a decreasing N-S hemispheric concentration difference3) decreased concentrations in pollution plumes reaching Hawaii

Preliminary NOAA results

225

230

235

240

2010 2012 2014 2016 2018 2020225

230

235

240

2010 2012 2014 2016 2018 2020

225

235

245

255

265

1995 2000 2005 2010 2015 2020CFC-

11 c

once

ntra

tion

(ppt

)

Year

Updated results

Updated measurements (post 2017) hint at changes in three respects:

1) an accelerating global concentration decline2) a decreasing N-S hemispheric concentration difference3) decreased concentrations in pollution plumes reaching Hawaii

Preliminary NOAA results

à These observations of declining concentrations suggest that global CFC-11 emissions may be on the decline

Key features of the Montreal Protocol enabling it’s success:It has repeatedly been revisited, adjusted, and amended by Parties.

Revisions are based on current knowledge that is assessed in quadrennial reports by three panels (Science, Technology and Economics, Environmental Impacts) drawn from worldwide experts.

Initial agreements would not have saved the ozone layer…

Only with subsequent revisions was ozone layer recovery a possible outcome

Key features of the Montreal Protocol enabling it’s success:It has repeatedly been revisited, adjusted, and amended by Parties.

Revisions are based on current knowledge that is assessed in quadrennial reports by three panels (Science, Technology and Economics, Environmental Impacts) drawn from worldwide experts.

Decisions are typically made by unanimous consent—everyone has to agree in order for decisions to move forward…

A 10-year grace period was allowed for developing countries for meeting the phase-out schedules.

Compliance by developing countries is assisted by a Multi-Lateral Fund for the Implementation of the Montreal Protocol, which is maintained by developed countries (>$3 billion USD distributed between 1995 and 2005). Management of the fund is by an executive committee with 14 member states (7 developed, 7 developing).

Concluding remarks:* The Montreal Protocol represents a unique situation in which scientists and policymakers are working constructively to enable ozone layer recovery

* The information feedback loop (quadrennial reports) has led to strengthening of controls, thereby improving chances for success

* Scientists have input to Parties without interference via assessed scientific understanding communicated in quadrennial reports

à providing “policy-relevant but policy-neutral information”* Leverage on compliance by Parties enabled by multiple mechanisms

à particularly MLF funding of projects

State-of-the-art atmospheric monitoring and modeling enabled:à the early detection of a significant violation of the Montreal Protocolà an indication of the location of the renewed source of CFC-11

Many questions remain for CFC-11, but early signs suggest the path to rapid and lasting mitigation remains unclear

Considering the future:is our atmospheric “vigilance” capability up to the task (obs., models)?

à HCFC phase-out, HFC phase down… greenhouse gases??

Thanks for your attentionThe Science Assessment Panel (+1) at MOP30

guiding ozone layer recovery with effective science and

Documents