Geophysical Fluid Dynamics Laboratory

Key drivers of Western U.S. surface O3 variability during 1980-2050: From extreme transport events to background trends

Meiyun Lin

Presented at 2014 AGU Fall Meeting, A14E-01

Acknowledgements: NOAA: L.W. Horowitz, O.R. Cooper, A.O. Langford, S. J. OltmansOthers: A.M. Fiore and H. Rieder (Columbia), D. Tarasick (Environment Canada)

Mae Gustin and Rebekka Fine (Univ of Nevada)

Geophysical Fluid Dynamics Laboratory

Major challenges for Western U.S. air quality managements

2

MAJOR CHALLENGES:-Frequency of stratospheric intrusions in spring [e.g. Langford et al., 2009; Lin et al., 2012b]

-Rising Asian emissions and global CH4 [e.g. Jacob et al. 1999; Cooper et al. 2010; Fiore et al. 2012]

-More frequent wildfires in summer [e.g. Westerling et al. 2006; Pfister et al. 2008; Jaffe 2011]

-High winter O3 pollution in an oil and natural gas basin [e.g. Edwards et al. 2014]

NEED PROCESS-LEVEL UNDERSTANDING ON DAILY TO MULTI-DECADAL TIME SCALES

Annual 4th Hi MDA8 O3

(2011-2013 AQS)Mean background, Apr-Jun

(NA anthrop emis off in GFDL-AM3 model)

Future NAAQS?

Current NAAQS

Sparse monitoring

Geophysical Fluid Dynamics Laboratory Lin M.Y. et al (in prep, 2014c)

Stratospheric intrusions push observed surface ozone in S. Nevada above the 75 ppbv NAAQS threshold

New Measurements: -Nevada Rural Ozone Initiative (NVROI), July 2011-present (M. Gustin & R. Fine)-Las Vegas Ozone Study, May-June 2013 (A.O. Langford)

May 25 (STT)

How do such events vary from year to year and what controls its variability?

Observed MDA8 O3 on May 25, 2013

ppb

3

Jun 2 (LA/BB)

Consistent with modeling results from Lin et al (JGR, 2012b)

NVROI Echo Peak (2286 m altitude)

Geophysical Fluid Dynamics Laboratory

Strong stratospheric influence on year-to-year variability of high-elevation Western U.S. surface O3 during April-May

Lin M.Y. et al (in review, 2014b)

75th

25th

50th

O3Strat

1990

Emissions held constantNudged to “real” winds

4

In contrast, the influences from wildfire emissions and Asian pollution are minimal (See paper) Large IAV may complicate the attribution of observed O3 trends in short records (See also Lin

et al.; Nature Geosci, 2014a)

Geophysical Fluid Dynamics Laboratory

The high tail of the observed daily surface O3 distribution over Western U.S. increases during La Niña springs

Observed daily max 8-h average (MDA8) ozone

Neutral, µ = 56.5,σ = 7.5El Niño, µ = 56.9,σ = 7.3La Niña, µ = 58.5,σ = 7.9Pinatubo, µ = 54.3,σ = 6.5

5Lin M.Y. et al (in review, 2014b)

See my talk A51Q-07 on Friday (9:30am, Moscone West 3006) for explanations on the physical mechanisms

Changes in the high tail are statistically significant

Geophysical Fluid Dynamics Laboratory 6

1999 (La Niña)

1992 (Pinatubo)

Frequent deep STT events

Weaker events, lower mean values

Gothic in the Colorado Rocky Mtn (2.9 km altitude)

NAAQS

O3Strat

BGO3

Total

Following La Niña conditions, deep STT may occur with sufficient frequency as to confound NAAQS attainment

NAAQS

BGO3

Total

MD

A8

O3

(pp

b)

MD

A8

O3

(pp

b)

O3Strat

Geophysical Fluid Dynamics Laboratory 7

Following La Niña conditions, deep STT may occur with sufficient frequency as to confound NAAQS attainment

2011 (La Niña)

2007 (Neutral)

Chiricahua NM in Arizona (1.5 km altitude)

NAAQS

BGO3

Total

NAAQS

BGO3

Total

MD

A8

O3

(pp

b)

MD

A8

O3

(pp

b)

O3Strat

O3Strat

Geophysical Fluid Dynamics Laboratory 8

More frequent stratospheric intrusions expected in the following spring over WUS?

Developing seasonal predictions with a few months of lead time to aid Western U.S. AQ planning?

SST (C)

Lin M.Y. et al (in review, 2014b)

The time lag could allow regional preparations, e.g. Conducting daily forecasts for public health alerts Deploying targeted measurements aimed at identifying “exceptional events”

MD

A8

O3

(pp

b)

Gothic, Colorado

Tropical SST cooling typically peaks in winter

Geophysical Fluid Dynamics Laboratory 9

GFDL-AM3 BASE(emis+met vary over time)

Long-term U.S. surface ozone trends: Did regional NOx reductions work?

Observed

MDA8 O3 trends (ppb yr-1)

95%

50%

Larger circles indicate statistically significant trends

MAM 1988-2012

NASA Project PIs (M.Y. Lin and O.R. Cooper)The model is nudged to “real” winds, sampled at site elevation and filtered to remove local influence -- more representative of observed conditions at remote WUS sites

Geophysical Fluid Dynamics Laboratory

Must consider variability in atmospheric circulation when interpreting observed changes in baseline O3

NASA PIs (M.Y. Lin and O.R. Cooper) 10

1988 1992 1996 2000 2004 2008 2012

Me

dia

n O

3an

om

aly

in

pp

b r

ela

tive

to

20

00

Pinatubo STT+

Weaker variability in the 2000s attributed to changes in background rather than in N.A. pollution

BGO3

Nudged to real winds

Larger growth of Asian emissions in the 2000s, but weaker rate of O3

changes observed at Lassen as compared to the 1990s

Lassen Volcanic NP, Calif. (1.7 km), Mar-Apr-May

E. China NOx Emissions

No

rma

lize

d t

o 2

00

0

Model Satellites

1980 1990 2000 2010

Geophysical Fluid Dynamics Laboratory

Projections of near-term changes in WUS lower trop. ozone in spring (March-April-May)

11

Me

dia

n O

3an

om

aly

in

pp

b r

ela

tive

to

20

10

No

rmal

ize

d t

o 2

00

0

Satellites

RCP8.5

RCP4.5

RCP8.5

RCP4.5

Global CH4 (ppb)3000

2500

2000

1500

E. China NOx emissions

1980

OBS (GRB, LAV, ROM)AM3 BASE, nudged

11

Geophysical Fluid Dynamics Laboratory

Summarizing drivers of WUS O3 variability

• The frequency of deep STT affects year-to-year variability in springtime high-O3 events Linked to La Niña in the previous winter Potential for seasonal forecast

• Rising Asian emissions and global CH4 are likely the major drivers of long-term changes in mean baseline ozone, but ….

• Ozone measurements contain signatures of circulation variability: Short records may complicate the attribution of trends Must ensure an apple-to-apple comparison btw model and obs

Contact: Meiyun.Lin@noaa.gov

12

Geophysical Fluid Dynamics Laboratory

Additional slides for Q & A

Geophysical Fluid Dynamics Laboratory

The GFDL AM3 Model Hindcasts for 1978- 2012

Experiment Meteorology Forcings(radiation)

CH4

(chemistry)Aerosol and O3

precursorsFire emissions

BASE Nudged to NCEP U&V

Historical Historical Historical and RCP 8.5 beyond 2005

Historical

PRB as BASE Historical Historical Shut off over NA;

as BASE elsewhere

Historical

FIXEMIS as BASE Historical 2000 Fixed Fixed

IAVFIRE as BASE Historical 2000 Fixed Historical

IAVASIA as BASE Historical 1980 Varying in Asia only Fixed

IAV_CH4Chem

as BASE Historical Historical Fixed Fixed

Designed to isolate the role of changes in meteorology, background, STE, wildfires, methane, and Asian vs. North Americananthropogenic emissions

Model setup similar to Lin et al. [2012a, b], but at 2ºx2º resolution

Geophysical Fluid Dynamics Laboratory 15

Summarizing drivers of U.S. ozone trends In the historical period (1980-2012)

Increases in total ozone over WUS despite reductions in local emissions

GFDL AM3 (2003-2012 minus 1981-1990)

PRB

AsianIncreases from boreal fires

Total

CH4

PRB

PRBEA

Total (filtered)

Rising Asian emissions contribute ~50% of total background increases

Geophysical Fluid Dynamics Laboratory

Western US surface O3 variability correlates poorly with O3

burdens in the UTLS but strongly with that in the Free Trop

Niño 3.4 Index

Apr-May 1995

Apr-May 2000

Apr-May 2005

Apr-May 2010

12-mon running mean UTLS O3

anomaly (%) r2(UTLS, Surface) = 0.07

98/99 07/08 10/11

Jan 1990

Jan 1995

Jan 2000

Jan 2005

Jan 2010

Jan 1990

Jan 1995

Jan 2000

Jan 2005

Jan 2010

Trinidad

-20

0

20

40

10

5

0

-5

Apr-May mean FreeTrop & SurfaceO3 anomaly (ppb)r2(FreeTrop, Surface) =0.74

Apr-May 1990