Multi-model ensemble simulations of Air pollution and ... · • MOZ2-GFDL • MOZART4 • MOZECH • MOZECH2 • p-TOMCAT • STOCHEM-HadAM3 • STOCHEM-HadGEM • TM4 • TM5 •

JRC- Brussels- PF

JRC Brussels JRC Brussels 1Unit_26012006

JRC Brussels 1

Multi-model ensemble simulations of Air pollution and Climate Change in

2030

Frank Dentener, David Stevenson, Drew Shindell, Maarten Krol,Rita van Dingenen, Kjerstin Ellingsen, Twan van Noije

and ca. 30 persons contributors

JRC- Brussels- PF


JRC Brussels 2

Background

• ‘OxComp’ IPCC TAR model intercomparison sampled models in ~1999

• OxComp focussed on SRES A2 in 2100.• New scenarios from IIASA include AQ legislation measures

(not in SRES)• Models and emissions have developed in the last 5 years • New sets of biomass burning and ship emissions• Preparation of IPCC-AR4, but also useful for e.g. TF HTAP

JRC- Brussels- PF


JRC Brussels 3

Emission scenarios:CLE: Current Legislation 2030MFR: Maximum Feasible Reduction 2030SRES A2: high emission Climate Change scenario:

IPCC92a (ca 1 K temperature increase) by 2030

•Focus on the year 2030; ‘the inter-mediate’ future which is of direct relevance to policy makers; compare to the year 2000

•Emphasis on the synergetic effect of air quality (human health and vegetation exposure) and climate (RF)

•26 models from 15 different model “families”; model resolution from 1°x1°to 2.5 °x2.5 ° to 5° x5°. Model statistics- and model response to emissions

•Currently 1 accepted paper; 4 submitted and 1 in prep.

•Baseline + 3 emission scenarios + climate change scenario

PhotoComp (2005)

JRC- Brussels- PF


JRC Brussels 4

26 Participating Models

• CHASER_CTM• CHASER_GCM• FRSGC/UCI• GEOS-CHEM• GISS• GMI/CCM3• GMI/DAO• GMI/GISS• IASB• LLNL-IMPACT• LMDz/INCA-CTM• LMDz/INCA-GCM• MATCH-MPIC/ECMWF

• MATCH-MPIC/NCEP • MOZ2-GFDL• MOZART4• MOZECH• MOZECH2• p-TOMCAT• STOCHEM-HadAM3• STOCHEM-HadGEM• TM4• TM5• UIO_CTM2• ULAQ• UM_CAM

CTMs driven by analysesCTMs driven by GCM output

CTMs coupled to GCMs

JRC- Brussels- PF


JRC Brussels 5

Global NOx emission scenarios

0.0

40.0

80.0

120.0

160.0

200.0

1990 2000 2010 2020 2030

Europe North AmericaAsia + Oceania Latin AmericaAfrica + Middle East Maximum Feasible Reduction (MFR)SRES A2 - World Total SRES B2 - World Total

Figure 1. Projected development of IIASA anthropogenic NOx emissions by SRES world region (Tg NO2 yr-1).

CLE

SRES A2

MFR

2000 2030

JRC- Brussels- PF


JRC Brussels 6

Analysis of oxidation capacity, surface ozone, deposition, satellite observations of NO2 and CO

Ozone and oxidation capacity:

•3D monthly mean fields for O3, CO, CH4, NO, NO2, and OH•Daily average tropospheric column ozone •3D monthly budgets of ozone production and destruction,

surface deposition, stratospheric O3 influx•Hourly surface ozone [ppbv] •3D monthly mean fields of the CH4+OH destruction (lifetime)=>OH

Ecosystem inputs: Oxidized and reduced nitrogen, and sulfur depositionBiodiversity, Eutrophication, Acidification

Comparison with satellite date:NO2 and CH2O column (molec/cm2). GOME;10:30 Local TimeCO (MOPITT of CO columns)

JRC- Brussels- PF


JRC Brussels 7

Year 2000:tropospheric O3

JRC- Brussels- PF


JRC Brussels 8

Year 2000Ensemble meanof 25 models

AnnualZonalMean

Annual TroposphericColumn

D. Stevenson, JGR, in press

JRC- Brussels- PF


JRC Brussels 9

Year 2000Inter-modelstandard deviation (%)

AnnualZonalMean

Annual TroposphericColumn


JRC- Brussels- PF


JRC Brussels 10Ensemble mean model closely resembles ozone-sonde measurements

UT: 250 hPa

MT: 500 hPa

LT: 750 hPa

J F M A M J J A S O N D

Sonde± 1SD

Model± 1SD

90-30S 30S-EQ EQ-30N 30-90N

Sonde data: Logan (1999) + SHADOZ (Thompson et al, 2003)

JRC- Brussels- PF


JRC Brussels 11

Annual Zonal Mean∆O3 / ppbv

Annual Tropo-sphericColumn∆O3 / DU

‘Likely’IIASA CLE

SRES B2 economy +Current AQ Legislation

‘Optimistic’IIASA MFR

SRES B2 economy +Maximum Feasible

Reductions

‘Pessimistic’IPCC SRES A2

High economic growth +Little AQ legislation

Multi-model ensemble mean change intropospheric O3 2000-2030 under 3 scenarios


JRC- Brussels- PF


JRC Brussels 12

O3 in 2030, radiative forcing

& influence of climate change

JRC- Brussels- PF


JRC Brussels 13

Radiative forcing implications

-500

0

500

1000

1500m

W /

m2

CO2 795 795 1035

CH4 116 0 141

O3 63 -43 155

CLE MRF A2

Forcings (mW m-2) 2000-2030 for the 3 scenarios:

-23% +37%

CO2

CH4

O3


JRC- Brussels- PF


JRC Brussels 14

O3 air quality

JRC- Brussels- PF


JRC Brussels 15

Monthly average Surface ozonecomparison with measurements

Max models+ 1 sd-1 sd-Min models

S.W. US

Brazil

N. India; NepalC. EuropeC. Mediterranean

Middle East S. India S.E. Asia

C.E. Africa S. Africa

S.E. US

Great Lakes N. China

Ellingsen et al, in prepR. Van Dingenen

JRC- Brussels- PF


JRC Brussels 16

y = 1.669E-02x - 1.474E+01R2 = 0.95

y = 0.002x - 0.8641R2 = 0.55

0

20

40

60

80

100

120

0 1000 2000 3000 4000 5000 6000 7000 8000

SOMO35 (ppb.days)

exce

eden

ce d

ays

EEAUSEPALimit level EEAlimit level USEPALinear (EEA)Linear (USEPA)

EC 25 days should not be exceeded

US EPA 3 days should not be exceeded

Statistical Relationship betweenSOMO35, EC, and EPA standards

Ellingsen et al, in prep.R. Van Dingenen

JRC- Brussels- PF


JRC Brussels 17

Increase in exceedances:2030 CLE versus 2000S2 - S1

-10000

1000200030004000500060007000

ppb.

days

-50050100150200250300350

days

of e

xcee

denc

e

SOMO35 756 896 372 169 823 1018 234 341 3726 463 3453 743 479 50

EEA 28 27 6 2 12 15 -2 -4 106 12 89 20 11 0

USEPA *5 20 35 10 0 15 5 -15 0 210 0 180 20 20 10

S.W. USA

S.E. USA Great Lakes

Latin America/

Central-East

Southern Africa

Central Mediterra

Central Europe

North India

Middle East

Southern India

S.E. Asia Northern China

Australia

Ellingsen et al, in prep.R. Van Dingenen

JRC- Brussels- PF


JRC Brussels 18

Oxidation capacity, CO and OH

JRC- Brussels- PF


JRC Brussels 19

O3

chem

ical

loss

/ Tg

(O3)

/yr

O3 budget and CH4 lifetime

IPCC TAR8.4 years

CH4 lifetime / years

Results for asingle model,several scenarios

Colours signifydifferent models

Ensemble mean model (offset)

Models with longerτCH4 have lower

O3 destruction rates:O(1D) + H2O → 2OH

Climate changereduces τCH4

Emissions haveminor influence

on τCH4

What causes the inter-model differences?Water vapour?Lightning NOx?Photolysis schemes?


JRC- Brussels- PF


JRC Brussels 20

CO comparison with MOPITT

55

60

65

70

75

80

85

90

95

100

105

1 2 3 4 5 6 7 8 9 10 11 12

month

50

60

70

80

90

100

110

120

1 2 3 4 5 6 7 8 9 10 11 12

month

55

65

75

85

95

105

115

1 2 3 4 5 6 7 8 9 10 11 12

month

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10 11 12

month

GLOBALNH

TROPICS SH

Shindell et al, JGR, submitted

JRC- Brussels- PF


JRC Brussels 21

OH zonal average [106 molec cm-3 ]

Krol et al., in prep.

JRC- Brussels- PF


JRC Brussels 22

NO2 column: retrievals and models

Ensemble model

Ensemble retrieval

Van Noije et al, submitted to ACPD

JRC- Brussels- PF


JRC Brussels 23Van Noije et al, submitted to ACPD

NO2 column: retrievals and models

Ensemble model

Max model diff

Max model diff

Max meas diff

Max meas diff

JRC- Brussels- PF


JRC Brussels 24

Nitrogen deposition

JRC- Brussels- PF


JRC Brussels 25

HNO3 wet deposition: models and measurements

Dentener et al., submitted GBC, 2005

NADPEMEP

IDAF

EAnet

Kulshrestha

Various

Galloway

JRC- Brussels- PF


JRC Brussels 26

Fraction Nr deposition in excess of 1 g/m2/yearon natural ecosystems

0102030405060708090

100

CA

NA

DA

US

A

CE

NTR

AL A

ME

RIC

A

SO

UTH

AM

ER

ICA

NO

RTH

ER

N A

FRIC

A

WE

STE

RN

AFR

ICA

EA

STE

RN

AFR

ICA

SO

UTH

ER

N A

FRIC

A

OE

CD

EU

RO

PE

EA

STE

RN

EU

RO

PE

FOR

ME

R U

SS

R

MID

DLE

EA

ST

SO

UTH

AS

IA

EA

ST A

SIA

SO

UTH

EA

ST A

SIA

OC

EA

NIA

JAP

AN

NH

SH

WO

RLD

%

S1S2S3S4

2000

CLE

MFR

A2

Fraction of Reactive Nitrogen deposition in excess of 1 g/m2/year on non-agricultural vegetation

Dentener et al., submitted GBC, 2005

JRC- Brussels- PF


JRC Brussels 27

Final Remarks

• Relatively simple experiment; but rich in results.• Air quality and climate versus emissions: emissions dominate the signal.

Influence of climate change uncertain.• Quantitative assessment of 2030 scenarios provide clear options for

policymakers (radiative forcing and AQ)• In free troposphere: Ensemble mean model O3 closely resembles

observations. Surface: larger problems.• Inter-model standard deviations highlight where models differ the most• Global budgets reveal interesting and fundamental model differences• Ozone, climate, deposition are coupled problems: integrated view

presented.• Useful baseline for TF HTAP studies on ozone?

JRC- Brussels- PF


JRC Brussels 28

JRC- Brussels- PF


JRC Brussels 29

Impact of Climate Change on Ozone by 2030(ensemble of 9 models)

MeanMean - 1SD Mean + 1SD

Negative watervapour feedback

Positive stratospheric

influx feedback

Positive and negative feedbacks – no clear consensusD. Stevenson, JGR, in press

JRC- Brussels- PF


JRC Brussels 30

JRC- Brussels- PF


JRC Brussels 31

Geographical differences in NOydeposition over Asia

Mean CHASER

JRC TM5 STOCHEM

baseline 2000baseline 2000

JRC- Brussels- PF


JRC Brussels 32

JRC- Brussels- PF


JRC Brussels 33

Multi Model Annual Average O3 in 2000

2000 base-line2000 base-line

JRC- Brussels- PF


JRC Brussels 34

CLE2030 minus baseline 2000+Current legislationCLE2030 minus baseline 2000+Current legislation

•Indian subcontinenttransportenergy

•Reductions in Europeand USA offset by increase of background

JRC- Brussels- PF


JRC Brussels 35

MFR2030- baseline 2000:Optimistic TechnologyMFR2030- baseline 2000:Optimistic Technology

•Worldwide decrease by2 ppbv

•Regionally -8 ppbv

JRC- Brussels- PF


JRC Brussels 36

SRES A2 2030- baseline; Yearly increase in surface ozone“No control case”SRES A2 2030- baseline; Yearly increase in surface ozone“No control case”

•Global increase by 5 pbbv•Africa, Asia 10-15 ppbv

JRC- Brussels- PF


JRC Brussels 37

Effect of climate change on surface ozone in 2030

Documents

Multi-model ensemble simulations of Air pollution and ... · • MOZ2-GFDL • MOZART4 • MOZECH • MOZECH2 • p-TOMCAT • STOCHEM-HadAM3 • STOCHEM-HadGEM • TM4 • TM5 •