Hydrological controls on the tropospheric ozone greenhouse gas effect

Le (Elva) Kuai1, Kevin W. Bowman2, Helen Worden3, Robert L. Herman2, Susan S. Kulawik4

1

NASASounderScienceTeamMeeting13-16September,2016|Greenbelt

1. JIFRESSE/UCLA;2. JPL/Caltech;3. NCAR;4. BAERInstitute/NASAAmes;

2

Largeuncertaintiesintroposphericozoneradiative forcing

08:32 2

WallingtonTJetal.PNAS2010;107:E178-E179

IPCCAR5

• Theestimatedradiative forcing(RF)oftroposphericO3 rangewidelyfrom+0.2to+0.6Wm-2.

• Thisrangeiscomputedusingvarietiesofchemical-climatemodels.• 97%oftheO3 longwaveRFisduetotheozoneabsorptioninthe9.6 μmband[Rothman

etal.,1987].

3

InstantaneousRadiative Kernels(IRK):

ObjectivesandMotivations

• Attribute the TOA band flux change/bias due to dominant physical quantities.

ΔFTOA =∂FTOA∂O3

ΔO3 +∂FTOA∂Tsur

ΔTsur +∂FTOA∂Tatm

ΔTatm +∂FTOA∂H2O

ΔH2O +∂FTOA∂τ cloud

Δτ cloud + rs

9.6μmbandfluxchange

O3 Surfacetemperature

Atmos.temperature

Watervapor

Cloud residual

IRKO3(z) =

∂FTOA(q)∂O3(z)

CO2

H2O, N2O O3

AURATES(TroposphericEmissionSpectrometer)

L1BOzoneBandflux

4

InstantaneousRadiative Kernels(IRK):

ObjectivesandMotivations• Attribute the TOA flux change due to

dominant physical quantities.

• Understand the dependence of O3 IRK variation on H2O, temperature, and clouds.

IRKO3(z) =

∂FTOA(q)∂O3(z)

ΔFTOA =∂FTOA∂O3

ΔO3 +∂FTOA∂Tsur

ΔTsur +∂FTOA∂Tatm

ΔTatm +∂FTOA∂H2O

ΔH2O +∂FTOA∂τ cloud

Δτ cloud + rs

9.6μmbandfluxchange

O3 Surfacetemperature

Atmos.temperature

Watervapor

Cloud residual

HydrologicalCycleTroposphericO3GHGeffect

ΔFTOA =∂FTOA∂O3

ΔO3 +∂FTOA∂Tsur

ΔTsur +∂FTOA∂Tatm

ΔTatm +∂FTOA∂H2O

ΔH2O +∂FTOA∂τ cloud

Δτ cloud + rs

9.6μmbandfluxchange

O3 Surfacetemperature

Atmos.temperature

Watervapor

Cloud residual

5

InstantaneousRadiative Kernels(IRK):

ObjectivesandMotivations

IRKO3(z) =

∂FTOA(q)∂O3(z)

• Attribute the TOA flux change due to dominant physical quantities.

• Understand the dependence of O3 IRK variation on H2O, temperature, and clouds.

RH = ew (H2O,P)e*w (T,P)

HydrologicalCycleTroposphericO3GHGeffect

5-angleGaussianQuadratureintegrationmethod

FTOA = Lν0

π2

∫0

2π

∫∫ (θ )cosθ sinθdθdφdν

IRK(zl ) =∂FTOA∂ql (zl )

LIRK(zl ) =∂FTOA

∂lnql (zl )

LWRE = ΔFTOA = (∂FTOA∂ql (zl )

)qll=surface

tropopause

∑ (zl )

Topofatmosphericflux(9.6μmozoneband):

InstantaneousRadiative Kernel(mW/m2/ppb):

LogarithmIRK(mW/m2):

LongWaveRadiativeEffect(Troposphericcolumn)(W/m2):

FullIntegration AnisotropyIRKwi θiNadir (°)

0.015748 63.6765

0.073909 59.0983

0.146387 48.1689

0.167175 32.5555

0.096782 14.5752K(θNadiri ) = [∂L(ν,θNadir

i )∂q(zl )

]Δν∑

couldbeanyatmosphericstate,suchasprofilesofO3,Tatm,H2O,orTsur,cloudOD,emissivity,etc. [Wordenetal.,2011]

[Doniki etal.,2015] 6

Trop

osph

eric

O3GH

Geffect

q(zl )

TroposphericozoneGHGeffect

• TwosecondarystrongfluxsensitivityinLIRKisnearsubtropicalmidanduppertroposphereinbothhemispheres.

• HighestLWREoverMiddleEast duringborealsummer(>1Wm-2).

• SubtropicalmaximumandtropicallowinLWRE.

7

2006

O3 LWREandRH

L L L

MediterraneanBasinandMiddleEast:MildrainywintersHot,drysummers

L L L

8

H H H

L

• SimilarspatialpatterninLWREandRH• Spatiotemporalchangeoppositely

LWRE(Wm2) RH(%)

High >0.6 >80

Low <0.4 <30

• Low LWRE within ITCZ deep convection zones.

• High LWRE over subtropical low RH regions.

H H HH

500-hP

aRe

lHLW

RE

L LL L

HLH H H H

H

H

Jan. Jul. Jul.– Jan.

L L L

Lesssaturateinsummer

O3 LWREandRH

L L L

MediterraneanBasinandMiddleEast:MildrainywintersHot,drysummers

L L L

9

H H H

L

Higherwateramountinsummer

H H H

• SimilarspatialpatterninLWREandRH• Spatiotemporalchangeoppositely

H2Ocol50

0-hP

aRe

lHLW

RE

L LL L

HLH H H H H

H

Jan. Jul. Jul.– Jan.

L L L

Lesssaturateinsummer

O3 LWRE,TroposphericO3 column,&Tsur

L

10

Jan. Jul. Jul.– Jan.

• Australia highLWREinJan.isduetohigherTsur becauselargethermalcontrastamplifythesensitivity.• MiddleEastLWREmaximumalsorelevanttosummerO3 enhancement(Lietal.,2001;Liuetal.,2009)and

highTsur.• AfricasavannahighLWREinJan.isrelatedtobiomassburningandO3 enhancement.• CongobasinhighLWREinJul.isduetoO3 enhancement.

T sur

O3TR

OPcolumnLW

RE

S.Tropicsbelt(0°~25°S)(January)

11

In January, at central Pacific, Amazon, Congo basin, and Indonesia, deepconvection zones correspond to low ozone flux sensitivity.

The Walker circulation is the primary driver for the deep convection zones attropical central Pacific.

Longitude– altitudeview

L

RH LWRE

• ITCZinS.Tropics |Jan.• ITCZinN.Tropics |Jul.

CentralPacific

Amazon Congobasin

Indonesia

LLL

N.Tropicsbelt(0°~5°N)(July)

E.TropicalPacific

SoutheastAsia

12

L

Africasavanna

Longitude– altitudeview

RH

High RH at E. Tropical Pacific and moderate high RH at Africa savanna areanother two places corresponding to low LIRK.

In July, Asian monsoon is the primary driver to bring deep convection andheavy precipitation to India and southeast Asia, where LIRK are found low.

• ITCZinS.Tropics |Jan.• ITCZinN.Tropics |Jul.

HLL

LWRE

H HH L

Latitude– altitudeview

• Similar anti-correlation between RH and Ozone LIRK.

• Two mid tropospheric maximum in Ozone LIRK correspond to thesubtropical arid regions where the tropopause tends to sink and thedownwelling of Hadley cell dominants.

13

14

TroposphericO3 GHGeffect

H2O& Cloud

Tsur andTatm

Atmosphericopacity

Thermalcontrast

RelativeHumidity(RH)

TroposphericO3

MiddleEast(Jul.)

• C.TropicalPacific,Amazonbasin,Congobasin,Indonesia(Jan.)

• E.TropicalPacific,Savanna,SoutheastAsia(Jul.)

Australia(Jan.)

Savanna(Jan.)Congobasin(Jul.)

StrongGHGeffectRHlow

WeakGHGeffectRHhigh

H2O,cloud,T,O3 signaturesonO3 GHGeffect

High

Low

Attenuate

Strengthen

Conclusions• The tropospheric O3 GHG effect is low in tropics but maximized in

subtropics in both hemisphere.• RH is a useful quantity to help identify the primary driver, the large-

scale circulation, that determine H2O, temperature and clouddistribution. It also helps to understand the hydrological control on thetropospheric O3 GHG effect.

• Tropics:– H2O and clouds cause the low O3 GHG effect.– The primary drivers are walker circulation and Asia summer

monsoon for the deep convection.• Subtropics:

– Surface temperature and O3 enhancement contribute to high O3GHG effect.

– The primary drivers are the descent of tropopause height anddownwelling of Hadley cell.

– The maximum O3 GHG effect are found at Middle East during itshot dry summer (>1 W/m2). Ozone enhancement and high Tsur overdry desert with clear sky.

15

• Hadley cell expansion (Seidel and Randel, 2007)– The width expanding; poleward shift of the downward branch– A shift in the ITCZ farther away from the equator due to the response

to CO2 forcing (Held, 2000; Kang and Lu, 2012; Lu et al., 2007)– Increase of global T and pole-to-equator T gradient (Frierson et al.,

2007)

• Inhabitability of Middle East due to global warming (Pal et al., 2016)– Additional O3 radiative forcing to this region

• The Asia monsoon strengthen (Li et al., 2010; Singh et al., 2014)– Another positive feedback to the Middle East O3 GHG effect

16

Future outlook

Kuai etal.2016submittedtoELEMENTA

17

Thankyou!

18

IASI-TESLWREcomparisons

• IASI=TES ±0.5 ° lat/lon• <6hourtimedifference,2011.07.15

CloudeffectonozoneIRK

19

Wordenetal.,2011:Fig.3

• Clouds significantly reduce the TOA flux sensitivity to O3 in the lowertroposphere compared to the clear sky kernels (Soden et al., 2008).

• Tropical clouds also greatly reduce the mid tropospheric maximum in O3IRK and contribute to tropical low LWRE.

Shrinktoathinnerlayershiftedupwards

ITCZshiftfromsouthofequator tonorthofequatorfromJanuary toJuly.

• InsideITCZbelt:Ø DeepconvectionØ Wet,rainyseason,and

cloudysky• OutsideITCZbelt:

Ø SubsidenceregionØ Aridandclearsky

• January:deepconvectionzoneatcentralPacific,Amazon,S.Africa(Congobasin),andIndonesia.

• July: deepconvectionzoneoccurnorthofequatoratE.TropicalPacific,AfricaSavanna,southeastAsia.

January July

20

TheInterTropicalConvergenceZone(ITCZ)inRHTO

A flu

x

5

00-h

Pa R

H

RelativeHumidity(RH)

Theamountofwatervaporpresentinairexpressedasapercentageoftheamountneededforsaturationatthesametemperature.

Theratioofthepartialpressureofwatervaporinthemixturetotheequilibriumvaporpressureofwateratagiventemperature.

RHdescribesthestateofatmosphericsaturationandsuggeststheclouddistributionbasedonthecombinationofwatervaporandtemperature.

RH = ew (H2O,P)e*w (T,P)

e*w (T,P) = (1.0007+ 3.46 ×10−6P)× (6.1121)e

( 17.502T240.97+T

)

21