Radia%veInfluencesofNaturalVariabilityinTropicalLowerStratosphericWaterVaporandOzoneDanielM.Gilford(dgilford@mit.edu)andSusanSolomon

Mo8va8on

January2016

Acknowledgements

DepartmentofEarth,Atmospheric,andPlanetarySciences,MIT

This work was supported by NASA Headquarters under the NASA Earth and Space Science Fellowship

Program - Grant NNX14AK83H. It was also supported by NSF Grant AGS-1342810 and 1461517.

SelectedReferences

Methods

SeasonalCycleResults QBOResults

Conclusions

•  Mote,P.W.,andcoauthors,1996.J.Geophy.Res.,101,3989–4006.•  Conley,A.J.,J.-F.Lamarque,F.ViW,W.D.Collin,andJ.Kiehl,2013.Geosci.ModelDev.,6,469–476.•  Folkins,I.,P.Bernath,C.Boone,G.Lesins,N.Livesey,A.M.Thompson,K.Walker,andJ.C.WiWe,2006.Geophys.Res.LeB.,

33,doi:10.1029/2006GL026602.•  FreeUniversityofBerlin,2015.[Availableat:hWp://www.geo.fu-berlin.de/en/met/ag/strat/produkte/qbo/]•  Fueglistaler,S.,P.H.Haynes,andP.M.Forster,2011.Atmos.Chem.Phys.,11,3701–3711,doi:10.5194/acp-11-3701-2011.•  Gilford,D.M.,S.Solomon,andR.W.Portmann,inpress.J.Climate,doi:10.1175/JCLI-D-15-0167.1.•  Livesey,N.J.,andcoauthors,2011.JetPropulsionLaboratory,Pasadena,California.[Availableat:

hWp://mls.jpl.nasa.gov/data/v3-3_data_quality_document.pdf]

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

12 17 26 38 56 82121177261

Press

ure (

hPa)

Month (Jan−Dec)

−5−4−3−2−1012345

12 17 26 38 56 82121177261

−25

−20

−15

−10

−5

0

5

10

−25−20−15−10−50510152025

12 17 26 38 56 82121177261

Press

ure (

hPa)

1 2 3 4 5 6 7 8 9 10 11 12Month (Jan−Dec)

1 2 3 4 5 6 7 8 9 10 11 12Month (Jan−Dec)

T (K) H2O (%) O3 (%)

0 1 2 3 4 5 6 7 8 9 10 11Lag (months)

O3/QBO β values at each Level/Lag

−1−0.8−0.6−0.4−0.200.20.40.60.81

0 1 2 3 4 5 6 7 8 9 10 11

12 17 26 38 56 82

121177261

Lag (months)

Press

ure (

hPa)

H2O/QBO β values at each Level/Lag

2006 2008 2010 2012

12 17 26 38 56 82121177261

Time

Press

ure (

hPa)

QBO O3 Regression (%)

−10

−5

0

5

10

2006 2008 2010 2012

12 17 26 38 56 82121177261

Time

Press

ure (

hPa)

QBO H2O Tape Recorder (%)

−15

−10

−5

0

5

10

15

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

23

53

85118163226313

O3 Low QBO Tadj (K)

Hybr

id Sig

ma/P

ressu

re

Month (Jan−Dec)

−0.75−0.6−0.45−0.3−0.1500.150.30.450.60.75

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

23

53

85118163226313

H2O Low QBO Tadj (K)

Month (Jan−Dec)

−0.15−0.12−0.09−0.06−0.0300.030.060.090.120.15

Hybr

id Sig

ma/P

ressu

re

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

23

53

85118163226

Ozone Radiative Seasonal Cycle (K), Tadj

Month (Jan−Dec)

−1.25−1−0.75−0.5−0.2500.250.50.7511.25

Hybr

id Sig

ma/P

ressu

re

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

23

53

85118163226

Water Vapor Radiative Seasonal Cycle (K), Tadj

Month (Jan−Dec)

−0.5−0.4−0.3−0.2−0.100.10.20.30.40.5

Hybr

id Sig

ma/P

ressu

re

Aura Microwave Limb Sounder (MLS), version 3.3 [Livesey et al. 2011] •  H2O, O3, and Temperature measurements, over 20S – 20N. •  5° x 5°, 316–0.02 hPa, 39 vertical levels QBO Index: Normalized 50hPa Singapore Winds [Free University of

Berlin 2015]

How do tropical lower stratospheric H2O/O3 seasonal cycles and QBO structures radiatively impact temperatures in the

upper troposphere and lower stratosphere (UTLS)?

Why is this important? •  Increase understanding of UTLS radiative controls •  Predictability and model representations of variability •  Upper tropospheric stability (convection and hurricanes) •  Stratospheric H2O important for surface climate and very

sensitive to temperatures •  Further applications exploring UTLS trends

Wave-driving and temperatures

vary seasonally and with QBO

Associated vertical structure in H2O and O3

anomalies

Radiative impact on UTLS

temperatures

Parallel Offline Radiative Transfer (PORT) model [Conley et al. 2013] •  10° x 15°, 992–3.5 hPa, 26 Hybrid vertical levels •  Seasonally Evolving Fixed Dynamical Heating (SEFDH) assumption,

Q à Heating Rates, T à Temperature, c à Constituents, t à Time: •  Tadj = T – Tp à Radiative Temperature Adjustment from Perturbation (p) •  One-year simulations with 4 month spin-up time (16 months total)

dTadjdt

=Q(T,c)−Q(Tp,cp )

Data

Model

Runs

MLS Observed Seasonal Cycles:

Temperature Adjustments (Tadj):

Lag Correlations (β) between H2O / O3 and QBO:

QBO-Regressed Anomalies at Optimal Lag:

Associated Tadj:

For the Seasonal Cycle and QBO, we perform the following calculations to test the importance of anomaly vertical structures: •  Full runs − Perturbations applied

everywhere above tropopause

•  Cutoff runs − Perturbations applied at and below the cutoff pressure level and above the tropopause

Determine H2O and O3

perturbations

Apply perturbations to PORT

background

Run PORT with SEFDH

Analyze resulting Temperature Adjustments

Seasonal Temperature Range (K) H2O Full Structure

O3 Full Structure

H2O Cutoff ~85hPa

O3 Cutoff ~85hPa

H2O Cutoff ~53hPa

O3 Cutoff ~53hPa

85 hPa 0.50* 1.77 0.41* 0.94 0.48* 1.48

118 hPa 0.44 0.51 0.48 0.15 0.44 0.36

Tropopause

•  Lower stratosphere: H2O Tadj offset seasonal cycle; O3 Tadj amplify the cycle

•  Upper troposphere: H2O and O3 Tadj constructively amplify/positively shift the seasonal cycle

•  H2O Tadj ~insensitive to cutoff altitude. Nearly all radiative influences due to local lower stratospheric anomalies

•  Lower stratospheric O3 Tadj strongly depends on nonlocal radiative influences. ~46% of 85hPa Tadj and ~66% of 118hPa Tadj due to O3 anomalies above 85hPa

•  UTLS Tadj lag H2O / O3 anomalies by 2-3 months

Fig. 1: Example H2O profile to illustrate run methodology

•  A “tape recorder” signal is apparent in H2O lag correlations; significant in lower stratosphere

•  O3 lag correlations follow QBO signal; significant at higher altitudes

•  Structures are consistent with correlations; a “low” QBO period ( ) is selected for radiative runs

•  QBO-related Tadj much smaller than seasonal cycle, especially below tropopause

•  O3 impacts are strong locally—away from tropopause—inconsistent with weaker H2O signal

Fig. 4: Seasonal cycles of H2O and O3 Tadj on 85, 118hPa hybrid surfaces

Table 1: Seasonal cycle temperature ranges for radiative Tadj. (*) indicates cycle offsets sign of the observed temperature cycle

Fig. 3: Mean tropical seasonal cycles of Tadj for H2O and O3 vs. pressure level. Black dashed curves are the levels analyzed below.

Fig. 2: Mean tropical seasonal cycles of Temperature, H2O and O3 vs. pressure level

Fig. 5: Pearson Correlation coefficients, hatching indicates >90% confidence

Fig. 6: Regression time series at each pressure level’s optimal lag

Fig. 7: Mean tropical QBO signals of Tadj for H2O and O3 vs. pressure level.

1.  Stratospheric seasonal cycles of H2O and O3 act to radiatively cool the upper troposphere in the boreal spring, and warm it in the boreal fall

2.  Anomalies very close to the tropopause (~85hpa) dominate the H2O radiative signal, with little influence from the overlying structure

3.  About half of the O3 radiative influences in the UTLS result from anomalies above the

lowermost stratosphere (p<85hPa)

4.  QBO-related H2O anomalies result in UTLS radiative influences smaller than those of the seasonal cycle; O3 influences are larger, but primarily focused at higher altitudes

−40 −30 −20 −10 0 10 20

313846566882

100121146

Anomaly (%)

Press

ure (

hPa)

Full StructureCutoff 82hPaCutoff 56hPa

1 2 3 4 5 6 7 8 9 10 11 12−0.25−0.15−0.050.050.150.250.35

Tadj

(K)

Month (Jan−Dec)

1 2 3 4 5 6 7 8 9 10 11 12−0.75−0.5

−0.250

0.250.5

0.751

1.25

Tadj

(K)

Month (Jan−Dec)

118 hPa (Below Tropopause)

85 hPa (Above Tropopause)

H2O FullO3 FullH2O Cutoff ~85hPaH2O Cutoff ~70hPaH2O Cutoff ~53hPaO3 Cutoff ~85hPaO3 Cutoff ~70hPaO3 Cutoff ~53hPa

H2O FullO3 FullH2O Cutoff ~85hPaH2O Cutoff ~70hPaH2O Cutoff ~53hPaO3 Cutoff ~85hPaO3 Cutoff ~70hPaO3 Cutoff ~53hPa

How do the anomalies at higher levels affect temperatures below?

Tropopause