[PPT] - How robust are stratospheric H 2 O and AoA trends derived from PowerPoint Presentation

SLIDE 1

How robust are stratospheric H2O and AoA trends derived from different re-analysis products?

Paul Konopka, Felix Ploeger, Bernard Legras, Mengchu Tao, Liuba Poshyvailo, Xiaolu Yan, Jonathon Wright, Rolf M¨ uller and Martin Riese

Forschungszentrum J¨ ulich, Germany, Institute for Energy and Climate Research - Stratosphere (IEK-7) P .Konopka@fz-juelich.de http://www2.fz-juelich.de/icg/icg-i/www export/p.konopka .

SLIDE 2

Outline

CLaMS - Lagrangian Chemistry Transport Model (Lagrangian mixing, diabatic heating rates) ERA-Interim versus JRA-55 (1979-2013) Stratospheric water vapor Mean age of air (AoA) and the Brewer-Dobson circulation Conclusions

SLIDE 3

Greenland from space shuttle (NASA) CLaMS - Lagrangian Chemistry Transport Model with ≈ 106 air parcels air parcel = “pivotal point with mixing ratios µi, of m species with i = 1, ..., m” Atmosphere below 0.1 hPa is resolved, 100km/400 m – hor./vert. resolution Horizontal meteor. winds (ERA-Interim, JRA55, NCEP) Vertical velocity: diabatic heating rates from radiation, latent heat rather than from ˙ p 3-D forward trajectories and Lagrangian mixing simplified chemistry and dehydration scheme McKenna et al., JGR, 2002, Konopka et al., JGR, 2004, Grooß et al., 2005, ACP , Konopka et al., 2007, ACP , Ploeger et al., 2010, 2013 JGR, Pommrich et al., 2014, GMD

Mixing Trajectory Chemistry Sedimentation

SLIDE 4

species lower boundary upper boundary (∼50 km) CH4 CMDL/AIRS HALOE Mean Age linear source MIPAS (SF6) CO2 CMDL Mean Age CO MOPITT/AIRS Mainz-2D O3 HALOE, θ ≥ 500 K O3 (tracer) HALOE, θ ≥ 500 K HCl HALOE, θ ≥ 500 K H2O ECMWF , ζ ≤ 250 K HALOE N2O, F11,F12 CMDL HCN MODIS Simplified chemistry CH4 ⇒ (OH, O(1D), Cl) ⇒ H2O, CO ⇒ (OH) ⇒ CO2 (hν) ⇒ O3 ⇒ (HOx) ⇒, N2O, F11, F12 ⇒ (O(1D), hν) ⇒ HCN ⇒ (OH, O(3D), uptake by the ocean)⇒

Multi-annual CLaMS simulations (1979-today)

HALOE - Climatology:

Grooss and Russell, ACP , 2005

CMDL: GLOBALVIEW, 2015

CO2/CH4/CO since 1979/84/91 P . Tans, K. Masarie, P . Novelli

CMDL: CATS (5 stations)

N2O, F11, F12 - J. Elkins

MIPAS, SF6-Age

Stiller et al., ACP , 2008

MOPITT (V3, V4)/AIRS

Pommrich at al., GMD, 2014

HCN

Pommrich at al., GRL, 2010

SLIDE 5

Lagrangian mixing and transport barriers

Hoppe et al, GMD, 2014

Southern Hemisphere winter: Antarctic vortex edge as N2O gradient (in pppbv/m) at θ = 450K. Eulerian (EMAC-FFSL, Lin and Rood, 1996) versus Lagrangian transport (CLaMS) versus MLS climatology at θ = 450K in September. ⇒ more realistic transport barrier in CLaMS

SLIDE 6

...diabatic thinking

SLIDE 7

diabatic rather than kinematic vert. velocities

potential temperature θ defines the vertical coordinate Cross isentropic velocity ˙ θ derived from the ERA-Interim forecast total diabatic heating (long- and shortwave radiation with clouds + latent heat +,..., see Ploeger et al., 2010) + annually averaged mass conservation (Rosenlof et al., JGR, 1995) ⇒ σ-θ, hybride ζ-coordinate (Mahowald et al., JGR, 2002)

SLIDE 8

diabatic rather than kinematic vert. velocities

potential temperature θ defines the vertical coordinate Cross isentropic velocity ˙ θ derived from the ERA-Interim forecast total diabatic heating (long- and shortwave radiation with clouds + latent heat +,..., see Ploeger et al., 2010) + annually averaged mass conservation (Rosenlof et al., JGR, 1995) ⇒ σ-θ, hybride ζ-coordinate (Mahowald et al., JGR, 2002)

p [hPa]

340 360 380 310 280 100 200 300 500 1013

[K]

0.12 0.25 0.80 0.40 1.00

θ

=p/p

s 10 50 70 30

[deg N] Subtropical Jet σ

ζ = θ (pot. Temp.) above 300 hPa

dζ dt = dθ dt

SLIDE 9

diabatic rather than kinematic vert. velocities

potential temperature θ defines the vertical coordinate Cross isentropic velocity ˙ θ derived from the ERA-Interim forecast total diabatic heating (long- and shortwave radiation with clouds + latent heat +,..., see Ploeger et al., 2010) + annually averaged mass conservation (Rosenlof et al., JGR, 1995) ⇒ σ-θ, hybride ζ-coordinate (Mahowald et al., JGR, 2002)

p [hPa]

340 360 380 310 280 100 200 300 500 1013

[K]

0.12 0.25 0.80 0.40 1.00

θ

=p/p

s 10 50 70 30

[deg N] Subtropical Jet σ

ζ = θ (pot. Temp.) above 300 hPa

dζ dt = dθ dt

ζ ∼ σ = p/ps, ps - surf. pressure below 300 hPa

dζ dt = ˙

σ

SLIDE 10

Diabatic versus kinematic

200. 400. 600. O3 [ppbv] 370 380 390 400 410 420 θ [K]

HALOE FOZAN ERA−Int. assim kinematic diabatic

0.0

Trajectory-based reconstruction of O3 diabatic significantly better than kinematic but why ? (from Ploeger et al., ACP , 2011)

SLIDE 11

Diabatic versus kinematic

isen_model/DIA

−50 50 Latitude, deg N 360 370 380 390 400 410 420 ZETA [K]

era_interim

isen_model/KIN

−50 50 Latitude, deg N 360 370 380 390 400 410 420 ZETA [K]

era_interim

...because diabatic approach is less dispersive, mainly due to assimilation errors in the ˙ σ ≈ dp/dt fields ! (Eluszkiewicz et al, 2000, Schoeberl et al., 2003, 2005, Diallo et al., 2012) kinematic diabatic

SLIDE 12

H2O-taperecorder

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K]

MLS (HALOE) MLS (HALOE)

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K] 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 H2O (ppmv)

3 40 50 70 1

MLS climatology, 2005-12 white line - HALOE

SLIDE 13

H2O-taperecorder

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K]

MLS (HALOE) MLS (HALOE)

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K] 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 H2O (ppmv)

3 40 50 70 1

MLS climatology, 2005-12 white line - HALOE

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K]

ERA−Interim ERA−Interim

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K]

3 40 50 70 1

ERA-Interim climatology, ±15N, 2002-12 (-) sligthly too dry during winter/spiring (-) slightly too wet during summer (Fueglistaler et al., JGR, 2013) (-) ...but much too fast tropical upwelling ! (Dee et al, QJRMS, 2011)

SLIDE 14

...and from CLaMS (diabatic transport)

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K]

MLS (HALOE) MLS (HALOE)

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K] 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 H2O (ppmv)

3 40 50 70 1

MLS climatology, 2005-12 white line - HALOE

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K]

CLaMS 2002−12 CLaMS 2002−12

J F M A M J J A S O N D Month 350 400 450 500 550 600 Potential Temperature, θ, [K]

3 40 50 70 1

CLaMS climatology, ±15N, 2002-12 ...tropical upwelling is much better represented (diabatic heating rates from ERA-Interim)

SLIDE 15

How robust are CLaMS simulations with respect to the used re-analysis ? (diabatic heating rates)

SLIDE 16

Zonal mean diabatic heating (2001-10) from different reanalisis products: top - total, middle

radiation, bottom - residuum (laten heat + ..), Wright and Fueglistaler, ACP

, 2013

SLIDE 17

Zonal mean diabatic heating (2001-10) from different reanalisis products: top - total, middle

radiation, bottom - residuum (laten heat + ..), Wright and Fueglistaler, ACP

, 2013 Zonal mean diabatic heating (2001-10) from different reanalisis products: top - total, middle

radiation, bottom - residuum (laten heat + ..), Wright and Fueglistaler, ACP

, 2013

SLIDE 18

H2O/AoA from CLaMS driven by: ERA-Interim (Dee et al, 2011) JRA-55 (Kobayashi et al, 2015)

SLIDE 19

Linear trends (35 years)

−4 −2 2 4 Age [month] −4 −2 2 4 Age [month] −1.0 −0.5 0.0 0.5 1.0 H2O [ppmv] −1.0 −0.5 0.0 0.5 1.0 H2O [ppmv]

CLaMS_ERA CLaMS_JRA HALOE

1980 1990 2000 2010 TIME [year] 1982 1984 1986 1988 1992 1994 1996 1998 2002 2004 2006 2008 2012

MLS MW during eQBO MW during wQBO

adapted from Tao et al., 2015, with added JRA-related analysis

Evolution of H2O (top) and of the mean age of air AoA (bottom) in the tropics (±10N) at θ = 400 K ( 18km) shown as the deseasonalized anomaly with respect to the 35 year climatology (15 days running mean).

SLIDE 20

Linear trends (35 years)

−4 −2 2 4 Age [month] −4 −2 2 4 Age [month] −1.0 −0.5 0.0 0.5 1.0 H2O [ppmv] −1.0 −0.5 0.0 0.5 1.0 H2O [ppmv]

CLaMS_ERA CLaMS_JRA HALOE

1980 1990 2000 2010 TIME [year] 1982 1984 1986 1988 1992 1994 1996 1998 2002 2004 2006 2008 2012

MLS MW during eQBO MW during wQBO

adapted from Tao et al., 2015, with added JRA-related analysis

Evolution of H2O (top) and of the mean age of air AoA (bottom) in the tropics (±10N) at θ = 400 K ( 18km) shown as the deseasonalized anomaly with respect to the 35 year climatology (15 days running mean).

(-) well-resolved variability on a time scale from weeks to few years (-) significant differences in the decadal variability (-) largest differences in the 80s and after 2000

SLIDE 21

Stratospheric water vapor (seasonality)

SLIDE 22

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 400 450 500 550 600 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 400 450 500 550 600 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 400 450 500 550 600 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 400 450 500 550 600 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 400 450 500 550 600 Potential Temperature, [K] Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 400 450 500 550 600 Potential Temperature, [K] 2.5 2.8 3.0 3.5 4.0 4.5 5.0 5.5 6.0 H2O [ppmv]

ERA/CLaMS JRA/CLaMS MLS

10S-10N, 2004-13

SLIDE 23

−50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K] −50 50 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K] 2.5 2.8 3.0 3.5 4.0 4.5 5.0 5.5 6.0 H2O [ppmv]

ERA/CLaMS JRA/CLaMS MLS

DJF , 2004-13

SLIDE 24

−50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. −50 50 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K] −50 50 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K] 2.5 2.8 3.0 3.5 4.0 4.5 5.0 5.5 6.0 H2O [ppmv]

ERA/CLaMS JRA/CLaMS MLS

JJA, 2004-13

SLIDE 25

2.0 2.5 3.0 3.5 4.0 4.5 5.0 6.0 7.0 8.5 10.0 H2O [ppmv]

ERA/CLaMS JRA/CLaMS MLS

DJF , θ = 380K, 2004-13

SLIDE 26

2.0 2.5 3.0 3.5 4.0 4.5 5.0 6.0 7.0 8.5 10.0 H2O [ppmv]

ERA/CLaMS JRA/CLaMS MLS

JJA, θ = 380K, 2004-13 ERA/CLaMS slightly drier than JRA/CLaMS, but why ?

SLIDE 27

Temparature differences

−50 50 Latitude [deg] 400 500 600 700 800 900 1000

Pot. temperature [K]

TEM/abs: TEMP, djf TEM/abs: TEMP, djf

−50 50 Latitude [deg] 400 500 600 700 800 900 1000

Pot. temperature [K]

−50 50 Latitude [deg] 400 500 600 700 800 900 1000

Pot. temperature [K]

TEM/abs: TEMP, jja TEM/abs: TEMP, jja

−50 50 Latitude [deg] 400 500 600 700 800 900 1000

Pot. temperature [K]

−3. −2. −2. −1. −1. 0. 1. 1. 2. 2. 3. ∆TEMP, K

DJF JJA ∆T := JRA − ERA

Zonally averaged temperature difference between JRA and ERA (1979-2013). lower tropical stratosphere ⇓ JRA is warmer if compared with ERA. ⇓ more H2O in JRA/CLaMS

SLIDE 28

Mean age of air (AoA) ERA versus JRA versus

bservations

SLIDE 29

Age of Air (AoA)

−50 50 Latitude [deg] 400 600 800 1000 1200

Pot. temperature [K]

−50 50 Latitude [deg] 400 600 800 1000 1200

Pot. temperature [K]

AoA [year]

10 14 1 4 1 8 18 22 22 26 26 30 3 34 34 38 3 8

ERA-Interim

−50 50 Latitude [deg] Latitude [deg] 400 600 800 1000 1200

Pot. temperature [K]

−50 50 400 600 800 1000 1200

Pot. temperature [K]

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 AoA [year]

10 14 14 18 18 22 22 26 26 30 30 34 34 38 38

JRA-55

−50 50 Latitude [deg] 400 600 800 1000 1200

Pot. temperature [K]

−50 50 Latitude [deg] 400 600 800 1000 1200

Pot. temperature [K]

−25 −20 −15 −10 −5 5 10 15 20 25 ∆AoA [%]

10 14 1 4 1 8 18 22 22 26 26 30 3 34 34 38 3 8

Difference: JRA - ERA

1979-13

(-) ERA and JRA-based AoA climatology very similar (-) Differences up to 20% in the lowest stratosphere (-) asymmetry between the the SH (older air) and NH (younger air) well resolved

SLIDE 30

Age spectrum

(a) (b)

1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg]

Age spectrum @ 600K (DJF), ERA Age spectrum @ 600K (DJF), ERA

1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg] 0.0005 0.0011 0.0023 0.0050 0.0108 0.0232 0.0500 PDF 1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg]

Age spectrum @ 600K (JJA), ERA Age spectrum @ 600K (JJA), ERA

1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg] 0.0005 0.0011 0.0023 0.0050 0.0108 0.0232 0.0500 PDF

(c) (d)

1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg]

Age spectrum @ 600K (DJF), JRA Age spectrum @ 600K (DJF), JRA

1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg] 0.0005 0.0011 0.0023 0.0050 0.0108 0.0232 0.0500 PDF 1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg]

Age spectrum @ 600K (JJA), JRA Age spectrum @ 600K (JJA), JRA

1 2 3 4 5 6 Transit time [year] −50 50 Latitude [deg] 0.0005 0.0011 0.0023 0.0050 0.0108 0.0232 0.0500 PDF

Age spectra at 600 K from CLaMS–ERA (top) and CLaMS–JRA (bottom) for December–February (left) and June–August (right) with AoA (solid white lines) and modal AoA (white diamonds). Adapted from Ploeger and Birner, ACP , 2016.

SLIDE 31

AoA versus experiment

20km

−50 50 Latitude [deg] 2 4 6 8 Mean Age [year]

MIPAS CLaMS−JRA CLaMS−ERA CO2 SF6

Zonal means of AoA at 20 km. CLaMS/ERA (red), CLaMS/JRA (blue), MIPAS (gray), 2002–

2012. In-situ observations of CO2/SF6 (symbols) Shadings: range of monthly mean values

for CLaMS/MIPAS observations. Error bars: range between maximum and minimum CO2

bservations in each latitude bin. Adapted from Ploeger at al., 2015, with JRA added.

SLIDE 32

Variability of AoA

(a) .

2005 2006 2007 2008 2009 2010 2011 2012 −60 −40 −20 20 40 60 Latitude 2005 2006 2007 2008 2009 2010 2011 2012 −60 −40 −20 20 40 60 Latitude −0.95 −0.30 −0.20 −0.10 0.00 0.10 0.20 0.30 0.40 [year]

− 1 5 −15 − 1 5 − 1 5

(b)

−60 −40 −20 20 40 60 Latitude −60 −40 −20 20 40 60 Latitude 2006 2008 2010 2012 −0.95 −0.30 −0.20 −0.10 0.00 0.10 0.20 0.30 0.40 year

− 1 5 −15 − 1 5 − 1 5 −15

(c)

−60 −40 −20 20 40 60 Latitude −60 −40 −20 20 40 60 Latitude 2006 2008 2010 2012 −0.95 −0.30 −0.20 −0.10 0.00 0.10 0.20 0.30 0.40 year

− 1 5 −15 − 1 5 −15 − 1 5

CLaMS-ERA CLaMS-JRA MIPAS AoA-anomaly (25km)

AoA deseasonalized anomalies at 25 km from MIPAS

bservations,

2005–2012 (a), from CLaMS/ERA (b) and from CLaMS/JRA (c), (CLaMS AoA has been averaged between 22–28 km.) Black line shows easterly zonal wind (−15 m/s). Adapted from Ploeger et al., 2015 with JRA added.

SLIDE 33

Empirical Orthogonal Function (EOF) analysis

−50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K]

EOF1, rel. var.=70.9%, season:ALL/smooth: ±02 years EOF1, rel. var.=70.9%, season:ALL/smooth: ±02 years

−50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K] −8.0 −6.5 −5.0 −4.0 −3.0 −1.5 0.0 1.5 3.0 4.0 5.0 6.5 8.0 ∆AoA [months] 1980 1985 1990 1995 2000 2005 2010 Year −1.0 −0.5 0.0 0.5 1.0 Fraction of EOF1

c1(t)

−50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K]

EOF1, rel. var.=96.1%, season:ALL/smooth: ±02 years EOF1, rel. var.=96.1%, season:ALL/smooth: ±02 years

−50 50 Latitude [deg N] 380. 420. 450. 500. 600. 800. 1000. Potential Temperature, [K] −8.0 −6.5 −5.0 −4.0 −3.0 −1.5 0.0 1.5 3.0 4.0 5.0 6.5 8.0 ∆AoA [months] 1980 1985 1990 1995 2000 2005 2010 Year −1.0 −0.5 0.0 0.5 1.0 Fraction of EOF1

c1(t)

ERA JRA

Variability in the zonal and annual means of the AoA after the QBO signal was removed (smoothing over 4 years). Because the first EOF contributes to more than 60% of the AoA anomaly, (∆AoA), we approximate: ∆AoA(x, t) =≈ c1(t)EOF1(x) EOF1(x)/c1(t) – first EOF and principal comp. ⇓ (-) both reanalysis resolve the impact of Pinatubo in 1991 (increasing age of air) (-) a clear effect of El Chichon in 1982 is only present in JRA/CLaMS (-) both reanalysis show that the stratosphere becomes younger over the time with strongest contribution in the SH.

SLIDE 34

Differences in trends

f AoA

SLIDE 35

Decadal trends of AoA (2002-12)

(a)

−60 −40 −20 20 40 60 Latitude 400 500 600 700

Pot. temperature [K]

−60 −40 −20 20 40 60 Latitude 400 500 600 700

Pot. temperature [K]

−20 −16 −12 −8 −4 4 8 12 16 18 [%/dec]

1 2 12 1 4 14 16 16 18 18 20 2 22 2 2 24 24 26 26 28 28

(b)

−60 −40 −20 20 40 60 Latitude 400 500 600 700

Pot. temperature [K]

−60 −40 −20 20 40 60 Latitude 400 500 600 700

Pot. temperature [K]

−20 −16 −12 −8 −4 4 8 12 16 18 [%/dec]

1 2 12 1 4 14 16 16 18 18 20 2 22 2 2 24 24 26 26 28 28

(c)

−60 −40 −20 20 40 60 Latitude 400 500 600 700

Pot. temperature [K]

−60 −40 −20 20 40 60 Latitude 400 500 600 700

Pot. temperature [K]

−20 −16 −12 −8 −4 4 8 12 16 [%/dec]

12 1 2 14 14 16 1 6 1 8 18 2 20 22 22 2 4 24 26 26 28

AoA trend CLaMS-ERA (2002-12) AoA trend CLaMS-JRA (2002-12) AoA trend MIPAS (2002-12)

Decadal change (in percent) of zonal mean AoA, 2002– 2012 from MIPAS (a), from CLaMS/ERA (b) and from CLaMS/JRA (c) (linear regression). The significance of the linear trend (multiples of the standard deviation σ) is shown as gray contours. ⇓ (-) decadal AoA trend from CLaMS/ERA is consistent with MIPAS observations, regarding the NH/SH asymmetry (age increase/decrease in NH/SH) and the age decrease in the lowest part of the stratosphere (-) CLaMS/JRA trends are very different and largely not consistent with MIPAS (e.g., positive trends in the SH). Adopted from Ploeger et al., 2015, with JRA added.

SLIDE 36