Interannual variations in global OH radicals over the period 1987-2005 in GEOS-Chem, and preliminary comparisons to other models I. Bey 1, S. Koumoutsaris.

Interannual variations in global OH radicals over the period 1987-2005 in GEOS-Chem,

and preliminary comparisons to other models

I. Bey1, S. Koumoutsaris1, S. Generoso1,J. Drevet1, A. Fiore2, T. van Noije3, S. Rast4, M. Schultz5, and S. Szopa6

Contact: [email protected], http://lmca.epfl.ch

1EPFL, 2GFDL, 3KNMI, 4MPI-Met, 5FZ Jüelich, 6IPSL Paris

GEOS-Chem simulation set up

•GEOS-Chem 3D Chemical-Transport Model Version 7.02.04 – Meteorological fields from NASA GMAO GEOS-4 – 4°x5°, 30 levels (from surface to 0.01 hPa) – 24 tracers for tropospheric O3-NOx-HC chemistry

•Interannual variation of input parameters– Biomass burning emissions

IAV prescribed using TOMS AI (1987-07/1996) [Duncan et al., 2003] andAATSR (08/1996-12/2005) [Generoso et al., 2003]

– Anthropogenic emissionsFossil fuel emissions based on climatological inventories scaled using

CO2 trends (-> 1998) and change in aircraft emissions (+3%/yr with a base year in 1992) – Methane concentrations

Prescribed using CH4 CMDL data (latitudinal bands) up to 2004

– Total O3 column

Prescribed TOMS satellite data up to 2002 (Nimbus 7, Meteor 3, Eptoms and climatology when data are missing, e.g. 1995)

CO anomalies (ppb)

Barrow, Alaska 71°N - 20m Mauna Loa, Hawaii 19°N - 3397m

Niwot Ridge, Colorado 40°N - 3475m Alert, Canada 82°N - 210m

Mace Head, Ireland 53°N - 25m Tutor Hill, Bermuda 32°N - 30m

1987 1990 1995 2000 2005 1987 1990 1995 2000 2005

Observations — monthly mean • 12-month running meanModel — monthly mean • 12-month running mean

Mauna Loa, Pacific 19°N - 3397m

Mace Head, Ireland 53°N - 15m

O3 anomalies (ppb)

Observations — monthly mean • 12-month running meanModel — monthly mean • 12-month running mean

Stratospheric contribution • 12-month running mean

Jungfraujoch, Switzerland 46°N - 3578m

See also Ordóñez et al., GRL, 2007: Changes in trop O3 driven by changes in O3 in the lowermost stratosphere (due to changes in dynamics)

Interannual variability in global mean OH in GEOS-Chem

OH global air mass weighted mean (1e5 molec. cm-3)

Annual anomalies (1e4 molec. cm-3)

=> Implications for CH4 trends

Extra tropics NH Tropics NH Tropics SH Extra tropics SH

Column(surface-250 hPa)

Surface-750 hPa

750-500 hPa

500-250 hPa

Interannual variability in OH in different regions (1e4 molec. cm-3)

Parameters contributing to OH interannual variability

standard – simulation with fixed anthropogenic emissionsstandard – simulation with fixed meteorology (1988 recycled)standard – simulation with fixed lightning emissions standard – simulation with fixed TOMS columnstandard – simulation with fixed methane concentrationsstandard – simulation with fixed biomass burning emissions

OH

(1

e5 m

ole

c cm

-3)

OH

(1

e5 m

ole

c cm

-3)

Interannual variability in J(O1D)

Standard simulationSimulation with a fixed total ozone column

J(O

1D

) (s

-1)

J(O

1D

) (s

-1)

Interannual variability in global P(OH)= J(O1D) × kH2O[H2O]/ kX[X](s-1)

Standard simulationSimulation with a fixed total ozone column

J(O

1D

) ×

kH

2O[H

2O

]/

kX[X

] (s

-1)

Interannual variability in water vapourQ

(g k

g-1)

Q (

g k

g-1)

Trenberth et al., Trends and variability in column-integrated atmospheric water vapor, Climate Dynamics, 2005: “Precipitable water variability for 1988–2001 is dominated by the evolution of ENSO and especially the structures that occurred during and following the 1997–98 El Niño event. … but … Users of these data (ECMWF ERA-40, NCEP, etc.) should take great care in accepting results as real. ”

Available observations during the reanalysis

SSM/I data TOVS data

Total precipitable water and surface winds

No data before 1987

Temperature, moisture, …

TOVS (HIRS, MSU, SSU) to ATOVS (HIRS-2, AMSU) in 1998

Courtesy: Steven Pawlson

=> Impact of the various observations used in the re-analysis on the water vapor variability & trends?

Variability in OH in GEOS-Chem compared to other models

GEOS-Chem (GEOS-4, CTM)TM4 (CTM, ECMWF, RETRO emissions, T. van Noije)MOZECH (GCM, ECMWF, RETRO emissions, S. Rast & M. Schultz)LMDzINCA (GCM, ECMWF, RETRO em., S. Szopa & D. Hauglustaine)

OH

anom

aly

(1

e4

mole

c cm

-3)

Variability in OH in GEOS-Chem compared other models (no change in anthropogenic emissions)

GEOS-Chem (GEOS-4)MOZART-2 (CTM, NCEP, A. Fiore)

OH

anom

aly

1e4

mole

c cm

-3)

Variability in water vapor in GEOS-Chem & other models

GEOS-Chem (CTM, GEOS-4) MOZECH (GCM, ECMWF)TM4 (CTM, ECMWF) LMDzINCA (GCM, ECMWF)

Q (

kg k

g-1)

Q (

kg k

g-1)

Conclusions

• GEOS-Chem has some success in reproducing the CO anomalies over the 1987-2005 period.

• GEOS-Chem has some success in reproducing the O3 anomalies over the 1987-2005 period, but not in regions influenced by a too large interannual variability in stratospheric-tropospheric exchange.

• Global air mass weighted mean OH shows some rather large interannual variability (with implications for the methane and CO variability and trends).

• Interannual variability in global air mass weighted mean OH are mainly driven by changes in NOx emissions (anthropogenic and lightning) and changes in water vapor.

• Different models indicate very different OH interannual variability over the period 1987-2000. This is probably associated to the simulated changes in water vapor and in lightning emissions. => An issue to be further investigated.