Ray Nassar, Jennifer Logan, Lee Murray, Lin Zhang, Inna Megretskaia

1
Investigating Tropical Tropospheric O3 and CO
during the 2006 El Niño using TES observations
and GEOS-Chem

• Ray Nassar, Jennifer Logan, Lee Murray, Lin
  Zhang, Inna Megretskaia
• Harvard University
• COSPAR, Montreal, 2008 July 13-19

2
El Niño Southern Oscillation (ENSO)
Oceanic-Atmospheric phenomenon warm phase
El Niño cold phase La Niña SST anomalies
and changes in ocean circulation induce changes
in atmospheric convection, precipitation and
chemical composition also Indian Ocean Dipole
(IOD) both ENSO IOD influence Indonesian
region, but warm phases rarely coincide 1963,
1972, 1997 and 2006
Niño 3.4
5N-5S, 170-120W
http//www.cgd.ucar.edu/cas/
ENSO
IOD
Saji et al. (1999) Nature
3
Ozone during the 1997 El Niño
1997 1996 Anomalies
Asymmetric Dipole Anomalies TOMS Tropospheric
Column Ozone (TCO) residual MLS H2O at 215
hPa NOAA Outgoing Longwave Radiation (OLR)
O3
H2O
OLR
http//ggweather.com/enso/nino_regions.gif
Chandra et al. (1998) GRL
4
Modeling O3 during the 1997 El Niño
Hauglustaine, Brasseur Levine (1999)
GRL MOZART model Sudo Takahashi (2001)
GRL CHASER model Trop Column Ozone from TOMS
using Convective Cloud Differential technique
Chandra et al. (2002) JGR GEOS-Chem model TCO
using CCD, TOMS TOMS Aerosol Index for biomass
burning -gt Biomass Burning component -gt
Meteorology/Dynamics/Convection component Duncan
et al. (2003) JGR GEOS-Chem, focus on biomass
burning and lightning
Observations
Model
Chandra et al. (2002) JGR
5
Tropospheric Emission Spectrometer (TES)
High resolution Fourier Transform Spectrometer
(FTS) on Aura, measures nadir IR emission,
launched 2004 July 15, 705 km sun-sync orbit
GEOS-ChemTropospheric Chemical Transport Model
6
Changes to O3 and CO during the 2006 El Niño
TES
TES has well-characterized O3 with 2 DOFS in the
troposphere and simultaneous coincident CO as
proxy for biomass burning Logan et al. (2008) GRL
1) Does GEOS-Chem properly simulate CO and O3
distributions during the El Niño? 2) How do
biomass burning, lightning and transport
contribute to enhanced tropospheric CO and
O3? 3) How can the model simulations be improved?
7
TES GEOS-Chem LT CO October 2006
2ºx2.5º resolution

• TES v02 CO cloud and data quality flag filtered
• Constant TES prior based on 30ºS-30ºN July mean
• Horizontal 2ºx2.5º, Vertical average of 6 TES
  levels LT (825-511 hPa)
• Differences exceed TES CO biases of 10, Luo et
  al. (2007) JGR

8
TES GEOS-Chem LT O3 October 2006
2ºx2.5º resolution

• TES v02 O3 cloud, data quality and emission layer
  flag filtered
• Constant TES prior based on 30ºS-30ºN July mean
• Horizontal 2ºx2.5º, Vertical average of 6 TES
  levels LT (825-511 hPa)
• Differences exceed TES O3 biases of 3-10 ppb,
  Nassar et al. (2008) JGR

9
20062005 CO Differences
GEOS-Chem
TES Observations
October
November
December
10
Biomass Burning impact on CO
GEOS-Chem GFEDv2 2005 2006
GEOS-Chem GFEDv2 2005 both years
TES Observations
October
November
December
GFEDv2 Global Fire Emissions Database (version
2) 8-day temporal resolution
11
20062005 O3 Differences
GEOS-Chem GFEDv2 2005 2006
GEOS-Chem GFEDv2 2005 both years
TES Observations
October
November
December
12
Evolution of Indonesian CO plume
ppb
13
CO and O3 Lower Troposphere (LT) Timeseries
TES
TES
GEOS-Chem wAK
TES corrected
GEOS-Chem wAK
GFEDv2
Subtracted 6 ppbv from TES O3 to account for bias
determined in validation (Nassar et al., 2008 JGR)
14
Global Fire Emission Database v2 Methodology
1) MODIS fire counts (8-day) for timing and
spatial distribution
2) Emission factors for each land type (savanna,
tropical forest, temperate forest) and each
chemical species
15
Updated 2006 Lower Trop Timeseries
GFEDv2
November GFEDv2 emissions increased 3x to account
for smoldering peat fires
16
GEOS-Chem Lightning Imaging Sensor (LIS)
NOx from lightning reacts with CO or hydrocarbons
to form tropospheric O3
LIS Observations 2006-2005
GEOS-Chem 2006 Lightning Flashrate
GEOS-Chem 2006-2005
October
November
December
Note Flashrates below a given absolute threshold
were omitted for differences
Hamid et al. (2001) GRL, discuss lightning
enhancement over Indonesia during 1997 El Nino
17
Updated 2006 Lower Trop Timeseries
Scaling model lightning to LIS observations
improves the O3 timeseries but discrepancy remains
18
TES GEOS-4 UT H2O comparison with O3 anomalies
19
Outgoing Longwave Radiation (OLR) from NOAA and
GEOS-4
NOAA Indonesian OLR anomaly OLRYYYY
OLRclimatology From Australian Bureau of
Meteorology http//www.bom.gov.au/bmrc/clfor/cfsta
ff/matw/maproom/index.htm
High OLR Low Convection
NOAA OLR 2006-2005
GEOS-4 OLR 2006-2005
W/m2
W/m2
20
Summary and Conclusions

• GEOS-Chem can simulate the main CO and O3
  features of the 2006 El Niño
• Biomass burning, lightning and transport are all
  important contributors to enhanced tropospheric
  O3 during El Niño
• GFEDv2 must account for CO from smoldering fires
• GEOS-Chem should move away from climatological
  approach to lightning
• Improvements to GEOS meteorological fields such
  as H2O and deep convection fields will result in
  better simulations of atmospheric composition

Acknowledgments Work was funded by a NASA grant
to Harvard University