Dylan Millet

1
Distribution of formaldehyde over North
America Implications for satellite retrievals
and mapping of precursor emissions
Dylan Millet D.J. Jacob, D.R. Blake, K. Chance,
A. Fried, B.G. Heikes, R.C. Hudman, T.P. Kurosu,
H.B. Singh, S. Turquety, S. Wu
ACCESS VIII Yellowstone National Park September
1-4, 2005
2
Can we use WHCHO as a proxy for VOC emissions?

• Volatile organic compunds (VOCs)
• Precursors of ozone
• Precursors of organic aerosol
• Sinks of OH

3
HCHO slant columns measured by GOME (K. Chance,
T.P. Kurosu et al.)

• Key Questions
• Measurement
• 1) What is the uncertainty and bias in HCHO
  columns measured from satellites?
• Interpretation
• 2) What are the main precursors contributing to
  HCHO columns and variability over North America?
• 3) What are the implications for retrieving VOC
  emissions from space?
• ? Address using aircraft measurements

HCHO slant columns measured by OMI (K. Chance,
T.P. Kurosu et al.)
4
Uncertainty in solar backscatter retrievals of
HCHO

• 1) Fitting uncertainty
• ( 4 x 1015 molec/cm2)
• 2) Relating slant columns to vertical columns
• Air mass factor (AMF)
• AMF depends on
• Atmospheric Scattering
• Cloud effects
• Aerosol effects
• HCHO vertical profile

HCHO
Model
5
AMF approach Calculate measured and modeled
AMFs using aircraft profiles
scattering
HCHO vertical distribution
Viewing geometry SZA
measured vs. modeled HCHO, aerosol

• DC-8 flight tracks
• (ICARTT, summer 2004)

measured vs. satellite clouds
6
AMF results

• Mean 1.2 1.3
• Range 0.11 - 2.42
• Model bias
• -5 over continents (-57?70)
• 13 over ocean (-14?72)
• 25 uncertainty for a single scene
• What drives the variability?

7
Errors in satellite retrievals of HCHO due to
clouds aerosols

• Model HCHO vertical profile
• -4 mean bias over land
• 7 over ocean
• Aerosols
• Increase the AMF (i.e. sensitivity to HCHO) by
  16
• Clouds
• Can ? or ? AMF
• Major source of error
• 2X uncertainty for a single scene
• Recommend cloud cutoff
• AMF error of 25 _at_ 40 cloud fraction

8
Interpreting HCHO columns

• Methane, anthro. biogenic VOCs
• What drives variability in WHCHO?

9
WHCHO sources variability

• Methane OVOCs main HCHO precursors in most of
  the atmosphere
• But variability in column production rate is low
• Satellite LOD / tHCHO
• 5x1011 molec/cm2 /s

Probability
Measured column HCHO production rate
10
Relating HCHO columns to precursor emissions
Column HCHO

• Isoprene dominant source when WHCHO is high
• Variability in WHCHO over N. America driven by
  isoprene

Measured column HCHO production rate
11
HCHO production yield from isoprene
HCHO column mass balance
MOD
MOD
From measured (WHCHO vs. Wi) modeled
(ki/kHCHO) Y 1.61
MOD (flighttrack)
OBS
12
Conclusions

• Uncertainty in HCHO columns measured from space
• Clouds major source of error
• Increase AMF uncertainty by 2x
• ? Uncertainty in satellite HCHO columns due to
  the AMF
• Mean bias -5 over continents
• Uncertainty 25 (1s) for individual scene (less
  for time averages)
• Mapping VOC emissions
• Variability in WHCHO over N. America driven by
  isoprene
• ? Satellite retrievals of WHCHO can be used as a
  proxy for isoprene emissions over N. America
• Estimated average HCHO yield from isoprene
  oxidation 1.61
• Upper end of GEOS-Chem (0.9-1.9) NOx-dependent
  yield

13
(No Transcript)