Transcript Review: Constraining global isoprene emissions with GOME

Review:
Constraining global isoprene
emissions with GOME formaldehyde
column measurements
Shim et al.
Luz Teresa Padró
Wei-Chun Hsieh
Zhijun Zhao
Isoprene
lifetime ~ 1- 2 hr

Dominant Volatile Organic Compound (VOC) globally

Play important role in oxidant chemistry in troposphere

Non-Methane Hydrogen Carbon

VOC’s possibility of climate change
Isoprene II

Biogenic emission: vegetation

Emissions are favored by:





Vegetation types
Light intensity
Temperature
Leaf area index (LAI)
Oxidizes with OH and O3 to produce formaldehyde

OH oxidation occurs faster and has higher yield
Isoprene Oxidation
Formaldehyde (HCHO)
lifetime ~ 1.5 days




High-Yield Product of isoprene oxidation
Source: methane
Sink: photolysis and atmospheric OH
Tracer for isoprene emissions
HCHO Column Observations

Obtained by the Global Ozone
Monitoring Experiment (GOME)


September 1996 – August 1997
Contributions to HCHO



10 biogenic sources
Biomass burning
Industrial sources
Modeling Methods

A priori (forward model)


A posteriori (inverse modeling)


Uses global GEOS-CHEM chemical transport model
Uses GOME measurements of formaldehyde to do
inverse modeling to obtain isoprene emissions
GEIA

Shows annual global isoprene distribution for 1990
inventory
GOME1 HCHO Column Measurements
and the Uncertainties



Affected by the South Atlantic Anomaly
(SAA) do not include this region in
inverse modeling
AMFs (air mass factors) : covert slant
columns to vertical columns
AMF uncertainties due to uncertainties
in UV albedo, vertical distribution of
HCHO, and aerosols
1 Global Ozone Monitoring Experiment
Model






GEOS-CHEM
Global 3-D chemical transport model
Comprehensive tropospheric O3-NOx-VOC
chemical mechanism
Oxidation mechanisms of 6 VOCs (ethane,
propane, lumped >C3 alkanes, lumped >C2
alkenes, isoprene, and terpenes)
4x5 a maximum of 15 ecosystem types (area
fraction, base emission)
Seasonality (light intensity, temperature, and
LAI)
Inverse Modeling





Only consider high signal-to-noise ratios in
GOME measurements
Criteria define regions for inverse modeling
Daily GOME HCHO slant columns are > 4
d(1.6*1016 molecules cm-2)
Observations satisfy more than one season
North America (eastern U.S.), Europe
(western Europe), East Asia, India, Southeast
Asia, South America (Amazon), Africa,
Australia
Inverse Modeling Regions
Inverse Modeling II



Using monthly GOME measurements
with GEOS-CHEM as the forward model
to estimate the source parameters of
HCHO (state vector)
Observation vector y and state vector x
y=Kx+e
Global Distribution of the
Vegetation Groups
Tropical rain forest(V1), grass/shrub(V2), savanna(V3), tropical
seasonal forest& thorn woods(V4), temperate mixed& temperate
deciduous(V5), agricultural lands(V6), dry evergreen& crop/woods
(warm) (V7), regrowing woods(V8), drought deciduous(V9), the rest
of ecosystems(V10)
Result Analysis
Isoprene Emission
Result Analysis
Formaldehyde
Column concentration
Result Analysis
Regional isoprene emission (N.America, Europe, E.Aisa, India)
Result Analysis
Regional isoprene emission (S.Asia, S.America, Africa, Australia)
Result Analysis
Effect of the Isoprene emission change
On global OH concentration
Decreased by 10.8%
1.Convective transport
2.NOx reduce
Emission
over N.
America
Effect of the Isoprene emission change
On global NOx concentration
Formation of PAN
Summary



1.The a priori simulation greatly
underestimates global HCHO columns and the
a posteriori results show higher isoprene and
biomass burning emission.
2.The a posteriori estimate a 50% larger
annual isoprene emission than the a priori,
decreasing global OH by 10.8%.
3.The a posteriori global isoprene annual
emission are higher at mid latitude and lower
in tropics compared to the GEIA inventory.
Thank you!