Transcript 2005 SOLAS Summer School Introduction to Marine Aerosols Eric S. Saltzman Earth System Science Univ.

2005 SOLAS Summer School
Introduction to Marine Aerosols
Eric S. Saltzman
Earth System Science
Univ. of CA, Irvine
What driving marine aerosol research?
•
geochemical cycles
– metals, nutrients, organics
•
acidification (sulfur, nitrogen)
•
climate change
– direct/indirect effects
– aerosol optical properties, aerosol/cloud interactions
•
nutrients
– N deposition
– coastal, HNLC
– desert dust and iron deposition
•
human health
– air quality, airborne pathogen transport
Earth’s energy balance...
Aerosol effects on radiation budget radiation
Global mean radiative
forcing of climate
for year 2000 relative to
1750 (IPCC)
Direct effects (cloud-free):
•scatter  cooling
•absorption  heating
Indirect effects (clouds):
•more (but smaller) droplets  scatter (Twomey)
•more droplets  longer cloud lifetime (Albrecht)
•absorption  heating  evaporates clouds
The life cycle of marine aerosols...
•What are marine aerosols made of ?
•Where do they come from?
•How long do they stay in the atmosphere? How are they removed?
•How do they evolve while in the atmosphere?
•How do they interact with the climate system?
•forcing
•feedback
Note on observational constraints:
intensive – many param., poor coverage
network – good coverage, few param.
satellite – excellent coverage, limited interpretability...
Things we “know we know”... (or don’t)...
“...there are known knowns. These are things we
know that we know. We also know there are
known unknowns; that is to say we know there
are some things we do not know. But there are
also unknown unknowns. These are things we
don't know we don't know.”
Donald Rumsfeld
Unknown knowns...?
• “nucleation is rare”
• “sulfate is the source of CCN (cloud condensation nucleii) over
the oceans”
• “sea spray doesn’t make submicron aerosols"
Terminology...
•aerosol - a dispersion of solid and liquid particles suspended in gas (air).
note: in common practice, “aerosol” is used to refer to the particles only!
•primary aerosol - emitted directly into the atmosphere.
•Saharan dust, sea spray, pollen, plant waxes, soot
•secondary aerosol - created by nucleation of new particles, aggregation of
existing particles, or growth of preexisting particles from gas phase molecules
(gas to particle conversion).
•either type → natural, anthropogenic, or both
•internal vs. external mixtures – in an internally mixed aerosol all particles have
the same composition
•CN – condensation nucleii - aerosols that scatter light at very high
supersaturation levels
•CCN – cloud condensation nucleii - aerosols that scatter light at very low
superaturation levels
How much aerosol is there?
typically ~10’s 100’s of ug/m3 (air density ~1kg/m3)
Aerosol size distributions...
number
distribution
surface area
distribution
volume
distribution
(Seinfeld and Pandis)
the log-normal aerosol size distribution...
number
distribution
surface area
distribution
Aitken mode
accumulation
mode
coarse
mode
volume
distribution
(Seinfeld and Pandis)
The aerosol modes...
•
•
•
Aitken mode – 0.01-0.1 um
accumulation mode – 0.1-1 um
coarse mode - >1 um
and sometimes, the elusive
•
nucleation mode <0.01 um
(C. Leck)
Humidity and aerosol size...
• hygroscopic aerosols
grow/shrink with RH (with
hysteresis!)
deliquescence
efflorescence
• aerosol size strongly affects light
scattering cross-section
a process-oriented view of the size distribution ...
... reflecting competition between production/transformation/removal
no ultrafines here...
at the time, there was
no instrumentation to
detect them
why “accumulation” mode?
impaction, settling
diffusion,
coagulation
removal mechanisms... gravitational settling
coarse particles
• 10 um particle  1000 cm hr-1
• 1 um particle  10 cm hr-1
Diffusion ...
You can estimate the distance a
particle will diffuse in a given time
from the equation:
distance (cm)= Dt
where D is the diffusion
coefficient
Diameter (μm)
.001
.01
.1
1
10
Distance diffused in 1 s (cm)
0.2
0.02
.002
.0004
.0001
fine particles
The chemical perspective ... a chemical size distribution
2. sulfate and organics
dominate the accumulation
mode, but there’s a
surprising amount of seasalt
Mass
1. chemical size distributions
resemble mass, not number
3. there are a lot of unidentified
organics
4. the coarse mode has the
expected mechanically
generated aerosols, but also
nitrate and sometimes
sulfate
(C. Leck)
Marine aerosol system...
cloud processing
nucleation
Mineral Dust
Dust (mineral aerosols)
diameter size: 2-300 µm
main material: sand, silt, clay
includes essential trace metals such as Fe
consists of insoluble and soluble fractions
Seasalt aerosols...
wind bubbles spray
whitecap coverage W α U3+!
seasalt production via
bubble bursting...
• film drops (many, small, drags
along surface organics)
• jet drops (fewer, larger, bulk)
• spume drops (larger still)
Seasalt aerosols...
Film Drops
Jet Drops
Number
20 nm
80 nm
1 m
10 m
Particle Diameter
(Dry)
(from C. Leck)
Seasalt – number and mass as a function
of surface wind speed
seasalt particle number
seasalt mass
(compiled by Lewis and Schwartz, 2004)
Seasalt – mass fraction as a function of size
(Lewis and Schwartz, 2004)
Marine aerosol mass fractions...
The sulfur story (in brief) ...
•
•
emissions: fossil fuel SO2, volcanic SO2, oceanic DMS
Atmospheric Reaction Pathways for Dimethylsulfide
DMS oxidation ... gas phase ... complex!
OH
CH3SCH3
OH
O2
O
O
CH3SCH3
OH
O
CH3SCH3
O2
CH3SCH3
O
OH
dimethylsulfone
CH3SCH3
HO2 , CH 2 O
CH3SO 2
CH3SO 3
CH3SO 3 H
methanesulfonic acid
CH3SOH
.
CH3SO
.
.
SO
OH
SO 2
+ CH 3
HO2
H O
SO 3 2
H2 SO4
sulfuric acid
O3 , NO2
OH
H2 O
. O2
CH3SCH2
. NO
CH3SCH2 OO
HO2
NO3
ONO 2
CH3SCH3
O2
CH3SCH2 O
. M
CH3S
+ CH 2 O
CH3SCH2 OOH
(mod. from Yin et al., 1990)
Happily, SO2 oxidation in the gas phase is simple...
M
SO2  OH 
 HOSO2
M
HOSO2  O2 
 HO2  SO3
SO3  H 2O  H 2O 
 H 2SO4  H 2O
but most SO2 oxidation occurs in the aqueous phase...
some basics...
SO ( g )  SO ( aq )
2
2
pK1 ~ 4
SO2  H 2O 
 HSO3  H 
pK 2 ~ 8
HSO3  H 2O 
 SO3  H 
heterogeneous oxidation of SO2
• in-cloud oxidation
– weakly buffered, pH ~4
– oxidation by H2O2
– growth of CN, split Aitken mode
• seasalt aerosols
–
–
–
–
strongly buffered by carbonate system
rapid oxidation by O3
slower oxidation by H2O2 (also OH, halogen radicals...)
growth of existing particles, inhibits nucleation of new
particles
(Chameides and Stelson, 1992)
the single-particle view...
Murphy et al., 1998; Buseck and Posfai, 1999
Organic aerosols - burning
soot – “elemental carbon”
formed in flames
little spectral abs. dependence
carbon-only
“brown carbon”:
sugars
alcohols
aromatics
di/tri acids
ketoacids
hydroxyacids
Marine-derived organic aerosols...
(O’Dowd et al., 2004)
A surprise from the Arctic sea ice...
Organic cluster
Virus like
Diatom
Organic film
Bacterium
Bacterium
C. Leck
More organics in marine
aerosols ...
SEM micrograph of an Arctic marine aerosol
sample. Single particles (diameter 1 μm) and
groups of particles are lying between the fibers
of the filter. About half of the particles are
coated by an organic layer.
Negative TOF-SIMS spectra of marine
aerosol.
C15
mass 255  C15H31COOH (palmitic acid)
C14 C15
C17 C18
Tervahattu et al., 2002
Iodine and nucleation...
• coastal I emissions
 I2O5
• macroalgae
• CH2I2, I2 ?
• broader importance?
O’Dowd et al., 2002a,b
Multi-year time series of bulk aerosol
chemistry at island stations...
eg. Prospero et al., 2003
Modeled aerosol sources...
annual average sources
(kg km-2 hr-1)
IPCC, 2001
Aerosol properties
from space...
•mineral dust
•urban pollution
•biomass burning
•aerosol size info
Data from POLDER-1
(IPCC, Deuze et al., 1999)
Understanding forcing/feedback requires realistic
aerosols in climate models...
What’s left to do...?
• in situ aerosol/cloud radiation experiments!
•
•
•
•
•
•
•
shape – non-sphericity
mixing state
“life cycle” of marine aerosol organics, marine microlayer...
gas phase  aerosol chemistry (sulfur, iodine, organics)
aerosol  gas phase chemistry (organics, halogens)
depositional fluxes
coupled aerosol/chemistry/climate models
• marine aerosols are woefully undersampled!
The end.