Embed Size (px)
Citation preview
Atmospheric AerosolBasics
AOSC 637
RUSSELL R. DICKERSON
Atmospheric Aerosols
Bibliography
Seinfeld & Pandis, Atmospheric Chemistry and Physics, Chapt. 7-13
Finlayson-Pitts & Pitts, Chemistry of the Upper and Lower Atmosphere, Chapt. 9.
Classic papers: Prospero et al. Rev. Geophys. Space Phys., 1607, 1983; Charlson et al. Nature 1987; Charlson et al., Science, 1992.
Recent Papers: Ramanathan et al., Science, 2001; Andreae and Crutzen, Science, 1997; Dickerson et al., Science 1997; Jickells et al., Global Iron Connections Between Desert Dust, Ocean Biogeochemistry and Climate, Science, 308 67-71, 2005.
EPA REGIONAL HAZE RULE: FEDERAL CLASS I AREAS TO RETURN TO “NATURAL” VISIBILITY LEVELS BY 2064
•
Acadia National Park
clean day moderately polluted day
http://www.hazecam.net/
…will require essentially total elimination of anthropogenic aerosols!
INTRODUCTION: Particles are one of the most important and certainly the most visible aspects of air pollution. The effects span the areas of health (1% increase in mortality per 10 μg m-3); acid rain, visibility degradation, radiation and photochemistry and cloud microphysics changes (and thus climate changes), and the Antarctic ozone hole. For a view into the "bad old days" see Killer Smog by William Wise.
NOMENCLATURE:Particle refers to a solid or liquid, larger than a molecule, diameter > 0.01 μm, but small enough to remain in the atmosphere for a reasonable time, diameter < 100 μm.
Particulate is an adjective, in spite of what EPA tries to say.
Aerosol is a suspension of particles in a gas.
Particles, like gases, are characterized by chemical content, usually expressed in g m-3, but unlike gases, particles also have a characteristic size. We may want to start discussion the characteristics of atmospheric aerosols by addressing the question "What is the mean diameter of the particles?" The answer to this question changes with your point of view.
A. Size Number Distribution
If your concern is the mass of some pollutant that is being transported through the air for biogeochemical cycles, then you want to know the mean diameter of the particles with the mass or volume. In other words, "What size particles carry the most mass?”
If your concern loss of visibility then you want to know the diameter of the particles that have the largest cross section or surface area. In other words, "What size particles cover the largest surface area?"
If your concern is cloud formation or microphysics then you want to know the range of diameters with the largest number of particles. In other words, "What is the size of the most abundant particles?"
If your concern is human health then you need to know about both the mass and number of the particles, because only a certain size particle can enter the lungs.
Here we define the number distribution function, fn (Dp), and the number of particles with diameter between Dp and Dp + dDp in a cm3 of air as follows:
fn(Dp) dDp (particles cm-3/m) The total number of particles, N, is given by the following integral (everywhere we integrate from 0 to infinite diameters):
N = fn(Dp) dDp (particles cm-3 )
We can define a surface area distribution function, fs(Dp), for spherical particles as
follows:
fs(Dp)dDp = Dp2 fn(Dp ) (m2 m-1
cm-3 )
This gives the surface area of particles in a size range per cm3 of air. The total surface area of the particles, S, is given by the integral over all diameters:
S = fs(Dp) dDp = Dp2 fn (Dp) dDp (m2 cm-3)
Likewise the volume distribution function and the total volume:
fv (Dp) dDp = {/6} Dp3 fn (Dp) (m3 m-1 cm-3 )
V = fv(Dp) dDp = /6 Dp3 fn(Dp) dDp (m3 cm-3)
The distributions based on log Dp can be defined in a similar manner, where
n(log Dp)dlogDp is the number of particles in one cm3 with diameter from Dp to Dp + log Dp.
The total number is:
N = n(log Dp) d(logDp) (particles cm-3 )
The normalized distribution functions based on log Dp for surface area and volume are
similar. For the differential number of particles between Dp and Dp + dDp we use the
notation dN, and likewise dS and dV, we can represent the size distribution functions as -
n (log Dp) = {dN} / {N dlogDp }
ns (log Dp) = {dS} / {S dlogDp }
nv (log Dp) = {dV} / {V dlogDp }
This is the common notation for expressing the variation in particle number, surface area or volume with the log of the diameter.
B. Chemical Composition
The bimodal nature of the size-number distribution of atmospheric particles suggests at least two distinct mechanisms of formation, and the chemical composition of the particles reflects their origins.
Fine particles have a diameter smaller than about 2.5 m, and are produced by the condensation of vapors, accumulation, and coagulation. They have a chemical composition that reflects the condensable trace gases in the atmosphere: SO2, NH3, HNO3,
VOC’s, and H2O. The chemical composition is water with SO4-2, NO3
-, NH 4+, Pb, Cl-, Br-,
C(soot), and organic matter; where biomass burning is prevalent, K+.
Coarse Particles have a diameter greater than about 2.5 m, are produced by mechanical weathering of surface materials. Their lifetimes, controlled by fallout and washout, are generally short. The composition of particles in this size range reflects that of the earth's surface - silicate (SiO2), iron and aluminum oxides, CaCO3 and MgCO3; over the
oceans , NaCl.
ORIGIN OF THE ATMOSPHERIC AEROSOL
Soil dustSea salt
Aerosol: dispersed condensed matter suspended in a gasSize range: 0.001 m (molecular cluster) to 100 m (small raindrop)
Environmental importance: health (respiration), visibility, radiative balance,cloud formation, heterogeneous reactions, delivery of nutrients…
AEROSOL NUCLEATION
# molecules 1 2 3 4
G
cluster sizeCritical cluster size
Surface tension effect
Thermo driving force
Atmospheric AerosolsAtmospheric Aerosols
Aerosol DistributionsAerosol Distributions
Number
• cloud formation
Surface
• visibility
Volume
• mass
Mass & Number
• human health
TYPICAL U.S. AEROSOL SIZE DISTRIBUTIONS
Freshurban
Agedurban
rural
remote
Warneck [1999]
SAMPLE AEROSOL SIZE DISTRIBUTION (MARINE AIR)
Seasalt
Sulfate(natural)
COMPOSITION OF PM2.5 (NARSTO PM ASSESSMENT)
Dust
C. Optical Properties and Visibility
The optics of aerosol science follow the most rigorous physics, but traditionally defined visibility is the distance at which a large dark object, such as a hill or a barn can just be seen. A more quantitative definition can be obtained by considering the change in intensity of light reflecting off an object as a function of the scattering of light by the atmosphere.
I/I = e(-bX)
Where I is the intensity of light, b (or bext) is the extinction coefficient with units of m-1, and
X is the distance in m. The limit to visibility for the human eye is a 2% change in intensity relative to the background or:
I/I = 0.02
Radiation and fine particles
Atmospheric VisibilityAtmospheric Visibility (absorption & scattering)
1.Residual
2.Scattered away
3.Scattered into
4.Airlight
The extinction coefficient represents the sum of the extinctions from gases and particles, each of which can in turn be divided into extinction due to absorption or scattering.
bext = bgas + bparticles
bext = babs + bscatt
babs (gases) = Beer's Law absorption
bscatt (gases) = Rayleigh Scattering
babs (particles) = Usually < 10% of extinction
bscatt (particles) = Mie Scattering = (bsp)
The ultimate limit to visibility in a very clean atmosphere is Rayleigh scattering, but Mie scattering usually dominates. The range of bsp is 10-5
m -1 to 10-3 m
-1.
Aerosol optical depth (AOD), is a measure of the extinction caused by scattering and absorption:
I/I0 = exp (-bext x) = bext x
= bext(z) dz (unitless)
Single scattering albedo, , is a measure of the fraction of aerosol extinction caused by scattering:
= bsp/(bsp + bap)
Angstrom exponent, , is a measure of the size number distribution. the name of the exponent in the formula that describes the dependency of the aerosol optical thickness on wavelength. Fresh urban aerosols for dust storms
Optical Properties of Small Particles
m = n + ikm = complex index of refraction
n = scattering (real part)
k = absorption (imaginary part)
The real part of the index of refraction is only a weak function of wavelength, while the imaginary part, ik, depends strongly on wavelength.
Refractive indices of aerosol particles at = 589 nm
m = n + ik
Substance n kWater 1.333 10-8
Ice 1.309 10-8
NaCl 1.544 0
H2SO4 1.426 0
NH4HSO4 1.473 0
(NH4)2SO4 1.521 0
SiO2 1.55 0
Black Carbon (soot) 1.96 0.66
Mineral dust ~1.53 ~0.006
The scattering cross section is the product of the mass loading, and the surface area per unit mass; note the ln of 0.02 is about -3.9, thus:
Visibility ≈ 3.9(bsp)-1
bsp = Sm
Where
bsp is the scattering coefficient in units of m-1
m is the mass loading in units of g m-3
S is the surface area per unit mass in units of m2g-1
For sulfate particles, S is about 3.2 m2 g-1 where the humidity is less than about 70%; for other materials it can be greater.
Visibility = 3.9/(3.2 m)
= 1.2 /(m)
Example: Visibility improvement during the 2003 North American Blackout
Normal conditions over Eastern US during an air pollution episode:
bsp ≈ 120 Mm-1 = 1.2 x 10-4 m-1 at 550 nm
bap = 0.8 x 10-5 m-1
bext = 1.28 x 10-4 m-1
Visual Range ≈ 3.9/bext = 30 km
During blackout
bsp = 40 Mm-1 = 0.4 x 10-4 m-1
bap = 1.2 x 10-5 m-1
bext = 0.52 x 10-4 m-1
Visual Range = 3.9/bext = 75 km
Example: Visibility improvement during the 2003 North American Blackout
Single scattering albedo, , normal = 1.20/1.28 = 0.94
Blackout = 0.4/0.52 = 0.77
With the sulfate from power plants missing, and the soot from diesel engines remaining the visual range is up, but the single scattering albedo is down. Ozone production inhibited.
See: Marufu et al., Geophys Res. Lett., 2004.
Extinction Coefficient as a PM2.5 Surrogate
PM2.5 = 7.6 g/m3 PM2.5 = 21.7 g/m3
PM2.5 = 65.3 g/m3
Glacier National Park images are adapted from Malm, An Introduction to Visibility (1999) http://webcam.srs.fs.fed.us/intropdf.htm
ANNUAL MEAN PARTICULATE MATTER (PM) CONCENTRATIONS AT U.S. SITES, 1995-2000
NARSTO PM Assessment, 2003
PM10 (particles < 10 m) PM2.5 (particles < 2.5 m)
Red circles indicate violations of national air quality standard:50 g m-3 for PM10 15 g m-3 for PM2.5
Aircraft obs. →
AOD ~ 0.1
←Model calcs.
NO2 + hυ (+O2) → NO + O3
AEROSOL OPTICAL DEPTH (GLOBAL MODEL)
Annual mean
AEROSOL OBSERVATIONS FROM SPACE
Biomass fire haze in central America (4/30/03)
Fire locationsin red
Modis.gsfc.nasa.gov
BLACK CARBON EMISSIONS
Chin et al. [2000]
DIESEL
DOMESTICCOAL BURNING
BIOMASSBURNING
RADIATIVE FORCING OF CLIMATE, 1750-PRESENT
“Kyoto also failed to address two major pollutants that have an impact on warming: black soot and tropospheric ozone. Both are proven health hazards. Reducing both would not only address climate change, but also dramatically improve people's health.” (George W. Bush, June 11 2001 Rose Garden speech)
IPCC [2001]
Climate radiative forcing – revised 2002.
What does the UN Inter-government Panel on Climate Change say?
ASIAN DUST INFLUENCE IN UNITED STATESDust observations from U.S. IMPROVE network
April 16, 2001Asian dust in western U.S.
April 22, 2001Asian dust in southeastern U.S.
GlenCanyon, AZ
Clear day April 16, 2001: Asian dust!
0 2 4 6 8g m-3
LONGITUDE
AL
TIT
UD
E (
km)
100E 150E 150W 100W
TRANSPACIFIC TRANSPORT OF ASIAN DUST PLUMES
Subsidenceover western U.S.
Source region(inner Asia)
Asian plumes over Pacific
GEOS-CHEM Longitude cross-section at 40N, 16 April, 2001
0
5
10
ASIA UNITED STATES
T.D. Fairlie, Harvard
Trends NE China versus NE US
50
100
150
200
250
300
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Ele
ctr
icit
y G
en
era
tio
n o
r S
O2
Em
issio
ns
0.2
0.4
0.6
0.8
1
1.2
1.4
SO
2 C
olu
mn
Co
nte
nt
(No
rmali
zed
)
China Elect. Generation USEPA Emissions
Satellite China Satellite USA
GOME & SCIA NO2: US Power Plant NOx reductions
• After 2000, clear decrease (> 30%) in NO2 in Ohio River Valley area. (SIP-Call)
• Little change over East Coast
1996 2000 2005
NOx Emissions Trends and Satellite NO2 Data, 1996 - 2005
R2 = 0.84
2.00
2.50
3.00
3.50
4.00
4.50
5.00
0 100000 200000 300000 400000 500000 600000
Power Plant Emissions in Ohio River Valley (tons/season)
GO
ME
/SC
IA c
olu
mn
NO
2
Point Source
Emissions
1999
Frost et al.,
JGR, 2006
Emissions rates (CEMS) for the third quarter (July to September) in 1999 (open squares) and 2003 (solid circles). Average emission rates at a particular plant are the ratio of total quarterly mass emissions (in pounds) divided by total quarterly heat input (in million British Thermal Units).
Frost et al. JGR, 2006.
Aerosols in the Atmosphere
Abundance and size
• Aerosol concentration is highly variable in space and time. Concentrations are usually highest near the ground and near sources.
• A concentration of 105 cm-3 is typical of polluted air near the ground, but values may range from 2 orders of magnitude higher in very polluted regions to several lower in very clean air.
• Radii range from ~ 10-7 cm for the for small ions to more than 10 µm (10-3
cm) for the largest salt and dust particles.
• Small ions play almost no role in atmospheric condensation because of the very high supersaturations required for condensation.
• The largest particles, however, are only able to remain airborne for a limited time
Aerosol Size Naming Convention
Usually divided into three size groups ( D - diameter)
1. Aitken Nuclei 2 x 10-3 µm < D 2 µm
2. Large Nuclei 2 µm < D 2.5 µm
(also called the accumulation mode)
3. Giant Nuclei D > 2.5 µm
Other Naming Convention
• Nucleation mode D ≤ 0.1 µm
• Accumulation or coagulation mode 0.1 µm < D ~ 1 µm
Thought to be most important in natural cloud formation
• Coarse Particle Mode D ~ 1 µm
Origins of Atmospheric Aerosols
1. Condensation and sublimation of of vapors and the formation of smokes in natural and man-made combustion.
2. Reactions between trace gases in the atmosphere through the action of heat, radiation, or humidity.
3. The mechanical disruption and dispersal of matter at the earth’s surface, either as sea spray over the oceans, or as mineral dusts over the continents.
4. Coagulation of nuclei which tends to produce larger particles of mixed constitution
Aerosol Makeup
• Typical substances formed in large quantities by condensation following combustion include ashes, soot, tar products, oils as well as sulfuric acid and sulfates. These particles are primarily within the range of Aitken nuclei.
• Mechanical disintegration, by wind and water, of rocks and soil produces particles with diameters > 0.2 µm. These fall primarily in the large nuclei range.
• According to Jaenicke (Science, 308 p. 73, 2005) about 25% of the number of particles with diameter greater than 0.2 µm are biogenic. (remains to be verified).
Aerosol Makeup - continued
• Chemical reactions between nitrogen, oxygen, water vapor and various trace gases (e.g., sulfur dioxide, chlorine, ammonia, ozone, and oxides of nitrogen) primarily produce particles in the Aitken and Large range.
Examples
Formation of ammonium chloride from NH3 and HCl
Oxidation of SO2 to H2SO4
Reaction of sulfur dioxide, ammonia, and water to produce ammonium sulfate particles.
Production of higher oxides of nitrogen through the action of heat, ozone or ultraviolet radiation
Cloud Condensation Nuclei - CCN
• Comprises a small fraction of the total aerosol population
• Sea salt is the predominant constituent of CCN with D > 1µm
• For 0.1 µm < D < 1 µm, the main component is thought to be sulfate, which may be present as sulfuric acid, ammonium sulfate, or from phytoplankton produced dimethylsulfide (see Charlson et al., Nature, 326, 655-661).
Activity Spectrum
Let Nc be the number of particles per unit volume that are activated to become cloud droplets.
Data from cloud chamber measurements are often parameterized as
Nc = C (S-1)k
where C and k are parameters that depend on air mass type.
Rogers gives:
Maritime air: 30 < C < 300 cm-3; 0.3 < k < 1
Continental air: 300 < C < 3000 cm-3; 0.2 < k < 2
Thus, for the same saturation ratio, one would expect to find small numbers of CCN per unit volume in maritime air and large numbers per unit volume in continental air.
Long term trend in PM concentrations in Mid Atlantic Region from the IMPROVE and
CASTNET samplers
OC trend – 0.05 g m-3 yr-1
OM trend ~ – 0.08 g m-3 yr-1
1988-2004 EC trend -0.05 g m-3 yr-1
1988-2006 EC trend -0.04 g m-3 yr-1
DRI doesn’t trust the data from 2000-2004.
Washington, DC Nitrate
IMPROVE every 3rd day observations (1988-2006) linear trend and 30-point running mean.
0
1
2
3
4
5
6
7
8
9
10
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
Nitr
ate
(ug/
m3 )
1988-2006 NO3- trend -0.02 g m-3 yr-1
Washington, DC IMPROVE 1989-2006 Sulfate
0
5
10
15
20
25
30
35
J an-
89
Jul-8
9
J an-
90
Jul-9
0
J an-
91
Jul-9
1
J an-
92
Jul-9
2
J an-
93
Jul-9
3
J an-
94
Jul-9
4
J an-
95
Jul-9
5
J an-
96
Jul-9
6
J an-
97
Jul-9
7
J an-
98
Jul-9
8
J an-
99
Jul-9
9
J an-
00
Jul-0
0
J an-
01
Jul-0
1
J an-
02
Jul-0
2
J an-
03
Jul-0
3
J an-
04
Jul-0
4
J an-
05
Jul-0
5
J an-
06
Jul-0
6
Su
lfat
e (u
g/m
3)
Trend – 0.094 g m-3 yr-1
1989-2006 Dust – 0.022 g m-3 yr-1
1989-2006 PM2.5 – 0.22 g m-3 yr-1
Seasonal cycle in PM concentrations in Mid Atlantic Region from the IMPROVE and
CASTNET samplers
