EPA-600/4-76-056
November 1976
Environmental Monitoring Series
MEASUREMENT OF DRY DEPOSITION OF
FOSSIL FUEL PLANT POLLUTANTS
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1 Environmental Health Effects Research
2 Environmental Protection Technology
3. Ecological Research
4 Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and instrumentation for the identification and quantification of environmental
pollutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
July 1976
MEASUREMENT OF DRY DEPOSITION
OF FOSSIL FUEL PLANT POLLUTANTS
By
J. G. Droppo, D. W. Glover and 0. B. Abbey
Battelle, Pacific Northwest Laboratory
Richland, Washington 99352
C. W. Spicer
Battelle, Columbus Laboratories
Columbus, Ohio 43201
John Cooper
ORTEC
Oak Ridge, Tennessee 37830
Contract No. 68-02-1747
Project Officer
Herbert J. Viebrock
Meteorology and Assessment Division
Environmental Sciences Research Laboratory
Research Triangle Park NC 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, NC 27711
-------
This report has been reviewed by the Environmental Protec-
tion Agency and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Agency, nor does mention of trade names or
commercial product constitute endorsement or recommendation for
use.
-------
CONTENTS
Page
ABSTRACT „ iii
LIST OF FIGURES, j_v
LIST OF TABLES . v
ACKNOWLEDGMENTS vii
SECTIONS
I. CONCLUSIONS 1
II. RECOMMENDATIONS 3
III. INTRODUCTION. 5
IV. DRY DEPOSITION MODELS 7
V. EXPERIMENT DESIGN ....... . . 33
VI. INSTRUMENTATION 44
VII. DATA COLLECTION 60
VIII. RESULTS 65
IX. DISCUSSION 89
X. REFERENCES 94
XI. NOMENCLATURE 1Q4
XII. APPENDICES 106
-------
ABSTRACT
Dry removal of air pollutants from fossil fuel plants is con-
sidered both from a modeling and measurement viewpoint. Litera-
ture on dry deposition rates is summarized. The processes in-
volved in dry deposition are discussed. The need for the
removal rates of specific pollutants is identified; this report
considers SO,,, 0^, N0x, and NO, as well as the total sulfur and
lead content of airborne particles.
A prototype field data acquisition system based on a combina--
tion of flux gradient relationships was developed, assembled,
and tested. The results of the field tests are described.
Significant progress was made in developing the aerosol dry
deposition measurement system. Analysis of the sulfur and
lead content of aerosols on the filters achieved the required
level of relative accuracy for application of the method. Flow
rate inaccuracy precluded deposition rate computation.
The vertical profiles of the gases were reasonably linear when
considered as a function of the logarithm of .height and were
consistent with the assumption of dry removal occurring at the
surface. A deposition velocity and its experimental accuracy
are computed for each field test of gaseous pollutants. These
results show generally wide ranges of deposition velocities.
The capability of the method to predict the removal rates in
certain cases, and its inability in certain other cases is
demonstrated.
Recommendations are made for improvement of the capability of
the field data acquisition systems and for future applications.
111
-------
FIGURES
Page
1. Schematic of Pollutant Concentration Profiles
Downwind of a Stack 34
2. View of Instrumented Tower and Mobile Van with
Cooling Tower Plume in Background 45
3. Nuclepore Filter Mounted on Tower 46
4. Gas Sampling Systems 39
5. Schematic of Instrumentation Location . 57
6. Diagram of Fetch Around Site. Distances are
not to Scale . 61
IV
-------
TABLES
Page
Summary of Deposition Velocities in Literature
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Normalized Wind and Pollutant Profiles (Based on
Data in References 8, 66)
Summary of Field Tests During March 1975 at
Centralia
Total and Normalized Sulfur Aerosol Concentrations
Uncorrected for Flow Rate
Summary of Computed Aerosol Air Concentrations . . .
Flow Rate Adjustment (m s )
Summary of SO Data Results
Least Squares to S00 Data
Summary of NO Profile Results
J X
Summary of NO Profile Results
Summary of Variation Noted in NO Profile Data
at 0909-0948
Least Squares Fit to Ozone Data. ..........
Summary of Meteorological Data for S02 Runs 1-3 . . .
Summary of Meteorological Data for NO Runs 1-3 . . .
Summary of Meteorological Data for NO Runs 4-6. . .
X
Summary of Meteorological Data for NO Runs 1-3 . . .
Summary of Meteorological Data for 0., Runs 1-3 . . .
Summary of Meteorological Data for 0^ Runs 4-6 . . .
Computed Eddy Diffusivities for Momentum K and
and Sensible Heat K
j-o
26
62
67
69
70
70
72
73
73
74
75
76
77
78
79
80
81
fn
v
-------
Page
21. Deposition Computation for SO-. ........... 84
22. Deposition Computation for NO . .... • • 85
X.
23. Deposition Computation for NO ............ 86
24. Deposition Computation for 0-, ............ 87
VI
-------
ACKNOWLEDGMENTS
The authors are grateful to Rose Morrill for use of her pasture
in Centralia for a test site.
The authors wish to thank Dr. J. M. Hales who made many useful
contributions to the progress of the current study. Other
Battelle personnel who contributed significantly to the project
are Dick Lee, Darrell Joseph, and Rodger Woodruff.
-------
SECTION I
CONCLUSIONS
The surface layer profile method has been shown in these
initial tests to have the capability of estimating dry removal
rates of pollutants under actual field conditions.
Careful analysis of the experimental accuracy of the computed
deposition parameters is a valuable tool for identification
of the relative reliance of individual field tests. This
technique clearly demonstrates strengths and limitations of
the system under initial field tests.
The aerosol test restuls are encouraging. Independent
analysis of the sulfur aerosol on the filters demonstrates
that the 1 to 2% relative accuracy in concentration needed
for dry deposition computation has been achieved. This is
a significant breakthrough; it implies that by upgrading the
flow rate determination it will be feasible to compute dry
removal rates for sulfur-bearing aerosols.
The significant gaseous pollutant profiles indicate dry
deposition in a consistent fashion. This is an indication
but not proof that the profile accuracies are sufficient
for resolution of dry removal rates.
The gaseous pollutant test results clearly demonstrate the
importance of considering the specific sources of experimental
error. The computation was found to be limited by either or
both the pollutant gradient and eddy diffusivity accuracy.
The pollutant dry removal rates are expressed as a range of
deposition velocities. The resolution was satisfactory only
in a few cases.
-------
Sulfur dioxide was computed to deposit at a rate comparable to
values in the literature. The results indicate that NO also
deposits at a comparable or lower rate than SO- but the range
is too great to support a definitive conclusion for NO.
-------
SECTION II
RECOMMENDATIONS
Application of the surface layer profile method shows promise
of providing considerable information on removal rates of
common pollutants, and the relative importance of various
processes controlling dry deposition. Modeling efforts should
be aimed at developing methods of treating the interaction of
these variously limiting processes.
The consistency of the pollutant and micrometeorological
profiles was encouraging in the current proof tests of the
system. The expansion of the surface layer pollutant profile
data base in future experiments should provide a clear indica-
tion of the applicability of this approach to determination
of dry deposition rates.
Several experiment design modifications are possible to
maximize the efficiency of the field effort. First, the use
of multiple sites is recommended for future studies. The use
of a single tower site is inefficient as a result of extended
down times during unfavorable conditions. The efficiency is
better as a result of optimum use of field time by sequential
use of the sites as dictated by prevailing conditions such as
winds,turbulence and pollutant sources. Also in this way a
number of combinations of surfaces, conditions, and pollutants
can be studied in a single field effort.
Second, it is recommended that turbulence parameterization
methods be investigated to calibrate each site. The apparent
need to consider each site for meteorological flux computations
was discussed recently by Thorn, et al . Data sets should be
extensive enough to derive a site dependent parameterization of
-------
turbulence. This formulation can be used where inherent
inaccuracy of the turbulence computation alone limits the
removal rate determination.
The overall performance of this method may be improved by
specific changes in the detail of the instrumentation systems.
Further development in the determination of flow rate in the
aerosol sampling system is necessary- Calibrated orifices
should be used to reduce the variations in flow rates between
levels. Inclusion of inline temperature and pressure measure-
ments at the flow meters is also needed to insure sufficiently
accurate determination of the flow rates.
The number of SO,, samples at each level should be increased to
provide better profile accuracy. Multiple monitors should be
incorporated into the system for simultaneous sequencing between
measurement heights.
For future studies of ozone or other highly reactive gases a
roving profiler system with a single line input is recommended
to minimize measurement problems.
Further effort is necessary to determine fluxes on the accuracy
of the NO profiles obtained by the fixed level sequencing
X
method. If this method is not acceptable, then the roving pro-
filer system is considered the best alternative. NO is also a
highly reactive gas and a roving profiler system is recommended.
The overall objective in future research efforts should be to
develop a semi-empirical relationship for the dry removal of
each pollutant. This should be potentially a function of the
pollutant, receptor, meteorological variables, and in certain
cases the concentration of other pollutants.
-------
SECTION III
INTRODUCTION
Concern with the fate of atmospheric pollutants has intensified
interest in removal processes. Transport and diffusion alone
have been found insufficient to model ambient pollutant concen-
trations. Quantitative removal estimates are needed to deter-
mine pollutant budgets on large scales as well as to assess
local impact from single sources. Possible removal mechanisms
include the very slow escape to space from the atmosphere; preci-
pitation and cloud scavenging; chemical transformations; and
dry deposition. The current analysis is concerned with the
dry deposition processes, emphasizing pollutants in the atmo-
spheric plume from a fossil fuel electrical power station.
Both gases and aerosols are considered.
Dry deposition is the direct transfer of a material from the
atmosphere to the "earth's surface". Pollutants deposit on
soil or vegetation or any other material which comprises the
surface of the earth. The adjective dry serves only to exclude
transfer by liquid water droplets. No condition is placed on
the relative dryness or wetness of the surfaces; dry deposition
occurs on deserts as well as oceans.
Dry deposition reduces ambient air pollutant concentrations
and determines the potential for an increase in the pollutant
concentration on or within the receptor. Processes affecting
dry deposition of a specific pollutant include gravitational
settling, transport by atmospheric turbulence, impaction,
concurrent surface fluxes, and chemical reactions. The quanti-
fication of these processes is necessary to evaluate potential
pollutant impacts both in the air and on the receptors. This
information is needed to develop emission control strategy.
-------
The objective of this program is to develop a comprehensive
model which evaluates dry deposition from specific pollutant
plumes. The model development requires both theoretical and
empirical input. Information based on field studies of the
dry deposition of pollutants is relatively scarce. The
following section defines the type of information needed to
further develop dry deposition models. Subsequent sections
describe the development and use of instrumentation, and the
field effort which have provided input to the model.
-------
SECTION IV
DRY DEPOSITION MODELS
The earliest understood aspect of dry deposition was gravita-
tional settling of large particles. Early estimates, of dry
deposition downwind of a stack were based primarily on gravita-
tional settling velocities. However, particles smaller in
diameter were predicted to have extremely low deposition rates
contrary to experimental evidence. Consequently models were
developed which included the effect of atmospheric turbulence
on the deposition of intermediate-size particles, for which
both gravitation and turbulence are important. Smaller particles
still deposited at rates faster than predicted. Molecular
transport processes were used to account for these higher
rates.
Gravitational models are sufficient for large particles. For
particles of an intermediate size gravitational settling can
2
often be important. For example Aylor found that in the low
turbulence region within and under vegetation canopies the
deposition rate of pollens equals the gravitational settling
rate. However, the current study will be limited to particles
whose settling velocities are small compared to typical atmo-
spheric eddy velocities. For these smaller particles gravita-
tional settling will not be significant and the gravitational
term will be included only as a parallel lower limit for dry
deposition.
For gases, gravitational settling rates in the troposphere are
unimportant for the normal magnitude of pollutant concentrations.
The modeling of dry deposition in pollutant transport has been
incorporated into a number of recent efforts. Examples of these
efforts are given in references . The methods of modeling
will be considered later in this section.
-------
Modeling of the dry deposition of smaller particles and gases
has been limited by a lack of data defining the relative
importance of various processes. The following model descrip-
tion provides a conceptual framework, which organizes these
processes for subsequent study based on the synthesis of a
7-14
number of references
The dry deposition of a substance may be a function of many
processes, or one process may dominate all the others. A
typical list of interrelated factors includes:
Atmospheric: Wind speed, turbulence, temperature structure,
moisture, pollutant concentration, concentra-
tion of other pollutants, solar radiation and
long wave radiation.
Receptor: Surface roughness, moisture, physical and
chemical nature, physiological state, membranes,
internal and surface pollutant concentration,
and status of stomata.
Pollutant: Chemical and physical properties; solubility,
density, size, shape, settling velocity.
Dry deposition involves flux by many different transport
mechanisms. These include gravitational settling; atmospheric
eddy transport; and molecular transport in the air near the
surfaces and from the surface into the receptor. These
mechanisms may be grouped into regimes where various processes
dominate the transport. These are 1) the atmospheric regime,
2) the surface air layer regime, and 3) the surface into the
receptor regime.
The atmospheric regime refers to the air layer over the sur-
faces, in which the pollutant transport is primarily by eddy
motions of air. The surface air layer regime refers to the
thin air layer immediately over the receptor, where molecular
-------
transport is important. The surface into the receptor regime
refers to the rate at which pollutants are transported into the
receptor.
The change of a pollutant to another species is a sink that may
be a parallel process in any of the regimes. Chemical reactions
within the atmospheric regime are not normally rapid enough to
be significant when considering the constant flux layer. React-
ions at the surface (such as destruction of ozone) and within
the surface (such as physiological uptake of SO-) can be
limiting in some situations. Not included in this scheme are
components representing potential saturation of various sinks.
For example, the deposition to a plant may be a function of the
concentrations within the plant. For simplicity such factors
are not included in the mathematical model since their inclu-
sion would require retention of time as a variable in the
analysis.
Each regime, is assigned a resistance (r) which defines the
extent to which that regime limits the deposition rate. For
a given situation:
-1- = 1 + -A- (1)
rT " ra + r£ + ri rg '
where the subscripts T, a, £, i and g refer to total, atmo-
spheric, surface air layer, internal (surface into receptor),
and gravitational resistances respectively. Once total resistance
is defined, boundary condition inputs into diffusion models are
possible.
Resistance Terms
Each of the resistance terms on the right of Equation 1 will be
considered separately below.
-------
Atmospheric Resistance. The processes occurring in the atmo-
spheric regime can be described if the transfer processes for
momentum and mass are assumed to be identical. The atmospheric
resistance to mass transfer may be assumed to equal that of
momentum ( T ) ,
r = r = Hf. . . . (2)
a T U*2
where u(z) is the wind speed at height z and u* is the fric-
tion velocity. This is the reciprocal of a relationship
1 5 —19
developed by Chamberlain for a "deposition velocity".
The equation for momentum resistance has been proposed to
apply to all fluxes to surfaces. However, mechanisms of trans
port at the surface for momentum and mass differ. Momentum
is transferred by pressure forces on the receptor, while
mass must be carried to the surface by gravity, impaction,
or molecular transport mechanisms. Equation 2 gives the
total resistance to momentum flux.
Surface Resistance. The dependence of deposition on near-
surface mechanisms comprises the surface resistance term,
r „ . Owen and Thompson defined this resistance, arising from
molecular interaction in the layer just over the surface, in
terms of a nondimensional Stanton number (B ) .
Internal Resistance. For plant tissue, resistance will be a
more complicated function of the plant properties and process-
es. Factors such as membrane pressures and pollutant concen-
trations may limit transport. The opening and closing of
stomata is important for gases. The control of the stomata
openings for certain plants has been shown to depend in a
10
-------
complicated fashion on the ambient air concentration of CO , H,,0
^ £•
and S0~ . Hence, the resistance in plants appears to be the most
difficult resistance to assess, since the properties of the
receptor must be defined.
Rates of Reaction Resistances. The limiting processes may be
different for gases such as ozone which are highly reactive.
Destruction of the gas will occur not only at the surfaces but
also in areas shaded by the surfaces. This amounts to a re-
duced resistance with some material being lost directly from
the free atmosphere.
Gravitational Resistance. The gravitational resistance term
(r ) gives a lower limit for the rate of deposition. The
resistance is the inverse of the appropriate settling velocity
vs'
rg = ^ -
Effective Surface Resistance. An effective surface resistance,
r , is to be used to include the processes other than atmo-
s
spheric and gravitational. Assuming the gravitational resis-
tance is large enough to ignore,
rT ra + rs
(4)
Hence, for computational purposes the effective surface resis
tance is assumed to be the incremental resistance over the
atmospheric resistance to account for the total resistance.
Deposition Velocity.
The simplest expression for dry deposition of either gases or
aerosols is based on an assumption relating the ambient air
11
-------
concentrations (x) and the flux through a surface at a given
height (G) . At the surface this is the rate of deposition,
which is assumed to be directly proportional to the ambient
air concentration at a defined height over the surface (z) .
The constant of proportionality has the units of length per
time and is referred to as a "deposition velocity", (V^)
The equation is:
G = Vd/xz (5)
with the height traditionally assumed to be 1 meter. This
assumes the constant flux layer is one meter deep.
The aim of the modeling effort is to relate dry deposition
to processes that are rate limiting. For this purpose the
resistance regime concept is preferable to the deposition
velocity concept. However, the deposition velocity has been
a framework for much of the research in dry deposition, and
will be used for summarizing these previous efforts. The
deposition velocity through all regimes is the inverse of the
total resistance;
Table 1 summarizes deposition velocity estimates for gases and
particles to be discussed in the subsequent text. These values
are applicable as inverse resistances to the regimes whose
effects were included in their determination. For example,
values obtained in wind tunnels or static chambers may only
partially include the first regime resistances or not. Values
derived from atmospheric turbulence alone will apply only to
the atmospheric eddy transport regime".
12
-------
Table la.
SUMMARY OF DEPOSITION VELOCITIES IN LITERATURE
FOR VARIOUS GASES AND PARTICLES.
U)
Deposition
Velocity
Type of
Surface
(number of
Substance
Iodine
Iodine
Iodine
Iodine
Iodine
Iodine
Form (cm/sec)
Gas* Range:
1.1 - 3.7
Mean: 2.1
Gas* Range:
0.3 - 1.3
Mean: 0.75
Gas* Range:
0.6 - 2.0
Mean : 1.3
Gas* Range:
0.3 - 0.9
Mean: 0.62
Gas* Range:
0.09 - 2.43
Mean : 0.87
Gas* Range:
0.6 - 0.7
Mean: .65
experiments)
Grass (7)
Clover
Leaves
(4)
Paper
Leaves
(4)
Filter
Paper
(5)
Grass (19)
Carbon (12
Methodology
Comments
V^ relative
to 1m height
Reference
Chamberlain
16,19
Cert Tests
21-26
There is some question relative to how long Iodine remains
a gas before becoming attached to particles.
-------
Table Ib.
Substance
SO,,
SO,
Form
Gas
Gas
SUMMARY OF DEPOSITION VELOCITIES IN LITERATURE
FOR VARIOUS GASES AND PARTICLES .
Deposition
Velocity
(cm/sec)
2.8
1.0 - 2.06
Mean; 1.3
Type of
Surface
(number of
experiments)
alfafa canopy
in
chambers
Grass (14)
Methodology
Comments
Wind speed:
1.8 to
2.2 m/s
Surface flux
determination
for SO^
Reference
Hill (1971)
Garland,
et al27
8
35
SO,
Gas 0.5 - 2.6
Mean: 1.3
Grass (3)
Radioactivity
labeled S02
release used
as tracer in
field experi-
ments .
Owers and
Powell28
*
35
SO,
SO,
Gas
Gas
0,5
0.27 - 1.5
Mean: 0.76
Water (1)
Grass (5)
Radioactivity
labeled SO2
release used
as tracer in
field experi-
ments .
Owers and
Powell28
Shepherd
29
Estimates of surface resistance are also given in the paper.
-------
Table Ib. (continued)
Deposition
Velocity
Substance
S02
II
IV
IV
VI
so2
Form (cm/sec)
Gas 1.8
(0,84)
1.3
0.91
2.2
<0.50
Gas 2,4
2.6
0.5
1.6
0.52
0.05
4.0
2.2
0.16
Type of
Surface
(number of
experiments)
Short
Grass
Medium
Grass
Bare
Calcareous
Soil
Fresh water
(ph = 8)
Forest Pine
Over Grass
Snow
Water
Methodology
Comments
Surface flux
(radioactive
tracer)
Reference^
Garland
(1974)9
Lapse
Neutral
Stable
Lapse
Neutral
Stable
Lapse
Neutral
Stable
Whelpdale
and Shaw30a
Estimates of surface resistance are also given in the paper.
-------
Table Ic.
Substance
SUMMARY OF DEPOSITION VELOCITIES IN LITERATURE
FOR VARIOUS GASES AND PARTICLES.
Form
Deposition
Velocity
(cm/sec)
Type of
Surface
(number of
experiments)
Methodology
Comments
Reference
NO
NO
NO
X
N02
HF
Gas
Gas
Gas
Gas
Gas
0
0.1
0.5
2,0
1,6
All surfaces
alfalfa
canopy
alfalfa &
oats
alfalfa
canopy
alfalfa
Not a field
result
In special
chambers
derived from
data in
Tingey32
Special
chamber
Fumigation
Robinson and
Robbins3ob'31
Hill8
Rassmussen
et al13
.Hill8
Israel33
3.1
field crops
with constant
concentrations
Fumigation with
constant
concentrations
-------
Table Id.
SUMMARY OF DEPOSITION VELOCITIES IN LITERATURE
FOR VARIOUS GASES AND PARTICLES.
Substance
O.,
O,
O.
Deposition
Velocity
Form (cm/sec)
Gas 0.7 - 1,7
Mean: 1.13
Gas 0.6 - 1.8
Gas 1.4 - 6.3
Gas 1.7
Type of
Surface
(number of
experiments)
Several
Soil, peat,
grass
Several
Alfalfa
canopy
Methodology
Comments
Review of
research by
several
authors34"39
Wind tunnel
experiments
Decay rate
experiment
in box
Special
chambers
Reference
Galbally39
Garland (1974)*
Garland (1974)9
Hill8
-------
Table le,
SUMMARY OF DEPOSITION VELOCITIES IN LITERATURE
FOR VARIOUS GASES AND PARTICLES.
Substance
Fission
Products
Form
Particle
137
Cs
CO
103
Ru
93
95
Zr,
Nb
Particle
Particle
Particle
Deposition
Velocity
(cm/sec)
0.07
0.9
0.04
0.2
0.2
2.3
0.4
0.6
0.4
5.7
2.9
1.4
Type of
Surface
(number of
experiments)
Gummed paper
Water (5)
Soil (15)
Grass (21)
Sticky
paper (117)
Water (9)
Soil (16)
Grass (20)
Sticky
paper (8)
Water (6)
Soil (6)
Sticky
paper (10)
Methodology
Comments
Surface
activity
measurements
particles are
formed by an
electric core.
Field release
test measure-
ments.
Reference
Megaw and
Chadwick4^
Convair
41-43
-------
Table le. (continued)
Substance
Te
Pb
Form
Deposition
Velocity
(cm/sec)
Type of
Surface
(number of
experiments)
Methodology
Comments
Reference
141^
Ca
1 2Q
Particle
Particle
0.7
0.7
Sticky paper
Sticky
paper (8)
. 41-43
Convair
Trace
Elements
Trace
Elements
Zinc
Sulfide
Particles
of O.lym
Natural
Particulate
Material
Particle
Particles
Particles
Particles
(diameter
of 2.5ym)
0.13
.04 to 0.3
0.3 to 1.0
0.5
0.2 - 9.2
0.8 - 7.6
(7)
Dilute HC1
solution
Deposition
plates
Deposition
plates
Sage brush
Desert
Great Lakes
(water)
44
Plastic Servant
pluviometer
Industrial Cawse &
substances Peirson^
Soil-derived
substance
46
Stable atmo- Simpson
spheric condi-
tions .
Islitzer &
Dumbauld47
Whelpdale
48
-------
Several recent reports have reviewed the dry deposition
processes for common pollutants13'14'49. The reader is
referred to these for additional information. References
13 and 14 include discussion of chemical reactions that
appear to be controlling the deposition rates.
Although the emphasis in this review is on certain fossil
fuel pollutants, other substances have been included to
demonstrate the general range of deposition velocity values.
Dry deposition of radioactive substances comprises a large
part of the available information.
Particles
Reported values of dry deposition of particles from the
atmosphere also shows a wide variation. In England, deposi-
tion studies of atmospheric trace elements onto artificial
surfaces produced deposition velocities which varied from
less than 0.04 cm/s to 0.3 cm/s for apparently industrial
45
substances, and 0.3 to about 1.0 for soil-derived substances
The deposition of small fluorescent particles was studied by
46 47
Simpson at Hanford, and Islitzer and Dumbauld at the
National Reactor Testing Station. Simpson's data for near
ground release of zinc sulfide particles (mean diameter 2.5
ym) under stable conditions were used by Islitzer and Dumbauld
to calculate an average deposition value of 0.5 cm/s.
Islitzer and Dumbauld, using particles of 1 ym, obtained
deposition velocities that ranged between 0.2 and 9.2 cm/s,
with a mean value for all experiments of 4.6 cm/s. Inversion
conditions consistently had the lowest values (0.2 cm/s) which
were comparable to Simpson's results. Deposition velocities
of 2.4 and 7.1 cm/s were computed for the neutral and lapse
conditions respectively.
20
-------
Table 1 includes results of U. S. Air Force studies of radio-
41 42
nuclide deposition ' . The properties of these radionuclides
are not clearly defined; as a result, they cannot be readily
applied to deposition of other substances. However, they do
indicate a wide range of possible values.
In a review paper, Gifford and Pack conclude that the average
deposition velocities for radioactive materials ( I, S0_, Ru)
are less than 1 cm/sec on flat plates or bare soil, and between
1 and 3 cm/sec on vegetation. For inert material he gave values
of less than 0.1 and 0.1 to 0.2 cm/sec for deposition on flat
plates or bare soil and vegetation, respectively.
40
A study by Megaw and Chadwick of the deposition of fission
products using surface activity measurements yielded deposi-
tion velocity values ranging from 0.07 cm/sec on gummed paper
to 0.2 cm/sec on grass. These fission products were submicron
particles produced by an electrical arc.
In a recent paper, Nickola and Clark reported about 7% of
a ZnS particle plume deposited within 842 m downwind of
release, in a moderately stable atmosphere. For two other
runs under very stable and unstable conditions, 1% or less
deposited within distances of 812 and 200 m respectively.
Estimates of deposition velocities were not made in this paper,
although these dual tracer results suggest that some earlier
studies may overestimate deposition rates.
Although there is considerable information on the rate of fall-
out of particles from upper atmospheric sources, the fact that
washout has been identified as a major mechanism for removal
precludes the use of this data except where wet and dry removal
can be separated. For example, a study of fallout materials
where dry and wet were separated at Kjeller gave average annual
dry deposition velocities in 1957 and 1958 of 0.51 and 0.75
21
-------
cm/sec. Individual average monthly values ranged between
52
3.4 and 0.2 cm/sec
The point of emission of a particle in the atmosphere can
influence the importance of certain removal processes.
Marenco and Fontan demonstrate by numerical models that a
particle produced at ground level will be eliminated chiefly
by dry deposition, whereas a particle produced near the
tropopause will be removed primarily by rainout. These
authors also point out that the residence time differs for
particles produced near the ground or near the tropopause
because of the change in process importance.
The surface and receptor resistances of particles have been
54 — 57
studied in wind tunnels. Sehmel has considered the
effects of variations in wind speed, particles sizes and
surfaces; and shows that deposition velocities change
significantly with wind speed and particle size. For 0.3
ym particles, (p = 1.5 g/cm ), the wind tunnel deposition
velocity on a simulated grass surface was .08 cm/s for a
5 mph wind and about 0.2 cm/s for a 30 mph wind. A field
experiment using the same surface and 0.7 ym mean diameter
rhodamine particles gave a total deposition velocity of
0.2 cm/s which compared well with a total deposition velocity
of 0.17 cm/s extrapolated from the wind tunnel data for this
size particle. In the wind tunnel studies, the deposition
velocities were found to be a function of particle size.
The minimum deposition velocities were fairly constant
within the range 0.1 to 1 ym particle diameters. Particles
larger or smaller than this deposited more rapidly. The
effect of a change in surface roughness was studied using
two sizes of crushed gravel. For 0.3 ym diameter particles and
a wind speed of 5 mph, the deposition velocities were measured
on gravel between 1.6 cm and 3.3 cm diameter. The deposition
velocities varied from 0.01 cm/s to 0.04 cm/s respectively.
22
-------
Sehmel ' has developed a model based on wind tunnel data.
When deposition velocity is considered as a function of particle
size, a minimum value is found in the relationship. His results
indicate that this minimum deposition velocity increases rapidly
as both wind speed and surface roughness increase; it is a
minor function of atmospheric stability. The model can be used
to predict deposition velocities (for a natural roughness height
of 3 cm) of about 0.20 cm/s for 1 m/sec winds and 3.3 cm/s for
7 m/sec winds.
Hicks has proposed a model for deposition velocity of rapidly
5 8
depositing gases based on the mean wind speed at a fixed height - .
Gases
The resistance can be expected to be the same for all gases in
the atmospheric transport regime. The resistances in the
layer over and on plant surfaces are functions of the molecular
transport processes over the surface and the potential for the
uptake of gas on or within the surfaces. Hence, surface resis-
tances of each gas must be considered separately.
The uptake of noble gases by natural surface will be very slow.
Therefore, resistance in the third regime is very high, and
significant deposition does not occur.
A comprehensive survey of the sources and natural removal
processes for atmospheric gaseous pollutants (including sul-
fur, carbon oxidants and organic-containing gases) has been
compiled . This survey contains considerable information on
the gases considered in this report, as well as on ammonia and
carbon monoxide.
Gases such as SO- and iodine, which are absorbed on most
natural surfaces, have a low surface regime resistance.
Studies show that deposition of iodine often occurs at a rate
23
-------
, . 6,16,19,21-26
primarily controlled by atmospheric turbulence
This implies the surface is nearly a perfect sink.
Because of its relatively rapid absorption onto surfaces, it
is not certain whether iodine should be considered a gas or
a particle. The release may be a gas but it is uncertain how
fast the iodine will attach to particles in the atmosphere.
The deposition velocity and the biological half-life of
elemental iodine on grass and foliage have been the subject
of recent investigations at the Julich Research Center in
5 9
Germany . Results indicate the deposition velocity (or
inverse of total resistance) is a function of biological and
meteorological variables. A semi-empirical relationship was
developed which includes the dry weight of vegetation per
unit area, the friction velocity, the relative humidity and
a biological factor.
Studies of SO,., deposition show that either the atmospheric
or surface regimes can be limiting. The state of stomata
openings, surface radiant energy budget, and the wetness
of the surfaces are surface properties that may control
resistance in the surface regime. Information on the actual
concentrations of SO,, in plants is contained in a study by
Pyatt . Martin and Barber documented the loss of SO0 near
f- n 2
a hedge. Mansfield and Heaten studied the changes in stomatal
openings at low SO- concentrations in Xanthium.
Spedding studied the effects of relative humidity and stomata
conditions for the uptake of S0« on barley. He found a
considerable range of resistance to uptake. These results may
be particularly useful in modeling efforts since they are
summarized in terms of deposition based on plant surface area.
Spedding confirmed that stomatal intake at high humidities
and absorption on thin water films can both be significant
in the rate of S02 uptake. His measurements were based on
24
-------
experiments with radioactively labeled SO-. Majernik and
64 2
Mansfield state that even though conditions are sufficient
for SO- to stimulate stomatal opening, the presence of suffi-
cient C02 can result in stomatal closing. This indicates that
the removal rate of SO may be a function of concurrent C02
concentration over a plant layer.
Relative rates of uptake of various pollutants in the soil are
factors in the surface resistance term. Abeles, et al.
present data supporting the hypothesis that microbial or chemi-
cal reaction rates of ethylene, other hydrocarbons, S0_, and
N02 are sufficient for dry removal to be a significant sink.
Ethylene uptake was not found with sterile soil or in soil
lacking oxygen. Significant uptake did occur with ethylene in
air over soil. The decay time for ethylene in air was such
that uptake was nearly complete in four days, indicating that
microbial reactions are important. For SO- and NO- the rates
were faster. In 15 minutes SO- was reduced from 100 ppm to
8 and 20 ppm for soil and autoclaved soil, respectively.
Nitrogen dioxide reduced from 0.100 ppm to 3 and 13 ppm after
24 hours for soil and autoclaved soil. These results suggest
that chemical reaction is dominant, although some allowance
should be made for microbial reactions.
The absorption and resultant profiles of gaseous air pollutants
by alfalfa canopies were studied in an environmental growth
p c c
chamber ' . Hydrogen fluoride, S02 and N02 profiles suggested
efficient removal by both the upper and intermediate surface
vegetation. Interpolation of their profile data suggests that
none of the gases were transferred to the canopy as efficiently
as momentum. Table 2 has been derived from their data to show
the efficiency of transfer of momentum and gases as a function
of height. Their data has been normalized to the top height to
allow comparison of the shape of the profiles. Assuming the
transport mechanisms are the same for all properties within the
25
-------
Table 2. NORMALIZED WIND AND POLLUTANT
PROFILES (BASED ON DATA IN
REFERENCES 8,66)
*
Height (cm) % Momentum % SO 2
60
50
40 (canopy)
30
20
10
100%
91
37
13
8
5
100
92
69
47
37
22
% 02
100
94
74
54
42
36
% HF
100
94
75
59
53
51
% NO2
100
96
90
75
62
57
% NO
100
98
100
93
92
94
9
54
24
5
3
8
23
22
10
15
6
20
20
12
6
6
19
16
6
2
4
6
15
13
5
2
-2
7
1
-2
Changes
60-50
50-40
40-30
30-20
20-10
*(Value at height/Value at 60 cm) x 100
air, the profiles may be interpreted as reflecting fluxes.
These profiles indicate that momentum transfer is more
efficient than any of the pollutants. Use of momentum
deposition velocities will overestimate the flux and the
location of the deposition. Although significant deposition
of the gases (except NO) occurs, the deposition is clearly
occurring deeper into the canopy than is the momentum trans-
fer as indicated by the changes in concentration with height.
Alfalfa canopies removed gaseous pollutants in the following
order: hydrogen fluoride > sulfur dioxide > chlorine >
nitrogen dioxide > ozone > peroxyacetyl nitrate > nitric
oxide > carbon monoxide. These results show the variation
26
-------
in deposition rates for different gases.
The exchange over water surfaces was considered from a phase-
resistance view by Liss , who tentatively concluded that the
vertical gradients of SO and water vapor close to the sea
surface should be similar, and the mean residence time of
S02 and a water vapor molecule should be of the same order
(10 days). He compared the resistances in the liquid and gas
phases and found the gas phase to be limiting at a pH greater
than about 2.8. He noted that this transition pH could vary
from 2.5 to 4 for the variation between calm and turbulent
conditions (in both phases). In a more recent paper, Liss
6 8
and Slater consider the relative resistances in gas and
liquid phases for SO, N20, CO, CH , CC1., CC1-F, Mel, (Me)_S.
For all except S02 the surface resistance was dominated by the
liquid phase. Estimates of exchange coefficients are given
based on available literature data.
Flux determinations of SO- deposition rates over specific
surfaces has been the topic of several recent papers. Garland
27
et al. used surface layer flux techniques to study the
deposition rates of S00. They computed the relative atmo-
9
spheric and "surface" resistance terms. Garland compared
surface layer flux techniques with deposition of a radio-
29
active tracer. Shepard used the surface layer flux technique
to study dry deposition over grass and water.
2 p
Owens and Powell used the radioactive tracer method to
study S09 dry deposition over land and water. Their paper
presents SO_ dry deposition rates which are fairly consistent
with those summarized in Table 1. References 69-71 provide
additional information on S02 uptake over water.
The capacity for various soils to take up S02 and other pollu-
tants under a variety of conditions has been the topic of sev-
27
-------
eral papers. Situations where soil absorption may be domin-
ant will require modeling a soil resistance term. The results
of Bonn98, Payrissat and Beilke72 (SO2), Fisher (S02),
Smith, et al.74 (S02/ H2S, CH3SH, C^, C^2 and CO), and
Terragilo and Manganelli75 (S02) will be useful in this task
for soils.
The vertical flux of 0., has been considered by a number of
authors as given in Table 1. 0_ appears to have a fairly
rapid removal rate, suggesting the atmospheric transport will
be primarily limiting.
3 8
Regener and Aldaz suggest 0., as an alternate flux measure-
ment to determine atmospheric turbulence when standard
techniques for momentum or sensible heat do not apply. Ozone
was assumed to be destroyed at the surface, with atmospheric
turbulence limiting the rate of transport to the surface and
hence determining the flux rate for nocturnal conditions. A
destruction coefficient was defined and is taken here to be
equivalent to a deposition velocity or inverse resistance.
In a paper which considers the sources and sinks of gaseous
nitrogen compound pollutants, Robinson and Robbins state
that gaseous deposition of NO is negligible. Gaseous deposi-
tion for N02 is calculated based on a deposition velocity
of 1 cm/sec31. This deposition velocity was an approximation
from experimental data over alfalfa and oats in Reference 32.
Application of Model
Although primitive, the present deposition model can be used
to improve modeling efforts requiring dry deposition
estimates. The basic dry deposition values (deposition
velocities or resistance values) are chosen from the literature
and vary as physical processes dictate. The deposition of SO
will be used as an example of a reactive gas that deposits
28
-------
rapidly on surfaces. The literature shows high deposition
velocities (low resistances) for S02. There is some informa-
tion on the relative values of the atmospheric and total
resistances for SC>2. In a simple model, the total resistance
to deposition can be defined as the sum of the surface
resistance (a constant estimated from the literature) and the
atmospheric resistance. The latter can be varied with wind
speed and roughness length using the relationship for atmo-
spheric resistance given in the model section.
Sulfur-bearing particles are generally about 0. 2um in diameter, anc
are thought to have deposition rates an order of magnitude
smaller than S0?. The order of magnitude higher resistances
for sulfur aerosols imply that surface resistance is con-
siderably more limiting for sulfur aerosols than for SO-.
The same approach to estimating deposition can be used, that
is summing a surface resistance and atmospheric resistance.
The variation in dry deposition of sulfur aerosols because
of atmospheric changes will be considerably less, perhaps
often negligible because the surface is limiting. In this
case, a more detailed surface model is necessary to estimate
the surface resistance as a function of ambient conditions.
A model accounting for impaction and surface area appears
applicable. The current field effort is designed to provide
estimates of the variation of total and atmospheric resist-
ances of sulfur aerosols under various field conditions.
Applications in Computational Models
In the gaussian diffusion model the total resistance may be
used directly to estimate deposition over a given point based
on the air concentration at that point:
G = X/rT .... (7)
29
-------
This technique fails to account for progressive depletion
of the plume downwind. To the extent that diffusion nomograms
can be expected to include deposition effects inherent in the
field data on which they are based, this technique is justi-
fiable. Where deposition for extended distances is considered,
the model may account for mass loss by deposition by modifying
the nomograms as a function of distance. Without this correc-
tion it is possible to compute more pollutant removed than
there is mass of pollutant. When great distances are con-
sidered, the failure of the model to account for mass loss
by deposition can cause large errors.
A single point relationship may be used as a boundary condi-
tion when a finite difference K theory model is utilized.
At each grid point the flux to the surface is given by:
G(x,y,z=0) = x(x,y,z=0/rT(x,y,...) (8)
This formulation presents the total resistance as a function
of the spatial coordinates of the surface, allowing incor-
poration of additional variables that research may reveal.
These variables can be specific to a gas or particle and may
include processes in the various regimes discussed earlier.
When the gaussian model is used for extended distances, the
total resistance may be used directly to estimate depletion
by a source depletion model. Following the notation of
Horst the downwind air concentration including loss by
deposition and diffusion is given by:
X(x,z) = Q D (x,z,h) exp -
where,
30
-------
-2
D = diffusion function (sec m )
_2
X = crosswind-integrated air concentration (units m )
Q = source strength (units sec )
x = downwind distance coordinate (m)
z = height of receptor (m)
h = height of diffusion source (m)
rT = total resistance
£ = downwind coordinate of surface source,
variable of integration (m)
where D is defined as:
D(x,z,h) = x(XgZ) - (10)
This is not difficult to apply once the total resistance is
defined, but it fails to consider that depletion occurs on the
surface. With this model the depletion occurs throughout the
entire vertical extent of the diffusing cloud.
Horst developed an alternative model which accounts for sur-
face depletion within the framework of a gaussian model. The
calculated downwind air concentrations include the effects of
resuspension, weathering and deposition as a function of time.
Considering the special case for a time independent source and
deposition processes only results in the relationship for the
depleted and diffused air concentration;
:,z,h) - J
X(5rZ,,h) D(x-g, z, h = o)
X(x,z) = Q D(x,z,h) - | * , d£ (11)
T
where the symbols are the same as for Equation 9.
Horst made a comparison between the two approaches (Equations
9 and 11) for various initial release heights and stabilities
31
-------
i _9
and assuming —— = 10 where u = mean wind speed. The
r'j'U
results show that the source depletion model overestimates
the surface air concentration by factors of 2 to 4 in some
cases. This effect results in the rate of decrease of mass
downwind in the surface depletion model being smaller, and
hence at large downwind distances the surface depletion
model eventually has surface air concentrations that are
greater than those of the large depletion model. The cross-
over distance where the concentrations (at 1m height) are
equal is almost 10 km downwind for very stable conditions.
The differences for unstable and neutral conditions are not
as great.
Elderkin et al. have modeled deposition processes in terms
of the turbulence parameters and compared their model with
field results.
32
-------
SECTION V
EXPERIMENT DESIGN
The objective was to design a field experiment to provide
information on the rates of removal of various pollutants.
The gases studied were S00, 0_, NO and NO. The aerosols con-
^ j X
tained lead and sulfur. The surface layer profile flux deter-
mination technique appeared best able to obtain quantitative
information on removal rates. The initial experiment design
was presented in Reference 29, which was published in the
proceedings of Atmosphere-Surface Exchange of Particulate
and Gaseous Pollutants.
A primary requirement of this experiment was that the measured
profiles be controlled by the local surface fluxes. An ideal
conceptualization of the processes influencing the vertical
concentration profile of a pollutant downwind of a stack is
given in Figure 1. Immediately downwind of the stack is an
area where the pollutant does not effectively reach ground
level. At greater distances, the deposition rate for smaller
particles becomes progressively a function of both the settling
velocity and the eddy movements of the particles. Then for
particles with settling velocities much smaller than typical
eddy velocities of the air, the deposition becomes a function
of turbulence and the surface properties.
Gases can be expected to be diffused with these smaller par-
ticles. Gaseous pollutants will not normally undergo the
gravitational settling phase. Certain conditions could exist
that could skew downwind the vertical profile of various gases
in any of the phases in Figure 1. Particles or liquid droplets
falling through a plume may redistribute gases in the vertical
79
profile, as shown in a study by Hales, et al.
33
-------
A
B
D
U)
GRAVITATIONAL
SETTLING
SOURCE
DEPENDENT
PROFILE
STACK RELEASE
SOURCE«<
I
\
CONCENTRATION
PROFILES
DEPOSITION
DEPENDENT
TRANSITION PROFILE
DRY DEPOSITION SURFACES
FIGURE 1. SCHEMATIC OF POLLUTANT CONCENTRATION
PROFILES DOWNWIND OF A STACK.
-------
For both gases and particles, the source dependent profile is
in the region where the vertical profile is still determined
by the source configuration. For hot plumes the buoyant char-
acteristics of the plume control the initial dispersion. How-
ever, as the plume mixes with ambient air both the pollutants
and the excess heat become diluted. The transition stage is
where the effects of both source and dry deposition appear in
the vertical structure of the air concentration profile. Fur-
ther downwind the source profile configuration is lost and the
vertical profile reflects only the effect of removal by dry
deposition.
This is an idealized discussion to illustrate the progression
of processes. Diurnal changes and local variations in the
atmosphere can greatly influence the processes,
Computational Basis
Dry deposition may be considered as a flux of the pollutant
from the air to a receptor. Although in practice this flux may
vary with distance from the receptor, horizontal homogeneity is
assumed. That is, the average characteristics of the receptors
and the atmospheric delivery processes are assumed constant
horizontally. This requires a sufficient upwind fetch to
allow development of a horizontal equilibrium of the dry
deposition processes. Any vertical variation represents the
flux of material from various heights to the receptors at
ground level. The current effort requires that measurements
be in the constant flux layer over the surface where the flux
does not change significantly with height.
The vertical flux (G), of the pollutant with air concentra-
tion x maY ke described in a gradient flux relationship:
35
-------
G = -K 4*
a Z
where K is the eddy diffusivity for the pollutant. When
dry deposition is occurring, the gradient -^ would be posi-
tive. This results in a flux towards the surface.
Equation (12) is applied to a layer of air over the receptors
where eddy flux is the dominant mechanism for transport.
Once the flux for the pollutant has been determined, the
resistance rT (inverse deposition velocity) at height z is
given:
K
(13)
In the constant flux layer over the surface the total resis-
tance appears to be a function of height. Consider the
middle expression in Equation 13. The denominator, G, is a
constant with height in the constant flux layer, and the
numerator, x ' wiH decrease near the surface when dry
Z
deposition is occurring. Hence, higher and lower apparent
resistances can be computed from the same data set using higher
and lower heights to define the air concentration. The expo-
nential nature of the surface layer profiles suggests the effect
will be greatest near the surface and decrease with height.
The upper level of data acquisition has been adopted for the
current study to minimize this effect. Adjustment of results
to a standard 1m level will be a function of prevailing micro-
meteorological conditions.
36
-------
Equation 13 defines the total resistance for the flux of mater-
ial to the surface. The effects of all regimes are included.
The atmospheric pollutant gradient reflects the combined effects
of atmospheric turbulence and rates of removal near or on the
environmental surfaces. The total resistance equals the atmo-
spheric resistance only when all other resistances are suffi-
ciently small.
The application of Equation 13 requires defining diffusivity
K • In finite difference form Equation 13 becomes:
-L = r = 5./Az\ = 1 _ Az (x) ,,4v
vd rT K \AX/ iT~7AX/Az\ K (AX)
\ X /
where clearly the resistance is a function of the eddy diffusi-
vity and the fractional change of concentrations with height
approximated over a finite layer. Direct determination of the
eddy diffusivity for pollutant particles or gases is generally
not possible. Instead, eddy diffusivities for concurrent fluxes
are principally momentum, sensible heat, and latent heat,
although literature exists on the possibility of using ozone
, 39,80
or radon
To determine a particular eddy diffusivity two flux relation-
ships are equated and solved for the eddy diffusivity. The
first (Equation 12) is the Fickian relationship between the
vertical gradient of the concentration of the material to be
3 Y
transported, -?p-, and the flux of the material, G. The second
relationship is the eddy flux equation:
G = () (15)
37
-------
where w is the vertical velocity and primes refer to
departures from the mean. The application of these requires
definition of terms. For momentum flux, G = T and x = PU'
(T = momentum flux, p = density of air, u = mean wind speed);
for sensible heat flux G = H, and x = PCpG, (H = sensible heat
flux, Cp = specific heat at constant pressure for air, 9 =
potential temperature) ; and for latent heat, G = LE and x = PLc3
(E = water vapor flux, L = latent heat of evaporation for water,
q = specific humidity). Equating the two flux relationships and
substituting definitions of x/ the following relationships for
eddy diffusivities may be obtained:
=
m
3z
,17,
3z
and
_ (q'W)
q -- ~
8z
K
H and K are the eddy dif fusivities for momentum, sensible
heat and latent heat respectively. The use of and the relation-
ship between these eddy dif fusivities are currently under
discussion. When any of the vertical gradients are near zero
(or below measurement capability) the eddy diffusivity is not
defined, and the technique cannot be used. When the flux of
the material is zero the eddy diffusivity is also undefined.
In addition, there is no requirement that various fluxes are
necessarily zero at the same time.
-------
In addition there may be basic differences in the transport
for various properties or substances. The sensible heat trans-
port is affected by the inherent tendency for updrafts to be
warmer air and downdrafts cooler air, with no corresponding
mechanism for momentum flux. Whether a certain similar inher-
ent bias may occur in pollutant fluxes is unknown.
Exact analogies between the various fluxes require that the
transport mechanisms be identical. To the extent that these
are similar, the corresponding vertical profiles will exhibit
similar shapes. The vertical fluxes of sensible heat, latent
heat, momentum and a pollutant are reflected in the mean
vertical profiles of potential temperature, water vapor, wind
speed, and pollutant concentration. An irregular vertical
change such as caused by a discontinuity (inversion layer,
etc.), will normally be reflected in some manner in each of
the profiles. In this way the similarity of the shapes of
profiles is useful for excluding cases with obvious vertical
flux divergence.
Without detailed study, the choice of the appropriate eddy
diffusivity must be qualitative and to a certain extent arbi-
trary. One convention is assuming that momentum eddy diffusi-
vity is applicable. The additional surface resistance caused
by molecular processes on the gas deposition surfaces may be
81
considered using the dimensionless reciprocal Stanton number
This is equivalent to the use of a surface resistance in the
current study. The surface resistance should also be applicable
to aerosol deposition with different processes producing a
surface resistance.
Another convention is assuming that an eddy diffusivity for
heat or water vapor is appropriate. A special case is the
use of an apparent eddy diffusivity obtained by an energy
budget method which assumes the heat and water vapor eddy
39
-------
diffusivities are equal, or that their ratio is known. The
surface resistance should apply with use of the heat or mois-
ture eddy diffusivity in a similar fashion.
There is no reason to suspect the source-sink relationships
between any of these fluxes and the pollutant flux are
identical under all conditions. Results based on either
eddy diffusivity should be interpreted as relative fluxes
because the estimated value of eddy diffusivity may be in
error.
If momentum eddy diffusivity is assumed, then the flux, G,
of a pollutant with air concentration x and corresponding
resistance r , may be expressed as follows with values defined
at level z,
G =\^^- (£) 1 (19)
II £*
(—}
-L = r = z ^Z^Z (20)
Vd T (u'w') /^_\
\3Z/Z
This may be written in finite difference form for computation;
_
G ~ -
I— }
1 _ „ _ x \Az/z
:= 3T ^ A ' ^* / n o \
Vd rT (u-r^T) _ (22)
,Az; z
40
-------
The variables that need to be measured to determine G and r
are: the horizontal and vertical wind speed (at a sampling
rate and with an instrument time response sufficient for eddy
flux calculation); the .vertical gradient of mean wind speed;
the vertical gradient of air pollutant concentrations; and the
air pollutant concentration.
Using the alternate heat flux analogy, the air temperature
and vertical wind velocity must be measured at a sampling rate
and with an instrument time response sufficient to calculate
an eddy flux plus a vertical gradient of potential temperature,
The definition of the conditions where the surface flux method
is applicable places constraints on the location and sampling
periods which can be used. The fetch conditions for develop-
ment of a surface layer profile of wind and temperature have
been defined in experimental studies of momentum and heat
transfer. The literature contains estimates based on canopy
heights and local roughness elements. Ultimately each site
must be considered individually including the prevailing
conditions during the experiments. As a general rule, the
surface flux approach cannot be applied unless 1) air contain-
ing the pollutant moving over a site has been well mixed
vertically up to a sufficient depth and 2) the site meets
equilibrium surface layer fetch requirements defined by the
site characteristics. These are the minimum conditions for
application of the flux model at a specific location.
A study of carbon dioxide profiles over a wheat crop showed
that the horizontal variation in the concentration difference
8 2
was small enough that a single tower should be adequate . If
the pollutant uptake can be assumed to be as uniform as the
CCU emissions, then this supports the current design of using
a single instrumented tower to determine profiles over repre-
sentative surfaces.
41
-------
Experimental Accuracy
The experimental accuracy is estimated by application of the
theory of sampling. The experimental error for each mean
value is estimated. These are then propagated through the
computation of a deposition velocity (inverse total resistance),
The accuracy of the pollutant gradient and the eddy diffusivity
are also limiting factors in the computation.
The initial estimates of accuracy require judgements of the
magnitude and randomness. As a result, the final accuracy
estimates are partly arbitrary, but they are a useful measure
of the relative accuracy between experiments.
Gradients. Gradients are computed from differences in con-
centrations divided by differences in heights. The accuracy
of the height differences is estimated to be better than 0.3%.
As a rule, the accuracy of the pollutant concentrations be-
tween measurement heights is estimated to be up to an order
of magnitude better than the absolute accuracy. Direct
assessment of the relative accuracy is not possible with the
current experiment design, although relative accuracy may be
indicated by the consistency of the vertical pollutant grad-
ients with concurrent meteorological gradients.
Eddy Diffusivity. The eddy diffusivities are computed for
Equations 16 and 17 for momentum and heat respectively. The
expression for the experimental error is
Exp. Error =
1/2
(23)
42
-------
where for heat a=Q and g=A0/Az, and momentum a=u and
g=Au/Az. The cr' s refer to the estimates of the experimental
standard deviations for the subscripts. This assumes the errors
are random. In practice, the value of a— was assumed to equal
aw ^
±.05% of uw or ±.02% of Gw.
The advantage of this analysis is that it keeps track of the
accuracy in automated computations. Note that, the terms
u'w' and 6'w1 in Equations 16 and 17 are computed from the
difference of two terms.
u ' w' = uw - u w , 2 4 \
0'w' = 0w - 0" w
These difference terms can generate an answer which is "noise"
when the terms are nearly the same size. Equation 23 combines
this and other computations into a single estimate of the
accuracy of the K values.
43
-------
SECTION VI
INSTRUMENTATION
Field Instrumentation
Development and field demonstration of measurement systems
applicable to the surface flux measurement concept is the
current goal. These systems involved areas of surface layer
micrometeorology and air pollution chemistry.
This section will discuss the systems that were designed,
assembled and tested. The systems that were chosen for the
field testing and demonstration effort will then be described.
The basic field system is composed of a tower for placement of
filters, sensors, and air intake lines at various heights to
obtain vertical profiles of pollutants and meteorological vari-
ables. A mobile laboratory is used as a base of operation and
contains the analyzers and recorders. The gas and aerosol
systems will be considered separately after a discussion of
common elements. The micrometeorological measurement system
used in conjunction with both the gas and aerosol systems will
be discussed in the final section. Photographs of the complete
field instrumentation and filters are given in Figures 2 and 3.
Common Components
The air intake lines between the tower and flow meters were 12.7
mm diameter Teflon tubing with stainless steel swage lock
fittings. Teflon wedges and collars were used in the Swage lock
fittings. Teflon tubing 6.35 mm in diameter was used for all
connections from the tower to the mobile laboratory. Stainless
steel or all-teflon Swage lock fittings were used. The sampling
lines were enclosed in a heated, insulated aluminum casing. The
lines can be heated in situations where condensation may be
problem.
44
-------
*>.
LTI
FIGURE 2. VIEW OF INSTRUMENTED TOWER AND MOBILE VAN
WITH COOLING TOWER PLUME IN BACKGROUND.
-------
FIGURE 3. NUCLEPORE FILTER MOUNTED ON TOWER.
46
-------
A 17 m crankup triangular tower was used to mount sensors and
sample input lines. The gas and aerosol systems required
inputs from various heights. Three heights, 0.91, 6.1, and
15.2 meters are used for intakes in the fixed height applica-
tions. Separate sampling lines, flow meters and pumps were
used for each height. A single input tube, flow meter, and
pump was used in applications requiring a single direct input.
Aerosol System
Aerosol samples were collected in the field for subsequent
laboratory analysis. Samples were collected on nuclepore
filters with a pore diameter of 0.8 ym. Moderate rates were
employed to minimize sub-isokinetic sampling errors. Stain-
less steel filter holders with an effective collection area
2
of 11.4 cm were used. Figure 3 shows a filter mount, and
a filter at the 0.91 m measurement height. Other filters are
at 6.1 and 15.2 m. Air was drawn through the nuclepore filters
for 1 to 3 hours at each of the three levels. The samples
were analyzed for sulfur and lead by dispersion x-ray fluore-
scence. The combination of a species density on a filter and
the total volume of air sampled were used to calculate the air
concentration.
The flow rates through the sampling system were monitored
by ROOTS flow meters (model 1.5M125). There are positive
displacement meters that accurately measure the volume passing
through the meter at a given temperature and pressure. If the
sample is heated or cooled, its density changes as does the
volume indicated. One temperature-compensated unit was
obtained for initial testing- The variation did not appear
to justify the use of such a unit in the current application.
Running under identical conditions uncompensated meters gave
essentially identical results for time periods and flow rates
comparable to the current application. In addition, the three
47
-------
flow meters were tested by running them in various orders in
series. The calibration differences were found to be negli-
gible. The flow meters produce a signal for an event recorder
each time 0.028 m3 (1 cu ft) of air passes through. The ROOTS
meter output from all three units was recorded on the same
recorder to allow close comparison of both total flow and flow
rate changes during experiments.
Gas - General
Three basic designs were proposed for the gas sampling effort;
System A - Alternating Collector Bags, System B - Direct
Sampling, and System C - Integrated Systems. These are shown
diagramatically in Figure 4, and were tested in varying con-
figurations as described below.
Each of the systems in Figure 4 has its own inherent advant-
ages and disadvantages. The use of the alternative collector
bag system can give a sample representative of the entire
period of time. However, storage of certain highly reactive
gases such as 0., and NO is not possible, so the technique
cannot be applied to these gases. The direct sampling system,
however, is designed for such reactive gases. The high flow
rate in the sampling lines yields residence times on the order
of a few seconds. Only a small portion of the gas flowing in-
to the mobile laboratory is used for the actual analysis. The
small residence time in the sample lines is inherent in both
Systems A and B, and may be true in System C. System B can
only give sequential samples from different intake lines in
our present system; its limitation is a single analyzer for
each of the gases. This disadvantage could be overcome by
4 8
employing several analyzers, as did Whelpdale , System B
has the advantage of direct sampling with minimum storage time
This system was used in several modes of operation for SO ,
NO , NO and 0 in the current field effort.
X -3
48
-------
SYSTEM A - ALTERNATING COLLECTOR BAGS
AIR SAMPLE
INLETS
BAG FILL RATE
CONTROL VALVE
REVERSING
CIRCUIT
1
ROOTS FLOW
METER
SAMPLE
MONITOR
SYSTEM B - DIRECT SAMPLING SYSTEM
AIR SAMPLE
INLETS
i
t
SAMPLE
MONITOR
ROOTS FLOW
METER
AIR PUMP
SYSTEM C - INTEGRATED SAMPLE
AIR SAMPLE
INLETS
SAMPLE
INTEGRATION
SYSTEM
I ON TOWER I I
ROOTS FLOW
METER
AT MOBILE LABORATORY
FIGURE 4. GAS SAMPLING SYSTEMS.
49
-------
The third, integrated, sample system is most applicable for
pollutant concentrations at or below the instrument detection
limits. Although not used in the current field effort, prepa-
rations were made to test this system for SO-. SO- deposition
83
tubes were prepared. These tubes use solid absorbents to
effectively trap all the SO- passing through the tube at
ambient conditions. The SO.-, can be thermally dissolved later.
By purging a heated tube with a known volume of dry nitrogen
into a collector bag, a sample may be prepared that is concen-
trated by the ratio of the volume used for sampling and purging,
This ratio can be varied to provide concentrations within the
detection limits of the analyzer. The lack of time and
sufficiently favorable conditions during the field studies
was the only reason this technique was not tested. Other
techniques for concentrating samples of gas including SO- and
NO were considered but not chosen for the present effort.
When considering the relative merits of each of these systems,
an important feature is the relative accuracy that is attain-
able between different measurement heights on the tower, since
the accuracy of the model to be developed depends directly on
the accuracy of the vertical profile measurements. The use of
a separate System A for each measurement height should yield
accurate gradient data as long as the rates filling the bags
are nearly equal and there is no differential loss of the
pollutant between the systems (either in the sampling lines or
during storage in the bags). System B, when used to sequence
air flow intake from different heights, offers the potential
for good relative accuracy. However several factors must
be considered: a question arises in the time sequencing of
the samples between measurement heights; the inlet pressure
can affect the calibration of certain analyzers and switching
between different inlet lines can change the pressure; slight
absorbtion of the pollutant may occur in one line compared to
50
-------
another; the shock of connect-disconnect can have transient
effects. Any of these could cause a spurious difference in
readings between levels. An alternative might be to use a
sampling system with a single sampling line. This would solve
the problem of changes in sample lines. Any errors in the line
would be the same for all heights. The sample intake line need
only be long enough to reach between the different sampling
levels on the tower. Current tests were performed by manually
moving the line, although automation of such a system is not a
difficult task84'85.
The bag collection system appeared to be the most promising for
S0« and NO measurements, although it was not used in the
^ X
current field effort. We planned to use three system A sampling
systems, one for each sampling height. Air samples would
slowly fill the collector bags I for each height for periods on
the order of 1/2 to 1 hour. Then, the plumbing would auto-
matically switch collector bags I and II. Collector bags II
would start to fill, while collector bags I were sequentially
sampled by the SO0 and NO analyzers. The collector bags I
^ x
would then be emptied, purged with dry nitrogen and emptied,
to prepare the system for another cycle of sampling.
The collector bags were specially fabricated out of Tedlar in
a cube shape to permit maximum sample volume. Tedlar has
acceptable properties for NO and SO,, measurements. The
X ^
collector bags were mounted inside a plexiglass box which sit
inside a metal rack with a metal face in case of accident.
The sequencing of the system operated satisfactorily in each
test at the laboratory. A FX system model programmable control
ler was used to control the sequencing. Stainless steel sole-
noid valves were used in all switching applications.
51
-------
The bag collection system had a fault that precluded its use
in the current field program. The boxes containing the
collector bags imploded several times in mockup field tests in
the mobile laboratory. The Tedlar bag was damaged in each
implosion. Lowering the flow rate to the lowest acceptable
sample line residence time was attempted, but the vacuum
created was still sufficient to implode the box. As a result,
this system was not available for the field tests.
In the field experiments, slightly different techniques were
used for each gas. These are described along with the analysis
method in the following section.
S02
Fixed tower heights and separate sampling lines were used for
the SO- profile experiments. Samples from each height were
sequenced as fast as possible through the S0« analyzer at a
rate of about 1 every 2 minutes. A portable gas chromatograph
manufactured by Analytical Instruments Development, Inc. (AID)
was utilized. This instrument was modified with an electro-
mechanical switch to automate the sampling procedure. The FX
system programmable controller was used to sequence the sampling
with other operations. The gas samples were sequenced through
stainless steel and Teflon valves.
A flexible sample line was adopted for measuring the ozone
profile. Ozone was determined using a Bendix Model 8002 Ozone
Analyzer. The following description is taken directly from the
Operation and Service Manual B1261TM, Bendix Process Instru-
ments Division.
-------
"The Bendix Model 8002 Ozone Analyzer operates over a wide
range of temperature and humidity variations without adverse
degradation to measurement accuracy. The system provides re-
liable operation and is designed for ease of maintenance.
Electronic components are mounted on plug-in printed circuit
cards, wherever practical, so that equipment operation can be
restored quickly in the event of a component failure. Chemical
and particulate filters, subject to periodic replacement, are
mounted on the rear panel for ease of replacement.
The Ozone Analyzer provides a direct read-out of ozone concen-
tration on a continuously exchanged ambient air sample. The
system reacts quickly to quantitative changes in ozone content
of the sample, while providing a highly stabilized measurement
capability for extended periods, without repeated adjustments.
In addition to the front panel meter indication, an output
signal is provided through the front panel RECORDER jacks and
rear panel terminal connections.
The Ozone Analyzer utilizes the principle of photometric
detection of the chemiluminescence resulting from the flameless-
gas-phase reaction of ethylene with ozone. This reaction pro-
duces light waves or photons that are transmitted to the photo-
multiplier tube. The photomultiplier tube converts the light
waves into electrical energy. This electrical energy is
multiplied or amplified by the photomultiplier tube, and is
further amplified in the electrometer amplifier to provide the
proper drive voltages for the front panel meter and the exter-
nal strip-chart recorder. Thus, the resultant meter reading or
the output to the recorder, is proportional to the light waves
produced by the ozone-ethylene reaction. The degree of reaction
is in turn proportional to the amount of ozone in the air
sample.
53
-------
The minimum performance parameters for the Ozone Analyzer
are presented below. The instrument will operate within
these stated performance parameters under the conditions listed,
a. Range: 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, and 1 ppm
full scale
b. Minimum detectable sensitivity: 1 ppb
c. Zero drift: ±1% per day to ±2% per three days
d. Span drift: ±1% per day to ±2% per three days
e. Precision: ±2% from mean value on the 0 to 0.1 ppm range
f. Linearity: ±0.5%
g. Noise: <_±1% on the 0 to 0.1 ppm range with 10 second
time constant
h. Interference equivalent: £0.01 ppm
i. Lag time: £3 seconds
j. Rise time: £7 seconds to reach 90% of ultimate indication
k. Fall time: £7 seconds to reach 90% of ultimate indication
1. Response .time : <_10 seconds to reach 90% of ultimate
indication after input change (lag time plus
response time)
m. Operational period: 7 or more days unattended
n. Ambient temperature fluctuations: ±5°C
o. Operating temperature extremes: 5°C and 40°C
p. Time constants: 1, 10 and 40 seconds
q. Power requirement: 105-125 volts, 60 Hz (to Hz optional),
350 watts
r. Outputs: 0 to 10 mV (recorder) and 0-1 Vdc (other
outputs optional)."
NO /NO
The N0x/N0 instrument functions using chemiluminescence , a
phenomenon in which light is produced as a byproduct of a
chemical reaction. The intensity of the light produced is
directly proportional to the concentration of the gas (NO)
54
-------
being measured, with the concentration of the other reactant
(ozone) being held constant. Thus, the instrument provides for
directing the sample gas into a reactor chamber. A selection
is made on the front panel for the instrument to measure NO or
NO . When NO is selected, the gas is directed into a cata-
J\. i£
lytic converter that changes NO into NO. Ozone, generated
X
within the instrument from the oxygen in ambient air, is also
directed into the reactor chamber. The resulting chemical
reaction in the reactor chamber produces light, which is sensed
photoelectrically. The signal thus produced is amplified and
drives a meter on the front panel. The meter indicates the
concentration of NO or NO directly in ppm. The output is
X
recorded on chart paper.
The specifications of
given as follows:
Sensitivity:
Range:
Precision:
Linearity:
Zero Drift:
Span Drift:
to NO" Converter:
Response Time: (10-
90% of Steady State)
Reaction Chamber:
Data Display:
the REM Model 642 NO and NO Monitor are
x
5-ppb (S/N = 1 when C = 5-ppb).
0-.5, 0-2, 0-10 ppm.
±1% of full scale.
1% for given range.
±1% of full scale between ranges.
±1% per day; ±2% per 3 days, ±3% per
month.
±1% per day; ±2% per 3 days, ±3% per
month.
Provides specific 100% conversion
of N02 to NO.
0-.5 ppm. RANGE: 15 Sec.
0-2, 0-10 RANGES: 9 Sec.
Operated at atmospheric pressure -
no vacuum pump necessary.
Analog front panel meter, readout
in parts per million. Digital volt-
meter optional.
55
-------
Data Output: Front panel analog output terminals.
Output voltage adjustable from 10
mv to 5 volts full scale on all
ranges (100% overrange capability
on analog output for each range).
Mode Control: Manual front panel switch to change
NO to NO mode. Remote mode switch-
^C
ing optional. Automatic mode switch-
ing and determination of N02 concen-
tration by difference method optional,
Zero Adjust: Front panel 10-turn potentiometer.
Power Requirement: 350 Watts.
Operating Temperature: Continuous duty at full load from
0 to 40C (32 to 104°F) .
Inputs: Separate sample and span/zero input
selected by front panel switch.
Micrometeorology Data System
All components and location are shown in Figure 5.
Wind. Three dimension component wind speeds (Gl, G2, G3) were
measured at three heights (1.79, 7.77, and 17.1 m) by Gill UVW
anemometers with shaft extensions. A detailed description of
this system and its range of application may be found in
references 84-88. The u and v component arms were pointed
south and west respectively. The Gill wind data were recorded
in digital form. The angle correction factors in reference 85
are used in the data reduction program to account for the
"cosine" error.
Wind speed measurements at intermediate heights were
obtained by cup sensors with the output recorded on strip
chart recorders. Wind speed and direction were also monitored
adjacent to the tower (15 m south) with a Climet Model
56
-------
17.1
15.2
•E 11.2
o
7.8
6.1
4.8
1.8
0.92
0
r GILL(u,v,w), T3
NUCLEPORE FILTER/AIR INTAKE
~ CUP SENSOR
' GILL(u,v,w), T2
- NUCLEPORE FILTER/AIR INTAKE
' CUP SENSOR
GILL(u,v,w), Tl
NUCl£PORE FILTER/AIR INTAKE GROUND COVER - GRASS
SCHEMATIC OF INSTRUMENTATION
FIGURE 5. SCHEMATIC OF INSTRUMENTATION LOCATION,
57
-------
weather vane and cup system. These were recorded on chart
paper.
Air Temperature. The primary air temperature system (Tl, T2,
T3) used aspirated thermistor sensors at three heights. These
were mounted in the lower portion of the Gill wind systems, and
used bridge circuits with matched components. The multivolt
output was recorded directly on the digital recording system.
Each height has two identical but independent systems. The six
thermistors were selected to matched resistance characteristics
The redundancy at each level provided additional field checks
that circuit components were operating correctly. The time
response of the thermistor in still air was about four seconds.
The recorded values are changed to standard units as part of
the computer program to reduce the data.
Additional onsite weather conditions were continually recorded
using a hydrothermograph for temperature and relative humidity
in a standard weather screen; a Solimeter (Model R401, Serial
134) for incoming solar radiation; a miniature net radiometer
(Micromet Inst. Serial 276) for net radiation; a heat flux
plate for heat flux into the ground; and barographs to record
changes in atmospheric pressure. These were all recorded on
chart paper, except net radiation and soil heat flux which were
included in the digital recording system. These additional
measurements were used in the field data acquisition and as
independent checks on the primary systems data but are not
reported as part of the current data set.
The digital recording system was a iletrodata Model digital tape
recorder. Twenty channels of three decimal places and a sign
were used. Data were recorded on a cassette tape at the rate
of slightly over 2 sets per second. The contents of the
channels are shown below.
58
-------
Channel Content Channel Content Code for Channels
11 T 2 HMS hour, minute,
1, 2 HMS 12 T 2' seconds
3 u 1 13u3 +u= south wind
4 v 1 14 v 3 +v= west wind
5wl 15w3 +w= wind upward
6 T 1 16 T 3 1 171 m
7 T 1 17 T 3 2 7.77 m
8 u 2 18 C 3 1.79 m
9 v 2 19 Tn ' second sensor system
10 w 2 20 G/Ozone C = Thermistor battery
voltage
G = ground heat flux
Rn = net radiation
The Metrodata cassette data tapes were transferred to computer
compatible tapes on a NOVA computer. Then an computer is
used to reduce the data-and provide appropriate summaries.
Intermediate data translations by Metrodata and Battelle
Columbus were necessary to recover the meteorology data.
59
-------
SECTION VII.
DATA COLLECTION
Field Tests
Brief field tests were performed to test instruments determining
the vertical fluxes of SO.-,, 0.,, NO , NO, sulfur and lead. The
£. j X
instrumentation was described in the previous section. These
tests were designed only to evaluate the operation of the equip-
ment and demonstrate the possible resolution of deposition
rates under field conditions. Extensive field data acquisition
was not the goal of these experiments.
The site chosen was 10 km north of a fossil fuel plant at
Centralia, Washington. The site elevation is approximately
85.3 meters (280 ft MSL), longitude 122° 55' 30" and latitude
46° 50' 5" N. The site is located near the center of T-16N,
R-l-W block #25. Figure G summarizes the fetch in various
directions around the site. A thirty degree sector is con-
sidered to be affected by the mobile laboratories (315 to 545°).
Figure 2 shows the site with the tower, instrumentation and
mobile laboratory looking towards the Centralia plant. The
cooling tower plume may be seen near the horizon.
The stack plume from the Centralia plant was not identifiable
at the site during the tests of systems. Instead the tests
were made on ambient pollutants from apparently more distant
sources. The aerosol data has higher than normal copper
values suggesting that some of the pollutants may be origina-
ting in the Tacoma area. This is consistent with wind trajec-
tories.
A summary of field tests performed at this site is included in
Table 3 with the date, time and purpose. The early tests in
this table were valuable in evaluating the performance of the
60
-------
MOBILE
LABORATORY
270
WEST
OPEN TO
ROAD AT 5 Km
TREES
AND HILL
AT 3 Km
- HILLS 1.5 Km
ROAD IKm
*
4 TO 6 Km
SMALLTOWN
(TEN I NO)
FEW
HOUSES
AT 1-2 Km
90
\ TREES AT
* 2Km
TREES AT
~2Km
EAST
180
SOUTH
FIGURE 6. DIAGRAM OF FETCH AROUND SITE
DISTANCES ARE NOT TO SCALE.
61
-------
Table 3. SUMMARY OF FIELD TESTS DURING
MARCH 1975 AT CENTRALIA.
Test
#
1
2
3
4
5
5b
6
7
8
9
10
11
12
13
14
15
Day of
Month
12
12
12
12
12
13
13
13
13
13
13
14
14
14
14
15
Time
1134-1332
1339-1429
1605-1825
1840-2029
2309-2355
0045-0311
0715-0910
1200-1315
1330-1358
1506-1546
2001-2045
0845-1020
1021-1123
1244-1420
2030-2118
0317-0417
Purpose
Test aerosol system
Test ozone
Test ozone
Test ozone and S02 horizontal
Test ozone and S02 vertical
S09 gradients
S09 gradients
S02 , 0_, aerosol tests
S09, 0 , aerosol tests
SO , 0.,, aerosol tests
0-, , S09 gradients fixed levels
«J ^
SO9 gradients fixed levels
Aerosol tests S09, O^ middle
level
0-, at upper level
0., S09 gradients fixed levels
S09 , 0-, times manually extende
16
17
18
19
20
21
22
23
24
15
15
16
16
16
16-17
17
17
17
0340-0355
0634-0805
0555-0615
1045-1223
1449-1608
2322-0440
0836-1026
0909-1345
1320-1350
SO,
SO,
fixed levels
0., times manually extended,
fixed levels
03 times manually extended,
fixed levels
Ozone single line
Ozone single line/SO product
tests 2
Aerosol test
NO /NO tests
S09 gradients, fixed levels
NO /NO gradient tests
.X.
Calibration tests
62
-------
sampling and analysis systems and development of data acquisi-
tion procedures, but relatively few of these experiments are
acceptable for complete analysis. During many of these tests
the winds were unacceptable (i.e., from the mobile van direction,
or a poor response direction for the wind system) or some other
ambient condition was unsatisfactory- The tests revealed many
factors relating to the potential performance of systems under
field conditions. Once a system appeared to be operating sat-
isfactorily, several experiments suitable for dry deposition
determinations were performed as ambient conditions allowed.
The aerosol system operated as expected. The systems were
leak tested by sealing the input ports on the tower and check-
ing for flow. For each run nuclepore filters were mounted on
the inlet ports. Air was drawn through the filters for iden-
tical periods while simultaneous micrometeorological data were
collected. Some difficulty was encountered in matching the
flow rates. The pump pressures were matched before starting
the flow through the filters in each experiment. It was noted
that during the experiments the middle level indicated consis-
tently lower flows than the other levels. Attempts to match
the flow were only successful in the last aerosol experiment.
The gas measurement system required considerable development
during the field experiments. Ozone was destroyed or absorbed
by the solenoid valves, so the automatic switching system was
useless. The solenoid valves decreased the 0., concentration
by an order of magnitude. It was assumed that S0_, NO , and
^ X
NO could not be safely switched by these solenoids without
considerable further testing.
A stainless steel Teflon ball-valve system was tested for
switching the gases. Ozone was still removed, but at a lower
rate. An attempt was made to "saturate" the system by running
ozone from the ozone calibration instrument through the valves
63
-------
for a day. Tests just after this procedure indicated that
measurable 0., decay was no longer taking place. Several sets
of vertical gradients were taken which looked promising. How-
ever, tests several days later indicated unrealistic gradients.
The valves were tested again and found to be removing much of
the ozone. It was concluded these valves could not be used
for switching ozone. Additional tests showed that even the
stainless steel Swage lock fittings were destroying enough
ozone to produce significant differences between the fixed
input lines. A single line system was tested for ozone pro-
files and found to give consistently reasonable results.
Problems of matching flows and of differential loss in the
intake lines were thereby avoided.
The ball-valves were tested for absorption of SO,,. The results
indicated that no significant absorption was occurring.
Fixed height profiles of SO,., were taken whenever the meteoro-
logical conditions allowed. The SO from the various intake
lines was manually switched by the ball valves. The SO
levels were normally near or below the threshold of the
instrumentation during the field effort.
The ozone generator in the REM NO /NO monitor failed at the
^"C
beginning of the field effort. A replacement was obtained
as quickly as possible and the NO , NO instrumentation was
X
tested during the final days of the field effort. Data for
both ball-valve switching of the fixed levels and single line
profile were obtained.
64
-------
SECTION VIII
RESULTS
The purpose of the current tests is to study the feasibility of
using the surface flux profile method to determine atmospheric
removal rates under field conditions. This involves the logis-
tics of the instrumentation and data collection described in
the preceeding sections. The ultimate proof is the demonstra-
tion of the ability to compute deposition rates using these
systems. In certain situations, such as rainfall or very high
humidities, certain of the experiments are obviously excluded
and data are not collected. Less obvious excluding cases
occur when conditions appear to be favorable, but the limita-
tions of the instrumentation preclude determination of an
accurate removal rate. This applies to both the meteorological
and pollutant data acquisition systems. To demonstrate the
magnitude of the errors under various conditions, the experi-
mental accuracy will be estimated for the field data tests.
As a result of the briefness and preliminary nature of the field
effort, data will be included in the analysis which were taken
under less than ideal conditions. These data will be valuable
in assessing the overall capability of the system.
Aerosol
The five sets of three samples collected on nuclepore filters
were analyzed for total sulfur and lead. The methodology
development for this analysis was the responsibility of John
Cooper at ORTEC. His report is attached as Appendix A. Based
on his recommendation, the samples were analyzed for sulfur by
two methods; Energy Dispersive X-Ray Fluorescence analysis per-
formed at ORTEC and Wavelength Dispersion X-ray Fluorescence
analysis performed by Ben Paris at Battelle Columbus Laboratory.
65
-------
John Cooper's results are included in Appendix A and the
Battelle Columbus results in Appendix B.
The results are summarized for total sulfur in Table 4. The
first two columns are the results supplied by Cooper and Paris.
The last two columns are these values normalized to the read-
ing at the top level. There is a close correspondence of the
relative values between levels for the sulfur concentrations.
The BCL summary does not include lead data.
The aerosol data for sulfur and lead from John Cooper (Appendix A)
and uncorrected flow rates are given in Table 5. These are the
raw data for computation of air concentrations. The flow rates
need to be adjusted for the reduced pressures under which the
volume flow rate was obtained.
Table 5 contains the air concentrations of the aerosols com-
puted by two assumptions; equal internal pressure on all three
lines in each experiment and different internal pressure in
each line. Both assumptions are included because of the
uncertainty involved in the pressure-temperature correction
factors for the volume of air measured in each line. These
factors are based on an empirical relationship for the correc-
tion determined in laboratory tests based on flow rates for a-
short intake line with a nuclepore filter. The correction
factor derivation was performed for flow greater than 3 x 10
m /sec. Values below this are based on extrapolation of higher
values.
Table 6 lists the correction factors for raw flow rate as a
function of adjusted flow rate. These were used to determine
specific correction factors in Table 5. The correction must
be considered approximate because other influences such as
lengths of lines, number of fittings and variations between
laboratory and field conditions were not included.
66
-------
Table 4. TOTAL AND NORMALIZED SULFUR AEROSOL
CONCENTRATIONS UNCORRECTED FOR FLOW
RATE.
Cooper Paris
(Appendix A) (Appendix B) Cooper Paris
Sample yg/cm^ Total yg v#3 v#3
Ml 3 1,29 ± .02 4.0 1.00 1.00
2 1.27 ± .02 3.8 .98 .95
1 1.43 ± .02 4.2 1.11 1.05
M7 3 2.25 ± .02 6.7 1.00 1.00
2 2.18 ± .02 6.4 .97 .96
1 2.56 ± .02 7.4 1.14 1.10
M8 3 2.81 ± .02 7.6 1.00 1.00
2 2.79 ± .02 7.8 .99 1.03
1 3.02 ± .02 8.7 1.07 1.08
M12 3 2.33 ± .02 6,7 1.00 1.00
2 2.34 ± .02 6.7 1.00 1.00
1 2.64 ± .02 7.4 1.13 1.10
M23 3 0.45 ± .02 1.9 1.00 1.00
2 0.46 ± .02 2.1 1.02 1.11
1 0.41 ± .02 2.1 .91 1.11
67
-------
Table 5. SUMMARY OF COMPUTED AEROSOL AIR CONCENTRATIONS1
CTl
00
Run
No.
M-l
M-7
M-8
M-12
M-23
Height
(m)
15.
6.
•
15.
6.
•
15.
6.
v
15.
6.
•
15.
6.
.
2
1
914
2
1
914
2
1
914
2
1
914
2
1
914
Sample Rate
(m3/s)
Sulfur Air
Concentrations2
(xlQ3)
2.073
1.84
2.25
1.63
1.37
1.71
2.19
2.00
2.31
1.32
1.15
1.44
1.99
2.24
2.26
2.064
1,95
2.15
1.60
1.48
1.63
2.18
2.08
2.23
1.32
1.18
1.41
2.08
2.20
2.21
2
2
2
6
6
6
1
1
1
0
0
0
Lead Air
Concentrations2
yg/m3
.6133
.681
.626
.43
,80
.62
,00
.57
.18
.55
.78
,60
.333
.343
.343
0
0
0
2
2
2
6
6
6
1
1
1
0
0
0
.616*
.641
.656
.47
.60
,76
.02
.31
.39
,55
.73
,64
.319
.349
.351
0
0
0
0
0
0
1
1
1
0
0
0
0
0
0
jag/m3
.233
.25
.29
.67
,80
.69
.22
.39
.21
.33
.33
.29
.37
.28
.29
0.231"
0.237
0.301
0.68
0.74
0.72
1.22
1.34
1.25
0.33
0.32
0.30
0.35
0.29
0.30
lThe accuracy of these data is uncertain, so three places are
listed recognizing that the actual accuracy may be less than this,
2based on a filter area of 7.92 cm^ .
3equal internal pressure assumed for all three lines in each run
(average raw flow rate used).
^different internal pressures assumed for each sample line (raw
flow rate used).
-------
Table 6. FLOW RATE ADJUSTMENT (m3s 1),
Raw Flow Rate Adjusted Flow Rate
1.00 x 10-3 1.00 x 10-3
1.50 x 10-3 1.39 x 10-3
2.00 x 10-3 1.58 x 10-3
2.50 x 10-3 1.73 x 10-3
3.00 x 10-3 1.94 x 10-3
3.50 x 10-3 2.08 x 10-3
4.00 x 10-3 2.22 x 10-3
4.50 x 10-3 2.34 x 10-3
S02
Table 7 contains the average air concentrations of S02 measured
in three tests in the field. The data, time and run number
are listed on the left. The average air concentrations in ppb
along with standard deviations and number of points are given
for the indicated levels. The small number of measurements at
each level combined with the large standard deviation limits
the accuracy of the profiles. The differences between levels
can be expected to be more accurate than indicated by the
standard deviations. These standard deviations include the
effect of natural scatter as well as trends in the concentra-
tions.
The least squares coefficients for all three data sets are
given in Table 8. In all tests, the best fit profiles indicate
deposition of S02. Only in run 2 are the gradients consistent
with a reasonable log profile and large enough to be considered
significantly non-zero gradients. The other two tests have
very small gradients.
69
-------
Table 7. SUMMARY OF SO DATA RESULTS.*
Run Date
1 051375
2 051375
1 CtC. 1 f,7 £.
Time
1330-1358
1506-1546
ninn TOT/I
(
2
3
A
.0.91
Leve]
.4 ±
.8 ±
c; +
m)
. 1
.2(4)
.4(5)
f. M Q ^
(6.1 m)
Level 2
2.5 ± .2 (6)
4.3 ± .7 (8)
2
4
A
Leve
.3 ±
.5 ±
c; +
S m
2l
.2
.5
c;
)
3
(4)
(4)
f\ 7 \
^ *Each set has the average concentration, its standard deviation
o in ppb and the number of points used.in para
Table 8. LEAST SQUARES TO SO2 DATA.*
Run B m
1 2.38 .00276
2 3.86 .251
3 4.48 .0142
*X = mlnz+B, z in meters.
-------
NO , NO
2\,
The NO and NO profile experimental results are summarized in
A
Tables 9 and 10. The run number data, starting time, and dura-
tion of the run are given on the left. The averages, standard
deviations and number of points are also shown for each of the
indicated levels. These average values are stated to ±0.1 ppb
although the excepted absolute accuracy of chemiluminenscent
instruments are an order of magnitude less accurate. The cur-
rent interest is in relative accuracy between values which may
be up to an order of magnitude better than absolute accuracy.
The last column is the ratio of the concentration gradient to
the upper level concentration multiplied by 100. Both short
and long duration tests were run. The recording level was
switched every 3 to 4 minutes. One minute average last value
readings were taken off the chart. The last value was defined
as the one minute period just before the level was changed.
This should be the best value in terms of instrumental equili-
brium. Table 11 contains a series of NO profiles computed from
J\.
intermediate one-minute values read in the 0909-0948 run.
Ozone
The ozone profile results are given in Table 12. These results
are all based on the roving line system, allowing at least two
minutes for an equilibrium reading at each level. The data and
time for each experiment are given on the left. The runs are
numbered for later reference. After each average value the
standard deviation and number of points are listed. The accu-
racy retained in this table is an order of magnitude better than
the stated accuracy of the ozone monitor (±0.001 ppm). This
reflects both the fact these are average values and the current
interest in relative values. The linear regression coefficients
for these data are given in Table 13.
71
-------
Table 9. SUMMARY OF NO PROFILE RESULTS.*
Run
1
2
3
4
5
6
Date Time Start
051675
051675
051775
051775
051775
051775
2322
2340
0419
0909
0945
1139
Duration
(min)
6
44
21
36
41
36
Heights (m)
0.304
51
15
4
18
22
30
.3
.6
.4
.9
.3
.7
± 3
± 2
± 1
± 1
± 3
±
,8
.5
.1
.4
.7
.7
(6)
(15)
(10)
(6)
(5)
(26)
50
15
4
20
23
31
7.
.7
.3
.5
.1
.4
.1
.62
± 4.
± 2.
± 0.
± 2.
± 2.
± 0.
3 (5)
9 (16)
9 (10)
1 (5)
5 (6)
7 (20)
%A
1.3
1.8
-1.7
-6.2
-4.9
-1.4
-J
M
*Each set has the average air concentration in ppb,
the standard deviation, and the number of values in para.
The reason for retension of ±0.1 ppb accuracy is
explained in the text. The %Acolumn is the vertical
percentage change in concentration relative to the upper level
concentration.
-------
Table 10. SUMMARY OF NO PROFILE RESULTS.*
U)
Run
1
2
3
Date
051675
051775
051775
Time Start
2332
1215
1322
Duration
(min)
8
17
12
Heights (m)
6
18
23
0.
.1
.7
.0
304
1.9
5.2
4.1
(4)
(5)
(4)
7
6.7
19.1
24.4
.62
1
5
3
.6(3)
.0(4)
.9(3)
%
-9.5
-2.1
-5.8
*Each set has the average concentration in ppb, the standard deviation
and the number of values. See text for reason for retention of 0.1 ppb
accuracy. The % column is the ratio of the difference of the concentra
tions to the upper level concentration multiplied by 100.
Table 11. SUMMARY OF VARIATION NOTED IN NO PROFILE DATA AT 0909-0948
X
Height
Last values
Second last values
Third last values
.304m
11.9
19.3
18.5
7.62m
20.1
20.0
19.8
Change
-6.2
-3.5
-4.0
*A11 values are NO air concentrations in ppb.
x ^r
-------
Table 12. OZONE PROFILE SUMMARY*.
Heights
Date
051775
051775
051775
051775
051775
051775
Time
0355-500
0504-525
0836-930
0930-1030
1139-1225
1226-1320
Run
#1
#2
#3
#4
#5
#6
0
0
0
0
0
0
0,
.0217
.0217
.0351
.0392
.0442
.0435
.304m
+
+
+
+
+
+
.0023
.0008
.0033
.0010
.0014
.0015
(15)
(3)
(6)
(9)
(6)
(8)
no
0.
0.
0.
0.
0.
2.44m
data
0231 ±
0372 ±
0403 ±
0449 ±
0448 ±
.0010
.0034
.0010
.0014
.0016
(6)
(13)
(15)
(14)
(15)
7.62m
0.0249
0.0241 ±
0.0372 ±
0.0407 ±
0.0455 ±
0.0456 ±
.0018 (2)
.0033(7)
.0012(8)
.0017(7)
.0013(7)
*Values in ppm with standard deviation and number of data
points after each value. See text for reason for retention
of ±0.0001 ppm.
-------
Table *13. LEAST SQUARES FIT TO OZONE DATA*
Run # B m
1 .0229 .000993
2 .0225 .000736
3 .0361 .000696
4 .0398 .000473
5 .0447 .000362
6 .0443 .000648
*X = m(lnz)+B, where x = concentration in ppm
and z is height in meters
Meteorological Data
Meteorological data collected during the field tests were
reduced and summarized for the periods of interest. Tables
14 thru 19 contain these results for the experiments identi-
fied for possible deposition computations in the previous
sections. Each table is labeled relative to the applicable
experiments. The averages of wind speed (u) , vertical velo-
city (w), products of wind speed and vertical velocity
(uw), air temperature (T) and products of air temperature
and vertical velocity ("5w) are given as a function of measure-
ment height; 3 = upper level, 2 = middle level, 1 = lower level,
Omitted lower level values do not apply to the current analysis,
Notation is made of the wind direction mode with the extremes
on both sides in ten degree sector units. Also listed are the
gradients of temperature and wind speed, as well as the
Richardson number; all based on levels 1 and 3. The 0w values
are based on w lagged 2.0 seconds to allow for the 0 response
time.
75
-------
Table 14. SUMMARY OF METEOROLOGICAL DATA FOR SO2 RUNS 1-3.*
S02
Run#
(28 min)
(40 min)
Ch
(74 min)
Height
(m)
3
2
1
3
2
1
3
2
1
u
(m/s)
4
3
3
3
2
2
2
1
1
.29
.70
.30
.12
.69
.47
.24
.92
.70
w
(m/s)
-.060
.036
.061
-.026
.039
.048
.043
.050
.060
uw
(m2/s2)
-.307
.0445
-.153
-.024
.113
.0929
-
T
(°C)
20.69
21.09
21.76
23.11
23.51
24.15
14.08
14.17
14.86
W0
(°A m/s)
-17.8
- 7.87
12.5
Notes
DM = 300° - 330° - 010'
DT = .0605 °C/m
DU = .0647 m/s m
RI = .482
DM = 300° - 320° - 40'
DT = .0584
DU = .0425
RI = 1.07
DM = 20° - 150° - 260'
DT = -.0411
DU = .0357
RI = -1.10
u is mean wind speed, w is mean vertical velocity, uw_is the average of the
products of u and w, T is the mean air temperature, w9 is the average product
of W and T, DM are the extremes of wind direction with the mode in the middle,
DT and DU are average temperature and wind speed gradients, RI is Richardson
number. DU, DT and RI are defined by heights 1 and 3.
-------
Table 15. SUMMARY OF METEOROLOGICAL DATA FOR NO RUNS 1-3.*
N0x
Run*
1
6 min)
(33 min)
(21 min)
Height
(m)
1
2
3
1
2
3
1
2
3
u
(m/s)
3.47
2.88
2.24
2.67
2.32
1.76
2.29
1.91
1.16
w
(m/s)
.071
.080
.100
.049
.054
.079
.043
.034
.036
uw
(m2/s2)
.242
.194
.133
.127
.0951
.0648
—
T
(°C)
9.33
9.44
9.41
9.29
9.36
9.24
7.45
7.18
6.45
wG
(°A m/s)
19.7
13.7
12.1
DM
DT
DU
RI
DM
DT
DU
RI
Notes
DM = 30° - 100° - 120°
DT = .0045 °c/m
DU = .0813 m/s m
RI = .0236
= 90° - 120° - 150°
= .0132
= .129
140° - 180°
.0755
.0742
.480
- 190
*Notation is defined in Table 14.
-------
Table 16. SUMMARY OF METEOROLOGICAL DATA FOR NO.. RUNS 4-6.*
CO
x Height
Run# (m)
4
(35 min)
5
(38 min)
6
(36 min)
1
2
3
1
2
3
1
2
u
(m/s)
1
1
1
1
1
1
2
1
.76
.59
.49
.90
.70
.31
.21
.99
w
(m/s)
.018
.052
.090
.049
.047
.083
.019
.050
uw
(mVs2)
.0789
.110
-
.0905
.0679
.-
.0355
.0955
T w9
(°C) (°A m/s)
10
11
11
11
11
12
14
14
.95 5.27
.02
,42
.86 14.0
.91
.40
.18 5.64
.13
DM =
DT =
DU =
RI =
DM =
DT =
DU =
RI =
DM =
DT =
DU =
RI =
Notes
90° - 130° - 270°
-.0210 °c/m
.0179 m/s m
-2.26
60° - 180° - 230°
-.0260
.0252
-1.41
90° - 150° - 230°
-.0323
.0278
-1.43
*Notation is the same as Table 14
-------
Table 17. SUMMARY OF METEOROLOGICAL DATA FOR NO RUNS 1-3.*
NO
Run#
Height u w
(m) (m/s) (m/s)
uw
(m2/s2)
T
W0
(°A m/s)
Notes
( 8 min)
1
2
3
3.31
2.76
2.12
.018
.053
.080
,0502
.119
9.33
9.44
9.40
4.98
DM = 70° - 100° - 120°
DT = .0775 °c/m
DU = .00506 m/s m
RI = .0292
(17 min)
1
2
3
2.44
2.05
1.78
019
036
095
0239
0654
14.79
14.79
15.48
-5.72
DM = 70° - 130°
DT = -.0361
DU = .0439
RI = -.640
- 22'
(12 min)
1
2
3
2.20
1.75
1.54
020
040
082
,102
,0598
16.19
16.17
16.85
6.17
DM
DT
DU
RI
0° - 110°
-.0336
.0433
-.606
- 180'
*Notation is defined in Table 14.
-------
Table 18.
SUMMARY OF METEOROLOGICAL DATA FOR 0., RUNS 1-3.*
°3
Runt
Height u w
(m) (m/s) (m/s)
uw
(m2/s2)
T
(°A m/s)
Notes
(65 min)
1
2
3
2.18
1.80
1.13
.062
.036
.040
,138
.066
7.48
7.24
6.68
17.4
DM
DT
DU
RI
100° - 180°
.0626 °c/m
.0684 m/s m
.468
- 230'
(21 min)
00
o
1
2
3
3.27
2.86
1.83
192
,093
097
628
261
171
7.73
7.48
7.19
53.
DM - 160° - 200°
DT = .0448
DU = .0946
RI - .175
- 210'
1
2
3
2.04
1.82
1.63
,051
066
,092
129
128
156
10.42
10.51
10.86
14.4
DM = 80° - 130°
DT = -.0191
DU - .0267
RI - -.925
- 230'
*Notation is defined in Table 14.
-------
Table 19. SUMMARY OF METEOROLOGICAL DATA FOR
RUNS 4-6.*
Run#
(60 min)
Height
(m)
u
(46 min)
CO
1 1.82
2 1.61
3 1.45
1
2
3
1
2
3
2.21
1.99
1.76
2.18
1.85
1.56
w uw
,2 /r,2
(m/s) (m/s) (mVs2)
T W0
(°C) (°A m/s)
.032
.048
.087
-.007
.032
.103
.017
.057
.075
.0797
.0793
-.0065
.061
.0383
.113
-
11.77
11.82
12.30
14.27
14.23
14.89
15.44
15.47
16.75
9.01
-
-1.96
5.13
Notes
DM
DT
DU
RI
DM
DT
DU
RI
DM
DT
DU
RI
20° - 160° - 270°
-.0253 °c/m
.0243 m/s m
-1.47
80° - 130°
-.0310
-.0294
-1.22
20° - 130°
.0431
.0405
-.894
- 220
- 300
*Notation is defined in Table 14.
-------
Table 20 summarizes the diffusivities and their relative experi'
mental accuracy, The resolution shows considerable variation
and is generally poor. Several of the momentum values are
significant, but all the heat values have such large experi-
mental errors that they cannot be considered for computations.
Deposition Parameter Computations
Pollutant profiles and meteorological summaries given in pre-
ceeding sections are used for computation of deposition para-
meters. The computation is carried through to the final pro-
duct in all cases to identify the sources and magnitude of
experimental error. Anomalous results may be reasonable when
the experimental accuracy is considered. More important,
determining error magnitude clearly identifies the most
reliable cases. It is in these latter cases where the results
will demonstrate the consistency and applicability of the
approach.
Separate tables of the deposition computations are given for
each of the gases using the momentum eddy diffusivity (Tables
21 to 24). Run numbers on the left correspond to numbers
identified in the previous sections for each gas. The frac-
tional pollutant gradient values in the next columns on the
left are computed from values taken directly from previous
tables. The Y^ and K^ columns contain eddy diffusivities
adjusted linearly with height to correspond to the same geo-
metric height as the pollutant concentration gradient. The
Vm and Vh columns contain the computed values of deposition
velocity using the previous columns. As noted earlier, esti-
mates of experimental accuracy have been carried through the
computation to identify the most reliable case. As noted, the
accuracy of the gradients has been carried through with several
values. One percent is the best value possible. The other
values are given to demonstrate the change in reliance with
82
-------
Table 20.
COMPUTED EDDY DIFFUSIVITIES FOR MOMENTUM
Km AND SENSIBLE HEAT Kh AT 5.53m.
Gas
S02
NO
X
NO
0
3
Run #
1
1*
2
2*
3
3*
1
1*
2
2*
3
3*
4
4*
5
5*
6
6*
1
1*
2
2*
3
3*
1
1*
2
2*
3
3*
4
4*
5
5*
6
6*
— 2
K (m /s)
.767
1.33
1.65
2.18
- .467
.843E-01
.538E-01
.340
- .103
- .282E-01
.454E-01
.112E-02
-2.64
-1.63
.103
.378
.233
.120
.121
.255
-1.60
.186
-1.34
.290
- .415E-01
- .107E-01
- .248E-02
.621E-01
- .935
- .755E-01
- .883
- .904E-01
- .307
.593E-01
- .306E-01
- .156
2 / >
.382
.257
.353
.641
.389
.802
.226
.234
.200
.174
.103
.549E-01
.827
3.95
.646
.840
.226
.999
.534E-01
.146
.895E-01
.396
.215
.606
.167
.676E-01
.830
.132
.744
1.25
.531
1.25
.605E-01
.487
.111
.541
K, (m /s)
- 5,76
.431
-2.31
85.8
7.40
- .756E-01
1.38
3.50
-6.13
24.3
-14.2
9.71
- .499E-01
-1.82
6.76
.433
-1.40
5.10
a o
Kh(m /s)
97.4
.101E 03
.140E 03
.160E 04
.436E 03
74.4
.271E 03
.220E 03
.178E 03
.113E .04
.160E 03
.172E 03
89.9
.130E 03
.300E 03
.226E 03
.185E 03
.134E 03
*meteorology data from middle and lower level
used.
83
-------
Table 21. DEPOSITION COMPUTATION FOR SO,
±a.
v
Run#
f1
-1
m
-2 ± 7
-109 ± 7
-6 ± 7
dm
0
1
1
2
0
K 2
m
2 -]
m s
.517 ±
.33 ±
.11 ±
.18 ±
.314 ±
.084 ±
\T
"dm
.335
.26
.24
.64
.262
.80
cm
0.
0.
1.
2.
_
.
^1
s
10
27
21
38
19
05
Froru
1%
-1
cm s
0.37
0.95
.27
.66
.27
.81
ue Accuracy
3%
_^L
cm s
0.70
2.8
.35
.77
.68
.85
f ASSU
5
cm
1.
4.
•
1.
.
mpric
Dn
% 10%
-1
s
8
7
47
95
12
94
cm
3.
9.
1.
1.
2.
1.
-1
s
5
5
00
54
22
27
CO
Footnotes:
i n
f =
X
x
z2
Az
x 10 where z± = 0.914 m and 22 = 15.2 m. Accuracy is based
on ±1% gradient error.
)
Values of K adjusted to correspond to same heights as the pollutant gradients.
The first Km is based on the upper and lower level data and the second Km is
based on the middle and lower data.
-------
Table 22. DEPOSITION COMPUTATION FOR NO,
V
dm
oo
t_n
Run#
1
2
3
4
5
6
f1
m-1
17 ± 14
25 ± 14
-24 ± 14
-85 ± 14
-68 ± 14
-19 ± 14
Footnotes :
i
f =
I x AX.
X ,, Az
K 2
m
2
m £
+ .014
+ .139
-.028
-.012
+ .013
+ .001
-.73
-.67
.03
.15
.064
.05
r1
+
+
+
+
+
+
+
±1
+
+
+
+
.062
.095
.055
.071
.028
.022
.23
.6
.18
.34
.062
.41
dm
cm s
.00238
.0236
-.0070
-.0030
.0031
.0002
-.62
-.57
.02
.02
.01
.01
Frori.
1%
cm s
.011
.025
.014
.018
.007
.004
.22
1.36
.12
.23
.012
.082
ie Accurac
3%
cm s
.012
.061
.018
.018
.009
.004
.37
1.39
.12
.23
.024
.085
-y Assumpi
5%
_T
cm s
.014
.100
.024
.020
.011
.004
.55
1.44
.12
.25
.038
.084
tion
10%
cm s
.022
.20
.042
.024
.019
.005
1.04
1.65
.13
.31
.074
.11
x 10 where z-, = 0.304 m and Z2 = 7.62 m. Accuracy is
based on ±1% gradient error.
>
Values of Km are adjusted to correspond to the same height as the pollutant
gradients. The first Km for each run is based on the upper and lower level
data, and the second Km for each run is based on the middle and lower level
data.
-------
Table 23. DEPOSITION COMPUTATION FOR NO
V
dm
Run#
1
2
3
f1
_ i
m m
-129 ± 14 .033
.104
-294 ± 14 -.43
,076
-789 ± 14 -.36
-.118
K 2
m
2
s
+
+
+
+
+
+
-1
.014
.060
.024
.162
.06
.247
dm
-1
cm s
.043
.13
-1.26
.22
-2.84
.931
fror u_
1%
-1
cm s
.019
.08
.09
.48
.48
1.95
e Accuracy
3%
_n
cm s
.023
.09
.19
.48
.50
1.95
Assumptio
5%
-1
cm s
.030
1.11
.31
.48
.54
1.95
n
10%
-1
cm s
.050
.16
.61
.49
.69
1.96
00
Ol
Footnotes:
X
x -r-^- x 10 where z.. = 0.304 m and z_ = 7.62 m. Accuracy is
Az 12
z2 ziz2
based on ±1% gradient error.
>
Values of Km are adjusted to correspond to the same height as the pollutant
gradients. The first Km for each run is based on the upper and lower level data,
and the second Km for each run is based on the middle and lower level data.
-------
Table 24. DEPOSITION COMPUTATION FOR O,
±0
V
00
•-J
Run# f1
— ]
1 -176 ±
2 -135 ±
3 -81. 7±
4 51.1 ±
5 -38.3 ±
6 -62.3 ±
Footnotes :
lf - l
X __
L
14
14
14
14
14
14
AX
Az
m
2 -1
m s
.011 ±
-.004 ±
-.001 ±
.025 ±
.253 +
-.0311±
-.24 +
-.04 ±
-.083 ±
.024 ±
-.008 ±
-.064 ±
.045
.028
.22
.054
.20
.51
.14
.51
.016
.20
.030
.22
vdm
cm s
.019
.007
.001
.034
-.20
-.02
-.12
-.02
-.032
.01
-.005
-.04
1%
cm s
.078
.049
.22
.073
.17
.42
.08
.26
.013
.03
.019
.14
3%
-1
cm s
.078
.049
.22
.074
.19
.42
,12
.26
.035
.08
.019
.14
5%
cm s
.078
.049
.22
.075
.24
.42
.18
.26
.058
.13
.020
.14
10%
cm s
.079
.049
.22
.081
.39
.42
.28
.27
.12
.25
.022
.16
x 10 where z. = 0.304 m and
= 7.62 m. Accuracy is
z2 ziz2
based on ±1% gradient error.
Values of Km are adjusted to correspond to the same height as the pollutant
gradients. The first Km for each run is based on the upper and lower level
data, and the second Km for each run is based on the middle and lower level
data.
-------
the decrease in profile accuracy under various conditions.
Using the best accuracy possible indicates the applicability
of the methodology in a given situation; that is, if a cer-
tain percentage profile accuracy is too limiting, then the
methodology cannot be realistically applied under that set
of conditions. In many cases the relative accuracy is not
as good as ±1%.
88
-------
SECTION IX
DISCUSSION
Aerosol
The aerosol results obtained by two independent laboratories
demonstrate that the necessary relative accuracy on the order
of 1 to 2% of the concentrations has been achieved for the
filter analysis. The relative accuracy applies in computing
the deposition velocity. An actual aerosol flux computation
will depend on the absolute accuracy of the pollutant concen-
tration. The achievable accuracy is on the order of ±20% with
application of corrections not made in the current study. These
results show a reasonable factor between the achievable absolute
and relative accuracy in the filter concentration data.
Much of the consistent variation in the filter concentration
data was traced to pressure variations within the air intake
lines. These variations could not be adequately accounted for
in the current tests. The pressure in the air intake lines is
a function of the air temperature within the line, the air
pump setting, the filter resistance to flow, and the air intake
line resistance to flow. The latter varies with the length
and size of the air line as well as the number of flow restric-
tive fittings in the air line. The flow rate-based correction
factors for pressure within the lines accounts only for the
relationship between the air pump setting and the filter resis-
tance.
The sulfur aerosol air concentration is derived assuming two
pressure correction factors. First the factor is based on the
average flow rate for all three lines. This results in profiles
that are as badly kinked as the raw data. Second, the factor
is based on the actual flow rates for each data point. This
89
-------
results in reasonable vertical profiles. The computed air
concentration decreases with height. This is felt to be a
result of the shortcoming of the pressure correction factor.
The suggested explanation is that the line pressure is signi-
ficantly affected by factors such as the resistance to flow
in the air intake line. In such case the pressure in the
line is progressively overestimated from the bottom to the
top data levels on the tower. This results in an under-
estimation of air concentration at higher tower levels, as
a result of progressively greater overestimation of the total
sample volume. Hence the profiles are not felt to be indicat-
ing either the direction or magnitude of sulfur aerosol re-
moval, but rather a result of the pressure correction.
The comments on the sulfur aerosol results apply also to the
lead aerosol results. The lead profiles appear to have con-
siderably more scatter than the sulfur, and the implication
is that they are not as accurate. Flow rate inaccuracy precluded
deposition velocity computation in the current effort.
Meteorology
The use of the upper level of meteorology data for eddy flux
computations is considered superior in most cases than is
the middle level. The response time of the wind sensors for
flux computations is less limiting at the upper level. Eddy
flux estimates made using the middle level may have an addi-
tional source of error not considered in the estimates of
experimental accuracy, this is the lack of response to higher
frequency atmospheric motions. The error could be estimated
and corrected by spectral analysis, but this was beyond
the scope of the current effort. The upper level values
are considered most reliable except when the constant flux
surface layer appears to be below the upper level. Then
the middle level is considered better.
90
-------
Both the upper and middle levels were presented in the results.
The values were consistent with larger experimental errors esti-
mated for the middle level data. The large experimental errors
in the eddy diffusivity determination are not suprising. A cer-
tain correspondence can be expected between the accuracy of the
pollutant profiles and the eddy diffusivities. The conditions
that lead to difficulties in accurate pollutant profile deter-
mination also lead to similar difficulties in accurate wind
velocity profile determination. In addition, the eddy diffusi-
vity determination requires direct flux estimates. This was the
major source of error in many cases in the current results.
The low accuracy of a term resulting from the substraction of
two numbers of near equal magnitude occurred in all the heat
computations.
Gases
SO
2. The SO profiles were summarized in Tables 7 and 8 in
Section VII. Although the gradients are negative in all cases,
only in Run #2 is it significantly non-zero. This is reflected
in the large estimated errors in the deposition velocities even
at the 1% assumed profile accuracy compared to the computed
deposition velocity for Runs #1 and 3 in Table 21. The magni-
tude of the error is the same for Run #2, but the relative
accuracy is much better. Runs 1 and 2 had significant winds
from the mobile van sector direction. The results must be
considered suspect since the mobile van may influence both pol-
lutant and wind characteristics at the tower.
The deposition velocity obtained for Run #2 has a reasonable
magnitude based on previously determined values as well as an
experimental accuracy that is within reasonable limits. For
this case the "aerodynamic" deposition velocity (using Equation
14) is 2.2 cm/s. In terms of resistances to SO2 flux these
91
-------
produce 0.44, 0.90, and .46 s/cm for the atmospheric (inverse
of aerodynamic resistance), total resistance (inverse of
computed V ), and surface resistance (difference of first two).
The surface and atmospheric terms are nearly equal in this
case.
x' an 3. The NO and NO results and notes were summari-
x
zed in Tables 9 and 10. A number of short experiments during
the night are included along with longer tests. These former
were part of the NO , NO system shake-down and were not taken
specifically for gradients. However, since they were inter-
esting they have been included. The low turbulence condi-
tions during the night may shorten the time required to obtain
a significant profile. The data based on the longer time
periods indicated dry deposition of NO without exception.
Some of the shorter periods indicate the ground as a source
of NO but the short averaging time limits the application
of the flux model and the gradient is of the same order as
the 1% profile accuracy value. It is clear that little signi-
ficance can be placed on the positive sign of these gradients.
Tables 22 to 24 show that the diffusivity values were limiting
in all NO , NO and 0 runs. The variation was so great that
•A. «J
positive and negative values were nearly equally common. NO
runs 4 and 5, all three NO runs, and most of the ozone runs
had gradients that appear to be significant. These may yield
further information with parameterization of the eddy diffusi-
vity values in subsequent analysis. Considering the large
experimental error, little significance can be placed on most
of the computed deposition velocities in these tables. One
exception is the NO runs where the relative errors are lowest.
All three NO runs have gradients that indicate significant
deposition is occurring. The middle levels are felt to be the
92
-------
best for estimating the diffusivity for these runs. In the NO
runs 2 and 3 (Table 23) the use of upper level momentum flux
values produced negative diffusivities and the middle level
positive diffusivities. Inspection of the meteorology summaries
for the NO runs leads us to conclude that for these cases the
constant flux surface layer is below the upper level and the
middle level data are more appropriate. This produces deposi-
tion velocities for NO in the range of 0.08 to 0.12 cm/s. A
low level of confidence in these numbers is primarily a function
of the accuracy of the eddy diffusivity determination. The
relative errors show only small changes based on several values
of assumed profile accuracy.
With the limited size of the current data a definitive resolu-
tion of the accuracy of the pollutant profile is not possible.
The listed accuracy of the gas concentrations is not claimed
to be obtained in an absolute sense, but rather is stated to
study the relative accuracy obtained by consideration of the
profiles. The vertical consistency of the profiles indicate
that the relative accuracy is better than the absolute accuracy
and is sufficient for deposition computations.
The accuracy of profiles and the turbulence is interrelated.
Small vertical pollutant gradients may be expected under great-
er turbulence. Pollutant gradients which are not significantly
different from zero do not necessarily imply a zero flux. We
have attempted to show this in the error analysis. That is,
the range of possible values for the deposition velocity may
be significantly non-zero. If it is not, then the measurements
are implying a near zero flux to the surface. The computed
ranges for deposition velocities do generally include the
reasonable range of values that might be expected for the depo-
sition velocity, and in cases where the computed range is nar-
row the magnitude of the deposition velocity is reasonable in
most cases.
93
-------
SECTION X
REFERENCES
Thorn, A.S., J. B. Stewart, H. R. Oliver, and J. H. C.
Gash. Comparison of Aerodynamic and Energy Budget
Estimates of Fluxes over a Pine Forest. Quart. J. R.
Meteor. Soc. 101: 93-105, 1975.
Aylor, D. E. Deposition of Particles in a Plant
Canopy. J. Appl. Meteorol. 14: 52-57, 1975.
Bolin, B., G. Aspling, and C. Persson. Residence
Time of Atmospheric Pollutants as Dependent on Source
Characteristics, Atmospheric Diffusion Processes and
Sink Mechanisms. Tellus 24: 1-2, 1974.
Scriven, R. A. and B. E. A. Fisher. The Long Range
Transport of Airborne Material and Its Removal by
Deposition and Washout - I. General Considerations,
II. The Effect of Turbulent Diffusion. Atmos. Environ,
9: 49-58. 1975.
Hosker, Jr., R. P. Estimates of Dry Deposition and
Plume Depletion over Forest and Grassland. (Presented
at IAEA Symposium on the Physical Behavior of Radio-
active Contaminants in the Atmosphere. Vienna.
November 12-16, 1973).
Angeletti, L. La Contamination Des Paturages par
L'iode-131 1'ere Purtie: Vitesse de Depot de L'iode-
131. EUR-4872. (In French).
Droppo, J. G. Dry Deposition Processes on Vegetation
Canopies. Proc. Atmosphere-Surface Exchange Particulate
and Gaseous Pollutants - 1974 Symposium. Richland,
WA. Septebmer 4-6, 1974. ERDA Symposium Series 38.
CONF-740 921.
Hill, A. C. Vegetation: A Sink for Atmospheric Pollu-
tants. J. Air Pollu. Con. Assoc. 2_1(6): 341-346,
June 1971.
94
-------
9 Garland, J. A. Dry Deposition of S02 and Other Gases .
Proc. Atmosphere-Surface Exchange Particulate and Gaseous
Pollutants - 1974 Symposium. Richland, WA. September 4-6,
1974. ERDA Symposium Series 38. CONF-740 921.
10 Slinn, W. G. N. Atmospheric Aerosols: Sources, Concen-
trations and Budgets. Proc. Atmosphere-Surface Exchange
Particulate and Gaseous Pollutants - 1974 Symposium.
Richland, WA. September 4-6, 1974. ERDA Symposium Series
38. CONF-740-921.
11 Hicks, B. B. Some Micrometeorological Aspects of Pollu-
tant Deposition Rates Near the Surface. Proc. Atmosphere-
Surface Exchange Particulate and Gaseous Pollutants -
1974 Symposium. Richland, WA. September 4-6, 1974.
CONF-740-921
12 Bennett, J. H., A. C. Hill, and D. M. Gates. A Model for
Gaseous Pollutant Sorption by Leaves. J. Air Pollu. Con.
Assoc. 23(11): 957-962, November 1973.
13 Rasmussen, K. H., M. Taheri, and R. L. Kabel. Sources
and Natural Removal Processes for Some Atmospheric Pollu-
tants. Center for Air Environ. Studies. Penn State
University. June 1974. EPA-650/4-74-032.
14 Select Research Group in Air Pollution Meteorology.
Second Annual Progress Report: Volume II. Penn. State
University. Grant No. R-800397. Office of Research and
Development, Environmental Protection Agency, Washington,
D.C. PB-241 391, EPA-650/4-74-045-6. September 1974.
15 deCormis, L. Some Aspects of Sulfur Absorption by Plants
Subjected to an Atmosphere Containing S02. Proc. Air
Pollution First European Congress. Netherlands. 1968.
16 Chamberlain, A. C. and R. C. Chadwick. Deposition of
Airborne Radioiodine Vapor. Nucleonics ILL (8) : 22-25,
August 1953.
95
-------
17 Chamberlain, A. C. Aspects of the Deposition of Radio-
active and Other Gases and Particles. Aerodynamic
Capture of Particles. New York, Pergamon Press,
1960. p 63-68.
18 Chamberlain, A. C. Transport of Gases to and from
Grass and Grass-like Surfaces. Proc. Roy. Soc. (A)
London. 290: 236-265, 1966.
19 Chamberlain, A. C. and R. C. Chadwick. Transport of
Iodine from Atmosphere to Ground. Tellus 18: 226-
237, 1966.
20 Owen, P. R. and W. R. Thompson. Heat Transfer Areas
Rough Surfaces. J. Fluid Mech. 15: 321, 1963.
21 Markee, Jr., E. H. Studies of the Transfer of Radio-
iodine Gas to and from Natural Surfaces. Environmental
Research Laboratories, Air Resources Laboratories,
Silver Spring, MD. April 1971. ERL-ARL-29.
22 Bunch, D. F. Controlled Environmental Radioiodine
Tests (CERT), Progress Report No. 2. AEC Operations
Office, Idaho Falls, ID. IDO-12053. 1966. p. 33ff.
23 Bunch, D. F. Controlled Environmental Radioiodine
Tests (CERT), Progress Report No. 3. AEC Operations
Office, Idaho Falls, ID. IDO-12063. 1968. p. 77ff.
24 Hawley, C. A. Controlled Environmental Radioiodine
Tests (CERT) at NRTS, 1965 Progress Report. AEC
Operations Office, Idaho Fall's, ID. IDO-12047.
1966. p. 41ff.
25 Hawley, C. A., C. W. Sill, G. L. Voelz, and N. F.
Islitzer. Controlled Environmental Radioiodine Tests
(CERT) at NRTS. AEC Operations Office, Idaho Falls,
ID. IDO-12035. 1964. p. 65ff.
96
-------
26 Zimbrick, J. D. and P. G. Voilleque. Controlled Environ-
mental Radioiodine Tests (CERT) at NRTS, Progress Report
No. 4. AEC Operations Office, Idaho Falls, ID. IDO-12065
1969. p. 78ff.
27 Garland, J. A,, D. H. F. Atkins, C. J. Readings, and
S. J, Cauhey. Technical Note: Deposition of Gaseous
Sulphur Dioxide to the Ground. Atmos. Environ. 8:
75-79, 1974.
28 Owers, M. J. and A. W. Powell. Deposition Velocity of
Sulfur Dioxide on Land and Water Surfaces Using a 35c
Tracer Method. Atmos. Environ. 8: 63-67, 1974.
29 Sheperd, J. G. Measurements of the Direct Deposition of
Sulphur Dioxide onto Grass and Water by the Profile
Method. Atmos. Environ. 8: 69-74, 1974.
30a Whelpdale, D. M. and R. W. Shaw. Sulphur Dioxide
Removal by Turbulent Transfer over Grass, Snow and
Water Surfaces. Tellus 26: 196-205, 1974.
3Ob Robinson, E. and R. C. Robbins. Gaseous Nitrogen Com-
pound Pollutants from Urban and Natural Sources.
J. Air Pollu. Con. Assoc. 20: 303-306, 1970.
31 Robinson, E. and R. C. Robbins. Sources, Abundance and
Fate of Gaseous Atmospheric Pollutants. Final Report,
SRI Project PR-6755. 1968.
32 Tingey, D. T. Foliar Absorption of Nitrogen Dioxide.
Dept. of Botany, University of Utah. M.A. Thesis. 1968
33 Israel, G. W. Deposition Velocity of Gaseous Fluorides
on Alfalfa. Atmos. Environ. 8: 1329-1330, 1974.
34 Kelley, Jr., J. J. and J. D. McTaggart Cowan. Vertical
Gradient of Net Oxidant Near the Ground Surface at
Barrow, Alaska. J. Geophys. Res. 73: 3328-3330, 1968.
97
-------
35 Galbally, I. Ozone Variation and Destruction in the
Amospheric Surface Layer. Nature, 218: 456-457. 1968
36 Galbally, I. Evaluation of the Ehmert Technique for
Measuring Ozone Profiles in the Atmospheric Surface
Layer. J. Geophys. Res. 74; 6869-6872. 1968.
37 Aldaz, L. Flux Measurements of Atmospheric Ozone Near
Land and Water. J. Geophys. Res. 74: 6934-6946.
1969.
38 Regener, V. H. and L. Aldaz. Turbulent Transport Near
the Ground as Determined from Measurements of the Ozone
Flux and the Ozone Gradient. J. Geophys. Res. 74:
6935-6942, 1969.
39 Galbally, I. Ozone Profiles and Ozone Fluxes in the
Atmospheric Surface Layer. Quart. J. R. Meteor. Soc.
97: 18-29, 1971.
40 Megaw, W. J. and R. C. Chadwick. Some Field Experi-
ments on the Release and Deposition of Fission
Products and Thoria. AERE-HP-M-114. December 1956.
41 Convair. Fission Products Field Release Test I.
Air Force Nuclear Aircraft Research Facility.
AFSWC-TR-59-44. 1959-
42 Fission Products Field Release Test II. Air Force
Nuclear Aircraft Research Facility. AFSWC-TR-60-26
1960.
43 Slade, D. H. Meteorology and Atomic Energy 1968.
U.S. Dept. of Commerce, Springfield, VA. TID-24190.
1968.
44 Servant, J. Deposition of Atmospheric Lead Particles
to Natural Surfaces in Field Experiments. Proc. Atmos-
sphere-Surface Exchange Particulate and Gaseous
Pollutants - 1974 Symposium. Richland, WA. September
4-6, 1974. ERDA Symposium Series 38. CONF-740 921.
98
-------
45 Cawse, P. A. and D. H. Peirson. An Analytical Study of
Trace Elements in the Atmospheric Environment.
AERE-R-7134. 1972.
46 Simpson, C. L. Some Measurements of the Deposition of
Matter and Its Relation to Diffusion from a Continuous
Point Source in a Stable Atmosphere. General Electric,
Richland, WA. HW-69292 REV. April 1961.
47 Islitzer, N. F. and R. K. Dumbauld. Atmospheric Diffu-
sion-Deposition Studies over Flat Terrain. Intern.
J. Air Water Pollution. 7(11-12): 999-1022, 1963.
48 Whelpdale, D. M. Dry Deposition over the Great Lakes.
Proc. Atmosphere-Surface Exchange Particulate and
Gaseous Pollutants - 1974 Symposium. Richland, WA.
September 4-6, 1974. ERDA Symposium Series 38.
CONF-740 921.
49 Atmosphere-Surface Exchange of Particulate and Gaseous
Pollutants. Oak Ridge, Division of Technical Information
Exchange, 1974. 988 p. CONF-740 921.
50 Gifford, F. A. and D. H. Pack. Surface Deposition of
Airborne Material. Nuclear Safety. 3_(4) : June 1962.
p. 76ff.
51 Nickola, P. W. and G. H. Clark. Measurement of Parti-
culate Plume Depletion by Comparison with Inert Gas
Plumes. Pacific Northwest Laboratory Annual Report for
1973 to the USAEC Division of Biomedical and Environ-
mental Research, Part 3: Atmospheric Sciences.
BNWL-1850 PT3, Battelle, Pacific Northwest Laboratories,
Richland, WA, April 1974. pp. 49-60.
52 Small, S. H. Wet and Dry Deposition of Fallout Materials
at Kjeller. Tellus. 12; 308-314, 1960.
53 Marenco, A. and J. Fontan. Influence of Dry Deposition
on Residence Time in Troposphere of Particulate Pollu-
tants. Proc. Atmosphere-Surface Exchange Particulate
and Gaseous Pollutants - 1974 Symposium. Richland, WA.
September 4-6, 1974. ERDA Symposium Series 38.
CONF-740 921.
99
-------
54 Sehmel, G. A. Turbulent Deposition of Monodispersed
Particles on Simulated Grass. II. Assessment of
Airborne Particles. Springfield, C. Thomas Publishing
Company. 1972. pp. 18-42.
55 Sehmel, G. A. and S. L. Sutter. Particle Deposition
Rates on a Water Surface as a Function of Particle
Diameter and Air Velocity- Pacific Northwest Labora-
tory Annual Report for 1973 to the USAEC Division of
Biomedical and Environmental Research, Part 3:
Atmospheric Sciences. BNWL-1850 PT3, Battelle, Pacific
Northwest Laboratories, Richland, WA, April 1974.
pp. 171-174.
56 Sehmel, G. A. and W. H. Hodgson. Particle and Gaseous
Removal in the Atmosphere by Dry Deposition. Proc.
Symposium on Atmospheric Diffusion and Air Pollution,
Santa Barbara, CA, September 9-13, 1974.
57 Sehmel, G. A. and W. H. Hodgson. Particle Dry Deposi-
tion Velocities. Proc. Atmosphere-Surface Exchange
Particulate and Gaseous Pollutants - 1974 Symposium.
Richland, WA. September 4-6, 1974. ERDA Symposium
Series 38. CONF-740 921.
58 Hicks, B. -B. Transfer of S02 and Other Reactive Gases
Across the Air-Sea Interface. (To be published in
Tellus).
59 Heinemann, K. and K. J. Vogt. Deposition and Biological
Heat-Life of Elemental Iodine on Grass and Clover.
Proc. Atmosphere-Surface Exchange Particulate and
Gaseous Pollutants - 1974 Symposium. Richland, WA.
September 4-6, 1974. ERDA Symposium Series 38.
CONF-740 921.
60 Pyatt, F. B. Plant Sulphur Content as an Air Pollution
Gauge in the Vicinities of Steelworks. Envirn. Pollut.
5_: 103-115, 1973.
61 Martin, A. and F. Barber. Some Measurements of Loss of
Atmospheric Sulphur Dioxide Near Foliage. Atmos. Environ
5: 345-352, 1971.
100
-------
62 Mansfield, T. A. and 0. V. S. Heath. An Effect of
"Smog" on Stomatal Behavior. Nature (London). 200;
596, 1963.
63 Spedding, D. J. Uptake of Sulfur Dioxide by Barley
Leaves at Low Sulfur Dioxide Concentrations. Nature.
224; 1229-1231, 1969.
64 Majernik, 0. and T. A. Mansfield. Stomatal Responses
to Raised Atmospheric C02 Concentrations During Exposure
of Plants to S02 Pollution. Environ. Pollut. 3:1-7,
1972.
65 Abeles, F. B., L. E. Craker, L. E. Forrence, and
G. R. Leather. Fate of Air Pollutants. Removal of
Ethylene, Sulfur Dioxide and Nitrogen Dioxide by Soil.
Science. 173: (4000). 914-916, 1971.
66 Bennett, J. H. and A. C. Hill. Absorption of Gaseous
Air Pollutants by a Standardized Plant Canopy. J. Air
Pollut. Control Assoc. 23:203-206, 1973.
67 Liss, P. S. Exchange of SC>2 Between the Atmosphere
and Natural Waters. Nature. 233:327-329, 1971.
68 Liss, P. S. and P. G. Slater. Flux of Gases Across
the Air-Sea Interface. Nature. 247:181-184, 1974.
69 Hales, J. M. and S. L. Sutter. Solubility of Sulfur
Dioxide in Water at Low Concentrations. Atmos. Environ,
7:997-1001, 1973.
70 Spedding, D. J. SC>2 Uptake by Limestone. Atmos.
Environ. 3:683, 1969.
71 Spedding, D. J. Sulfur Dioxide Absorption by Sea
Water. Atmos. Environ. 6:583-586, 1972.
72 Payrissat, M. and S. Beilke. Laboratory Measurements of
the Uptake of Sulfur Dioxide by Different European Soils,
Atmos. Environ. 9^:211-217, 1975.
101
-------
73 Fisher, B. E. A. Discussion of Reference 72. Atmos.
Environ. 9_:553, 1975.
74 Smith, K. A., J. M. Bremer, and M. A. Tabatabai. Sorption
of Gaseous Atmospheric Pollutants by Soils. Soil Science.
116 (4) :313-319, 1973.
75 Terraglio, F. P. and R. M. Manganelli. The Influence of
Moisture on the Adsorption of Atmospheric S02 by Soil.
Int. J. Air & Water Poll. 783-791, November-December 1966.
76 Horst, T. W. A Surface Flux Model for Diffusion, Deposition,
and Resuspension. Pacific Northwest Laboratory Annual
Report for 1973 to the USAEC Division of Biomedical and
Environmental Research, Part 3, Atmospheric Sciences.
BNWL-1850 PT3. Battelle, Pacific Northwest Laboratories,
Richland, WA, April 1974. pp. 63-67.
77 Elderkin, C. E., D. C. Powell, G. H. Clark, and P. W.
Nickola. Comparison of Diffusion-Deposition Model Compon-
ents with Experimental Results. Pacific Northwest
Laboratory Annual Report for 1973 to the USAEC Division
of Biomedical and Environmental Research, Part 3, Atmo-
spheric Sciences. BNWL-1850 PT3. Battelle, Pacific
Northwest Laboratories, Richland, WA, April 1974. pp. 38-44.
78 Droppo, J. G. and J. M. Hales. Profile Methods of Dry
Deposition Measurement. Proc. Atmosphere-Surface Exchange
Particulate and Gaseous Pollutants - 1974 Symposium.
Richland, WA, September 4-6, 1974. ERDA Symposium Series
38. CONF-740 921.
79 Hales, J. M., J. M. Thorp, and M. A. Wolf. Field Investi-
gation of Sulfur Dioxide Washout from the Plume of a Large
Coal-Fired Power Plant by Natural Precipitation. Final
Report. Environmental Protection Agency No. CPA22-69-150.
1971.
80 Shaffer, W. A. Atmospheric Diffusion of Radon in a Time-
Height Regime. Ph.D. Thesis. Drexel University. June 1973
102
-------
81 Pasguill, F. Atmospheric Diffusion, 2nd Edition,
New York. John Wiley & Sons. 1974.
82 Pearman, G. I. and J. R. Garrett. Carbon Dioxide
Measurements Above a Wheat Crop. 1. Observations of
Vertical Gradients and Concentrations. Agr. Meteor. 12:
13-25, 1973.
83 Zlatkis, A., H. A. Lichtenstein, A. Tishbee, Concentra-
tions and Analysis of Trace Volatile Organics in Gases
and Biological Fluids with a New Solid Adsorbent.
Chromatographia. 6^:67, 1973.
84 Droppo, J. G. and H. L. Hamilton. Experimental Variability
in the Determination of the Energy Balance in a Deciduous
Forest. J. Appl. Meteor. 12J5):781-791, August 1973.
85 Paulson, C. A. Profiles of Wind Speed, Temperature
and Humidity Over the Sea. Ph.D. Thesis, University
of Washington, Seattle, Washington.
86 Gill, G. C. Development and Use of the Gill UVW
Anemometer. Proc. 3rd Symposium on Meteorology and
Observations and Transformation, Washington, DC.
February 10-13, 1974.
87 Horst, T. W. Corrections for Response Errors in a
Three-dimensional Propellor Anemometer. J. Appl.
Meteor. L2:716-725, 1973.
88 Garratt, J. R. Limitations of the Eddy-Correlation
Technique for the Determination of Turbulent Fluxes
Near the Surface. Boundary-Layer Meteorology. 8^255-
259, 1975.
89 Whelpdale, D. M. and R. W. Shaw. Sulphur Dioxide
Removal by Turbulent Transfer Over Grass, Snow, and
Water Surfaces. Tellus. 24:1-2, 1974.
103
-------
SECTION XI
NOMENCLATURE
Units:
Symbol
B
D
f
G
h
K
K_
K
K.
m
h
m
Q
q
T
u
L = length; m = mass; t = time; °C = °C;
°A = °A; units = as defined by quantity
of interest; none = nondimensional;
E = energy.
Units
units
none
E M~~°A
'""1
E M~
units
t"1
none
°C
L t~
Definition
"y" intersect in least squares
fit
Stanton number
Specific heat at constant pressure
for air
Diffusion function
Gradient factor
Flux to surface
Release height
Eddy diffusivity
Eddy diffusivity for momentum
Eddy diffusivity for heat
Eddy diffusivity for water
vapor
Latent heat for evaporation
of water
Slope in least squares fit
Source strength
Specific humidity
Air temperature
Wind speed
104
-------
w
X
RI
z
( )
ra
rg
T
Vd
vs
X
p
cr,
L t-1
L
none
L
units
units
t L-1
t L-1
tL
-1
tl
-1
tL
-1
t L
-1
t L
-1
L t
-1
L t
-1
units L
M L"3
units
°A
L
-3
Vertical velocity
Downwind distance
Richardson number
Height
Indicates departure from
mean value
Indicates mean value
Atmospheric resistance
Gravitational resistance
Internal resistance
Surface layer resistance
Effective surface layer
resistance
Total resistance
Momentum flux resistance
Deposition velocity
Settling velocity
Air concentration
Density of air
Standard deviation of variable
Potential temperature
Downwind surface source
coordinate
105
-------
SECTION XII
APPENDICES
Page
A. NUCLEPORE FILTER ANALYSIS RESULTS FROM ORTEC 107
B. NUCLEPORE FILTER ANALYSIS RESULTS FROM BATTELLE
COLUMBUS LABORATORIES 121
106
-------
APPENDIX A
NUCLEPORE FILTER ANALYSIS RESULTS FROM ORTEC
107
-------
HIGH PRECISION ENERGY DISPERSIVE
X-RAY FLUORESCENCE ANALYSIS of SULFUR
IN AEROSOLS DEPOSITED ON FILTER PAPERS
By
JOHN A. COOPER
ORTEC,. Incorporated
100 Midland Road
Oak Ridge, Tennessee 37830
September 11, 1975
108
-------
HIGH PRECISION ENERGY DISPERSIVE
X-RAY FLUORESCENCE ANALYSIS OF SULFUR
IN AEROSOLS DEPOSITED ON FILTER PAPER
John A. Cooper
ORTEC, Incorporated
100 Midland Rd.
Oak Ridge, Tennessee 37830
September 11, 1975
INTRODUCTION
Dry deposition of atmospheric aerosols is a significant mechanism
for removal of atmospheric contaminants which may exceed precipitation
scavenging under some climatological conditions. The development of a
reliable model for predicting the contribution of dry deposition to
the removal of material from fossil fuel power plant effluents by
profile measurement techniques requires accurate determination of small
differences (a few percent) in chemical composition as a function of
altitude. These differential composition measurements require both
accurate sampling procedures and precise analytical methods.
This investigation is part-of a larger study which includes the
collection of aerosol and gaseous samples at different altitudes and
the measurement of the vertical profile for the gaseous components CO,
NO, NO-/ H_S, SO_, O_, hydrocarbons as well as SO4=, N02~» NO3~ ajn^-
organic particulates. The purpose of this part of the study is to
develop a precise method for the analysis of S in aerosols deposited
on filter paper by energy dispersive x-ray fluorescence analysis and
to analyze a series of aersol sample collected at different altitudes.
THEORETICAL CONSIDERATIONS
Energy dispersive x-ray fluorescence analysis of S in aerosols is
complicated by spectral interferences and uncertainties due to x-ray
absorption in the collection filters and deposited particulate material.
Sulphur emits a 2.31 keV K x-ray which may be spectrally interfer-
red with by unresolved Mo L line at 2.29 keV and a Pb M line at 2.35
keV. The Mo L line is not a common interference in aerosol analyses
because of its low abundance. Lead, however, is typically as abundant
as S (Figure 1) and must be corrected for in any S analysis.
The analysis procedure used in this study included measurements
of the Pb and Mo concentration in each filter and the correction of the
109
-------
2.3 keV x-ray peak intensity for the presence of Pb. (Mo was not corrected
for because of its consistently low levels.)
Accurate analysis of S in deposited aerosols will also require cor-
rections for x-ray absorption in the filter and the deposited particulates.
The magnitude of these errors, however, is small for this study because
of the interest in relative concentration differences. Errors associated
with the sample would thus be caused mainly by differences in significant
particle size distributions and/or differences in deposition patterns in
addition to obvious uncertainties in air volumes sampled per unit filter
area.
EXPERIMENTAL CONDITIONS
Sample Description
The samples analyzed consisted of aerosols deposited on Nuclepore
filters. The filter diameter was about 1 3/4 inches in diameter while
the exposed area was only 1 1/4 inches. The exposed area represents a
collection of about 5 cfm over a period of about one hour.
The samples were .received in labeled plastic pill boxes. It appeared
as though the aerosol was deposited on the dull side of the filter. Stand-
ards prepared by Columbia Scientific Inc., by depositing known amounts of
material in solution form on millipore filters were used to determine the
relative counting rates for Pb using the L and M lines and estimating the
absolute counting rates of Pb and S.
A piece of Nuclepore filter from a different batch was used as a
blank. The S standard contained 75 p.g/cm of elemental S while the Pb
standard contained 24 /ig/cm .
Sample Preparation
The filters were taped to 2 inch diameter flat plastic rings with the
dull side of the filter exposed to the x-ray tube and detector. The
plastic holders were mounted in an Al wheel with a positioning accuracy
of +.002 in between the x-ray tube-sample and sample-tube distance.
Instrumentation
An ORTEC Model 6llO Tube Excited Fluorescence Analyzer was used for
the analysis. The samples vere analyzed in vacuum with the excitation
conditions listed in Table 1.
110
-------
TABLE 1
Anode
Mo
W
W
W
Mo
Filter
—
—
—
Al
Mo (3)
Voltage (KV)
10
10
15
15
40
Current (UA)
200
200
200
200
200
Software
The samples were analyzed automatically using the intensity-only ver-
sion of the XRFCALC program which determines the net x-ray intensities above
background for each element defined with a region of interest.
RESULTS
The major elemental components (Al, Si, S (Pb) , Cl, K, Ca and Fe) pre-
sent in the aerosol samples are illustrated with the low energy x-ray
spectra of Sample M7-1 shown i*n Figures 1 to 3. Low intensity peaks asso-
ciated with Na, Mg, P and Ti were also observed. Figures 1 and 2 contrasts
the differences between spectra excited with filtered and unfiltered
radiation from a W anode.
The high energy portion of the x-ray spectra are illustrated in Figures
3 to 6 which show the clear presence of Fe, Ni, Cu, Zn, Pb and Br. The
logarithmic plot shown in Figure 4 is contrasted with the linear plot in
Figure 5 and the spectra of an unrelated rural aerosol shown in Figure 6.
A spectrum of scattered radiation off the plastic sample holder is
shown in Figure 7 and illustrates the low system blank.
The samples were analyzed eight times under several different excita-
tion conditions and each sample was rotated between analysis. The averages
from these analyses are listed along with expected errors in Table 2.
Ill
-------
1C
o
o
o
TEFA SPECTRUM OF SAMPLE M7-1
(0 - 20 keV)
K
Si
ogn
Al '
Mg
(U
Na
\ /
Anode
Scattered Lines
EXCITATION CONDITIONS
Tube Anode: W Filter: Al
Anode Voltage: 15 kV
Anode Current: 200 uamps
Livetlme: 800 seconds
Figure 1
CHANNEL NUMBER (ENERGY)
-------
TEFA SPECTRUM OF SAMPLE M7-1
(0 - 10 keV)
SI _
Al
fc?
Mg.
(K )
\ /
Na_
'
~Ca
T1
EXCITATION CONDITIONS
Tube Anode: W Filter:
Anode Voltage: 15 kV
Anode Current: 200 uamps
Livetime: 400 seconds
None
Figure 2
CHANNEL NUMBER (ENERGY)
-------
Si
TEFA SPECTRUM OF SAMPLE M7-1
(0 - 10 keV)
Pb
(M)
EXCITATION CONDITIONS
Tube Anode: W Filter: None
Anode Voltage: 10 kV
Anode Current: 200 ylamps
Livetime: 800 seconds
A! —,
Cl _
(Ka)
1/1
I—
o
o
Mg-
(Ka)
Na
Figure 3
CHANNEL NUMBER (ENERGY)
-------
TEFA SPECTRUM OF AIR FILTER SAMPLE M23-3
(0 - 20 keV)
Fe
o
CO
cC
CJ
C£
LU
Q.
l/l
O
O
' \.
s-
V/A. 0
Fe
V
r-N1
j
rCu
Cu
(K*
/
^Ka
}
r
Zn
(K
'. /
a
r-.Ga.
N .>
(Kp)
r^»
1
A
''*/
EXCITATION CONDITIONS
Tube Anode: Mo Filter: Mo(3)
Anode Voltage: 40 kV
Anode Current: 200 yamps
Livetime: 2000 seconds
Figure 4
CHANNEL NUMBER (ENERGY)
-------
TEFA SPECTRUM OF AIR FILTER SAMPLE M23-3
(0 - 20 keV)
EXCITATION CONDITIONS
Tube Anode: Mo Filter: Mo
Anode Voltage: 40 kV
Anode Current: 200 uamps
Llvetime:' 2000 seconds
Fe _
t Lt \
\^l
Fe
/ |y \ L J|
.
1
\
\ -\
--X.VY \
'"""•*•' '•, *
• •• s • «
.1 - V H, . t
>
V'*
N
(K
""'
i
a)
rc
(K(
c
(K
z
/ )/
» /
V ,
Vv'1
J
Jl'
J-
3)
1
Ga-
(Kp)
Ga
(K°
,'.S . - w"
l
|— Pb ' '. /
\
- y
?
.'•
\'
/
•/'
'.\v'
CHANNEL NUMBER (ENERGY)
Figure 5
-------
TEFA SPECTRUM OF RURAL AEROSOL DEPOSITED
ON NUCLEPORE FILTER
(0 - 20 keV)
o
01
13
O
o
o:-
LU
a.
CO
o
CJ
Ar
(Air)
v..
Fe
(ty-
(K )
a
'
Pb
Pb
Zn
/ V
V **
Cu_
Br
;A
/\A
EXCITATION CONDITIONS
Tube Anode: Mo Filter: Mo(5)
Anode Voltage: 50 kV
Anode Current: 200 amps
Livetime:
Figure6
CHANNEL NUMBER (ENERGY)
-------
TEFA SPECTRUM OF THICK PLASTIC
(0 - 20 keV)
S —,
(K
» iv
/ \
Al—,
a
EXCITATION CONDITIONS
Tube Anode: W Filter: Al
Anode Voltage: 15 kV
Anode Current: 20 uamps
Uvetime: 118 seconds
Figure 7
CHANNEL NUMBER (ENERGY)
-------
TABLE 2
AVERAGE SULPHUR AND LEAD CONCENTRATIONS ON NUCLEPORE
FILTERS AND RELATIVE CHANGE IN SULPHUR CONCENTRATIONS
Sample No.
1-3
1-20
1-50
M7-1
M7-2
M7-3
M8-1
M8-2
H8-3
Ml 2-1
Ml 2- 2
Ml 2-3
M23-1
M23-2
M23-3
Number of S
Analyses
10
10
8
6
6
5
9
9
9
10
10
10
6
6
6
Average S *
Cone. (ug/cm2)
1.43 + .02
1.27 + .02
1.29 +. .02
2.56 + .04
2.18 + .04
2.25 + .04
3. 02+. .05
2.79 + .04
2.81'+ .04
2.64 + .04
2.34 + .04
2.33 +_ .04
0.45 +_ .01
0.46 +_ .01
0.41 + .01
Average Relative
Change in S Cone.
1.00
0.89 + .02
0.90 + .02
1.00
0.85 +, .02
0.88 + .02
1.00
0.92 + .02
0.93 +_ .02
1.00
0.89 + .02
0.88 + .02
1.00
1.02 + .02
0.91 + .02
Pb
Cone, (uq/cm2)
0.66 + .02
0.47 +_ .02
0.48 + .02
0.67 +. .02
0.62 + .02
0.62 i .02
0.59 +_ .02
0.59" i .02
0.57 + .02
0.48 +_ .02
0.45 +_ .02
0.43 + .01
0.50 +_ .02
0.38 + .01
0.35 + .01
The errors listed in this table are precisional errors based upon repetitive analyses.
* This concentration will be low by a constant factor of about 20% due to differences
in the deposition patterns between the standard and the specimens.
119
-------
REFERENCES
1. Plesh, R. "X-Ray Fluorescence Analysis of Sulfur Precipitates
on Air Filters" Siemans X-Ray Analytical Application Notes No. 12.
2. Bonnevie-Svendsen, M. and A. Folio, "Evaluation of filters,
Standardization and Measuring Procedures for X-Ray Fluorescence
Analysis of Sulphur in Air-Born Matter," Norwegian Institute
for Nuclear Pnergy report number DH-98, June 1972.
3. Dzubay, T. G. and R. 0. Nelson, "Self Absorption Corrections for
X-Ray Fluorescence Analysis of Aerosols" Denver X-Ray Applications
1974
120
-------
APPENDIX B
NUCLEPORE FILTER ANALYSIS RESULTS FROM
BATTELLE COLUMBUS LABORATORIES
121
-------
llBalfeile
Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
Telephone (614)299-3131
October 6, 1975 Telex24-5454
Dr. J. G. Droppo
Battelle, Pacific Northwest Laboratories
Battelle Boulevard
P.O. Box 999
Richland, Washington 99352
Dear Jim:
This letter confirms the results relayed to you earlier regarding the
analyses of 15 nucleopore filter catches for sulfur. The conditions to be
achieved in this effort were a total sulfur content determination having
a precesion of 17« and a quantitative estimate of about 20%. In addition,
a direct comparison of the results were to be made after analyses of the
same samples using energy-dispersion (ortec) in one case and wavelength
dispersion (Battelle - Columbus) in the other.
The analyses were performed at Battelle using a conventional wavelength-type
X-ray fluorescence unit. Sufficient counts were accumulated at the "Of sulfur
peak to provide counting statistics of 1% for all samples. For the M-23
series, counting times of 400 seconds were necessary for the desired precision,
where as for all other samples only 200 seconds were required.
Synthetic standards were prepared by drying appropriate amounts of an aqueous
solution of Na2SO^ onto unused nucleopore filters. The prepared standards
provided a linear working curve from 2 to 200 micrograms/3cnr.
The results of these analyses are shown in Table 1 as total sulfur per active
filter area (~11.4 cm-). The condition of 1% percision among filter values
appears to have been substantiated from a comparison of these and the preliminary
Ortec data you have received. The accuracy of these analyses however is
difficult to assess and may require complete filter destruction for analysis
by some absolute method.
During our last phone conversation I mentioned attempting to determine the
sulfur state on these catches using x-ray fluorescence. Sample M-8(l) was
used for this investigation of the possible definitive variations of the
sulfur peaks. The results of this examination indicated that sulfur was
present solely as the sulfate. Also, the peak anomalies point toward calcium
122
-------
Dr. J. G. Droppo 2 October 6 1975
as the cation. Subsequent X-ray scan revealed that calcium is present in
sufficient amounts that may approximate the sulfate molar concentration,
I want to thank you for successfully extending the project so that I could
recover most of the costs. I still have the samples and will await word
regarding their dispensation.
Please let me know if we can be of further service to you.
Very truly yours,
Ben Paris
Research Chemist
Analytical and Physical Chemistry Section
BP:lj
Enc- 1 Table
123
-------
TABLE 1. TOTAL SULFUR CONTENT OF NUCLEOPORE FILTER CATCHES
(Results in total micrograms)
Sample
5/12/75
3
20
50
feet
feet
feet
(1)
(2)
(3)
Total
4.2
3,8
4.0
S
5/13/75
5/14/75
5/16/75
M-7
M-7
M-7
M-8
M-8
M-8
M-12
M-12
M-12
M-23
M-23
M-23
(1)
(2)
(3)
(1)
(2)
(3)
(1)
(2)
(3)
(1)
(2)
(3)
7.4
6.4
6.7
8,2
7.8
7.6
7.4
6.7
6.7
2.1
2.1
1.9
124
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO.
2.
3. RECIPIENT'S ACCESSION«NO.
4. TITLE AND SUBTITLE
MEASUREMENT OF DRY DEPOSITION OF FOSSIL FUEL
PLANT POLLUTANTS
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
J. G. Droppo, D. W. Glover, A. B. Abbey, C. W. Spicer,
and J. Cooper
9. PERFORMING ORGANIZATION NAME ANO ADDRESS
Battelle-Pacific Northwest Laboratories
Batte!!e Boulevard
Richland, Washington 99352
10. PROGRAM ELEMENT NO.
1AA009
11. CONTRACT/GRANT NO.
68-02-1747
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory
Office of Research and Development
Environmental Protection Agency
Research Triangle Park, NC 27711
13, TYPE OF REPORT AND PERIOD COVERED
Final 6/24/74-6/24/76
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Dry removal of air pollutants from fossil fuel plants is considered from both
a modeling and measurement viewpoint. Literature on dry deposition rates is summa-
rized and the processes involved in dry deposition are discussed. The dry deposition
of S02, 03, NOX, and NO, as well as total sulfur and lead particles are considered.
A prototype field data acquisition system was developed, assembled, and tested.
Deposition velocities were computed for each field test.
The sulfur dioxide profiles gave reasonable estimates of the dry deposition
values, comparable to those in the literature. They varied from 0.10 to 2.38 cm/sec
for the test runs. Values for the 03 deposition velocities were very small. The
results for NO varied over a wide range, with the number of profiles measured in the
test runs insufficient for reaching a definitive conclusion.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFlERS/OPEN ENDED TERMS C. COSATI Field/Group
*Mathematical Models
Reaction Kinetics
Electric Power Plants
Air Pollution
12A
07D
10B
13B
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport}'
UNCLASSIFIED
21. NO. OF PAGES
177
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
Z LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE . , ., , . „
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, tf used in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S) ...
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code. " r
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered. - •
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less} factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited."' Cite any availability to
the public, with address and price.
19. &20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |