"SEPTEMBgfcdSSS
LOCAL SOURCE IMPACT ON WET DEPOSITION
ATMOSPHERIC SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARC, NORTH CAROLINA 27711
-------
LOCAL SOURCE IMPACT ON WET DEPOSITION
by
Aristides A. N. Patrinos
Brookhaven National Laboratory
Upton, Long Island, New York 11973
DW89006701
Project Officer
Francis S. Binkowski
Meteorology and Assessment Division
Atmospheric Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ATMOSPHERIC SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
The information in this docunent has been funded wholly or in part
by the United States Environmental Protection Agency under Interagency
Agreement Number DW89006701 to the Brookhaven National Laboratory. It has
been subject to the Agency's peer and administrative review, and it has
been approved for publication as an EPA document. Mention of trade names
or comercial products is does not constitute an endorsement or recom-
mendation for
use.
-------
ABSTRACT
Precipitation chemistry measurements over a network of samplers
upwind and downwind of Philadelphia, PA show that a major contribution of
the local sources can be discerned under certain conditions. For winter
frontal storms with low level winds from the south east, up to as much as
a factor of two increase over upwind values has been observed for downwind
nitrate deposition. Sulfate deposition shows an increase of about a
factor of one and one half. The nitrate deposition increases toward the
downwind direction away from the urban-industrial sources, indicating
that the maximum is likely to have been beyond the sampling network for
these case studies. One storm had no increase in nitrate or sul fate
deposition but did have an increase in total sulfur content in the pre-
cipitation. Reasons for this difference are being sought.
n i
-------
CONTENTS
Abstract iii
Figures vi
Tables vii
1. Introduction 1
2. Point Sources 5
3. Area Sources 17
4. The Philadelphia Field Study 29
Background 29
Exploratory Phase and Quality Assurance 30
The Main Field Study and General Results 35
Diagnostic Modeling of a Philadelphia Storm 52
5. Concluding Remarks 55
6. References 60
-------
FIGURES
*
Number Page
1 Sulfur IV Deposition for December, 1983 Storm 31
2 Sulfate Deposition for December, 1983 Storm 42
3 Nitrate Deposition for November, 1983 Storm 49
4 Nitrate (left) and Ammonium (Right) Deposition
for April, 1984 Storm 50
VI
-------
TABLES
Number Page
1 Field Studies Around Point Sources 8
2 Summary of Meteorological Characteristics of Storms ... 32
3 Statistical Summary of Deposition Data 38
4 Dissolved S02 Deposition Data for Sampled Storms 43
5 Estimates of SOX and NOX Emission Rates along the
Deleware Valley 46
6 Diagnostic Modeling Parameters and Inputs for Storm
0405 Simulation 54
VI 1
-------
-------
SECTION 1
INTRODUCTION
It has become increasingly apparent that the impact of local sources
on acid wet deposition is of considerable importance in developing source
receptor relationships and establishing fair and effective mitigation
strategies; these strategies may seek the apportionment of fractions of
sensitive receptor depositions to various source regions. It should be
emphasized that the "local source" issue has been somewhat controversial
with several studies alternatively overemphasizing or underestimating the
local source contribution. Earlier studies generally tended to disregard
local source contributions for several reasons. Public awareness of the
"acid rain" problem was associated with discovery of alleged ecological
damage (Likens et al., 1979) of remote and assumed pristine areas such as
the Adirondacks in New York, the White Moutains in New Hampshire, and in
several regions of Scandinavia. Claims of the rise in the acidity of
some lakes resulting in fish population decreases and reports of possible
forest deterioration (Johnson and Siccama, 1983) due to soil acidifica-
tion have been linked to the deposition of sul fate which originated as
sulfur dioxide (S02) at considerable distances upwind. For the North-
east, the candidate sources are the large power generating facilities of
the Ohio Valley. These "tall stack" point sources provide the framework
for the "long range transport" hypothesis since sulfur emissions at
higher elevations would be transported by the prevailing winds and with
longer atmospheric residence times would interact with available oxidants
producing the sulfates which would ultimately be scavenged at the sensi-
1
-------
tive receptor regions. The contribution of nitrates to the overall
acidity in the above context has been considered secondary. The "long
range transport" scenario was further bolstered by the evidence from
field studies of wet deposition around large point sources. As will be
presented in later sections, most of these studies point to a relatively
small contribution from the overall emissions to the wet deposition in
the near field. The equivalent results from large area sources, however,
are somewhat at odds with the point source studies. These results,
particularly from the recent Philadelphia field studies, have shown that
the diffuse low level emissions associated with areal sources have some
impact on the local scale and maximum impact on the mesoscale (within 100
km). This impact is mostly manifested with the deposition of nitrate
which implicates the significant transportation sources present in large
urban areas. The recognition that acidic deposition, in fact both wet
and dry, would be significant on the mesoscale around large urban and
industrial centers along with the realization that the eastern U.S. has
a sizable number of such centers have revived the importance of the local
source issue. Further support is prompted by the materials damage asses-
sment; acidic deposition has been suspected to cause damage to struct-
ures, masonry, paint, and generally materials (Baer et al., 1984) which
are in great abundance in and around area sources. The potential damage
to historical landmarks, statues, etc., is of particular concern. Impli-
cit support to the "local source" issue is provided by the realization
that acidic materials which are deposited locally are not available for
long range transport; thus an accurate accounting of near source deposi-
tion is important to the overall pollutant budget. Furthermore, studying
2
-------
wet deposition in the near field of a large pollution source presents a
convenient experimental framework with a potentially better opportunity
to comprehend some of the fundamental atmospheric processes involved in
acidic wet deposition", such as scavenging mechanisms and chemical react
ions; this would improve the understanding of source-receptor relation-
ships at longer scales. Having established the importance of the "local
source" contribution it should be pointed out that the uncertainties in
the extent, type, and frequency of this contribution are high. There are
two important reasons for these uncertainties. The first is due to the
paucity of available data on wet deposition in and around large area
sources. In the U.S. only two comprehensive, large scale field campaigns
were undertaken for the purpose of sampling wet deposition on an event
basis, and those only for limited time periods. The METROMEX study (AMS,
1981) around the city of St. Louis concentrated on summer, convective
storms. The Philadelphia studies were geared toward cyclonic and primar-
ily nonconvective storms. Furthermore, most of the existing precipita-
tion chemistry networks have purposely located their sites away from
large local sources; this was dictated by the desired regional nature of
these networks. As a result, it is felt that the integrated results
based on these networks have consistently underestimated the total deposi-
tion budgets, particularly for nitrate. The absence of long-operating
urban sites also make trend analyses such as at the Hubbard Brook Experi-
mental Station (Munn et al., 1982) impossible. The second cause of
"uncertainty" pertains to the nature of emissions. For urban and indus-
trial centers the pollutant precursor mix is characterized by consider-
able variability in both the type and temporal and spatial variability.
3
-------
The distribution of primary pollutants (mostly sulfur dioxide, nitric
oxide, hycrocarbons) may vary considerably from city to city depending on
the level of industrial activity, transportation sources, type of residen-
tial heating and others). The primary pollutant mix may also include a
certain amount of sulfate (particularly from residential oil burners) and
hydrogen chloride (HC1) from the combustion of materials containing
chloride impurities (lapalucci et al., 1969; Gregory, 1976; Hlavay and
Guibault, 1978). HC1 is a strong acid and can significantly affect cloud
and rainwater acidity near its source and is washed out rapidly when rain
occurs (Patrinos et al., 1983). The distribution of primary pollutants
and their temporal variability (in terms of time of day, day of the week,
season of the year, etc.) affects the generation of secondary pollutants
such as ozone (03) which is derived from photochemical reactions involv-
ing hydrocarbons and nitrogen oxides, and which in turn may play an
important role in acid forming reactions (Calvert and Stockwell, 1983;
Richards, 1983). Urban and industrial areas are also rich sources for
catalytic substances (Penkett et al., 1979) promoting these reactions.
-------
SECTION 2
POINT SOURCES
There exists a significant body of literature on the subject of wet
deposition around large point sources. Observed concentrations of the
major scavenged species vary considerably, presumably as a result of
variations in meteorological and background conditions and source charact-
eristics, but often insufficient detail is available to evaluate precise
scavenging characteristics. Earlier field studies were aimed toward
demonstrating the efficacy of taller stacks in lessening local impacts.
In fact point sources may be defined as those sources which have elevated
releases because of requirements to meet ambient air quality standards on
SOX, NOX, and particulates. They may include power plants, smelters,
pulp and paper mills, petroleum refineries, cement plants, etc. The
majority of the studies generally support this claim. The most notable
exception has been in the deposition of trace metals for which the par-
ticulate nature of the emissions promotes efficient wet scavenging in the
near field (within 25 km of the source). For the important inorganic
ions found in "acid rain," sul fates and nitrates, the wet deposition in
the near field represents but a few percentage points of the total emis-
sions during precipitation. Of the observed variabilities mentioned
above perhaps the most prominent is that of the background conditions
particularly for sulfur. Since most of the emitted sulfur is in the form
of sulfur dioxide ($02) its solubility in cloud or rain drops is very
much a function of pH (Hales and Sutter, 1973) and consequently the
-------
ultimate scavenging efficiency is highly dependent on background cloud
and precipitation acidity. At the same time the presence of certain
oxidants (such as hydrogen peroxide) (Penkett et al., 1979) or catalytic
substances may accelerate the S02 to sulfate process increasing the wet
deposition of sulfate well beyond what is estimated on the basis of
primary sulfate emissions.
Another source of variability is the nature of the precipitating
system. It appears that the scavenging of sul fate and nitrate may be
considerably enhanced during summer convective storms compared to winter
time frontal precipitation. This may be due to longer in-cloud residence
times of primary pollutants in convective situations actively promoting
faster acidifying processes and reactive scavenging.
Dana and Patrinos (1983) reviewed open literature results from wet
deposition studies around large point sources to, primarily, compare
simply calculated scavenging parameters. Table 1 is an expanded version
of their table presenting most field studies around large power plants,
smelters and refining installations; these projects are compared for
source characteristics, sampling network, and fractional removal. Re-
sults of measured SOo and H+ concentrations generally exhibited a vari-
ability of more than two orders of magnitude. The above mentioned vari-
ability in S02 deposition is increased further by the lack of sufficient
data at greater distances.
-------
One of the first sulfur washout experiments was conducted at the
Keystone power plant in western Pennsylvania (Hales et al., 1971) by Bat-
telle Pacific Northwest Laboratories (BNW). The primary objective was to
assess the effectiveness of precipitation washout as a scrubber of sulfur
compounds from the plume; the secondary objective was to evaluate the
relative importance of various atmospheric conditions on washout efficien-
cies. Although the high background levels of acidity and sulfur prevent-
ed accurate calculations of removal rates for the Keystone plume, it
became apparent that the old irreversible gas scavenging models were not
appropriate for S0£; indeed the samples showed very little S02 washout
compared with the sulfate aerosol washout. Another early study around
the Colbert County generating station in Alabama (Hutcheson and Hall,
1974) contrasted the Keystone results. A significant amount of the
sulfur wet deposition was attributed to S02 scavenging due to the relat-
ively "clean" background rain.
The studies around the Central ia power plant were also conducted by
BNW (Dana et al., 1975; Dana et al., 1976) in a region of low background
acidity and sulfur concentrations in rain. Results confirmed that the
method of calculating the washout of S02 from plumes, based on reversible
gas absorption phenomena, is applicable to circumstances involving power
plant plumes emitted from tall stacks, as well as for less complicated
situations; indeed, the possibility of a "negative" washout effect was
demonstrated. The Centralia results also indicated that the washout of
sulfur from the plume can be 1-5 times greater than that of S02 on a
molar basis, thus emphasizing that the important sulfur compound scaveng-
ing problem is that involving sulfate. A smaller scale study was conduct-
-------
(/I
00
T3
"QJ
UJ
I
CQ
a*
Qp
**
<+_
T3
I/I
CU
T
O
CO
Q-
^_^
<4
X
4- f
«k ^v^
t_ CO co
3 U CU
<4- t- .
3 O
3 0 S
CO OO &
*
to T3
CO CO
t- M
0 >>
CO
Q. to
00 C
-c
.0 to cn e
§»^
4-
3 z: u
o co
Q. r
CO
CO
o
c_
3
O
oo
CO
frj»
r-
OO
LO
£ -4-J
QJ C
5 T3
0 .
0- 0-
CM cn
CM
<-i co o o
II LO 1 1 I O I
co o LO oo i <-> vo cn
i ICMLOI r»» co i i i *-4
O CM i ' CO«-H
O O O O CO CO
II II II
1 t CM < 1 CM <4- CM «J-
O O O O O i O
coco oocoooooo coco co
1 1 i 1 i * «* LO O O «*
i 4 t « t 1 i t VO CM
CM
LO r-l CO CO <1- O
cn i co o CM < o
i cc?) co cn i cn i i cc?** cn
O LO LO O
^ cn cj i t
II 10 1
«* * " " "r- CO
O to to co+ +*} O
" co c c c ID ^rc *j z
II ^OOOZZOCO «
CO CM **~ "*~ * ""'" «1-
OOZ O t. t- t. Z Z t- O
ICO OO O O O Q. Q.OCMOO
CM * CM -1-5 CM'tr> CMr-j CM -r-j O *
cO"^ co + *o cO *^ co f^ co * fo oo cn
oozz z scoosoozioooo cos: z:
t I . i O LO
LO i 1 i 1 VO T 4
VO CM 1 1 «3- LO 1 1
t 1 *4" ^^ ^" r*H r)-r f- T CT1 g
O t« C re rt3 /O 3 f O
4-j co 3 L. c_ t_ c cn to o CT>
CO ^ Q -4^ 4J +J CO CC "^3 ^* > CO
>, i O C C C 5 COCl-CO ?
cu ocococoO4J3rei to
i^ C_)OC_3C_)OQOOtOi^ £ C C C *r~ C C ^ O
core re » » M cnco coco >-
ct- JD ^ ^i j= j_ -a -a-c
C C C
'CO <4-
£ CO
t75 Q£
vo ^ - '
»
^-1 VO f~- CM VO
1 « ^J- «
LO LO O OO O CO I -»e
^- CM CM
O CO
II
CM «*
O O
oo co co co co co oo
LO
o
O O O O CM O «3-
LO LO *3" "3- t 1 VO
O O O 00 CM
CM r-l CM l-» "
r^ co CM
1 CM
CO " "O
O to cn oo to
Z C C C 1
0 0 * 0 i
T- (/) 1 CO 1 C_)
CO r r
i O -t-> O 4-> O II
CMt_3 -r-> CO -r-> CO -r-j ^J-
o re £ re s re o
si s: z: z oo
o o o
r*» LO LO oo o
II 1 «*
00 CO CM CO 1
i I t 1 CiTi
O LO CM CO O
LO CM CM ^T ^-1
o o o
CO CO CO
CO CO CO ^O^
*^*fc ^"^ ^^, '**. CcT^
r 1 i 1 r « t 1
C_ CO
>> >i >> co re
t- t- t_ re +j o 4J
33 3 = r C
^"* <^ t^* o co i_ re
T3 T3 "O U g 3 i
3 3 3 re OO O Q-
oo oo oo 1 oo
c
o
1 »
+^
o o o cn re
T- T- T- C 4->
(- (- t- T- t_
re re re .c co
4^ 4^ 4^ tO o
c c c re .
O O O 3
-------
to
c
o
o
<5
fO
O)
'a?
4->
(O
i|
3
to
to
IQ
-c
0)
t
w.
3
to
O)
0
3
O
to
"§
-t-J
«J
U
r
*O
!-
<4-
o
*»-
a
Nw^
X
o
!->
Q.
3
-o
ID
C/1
o
o- e
rtj
OJ
-l->
fO
3
C
*
-M
C
O
o
NM"
i4
LLJ
_i
CQ
<
I
0)
>
r
O
en
to
r*
-a
o
c+
fft *
^ir1
(U Z
4->
«3 "
4- 1
r* !~
3 O
to
*
t- 1
O f
o
r Z
"
O+
4-> ^
*
O
Q. ifl
OJ C
£- O
r*
to +J
IT3 O
(O
(U S-
aiM-
-a
(O
i«-
c- o
o
c
4-» 0
C i-
OJ -l->
)£ "3
^ 3
t_ i
a; *o
Q. >
X 0)
(U
<4-
t- O
c
o
T
+->
fO
4->
l~"
a.
r~
U
01
E
O)
u
c_
3
O
to
-a
O
"s
I (->
i -a
3 3
(/) i
(/)
*
> -r- o e -i-*
c to s_ to .M e c
(O ! 3 S_ O 3 -r-
oi Q -a u- z oo 3
-------
ed by the University of Maryland (Li and Landsberg, 1975) around the
Chalk Point power plant with a sampling network encircling the plant (up
to 5 km) and with rainwater pH found to vary between 3.0 and 5.7. A
dependence of acidic washout from the plume on wind direction was noted,
but the degree of correlation was unclear.
Barrie (1980) investigated the nature and fate of emissions from an
isolated power plant in a shallow river valley of the Athabasca oil sands
area in western Canada. The study concentrated on "total" deposition
(wet and dry) with snow chemistry surveys using acrylic plastic snow
corers. The snowpack had accumulated pollutants without a melt for the
70-day period November 18, 1977, to January 26, 1978. Results are in
general agreement with the wet deposition studies showing a small per-
centage of the sulfur and nitrogen oxides emissions depositing in the
near field with an order of magnitude increase in the percentage of
deposited trace metals. Due to the soluble oxides of calcium and magnes-
ium in flyash, "total" deposition near the source was more alkaline than
in outlying areas.
Holt et al. (1983) used an oxygen-18 method to distinguish the pri-
mary sulfate in the total sulfate scavenged by rain around the Tennessee
Valley Authority (TVA) Widows Creek Steam Plant at Stevenson, Alabama.
The network of four automatic wetfall collectors (wet-only) was operated
on an event basis during the summer of 1981. Results demonstrated a
300-fold higher rain scavenging efficiency for sulfate compared with that
of S02- Of the scavenged sulfate downwind of the plant half was attrib-
uted to primary emissions.
10
-------
Patrinos et al. (1983) conducted a wetfall chemistry study around
the Bowen Electric Generating Plant in northwestern Georgia. The network
of automatic wetfall collectors was operated on an event basis during the
winter of 1981. Sulfur, hydrogen ion, and chloride ion were found to be
the predominant plume related species. Concentrations in the affected
regions exceeded the background levels by up to 100% in the case of
sulfur and of hydrogen ion and by up to 145% in the case of chloride ion.
Again, the percentage of emitted sulfur and nitrogen products which was
scavenged in the near field was small; for chloride, however, it was
determined that all the emitted Cl was depleted by scavenging in the near
field. Approximately 20% of the sampled storms exhibited substantial
plume S02 scavenging.
Several major field studies around oil-fired power plants have been
conducted in Europe. Granat and Rhode (1973) investigated the wet depos-
ition around the Stenungsund plant on the western coast of Sweden, while
Enger and Hdgstro'm (1979) conducted a study at the Karlshamm site on
the Baltic shore. The experimental plans were similar (sampling on an
event basis, frontal type storms), but the results were considerably
different in terms of observed deposition of plume related material and
background concentrations. Results around Stenungsund indicated that the
additional (above background) wet deposition of H+ would be no greater
than 10% to 15% within 15 km of the plant and the sulfur percentage would
be an order of magnitude less. For the Karlshamm study the authors
concluded that two thirds of the emitted material was deposited within
100-200 km from the source; for one case with high relative humidity they
11
-------
reported that 70% of the sulfur from the plant occurred as sul f ate and
was scavenged within the first 30 km. This result appears somewhat at
odds with the results from most other studies. Enger and Hflgstr&m
attributed the implied rapid S02 to SO^" conversion to ammonia catalysis.
A field study around the Flevo plant in the Netherlands (Slanina et al.,
1983) employed tracer techniques to detect the plume location and measur-
ed dissolved S02 in the rainwater samples. The experiment found that
most of the plume related sulfur wet deposition was in the form of SOg
out to 15 km.
In the U.S., Stamm et al. (1984) sampled wet deposition downwind of
an 1230 MWe oil-fired power plant in Oswego, NY, which burns high sulfur
oil (ca. 2.8%); magnesium oxide is added to the effluent to protect the
stack. Sampling was limited to two precipitation events on two consecu-
tive days associated with a cyclonic weather system. For both events the
plume was imbedded into a low stratus cloud. It was found that due to
the MgO the acidity immediately downwind of the stack decreased compared
with background and returned to background values at 4 km. Sulfate
values were somewhat higher than background values but not nitrate.
Vanadium (V) was used as a tracer for plume location.
The nickel smelting industries in Sudbury, Ontario, have emitted be-
tween 1.5 and 2.7 x 10° metric tons of S02 per year for a minimum of 25
years; the single 381m stack (INCO Ltd) represents the largest anthropo-
genic point source in the world (Summers and Whelpdale, 1976). Sudbury
has been the site of a series of wet deposition field studies. Wiebe and
Whelpdale (1977) examined precipitation concentration measurements of
S02, total sulfur, and trace metals out to 70 km from the INCO stack.
12
-------
They determined that less than 1% of the sulfur emitted during a precip-
itation event was deposited within 50 km of the source, whereas the
fraction for trace metals was at least an order of magnitude higher.
Slightly elevated nitrate concentrations were observed, but chloride
appeared to be at background levels under the plume. A later, more
extensive sampling, project examined 31 events over two years (Chan et
al., 1982; Chan et al., 1984a) . Samplers were of the bulk variety and
were replaced on a 24h basis. For most species the wet deposition was
attributed to sources other than the INICO smelter; the exceptions were Cu
and Mi for which the increase, above background, was an order of magni-
tude. Most particulate constituents (acids, sulfates, trace metals) were
scavenged quite efficiently; however, sulfur which was emitted mostly as
S02 had low scavenging efficiency. On the average, during precipitation
events, the INCO smelter was found to contribute ca. 70% of total wet
deposited Cu and Ni and ca. 2Q% of the other trace metals and sulfur
within 40 km of the stack. For events classified as cold front (west and
northwest trajectories) the ratios of plume-related to background wet
deposition of sulfate and nitrate were higher by a factor of two over the
warm front events (south and southwest trajectories). In terms of emis-
sions the percentage of wet deposited sulfur within 40 km was in line
with most other studies (ca. IX). The dry deposition contribution was
estimated to be considerably larger.
A parallel study was undertaken during the mid-1978 to mid-1980
period in the Sudbury basin. The study was aimed toward the cumulative
precipitation quality (Chan et al., 1984b) with a network of automatic
13
-------
wetfall collectors (wet only); collection was undertaken on a monthly
basis. During most of the first year of operation the INCO smelter was
shut down. Comparisons between the data corresponding to the shutdown
and the operation revealed no statistically significant differences in
the acidity and the concentrations of inorganic species and most trace
metals. It was argued that wet deposition was largely governed by long
range transport, local wind-blown dust and vehicular traffic.
Scheider et al. (1981) compared bulk deposition data (which included
both wet and dry inputs) at eight sites near Sudbury and eight sites in
Muskoka-Hal iburton ca. 225 km southeast of Sudbury. Comparisons were
made between data collected before and after October 1978 when smelter
operations were shut down due to strikes. They concluded that the bulk
deposition of acidity, sul fate and total Cu at Muskoka-Haliburton did not
significantly decrease during the shutdown supporting the view that at
least for H+ and S0^~ long range transport from many sources dominate the
local deposition. At the Sudbury area (within 12 km of the source),
however, they determined that sul fate decreased significantly (5-50%)
during the shutdown period; for Cu and Ni the decrease in deposition was
estimated to more than two orders of magnitude.
Remote sensing of tne S02 plume at Sudbury in addition to conven-
tional wet deposition measurements were employed by Mi 11 an et al. (1982).
Total sulfur wet deposition in the near field agreed with that of previ-
ous studies, although the S02 deposition "plume" appeared at times dis-
placed away from the sulfate and pH plumes, possibly a result of lowered
solubilty of S02 from the increased acidity of the central plume, wind
shear, or low-level source effects. Another Sudbury study (Lusis et al.,
14
-------
1983) involved the measurement of scavenging rates of sulfur and trace
metals from the smelter plume during the fall of 1980 and 1981. Bulk
collectors were used on a precipitation event basis and deployed along
two arcs, 14 km and 24 km downwind of the smelter. Most of the sampled
storms were frontal in nature. For trace metals, despite differences in
their particle size distributions comparable scavenging coefficients were
observed, probably, due to in-cloud modification of the emitted particul-
ate matter. For sulfur the scavenging rate was an order of magnitude
less because of the fact that, within the acidic environment of the
plume, little SO? (which makes up ca. 99% of the emitted sulfur) is
dissolved in rain or cloud drops and removed by precipitation.
A smelter in the state of Washington was the apparent major source
of excess pollutants measured in the Puget Sound area by Larson et al.
(1975); the removal rate estimated for sulfur was comparable to that for
Sudbury and most power plant studies. Mobile rain and snow sampling was
undertaken around a sour gas processing plant in Central Alberta (Summers
and Hi tenon, 1973). They deduced that 30-45% of the sulfur emitted was
removed during summer convective storms within 40 km, but less than 2%
during the winter.
A wetfall chemistry study employing sequential precipitation sam-
plers was undertaken by Pratt et al. (1983) at seven sites clustered
within 40 km of each other in the vicinity of a coal burning plant in
central Minnesota. This plant is located in a rural area 50 km northwest
of Minneapolis/St. Paul. Sampling was undertaken during the summers of
1977 through 1980 and analyses involved pH and major ions. In this
publication no effort was made to relate measurements to this local
15
-------
source. Instead, the data were subjected to statistical investigations
which demonstrated the absence of a correlation between H+ and either
S0/i~ or N0o~. It was argued that weak acids contributed most of the H+
^ O
due to alkaline dust, agricultural activities, and long range transport
from distant pollutant sources.
16
-------
SECTION 3
AREA SOURCES
In studying the impact of local sources on acid wet deposition it is
customary to separate these sources into point and area sources. Area
sources are usually urban and industrial centers. A fundamental distinct-
ion between the two types is the influence of the source on the local
meteorology. Point sources rarely affect the local meteorology; area
sources, on the other hand, are of sufficient size to significantly
affect the local climate and meteorology. This is due to changes in
surface roughness influencing the dispersion mechanisms, different radiat-
ional responses and the inputs of heat and pollutants at different
heights impacting local and downwind cloudiness and precipitation pat-
terns (AMS, 1981). This "weather modification" effect figures prominent-
ly in several aspects of the wet deposition and often necessitates a
different experimental approach compared with the point source studies.
Point source plumes are generally easy to locate and often "controlled"
type experiments may be attempted since emissions can be easily monitored
and detailed airborne and groundlevel measurements in a limited region of
space may provide an adequate picture of the transformation processes
(Gillani et al., 1978, 1981; Forrest et al., 1981) and the subsequent
scavenging characteristics.
Apart from the "weather modification" effect mentioned above, area
sources are also characterized by a complex and often intractable mix of
pollutants derived from the wide variety of transportation, industrial,
and residential activities on the urban scale. The complexity is com-
17
-------
pounded by the temporal variability in emissions (by time of day, day of
the week, season of the year, etc.). Since point sources are usually
coal- or oil-fired power plants, sulfur oxides (SOX) are their dominant
pollutants followed by nitrogen oxides (NOX). For area sources primary
emissions of SOX, NOX, and hydrocarbons (HC) often compete for the dominant
role with significant differences from one source area to another; even
for the same source the dominant pollutant may depend on the time of day,
season, etc. Knowledge of the magnitude and variability of urban emis-
sions becomes an important prerequisite to successfully studying wet
deposition on the urban scale.
It is important to identify the two distinct length scales associat-
ed with urban wet deposition. The first is the scale of the actual urban
area. The importance of that pertains primarily to materials damage
considerations since the density of materials at risk is highest within
urban areas. It may be that dry deposition (i.e. air quality) may be
considerably more important than wet deposition for materials damage.
One facet of wet deposition, fog and dew (Cadle, 1984) may be also impor-
tant in the urban environment due to accelerated oxidation of S02 to
H2S04 because of Mn catalysis (Penkett et al., 1979). Generally, however,
wet deposition in the urban setting will be dominated by below cloud
scavenging with variable net effects. The second length scale is the
suburban-rural one which defines the extent to which the urban plume
markedly impacts the wet deposition above regional background. This
scale may indeed approach 100 km often because the urban plume with some
downwind travel could mix thoroughly and may become incorporated into
18
-------
clouds thus promoting faster acid-forming reactions. Identifying this
second length scale is an important goal of source-receptor investiga-
tions particularly on the eastern coast of the U.S. and Canada where
sensitive receptors are proximate to urban and industrial centers.
Literature addressing urban wet deposition is comparable, in quan-
tity and quality, to that addressing wet deposition around point sources.
There is a general paucity of relevant wet deposition data and several of
the relevant publications have often relied on limited observations.
There are two major field efforts which have attempted to quantify the
urban source influence on downwind wet deposition in the U.S.: the
wetfall chemistry studies around St. Louis (Hales and Dana, 1979b) which
concentrated on summer convective storms and the Philadelphia field
effort (Patrinos and Brown, 1984) which addressed frontal wintertime
precipitation. Both studies concluded that contrary to the point source
results, urban area emissions significantly impact the quality of precip-
itation on the suburban-rural scale. This is true for both convective
and frontal meteorological situations and indeed a significant fraction
of the emitted urban pollutants are deposited within 50 km of the urban
center. Details on both studies will be given later.
The city of Uppsala in Sweden has been the focus of several early
investigations of wet deposition. It is a moderately sized city (popula-
tion 100,000) which is roughly circular in shape (radius ca. 7 km) and is
in a relatively flat area with very few other pollution sources within a
60 km radius. One controversial aspect of this study is the nature of
the urban emissions; ca. 65% of the S02 emissions originate at a 40m
stack of a district heating station. Most of the remainder comes from an
19
-------
additional 12 central heating refuse-burning facilities. Studies at
t
Uppsala are, therefore, more hybrid, a combination of point and area
sources, and not directly comparable to the studies around the large
urban centers of the U.S. Andersson (1969) sampled monthly deposition
(both wet and dry inputs) with a network of 24 stations in an area of ca.
400 km2 surrounding Uppsala to a distance of ca. 40 km during July to
October 1962. Sites within 4 km of the city center were considered
urban; beyond 4 km they were considered rural. At an addditional 3 sites
precipitation was sampled on a daily basis. Samples were analyzed for pH
conductivity, Cl~, Na+, K+, Ca2+ and total S. Urban sites showned
significantly higher Ca2+, Cl~, and S levels; although some of the excess
was due to dry deposition, wet deposition was considered largely respons-
ible. These results appeared to support the claims by Stevenson (1968)
who based his conclusions on data over the British Isles and Eire during
a six-year period. In both these studies acidity did not show a definite
pattern perhaps due to the conflicting influences of Ca2+ and S. Anders-
son attributed the urban effect to below cloud scavenging. Rhode (1970)
conducted a simple analysis during a 12-day period of February 1969 by
collecting fresh snow samples at 31 sites out to distances of 15 km from
Uppsala. The sites were away from main roads and the samples were anal-
yzed for total sulfur. Of the total emitted 240 tons of emitted sulfur
less than 5% was deposited within 15 km. Hdgstrttm (1974) studied the
wet fallout of sulfurous pollutants emitted from Uppsala during rain or
snow. A network of ca. 100 samplers was deployed at distances of 7, 20,
40, and 60 km from the city's center. Samples were collected on an event
basis and analyzed from H+ and sulfur. Based, primarily, on the data
20
-------
from eight winter storms (during the period April, 1972 to April, 1973)
Htigstrtim concluded that the characteristic scale of fallout appears
in most cases to be in the range 50-100 km. He also found that sulfur
was found to be deposited mostly as sulfate; a small amount was deposited
as S02 fairly close to the source. Despite the fact that their study did
not involve wet deposition measurements the work by Rodhe et al. (1972)
is relevant in this discussion. They measured soot, particulate sulfur,
and S02 in air on a daily basis at four remote coastal sites and at ten
towns in southern Sweden during September to December 1969. They con-
cluded that long range transport from source regions to the SE and SW was
responsible for a significant percentage of the sulfur found at both the
coastal and urban sites, particularly for sulfate; for soot and S0£ local
emissions dominated the urban sites. They also estimated a less than 5%
SOg to sulfate transformation rate on the urban scale.
In the U.K. Davies (1976) studied the precipitation scavenging of
S02 in the industrial region of Sheffield with significant iron and steel
industries and a power station. Although it had been reported (Bielke
and Georgii, 1968) that S02 contributes as much as 75% of the total
sulfur in precipitation Davies expected that this figure would fluctuate
widely. The study involved a single site with a sequential precipitation
sampler with a recording raingage and hourly S02 measurements over a
period of one year. Dissolved S02 was measured with the West and Gaeke
(1956) method following the treatment of the rainfall samples with
tetrachloromercurate (TCM). Results indicated that in absolute terms
precipitation removes only a very small fraction of the emitted S02 in
the industrial Don Valley as S02 washout even during the time of rainfall.
21
-------
For individual rainfalls, the gradient of the washout-rainfall amount
curve is smaller than the gradient of similar curves for sulfate. In
general, the washout S02 concentration is quite low for light summer
rains compared with light winter rains. Dissolved SOg values showed no
obvious correlation with the majority of presumably important meteoro-
logical parameters emphasizing the inherent complexities of the processes
in the industrial atmosphere. Following up on this study Davies (1979)
examined S02 and sulfate in urban and rural precipitation in Norfolk,
U.K. One sampling site was situated ca. 1 km west of Norwich (population
ca. 150,000 with some light industry). The other site was 16 km west of
Norwich, considered quite rural. Indeed most of Norfolk is quite rural
with major sulfur sources situated more than 150 km to the SW. Sampling
involved bottle funnel combinations (one with TCM for $03 fixing) as well
as recording raingages. Atmospheric S0£ concentrations were measured in
the urban area and at a rural site. Monitoring was undertaken on an
event basis between 10/77 and 8/78. Results are based on a total of 50
"urban-rural" pairs. Davies concluded that for the rural site ca. 15% of
precipitation sulfate is derived from reactive scavenging of $03; for the
urban site the corresponding percentage was 22%.
Another relevant study not involving wet deposition data is by
Benarie (1976). It is based mostly on air quality data for one year from
a station 37 km from the center of Paris, France. Based on existing
emission inventories Benarie concluded that on an annual basis not more
than half of the sulfur emissions leaves the 37 km radius boundary.
Gaseous and particulate dry deposition were considered mainly respons-
ible.
22
-------
In North American, Liljestrand and Morgan (1981) examined the
spatial variations of acid precipitation in southern California. Wet
deposition data were collected at 9 sites in the Los Angeles basin of
Southern California during the 1978-1979 hydrologic year. Sampling was
performed on an event basis and samples were analyzed for all major ions,
pH and conductivity. They determined that the major net acidity compon-
ents H2S04 and HN03 are partially neutralized by gaseous NH3 and alkaline
soil dust. Results indicate that the net acidity flux by wet deposition
is ca. IB% of wet deposition in the northeastern U.S. (ca. 50% of that in
the southeastern U.S.) due primarily to the arid climate and higher
contributions from alkaline sources. Based on emission estimates the
conclusions were that less than 2% of sulfur emissions and 1% of nitrogen
emissions were scavenged on an annual basis in southern California.
Nitric acid was the predominant acid however, particularly at the inland
sites; H2S04 showed maxima near the coast. Due to the distinctly different
climatologies these results have dubious application for the northeastern
U.S. and southeastern Canada where the emphasis of "effects" research has
been placed.
In Nova Scotia, Shaw (1982) combined back trajectory calculations
with analyses of air and precipitation quality to estimate the relative
importance of different source areas. Data were collected during 1979 at
a rural, coastal site ca. 25 km from the city of Halifax (population ca.
260,000). Aerosol sulfate was collected every third day (24h run) and
precipitation with an automatic wetfall collector on an event basis.
Chemical analyses of the precipitation samples involved only the major
anions (S0^~, N03~, Cl~) and pH. Nitrate concentrations in the rainwater
23
-------
were found to be 1/10 of the sulfate concentrations; it was speculated
that the nitrate concentrations were low because of delays in the anal
yses leading to biological degradation. Based on results from 56 indi-
vidual storm events and the relevant surface wind data Shaw concluded
that 50% of the annual wet deposition of H+ and SO^ within a 25 km rad-
ius of Halifax is due to the S02 Halifax emissions. These findings were
supported further by data from the Dutch Settlement ca. 40 km to the
northeast of Halifax. However, no effect of the city's emissions was'
felt at the monitoring sites of Truro and Kejimkujik ca. 100 km from
Halifax. Shaw also confirmed the highly episodic nature of wet deposi-
tion of atmospheric acid: eight of the 56 storms contributed ca. 45% of
the total annual wet deposition of S0^=; one storm, in fact, contributed
ca. 13% of this total. It is reasonable to conclude that wet deposition
in Atlantic Canada is more important that dry deposition. Shaw's results
confirmed the findings of Watt et al . (1979) who attributed the high
concentrations of H+ and S0^= in lake waters in the vicinity of Halifax
to the uroan emissions, and of Wiltshire (1979) who performed a sulfur
budget for Nova Scotia concluding that 25% of the total deposition of
sulfur in Nova Scotia is due to local emissions while 30% of local emis-
sions are deposited within that province.
A comparison of "Lagrangian" meteorology with pH measurements around
Washington, D.C., was undertaken by Draxler (1983). The pH measurements
represented daily samples at 6 sites within a 15 km radius of Washington,
D.C., during 1975. Approximately 75 sets of data were used in the anal-
yses which were compared against the Lagrangian averages of temperature,
rainfall, relative humidity, and normalized concentration from a multiple
24
-------
source trajectory model. Draxler found that the most consistent acidic
rainfall occured during an extended period from May to June when the flow
was from the north. He concluded that the most important predictor of pH
during the summer was the average rainfall along the trajectory. Based
on this result he claimed that complex chemical source-receptor models
were unnecessary for summer simulations; for the winter, results were
ambiguous.
The hypothesis that local sources impact local wet deposition was
tested by Dasch and Cadie (1984) and Dasch et al. (1984). Air and pre-
cipitation quality data were collected for a period of one year (June 81
to June 82) at two sites in S.E. Michigan, one in Warren which is ca. 7
km north of Detroit and the other ca. 50 km to the north of the first
site. The Warren site was considered "urban" while the second site was
considered "rural." Estimates of local emissions at the two sites showed
a 300-fold increase in terms of SOX (urban to rural), nine-fold increase
in terms of NOX and 56-fold for TSP. Precipitation was collected mostly
on a daily basis. Dryfall was collected on a weekly basis from the "dry"'
buckets of the automatic collectors. Although air quality concentrations
and consequently dry deposition, reflected higher values at the urban
site wetfall chemistry was remarkably similar at both sites with only
exceptions the Na+ and Cl ~ ions which showed higher concentrations at the
urban site; precipitation samples were analyzed for all major ions, pH
and conductivity. No evidence of local primary S04= was evident for both
winter and summer rainfall. These results are generally at odds with
most other urban deposition studies and underline the inherent difficulty
of studying wet deposition on the urban scale with a limited number of
25
-------
sites. It is conceivable that there was such a predominant transport
*
during the sampled events that the effect of the urban plume was not
detected at either site. Furthermore, the "urban" site may have been too
close to the urban sources with below cloud scavenging being the only
contribution to its wetfall chemistry.
The METROMEX study (AMS, 1981) and the RAPS effort (Schiermeier,
1978) were the first comprehensive field endeavors to investigate all
aspects of the urban influence on meteorology, air quality and deposit-
ion. St. Louis, Missouri, was chosen as the focus; it is a continental
city of negligble topographic relief with ca. 2.3 million people in the
metropolitan area and substantial industrial activity. The wet deposit-
ion efforts concentrated mostly on the summer periods of 1972 and 1973
(Hales and Dana, 1979b). The primary objective was to assess the effect-
iveness of convective storms in removing urban pollutants and to provide
a data base for scavenging model development. The network employed ca.
60 event-type samplers around St. Louis with different configurations
during 1972 and 1973. Results are based on ca. 12 convective precipita-
tion events; the samples were analyzed for most inorganic nonmetallic
species. Hales and Dana concluded that precipitation scavenging is a
highly efficient removal mechanism since the quantities of rainborne
material (S0^~ and NO-j") deposited on the network downwind of the urban
area because of the urban plume are comparable to the urban pollution
burden (for SOX and NOX). Much of the observed rainborne S04= and NC^'
appears to have been incorporated into the rain by scavenging of gaseous
precursors. It may be that rapid oxidation of S02 to sulfate occurs in
cloud systems in warm, polluted environments and leads to the observed
26
-------
seasonal trends in S0^~ levels. The St. Louis wet deposition pattern
showed a definite lack of organization in the H+ patterns and poor cor-
relations between H and S0^~ and H+ and N03~ ostensibly the consequence
of the competing effects of NH^"1" and the more abundant metallic cations.
The Hales and Dana results appeared to contrast the conclusions of Scott
and Laulainen (1978) whose analysis of winter-storm scavenging data from
Michigan suggest [in concordance with Scott (1978)] that direct nucleat-
ion of cloud droplets by particulate S0^~ is the principal mechanism for
scavenging. However, there are important differences between the two
sets of results. For example, temperatures during the St. Louis experi-
ments were 30°C warmer than the Michigan ones; increased photochemical
activity during the St. Louis storms would produce substantial amounts of
oxidants (63 and H202) which contribute to the reactivity of the scaveng-
ing mechanism. Furthermore the urban environment is most likely richer
in catalytic trace metals which accelerate certain oxidation reactions.
Regarding the deposition of nitrate, results disagree, in general, with
those of individual power plants possibly because of faster formation of
soluble products such as nitric acid. Even though the origin of nitric
acid is suspected to be photochemical observed patterns for both day and
night storms suggest other potential pathways to the nitric acid product-
ion (Richards, 1983).
Examining the relationships between measured sulfate concentrations
in air and rain in St. Louis on four days (during the summers of 1974 and
1975) Bridgman (1984) disputed the reactive scavenging claims of Hales
and Dana. Bridgman detected considerable variability in both media
27
-------
across the mesoscale network. Aerosol S04= varied by a factor of up to
nine from stations directly affected by point source emissions to those
most unaffected. Spatial variability of normalized sulfate in rain is on
the order of a factor of 3, although at occasional individual sites S04=
concentrations were much higher. In three of the four cases, sul fate
concentration patterns in air and rainwater were similar, and consistent
with wind direction and location of potential sources. The similarity in
patterns of SO^" in air and rain and the location of the highest concen-
trations of both close to the major sources of SOX emissions indicated to
Bridgman that nucleation must be the major cause of sulfate scavenging
with sub-cloud impaction perhaps having a small role.
Associated with METROMEX was the study by Gatz (1980) who examined
wet deposition data from ten convective rain events during the summers of
72, 74, and 75. Approximately 85 collectors covering 2200 km2 around St.
Louis were analyzed for Li, Na, Mg, K, Ca, Fe, Zn, Cd, and sul fate (sol-
uble and insoluble). Factor analysis was applied to these mesoscale data
and results showed that deposition patterns grouped consistently into
four main types (1) soluble soil elements, (2) insoluble soil elements,
(3) soluble pollutant elements, and (4) insoluble pollutant elements.
Gatz concluded that the differences between soil element and pollutant
deposition patterns reflect different source regions while the differ-
ences between soluble and insoluble element deposition patterns reflect
differences in scavenging and/or precipitation formation processes.
28
-------
SECTION 4
THE PHILADELPHIA FIELD STUDY
BACKGROUND
The Philadelphia field study was conceived partly in response to the
growing controversy regarding the impact of large area sources on local
wetfall chemistry (Spaite and Szabo, 1982). The only other study similar
in scope and objectives was the one conducted during METROMEX (Hales and
Dana, 1979b) at St. Louis, during the early 1970's before the emergence
of widespread awareness of the "acid rain" problem. The conclusions of
the St. Louis study were limited to summertime convective storms in the
Midwest. The geographical limitation is not considered important; it is
felt that insofar as the meteorological effect on wet deposition for
convective storms is concerned, the results could be accepted for area
sources in the eastern U.S. and Canada. The "data gap" existed primarily
for frontal and primarily nonconvective storms which are responsible for
the largest fraction of wet deposition, particularly in the Northeast
[according to one estimate, greater than 60% (Raynor and Hayes, 1982)].
Available estimates of the importance of area sources on local wet depo
sition have been based on spatially limited data (Dasch and Cadle, 1984;
Shaw, 1982) with conflicting conclusions. More complete data are avail-
able from Europe, particularly Sweden (Hflgstrtim, 1974), but there is
serious doubt whether these results are relevant to the U.S. since the
Swedish area sources are considerably smaller and the climatology some-
what different.
29
-------
Philadelphia was chosen as the test site because it is a reasonably
large city (population ca. 2.5 million) with considerable industrial
activity along the Delaware Valley. It is, also, sufficiently inland to
avoid the complications of a mostly coastal city (e.g. New York or
Boston). Philadelphia has been the target of several air quality studies
[e.g. the Philadelphia Aerosol Field Study (PAFS)] and consequently has
contributed to a reasonable emissions inventory, perhaps better on the
average than for most other cities. A possible complication is the
position of Philadelphia in the center of the heavily urbanized and
industrialized "Northeast Corridor" (from Washington, D.C. to Boston)
with possible multiple urban plume interactions. As it developed this
complication was not a serious one because of fortuitous meteorological
conditions during the sampled storms.
The field study was geared toward the sampling of frontal and pri-
marily nonconvective storms and was divided into two parts: an explor-
atory phase during the fall, 1983, period and the main field effort
during March and April of 1984. The primary objective was to sample a
sufficient number of storms in order to provide an adequate data base for
a quantitative assessment of the area sources impact on mesoscale wet
deposition.
THE EXPLORATORY PHASE AND QUALITY ASSURANCE
The exploratory phase of the field study was undertaken to investi-
gate the suitability of the candidate sampling sites and the density of
the sampling network, to establish a comprehensive quality assurance and
quality control (QA/QC) program, and to test the field study logistics.
Approximately 40 sites were selected on both sides of the Delaware Valley
around Philadelphia and within a 60 km distance from the river (Fig. 1).
30
-------
5
IV
Figure 1
204
TRENTON
PHILADELPHIA
' 33 * -*
20km 20km
31
-------
CM
CO
c
o
^
4->
ro
4_>
r"
o.
^
0
Oi
(_
Q.
i *-"
| Ol 00
5- ro
3 O
00 U
E
0)
oo
>.
GO
t_
01
f~
} *
ro
Ol
3
oo
O
f
4->
in
*^
L.
Ol
4_>
O
ro
L.
ro
e~
!_>
Ol
Q.
>^
1
C
O
^
4-J
fa
4-)
.^
Q.
1
U
Ol
u.
Q-
T3
O
g
Ol
Q.
E
c_
o
oo
C 00
r Ol Ol
C- JZ
SZ Q. 4->
Ol
1-20.
J= O 3
^J
ro Ol
C
O1T3 -
C C >
» ra O
C E
C 01
3 e E
<- LU Ol
t- 4->
H 2 to
> 0) >,
O Z to
A
^ Ol
rO Ol
» ra
4-J t-
ro O)
a. >
m ra
E " E
t. -3 U
O Z
<4 ^"
- C
C -r- OO
3
C-
>, Ol X
-r- O
(_ > £_
i- k
o
ra
Ol
J^J
c/)
M
c
fH
ra
QC
oo
CO
oo o
CO <-<
O i"~t
T 1 1 <
^^
r-* r
I-H GO
UJ
1
GO O
UJ O
CM
O CM
OO
CM O
-< 4-J
^H
r 4
1 1
OJ 4-J
t. 00
3 ro
OO O
in o
C 01
p- U 01
Q.X:
.C 4-J
0) 2
- o a.
-c i 3
ro Ol
C
OJTJ
C C =»
r- fO O
C ~- E
C Ol
3 C E
t- LU Ol
C» 4-J
Ol 2 00
> 01 >,
0 Z oo
*
Z U
c oo
M ,^ *
>> f
i OJ
TO -r- X
»- > o
J/ > TO L.
TO t^
>k
E Oi
U 4-J Ol
O .C ra
<4- O) t-
i *^- OJ
c >
ID 00 ra
01
S
ro
4-)
in
T3
C
j
o
rO
Ol
4^
OO
*
C
r
ro
OL
oo
CO
oo -sr
CO ^.
*^ f I
^N^
CVJ t
t i oo
UJ
I
GO o
UJ O
o
O 1-1
o
oo o
O 4->
^f
o
CM
Ol
C f
i 4-J
0 3
C 0
ro oo
>
-a oi
ro f~
4_)
C
(Q P~
O
01 t.
C <4-
!
C E
c t-
3 ro
«- 2
t_
O) LU
> t
O 3
X
O
t_ *
" 0.
E 0.
t-
i- C
C 3
3 O
E
plil.
i 0)
t3 O^
»
OO ra CM
1)
.0
ro
4-)
OO
*
^
O
c
in
oo to
rO 3
O
a 3
Ol C
+j f-
ra CT
4-J O
OO U
^f
CO
*^ {Q
00 ^
^* OO
\f>
^^ ^v
oo oo
UJ
r
OO O
LU O
OO
O O
oo
ui o
0 4->
t-O
o
oo
o
C 4
r- O
J= C
Ol O
f *
r- 4^
T3
a E
i t_
O 0
U LU
rO 2
O
Ol-O i
C C
- ro >>
C . U,
C Ol ro
3 C "O
L. LU C
i- O
Ol 2 U
> Ol Ol
O Z in
^ 01
TO O^
f ro
4-J CH
TO CU
a. >
(/^ TO
E" S
"3 0
O Z
H- OO
- C
C «3-
3
{_
>, a; x
1 !- O
!_ > S_
i- ro Q.
ro Ol Q.
CJU JZ ra
Ol
.a
ra
4VJ
OO
t
2
0
c
00
oo oo
O
oo o
CO
"V.^
^^ ^"
CO r-4
OO OO
t 1
^^^ ^_
oo oo
UJ
I
oo o
LU O
o
o o
o
«3- O
O 4-J
^3"
( *
oo
o
Ol
C -C
i 4-J
U 3
c o
ra to
>
a ai
ro JC
4-J
c
1
0 3
* * ^£
>> x a.
.-? 0
«- c
rO Q.-I-
i- a.
4-> ra Ol
ro Ol E
f-> * ^ ^j ^_j
00 Z t-
01 CO
E c >
l_ - ro O
0
<4- 01 -
- 01 E x
C ro O O
3 C- C-
C Ol OO Q.
o > a.
Z ra O ro
oo
u.
Ol
" 2
Oi O
> .c
4_>
U 4-J
Ol C
> Ol
C 4-J
0 4J
O -r-
E
« L.
C Ol
r 4-J
rO C
oo
^" * 1
CO CM
^^ ^^^
-t OO
CM
"^^ ^B
oo oo
UJ
I
oo o
uj o
'-O
O i-H
o
^^ ^D
0 4J
f 1
CM
oo
o
\7
a i
Ol 4-J
> 01
o c
E
E O
Ol
4-J JZ
tO 4->
>! 3
tO O
to
Ol
3 ro
to oi
OO OO
Ol
S- 0
a. 4->
2 4-> t-
030
102
* ^c
>, X CL
- 0
5- C
rO O.T-
i- O.
4-J ro O)
rO Ol
CL ^O *o
to Z u,
0)
E C >
!_ , ro
o
- Oi E
C ro O
3 «-
c 01 oo
o >
Z ra O
-Q*
rO
Ol
4_)
in
1 *
3
4-J
-C
Ol
.r*.
^_
O)
** ,
C .0
- ro
ra 4-J
o: to
00
^^
C^ (yO
CO CM
^^ *^^
ID OO
CM
*o
oo oo
UJ
I
oo O
UJ O
o
o o
o
^^ o
0 4->
**o
CM
oo
o
E
o
CO
o
X
o
CL
Q.
Ol
.c
o
OJ O 4->
4-J 4-> 4-J
O 3 tO
C Ol O ra
f *s O
o -a o
O ra Ol
Ol t- > -3
E r O Z
c- E
O Ol
14 « JZ
4-J 4-J 4>J
2 oo oo
O ro ra 4-
r O) O) v<_
JZ JZ O
Q. 4-J 4-J
Ol 3 t- ra
Ol O O Ol
O 00 C 00
E
u
^^
*
E ^>
t_
o
<4- X
r- 0
C i_
3 0.
a.
^M
01
ro Ol
4-J *
ra Ol
a. >
OO ro
o
2 4J
>, O Ol 0
i oo .a >
ro ro C
i- Ol 4-J O
4-) E 00 O
- o
C to C T3
t *t Ol
« ra TT
C >, £_ T3
i- > O>
ro ro 4-J .O
4- 01 JZ E
4J !
j= c p- ^:
Ol Ol 4->
i- J= O -i-
1 4-1 4-J 2
CO
^^
«s- cr>
CO CM
^^ ^^
r^ oo
CM
^**^ ^~
oo oo
UJ
I
00 O
LU O
o
O CM
o
0 0
CM 4->
en
CM
oo
o
4J
oo
Ol
2
LU O
3 oo
a 3
E Oi -1-3
U. JZ
ra o -a
2 "3 Ol
4-J OO
ra 4-J oo
ra ra
Ol O.
C 4->
- C JZ
coo
C L. -
3 «*- JZ
L. 2
t~ E
Oi c- 2
> ra O
0 2 r-
^- E
(O O
4-J CO
> 01
i ra
(_ t-
- O)
ro >
U- ra
*
>-
^3
ro
Ol
in
>.
,».
(_
liv
rO
Ol
C rQ
^- ro
ra 4-J
OC 00
^>
CO
*3- on
CO ^^
^
^^ ^_
«* oo
LU
h-
oo O
uj o
O 1-1
00
co o
O 4-J
Lf>
o
o
N^
C-
o
2
Ol
c
(4_
o
-------
For the exploratory phase the samplers were simple bottle-funnel combin-
ations (B samplers) which were deployed a short time before the onset of
precipitation and collected shortly after the cessation of precipitation.
Deployment and collection timing was, therefore, crucial in minimizing
dry fall contamination; these activities were closely coordinated with
the National Weather Service (NWS) forecast office in Philadelphia.
The exploratory phase involved the sampling of two precipitation
events during the fall of 1983. Table 2 presents the important meteoro-
logical features of these storms. Emphasis was placed on cyclonic-type
storms with minimum convective activity and as it developed the selection
was quite good.
A comprehensive QA/QC program was developed for this study
(Patrinos, 1983); it covered all aspects: the sampling instruments, the
sampling procedures and the chemical analyses of the collected rainfall
samples. The development of rigorous and scientifically defensible QA/QC
standards was an important goal of the Philadelphia study because of the
realization that QA/QC shortcomings had compromised a number of previous
wet deposition studies.
Interim QA/QC results from the field study have been satisfactory
(Patrinos, 1984). Several aspects of the QA/QC program are presented in
order to highlight the importance of QA/QC in wetfall chemistry studies.
0 The wetfall chemistry automatic wetfall collectors (A samplers),
which were used during the main field study, have two sensiti
vity controls. One establishes the intensity of the signal
which triggers the opening sequence, the other controls the
33
-------
degree of heating of the sensor head. The first controls the
degree of moisture sensing; if the setting is too sensitive,
the instrument will respond too quickly, potentially exposing
the samples to dry deposition. On the other hand, if it is not
sensitive enough, it may miss the first part of the rain which
may contain important chemical information. The second deter-
mines the speed with which the collector responds to the ces-
sation of rainfall. Both sensitivity controls should be ju-
diciously set for the "typical" circumstances expected during
the field operations. What is perhaps more important is that
all wetfall collectors used in the field be uniformly adapted
to identical sensitivities. This was accomplished during a
two-month period (December 1983 to January 1984) when all
collectors were deployed side-by-side and their behavior moni-
tored continuously during real and simulated rain events. This
allowed the periodic adjustments of the sensitivities to accom-
plish a uniformly acceptable performance by all collectors
prior to deployment.
Rainfall samples were collected into polyethylene bottle-funnel
combinations which fit in the standard 13-L bucket of the
automatic wetfal1 collector. Thus, each site was represented
by two independent samples and provided an assurance against
inadvertent contamination due to operator error or fugitive
debris. An advantage of the use of the bottle-funnel combina-
tion is the minimization of escape of dissolved gases in the
precipitation samples due to the narrow funnel stems.
34
-------
0 Each automatic wetfall collector was collocated with a raingage
geared for daily recording for the accurate estimation of
beginning and ending times of precipitation and of rainfall
rates. The presence of the raingages was a check for the
collection efficiency of the automatic wetfall collectors.
Collection efficiency was often doublechecked with the deploy-
ment of B samplers at several A sites. In fact, during the
main field study a B network was intertwined with the A net
work. Apart from increasing the sampling density this provided
an estimate of dryfall's contribution to the wetfall chemistry.
0 The chemical analyses of the collected rainfall samples repre-
sent one of the most important and costly activities of the
field effort. The QA/QC program involved the use of two inde-
pendent analytical chemistry laboratories (BNL and PNL), use of
dynamic blanks (standards) from both laboratories to be used as
dummy rainfall samples and blind analyses of parts of the same
rainfall samples from several sites by both laboratories. Upon
collection, rainfall samples were rapidly analyzed at a field
laboratory for pH and conductivity after aliquots were taken
for TCM fixing (for dissolved S02 estimation).
THE MAIN FIELD STUDY AND GENERAL RESULTS
The main field effort was conducted during March and April of 1984.
Most of the sites used during the exploratory phase were also used during
the main field effort. Twenty four of the ca. 40 sites were designated
as A sites; there were ca. 14 B sites. Sampling procedures and operat_
35
-------
i.onal details are presented elsewhere (Patrinos et al., 1983; Patrinos,
1983; Dana et al., 1984; Patrinos and Brown, 1984). Nine precipitation
events were sampled during the main two-month experimental period. With
two exceptions these events represent the period's total wet deposition.
The exceptions were, a snowstorm on March 8 and a number of convective
showers which occurred during the early part of the week of April 16.
The network was not activated during those times because of sampling
problems associated with entirely frozen precipitation and because the
showers were forecasted as too scattered. Summaries of the important
meteorological features of these storms are presented in Table 2. As
with the fall 1983 period, the majority of the storms were primarily
nonconvective in nature and were associated with overrunning. This is
typical of a dominant type of precipitation in the Northeast for this
period: strong cyclonic systems forming in the southeastern United
States and travelling to the northeast. A frequent occurrence is the
development of a secondary low pressure system off the coast of Virginia
which tracks to the north; this is particularly the case when the main
cyclonic system is located further inland. The meteorological character-
istics of this storm are responsible for the prevailing southeast to
northwest transport configuration during most of the sampled storms.
According to published estimates of source emissions (Benkovitz,
1980) and more current information from the air quality monitoring agen-
cies of the city of Philadelphia and the states of New Jersey (NJ) and
Pennsylvania (PA) approximately 80% of the SOX, NOX, and HC emissions in
the sampling region occur within 20 km of Delaware river. Consequently
36
-------
the prevailing transport configuration caused the NJ sector of the net-
work to be the upwind (or control) region while the PA sector was the
downwind (or target) region of the network. The determination of the
prevailing transport was made primarily on the basis of data supplied by
the NWS. Four of the sampled eleven storms (from both the exploratory
and main periods) were considered the "best" in establishing the "source
receptor" relationship for the following reasons: clear start and end in
precipitation, steady transport during the precipitation, fairly uniform
(spatially) and sufficient rainfall amounts, minimum convective activity
and satisfactory QA performance. These storms were 1111, 1204, 0405, and
0424. Storm 0314 was also quite useful although the transport veered ca.
180° towards the end of precipitation. Continuously veering transport
also compromised storm 0306 making a control-target delineation impos-
sible. The latter two storms would have been extremely valuable if
sequential precipitation sampling was implemented. Storms 0414 and 0415
are of limited value because, despite steady transport, they were part of
a stagnant weather system with convective activity and intermittent
periods of light rain. Storms 0321 and 0326 shed no light on the impact
of urban emissions because of very low rainfall amounts on the NJ and PA
sectors of the network respectively. Storm 0329 was compromised by poor
QA performance; this was an intense coastal storms with high winds re
suiting in widespread instrument malfunctions (power losses, tipped
sampling buckets, contaminated samples, etc.) and a limited number of
val id data.
Table 3 presents the statistics of deposition of the various species
for the various storms. The unit of deposition is in umol/m2. Sample
37
-------
I/I
eg
in
CO
o o
t- CL
OJ 10
c. o
T
It £
IQ sa
o
O
C. <«-
c
CU i-
ai U
o
T3 >*-
ai
CL 0)
CU t-
T3 as
c
3
g
o
c/1
LO
cn ^H
i i
cn LO
3- CO
i i
«* CO
CM t-H
p^*
LO O
r-t CO
1 1
CO VO
P** *^"
CO l-H
1 1
«J- CO
CM r-H
CM
t-H
r-H CO
1 r-H
LO 1
P^ CM
CM cn
1 1
^f CO
CM t-H
r-.
^J" r *
CM r-H
1 1
r-H LO
P- CO
ft CM
I I
«f CO
CM t-l
o cn
CO CO
i i
VO CM
CO »*
1 1
*J- CO
CM I
CO
r-H LO
^ l-H
1 1
CO VO
CO r-H
CM P-.
| |
|
LO *3-
I ^J-
CM 1
r-H CO
CO
CM 00
1 CO
CM 1
o cn
CM CO
1 r-H
CM 1
r-H CO
LO
CO
1 O
CM CM
vo i
r-H cn
i vo
CM 1
^H CO
VO CO
P-. I 1
1 r-H
CM 1
CM *J-
VO CO
1 LO
CM 1
r-H CO
t-H P-»
VO r-H
1 r-H
VO 1
vo cn
P-. LO
i p**
CM 1
t-H CO
CO
OO
1 LO
CM CO
CO 1
O CO
I-H cn
1 CO
CM 1
r-H CO
LO
CO P-.
I 1 r-»
1 CO
cn i
co en
en I-H
r-H !
1 CM
CM 1
ft CO
CO CO
OO O
1 CM
cn i
cn P--
Lf) CO
r-H VO
1 r-H
CM 1
r-H CO
vo
CO CO
1 CM
OO 1
VO
o
CM r-H
1 CM
CM 1
r-H CO
^-3 «£
Z 0-
o
i vo
cn i
d- cn
cn CM
i cn
CO 1
ft VO
VO
VO CO
l-H P-»
1 CM
CO 1
r-i cn
VO LO
^H ^^
1 t-H
CO 1
ft vo
CM
t-H P*^
t-H VO
1 t-H
LO 1
«3- CM
P^ r"H
t-H 00
i ft
CO 1
r-H VO
CO
LO
CO
1 I-H
CM 1
LO
CM
CM CM
1 CM
CO t
r-H VO
r3 «£
Z 0-
P^ rf^
CO -H
LO 1
i cn
«3r 0
CO CM
r-H 1
1 O
cn i
in
cn co
CO 1
i cn
CO O
VO r-H
CM 1
1 O
Cn r-H
cn en
r-«. CM
i i
-H VO
CM VO
CM 1
1 O
G\ ^H
I-H cn
co co
i i
^ i i
r-H cn
l-H 1
1 O
^^ ^^
o
*f VO
in i
i cn
LO CO
vo ^
CO 1
1 0
cn ft
o
CO VO
m i
1 VO
CM O
ft CO
CM 1
1 O
cn I-H
in
co *r
«! t
cn LO
CM O
i-H LO
1 i-H
CM 1
r-H CM
CM
en CM
1 CO
CM 1
^3" *3"
P*~ P*»
1 P>-
CM 1
r-H CM
en co
VO CM
1 l-H
I-H |
^-
CM
LO CO
1 CO
CM 1
i-H CM
-3
1 1
o cn
VO r-H
1 l-H
l-H |
r-H Cf>
^3>
CO l-H
i in
LO 1
CO r-H
I-H in
1 CM
I-H 1
ft cn
VO
CO VO
i vo
O CM
r-H LO
1 CM
r-H 1
r-i cn
>4~ <4-
r-H CO
VO CM
1 t-H
CO 1
CO LO
CO 1
1 LO
r-H 1
t-H cn
CO Tf
CO r-H
1 1
CM 00
LO t-H
r-H t-H
1 1
ft cn
I-H
rs
cn
*3- CM
^3" 1
1 CO
LO 00
m CM
1 1
in «*
^f
CO CM
^t" 1
1 P-.
LO VO
en CM
1 1
in ^p
r-H LO
vo r»»
i i
00 CM
CO CO
r-l CO
1 t
in «3-
cn cn
o o
t-H t-H
1 1
co o
*S- CM
CM *±
| |
LO «3-
vo in
p1^ P*»
i i
I-H cn
CM en
CM i-H
i i
cn P**
VO
CM O
1 1
LO LO
"
CM <-
i i
m^r
-s
-------
CU
m in
cu
<< T-
O LJ
, <"
<~ a.
C-
Q. CU
CU t-
^J fQ
3
iq CO
"3
O
o
O -r-
Q. >
CU CO
Q -a
"- T3
O t_
c_
>a
i O
o «
*-> (TS
VO
1-4 Cn
vo co
1 1
p- en
^"2"
z. o.
-a
cu en
> CO
CO r
i O.
c S
O i/>
4-> CU
CU 3
(-> r
CO O
T3 >
O
cu
CO
a.
CO T3
OJ.CC
[_ +J (T3
CO
S CT)_)
C Z
in M- CQ
c: 4->
O 'Q 4^
4-> CU
Z C t-
a_ a> a>
o >
>i C O
t- O
O O in
3
rQ CO -C
t- .e 4->
O 4J
-a
at
N 01
>, 0)
< O 0)
:SȤ
i O C i
(O (TJ
c
CO (C
CU CO
cu
01 p
co a.
"a. §
E VJ
ct: u- u- s.
u -o
CU
39
-------
volumes are given in mm of rain. The first most important conclusion of
the field study concerns the impact of fresh NOX emissions on the pro-
duction and deposition of nitric acid. If synoptic transport over the
lowest kilometers of the atmosphere remains steady during the storm and
the control-target distinction can be confidently made, target deposition
of nitrate may increase by as much as 200% compared with control deposit-
ion. This was the case for 1111 and 0405; storm 1111 in fact registered
the highest target-control increase both in absolute and relative terms.
The increase for 0424 was somewhat less (ca. 140%), while for 0314 the
veering transport during the tail end of the storm may have diluted the
final impact (ca. 50%). Veering transport is considered responsible for
tne uniform pattern of 0306 while for 0414 and 0415 no firm conclusions
can be reached because of less than ideal sampling circumstances and
insufficient rainfall amounts. Storm 1204 is a special case as will be
described later. Table 3 distinguishes between the results obtained from
the A and B networks; it also presents results of duplicate analyses at
BNL and PNL.
It is important to emphasize the difficulties of using event-type
sampling to estimate longer-term, (seasonal and annual) differences and
trends since a considerable fraction of total wet deposition may be
associated with less organized weather systems causing long, intermit-
tent, and spatially inhomogeneous rain events. Nevertheless, it is felt
that only event-type sampling and perhaps sequential within-event sam-
pling can provide the comprehensive information blocks required to deter-
mine the impact of local sources. Such information blocks, together with
a firm knowledge of precipitation climatology, may provide the necessary
40
-------
assessment of impact on longer time scales. The importance of event-
type sampling is highlighted with storm 1204 which contrasted the major-
ity of the sampled storms. This storm despite the remarkable meteoro-
logical similarities with 1111 showed no impact of the urban emissions on
most ionic species; in fact, a target reduction in nitrate and total
acidity was noted. However, this storm showed a widespread distribution
of dissolved S0£ gas (Fig. 1) and an over 200% target-control difference.
Table 4 shows the statistics of dissolved S02 for the sampled storms.
The dissolved S02 result for 1204 is particularly noteworthy because it
is derived entirely on the basis of B samples and despite good deployment
and collection timing some escape must have taken place. The seasonal
dependence of dissolved S02 is in line with the published literature
(Hales and Dana, 1979a; Dana, 1980; Dana, 1985).
Several explanations have been examined (Patrinos and Brown, 1984;
Patrinos et al. , 1984) regarding the unique features of storm 1204. It
should be noted that the storm occurred on a Sunday morning with most of
the rainfall ending before sunrise. Some of the considered explanations
are presented:
0 Despite apparent meteorological and transport similarities
between 1204 and the other storms, subtle yet important syn-
optic and microphysical differences (cloud base height, verti-
cal winds below clouds, cloud thickness, extent of mixing of
surface layer air with air from aloft during precipitation
formation, etc.) led to reduced S02 and NOX incorporation into
clouds and reduced conversion to H2S04 and HN03 for 1204.
41
-------
Figure 2
so
'4
z
241
266. 206*
236 232
180 290
*232
528
573
*266
263
* PHILADELPHIA
1204
TRENTON
305*
364
259
.434
327
,
200 324
279
219 381
355 *
346
20km 20km
385
42
-------
Table 4. Dissolved S02 Deposition Data for Sampled Storms (unit: umol/nr)
Storm
1111
1204
0306
0314
0326
0405
NJ
PA
NJ
PA
NJ
PA
NJ
PA
NJ
PA
NJ
PA
Sites3
2
3
25
13
-
5b
-
10
13
c
2
Mean
60
55
47
164
-
51
-
84
66
c
115
_
Standard Deviation
20
27
15
31
-
18
-
26
12
c
81
_
aMost samples were below detection level (1 uM); only samples with concentrations
above detection level are given.
^Both A and B samples are used in the computations.
cSample volumes in the PA sector were low but concentrations were similar
to those on the NJ sector; dissolved S02 analyses for storm were performed
at PNL.
43
-------
Weekday-weekend differences in emission rates leading to lower
oxidation processes which are first order in S0£ and N02- The
occurrence of precipitation during 1204 coincided with the
minimum in manufacturing and transportation emissions (EPA,
1982). Evidence of improved air quality near large cities
during weekends has been shown for ozone (Cleveland and McRae,
1978; Cleveland and Graedel, 1979); it is reasonable to assume
that the precursor pollutant mix emitted by the city may be
reduced for a weekend storm leading to the absence of signifi-
cant excess deposition in the target region. Although compar-
isons of air quality data from various local agencies for the
various storms do not generally support this explanation, lower
mixing heights for 1204 may have contributed to the observed
similarities in air concentration patterns.
In general terms, the concentrations of OH control the conver
sion of S02 and N02 below clouds; in-cloud conversions are
controlled by H202 (Penkett et al. , 1979; Schwartz, 1984).
Hence, since both are driven by radical species
2H02 > H202 + 02
they occur more rapidly in the daytime. This is supported by
the abundance of dissolved S02 for 1204. . The nighttime con-
version of N02 via reaction with ozone.
03 + N02 --> N03 + 02
N03 + N02 <--> ^05
N205 + H20 --> 2HN03
44
-------
may, in fact, be occurring but at a substantially slower rate than the
daytime OH reaction. Reduced ozone concentrations due to "weekend"
effects may, also, be slowing the above sequence.
It is possible that all three causes may be synergistically respons-
ible for the unique features of storm 1204. None of the storms of the
March-April experimental period exhibited the exclusive characteristics
of 1204 and, consequently, no further experimental support of an indivi-
dual expanation can presently be pursued. At the same time, the exclus-
ive nature of this storm may, from a statistical standpoint, deemphasize
its importance in estimating longerterm assessments of local source
contributions.
The impact of the urban and industrial emissions on the deposition
of 504" appears less striking compared to the deposition of N03~. The
maximum target control increase was seen for storm 0424 which showed an
ca. 90% rise. Among the others that showed an increase it averaged less
than 50%. As mentioned earlier, storm 1204 exhibited a 240% increase in
the-deposition of dissolved SOg; the presence of these unreacted
amounts in the rain may be another indication of the exhaustion of
in cloudwater which reduces the extent of S02 oxidation. The fact that
no excess sulfate in the target area was seen for 1204 (Fig. 2) supports
the hypothesis that primary sulfate emissions have a negligble contribu-
tion on downwind precipitation chemistry. After all, the suspected major
sources of primary sul fates are residential boilers and would not have
experienced the weekend emission reductions assumed for the manufacturing
and transportation sources. Therefore, it is reasonable to conclude that
whenever a target SO^" excess is observed, it is mostly the result of
rapid S02 to $04= oxidation, particularly in the liquid phase.
45
-------
Table 5. Estimates of NOX and SOX emission rates (in tons/year) along the
Delaware Valley
NOX SOX
Point Sources 110,000 170,000
Area! Sources 140,000 40,000
Total Sources 250,000 210,000
46
-------
The more substantial impact of the Valley emissions on HO^' rather
than S0^~ can be rationalized on several grounds. NOX are typically
oxidized to nitrate more rapidly than S02 is oxidized to sulfate in the
daytime (Spicer, 1982; Calvert and Stockwel1, 1983, PIatt et al., 1984)
and NOX oxidation potentially continues aloft at night while S02 oxid-
ation typically shuts off. Furthermore, gaseous nitric acid is removed
from the atmosphere much more rapidly than particulate sulfate, and some
fine particle nitrates can volatilize to release nitric acid (Richards,
1983). The quantity and nature of NOX vs. SOX emissions may also promote a
more vigorous HN03 production. This may be related to the more diffuse
and lower-level manufacturing, refinery, and transportation emissions of
NOX compared to S02 emissions which may be dominated by a few elevated
sources. The diffuse nature of the NOX emissions may provide longer
residence times, more efficient mixing with other pollutants and oxidants
(Gorham et al., 1984) and generally greater availability for reaction and
scavenging both in-cloud and below-cloud (Levine and Schwartz, 1982).
An attempt was made to estimate the relationship between the de-
posited excess N03~ and S04~ for the various storms and the emitted NOX
and SOX in the sampling region. The emission estimates are based on
published information (Benkovitz, 1982) and data supplied by local agen-
cies. It is unclear whether NOX anissions from local refineries are
accounted for; elevated NOX levels downwind of refineries have been
documented by several investigators (Sexton and Westberg, 1983; Parungo
and Pueschel, 1980; Parungo et al., 1980). As mentioned earlier, 80% of
all emissions in the sampling region occur within 20 km of the Delaware
47
-------
River. Table 5 presents these estimates for both area! and point sources.
Assuming uniform emission rates (from Table 5) and accounting for the
period of rainfall, the ratio of deposited excess N0o~ in the downwind
0
sector to emitted NOX is estimated to be ca. 50% for 1111, ca. 30% for
0405 and less than 10% for 0424 and 0314. For the corresponding sulfur
estimates (dissolved S0£ and SO^3), comparisons indicate ca. 20% for
1111, 1204, and 0405, and less than 10% for 0424 and 0314.
Storm 0424 showed a sizable target-control relative increase but a
low excess deposition to emission ratio for both nitrogen and sulfur.
Apart from the strong meteorological and transport similarities with 1204
there was a similarity with the time of occurrence as well. A consider-
able fraction of the 0424 rain occurred over a Sunday (in fact the Easter
Sunday) evening to Monday morning and consequently some emissions (pri-
marily manufacturing and transportation) may have been reduced during the
early part of the storm compared with the average weekday values leading
to the low excess deposition-to-emission ratios. Nevertheless, the
sizable target-control relative increases of NOo" and SO/T indicate that
^ *
the similarities with 1204 did not extend to the supression of HN03 and
H2S04 production.
It is important to emphasize that for those cases with an increased
target deposition of N03~, the effect appears to increase with distance
from the Delaware River, suggesting perhaps that the impact may indeed
peak beyond the distance of 60 km which represents the domain of the
sampling network. Figs. 3 and 4 present the deposition patterns for 1111
and 0405. This suggestion may have important implications regarding the
-------
NO
Figure 3
734
728
1026 676
550 497
562
814
* 893
635
PHILADELPHIA
TRENTON
270
242
215
226
231
258
198
177
243
23
247 223
143 '24,
218
288
20 km , 20 km
49
-------
LO
o
50
-------
estimates of deposition to emission ratios presented earlier. Assuming a
peak at 80 km with "mirror" decrease beyond and extrapolating the depos-
ition estimates to 100 km downwind of the Valley, almost 100% of the
emitted NOX may be transformed and deposited as nitric acid on the meso-
scale. It should be pointed out that the above mentioned 100% transform-
ation rate may occur only during the precipitation periods and, conse-
quently, for long-term estimates which include both wet and dry periods,
the importance of local contributions to the total acid deposition may be
deemphasized.
The observed patterns of NH^* deposition is one of the surprises of
the field results. It has generally been assumed that the distribution
of NH^+ is spotty (Hales and Dana, 1979b), perhaps as the consequence of
local ammonia sources. In fact, for most of the sampled storms, the
coefficient of spatial variation of NH4+ is comparable to the ones for
the other species. Furthermore, this ion registered an ca. 100% increase
in target deposition for 0424 and an ca. 70% increase for 1111 and 0405
(Fig. 4) thus implicating the urban and industrial emissions. The pro-
cesses responsible for this result are the subject of continuing studies.
With one exception, the deposition of Na+ mirrors the deposition of
Cl~. This is expected from the proximity of the sampling network to the
coast and the predominance of sea salt as the source of Na+ and Cl~. The
stratified patterns of Na+ and Cl~ deposition with the monotonic decrease
with distance from the coast (Patrinos and Brown, 1984) were another
confirmation of the southeasterly transport which dominated the sampled
events. Even the slight excess of Cl ~ is in agreement with regional
51
-------
results of the MAP3S/RAINE precipitation chemistry network (The MAP3S/
RAINE Research Community, 1982). The exception is storm 0424 which
showed considerably higher Cl to Na ratios over the entire network. The
cause for this exception is as yet undetermined but is of great interest
because of the rising concern regarding the importance of chloride in a
wide range of toxic cross media problems (Milot, 1985).
DIAGNOSTIC MODELING OF A PHILADELPHIA STORM
Modeling activities in Task Group C of NAPAP include the development
of an Eulerian chemical model to study wet deposition on the mesoscale.
The "mesoscale" model is intended to complement the Regional Acid De-
position Model (RADM) currently under development (NCAR, 1984) in its
subgrid parameter!zations and to identify those processes which are
important on the mesoscale. An "engineering" version of the mesoscale
model will be made available for assessment purposes. Current plans
envision the mesoscale model as the combination of a dynamic model
(Kaplan et al., 1982) and a transport-chemical model (Carmichael and
Peters, 1984).
As a prelude to the mesoscale modeling activities a series of
diagnostic model ing exercises were undertaken (Patrinos and Kleinman,
1984) using a multilayer Lagrangian photochemical model (Kleinman, 1984)
with simple scavenging parameterizations. These exercises concentrated
on the simulation of conditions prevailing during storm 0405. Their goal
was to determine whether chemical pathways thought to be significant both
in-cloud and below-cloud coupled with reasonable meteorological and
emissions inputs may explain the excess nitrate and sulfate in the
52
-------
downwind precipitation samples. The model employs a gas phase chemical
mechanism adapted from the ERT models (Atkinson et al., 1982; Godden and
Lurmann, 1983) and involves 40 chemicals and 65 reactions; an aqueous
phase chemistry module is included which involves the dissolution of
aerosol sulfate, the liquid phase reactions of S02 with 03 and ^02, the
dissolution of gas phase HN03 and the production of NOo" via the dis-
sol ution of ^05.
Storm 0405 was chosen as a test case because it had a substantial
"control-target" contrast in wet deposition for both NOg" and SO^" and
presented meteorological features amenable to simple parameterizations in
a Lagrangian framework perpendicular to the Delaware river. Steady
southeasterly transport prevailed during the storm with low lying stratus
clouds (base at 500 m) and with the frontal surface at significantly
higher altitude; the surface front remained well to the south during the
storm. The Delaware Valley emissions are assumed to enter the clouds
through low level convergence and the "control" deposition is due to
pollutants from NJ and from the southwest of the sampling area. Table 6
presents some details of "background" pollutants, emissions and model
parameters. The calculated excess deposition of N03~ and S0^~ was
derived by calculating the amount deposited in a four hour period down-
wind of the river and integrating over the trajectories spanning the
duration of the storm. For N03~, the model calculations explained 60% of
the observed excess with the ^05 dissolution dominating the HN03 dis-
solution three to one. For SO^", the model explained 75% of the observed
excess with the 03 reaction almost entirely responsible. Modeling of
additional storms is planned.
53
-------
Table 6. Diagnostic Modeling Parameters and Inputs for Storm 0405 Simulation
NO,
03.
CO
CHd
RHC
Background
Pollutants (ppb)
1 0.1 40 200 1700 19
SO? NO* HC
Emissions (ppb in
2.4 km layer)
2.1 4.1 9.3
Model Parameters
Wind Speed
Storm Duration
Cloud Temperatures
Cloud Depth
Cloud Liquid Water
Lifetime of Cloudwater
Vertical Velocity
20 km/h
24 h
1-10 °C
2 km
0.5 gr/kg
1 h
10 cm/sec
54
-------
SECTION 5
CONCLUDING REMARKS
This chapter has addressed the issue of the impact of "local"
sources on local and mesoscale wet deposition. A sizable portion of the
relevant literature (primarily the "open" literature) has been consulted
and reviewed and particular emphasis has been placed on recent findings
of urban plume effects on mesoscale wetfall chemistry (the Philadelphia
study) (Patrinos and Brown 1984; Patrinos et al., 1984). As expected the
reviewed scientific results are often contradictory regarding the magni-
tude and sometimes even the existence of a "local source" problem. Three
of the causes for this confusion will be presented and discussed:
1. The complexity of the physical and chemical processes involved in
the emission, reaction, scavenging, and deposition of the pollutants
implicated in acid wetfall. Examples of meteorological variability
seriously affecting the ultimate deposition include the degree of
convective activity during precipitation which may, for example,
accelerate the oxidation processes and increase the rate of acid
production; this may be due to longer residence times of pollutants
in-cloud and to the greater availability of oxidants resulting from
increased photochemical activity. Another example is the chemical
nature of the "background" rain. Since most sulfur emissions occur
in the form of S02 and its solubility in cloud and rain droplets is
a function of pH (and to a lesser extent temperature) the near
source wet deposition or liquid phase oxidation will be highly
dependent on the chemical history of the precipitating system.
55
-------
Particularly for area sources, such as urban or industrial centers
with rich pollutant mixes, the presence of certain substances may
accelerate certain reactions and lead to significant local effects.
2. The paucity of relevant data in the near field of point and area
sources. Regional studies on acid wet deposition have relied on
precipitation chemistry networks such as MAP3S (The MAP3S/RAINE
Research Community, 1982), NADP (Semonin and Bowersox, 1983), CANSAP
(Barrie and Sivois, 1982) and others. These networks have purposely
located their sites away from large sources (point and area) for
more regional representation. Data from these networks have contri-
buted to trend analyses and comprehensive attempts at source-receptor
characterizations on a regional scale. Nevertheless, the sink
terms based on these networks have been underestimated particularly
for nitrate whose characteristic fallout scale from its emission
point is shorter than that for sulfur. In recent years, some urban
monitoring has been initiated such as in the New York City area
(Volchok and Freeswick, 1981), in Philadelphia (Dugan, 1984) and
others, but these attempts have not matched the regional networks in
resources and funding commitment. Thus, the current state-of-the-
knowledge of local source impacts is based on limited wet deposition
data, often collected on a "campaign" basis: the deployment of a
sampling network during select periods and the extrapolation of the
results to seasonal or annual assessments. An alternative approach
has been the operation of a few sampling sites in the vicinity of the
sources for longer times. Although the results may be more reliable
56
-------
from a climatological standpoint the limited spatial coverage has
often failed to characterize the local source-receptor relationship.
3. Incomplete or inadequate analytical procedures or unsatisfactory
QA/QC standards. One of the common weaknesses of earlier studies
has been the incomplete chemical analysis of the precipitation
samples thus providing only partial and often confusing answers.
Insofar as QA/QC procedures are concerned it should be emphasized
that the "local source" environment requires more stringent QA/QC
standards compared with those for regional networks because of
generally poorer air quality (and consequently higher potential
dryfall contamination) and severe constraints on the siting of the
sampling instruments. Despite the above mentioned shortcomings some
general conclusions regarding the local source issue may be reached.
A. The impact of "tall stack" point sources on near field acid wet de-
position is small. Tall stacks may be defined as greater than 50 m and
the near field as within 30 km. On a budget basis the average percentage
of emitted SOX and NOX, during the precipitation, which is scavenged in the
near field is less than 5%. The percentage for trace metals, however,
particularly from smelters may be an order of magnitude higher.
Chloride, whenever present, is also almost totally scavenged in the near
field. Below cloud scavenging appears to dominate in this case. Further-
more, even for SOX the above percentage may under special circumstances
be higher. These may include faster oxidation and scavenging rates due
to, e.g., favorable convective activity in the vicinity of the source
leading to in-cloud scavenging or the presence of certain oxidants or
catalysts. As mentioned earlier, since most SOX is emitted as S0£ its
57
-------
near source deposition is very much a function of the background rain
chemistry. The opening statement of this conclusion needs to be quali-
fied further. Even though from a budget point of view the impact may be
small the contrast between "target" near field and "control" background
particularly in relatively pristine areas may be striking. Beyound the
30 km distance defined as near field for point sources it may be argued
that the source-receptor problem assumes regional characteristics.
B. The impact of area sources, such as large urban and industrial com-
plexes on mesoscale wet deposition is significant. Area sources may be
defined as population sources of ca. one million with SOX, NOX, and HC
emissions in the 10^ tons/year for each from a variety of industrial,
residential, and transportation sources. Two deposition scales have been
identified with regard to those sources. The first is the area source
itself. Wet deposition on that scale is dominated by below cloud scaveng-
ing. Due to higher pollutant air concentrations sul fates and nitrates in
rainwater would be higher than regional values. However, due to high
concentrations of neutralizing agents, such as Ca2+, net acidity may not be
substantially higher than regional values and may in fact be lower at
times.
The second deposition scale is the mesoscale defined as ca. 100km.
The impact of the area source emissions was found to be maximum on that
scale. It appears that, particularly for NOX, a significant percentage
(>50%) is transformed to nitric acid and deposited on the mesoscale. The
percentage for SOX is smaller, but still significant. The nitrate de-
position primarily implicates the significant transportation sources
which are at low elevations and diffuse. Therefore, they may mix
58
-------
thoroughly with other pollutants and oxidants and as they travel downwind
may become incorporated into clouds leading to faster transformation and
scavenging rates. Some evidence of the impact of the urban plume on
downwind wet NH4+ deposition is also present.
C. Preliminary indications, based on limited but reliable field data,
have shown that the impact of primary sul fates on local and mesoscale wet
deposition is minimal.
59
-------
REFERENCES
American Meteorological Society. 1981. METROMEX: A review and summary,
Meteorological monograph (edited by S.A. Changnon, Jr.), 18, 40, AMS,
Boston, MA, 181 pp.
Andersson, T. 1969. Small-scale variations of the contamination of rain
caused by washout from the low layers of the atmosphere, Tellus, XXI, 5,
685-692.
Atkinson, R., A.C. Lloyd, and L. Winges. 1982. A new chemical mechanism
for hydrocarbon/NOx/SOx photo-oxidation suitable for inclusion in atmos-
pheric simulation models, Atmos. Environ. 16, 1341-1355.
Baer, N.S., G. Kirmeyer, and J.E. Yocom. 1984. Critical Assessment
Review Papers, Chapter E-7, Vol. 2, 7.1-7.38, EPA-600/8-83-U16BF.
Barrie, L.A. 1980. The fate of particulate emissions from an isolated
power plant in the oil sands areas of western Canada, in Annals of the
New York Academy of Sciences, (Eds. T.J. Kneip and P.J. Lioy), 33~8~^
434-452.
Barrie, L.A., and A. Sirois. 1982. An analysis and assessment of preci-
pitation chemistry measurements made by CANSAP: 1977-1980, Report AQRB-82
-003-T of the Atmospheric Environment Service, 4905 Dufferin Street,
Downsview, Ontario, Canada
Benkovitz, C.M. 1980. MAP3S/RAINE emissions inventory progress report.
BNL 51378, Brookhav.en National Laboratory, Upton, NY 11973.
Benkovitz, C.M. 1982. Compilation of an inventory of anthropogenic
emissions in the United States and Canada. Atmos. Environ. 16:1551-1556.
Benarie, M. 1976. Transport of pollutants considered from the point of
view of a short and medium-range material balance, Water, Air and Soil
Pollut., 6_, 329-338.
Bielke, S., and H.W. Georgii. 1968. Investigation on the incorporation
of sulphur dioxide into fog- and rain droplets, Tellus, 20, 435-442.
Bridgman, H.A. 1984. Mesoscale spatial variabil ity .of sulfate in air
and rainwater at St. Louis, Water, Air and Soil Pollut., 22, 153-172.
Cadle, S.H. 1984. The composition of dew in an urban area, in
The Meteorology of Acid Deposition (editor, P.J. Samson), Air Pollution
Control Association, Pittsburgh, PA 15230, pp. 17-29.
Calvert, J.G., and W.R. Stockwell. 1983. Acid generation in the tropos-
phere by gas-phase chemistry, Environ. Sci. Technol ., 17, 9, 428A-443A.
60
-------
Carmichael, G.R. and L.K. Peters. 1984. An Eulerian transport/transform-
ation/ removal model for SOg and sul fate - I. Model development, Atmos.
Environ., 18_, 5, 937-951.
Chan, W.H., C.U. Ro, M.A. Lusis, and R.J. Vet. 1982. Impact of the INCO
nickel smelter emissions on precipitation quality in the Sudbury area,
Atmos. Environ., 16_, 4, 801-814.
Chan, W.H., R.J. Vet, C.U. Ro, A.J.S. Tang, and M.A. Lusis. 1984a.
Impact of INCO smelter emissions on wet and dry deposition in the Sudbury
area, Atmos. Environ., 18_, 5, 1001-1008.
Chan, W.H., R.J. Vet, C.U. Ro, A.J.S. Tang, and M.A. Lusis. 1984b.
Long-term precipitation quality and wet deposition fields in the Sudbury
basin, Atmos. Environ., 18, 6, 1175-1188.
Cleveland, W.S., R.J. McRae. 1978. Weekday-weekend ozone concentrations
in the Northeast United States, Environ. Sci. Techno!., 12, 5, 559-563.
Cleveland, W.S., and T.E. Graedel. 1979. Photochemical air pollution in
the Northeast United States, Science, 204, 1273-1278, 1979.
Dana, M.T., J.M. Hales, and M.A. Wolf. 1975. Rain scavenging of S02 and
sulfate from power plant plumes, J . Geophys. Res., 80, 4119-4129.
Dana, M.T., D.R. Drewes, D.W. Glover, and J.M. Hales. 1976. Precipita-
tion scavenging of fossil-fuel effluents, Report EPA-600/4-76-031,
Battelle Pac. Northwest Lab., Richland, Washington.
Dana, M.T. 1980. S02 versus sulfate wet deposition in the Eastern
United States, J. Geophys. Res., 85_, C8, 4475-4480.
Dana, M.T., and A.A.N. Patrinos. 1983. Scavenging measurements near
large point sources: a review and evaluation, in Precipitation Scaveng-
ing, Dry Deposition, and Resuspension, Vol. 1, Precipitation Scavenging
(Editors, H.R. Pruppacher, R.G. Semonin, and W.G.N. Slinn), Elsevier, New
York, pp. 335-347.
Dana, M.T., A.A.M. Patrinos, E.G. Chapman, and J.M. Thorp. 1984.
Wintertime precipitation chemistry in North Georgia, in Meterological
Aspects of Acid Rain, Vol. 1, Acid Precipitation Series (editor C.M.
Bhumralkar), Butterworth Publishers, MA. pp 119-144..
Dana, M.T. 1985. The contribution of sulfur-IV to sulfur wet deposition
in the eastern United States. Submitted to J. Geophys. Res.
Dasch, J.M., and S.H. Cadle. 1984. The effect of local emissions on wet
and dry deposition in southeastern Michigan, Atmos. Environ., 18, 5,
1009-1015.
61
-------
Dasch, J.M., S.H. Cadle, and G.T. Wolff. 1984. Summertime study of acid
deposition in the Detroit-area, Water, Air, and Soil Pollut., 21, 51-69.
Davies, T.D. 1976. Precipitation scavenging of sulphur dioxide in an
industrial area, Atmos. Environ., 10, 879-890.
Davies, T.D. 1979. Dissolved sulpher dioxide and sulphate in urban and
rural precipitation (Norfolk, U.K.) Atmos. Environ., 13_, 1275-1285.
DeWys, J.N., A.C. Hill, and E. Robinson. 1978. Assessment of the fate
of sulfur dioxide from a point source, Atmos. Environ., 12, 633-639.
Draxler, R.R. 1983. Lagrangian meteorology and measurements of acidic
precipitation at Washington, D.C., Atmos. Environ., 17, 12, 2525-2531.
Dugan, Neil. 1984. Personal Communication, Air Management Services
Laboratory, Philadelphia, PA.
Enger, L., and U. Hdgstrttm. 1979. Dispersion and wet deposition of
sulfur from a power plant plume, Atmos. Environ., 13, 797-810.
EPA. 1982. Emission inventories for urban airshed model application in
the Philadelphia AQCR. EPA-45U/4-82-005, Research Triangle Park. NC.
Forrest, J., R. Garber, and L. Newman. 1979. Formation of sulfate,
ammonium and nitrate in an oil-fired power plant plume, Atmos. Environ..
1£, 1287-1297.
Forrest, J., R.W. Garber, and L. Newman. 1981. Conversion rates in
power plant plumes -based on filter pack data: The coal-fired Cumberland
plume, Atmos. Environ., 15_, 10/11, 2273-2282.
Gatz, D.F. 1980. Associations and mesoscale spatial relationships among
rainwater constituents, J. Geophys. Res., 85 , CIO, 5588-5598.
Gillani, N.V., R.8. Husar, J.D. Husar, and D.E. Patterson. 1978. Pro-
ject MISTT: Kinetics of particulate sulfur formation in a power plant
plume out to 300 km, Atmos. Environ., 12, 589-598.
Gillani, N.V. , S. Kohli, and W.E. Wilson. 1981. Gas-to-particle conver-
sion of sulfur in power plant plumes - I. parameterization of the
conversion rate for dry, moderately polluted ambient .conditions, Atmos.
Environ., 15_, 10/11, 22932313.
Godden, D. and L. Lurmann. 1983. Development of the PLMSTAR model and
its application to ozone episode conditions in the South Coast Air Basin.
ERT Document No. P-A702-20U, ERT, Westlake Village, CA.
Gorham, E., F.B. Martin, and J.T. Litzau. 1984. Acid rain: ionic cor-
relations in the eastern United States, Science, 225, 407-409.
62
-------
Granat, L., and H. Rodhe. 1973. A study of fallout by precipitation
around an oil-fired power plant, Atmos. Environ., 7_, 781-792.
Gregory, G.L. 1976. NASAS Report TN D-8352, NTIS, Springfield, VA 22151.
Hales, J.M. 1978. Wet removal of sulfur compounds from the atmosphere,
Atmos. Environ., 12_, 389-399.
Hales, J.M., and S.L. Sutter. 1973. Solubility of sulfur dioxide in
water at low concentrations, Atmos. Environ., 7_, 997-1001.
Hales, J.M., J.M. Thorp, and M.A. Wolf. 1971. Field investigation of
sulfur dioxide washout from the plume of a large coal-fired power plar^t
by natural precipitation, Finalreport, EPA Contract 22-69-150, Battelle
Pacific Northwest Labs, Richland, Washington.
Hales, J.M., and M.T. Dana. 1979a. Regional scale deposition of sulfur
dioxide by precipitation scavenging, Atmos. Environ., 13, 1121-1132.
Hales, J.M., and M.T. Dana. 1979b. Precipitation scavenging of urban
pollutants by convective storm systems, J . Appl. Meteorol., 18, 3, 294-316.
Hegg, D.A., and P.V. Hobbs. 1978. Oxidation of sulfur dioxide in aqueous
systems with particular reference to the atmosphere, Atmos. Environ., 12,
241-253. ~
Hegg, D.A., P.V. Hobbs, and L.F. Radke. 1984. Measurements of the scav
enging of sulfate and nitrate in clouds, Atmos. Environ., 18, 9, 1939-1946.
Hflgstrbm, U. 1974. Wet fallout of sulfurous pollutants emitted from
a city during rain or snow, Atmos. Environ., 8^ 1291-1303.
Holt, B.D. E. Nielsen, and R. Kumar. 1983. Oxygen-18 estimation of pri-
mary sulfate in total sul fate scavenged by rain from a power plant plume,
in Precipitation Scavenging, Dry Deposition and Resuspension, Vol 1,
Precipitation Scavenging (editors, H.R. Pruppacher, R.G. Semonin, and
W.G.N. SI inn), Elsevier, New York, pp. 357-368.
Hlavay, J. and G.G. Guibault. 1978. Detection of HC1 gas in ambient air
with a coated piezoelectric quartz crystal. Anal. Chem. 50:965-967.
Hutcheson, M.R., P.P. Hall, Jr. 1974. Sul fate washout from a coal fired
power plant plume, Atmos. Environ., 8_, 23-28.
lapalucci, T.L., R.J. Demski and D. Bienstock. 1969. Chlorine in coal
combustion. Bur. Mines Rep. Invest., 7260, Pittsburgh, PA.
Johnson, A.M., and T.G. Siccama. 1983. Acid deposition and forest de-
cline, Environ. Sci. Techno!., 17, 7, 294A-305A.
63
-------
Kaplan, M.L., J.W. Zack, V.C. Wong, andJ.J. Tuccillo. 1982. Initial
results from a mesoscale atmospheric simulation system and comparison
with the AYE-SESAME I data set, Monthly Weather Review, 100, 1564-1590.
Kleinman, L.I. 1984. A regional scale Lagrangian nonlinear chemistry
model for acid deposition. Proceedings of Fourth Joint Conference of Air
Pollution Meteorology, October 16-19, 1984, American Meteorological
Society, Boston, MA.
Larson, T.V. , R.J. Charlson, E.J. Knudson, G.D. Christian, and
H. Harrison. 1975. The influence of a sulfur dioxide point source on
the rain chemistry of a single storm in the Puget Sound region, Water,
Air, and Soil Pollut., 4_, 319-328.
Levine, S.Z., and S.E. Schwartz. 1982. In-cloud and below-cloud scav-
enging of nitric acid vapor, Atmos. Environ., 16, 7, 1725-1734.
Li. T.Y. , and H.E. Landsberg. 1975. Rainwater pH close to a major power
plant, Atmos. Environ., 9_, 81-88.
Likens, G.E., R.E. Wright, J.N. Galloway, and T.J. Butler. 1979. Acid
Rain, Scientific American, 241, 4, 43-51.
Liljestrand, H.M., and J.J. Morgan. 1981. Spatial variations of acid
precipitation in Southern California, Environ. Sci. Techno!., 15, 3,
333-339.
Lipfert, F.W., L.R. Dupuis, and L. Alvarez. 1984. Urban and local
source effects on precipitation chemistry. BNL 34565, Brookhaven
National Laboratory, Upton, NY, 20 pp.
Lusis, M.A., K.G. Anlauf, L.A. Barrie, and H.A. Wiebe. 1978. Plume
chemistry studies at a northern Alberta power plant, Atmos. Environ., 12,
2429-2437.
Lusis, M.A. , W.H. Chan, A.J.S. Tang, and N.D. Johnson. 1983. Scavenging
rates of sulfur and trace metals from a smelter plume, in Precipitation
Scavenging, Dry Deposition, and Resuspension, Vol. 1 Precipitaton Scaven-
ging (editors, H.R. Pruppacher, R.G. Semonin, and W.G.N. Slinn), Elsevier,
New York, pp. 369-382.
Millan, M.M., S.C. Barton, N.D. Johnson, B. Weisman, M. Lusis, W. Chan,
and R. Vet. 1982. Rain scavenging from tall stack plumes: a new experi-
mental approach, Atmos. Environ., 16, 2709-2714.
Milot, C. 1985. Multimedia maneuvers: Shifting tactics for controlling
shifting pollutants. Science News 127:124-126.
64
-------
Munn, R.E., G.E. Likens, B. Weisman, J. Hornbeck, C.W. Martin, F.H. Borman,
G.W. Oehlert, R. Bloxam. 1982. A climatological analysis of the Hubbard
Brook (New Hampshire) precipitation chemistry data, Grant DE-FG01-SOEV-
10455, prepared for U.S. DOE, Washington, DC.
NCAR. 1984. Regional acid deposition: Models and physical processes,
the NCAR Acid Deposition Modeling Project, EPA Report EPA-600/3-84-0541,
Research Triangle Park, NC.
Parungo, P.P., and R.F. Pueschel. 1980. Conversion of nitrogen oxide
gases to nitrate particles in oil refinery plumes, J. Geophys. Res., 85,
C8, 4507-4511.
Parungo, P.P., R.F. Pueschel, and D.L. Wellman. 1980. Chemical charact-
eristics of oil refinery plumes in Los Angeles, Atmos. Environ., 14,
509-522.
Patrinos, A.A.M., M.T. Dana, and R.E. Say! or. 1983. Wetfall chemistry
studies around a large coal-fired power plant in the southeastern United
States, J_._ Geophys. Res., 88_, C13, 8585-8612.
Patrinos, A.A.N. 1983. Mesoscale acid deposition study. BNL 34494,
Brookhaven National Laboratory, Upton NY 11973, 70 pp.
Patrinos, A.A.M., 1985. The influence of urban and industrial emissions
on mesoscale precipitation quality, J. of Air Pollut. Control Assoc., 35,
719-727.
Patrinos, A.A.N. , a-nd R.M. Brown. 1984. Mesoscale wetfall chemistry
around Philadelphia during frontal storms, Geophys. Res. Lett., 11, 5,
561-564.
Patrinos, A.A.N. 1984. Interim quality assurance and quality control
activities during the Philadelphia field study.BNL 35715, Brookhaven
National Laboratory, Upton, NY 11973.16 pp.
Patrinos, A.A.N, and L.I. Kleinman. 1984. Diagnostic modeling of pol-
lutant transformation and wet deposition on the urban scale, Transactions,
American Geophysical Union, Vol. 65, 45, p. 844.
Penkett, S.A., B.M.R. Jones, K.A. Brice, and A.E.J. Eggleton. 1979. The
importance of atmospheric ozone and hydrogen peroxide in oxidising sulphur
dioxide in cloud and rainwater, Atmos. Environ., 13 , 123-137.
Platt, U.V., A.M. Winer, H.W. Biermann, R. Atkinson, and J.N. Pitts, Jr.
1984. Measurement of nitrate radical concentrations in continental air.
E.S. & T. 18:365-369.
65
-------
Pratt, G.C., M. Coscio, D.W. Gardner, B.I. Chevone, and S.V. Krupa. 1983.
An analysis of the chemical properties of rain in Minnesota, Atmos.
Environ., 17_, 2, 347-355.
Raynor, G.S. and J.V. Hayes. 1982. Variation in chemical wet deposition
with meteorological conditions, Atmos. Environ., 16, 1647-1656.
Richards, L. W. 1983. Comments on the oxidation of N02 to nitrate-day
and night, Atmos. Environ., 1_7, 2, 397-402.
Rodhe, H. 1970. On the residence time of anthoropogenic sulfur in the
atmosphere, Tellus, XXII, 1, 137-139.
Rodhe, H., C. Persson, and 0. Akesson. 1972. An investigation into
regional transport of soot and sulfate aerosols, Atmos. Environ., 6,
675-693. ~"
Scheider, W.A., D.S. Jeffries, and P.J. Dillon. 1981. Bulk deposition
in the Sudbury and Muskoka-Haliburton areas of Ontario during the shut-
down of INCO Ltd in Sudbury, Atmos. Environ., 15_, 6, 945-956.
Schiermeir, F.A. 1978. Air monitoring milestones: RAPS' field measure-
ments are in, Environ. Sci. and Technology, 12, 644-651.
Schwartz, S.E. 1984. Gas-aqueous reactions of sulfur and nitrogen
oxides in liquid-water clouds. In S02, NO, and N02 oxidation mechanisms:
atmospheric conditions, J.B. Calvert, ed., Butterworth, Boston, MA, pp.
173-208.
Scott, B.C. 1978. Parameterization of sulfate removal by precipitation,
J. Appl. Meteor., 17, 1375-1389.
Scott, B.C. and N.S. Laulainen. 1978. On the concentration of sulfate in
precipitation. J. Appl. Meteor., 18:138-147.
Semonin, R.G. , and Y.C. Bowersox. 1983. Characterization of the inorgan-
ic chemistry of the precipitation of North America, in Precipitation
Scavenging, Dry Deposition and Resuspension, Vol. 1, Precipitation Scaven-
ging (editors, H.R. Pruppacher, R.G. Semonin, and W.G.N. Slinn), Elsevier,
New York, pp. 191201.
Sexton, K. and H.H. Westberg. 1983. Ambient hydrocarbon and ozone con-
centrations near a refinery. EPA-600/S3-83-02b, Research Triangle Park,
NC.
Shaw, R.W. 1982. Deposition of atmospheric acid from local and distant
sources at a rural site in Nova Scotia, Atmos. Environ., 16, 2, 337-348.
66
-------
Slanina, J., F.G. Romer, and W.A.H. Asman. 1983. Estimation of wet de-
position- from a plume of a powerplant and the corresponding background.
concentration, in Precipitation Scavenging, Dry Deposition, and Resuspen-
sion, Vol. 1, Precipitation Scavenging (editors, H.R. Pruppacher, R.G.
Semonin, and W.G.N. Slinn), Elsevier,
New York, 349-356.
Spaite, P.W. and M.F. Szabo. 1982. Acid rain: commentary on controv-
ersial issues and observations on the role of fuel burning.NTIS
Publication No.DE82 016914. U.S. Dept. of Energy, Morgantown, WV.
Spicer, C.W. 1977. The fate of nitrogen oxides in the atmosphere,
Advances in Environmental Science and Technology (edited by Pitts &
Metcalf), Wiley, New York, 7, 163-261.
Spicer, C.W. 1982. Nitrogen oxide reactions in the urban plume of
Boston, Science, 215, 1095-1097.
Stamm, A.J., P. Cap! an and R.A. Hinrichs. 1984. Rain chemistry under
the plume from an oil-fired power plant - a case study, Atmos. Environ.,
18_, 4, 817-823.
Stevenson, C.M. 1968. An analysis of the chemical composition of
rainwater and air over the British Isles and Eire for the years 1959-1964,
Quart. J. Roy. Met. Soc., 94, 56-70.
Summers, P.W. and B. Hitchon. 1973. Source and budget of sulfate in
precipitation from central Alberta, Canada, J. of Air Pollut. Control
Assoc., 23_, 3, 194-199.
Summers, P.W., and D.M. Whelpdale. 1976. Acid precipitation in Canada,
Water, Air and Soil PoTlut.. 12_, 171-185.
The MAP3S/RAINE Research Community. 1982. The MAP3S/RAINE precipitation
chemistry network: statistical overview for the period 1976-1980. Atmos.
Environ. 16:1603-1631.
Volchok, H.L. , and D.C. Freeswick. 1981. Environmental Measurements
Laboratory Regional Baseline Station, Chester, New Jersey, EML-399,
Environmental Measurements Laboratory, U.S. Dept. of Energy, New York,
NY.
Watt, W.D., D. Scott, and S. Ray. 1979. Acidification and other
chemical changes in Halifax County Lakes after 21 years, Limnol. Oceanogr.,
24_, 1154-1161.
West, P.W., and G.C. Gaeke. 1956. Fixation of sulphur dioxide as
disulphiomercurate (II) and subsequent colorimetric estimation,
Analyt. Chem., 28_, 1816-1819.
Wiebe, H.A., and D.M. Whelpdale. 1977. Precipitation scavenging from a
tall stack plume, Precipitation Scavenging, Rep. CONF-741003, pp. 118-126,
U.S. Dep. Energ., Washington, DC.
67
-------
Wiltshire, J.F. 1979. The long range transport of atmospheric acids; a sulphur
budget for Nova Scotia. Rep. EPS-5-AR-77-8.Environmental Protection Service,
Halifax, Nova Scotia, Canada B3J 1M5, 114 pp.
68
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
\. REPORT NO.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
LOCAL SOURCE IMPACT ON WET DEPOSITION
5. REPORT DATE
6. PERFORMING ORGANIZATION COOE
7. AUTHOR(S)
Aristides A.N. Patrinos
S. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME ANO AOORESS
Atmospheric Sciences Division
Brookhaven National Laboratory
Upton, Long Island, NY 11973
10. PROGRAM ELEMENT NO.
CCVN1A/Q5 -3104 (FY-85)
11. CONTRACT/GRANT NO.
DW89006701
12. SPONSORING AGENCY NAME ANO AOORESS
Atmospheric Sciences Research Laboratory - RTF, NC
Office of Research and Development
U.S. Environmental protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT ANO PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Precipitation chemistry measurements over a network of samplers
upwind and downwind of Philadelphia, PA show that a major contribution of
the local sources can be discerned under certain conditions. For winter
frontal storms with low level winds from the south east, up to as much as
a factor of two increase over upwind values has been observed for downwind
nitrate deposition. Sulfate deposition shows an increase of about a
factor of one and one half. The nitrate deposition increases toward the
downwind direction away from the urban-industrial sources, indicating
that the maximum is likely to have been beyond the sampling network for
these case studies. One storm had no increase in nitrate or sulfate
deposition but did have an increase in total sulfur content in the pre-
cipitation. Reasons for this difference are being sought.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19 SECURITY CLASS (Tins Report/
Unclassified
21 NO. OF PAGES
20. SECURITY CLASS (Tins page/
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
------- |