Unit«d States
Environmental Protection
Agency
Environmental Research
Laboratory
Duluth MN 55804
EPA 600 3 80 071
July 1980
Research and Development
Environmental
Effects of Western
Coal Combustion
Part II
The Water Quality of
Rosebud Creek,
Montana
-------
REPORTING SERIES
Research reports of (he Office ul Research and Development, U S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories wen-' established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface m related fields
The nine series are
1 Environmental Health Effects Research
Environmental Protection Technology
Ecological Research
4 Environmental Monitoring
5 Socioeconomic Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8 "Special" Reports
9 Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials Problems are assessed for their long- and short-term influ-
ences Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms m the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-80-071
July 1980
ENVIRONMENTAL EFFECTS OF WESTERN COAL COMBUSTION:
PART III - THE WATER QUALITY OF ROSEBUD CREEK, MONTANA
by
R. K. Skogerboe, M. M. Miller, D. L. Dick
Department of Chemistry
Colorado State University
Fort Collins, Colorado 80523
and
R. V. Thurston, and R. C. Russo
Fisheries Bioassay Laboratory
Montana State University
Bozeman, Montana 59717
Grant No. R803950
Project Officer
Donald I. Mount
Environmental Research Laboratory
Duluth, Minnesota 55804
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
DULUTH, MINNESOTA 55804
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory-
Duluth, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views
and policies of the U.S. Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or recommendation
for use.
-------
FOREWORD
This study, one of many examining impacts of energy development in the
western U.S., documents pre and post operation contributions of a coal-fired
power plant on a small stream. The role of snowfall and subsequent melt
water as transport mechanisms from air to water is a finding of this study
that merits more attention at other sites.
Norbert Jaworski, Ph.D
Director
Environmental Research Laboratory-Duluth
-------
ABSTRACT
The results of a study on Rosebud Creek, Montana, designed to
assess the impacts on water quality of surface coal mining and/or coal com-
bustion at Colstrip are summarized herein. A general degradation of water
quality has been observed along the stream course but direct impacts of
groundwater from the mine areas have not been demonstrated. Influxes of ar-
senic, mercury, selenium, and polynuclear aromatic compounds during snow-
melt periods have been linked by analyses of snow samples to their accumula-
tion in snowfall via scavenging of the power plant plume and subsequent
delivery to the stream via surface runoff. Although this contaminant trans-
fer route may prove highly significant, it is emphasized thai this will
depend on site specific conditions.
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables viii
Acknowledgments ix
I Introduction 1
II Conclusions 6
III Recommendations 7
IV Site Description 8
Physical Features 8
Hydrology 8
Sampling Sites 10
V Methods and Procedures 14
VI Results and Discussion , 15
Stream Water Quality 15
Emission Levels 31
Atmospheric Transport and Dispersion 37
Deposition and Scavenging from the Plume 42
Stream Impacts 48
Accumulation of Atmospheric Constituents in Snowfall 53
References 54
-------
FIGURES
Number Page
1 Location map of water quality monitoring stations
sampled at least ten times 4
2 Cow Creek sampling sites 5
3 Average monthly flow of Rosebud Creek, 1948, 1953, and 1975. 9
4 Rosebud Creek sampling stations in relation to
wind exposure from Col strip 12
5 Comparisons of trace element concentrations in
Rosebud Creek 18
6 Spatial and temporal changes in specific
conductivity 20
7 Spatial and temporal changes in dissolved solids 21
8 Spatial and temporal changes in alkalinity 22
9 Spatial and temporal changes in hardness 23
10 Spatial and temporal changes in sulfate 24
11 Spatial and temporal changes in chloride 25
12 Spatial and temporal changes in magnesium 26
13 Spatial and temporal changes in sodium 27
14 Spatial and temporal changes in mercury 28
15 Spatial and temporal changes in arsenic 29
16 Comparison of mercury levels with stream flows 33
17 Schematic of transfer pathways between coal
combustion sources and aquatic ecosystems 34
18 Model calculation of mercury 41
19 Measured atmospheric mercury concentrations 43
vi
-------
20 Mercury levels in snow versus downwind distances
from the Col strip plant 46
21 Mercury levels in snow versus distance from the
mean plume centerline 47
22 Concentrations of arsenic and selenium in snow-
fall as functions of distance from the source 49
23 Relative concentrations of PNA compounds found
in Rosebud Creek during and after snowmelt runoff 51
-------
TABLES
Number Page
1 Description of Rosebud Creek water sampling sites
and their cumulative plume exposures 11
2 Summary of mean values and ranges observed at
selected Rosebud Creek stations 16
3 Comparison of calculated and observed metal
concentrations for Rosebud Creek 32
4 Emission test data for Col strip Unit No. 2 35
5 Element concentrations in fly ash emitted at
Col strip compared with concentrations in coal
burned 36
6 Typical dependence of the concentrations of
some trace elements on the aerodynamic particle
diameter of emitted coal fly ash 38
7 Mean concentrations of trace elements in U.S.
coals 39
8 Summary of mercury levels in snow 45
9 Summary of snow analyses for anionic species
which may originate from coal combustion 50
vm
-------
ACKNOWLEDGMENTS
The field sampling along Rosebud Creek was carried out with the coopera-
tion of the following landowners and ranchers: P- Kluver, D. Polich, W. D.
McRae, and J. Bailey.
Research was supported by funds provided to the Natural Resource Ecology
Laboratory, Colorado State University, and Fisheries Bioassay Laboratory,
Montana State University, by the Environmental Research Laboratory-Duluth,
U.S. Environmental Protection Agency, Research Grant No. R803950.
-------
SECTION I
INTRODUCTION
Increased reliance on coal as an energy source is accompanied by poten-
tial environmental impacts due to coal mining and combustion. It now appears
certain that mine-mouth power plants will become increasingly common for
various economic reasons. Since many of these are, and will be, located in
relatively fragile areas of the western states where water is scarce enough
to be treated as a commodity, it is particularly pertinent to assess the
potential impacts on the aquatic system.
The mining and combustion activities may affect the surrounding area in
various ways. It is known that strip mining may alter the hydrologic cycle,
and that backfilling, leveling, and revegetation are useful means of alle-
viating the associated problems (McWhorter et al. 1975; Dollhopf et al.
1977a,b, 1978; Van Hook and Shults 1977). Such studies have shown that drain-
age from western surface coal mines can have deleterious effects on surface
water quality. An evaluation of such reports indicates that the disruptions
caused by mining result in increases in the total dissolved solids and saline
constituents in groundwaters and eventually the receiving surface waters.
Although these investigations have also examined the influx to surface waters
of numerous heavy metals from mine operations, these metals do not appear
particularly significant in the western states. This is due, in part, to the
basic characteristics of ground and surface waters in the western states.
Such waters are typically buffered by carbonate-bicarbonate constituents
leached from mine overburden as well as the undisturbed strata. This buffer-
ing controls heavy metal solubilities through metal carbonate precipitation
(Skogerboe et al. 1979).
The possible impacts of coal combustion can be broadly assigned to two
categories: (i) those associated with the furnace (bottom) and precipitator
ash and the scrubber waters and (ii) those associated with the materials
emitted to the atmosphere. An overview of the problems which may be asso-
ciated with coal combustion has been presented by Van Hook and Shults (1977).
This workshop summary drew the following consensus conclusions.
1. There is a considerable body of knowledge pertaining to trace ele-
ments released by coal combustion. This information needs to be
evaluated in the context of source term characterization, dose-
response, and ecological transport/transformation investigations.
Understandings of the total problem will be significantly expedited
by this integrated approach.
2. Occurrence of detectable ecological or health effects due to atmos-
pheric releases of trace elements is unlikely in the near future.
-------
The elements As, Be, Cd, Cr, F, Hg, Ni, and Pb were excepted from
this generality and studies of the atmospheric transformations of
sulfur and nitrogen oxides were singled out in terms of pressing
importance.
3. There is little reason to expect acute effects from trace elements
on the aquatic ecosystem, but there is a lack of knowledge regarding
the impact(s) of ash, slag, and sludge disposal practices and
associated aquatic discharges. A better understanding of the
mobility of trace elements and the controlling factors was delineated
as essential to the health and ecological effects assessment question.
4. A pressing need for more information regarding the organic contam-
inants produced and released in coal combustion technologies was
emphasized.
Although these points or their emphasis may be subject to challenge,
they represent the consensus opinions derived from intensive discussions by
experts having widely variant interests and training. They serve as a general
means of focusing on pertinent problem areas. The present research program,
in toto, has dealt directly or indirectly with each of the four areas cited
above.
Reports by Natusch (1976), Natusch et at. (1975), Natusch and Taylor
(1980), Davison et al. (1974), and Linton et al. (1976) present a comprehen-
sive overview of the chemical and physical characteristics of fly ash emitted
from_ coal-fired power plants. The findings of that research indicate that
the more volatile trace elements and many organic contaminants are emitted
from a coal-fired power plant stack as gases and subsequently condense or
"sorb" onto the fly ash particles as they cool. Such processes affect the
atmospheric dispersion question, since particle rather than gaseous dispersion
laws exert control (Friedlander 1977). The presence of contaminants on
particle surfaces rather than within the relatively insoluble particle matrix
sharply affects their accessibility to the environment and their solubility
characteristics (Davison et al. 1974; Natusch et al. 1975; Natusch 1976;
Natusch and Taylor 1980; Linton et al. 1976; Van Hook and Shults 1977).
Finally, the coexistence of heavy metals and organics on fly ash particle
surfaces raises a concern not just about the potential carcinogenic, mutagenic,
and/or teratogenic hazards, but also about possible synergistic enhancements
of health and ecological effects resulting from heavy metal/organic inter-
actions (American Chemical Society 1969, National Academy of Sciences 1972,
Federal Power Commission 1974, U.S. Environmental Protection Agency 1976, Vf.n
Hook and Shults 1977). Investigations in this last area will clearly require
extensive efforts involving scientists from several disciplines. Meanwhile,
the present efforts have concentrated on a general evaluation of the transfer
of combustion emissions to surface waters.
Coal-fired power plants deliver both particulate matter and gaseous ma-
terials to the atmosphere. Atmospheric processes such as rainout, washout,
and fallout may deliver these materials directly to surface waters; they may
also be delivered indirectly via being incorporated into precipitation and
surface or groundwater runoff. The determination of the extents of such
-------
transfer processes is a difficult problem complicated by the fact that pre-
cipitator and bottom ash from the power plant must be disposed of. Such
disposal typically involves burial and/or settling pond treatment such that
the disposal site(s) may be accessible to surface and groundwaters and the
settling pond waters may be cycled back to surface impoundments (American
Chemical Society 1969, National Academy of Sciences 1972, Federal Power
Commission 1974, U.S. Environmental Protection Agency 1976). Thus, field
studies should ideally involve sites where the groundwater and surface im-
poundments directly associated with the coal combustion plant are reasonably
isolated from the watershed(s) which may be affected by the atmospheric
transfer processes. Only under such circumstances can the relative impor-
tance of these two general routes of surface water contamination be assessed.
This report summarizes the results of a study focusing on the water
quality of Rosebud Creek, the major surface drainage immediately east of the
power plant at Colstrip, Montana. The study was begun in October of 1975,
about two months before initiation of power generation at this site, and
continued through the spring of 1978. Concurrent with this study, separate
studies were conducted on the distribution of fishes and aquatic macroinverte-
brates in Rosebud Creek (Elser and Schreiber 1978; Baril et at. 1978). The
results of all these studies address short-term effects of the Colstrip power
plant operation and serve as references against which long-term effects may
be assessed.
Although mining and power generation occur at Colstrip, the effects of
mining on water quality have not been detected in eastward flowing creeks
having headwaters in the mine area (Van Voast et al. 1977), e.g., Cow Creek
(Figures 1 and 2). Rosebud Creek receives flow from intermittent Cow Creek,
which contributes surface water to Rosebud Creek during the snowmelt season
and in times of heavy rainfall (Figure 2). The location of Rosebud Creek in
the primary downwind direction from the power plant (Heimbach and Super 1973)
coupled with the reported lack of mining influence led to its selection as a
site for the evaluation of possible impacts due to atmospheric transport of
coal combustion emissions.
-------
Crow Indian
Reservation
Northern Cheyenne
Indian Reservation
N
milM
20
Figure 1. Location map of water quality monitoring
stations sampled at least ten times.
-------
Colstrip
Area of most frequent
sample collection
Kilometers
024
01234
Figure 2. Cow Creek sampling sites.
-------
SECTION II
CONCLUSIONS
1. Although the water quality characteristics of Rosebud Creek, Montana,
degrade between the headwaters and its confluence with the Yellowstone
River, the present study has not found changes directly attributable to
surface coal mining in the region.
2. Influxes of arsenic, mercury, and polynuclear aromatic compounds occur-
ring during snowmelt periods have been linked to scavenging of those
contaminants from the plume of the power plant located at Col strip,
Montana. In general, snow cover tends to accumulate plume constituents,
thereby semi-isolating them from the soil and terrestrial biota so that
both are denied the extent of contaminant accumulation operative during
warmer months. Soluble contaminants accumulating throughout the water-
shed may consequently be delivered to the stream via the surface runoff
associated with snowmelt.
3. Influxes of contaminants delivered intermittently to surface streams via
this general route are likely to accumulate in sediments due to normal
sedimentation processes and the generally alkaline characteristics of
streams in this region which will enhance precipitation of heavy metals.
4. Examination of the Rosebud Creek water quality data for the period of
this study indicates that the average concentrations of constituents are
at levels considered safe for maintenance of aquatic life.
-------
SECTION III
RECOMMENDATIONS
The results of this investigation have provided data indicative of the
water quality of Rosebud Creek for 3-4 months preceding and several months
following the beginning of coal-fired power plant operations in the area of
the watershed. These data should be used as reference points against which
long-term changes can be assessed. Continued monitoring of the stream for
this purpose is recommended.
This investigation has also demonstrated a route of transfer between a
coal combustion power plant and a surface stream. This transfer route has
heretofore not been widely studied but will surely be operative at other
sites. The potential impacts on water quality and the aquatic biota derived
from this means of transfer will be site specific to some extent. It is
recommended that studies be undertaken at other sites to demonstrate the
generality of the atmosphere-snowmelt transfer route and define the nature(s)
and extent(s) of associated impacts.
-------
SECTION IV
SITE DESCRIPTION
PHYSICAL FEATURES
Surface descriptions of the Rosebud Creek study area have been given by
Renick (1929) and Pierce (1936). The region is semi-arid with hot summers,
cold winters, and lasting snow cover. The average annual precipitation is
34.5 cm at Forsyth (north of Col strip and west of the Rosebud Creek-
Yellowstone River confluence) and 47 cm in the headwaters vicinity of Rosebud
Creek (south of Colstrip). Highly dissected plateau-like uplands, terraces,
and badlands are characteristic of the area. Irregular uplands, often with
buttes along their margins, divide the Tongue River and Rosebud Creek water-
sheds. The uplands tend to be sharply separated by steeply descending slopes
which flatten to irregularly dissected slopes that merge into valley bottoms.
Variations in vegetation occur in association with variations in soil
and precipitation. Some pine is dispersed throughout the area. Lower ele-
vation areas support various natural grasses with gramma predominating in the
drier areas and wheat grass predominating in the lowlands where flooding may
occur. The area along the creek supports hay meadows including alfalfa, but
the rest of the area is naturally adaptable only for grazing.
HYDROLOGY
In its 300 km course, Rosebud Creek enhances the value and livelihood of
the farms and ranches through which it flows. It is a meandering stream,
deep and slow moving enough to raise the water tables and provide subirriga-
tion in summer months. It is also used for flood irrigation, particularly
during high runoff months. The flow at its confluence with the Yellowstone
River varies significantly from year to year but all years show a common
pattern of low flow from July through February and then sharp increases with
two distinct maxima (see Figure 3) (Thorn et at. 1938). The first maximum
usually occurs between mid-March and early April due to snowmelt at lower
elevations; the second and higher maximum occurs in June or July due to snow-
melt at higher altitudes (headwaters) and enhanced by seasonal precipitation.
The first maximum will most likely include fallout, rainout, and washout from
the atmosphere that has accumulated in snowfall at the elevations near Col-
strip; its effects on the water quality of Rosebud Creek may be most prominent.
The second maximum is principally associated with melting of snowfall occur-
ring at more remote locations; it may have less effect on water quality.
-------
10-
o>
ro
CJ3
cc
tf.
I i
1975
1948
1953
j i ' ••!
0 N D J F
M A
MONTH
M J J A s
Figure 3. Average monthly flow of Rosebud Creek, 1948, 1953, and
1975. (Thorn at at. 1938)
-------
There are two primary aquifers in the region; they are associated with
the Rosebud and the McKay coal beds and are separated by clay, silt, and
sand (Van Hook and Shults 1977, Van Voast et al. 1977).
SAMPLING SITES
Ten sample collection stations were established along Rosebud Creek
(Figure 1). All stations were sampled at least ten times during the course
of this study; most were sampled monthly for an 18-month period. The col-
lections were initiated in October 1975, approximately three months before
the coal-fired power plant at Col strip became fully operational. As noted
in Figure 1, the collection stations along Rosebud Creek ranged from the
headwaters, about 80 km S,SW of Col strip, to its confluence with the
Yellowstone River; the most intensive sampling centered around the region
nearest Colstrip. A detailed description of the sampling sites and their
locations is given in Table 1. Collection stations were also established
along Cow Creek which intermittently drains the watershed immediately east
of Colstrip; these are shown in Figure 2.
The Rosebud Creek sampling sites are shown in Figure 4 in relation to
the "wind exposure" from Colstrip. The 10-year mean wind data (Heimbach
and Super 1973) were used to construct this figure. Thus, stations RC-2
and RC-3 lie in a quadrant which is downwind of Colstrip less than 3% of the
time (averaged over 10 years). Station RC-4 lies in a quadrant which is
downwind ~3% of the time while RC-6, RC-9, and CC-8 lie in a quadrant which
is downwind 11% of the time (see the differential plume exposure data in
Figure 4). A cumulative plume (wind) exposure profile is also shown in
Figure 4. The determination of this profile was based on linear addition
of the differential exposures for each quadrant and was used to estimate the
approximate exposure effect integrated over time. These data suggest that
the wind from Colstrip is in the direction of the Rosebud watershed more than
70% of the time. This clearly indicates that any atmospheric emissions from
the power plant are most likely to impact on the Rosebud watershed. It may
also be noted that the mean wind directions are such that the areas of
highest wind plume exposure are those immediately to the east of Colstrip
where Rosebud Creek is nearest the power plant. A further rationale for
considering the cumulative exposure data as presented in Figure 4 and Table
1 may be based on the fact that the stream flows north from a region of low
plume exposure to regions of successively higher exposure. Thus, if a
pollutant were delivered to the creek at RC-4, for example, and if that pol-
lutant is capable of being transported in the natural water system, it would
be carried northward into the region of higher exposure so that a "cumulative"
effect would be possible in the downstream direction. The sites selected
for the present study considered the above exposure factors.
The following points concerning the sample stations should be noted.
1. The RC-1 and RC-2 sites are basically reference point stations
which lie upstream from all mining activity and outside the regions
of most significant wind (plume) exposure. Water quality differ-
ences between these two stations probably reflect agricultural and
ranching impacts.
10
-------
Table 1. Description of Rosebud Creek water sampling sites and their cumulative
plume exposures.
Station (
*RC-1
RC-2
RC-3
RC-4
RC-6
RC-9
**CC-8
RC-10
RC-13
lamp ling
Itarted
10/75
10/75
10/75
10/75
10/75
7/76
3/76
3/76
10/75
Approximate
Site Description Distance from
Colstrip, km
Collected from a road culvert -^2 km 80 S,SW
south of Kirby township
Collected under a bridge ^2 km west 50 S,SW
of Busby township
Collected behind the Bailey house 30 S
(Box 42), along Rosebud Creek
Collected at the McRae Ranch (Box 15 S
28) ^100 m upstream from stables
Collected at the Kluver Ranch (Box 15 S,SE
24) along Rosebud Creek ^300 m
above RC-9
Collected at the Kluver Ranch ~20 m 15 E.SE
upstream from the confluence of
Cow Creek with Rosebud Creek
Collected at the Kluver Ranch ~200 m 13 E,SE
upstream from the Rosebud-Cow
Creek confluence
Collected at the Kluver Ranch -vLO m 15 E
downstream from the confluence of
Cow Creek with Rosebud Creek
Collected at the Kluver Ranch vL km 19 E
•(•Cumulative
Plume
Exposure
<3%
<3%
<3%
3%
14%
14%
14%
25%
25%
below the confluence of Cow Creek
with Rosebud Creek
RC-14 10/75 Collected at the Polich Ranch (Box
14) along Rosebud Creek
RC-15 10/75 Collected ^400 m south of the con-
fluence of Rosebud Creek with the
Yellowstone River
24 NE
40 N,NE
55%
73%
* Rosebud Creek.
** Cow Creek (intermittent stream).
t It is assumed that the plume exposure will be cumulative since the stream flow
is northerly and the water from each site, as it moves north, will reflect the
plume exposure from each quadrant of the wind rose through which it has passed.
11
-------
Differential
Plume Exposure
Cumulative
ume Exposure
16
N
Figure 4. Rosebud Creek sampling stations in relation to wind
exposure from Col strip (Heimbach and Super 1973).
(Station 1 located approximately 30 km south of
Busby)
12
-------
2. The Sites RC-3 and RC-4 are also not reflective of mining and/or
plume exposure but generally indicative of the effects of agri-
cultural activities which tend to increase in the downstream
(northerly) direction. Each successive station should reflect in-
creasing changes attributable to irrigation return flow, primarily
during the June-October irrigation season.
3. Comparison of RC-6 and RC-9 with RC-13 et seq. should reflect
changes which occur due to inputs from the Cow Creek drainage as
well as impacts associated with the cumulative effects of plume
exposure.
13
-------
SECTION V
METHODS AND PROCEDURES
Water samples were collected at each sampling station on a monthly basis
between October 1975 and April 1977. Samples were collected in acid-washed
polyethylene or polypropylene bottles. Temperature was measured in the
field. Dissolved oxygen was measured either in the field with a YSI Model
54-RC meter or by fixing samples in the field according to the azide modifi-
cation of the idometric method (APHA et at. 1971) and subsequently titrating
samples at Fisheries Bioassay Laboratory (FBL) with phenylarsine oxide sub-
stituted for sodium thiosulfate. Alkalinity, hardness, and conductivity were
determined according to analytical procedures described in APHA et al. (1971,
1976) on samples that were chilled immediately upon collection and trans-
ported to FBL for analysis within 24 hours. All other analyses were per-
formed on samples filtered at FBL through a 0.45 ym glass filter and preserved
according to recommended procedures described by the U. S. Environmental
Protection Agency (1974) and shipped to Colorado State University for subse-
quent analysis according to standard procedures (U. s. EPA 1974). The pre-
cision of the metal analyses was typically ±10% of the amount present. The
quality assurance check program of the analytical laboratory indicated
analytical accuracies that were typically within the ±10% range.
The data obtained were keypunched and placed in permanent tape file in
the Colorado State University computer system so that they could be summarized
and evaluated via any of several available procedures.
14
-------
SECTION VI
RESULTS AND DISCUSSION
STREAM WATER QUALITY
A synopsis of the general water quality characteristics at selected
sites in Rosebud Creek is presented in Table 2; these show average changes
which occur between the headwaters (RC-1) and the confluence with the
Yellowstone River (RC-15). A complete tabulation of the water quality
characteristics of Rosebud Creek has been presented for reference purposes
in a report by Skogerboe et al. (1980). The general chemistry evidenced
in Table 2 is very similar to that of numerous streams in southeastern
Montana and the Rocky Mountain states in general. The water is quite
basic, carbonate buffered, and relatively high in sulfate due to the
presence of appreciable amounts of gypsum and epsomite in the region. In-
creases in these characteristics, and others evident in Table 2, may be
largely attributed to groundwater and irrigation return flow influxes. The
higher concentrations of several constituents which may be noted for the Cow
Creek station (CC-8) reflect the fact that this stream is essentially stag-
nant for several months each year. As such, it reflects the rather poor
quality of the ground water in shallow aquifers in the region.
A summary of data for several elements is presented for 'Convenience in
Figure 5. In this figure, data for stations RC-2 and RC-3 have been pooled
to obtain upstream references; data for RC-4, RC-6, and RC-9 have been
pooled to represent stream conditions above the Cow Creek confluence; pooled
data for RC-10 and RC-13 reflect changes just downstream of Cow Creek; and
the pooled data for RC-14 and RC-15 indicate concentration levels as Rosebud
Creek approaches its confluence with the Yellowstone River. The lower end
of the bar in each case (Figure 5) indicates either the detection limit of
the analysis method or the lowest concentration measured for the element in
question; the upper end of the bar indicates the maximum concentration mea-
sured at each site; the percentage given above each bar indicates the frac-
tion of the total measurements for which the element was present above the
detection limit; and the star (-*-) indicates the mean for each case. This
was computed, however, using the detection limit as the actual concentration
for those samples where a particular element could not be detected; since
the concentrations were likely less than the detection limits, the averages
should be taken as upper limit estimates.
Examination of the data summarized in the figure suggests the follow-
ing general conclusions. The average concentrations of the elements are
generally low and within the criteria limits established by EPA for protec-
tion of aquatic life (U.S. Environmental Protection Agency 1977). A pre-
vious report (Skogerboe et al. 1979) has shown that, even though such heavy
15
-------
TABLE 2. SUMMARY OF MEAN VALUES AND RANGES OBSERVED AT SELECTED
ROSEBUD CREEK STATIONS-7
Water quality
parameter
PH
H buffer capacity
Specific conductivity
(u Siemens)
Hardness
(mg CaC03/liter)
Alkalinity
(mg CaC03/ liter)
Chloride
(mg/liter)
Sulfate
(mg/liter)
Total dissolved solids
(mg/liter)
Ca
(mg/liter)
Mg
(mg/liter)
Mean
RC-1
8.4
(8.0-8.8)
0.71
(0.41-1.38)
857
(647-979)
445
(345-484)
377
(270-425)
2.9
(1.0-7.0)
145
(90-260)
535
(252-780)
88
(58-110)
142
(54-930)
values and ranges observed at
RC-6
8.5
(8.0-8.8)
0.71
(0.42-1.49)
1244
(967-1450)
556
(400-659)
407
(326-488)
4.8
(2.5-8.0)
316
(175-375)
808
(551-1000)
84
(45-140)
170
(50-1200)
RC-13
8.4
(8.0-8.7)
0.75
(0.43-1.51)
1266
(856-1496)
560
(421-659)
409
(244-480)
5.0
(2.8-8.0)
332
(250-405)
836
(610-980)
86
(56-140)
180
(69-1200)
station numbers—
RC-1 5
8.4
(8.1-8.8)
0.74
(0.39-1.36)
1352
(558-1798)
542
(215-687)
418
(313-492)
7.3
(2.1-28)
388
(160-605)
926
(431-1367)
81
(45-150)
150
(30-900)
CC-8
8.0
(7.8-8.2)
1.66
(1.17-2.07)
5870
(5410-6510)
2790
(2530-3060)
470
(451-528)
25
(22-29)
2950
(710-3880)
5220
(4220-6220)
199
(120-350)
611
(350-810)
-------
TABLE 2. CONTINUED
Mean values and ranges observed at station numbers-
Water quality
parameter
Total Fe
(mg/liter)
Suspended Fe
(mg/liter)
Na
(mg/liter)
K
(mg/liter)
RC-1
0.66
(0.26-1.68)
0.63
(0.23-1.6)
21
(13-25)
5.9
(3.0-9.2)
RC-6
1.2
(0.25-3.8)
1.1
(0.2-3.7)
68
(35-88)
10
(3.3-22)
RC-13
1.2
(0.19-4.4)
1.1
(0.16-4.3)
68
(34-90)
11
(0.9-22)
RC-15
2.2
(0.33-24)
2.1
(0.25-22)
130
(60-740)
11
(4-23)
CC-8
0.44
(0.1-1.3)
0.41
(0.1-1.3)
570
(520-620)
29
(19-40)
— See Figure 1
•2/Average over 18-month monitor period; ranges given for same period.
-------
Zn
Se
Mil
500-
77*
Itfc
rid
. _
so?;
I—* -
i- r—
85V
t — 3
0.5-
•
-
85* 96?, 938
100",
_*_
- * -
- *—
-*-
1
*
200-
-
52K
41%
54;; 64:;
2-3 4-6-9 10-13 14-15
Hq
2-3 4-6-9 10-13 14-15
Fe (SuspendprQ
1C-
yj/j
—
-.-
96^
-*-
89'
-* —
7000 '
6000-
4000-
2000-
' lU'JJt IUU!t
-*-
.,_
1 OOi
3 '
100
-*-
2-3 4-6-9 10-13 14-15
Fe (Dissolved^
1 000 •
500-
'
-
742
44~ ri
y U
64
~*~
%.
74V.
_* _
2-3 4-6-9 10-13 14-15
Cti
100
50-
2-3 4-6-9 10-13 14-15
10-
2-3 4-6-9 10-13 14-15
95S
77'i
74*
5-
JQO^
ioos
IOC
2-3 4-6-9 10-13 14-15 2-3 4-6-9 10-13 14-15
CONCENTRATION, yg/1
Figure 5. Comparisons of trace element concentrations in Rosebud Creek.
[Lower limits of bars indicate detection limits; upper limits
of bars indicate maximum observed concentrations; -*- indicates
average concentration observed; and percentage numbers indicate
frequency of occurrence above detection limit. Station numbers
are given below respective bars (see text).]
18
-------
metals may be available to the surface waters, the basic characteristics of
such waters are likely to limit their solubilities through carbonate and/or
hydroxide precipitation processes. Comparison of the averages for each
element between the headwaters and downstream stations also suggests that
prominent influxes of soluble heavy metals are absent. The solubility
limitations mentioned above surely support this observation. Finally, pos-
sible influxes of dissolved Fe, Hg, and Zn in the region of the Cow Creek
confluence are suggested by the upper limit excursions of the bar graphs;
this is discussed below. It should also be noted that data are not pre-
sented for Al, Cd, Cr, Pb or Ni because the majority of samples had concen-
trations of these elements below the detection limits, i.e., generally less
than 10 ug/liter.
Further indications of the general water quality features of Rosebud
Creek and their implications are presented in three-dimensional graphs
(Figures 6-15). The results have been temporally and spatially blocked on
the basis of the following rationales.
A. Temporal blocks
1. January and February represent months of low stream flow and
low surface and subsurface runoff.
2. March and April reflect stream flow increases generally
caused by snowmelt at the elevations near Colstrip.
3. May and June are blocked to average the further stream flow
increases associated with snowmelt at elevations nearer the
headwaters and the increased spring precipitation.
4. July through September represent the average low flow summer
months during which irrigation return flows may influence
water quality.
5. October through December cover low stream flow periods during
which precipitation, irrigation return flows, and ground
water inflows should be generally minimal.
B. Spatial blocks
1. Sampling sites RC-1 through -3 serve as reasonable water
quality references because they are located upstream nearest
the headwaters above the primary regions of possible irri-
gation return flow impact, and where the Colstrip plume
impact should be less than 3% of the total.
2. Site RC-4 is also in a region where possible plume impact
should be small (-3%), but farming, ranching, and population
(Busby, Montana) effects on water quality may be significant.
19
-------
-1000
73*
t\vje
Figure 6. Spatial and temporal changes in specific conductivity.
[Percentages indicate cumulative wind exposure (see
Figure 4)].
20
-------
1000
r 1000
- 800
- 600
t^e
Figure 7. Spatial and temporal changes in dissolved solids.
[Percentages indicate cumulative wind exposure
(see Figure 4)].
21
-------
_ 500
r500
Figure 8. Spatial and temporal changes in alkalinity.
[Percentages indicate cumulative wind exposure
(see Figure 4).]
22
-------
500
300
Figure 9. Spatial and temporal changes in hardness. [Percentages
indicate cumulative wind exposure (see Figure 4).]
23
-------
300 n
h300
73%
Figure 10. Spatial and temporal changes in sulfate. [Percentages
indicate cumulative wind exposure (see Figure 4).]
24
-------
10
<3% UU^er%*OOS^eS
S^e ?-\u^e **V
Figure 11. Spatial and temporal changes in chloride. [Percentages
indicate cumulative wind exposure (see Figure 4).]
25
-------
lOOn
rlOO
73%
Figure 12. Spatial and temporal changes in magnesium. [Percentages
indicate cumulative wind exposure (see Figure 4).]
26
-------
MOO
Figure 13. Spatial and temporal changes in sodium. [Percentages
indicate cumulative wind exposure (see Figure 4).]
27
-------
Figure 14. Spatial and temporal changes in mercury.
28
-------
•10-,
15
1975
Figure 15. Spatial and temporal changes in arsenic.
29
-------
3. Sites RC-6 through -9 may reflect enhanced plume impacts
(-14%) and serve as a reference for water quality charac-
teristics prior to inputs from the Cow Creek drainage.
4. Sites RC-10 through -13 should show further plume impacts
combined with influx from the Cow Creek drainage.
5. Sites RC-14 and -15 should be indicative of the degree of
transport of contaminants from the stream segment closest
to Col strip combined with further plume impacts from north
of there on to the Yellowstone River.
The water quality parameter values presented in the following graphs
(see Figure 6 et seq.) are therefore averages for the above temporal and
spatial groupings. As such, they provide average indications of the tem-
poral and spatial changes which occurred over the study period. It must be
emphasized, however, that samples collected on a monthly basis can only be
rough approximations of the averages and that interpretations based on them
are also limited by the general validity of the monthly collections. Note
that the temporal blocks are arranged progressing from the high to the low
stream flow months.
The general water quality characteristics of Rosebud Creek may be infer-
red from Figures 6 through 13. The specific conductivity and dissolved
solids (Figures 6 and 7) tended to increase in the downstream direction
reaching highest values during the low streamflow months. The alkalinity
and total hardness values (Figures 8 and 9) show the same general trends and
tend to be reasonably constant (factors of 2 or better) due largely to the
geology of the region in which carbonate and gypsum are quite ubiquitous.
The pH (8.0-8.8) remained consistently in the range characteristic of car-
bonate-buffered natural water systems. The general influence of gypsum, and
other sulfate minerals such as epsomite, may be inferred from the sulfate
data (Figure 10) and parallel changes in sulfate-associated cations such as
magnesium (Figure 12) and sodium (Figure 13). Indeed, sulfate was the princi-
pal anion (other than carbonate and bicarbonate) while chloride (Figure 11)
was a minor anion and phosphate, nitrate, and nitrite were rarely observed
and then only at levels approximating 0.1 mg/liter (Skogerboe et al. 1980).
V
Examination of these summary figures for these general water quality
parameters does not indicate readily identified and/or definitive effects
on water quality that can be directly associated with the mining or power
plant operations at Colstrip. The changes observed can be rationalized on
the basis of other probable causes including the increases in irrigation and
population along the stream course.
As indicated above and in other reports (Mesmer and Baes 1975, Van Hook
and Shults 1977, Van Voast et al. 1977, Skogerboe et al. 1979) the alkaline
and carbonate buffer characteristics of western surface and groundwaters are
likely to exert prominent influences on the heavy metal concentrations in
streams such as Rosebud Creek. In essence, equilibrium calculations indicate
that many heavy metals, even if available, will typically be precipitated as
carbonate or hydroxide species under such conditions. Such computations
30
-------
would, of course, be rendered invalid if certain chemical entities capable of
forming soluble complexes with the heavy metals were present at significant
concentrations in the stream or the influent groimdwaier. The characteristi-
cally low levels of chloride found (Figure 11) and the fact that checks on
the total organic carbon produced values of 1-2 mg/liter or less tend to min-
imize the possibility of soluble entities being formed with chloride or or-
ganic ligands. Furthermore, evaluation of the heavy metal concentrations
observed provides reason for accepting the model calculations as generally
valid. The comparisons presented in Table 3 indicate that the maximum and
the mean soluble concentrations observed at all stations were generally equiv-
alent to or less than those predicted by calculation. This indicates, in a
circumstantial context, that either these metals were not being delivered to
the stream or that their concentrations were controlled by precipitation
equilibria. If the latter is the case, the precipitate species indicated in
Table 3 are those most likely to exert primary control.
The results for mercury presented in Figure 14 indicate the occurrence
of sharp excursions in the dissolved concentrations observed. A similar
profile was observed for arsenic (Figure 15). Although these increases
were observed in varying extents along the entire stream course, they
were most prominent in the region directly east of Colstrip (sites RC-6
through -15) and were apparent in a less pronounced way for selenium.
Examination of precipitation and temperature records, as well as the stream
flow data, indicated that the influx of mercury and arsenic coincided
temporally with snowmelt and surface runoff periods. This is illustrated
for mercury in Figure 16 where it may be noted that the primary influx was
associated with the March 1976 snowmelt at the low elevations but not with
that at high (headwaters) elevations (June 1976). These observations
suggested a mechanistic possibility for impacts of the power plant on sur-
face waters, i.e., accumulation of contaminants emitted in and on the
relatively clean and nonreactive snowfall followed by delivery to surface
waters during runoff (snowmelt) periods.
The routes via which this type of transfer may occur have been discussed
(Van Hook and Shults 1977); these are shown in Figure 17. The focus of the
present discussion is directed toward those routes involving the atmosphere
as a transfer medium.
EMISSION LEVELS
The 750 megawatt plant at Colstrip began test operation in late 1975 and
achieved essentially full-scale operations in early 1976. Emission test data
for Unit No. 2 (Montana Power Co. 1976a, b) are summarized in Table 4; the
values should be multiplied by two to estimate the emissions for both units.
These indicate operation within state and federal emission standards. Cal-
culations based on coal consumption and emission rates indicate that nomi-
nally 5-15% of the sulfur in the coal may be emitted as S02. Limited data on
the chemical composition of the fly ash (Montana Power Co. 1976b, Crecelius
et al. 1978) are given in comparison to the concentrations of the same ele-
ments in the coal (Gluskoter et al. 1977) in Table 5. These data verify
that the fly ash is primarily composed of "glassy" and relatively insoluble
matrix constituents as previously shown (Davison et al. 1974; Natusch et al.
31
-------
TABLE 3. COMPARISON OF CALCULATED AND OBSERVED METAL
CONCENTRATIONS FOR ROSEBUD CREEK
Concentrations (yg/1)
Metal
(oxidation
state)
Al(III)
Cd(II)
Cu(II)
Fe(II)
Fe(III)
Hg(H)
Mn(II)
Pb(II)
Zn(II)
Calculated^/
3
0.3
350
140
100
36,000
2,500
1
0.1
Mean value-'
<100
<5
5
<100^-/
<100-7
0.05
10
<10
6
Maximum ,
observed—
<100
<5
10
190^
190^
20
22
20
10
Chemical
limiting
the .,
solubility-7
A1(OH)3
CdC03
Cu(OH)2
FeC03
Fe(OH)3
Hg(OH)2
MnC03
PbC03
ZnC03
-/Calculated for: pH = 7.8, alkalinity = 250 mg/s. as CaC03, sulfate =
150 mg/ji, chloride = 5 mg/a, and phosphate = 0.1 mg/i.
-Average of all values measured over study period.
c/
-Maximum concentration observed at any sampling site over the 18-month
study period.
-Precipitate which limits solubility as predicted by the equilibrium
calculations for the above average stream conditions.
e/
-Analysis technique used does not differentiate between Fe(II) and Fe(III).
32
-------
o
UJ
co
Di
3.
o
a:
18
14
10
— Station RC-9
Station RC-4
10 12 2 4 6 8 10 12 2 4 6
MONTHS, BEGINNING OCTOBER 1975
Figure 16. Comparison of mercury levels with stream flows.
33
-------
THAI rriMDMCTTAM
1
A
i
\
Deposition
1 '
Runoff/Leaching
IR Jtesuspension bU1L _ Flooding " WH
t
Uptake,
Decay,
and
Leaching
Uptake 1 Uptake
ana
Decay BIOTA Decay
^ (AQUATIC ANP TFRRF^TRTfil \ ^
FER
Figure 17. Schematic of transfer pathways between coal
combustion sources and aquatic ecosystems
(Van Hook and Shults 1977).
34
-------
TABLE 4. EMISSION TEST DATA FOR COLSTRIP
UNIT NO. 2 (MONTANA POWER CO. 1976a, b)
Emitted species
Participate matter
S02
N0x (as N02)
Kg/hr
47.9
353.9
355.7
Concentration—
yg/m3
37,500
257,400
278,600
g/106 BTU
15.6
115.4
115.5
— Stack concentrations based on the average of three tests; plant operating
at 325 megawatts (3.08 x 108 BTU/min) gross load and using three scrubbers;
coal use rate was 181 tons/hr; mean percent sulfur * 0.7.
35
-------
TABLE 5. ELEMENT CONCENTRATIONS IN FLY ASH EMITTED
AT COLSTRIP COMPARED WITH CONCENTRATIONS IN COAL BURNED
Element
Al
As
Ba
Br
Co
Cr
Fe
Mn
Se
Emitted
fly ash-^
15,400
N.D.^7
10,900
72
10
178
12,300
410
370
Mean concentration
(yg/g)
& Coal^-/
17,800
5.7
390
17
5.2
19
0.81
41
3.3
-/Fly ash samples collected April 19, 20, and 21, 1976. Averages of three
samples reported. Analysis by neutron activation analysis (Crecelius
et al. 1978).
— Federal Power Commission (1974), U.S. Environmental Protection Agency (1974)
^/Crecelius et al. (1978).
Elements not detected in<
tions less than approximately 1 yg/g fly ash.
-Elements not detected indicated by N.D. including Cd, Hg, and V; concentra-
36
-------
1975, Natusch and Taylor 1980; Linton et al. 1976; Natusch 1976). It is signi-
ficant to note that the more volatile elements, e.g., As, Cd, and Hg, were not
detected. This is consistent with the typical size dependence results given
in Table 6 (Davison et al. 1974) and indirectly supports the tendency of vola-
tile elements to condense onto fly ash surfaces as the plume cools. The mean
aerodynamic diameters of the particles emitted are less than 4-5 ym (usually
around 1 ym) (Montana Power Co. 1976a,b); these are highly enriched in volatile
elements (see Table 6). These smaller particles are also those which enter in-
to long range transport (Friedlander 1977). In view of the fairly universal
presence of significant concentrations of such elements in U.S. coals (Table 7)
and their tendencies to be emitted and collect on aerodynamically small parti-
cles, the above observations are applicable on a widespread basis.
ATMOSPHERIC TRANSPORT AND DISPERSION
Current transport modeling techniques rely primarily on a Gaussian plume
formulation which in turn relies on an a priori specification of the local
wind velocity and turbulence levels (Turner 1974). The application of such
models to transport calculations must take the following caveats into account
(Hales 1973). First, the dispersion parameters used in the model were deter-
mined for open, flat terrain. Such parameters do not take into account the
effects of rough terrain on dispersion; thus the calculations must be regarded
as approximations only. Second, one cannot assume, in general, that power
plant site measurements of wind speed, direction, and atmospheric stability
are going to persist on a spatial or temporal basis beyond a few kilometers.
As a result, dispersion calculations based on such data must be considered
approximate for transport modeling over longer ranges.
In the present program, dispersion calculations based on the Gaussian
model have been carried out for approximation purposes. The example results
shown in Figure 18 are for emission and transport of mercury vapor to ground
level along the plume centerline. In view of the stack temperatures, mer-
cury is emitted in the vapor state. One report (Gluskoter et al. 1977) has
shown that more than 90% of the mercury in coal was lost by vaporization when
the coal was ashed at ~100°C in a low temperature ashing unit. Its vapor
pressure is such that it is also transported in the vapor state at typical
ambient temperatures. Mercury levels in the coal burned are nominally 0.05-
0.1 yg/g (Gluskoter et al. 1977). Based on the discharge and coal use
rates characteristic of the Colstrip plants and the assumption of 85% emis-
sion for Hg, concentrations of 5-10 yg Hg/m3 are estimated for stack exit
emission levels. The calculations summarized in Figure 18 assumed a conser-
vative level of 1 yg Hg/m3 at the stack exit. The results indicate: (i)
that the impact of the plume at ground level should occur at reasonable dis-
tances downwind of the plant and typically within the Rosebud Creek water-
shed when the wind is from westerly directions, and (ii) increasing wind
velocity should result in ground level impact of the plume closer in to the
plant. In addition, it should be noted that the presence of a temperature
inversion will affect the dispersion and that Gaussian reductions in the
concentrations are expected as a function of distance from the plume center-
line (Turner 1974). Finally, even though these model calculations are for a
gaseous entity, the fact that the particles emitted have aerodynamic dia-
meters approximating 1-2 ym or less (Montana Power Co. 1976a, 1976b)
37
-------
TABLE 6. TYPICAL DEPENDENCE OF THE CONCENTRATIONS OF SOME TRACE
ELEMENTS ON THE AERODYNAMIC PARTICLE DIAMETER OF EMITTED
COAL FLY ASH (DAVISON ET AL. 1974)
Concentration (yg/g)
Aerodynamic
diameter
(ym)
>11.3
11.3-7.3
7.3-4.7
4.7-3.3
3.3-2.1
2.1-1.1
As
680
800
1000
900
1200
1700
Cd
13
15
18
22
26
35
Cr
740
290
460
470
1500
3300
Mn
150
210
230
200
240
470
Ni
460
400
440
540
900
1600
Pb
1100
1200
1500
1550
1500
1600
Sb
17
27
34
34
37
53
Se
13
11
16
16
19
59
Tl
29
40
62
67
65
76
V
150
240
420
230
310
480
Zn
8100
9000
6600
3800
15000
13000
38
-------
TABLE 7. MEAN CONCENTRATIONS OF TRACE ELEMENTS IN U.S. COALS
(6LUSKOTER ET AD. 1977)
Mean concentration (yg/g
Element
Ag
As
B
Ba
Be
Br
Cd
Co
Cr
Cs
Cu
F
Hg
I
Mn
Mo
Ni
P
Pb
Sb
Se
Illinois-/
0.03
14
110
100
1.7
13
2.2
7.3
18
1.4
14
67
0.2
1.7
53
8.1
21
64
32
1.3
2.2
Appal achia-
0.02
25
42
200
1.3
12
0.24
9.8
20
2.0
18
89
0.2
1.7
18
4.6
15
150
5.9
1.6
4.0
for indicated region)
Western U.S.-/
0.03
2.3
56
500
0.46
4.7
0.18
1.8
9.0
0.42
10
62
0.09
0.52
49
2.1
5.0
130
3.4
0.58
1.4
Colstrip^-/
0.06
5.7
—
390
—
17
—
5.2
19
1.4
—
—
—
—
41
—
16
—
—
37
3.3
39
-------
TABLE 7. CONTINUED
Element
Sn
Th
Tl
U
V
Zn
Mean
Illinois-/
3.8
2.1
0.66
1.5
32
250
concentration (yg/g
Appalachia-/
2.0
4.5
—
1.5
38
25
for indicated region)
Western U.S.-/
1.9
2.3
--
1.2
14
7.0
Colstrip^-/
—
3.4
—
—
33
—
^/114 samples
-/23 samples
-/28 samples
^/Crecelius et aZ.(1978)
40
-------
1 - Wind at 2 m/sec
2 - Wind at 4 m/sec
40 -
20 -
5 10
DISTANCE FROM SOURCE, km
PREDICTED MERCURY CONCENTRATION, ng/nf
Figure 18. Model calculation of mercury. (Dispersion along
the plume center!ine at ground level.)
41
-------
indicates that their settling velocities are quite low (Friedlander 1977).
Consequently, these calculations also serve as approximations of the fallout
profiles for particulates. The concentration scale in these instances must
be considered only in qualitative terms, i.e., indicative of the fallout
trend(s).
A two-day atmospheric sampling program was run at Colstrip in October
1976 to determine the levels of mercury along the plume direction, to eval-
uate the ratio of gaseous-to-particulate mercury characteristic of the plume,
and to qualitatively check the model calculations. A 0.45 ym membrane
filter was used to remove the particulates. This was backed by a graphite
filter coated with a thin film of gold to collect gaseous Hg via formation
of the Au:Hg amalgam. Two such sampling trains were run in parallel at each
of several sites located at various distances downwind of the plant. A
control site was also run upwind (~2 km). Winds were relatively constant
in terms of direction and velocity on both sampling days so location of the
samplers near the plume centerline was relatively simple. The atmospheric
stability conditions were stable on the first day but moderately unstable
the second.
The results plotted in Figure 19 summarize the ranges and means observed
for the duplicate samples collected at each site -2-3 km downwind. These
data show high levels of gaseous Hg at the plume centerline which undergo the
expected (Gaussian) dropoff in moving away from the centerline. The values
near the plume centerline are clearly greater than the upwind control levels
of 5.5 ± 2.5 ng/m3. The differences observed between the two days reflect the
increase in the wind velocity on the second day as well as the increased at-
mospheric-instability for that day. Analyses of the membrane filters also
indicated that Hg associated with particulates was not detectable; the general
conclusion based on the data was that more than 95% of the Hg remained in the
gas phase during transport. This is consistent with the vapor pressure char-
acteristic of Hg.
The membrane filters were also analyzed for As and Se but neither of
these elements was detected. Based on the sampling and analysis conditions
used, the atmospheric As and Se levels were generally less than 50 ng/m3. In
view of the small aerodynamic size of the emitted particles, however, spa-
tially defined concentration profiles for particle associated elements com-
parable to that observed for mercury can be anticipated.
DEPOSITION AND SCAVENGING FROM THE PLUME
Scavenging of particles and gases by precipitation has been measured in
the neighborhood of coal-fired power plants; the best experimental results
relate to total mass scavenged (Friedlander 1977). Rainfall scavenging of
trace organics when the solubility coefficients are known has been theoreti-
cally modeled (Hales 1973). In effect, two types of wet deposition may be
operative: (i) rainout refers to the inclusion of the constituents in the
precipitation during its formation (nucleation) in the clouds; (ii) washout
deals with the removal of the contaminants below cloud base during the fall
of the precipitation. Junge (1963) and Gatz (1976) have evaluated the washout
ratio, i.e., the ratio of the concentration of the pollutant 1n the
42
-------
70
o
o
10
O- WIND AT ^2M/SEC; DAY 1
• - WIND AT MM/SEC; DAY 2
JL.
•1.0 -0.5
.< South
MEAN AND
RANGE FOR
UPWIND
CONTROLS
I
0 0.5 1.0
> < North —
APPROXIMATE DISTANCE FROM PLUME CENTERLIME, km
Figure 19.
Measured atmospheric mercury concentrations.
(Error bars indicate ranges for replicate
samples.)
43
-------
precipitation to its concentration in air. These ratios appear to be fairly
constant for trace contaminants, and approximate 3 x 105. Values this high
indicate that precipitation can effectively remove contaminants from the
atmosphere and that the process can "preconcentrate" the constituents in the
precipitation.
Dry deposition of particles to ideal smooth surfaces as influenced by
particle size and degree of atmospheric turbulence can be reasonably predicted
but deviations from ideal conditions cause the predictions to be less relia-
ble. Modeling of deposition of small particles to trees, grass, lakes, etc.
is largely based on guesswork (Friedlander 1977). The matter is further com-
plicated by the effects of rough terrain on meteorological conditions and at-
mospheric stability. As a result, a dry deposition velocity of 1 cm/sec for
all particles being delivered to real surfaces is often applied to calculate
the deposition flux (Friedlander 1977).
In spite of the uncertainties in models for wet and dry deposition, the
fact remains that constituents in plumes are deposited on the terrestrial
surface and may ultimately reach ground and surface waters. To estimate the
extent of such possible impacts, snow samples were collected in the watershed
areas lying in the mean winter wind direction from Colstrip and draining into
Rosebud Creek. Primary rationales for the choice of snow samples have been:
(i) snow serves as a relatively clean substrate for the accumulation of pol-
lutants; (ii) snow cover reduces the fugitive dust and soil contamination
problems so that the plume fallout can be more readily differentiated from
other contamination sources; (iii) the snow cover serves as an integrator of
the fallout flux over the time period(s) between successive snowfalls; (iv)
the temperature regimes involved may operate to preserve the identities of
the trace contaminants through the reduction of chemical transformations; and
(v) some of the trace contaminants may be included in the snowflakes during
the nucleation-growth and/or fallout processes. The latter may involve
particulates serving as nuclei for snow formation, condensation of gaseous
constituents with the nuclei at the low temperatures involved (<-20°C), or
adsorption onto the formed nuclei.
Snow samples collected in 1976, 1977, and 1978 were analyzed for mer-
cury. The results summarized in Table 8 are generally indicative of the
plume impacts which may be inferred from the profiles shown in Figures 18
and 19 and also suggest the integrative deposition of mercury over
time. The 1978 snowfall collections were also analyzed for other consti-
tuents (Ca, Fe, Mn, Zn) considered likely to be indicative of the plume im-
pact. The concentration patterns observed were not as pronounced as those
for mercury, probably due to the occurrence of significant amounts of those
elements in the soils and humic materials of the region.
Snow collections carried out during December 1977 and January 1978 were
analyzed for water soluble As, Cl~, F", Hg, HOl, Se, and SO^". The results
for mercury are summarized in Figures 20 and 21 using the mean easterly wind
direction for December and January to estimate the mean centerline of the
plume impact. The results are expressed in terms of yg Hg per square meter
of surface for a snow depth of 1 cm to provide an approximation of the mer-
cury flux integrated over the time that the snow surface was exposed. Figure
44
-------
TABLE 8. SUMMARY OF MERCURY LEVELS IN SNOW
Month and year
Sample
Collection
Locale
(number of samples)
yg Hg/liter
snowmelt
(mean value, range
in parentheses)
February 1976
March 1976
January 1977
March 1978
Upwind controls (4)
Downwind 4-6 km (4)
Downwind 8-10 km (4)
Upwind controls (4)
Downwind 4-6 km (6)
Remote controls (5)
Downwind 2-4 km (3)
Downwind 4-6 km (3)
Downwind 6-8 km (3)
Downwind 8-12 km (4)
Remote controls (5)
Downwind 4-7 km (8)
Downwind 10-12 km (5)
<0.02
0.67 (0.38-0.78)
0.16 (0.08-0.28)
0.05 (<0.02-0.1)
5.6 (4.3-6.8)
0.05 (0.04-0.07)
0.22 (0.17-0.28)
0.55 (0.44-0.67)
0.15 (0.11-0.18)
0.06 (0.04-0.08)
0.07 (0.04-0.09)
0.12 (0.09-0.19)
0.06 (0.04-0.08)
45
-------
0.8
0.6
i
sn
0.4
o
a:
0.2
5 10
APPROXIMATE DISTANCE FROM SOURCE, km
15
Figure 20. Mercury levels in snow versus downwind distance
from the Colstrip plant. (Each point is the
average of 3-5 samples; vertical bars indicate
the standard deviations.)
46
-------
0.2
0.1
g 0.4
i
CM
•E
CD
0.2
>-
CtL
CJ
o:
O.S
0.4
C. > 10 km
downwind
B.
8-10 km
downwind
A.
5-7 km
downwind
N
-2-101 2
APPROXIMATE DISTANCE FROM MEAN PLUME CENTERLINE, km
Figure 21. Mercury levels in snow versus distance from the mean
plume center!ine.
47
-------
20 indicates the general accumulation of Hg in the snow along the mean plume
direction dropping off with distance from the plant due to dilution and
scavenging of the plume during transport. The plots also demonstrate the
localization of the plume impact close in and dilution of its impact farther
out due to its lateral expansion (Figure 21).
Comparable results were observed for As and Se, as shown in Figure 22.
Finally, results for the anions determined are summarized in Table 9. These
results suggest fairly significant levels of plume impact which also tend to
drop off with distance.
It should be emphasized that the snow samples were allowed to melt at
ambient temperature (20°C) with the melt water being immediately filtered,
as it formed, through 0.45 ym membrane filters to remove particulates before
chemical preservatives were added to maintain the shelf lives of the solu-
tions. Thus, the element concentrations found in the melt solutions are rep-
resentative of only the water soluble fractions of each constituent and, con-
sequently, of those fractions which might be transported to the stream via
surface runoff. This is clearly an important distinction in considering
potential impacts on surface waters.
STREAM IMPACTS
Analyses of water samples collected at the Rosebud Creek sampling sta-
tions have been discussed above. The results summarized in Figures 14 through
16 quite clearly show that increased concentrations of dissolved As and Hg
occur concomitantly with snowmelt periods occurring near Col strip but not with
those occurring in the headwater regions remote from Colstrip. This clearly
implies the importance of accumulation of plume constituents in snowfall and
delivery to streams via surface runoff. The importance of other possible
sources of snow contamination is largely minimized for the following reasons.
1. The snow and stream measurements have relied exclusively on deter-
mining soluble (dissolved) constituents. The As and Hg associated
with soils, coal, and humic matter are generally insoluble.
2. Analyses for organic mercury indicated that more than 95-98% was
present in inorganic forms. Thus, although the historical use of
organic mercury to treat seeds might prove to be a source of mer-
cury during runoff, the relatively long environmental lifetimes of
such compounds and the failure to find them in surface waters sug-
gest that such sources are not primary.
3. Analyses of snow and stream samples using a fluorescence technique
(American Public Health Association Intersociety Committee 1977) to
approximate total concentrations of dissolved polynuclear aromatic
compounds (PNA) indicated their influx into Rosebud Creek in paral-
lel with As and Hg during the March 1978 runoff (Figure 23) and
(circumstantially) their deposition in snow downwind of Colstrip.
Although these concentration estimates approximate total PNA con-
centrations, these observations coupled with the demonstration that
48
-------
E
••- -i- (C
QJ
O)
L.
X) i/l
cot
O-
CO c£
O
(O
UJ
O
oo
>-H *i
t' O
X
O
Oi
Q-
O-
< '
O
if)
6
As or Se CONCENTRATION, ug/i/-cm
Figure 22. Concentrations of arsenic and selenium in snowfall as
functions of distance from the source. (Each point
is the average of 3-5 samples; vertical bars indicate
the standard deviation; arrows downwind indicate less
than concentration values.)
49
-------
TABLE 9. SUMMARY OF SNOW ANALYSES FOR ANIONIC SPECIES WHICH
MAY ORIGINATE FROM COAL COMBUSTION
Mean constituent concentration—
(yg/ml snowmelt water)
Approximate
distance from
source (km)
2.5-5
5-7.5
7.5-10
10-12.5
12.5-15
Controls-/
Cl"
2.6
1.8
1.3
0.69
0.66
0.88
F"
0.42
0.20
0.21
0.20
0.13
0.15
N03"
1.8
1.2
1.1
1.4
1.3
0.55
2-
2.6
1.5
0.90
1.2
0.89
0.62
-Average of three to four samples; individual sample results usually
consistent to ±20-30% of the mean concentration given.
-Collected approximately 80 km west of Col strip near Red Lodge, Montana,
50
-------
10
O—-O'
O - April 1977
• - May 1977
O O
6 8 10
STREAM STATION NUMBER
12 14
16
RF.IATIVE PNA CONC.
Figure 23. Relative concentrations of PNA compounds found in
Rosebud Creek during and after snowmelt runoff.
51
-------
PNA compounds tend to accumulate on the surfaces of emitted fly ash
particles (National Academy of Sciences 1972, Davison et al. 1974,
U.S. Environmental Protection Agency 1976, Van Hook and Shults
1977, Natusch and Taylor 1980) tend to.emphasize the plume as their
primary source.
The general rationale of the plume impact on Rosebud Creek via the snow
.ccumulation-runoff mechanism may also be roughly inferred on the basis of
nass balance calculations. The two coal-fired units at Colstrip consume nom-
inally 360 tons of coal per hour (Montana Power Co. 1976a, 1976b) having a
mean mercury concentration approximating 0.1 yg/g (Gluskoter et al. 1977).
Accepting 85% as a reasonable emission factor for mercury, it may be estimated
that 780 g of mercury are emitted each day giving a total of 23 kg/month.
Since the wind direction is toward Rosebud Creek about 70% of the time
(Heimbach and Super 1973), about 16 kg of mercury may be transported in that
direction each month.
It was noted (Figure 16) that the stream flow for February and March
ranged from about 1-5 m3/sec and may have averaged around 2 m3/sec. The peak
mercury concentration during this time approximated 18 yg/1. Use of this
value and the mean stream flow, and assuming that the snow runoff and the peak
mercury concentration was maintained for one week, it may be estimated that
the total mercury discharge in the stream would be approximately 2.2 kg.
This amounts to roughly 14% of the mercury transported in the direction of
the Rosebud watershed during one month, i.e., 16 kg. If the mercury accumu-
lated in the snowfall over a longer time period, a proportionately smaller
percentage accumulation from the plant emissions would be determined.
Mass balance calculations of this type serve only as crude approxima-
tions. Although it is generally agreed that gaseous entities such as mercury
and small particles are typically transported over distances approximating 50
kilometers or more (Friedlander 1977), it should be noted that the Rosebud
Creek watershed extends 50 to 100 km to the east of Colstrip depending on the
projection direction. Thus, atmospheric contaminants transported over dis-
tances in excess of 40-50 km may still impact on the stream during snowmelt
runoff. In essence, these mass balance approximations suggest that the
mechanism of impact presented above is reasonable.
A complete assessment of potential impacts of such plume constituents on
surface waters via the snow collection-runoff route must take into account
site-specific conditions which will affect the deposition and accumulation
processes and the runoff transfer process. The results discussed above apply
specifically to the Colstrip-Rosebud Creek site and the meteorological and
watershed parameters for the time periods embraced by this study. Surely the
effects observed and the magnitudes thereof will vary considerably from year
to year. Consideration of factors which may affect the accumulation processes
is germane to the present discussion.
52
-------
ACCUMULATION OF ATMOSPHERIC CONSTITUENTS IN SNOWFALL
As previously indicated, the accumulation of gaseous pollutants with snow
may occur via condensation, nucleation, or adsorption processes enhanced by
the fact that snow tends to form at temperatures below ~-20°C. Particulate
material can accumulate by serving as the nuclei for snow formation and
growth, by gravitational settling, and by impactive removal under (micro- or
macro-scale) turbulent atmospheric conditions.
The accumulation and retention of gaseous entities such as Hg vapor via
snow will clearly depend on the operative temperature regime. To demonstrate
this, known amounts of Hg vapor were generated in the lab by reduction of
standard mercury solutions with stannous ion and passed through loosely packed
columns of snow held at various temperature levels using a cryostatic chamber.
At levels between ~2-8°C, small fractions (10-25%) of the mercury vapor were
absorbed by the melted water on the snow crystal surface(s). Mercury
accumulated in this manner was retained when the snow was allowed to continue
melting and reach ambient temperature. Between ~0 and ~-15°C the retention
of Hg vapor by the snow was essentially negligible (<5%), but below this re-
tention increased rapidly to 90-100%. In view of the fact that the vapor
pressure of Hg drops from 185 x 1Q-6 to 18 x 10-6 to 4.8 x 10-6 mm as the
temperature drops to 0, -20, and -30°C, respectively, the condensation of Hg
at these lower temperatures can be anticipated. After Hg was condensed on the
snow at such temperatures, raising the temperature to 0°C resulted in the
recovery of most of the Hg in the vapor state. These laboratory experiments
verified that snow can serve as an accumulator for mercury vapor at low temp-
eratures (<-20°C) but that the mercury vapor may be released back to the
atmosphere when the temperatures rise. If the snow melts rapidly, a fraction
of the mercury may be "dissolved" in the product water.
The inclusion of gaseous constituents such as organics in rainfall has
been demonstrated by Junge (1963) and Gatz (1976). This process may also
occur in the cloud during the conversion of raindrops to snowflakes but data
supporting this have not been found. By analogy with the Hg case discussed
above, organic plume constituents having vapor pressures approaching 10~s mm
at temperatures characteristic of snow nucleation (<-20°C) may condense onto
snowflakes or be adsorbed onto the surfaces thereof. Most inorganic gases
(e.g., NO , SO , and CO ) exhibit much higher vapor pressures at such temp-
XX X
eratures; the probability of these condensing during nucleation should be
quite low. It is known, however, that gases such as S02, NO^, NH3, C02, and
C12 dissolve in water and hydrolyze to produce a variety of ionic species
including, for example, SO*- and NO^ (Junge 1963, Gatz 1976, Friedlander
1977). The prominence of such gas-to-particle conversion reactions (both
homogeneous and heterogeneous) in coal combustion plumes is widely accepted
as highly significant. Their occurrence during nucleation and growth of
snowflakes can also be readily rationalized.
53
-------
REFERENCES
American Chemical Society. 1969. Cleaning our environment: The chemical
basis for action. Rep. by Subcommittee on Environmental Improvement,
Committee on Chemistry and Public Affairs, Washington, D. C. 249 pp.
American Public Health Association, American Water Works Association, and
Water Pollution Contr&l Federation. 1971. Standard methods for the
examination of water and wastewater. 13th Edition. Washington, D. C.
874 pp.
American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1976. Standard methods for the
examination of water and wastewater. 14th Edition. Washington, D. C.
1193 pp.
American Public Health Association, Intersociety Committee. 1977. Methods
of air sampling and analysis. M. Katz (Ed.). 2nd Edition. Washington,
D. C. 480 pp.
Baril, S. F., R. J. Luedtke, and G. R. Roemhild. 1978. Environmental effects
of western coal combustion. Part II - The Aquatic Macroinvertebrates of
Rosebud Creek, Montana. EPA Ecol. Res. Ser. EPA-600/3-78-099, U. S.
Environmental Protection Agency, Duluth, Minnesota. 75 pp.
Crecelius, E. A., L. A. Rancitelli, and S. Garcia. 1978. Power Plant Emis-
sions and Air Quality, pp. 9-33. In The Potential Impact of Coal-Fired
Power Plants in the Northern Great Plains. Dept. of Energy Annual
Report, Ames Laboratory, Iowa State University, Ames, Iowa 243 pp.
Davison, R. L., D. F. S. Natusch, J. R. Wallace, and C. A. Evans, Jr. 1974.
Trace elements in fly ash. Dependence of concentration on particle
size. Environ. Sci. Technol. 8(13):1107-1113.
Dollhopf, D. J., W. D. Hall, W. M. Schafer, E. J. DePuit, and R. L. Hodder.
1977a. Selective placement of coal stripmine overburden in Montana, I.
Data base. Montana Agricultural Experiment Station Research Report 128.
Bozeman, Montana. 109 pp.
Dollhopf, D. J., W. D. Hall, C. A. Cull, and R. L. Hodder. 1977b. Selective
placement of coal stripmine overburden in Montana, II. Initial field
demonstration. Montana Agricultural Experiment Station Research Report
125. Bozeman, Montana. 98 pp.
54
-------
Dollhopf, D. J., J. D. Goering, C. J. Levine, B. J. Bauman, D. W. Hedberg,
and R. L. Hodder. 1978. Selective placement of coal stripmine over-
burden in Montana, III. Spoil mixing phenomena. Montana Agricultural
Experiment Station Research Report 135. Bozeman, Montana. 68 pp.
Elser, A. A., and J. C. Schreiber. 1978. Environmental effects of western
coal combustion. Part I - The Fishes of Rosebud Creek, Montana. EPA
Ecol. Res. Ser. EPA-600/3-78-098, U. S. Environmental Protection Agency,
Duluth, Minnesota. 34 pp.
Federal Power Commission. 1974. Report on Environmental Research. National
Power Survey, Technical Advisory Committee an Research and Development,
Washington, D. C.
Friedlander, S. K. 1977. Smoke, dust, and haze: Fundamentals of aerosol
behavior. John Wiley and Sons, Mew York, New York. 237 pp.
Gatz, D. F. 1976. Precipitation scavenging. ERDA Symp. Ser. 41. National
Tech. Information Sen/., CONF-741003. 78 pp.
Gluskoter, H. J., R. R. Ruch, W. G. Miller, R. A. Cahill, G. B. Dreher, and
J. K. Kuhn. 1977. Trace elements in coal. Illinois Geol. Surv. Circ.
No. 499. 154 pp.
Hales, J. M. 1973. Fundamentals of the theory of gas scavenging by rain.
Atmos. Environ. 6(9):635-659.
Heimbach, J. A. and A. B. Super. 1973. Air pollution potential determina-
tion program for Colstrip, Montana. Final Report, Part II,, Prepared
for the Montana Power Company by the Department of Earth Sciences,
Montana State University, Bozeman, Montana. 79 pp. + 3 appendices.
Junge, C. E. 1963. Atmospheric chemistry and radioactivity. Academic
Press, New York. 465 pp.
Linton, R. W., A. Loh, D. F. S. Natusch, C. A. Evans, Jr., and P. Williams.
1976. Surface predominance of trace elements in airborne particles.
Science 191(4229):852-854.
McWhorter, D. B., R. K. Skogerboe, and G. V- Skogerboe. 1975. Water quality
control in mine spoils, Upper Colorado River Basin. EPA-670/2-75-048.
U. S. Environmental Protection Agency, Cincinnati, Ohio, 122 pp.
Mesmer, R. E., and C. F. Baes, Jr. 1975. The hydrolysis of cations. Oak
Ridge National Laboratory Rep. Nos. ORNL-NSF-EATC-3, Parts I, II, and
III. Oak Ridge, Tennessee. 235 pp.
Montana Power Co. 1976a. Report to Montana Department of Health and Environ-
mental Sciences for tests run on December 15-17, 1976. Rep. prepared by
Yapuncich, Sanderson, and Brown Laboratories, Billings, Montana. 46 pp.
55
-------
Montana Power Co. 1976b. Report to Montana Power Co. by Yapuncich, Sanderson,
and Brown Laboratories, Billings, Montana. 26 pp.
National Academy of Sciences,. 1972. Participate polycyclic organic matter.
Rep. of Committee on Biological Effects of Atmospheric Pollutants,
Washington, D. C. 235 pp.
Natusch, D. F. S. 1976. Characterization of atmospheric pollutants from
power plants, pp. 114-129. In Proc. Second Federal Conference on the
Great Lakes. Public Information Office of Great Lakes Basin Commission,
Ann Arbor, Michigan.
Natusch, D. F. S., C. F. Bauer, H. Matusiewicz, C. A. Evans, Jr., J. Baker,
A. Loh, and R. W. Linton. 1975. Characterization of trace elements in
fly ash, pp. 553-575. In Proc. Int. Conf. on Heavy Metals in the Environ-
ment, Vol. II, Part 2. Toronto, Canada.
Natusch, D. F. S., and D. R. Taylor. 1980. Chemical and physical character-
istics and environmental behavior of coal fly ash. EPA Ecol. Res. Ser.,
Environmental Research Laboratory, U. S. Environmental Protection Agency,
Duluth, Minnesota (In prep.).
Pierce, W. G. 1936. The Rosebud Coal Field, Rosebud and Custer Counties,
Montana. U. S. Geol. Surv., Bull. 847-B.
Renick, B. C. 1929. Geology and Ground-Water Resources of Central and
Southern Rosebud County, Montana. U. S. Geol. Surv., Water-Supply Paper
No. 600. 140 pp.
Skogerboe, R. K., C. S. Lavallee, M. M. Miller, and D. L. Dick. 1979.
Environmental effects of western coal surface mining. Part III. The
water quality of Trout Creek, Colorado. EPA Ecol. Res. Ser. Rep. No.
EPA-600/3-79-008. Environmental Research Laboratory, U. S. Environ-
mental Protection Agency, Duluth, Minnesota. 57 pp.
Skogerboe, R. K., M. M. Miller, D. L. Dick, R. V. Thurston, and R. C. Russo.
1980. The water quality of Rosebud Creek, Montana, 1975-1977: Compila-
tion of chemical data. NREL-FBL Technical Rep. No. 4. Natural Resource
Ecology Laboratory, Colorado State University, and Fisheries Bioassay
Laboratory, Montana State University, 116 pp.
Thorn, W. T., Jr., G. M. Hall, G. H. Wegemann, and G. F. Moulton. 1938.
Geology of Bighorn County in the Crow Indian Reservation, Montana, with
special reference to the water, coal, oil, and gas resources. U. S.
Geological Survey Bull. No. 856. 88 pp.
Turner, D. B. 1974. Workbook of atmospheric dispersion estimates. Health
Services Publ. No. 995-AP-26. U. S. Department of Health, Education,
and Welfare, Washington, D. C. 185 pp.
56
-------
U. S. Environmental Protection Agency. 1974. Methods for chemical analysis
of water and wastes. EPA Rep. No. EPA-625/6-74-003. Office of Technol-
ogy Transfer, Washington, D. C.
U. S. Environmental Protection Agency. 1976. Proc. National Conf. on Health,
Environmental Effects, and Control Technology of Energy Use. February
9-11. EPA Rep. No. 600/7-76-002, Office of Res. and Dev., Washington,
D. C. 340 pp.
U. S. Environmental Protection Agency. 1977. Quality criteria for water,
Office of Water and Hazardous Materials. Washington, D. C. 256 pp.
Van Hook, R. I., and W. D. Shults. (Eds.). 1977. Effects of trace contami-
nants from coal combustion. U. S. Energy Res. and Devel. Admin. Rep.
77-64. Washington, D. C. 79 pp.
Van Voast, W. A., R. B. Hedges, and J. J. McDermott. 1977. Hydrogeological
conditions and projections related to mining near Colstrip; Southeastern
Montana. Montana Bureau of Mines, Bull. 102. 42 pp.
57
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/3-8Q-071
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Environmental Effects of Western Coal Combustion:
Part III - The Water Quality of Rosebud Creek,
Montana
5. REPORT DATE
JULY 1980 ISSUING DATE.
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
R. K. Skogerboe, M. M. Miller, D. L. Dick, R. V.
Thurston, and R. C. Russo
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Dept. of Chemistry, Colorato State University,
Fort Collins, Colorado 80523 and Fisheries Bioassay
Laboratory, Montana State University, Bozeman,
Montana 59717
10. PROGRAM ELEMENT NO.
1HE625, 1NE831
11. CONTRACT/GRANT NO.
R803950
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory-Duluth
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, Minnesota 55804
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/03
16. SUPPLEMENTARY NOTES
16. ABSTRACT
The results of a study on Rosebud Creek, Montana, designed to assess the impacts
on water quality of surface coal mining and/or coal combustion at Col strip are
summarized herein. A general degradation of water quality has been observed along
the stream course but direct impacts of groundwater from the mine areas have not
been demonstrated. Influxes of arsenic, mercury, selenium, and polynuclear aromatic
compounds during snowmelt periods have been linked by analyses of snow samples to
their accumulation in snowfall via scavenging of the power plant plume and subsequent
delivery to the stream via surface runoff. Although this contaminant transfer route
may prove highly significant, it is emphasized that this will depend on site specific
conditions.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Water quality
Heavy metals
PNA compounds
Coal combustion
Coal conversion
Atmospheric transport
07/B
10/A
10/B
18. DISTRIBUTION STATEMEN1
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
Unclassified
20. SECURITY CLASS (Thispage)
Unclassified
21. NO. OF PAGES
68
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
58
U.S. GOVERNMENT PRINTING OFFICE: 1980--657-165/0064
-------
United States
Environmental Protection
Agency
Center for Envir7>nm*ritIT ftoaelrch
Information
Cincinnati OH 45268
Fees Paid
Environmental
Protection
Agency
EPA-335
Official Business
Penalty for Private Use, $300
Special Fourth-Class Rate
Book
Please make all necessary changes on the above label,
detach or copy, and return to the address in the upper
left-hand corner
If you do not wish to receive these reports CHECK HERE a,
detach, or copy this cover, and return to the address in the
upper left-hand corner
EPA-600/3-80-071
------- |