IMPACT OF MINE DRAINAGE AND DISTRIBUTION OF METAL LOADING
SOURCES IN THE JAMES CREEK WATERSHED
ASSESSED UNDER LOW-FLOW CONDITIONS
University of Colorado at Boulder
Laura Harrington, Duke University
Ned Turner, University of Colorado
Advisor – Joe Ryan
Abstract
Increasing environmental awareness over the past several decades has led to widespread
concern about the adverse effects that heavy mining has had on the environment. These effects
are felt particularly in the form of water contamination from acid mine drainage and metal
dissolution from mining waste rock. The goal of the current study was to assess these effects on
the James Creek watershed located in northwestern Boulder County, Colorado, where mining
was prevalent from the 1850s – 1980s. Specifically, a conservative chloride tracer was injected
in two adjacent reaches of James Creek in order to quantify stream discharge, and synoptic
samples were taken to identify the sources that of metal and acid contamination in the creek and
to quantify the loading of these metals. The results of this study suggest that there are several
metal sources along the stretch of the Upper James Creek, including significant sources from the
Fairday Mine area, an unidentified zinc source at about 2.7 km downstream of the injection, and
a major metal source near the treatment plant. The Little James Creek, in addition to another
metal source located 4.5 km downstream of the injection, are the two major metal contributors in
Lower James Creek. Our results may be somewhat limited in identifying all potential
contamination sources in James Creek because of the low flow conditions in the creek when our
experimentation was carried out. Finally, our results suggest that the effects of metal loading in
the creek are spatially restricted by precipitation of metals.
2
Introduction
The environmental health of historically mined areas is significantly threatened by the
process of acid mine drainage and the resulting water contamination poses a problem for both
aquatic life and human drinking water. Acid mine drainage occurs when rocks containing
sulfide minerals from mines or tailings piles interact with air and water to produce sulfuric acid.
The production of sulfuric acid not only creates acidic conditions in the surface or groundwater
into which the sites drain, but it also causes the dissolution of heavy metals from waste rock,
which can also contaminate the water [Drever, 1997].
The watershed of interest in this study is the James Creek watershed, encompassing about
2
48 km in northwestern Boulder County, Colorado. The James Creek watershed contributes
directly to the water supply for Jamestown, Colorado, a town of approximately 250 residents,
and includes the James and Little James Creeks. James Creek is a tributary of Lefthand Creek,
and is thus considered to be a part of the Lefthand Creek watershed. Therefore, water quality
concerns are not only relevant to the 250 residents of Jamestown, but also to all 14,000 residents
in the Lefthand Creek watershed [Duren et al., 2001].
The area around the James and Little James Creeks is typically referred to as the Golden
Age Mining district, where mining for gold, silver, lead, fluorspar, and uranium took place from
the 1850s – 1980s [Duren et al., 2001]. Metal contamination in the James Creek watershed was
discovered when, in response to the Clean Water Act, Section 303(d), states were obligated to
identify water bodies with poor water quality and to propose standard contamination levels
above which the concentrations could pose a threat to either aquatic or human well being. The
Little James Creek was on the 1998 list for Colorado as not being in attainment for pH and metal
concentrations. Although James Creek was not included on this list because tests revealed it was
in attainment, it is still worthwhile to study in order to fully understand the affects the Little
James Creek has on James Creek after their confluence [Colorado Water Quality Control
Division, 2002].
Two portions of James Creek are under consideration in this study. The first is the 5 km
stretch starting at the road crossing upstream of John Jay Mine and ending at the Jamestown
water treatment plant (JWTP); the second portion is the 5 km stretch starting at the JWTP and
ending at the confluence of James Creek with Lefthand Creek. For the purpose of this study,
these two stretches of creek will be referred to as the Upper James Creek and the Lower James
3
Creek, respectively. The land adjacent to Upper James Creek contains several mining areas, as
well as a tailings pile near the JWTP. The two mining areas along the Upper James Creek are
John Jay mine and Fairday Mine. The land adjacent to Lower James Creek is used mostly for
residential purposes. There are no mines located in the immediate vicinity of the creek, but there
is an old tailings pile that is now the site of Elysian Park, the town park. Another site of interest
along this section of creek is Curie Springs, located downstream from Jamestown, where
uranium rich water was bottled for medicinal purposes as late as the 1950s. We are interested in
these areas, as well as any other inflowing water sources, to see if they contribute significantly to
metal concentrations in the creek.
In order to fully understand the effects of metal contamination, it is necessary to measure
metal loading. Metal loading rates (mass per time) are calculated as the product of metal
concentrations (mass per volume) and stream discharge rates (volume per time). Typically,
discharge has been measured using the cross-sectional area of the stream and a flow meter to
measure water velocity [Kimball, 1997]. However, flow meter measurements underestimate
stream discharge because they cannot account for water flow through streambed sediments and
cobbles. Such an underestimation would also lead to error in the calculation of metal loading
rates. Therefore, an alternative method has been devised to mitigate this error [Kimball, 1997;
Kimball et al., 2001a; Kimball et al., 2001b].
Experiments involving injections of conservative tracers (e.g., sodium chloride) and
synoptic sampling over the reach of the stream affected by mine drainage produce more accurate
calculations of stream discharge and more precisely identify potential sources of contamination
[Kimball, 1997; Kimball et al., 2001a; Kimball et al., 2001b]. These methods have been
successfully employed in studies examining the effects of acid mine drainage on the water
quality in Cement Creek (Colorado) and in Little Cottonwood Creek (Utah) [Kimball et al.,
2001a; Kimball et al., 2001b]. We also anticipated that the use of a tracer injection and synoptic
sampling in our study would help identify less obvious contamination sources, such as
contaminated groundwater, which cannot be seen by observing surface runoff or seepage into the
creek.
The objective of this study is to (1) identify the sources that might be responsible for
metal and acid contamination in the creek and (2) quantify the amount of metals being
introduced into the each reach of James Creek. This knowledge is important not only for the
4
consideration of the effects of metal contamination on aquatic life and human drinking water, but
it is also important to determine if the James Creek watershed should be included on the U.S.
Environmental Protection Agency’s National Priority List for remediation under the
Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA; i.e., the
“Superfund” list).
Methods and Materials
Field Experimentation
The methods used to evaluate metal contamination in the each reach of James Creek were
adapted from previous tracer injection and synoptic sampling experiments performed by Kimball
et. al.. To measure stream discharge, a conservative tracer (chloride) was injected above the
JWTP and monitored just upstream of the confluence of James Creek with Lefthand Creek. To
measure metal loading, the stream was sampled at intervals of about 200 m while the injected
tracer concentration was on its plateau.
In the first part of the experiment, a 3 M solution of NaCl (concentration determined
experimentally) was injected into Lower James Creek for 3 hours and 15 minutes (from 10:45
am until 2:00 pm). The injection solution was prepared in 2-44 gallon trashcans using 80 lbs.
NaCl in approximately 40 gallons of water for each barrel. After stirring the solutions for about
30 minutes, excess NaCl was removed from the barrels. During the injection experiment, six
samples were taken of the injection solution (three from each barrel), in order to accurately
determine its concentration. The second barrel of injection solution was determined to be at a
slightly higher concentration that the first. Therefore, chloride concentrations of samples taken
when the solution from the second barrel was in the stream were corrected to the concentration
of the first barrel (3 M).
The 3 M NaCl solution was injected into Lower James Creek at a rate of 1.26 L·min-1;
monitoring of the pump throughout the injection ensured the pump rate remained constant. Over
the course of the injection, six water samples were taken upstream of the injection site in order to
determine the background concentration of chloride in James Creek.
There were two sites on the Lower James Creek where water samples were collected in
order to monitor the chloride tracer concentration as a function of time. The upstream
monitoring site was located about 107 m downstream from the injection, just upstream from the
5
confluence of the Little James and James Creeks. The downstream monitoring site was located
5212 m downstream from the injection site, just upstream from the confluence of James Creek
with Lefthand Creek. At the upstream site, samples were taken every minute during the leading
and trailing edges of the tracer concentration, and were taken approximately every thirty minutes
while the tracer concentration was on its plateau. At the downstream monitoring site, samples
were taken approximately every five minutes for the duration of the experiment.
In the second part of the experiment, water samples were taken at synoptic sample sites
located 200 m apart over the stretch of the creek. In order to better bracket the area around
Elysian Park, synoptic samples in this area were taken at intervals of 100 m. Figure 1 shows a
map of where synoptic samples were taken along the stretch of Lower James Creek. An ion
specific electrode (ISE) with chloride probe was used to monitor the relative change in chloride
concentrations at the upstream and downstream monitoring sites in order to determine when the
plateau had been reached. Once the plateau was reached at the downstream sample site, synoptic
sampling began along Lower James Creek. In addition to synoptic samples, four potential
inflowing water sources were sampled for metal analysis. Two of the sampled inflows appeared
to be gulches flowing into Lower James Creek, most likely the Buffalo and Castle Gulches. The
third inflow was sampled from ponds located in the vicinity of Curie Springs. The final inflow
was a ground water source located across the road from the creek that could potentially affect the
Lower James Creek by flowing under the road toward the creek.
Figure 1 – Map of Synoptic Sample Sites along the Lower James Creek
6
A similar procedure was used for the Upper James Creek, as well. A 6.2 M NaCl
solution was injected into the upper stretch at a rate of 0.82 L·min-1 for 4 hours and 20 minutes.
The chloride tracer concentration was monitored as a function of time at two stations
downstream of the injection. The upstream site was located at approximately 90 m downstream
of the injection, and the downstream sampling site was located at 4940 m downstream of the
injection, at the JWTP. Sampling rates are similar to those of the Lower James Creek portion.
Synoptic samples were also taken at approximately every 200 m along the entire stretch.
. Figure 2 shows a map of where synoptic samples were taken along the stretch of Lower James
Creek. Colorimeters were used to monitor the relative change in chloride concentrations at the
upstream and downstream monitoring sites in order to determine when the plateau had been
reached. Once the plateau was reached at the downstream sample site, synoptic sampling began
along Upper James Creek. In addition to synoptic samples, two potential inflowing water
sources were sampled for metal analysis. These inflows were located at the Fairday Mine
tributary and at a stagnant pond at John Jay Mine.
Figure 2 – Map of Synoptic Sample Sites along the Upper James Creek
7
Laboratory Analysis
For analysis of chloride concentrations of samples taken in the field, an ISE with chloride
probe was used to make concentration measurements. Chloride concentrations were measured
for the samples taken at each chloride monitoring site, for all synoptic samples, for the samples
taken of the injection solutions and for the samples taken upstream of the injections. In addition
to chloride measurements, pH was also measured on the synoptic samples.
Metal analysis, which included measurement of both total and dissolved metal
concentrations, was performed on the 34 synoptic samples and the 4 samples taken of potential
inflowing water sources for Lower James Creek, and on the 29 synoptic samples and the 2
samples taken of potential inflowing water sources for Lower James Creek. The total metal
samples were prepared by acidifying the samples with 2-3 drops of trace metal grade nitric acid
(HNO3, VWR). The dissolved metal samples were prepared by first filtering through a 0.2 μm
cellulose acetate membrane and then acidifying with 2-3 drops of trace metal grade HNO3. Both
total and dissolved metal samples were analyzed for concentrations of Al, Cu, Fe, Pb, Mn, U,
and Zn using Inductively Coupled Plasma – Mass Spectrometry (ICP-MS).
Calculations
The chloride concentration results for the synoptic samples were used to calculate stream
discharge at each sample site. Stream discharge was calculated according to the following
equation:
Q AB =
Ci Qi
CB − CA
(1)
where QAB is the stream discharge (L/s) between sampling points A and B, Qi is the injection rate
(L/s), Ci is the injection chloride concentration (mol/L), CA is the upstream chloride
concentration (moles/L), and CB is the chloride concentration at downstream site B (moles/L).
Metal loading was calculated according to the method described earlier:
M B = C B Q AB
(2)
where MB is the metal loading of a particular metal at synoptic sample site B, CB is the metal
concentration of a particular metal at synoptic samples site B and QAB is the stream discharge
between the upstream sample site and sample site B.
8
Results
Results from Chloride Analysis
The purpose of monitoring chloride concentrations at the upstream and downstream sites
was (1) to calculate the travel time of the tracer (2) to ensure that the synoptic samples were
taken while the concentration of the tracer had reached a plateau throughout the creek. The
results of the chloride sampling for James Creek (Figure 3) show that the tracer arrived very
quickly at the upstream sampling site, but took between two and a half hours and three and a half
hours to reach the downstream monitoring site. This rate at which the tracer moved in the stream
was determined by observing the time difference between the arrivals of the tracer at the two
monitored sites, and was calculated to be 0.60 m/s for Lower James Creek and 0.40 m/s for
Upper James Creek.
Upper James Creek
3.00E-04
2.50E-04
2.00E-04
1.50E-04
1.00E-04
5.00E-05
0.00E+00
Concentration (M)
Concentration (M)
Lower James Creek
0
50
100
150
200
250
300
350
Time afte r inje ction (min)
Upstream Site (107 m)
Downstream Site (5212 m)
2.00E-04
1.50E-04
1.00E-04
5.00E-05
0.00E+00
0
100
200
300
400
500
600
Time afte r inje ction (min)
Upstream Site (92 m)
Downstream Site (4938 m)
Figure 3 – Chloride concentration as a function of time
Abnormal weather conditions in the months leading up to our experiment, including a
mild winter and a summer drought, caused the stream flow in James Creek to be much lower
than normal. Due to the underestimation of travel time, the trailing edge of the tracer was not
observed by samples taken at the downstream sampling site. However, it is clear from Figure 3
that a distinct concentration plateau was reached at the downstream monitoring site, and it was
during this time that synoptic samples were taken.
Under normal conditions, the tracer concentration should decrease between upstream and
downstream sample sites due to inflowing water sources. This behavior is exhibited in Figure 4
for the Upper James Creek. However, the chloride concentrations measured for our synoptic
samples did not reflect this dilution for the Lower James Creek because most of the potential
inflow sources had dried up as a result of the drought. The measured concentrations were
9
relatively stable however a jump in chloride concentration was observed for sites after 1800 m
downstream from the injection sites. Data obtained from chloride analysis of the synoptic
sample site was used to calculate stream discharge at each sample site, according to Equation 1.
Because there were no significant inflowing water sources along the stretch of the Lower James
3000
1000
-1
S tream D ischarge (L·s )
Creek, an increase in stream discharge was not observed as a function of distance downstream.
Discharge (L/s)
2500
800
600
2000
1500
1000
500
400
0
1000
2000
3000
4000
D istance D ow nstream (m)
Lower James Creek
5000
6000
0
0
1000
2000
3000
4000
5000
Distance Downstream (m)
Upper James Creek
Figure 4 – Results of chloride concentration measurements taken of the synoptic samples
Results from Metal Analysis
The results of metal analysis reported directly from the ICP-MS were in terms of metal
concentrations for both total and dissolved metal types. For the purpose of further metal
analysis, the concentration of colloidal metal was taken to be the difference between the total and
dissolved metal concentrations. Concentrations for each of the three metal types are shown in
Figure 5 as a function of distance downstream from the injection, for all of the metals analyzed
except for zinc in Lower James Creek. There was some question about the validity of the
measurements made for zinc, and they are therefore omitted from the results for the lower
portion of the creek. Also shown on the graphs are the ICP-MS detection limits for each metal.
For those samples with metal concentrations below the detection limit of the instrument, the
concentrations were taken to be the concentration of that metal at its detection limit.
10
Aluminum C oncentration C urves
3.00
Conce ntra tion Cu (ppb)
Conce ntra tion Al (ppb)
180.0
C opper C oncentration C urves
Unidentified M etal S ourc e
Little Jam es Creek
Curie S prings
E lys ian P ark
120.0
60.0
0.0
Unidentified M etal S ourc e
Little Jam es Creek
Curie S prings
E lys ian P ark
2.00
1.00
0.00
0
1000
2000
3000
4000
5000
6000
0
1000
2000
Dista nce Dow nstre a m (m )
Total A l
Dis solved A l
Colloidal A l
Detection Lim it
Total Cu
Unidentified M etal S ource
Dis solved Cu
Curie S prings
E ly sian P ark
200
100
5000
6000
Detec tion Lim it
Unidentified M etal S ource
0.60
Curie S prings
E lys ian P ark
0.30
0.00
0
0
1000
2000
3000
4000
5000
0
6000
1000
2000
Total Fe
Diss olved Fe
Colloidal Fe
Total P b
Detection Lim it
Manganese C oncentration C urves
20.00
Curie S prings
15.00
5000
6000
Dis solved P b
Colloidal P b
Detection Lim it
U ranium C oncentration C urves
Unidentified M etal S ourc e
Little Jam es Creek
4000
Little Jam es Creek
E ly sian P ark
10.00
5.00
0.00
Unidentified M etal S ource
0.80
Conce ntra tion U (ppb)
25.00
3000
Dista nce Dow nstre a m (m )
Dista nce Dow nstre a m (m )
Conce ntra tion M n (ppb)
Colloidal Cu
Little Jam es Creek
0.90
Conce ntra tion P b (ppb)
Conce ntra tion Fe (ppb)
300
4000
Lead C oncentration C urves
Iron C oncentration C urves
Little Jam es Creek
3000
Dista nce Dow nstre a m
Curie S prings
0.60
E ly sian P ark
0.40
0.20
0.00
0
1000
2000
3000
4000
5000
6000
0
1000
Dista nce Dow nstre a m (m )
Total M n
Diss olved M n
Colloidal M n
2000
3000
4000
5000
6000
Dista nce Dow nstre a m (m )
Detection Lim it
Total U
Diss olved U
Colloidal U
Detection Lim it
Figure 5a: Metal Concentration Curves from synoptic samples in Lower James Creek: Samples taken around
the park area enclosed in the rectangle, black arrows point to significant geographical areas (Little James Creek,
Curie Springs, and an unidentified metal source), and the green arrows point to sites just downstream from the three
remaining inflows were sampled.
It is important to note that because the dissolved metal concentrations are relatively
constant at all the synoptic sample sites, the shape of the colloidal metal concentration curve
greatly resembles the total metal concentration curve. The geographical significance of trends
observed on the metal concentrations curves will be discussed in further detail in the section
addressing metal loading. The results for metal concentration and metal loading were very
similar in nature, and it is more useful to consider the impacts of the metals on the stream from
the standpoint of metal loading.
11
Synoptic Mn Concentration
Synoptic Al Concentration
140
120
100
25
Mn Concentration (ppb)
160
80
60
40
20
15
10
5
20
0
0
500
1000
John Jay Mine
Fairday Mine
John Jay Mine
Al Concentration (ppb)
180
1500
2000
2500
3000
3500
4000
4500
Fairday Mine
30
200
0
5000
0
500
1000
1500
Distance Downstream (m)
Dissolved
Total
Dissolved
Colloidal
2500
3000
3500
4000
4500
5000
Total
4000
4500
5000
Colloidal
Synoptic Cu Concentration
Synoptic Fe Concentration
600
300
200
100
8
Fairday Mine
10
John Jay Mine
400
Cu Concentration (ppb)
500
Fairday Mine
12
John Jay
Mine
Fe Concentration (ppb)
2000
Distance Downstream (m)
6
4
2
0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0
Distance Downstream (m)
0
500
1000
1500
2000
2500
3000
3500
Distance Downstream (m)
Dissolved
Total
Dissolved
Colloidal
Colloidal
Synoptic Pb Concentration
Synoptic Zn Concentration
4.0
John Jay
Mine
300
100
50
2.5
Fairday Mine
150
3.0
John Jay
Mine
200
3.5
Pb Concentration (ppb)
250
Fairday Mine
Zn Concentration (ppb)
Total
2.0
1.5
1.0
0.5
0.0
0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Dissolved
Total
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Distance Downstream (m)
Distance Downstream (m)
Dissolved
Colloidal
Total
Colloidal
Synoptic U Concentration
1.0
Fairday Mine
0.8
0.7
0.6
John Jay
Mine
U Concentration (ppb)
0.9
0.5
0.4
0.3
0.2
0.1
0.0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Distance Downstream (m)
Dissolved
Total
Colloidal
Figure 5b: Metal Concentration Curves from synoptic samples in Upper James Creek: Green arrows point to
sampled inflows (John Jay Mine, and Fairday Mine).
12
Discussion
Metal Loading
Metal loading graphs (Figure 6) were generated for aluminum, copper, iron, lead,
manganese, zinc, and uranium according to Equation 2, using the measured metal concentrations
and stream discharge rates at each synoptic sample site. The accumulation curves are included
on the graphs in Figure 5 first, to identify points along James Creek where significant increases
in metal loading were observed, and second, to help illustrate the total amount of metal that
accumulates over the stretch of the creek. It is important to note that not all of the accumulated
metals remain in the water for the entire stretch of the creek, because some metals are being
removed through attenuation.
The graphs in Figure 6a emphasize that there are two distinct spikes in metal loading
along Lower James Creek shared by all six metals. The first spike occurred just downstream
from the confluence of Little James Creek with James Creek, and the second spike was observed
about 4.5 km downstream from the injection site. The source of this spike is at this point
unidentified, but appears to be a significant metal contributor.
Samples taken from Lower James Creek in the region that bracketed Elysian Park did not
contain as much metal loading as was expected prior to experimentation. Increases in the metal
loading of aluminum, iron, manganese, and uranium in this area were relatively small. However,
more significant increases in metal loading of copper and lead are seen at sites adjacent to the
park. The left over tailings pile could be the cause of these observed metal loadings in the Lower
James Creek.
The only other significant spike in metal loading was observed for copper in the vicinity
of Curie Springs. It was surprising that a corresponding increase in uranium loading was not
also observed, given the past use of the land. With respect to the remaining sampled inflow sites,
they do not appear to have contributed significantly to metal loading. With the exception of
copper, the other metals do not see any significant increases in metal loading between 500-4000
m and any observed variations in metal loading within this stretch of stream were relatively
small. It is interesting to note that sites where an increase in metal loading was observed were
generally followed decreased metal loading at sites immediately downstream.
13
C opper Loading
Aluminum Loading
0.40
30.00
E lysian Park
20.00
Unidentified M etal S ource
10.00
Cu Lo a d (kg/da y)
Al Loa d (kg/da y)
Curie S prings
Curie S prings
Little Jam es Creek
Little Jam es Creek
0.30
Elysian P ark
Unidentified M etal S ource
0.20
0.10
0.00
0.00
0
1000
2000
3000
4000
5000
0
6000
1000
2000
Total Al
Dissolved Al
Colloidal Al
Total Cu
Total A l A ccum ulation
Dissolved Cu
Iron Loading
40.00
E lysian Park
Unidentified M etal S ource
20.00
10.00
0.00
Pb Loa d (kg/da y)
Fe Loa d (kg/da y)
5000
6000
Colloidal Cu
Total Cu A ccum ulation
0.16
Curie Springs
Little Jam es Creek
30.00
4000
Lead Loading
Curie S prings
Little Jam es Creek
3000
Dista nce Dow nstre a m (m )
Dista nce Dow nstre a m (m )
0.12
Elysian Park
Unidentified M etal S ource
0.08
0.04
0.00
0
1000
2000
3000
4000
5000
6000
0
1000
2000
Dista nce Dow nstre a m (m )
Total Fe
Dis solved Fe
Colloidal Fe
3000
4000
5000
6000
Dista nce Dow nstre a m (m )
Total Fe A ccum ulation
Total Pb
Dissolved Pb
Mangnese Loading
Colloidal P b
Total P b Ac cum ulation
U ranium Loading
3.50
0.15
3.00
Curie S prings
Little Jam es Creek
Curie Springs
Little Jam es Creek
2.50
0.10
E ly sian P ark
2.00
Elysian Park
Unidentified M etal Sourc e
Unidentified Metal Source
1.50
0.05
1.00
0.50
0.00
0.00
0
1000
Total M n
2000
Dissolved M n
3000
Colloidal M n
4000
5000
Total M n A ccum ulation
6000
0
1000
Total U
2000
Dissolved U
3000
Colloidal U
4000
5000
6000
Total U A ccum ulation
Figure 6a – Metal Loading and Accumulation Graphs (Lower James Creek) – Metal loading curves for total,
dissolved, and colloidal metals are given in blue, yellow, and green, respectively. Metal accumulation graphs,
generated by adding all the consecutive increases in metal loading, are shown in red.
The graphs of Figure 6b emphasize the metal loading of Upper James Creek. All seven
metals observed shared a significant spike near sampling site 3. This site is located near the
JWTP, and could potentially be produced by the tailings pile north of the JWTP on a nearby hill.
Surprisingly, there is no evidence of metal loading from the John Jay Mine site, as there is no
spike in the graphs at this location. However, uranium, copper, and lead all exhibit an increase
in load near the confluence of Upper James Creek and the Fairday Mine tributary. Interestingly,
the concentrations of copper and lead are very low in the influent samples collected. These
metals are most likely introduced by a groundwater source separate from the influent tributary.
14
The only other significant metal loading source is between sites 13 and 14, at 2.7 km
downstream of the injection. This location sees a significant loading of zinc, with no obvious
source.
Synoptic Al Load
Synoptic Mn Load
3/4
8/9
8/9
20
7
6
5
4
Fairday Mine
30
John Jay Mine
8
Mn Load (kg/day)
40
Fairday Mine
28/29
50
John Jay Mine
Al Load (kg/day)
60
9
3/4
10
70
3
2
10
1
0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0
5000
500
1000
1500
Dissolved
Total
Colloidal
Total Accumulated
Dissolved
Synoptic Fe Load
3000
3500
4000
4500
5000
Total
Colloidal
Total Accumulation
40
20
19/20
22/23
Fairday Mine
60
2.0
John Jay Mine
Cu Load (kg/day)
80
2.5
Fairday Mine
John Jay
Mine
Fe Load (kg/day)
100
3.0
3/4
8/9
160
120
2500
Synoptic Cu Load
180
140
2000
Distance Downstream (m)
Distance Downstream (m)
1.5
1.0
3/4
0
0.5
0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0.0
0
Distance Downstream (m)
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
Distance Downstream (m)
Dissolved
Total
Colloidal
Total Accumulated
Dissolved
Total
20
3/4
19/20
2/3
30
0.8
Fairday Mine
40
1.0
John Jay
Mine
50
Pb Load (kg/day)
John Jay
Mine
70
60
1.2
Fairday Mine
80
8/9
13/1
4
90
Zn Load (kg/day)
Total Accumulated
1.4
100
0.6
0.4
0.2
10
0.0
0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0
5000
500
1000
Dissolved
Total
Colloidal
Dissolved
Total Accumulated
Al
Mn
Fe
Cu
Zn
Pb
U
3/4
8/9
21/2
0.25
15/1
Synoptic U Load
0.30
John Jay
Mine
Fairday
Mine
0.20
0.15
0.10
0.05
0.00
0
500
1000
1500
2000
2500
3000
3500
4000
4500
Distance Downstream (m)
Dissolved
Total
Colloidal
1500
2000
2500
3000
3500
4000
4500
5000
Distance Downstream (m)
Distance Downstream (m)
U Load (kg/day)
Colloidal
Synoptic Pb Load
Synoptic Zn Load
Total Accumulated
5000
2/3:
3/4:
8/9:
13/14:
15/16:
19/20:
21/22:
22/23:
28/29:
Total
Colloidal
Total Accumulated
Inflow Concentrations (ppb)
Dissolved Metals
Total Metals
Fairday John Jay Fairday John Jay
2.2
0.0
50.4
181.1
146.7
4547.3
246.0
12249.8
263.0
7340.2
1377.8
13735.3
2.2
4.3
2.9
5.5
118.1
41.0
102.5
39.3
0.0
0.0
0.9
0.0
25.5
65.2
46.0
72.3
Possible Sources
Eroded Hill Side
Eroded Gulley
Road Built on Cobbles
Moist Ground / Loose Rock
Weir / Culvert / Moist Ground / Loose Rock
Fairday Mine? (Inflow conc's don't indicate high Cu or Pb inflow)
Fairday Mine? (Inflow conc's don't indicate high Cu or Pb inflow)
Dry Gulley
???
Figure 6b – Metal Loading and Accumulation Graphs (Upper James Creek) – Metal loading curves for total,
dissolved, and colloidal metals are given in blue, yellow, and green, respectively. Metal accumulation graphs,
generated by adding all the consecutive increases in metal loading, are shown in red.
15
Evidence of Precipitation
The observed decreases in metal loading that usually accompanied any observed
increases in loading suggested that precipitation of metals into the stream sediments might be
happening. In order to determine if this process could possibly have taken place, an area of
stream about 4.5 km downstream from the injection of the Lower James Creek (where the
unidentified spike in metals was observed) was examined for evidence of precipitation. Figure 7
shows pictures of the streambed taken in this vicinity. Picture (b), taken just downstream from
the observed spike in metals, appears to have distinctly lighter sediment than sites examined
either upstream or downstream from this location (pictures (a) and (c), respectively). This could
be visible evidence of aluminum precipitation into the creek. Although other metals could not be
observed visibly for precipitation, it is reasonable to conclude that the reason the metal loading
decreases downstream from this site is the result of precipitation. This assumption could also
hold for other areas of the Lower James Creek where increased metal loading is immediately
followed by decreased metal loading.
(a)
(b)
(c)
Figure 7 – Evidence of Precipitation: Pictures show a progressive downstream view of stream around the
unidentified metal source about 4.5 km downstream from the injection site; (a) was taken upstream from the source,
(b) was taken downstream from the source, where precipitation was expected to be taking place, and (c) was taken
downstream of where the decrease in metal loading was observed.
Conclusions
The Fairday Mine area, the unidentified zinc source at 2.7 km downstream of the upper
injection, and the tailings pile near the JWTP appear to be the primary sources of metals in
Upper James Creek. The Little James Creek and the unidentified inflow located about 4.5 km
downstream from the lower injection site appear to be greatest sources of metal loading into
Lower James Creek. The effects of metals entering the creek at these and other inflows are
16
spatially controlled as a result of metal precipitation. It is surprising that Elysian Park and John
Jay Mine were not as significant as metal contributors as was expected prior to experimentation.
This does not mean that the reclaimed tailings area and the mine do not pose a threat as metal
sources to James Creek. Our results may be limited in identifying all contributing metal sources
due to the extremely dry conditions under which the experiment was carried out.
At this point, it is too early to recommend appropriate remediation efforts for the James
Creek watershed, but future research will contribute even more knowledge about the sources and
consequence of metal loading in the James Creek watershed. Current research in the watershed
includes conducting similar tracer on the Little James Creek when flow conditions will permit,
and sedimentation analysis to study the attenuation phenomenon. The results from this would
provide more complete picture about the effects of mining on water quality in the James Creek
Watershed.
17
Works Cited
Colorado Water Quality Control Division, 2002. Draft Total Maximum Daily Load Assessment
– Little James Creek. Department of Public Health and Environment. 1-23.
Drever, J.I., 1997. The Geochemistry of Natural Waters. Prentice Hall, 306-309.
Duren, S., Gilbert, H., Miller, B., Taylor, S., Ryan, J., 2001. Investigating the Effect of OffRoad Vehicles on Turbidity in James Creek.” Presented to the Environmental Engineering
Seminar, Department of Civil, Environmental, and Architectural Engineering, University of
Colorado, Boulder.
Kimball, B.A., 1997. Use of tracer injections and synoptic sampling to measure loading from
acid mine drainage. Report FS-245-96, U.S. Geological Survey, 4 pp.
Kimball, B.A., Runkel, R.L., Walton-Day, K., Bencala, K.E., 2001a. Assessment of metal loads
in watersheds affected by acid mine drainage by using tracer injection and synoptic
sampling: Cement Creek, Colorado, USA. Applied Geochemistry, in press, 25 pp.
Kimball, B.A., Runkel, R.L., Gerner, L.J., 2001b. Quantification of mine-drainage inflows in
Little Cottonwood Creek, Utah, using a tracer-injection and synoptic sampling study.
Environmental Geology, 40, 1390-1404.
18