U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings
of the Technical Meeting, Colorado Springs, Colorado, September 20-24, 1993,
Water-Resources Investigations Report 94-4015
Past and Present Research on Metal Transport In
St. Kevin Gulch, Colorado
Briant A. Kimball (U.S. Geological Survey, WRD, Salt Lake City,
To prepare for mitigation of the effects of acidic mine drainage on upland
watersheds, process-oriented research at St. Kevin Gulch has focused on
chemical reactions that affect metal transport and partitioning among phases.
We operationally defined dissolved and colloidal transport phases and the
speciation of iron because the cycle of iron can affect the cycling of other
metals. With these phases defined, our approach was to study chemical reactions
in the context of hydrologic transport. By establishing the hydrologic setting
with tracer-dilution injections, we studied temporal and spatial variability
in metal concentrations resulting from combined hydrologic and chemical
processes. Temporal and spatial aspects of variation were combined in an
instream pH- modification experiment to evaluate the kinetics of reactions
involving metals. These data are the focus of a simulation model that combines
transport and reactive chemistry of metals. Ongoing studies of the ecological
system are in anticipation of higher pH and lower metal concentrations that
should folow remediation by a passive treatement system.
Many streams in upland watersheds are affected by acidic mine drainage
(Moore and Luoma, 1990). Efforts are underway to mitigate the effects of
this acidic drainage. However, successful mitigation requires an understanding
of the processes that affect metals in these streams. Since 1986, the Toxic
Substances Hydrology Program of the U.S. Geological Survey has supported
field-based research on processes that affect metals in streams. Objectives
of this research include: (1) The study of instream chemical and biological
processes in the context of transport processes that affect them, (2) determination
of the role of sediment and colloids in the processes affecting metals,
(3) quantification of the temporal and spatial variability of processes,
and (4) simulation of the reactive transport of metals to evaluate field
observations. The purpose of this report is to present the major findings
from past studies of St. Kevin Gulch and to describe the focus of present
St. Kevin Gulch is an upland watershed in the Rocky Mountains;
its elevation is 2,500 to 2,800 m (meters) above sea level and
it has an area of about 102km (square kilometers,
fig. 1). Most annual precipitation falls as snow; snowmelt runoff
that occurs in late May or early June is the dominant hydrologic
event each year. Bedrock is a quartz-biotite-feldspar schist.
Mining of silver and zinc sulfides in vein deposits mostly
occurred around 1920. St. Kevin Gulch has been a valuable
research site because of its combination of physical and chemical
characteristics: (1) Metal concentrations are sufficiently
high to be quantified by routine methods and yet not so
high as to be serious health and safety concerns, (2) a 100-m
reach is a point source of acidic inflow, (3) physical and
chemical processes that affect metal concentrations occur
downstream from the principal acidic inflows, (4) a single algal
species, Ulothrix sp., predominates (McKnight, 198S),
and (5) a natural wetland is present just upstream from
the confluence of St. Kevin Gulch and Tennessee Creek (fig.
Figure 1. Location of (a) St. Kevin Gulch,
near Leadville, Colorado, and (b) location of pH-modification experiment in
Downstream changes in the chemical characteristics and discharge in St.
Kevin Gulch result from acidic inflows (fig. 2). Upstream from the principal
acidic inflows (0-363 m, fig. 1) the chemical composition mostly results
from natural weathering in the basin. Between 363 m and 501 m, discharge
of the acidic inflows is only about 0.8 L/s (liters per second) compared
to 6 L/s in the stream, but the inflow chemistry strongly affects instream
pH and metal concentrations (figs. 2a and 2b). The confluence with the nonacidic
Shingle Mill Gulch is a short distance downstream from the acidic inflows,
at 501 m. At the confluence, discharge approximately doubles (fig. 2c) and
pH increases from the nonacidic inflow. Downstream from 526 m there only
are a few acidic inflows from small mines; these inflows do not substantially
change instream metal concentrations upstream from the wetland. The mass
loading of manganese (Mn) (fig. 2d) shows that the acidic inflows between
363 m and 501 m are virtually a point source to the stream.
Figure 2. Downstream profiles of (a) pH,
(b) manganese, (c) discharge, and (d) manganese mass flow in St. Kevin Gulch,
near Leadville, Colorado, August 1987.
We have taken advantage of the characteristics of St. Kevin Gulch to
conduct process-oriented research, studying well-known chemical reactions
in the context of hydrologic transport. Using tracer-dilution injections
to establish the hydrologic background, we have carried out experiments
with intensive spatial and temporal sampling (table 1).
Table 1. Summary of experiments in St. Keven Gulch, 1986 through 1993.
LiCl injection for 36-hour diel, synoptic sampling, and hydrologic characterization.
1. Documentation of iron photoreduction reaction (McKnight and others,
2. Hydrologic characterization (Broshears and others, 1993).
3. Initial steady-state simulation (Kimball and others, 1991).
NaCl injection at high flow.
1. Hydrologic characterization during snowmelt runoff (unpublished data).|
LiCl injection for synoptic sample for filtered and particulate concentrations.
1. Comparison of rates for hydrologic and biogeochemical processes; evaluation
of particulate concentrations (Kimball and others, 1994).|
LiCl injection for diel sampling; nighttime synoptic sampling; pH-modification
1. Mechanisms of iron photoreduction reaction (Kimball and others, 1992b).
2. Synoptic data set without effects of photoreduction (B.A. Kimball,
U.S. Geological Survey, 1988, unpublished data).
3. Temporal and spatial data on metal response to increasesd pH (Kimball
and others, 1992a; Kimball and others, 1994).
4. Stream-side sorption experiments (Smith and others, 1991).
Multiple tracer injections to define loss of streamflow.
1. Quantification of losing reach (Zellweger and Maura, 1991).|
Hillslope interactions; injection of radioactive phosphorous for identification
of nutrient pathways.
1. Water exchange between stream and alluvium (Harvey and Bencala, 1993).
2. Phosphate uptake (Tate and others, 1991).
Seasonal sampling at fixed sites using natural conservative tracers.
1. Seasonal variation of metal concentrations (B.A. Kimball, U.S. Geological
Survey, 1990, unpublished data).|
Hillslope interactions; diel sampling of streamwater and alluvial water.
1. Effects of substream on diel patterns (B.A. Kimball, U.S. Geological
Survey, 1991, unpublished data).|
Effects of alluvium on instream pH modification.
1. (B.A. Kimball, U.S. Geological Survey, 1993, unpublished data).|
Methods for Assessing Iron-Rich Systems
Particular methods have been used to define processes that affect metals
in acidic mine drainage. We will review some of these methods and then discuss
how these methods aid the study of instream processes.
Setting the hydrologic framework--Use of experimentally injected
chemical tracers helps to define relevant physical characteristics of a
stream With the physical characteristics defined, it is possible to distinguish
between physical and chemical effects on instream metal concentrations (Stream
Solute Workshop, 1990; Bencala and others, 1990; Broshears and others, 1993).
However, a seasonal study has a greater temporal scale, and it is not always
practical to inject tracers to set the hydrologic framework. In such cases,
natural conservative solutes can substitute for injected tracers (Bencala
and others, 1987). Although this procedure does not establish absolute values
of discharge as does the tracer-dilution method, it accounts for the hydrologic
effects by a discharge ratio so that chemical effects can be studied.
Defining trasport phases--In addition to defining the hydrologic
framework, it is necessary to define transport phases for the metals. The
cycle of iron (Fe) affects many metals and is strongly affected by precipitation
(Kimball and others, 1991) and by photoreduction (McKnight and others, 1988;
Kimball and others, 1992b). At pH values greater than about 2.2, Fe oxyhydroxides
commonly precipitate to form colloids, which affect the cycling of other
metals by sorption (Smith and others, 1991). Colloids in St. Kevin Gulch
can contain as much as 130 ppm (parts per million) arsenic (As), 230 ppm
copper (Cu), 600 ppm lead (Pb), and 1,200 ppm zinc (Zn). However, Ranville
(1992) determined that most of the suspended sediments in St. Kevin Gulch
are primarily aggregates of colloidal (40-nanometer diameter) Fe oxyhydroxides
and Fe oxyhydroxysulfates. In opposition to precipitation, photoreduction
of Fe III can dissolve colloids and return ferrous iron (Fe II) to the stream
(McKnight and others, 1988). Other metals that may be sorbed to the colloidal
Fe also can be released. This dynamic cycling of Fe on a daily time scale
affects the transport and transformation of other metals and can be studied
only by adequate sampling for Fe phases and species.
Operational definitions of phases can be defined by using multiple filtrations.
The definitions that are most meaningful in terms of the Fe colloids include
the following: (1) filtered concentration (representing "dissolved"
solutes) determined by filtration through a 0.001-µm (micrometer)
pore-size membrane, (2) colloidal concentration, determined by filtration
through a 0.45-µm pore-size membrane and then subtracting the dissolved
concentration, and (3) suspended particulate concentration determined by
an unfiltered sample and then subtracting the colloidal and dissolved concentrations.
These definitions are time consuming to obtain when we could not take the
time for the sequence of filtrations. For intensive temporal and spatial
sampling, we generally obtained only an unfiltered and a 0.1-µm filtered
sample (McKnight and others, 1988; Kimball and others, 1992a).
With these definitions of transport phases, we have described instream
processes affecting metals in the context of transport.
Photoreduction of iron--Intensive temporal sampling showed the
importance of the photoreduction process on the cycling of Fe and other
metals. Quantification of discharge allowed the calculation of mass flow
and flux of Fe II from the streambed to the water column in response to
light intensity (see fig. 3 in Kimball and others, 1992b).
Figure 3. Variation of sampled and simulated
(a) filtered iron and (b) colloidal iron with downstream distance in St. Kevin
Gulch, near Leadville, Colorado.
Coupling of rates for hydrologic and chemical processes--Intensive
spatial sampling along a 1,800-m reach allowed us to compare rates of transport
to rates of chemical reaction by using steady-state solute transport simulation
(fig. 3). If chemical reactions are relatively fast compared to rates of
transport, the reactions can affect instream concentrations. The solid line
in figure 3a illustrates conservative transport of Fe. Adding first-order
rate constants to simulate Fe removal indicates the relative importance
of chemical reaction (dashed line, fig. 3a). The increase of colloidal Fe
(fig. 3b) corresponds to the decrease of filtered Fe. Loss of colloidal
Fe (dashed line simulation, fig. 3b) is from sedimentation of colloidal
Response to pH modification--Combining intensive temporal and
spatial sampling, we modified the chemistry of St. Kevin Gulch to doeument
the kinetics of meta1 reactions as pH was increased in two steps (fig. 4a).
With an increase of pH to near 6.0, aluminum (Al) was completely partitioned
from the filtered to the co11oida1 phases in the water column (fig. 4b).
This is likely from the rapid formation of an Al oxyhydroxysulfate complex
(Kimball and others, 1994). Sorption onto colloids affected the concentrations
of Cu. However, Mn and Zn were little affected at this level of pH (Kimball
and others, 1994).
Figure 4. Variation of (a) pH and (b)
filtered and colloidal aluminum with time during pH-modification experiment
in St. Kevin Gulch, near Leadville, Colorado, August 1988.
Exchange of streamwater and subsurface water--By injecting tracers
both in the stream and in the hillslope, the exchange of water between the
stream and the alluvium was documented (Harvey and Bencala, 1993). This
changes the traditional concept of the stream as a pipe of water leaving
a watershed. Instead, the stream continues to interact with the watershed,
which can affect instream metal concentrations because metals are present
in substantial concentrations in the alluvium (J.W. Harvey and B.A. Kimball,
U.S. Geological Survey, 1991, unpublished data, Salt Lake Oty, Utah). The
scale of this interaction can be on the order of 1-m stream segments.
Treatment of metals by natural wetlands--Before entering Tennessee
Creek, the metal-rich water from St. Kevin Gulch passes through a natural
wetland. Initial sampling of inflow and outflow water from this wetland
indicated that metals were being treated by interaction with organic matter
and minerals of the wetland. Metal flux and geochemical processes in this
wetland have been studied by Walton-Day (1992) to determine whether the
wetland is removing Fe, Mn, cadmium (Cd), Cu, Pb, and Zn from surface water
flowing through the wetland. Careful measurement of metal fluxes indicates
that only Fe was removed from surface water. Most metals were untreated
by the wetland, and essentially passed through despite prolonged physical
contact with the wetland environment.
In our current work, we have used mathematical models to provide a frame
of reference for evaluating data from field studies. We developed computer
simulations for conservative and reactive solute transport in St. Kevin
Gulch. Reactive transport simulation takes the study of biogeochemical processes
beyond batch experiments and places it in the context of hydrology.
Simulation of conservative transport--Physical effects on the
transport of solute mass must be quantified to distinguish the chemical
and biological effects. Simulation of transport in mountain streams presents
certain unique physical aspects, the most prominent being transient storage
(Bencala and Walters, 1983). The transport model developed by Bencala and
Walters (1983) was improved by using a new algorithm and by execution on
new generations of computers (Runkel and Broshears, 1992; Runkel and Chapra,
1993). These improvements were the basis for building the simulation model
for reactive solute transport.
Simulation of reactive solute transport--Data from the pH-modification
experiment in 1988 (as in fig. 4) provide a unique opportunity to evaluate
our understanding of rates for metal reactions. The complexity of reactive
chemistry is simulated by coupling MINTEQA2 (Allison and others, 1991) with
the transport code (Runkel and others, 1996; Broshears, 1996).
Our understanding of geochemical reactions in St. Kevin Gulch has provided
the background to study remediation of acidic mine drainage. The State of
Colorado plans to install a passive treatment system near the mine dump
in St. Kevin Gulch, creating the opportunity to study the changes that occur
in the recovering stream. Chemical changes in the water column, in the bed
sediment, and in the alluvium will combine to affect the stream ecosystem.
The current dominance of the blue-green algae, Ulothrix sp. (McKnight,
1988) will be affected by chemical changes, indicating the ecosystem response.
Combining this remediation study with our past and present studies will
complete a comprehensive process-oriented study at the St. Kevin Gulch site.
- Allison, J.D., Brown, D.S., and Novo-Gradac, K.J., 1991,
- MINTEQ2/PRODEFA2, a geochemical assessment model for environmental
systems-Version 3.0 User's Manual: EPA/600/3-91/021, U.S. Environmental
Protection Agency, 106 p.
- Bencala, K.E., McKnight, D.M., and Zellweger, G.W., 1987,
- Evaluation of natural tracers in an acidic and metal-rich stream: Water
Resources Research, v. 23, p. 827-836.
- Bencala, K.E., McKnight, D.M., and Zellweger, G.W., 1990,
- Characterization of transport in an acidic and metal-rich mountain
stream based on a lithium tracer injection and simulations of transient
storage: Water Resources Research, v. 26, p. 989-1000.
- Bencala, K.E., and Walters, R.A., 1983,
- Simulation of solute transport in a mountain pool-and-riffle stream-A
transient storage model: Water Resources Research, v. 19, p. 718-724.
- Broshears, R.E., Bencala, K.E., Kimball, B.A., and McKnight, D.M.,
- Tracer-dilution experiments and solute-transport simulations for a
mountain stream, Saint Kevin Gulch, Colorado: U.S. Geological Survey Water-Resources
Investigations Report 92-4081, 18 p.
- Broshears, R.E., Kimball, B.A., and Runkel, R.L., 1996,
- Simulation of reactive transport during a pH modification experiment
in a mountain stream affected by acid mine drainag, in Morganwalp, D.W.,
and Aronson, D.A., ed., U.S. Geological Survey Toxic Substances Hydrology
Program-Proceedings of the technical meeting, Colorado Springs, Colorado,
September 20-24, 1993: U.S. Geological Survey Water-Resources Investigations
- Harvey, J.W. and Bencala, K.E., 1993,
- The effect of streambed topography on surface-subsurface water exchange
in mountain catchments: Water Resources Research, v. 29, no. 1., p. 89-98.
- Kimball, B.A., Broshears, R.E., Bencala, K.E., and McKnight, D.M.,
- Comparison of rates of hydrologic and chemical processes in a stream
affected by acid mine drainage, in Mallard, G.E., and Aronson, D.A., eds.,
U.S. Geological Survey Toxics Substances Hydrology Program-Proceedings
of the technical meeting, Monterey, California, March 11-15, 1991: U.S.
Geological Survey Water-Resources Investigations Report 91-4034, p. 407-412.
- Kimball, B.A., Broshears, R.E., McKnight, D.M., and Bencala, K.E.,
- Preliminary evaluation of the effects of an instream pH modification
on transport of sulfide-oxidation products, in Alpers, C.L, and Bowles,
D., eds., The Environmental Geochemistry of Sulfide Oxidation: Washington,
D.C., American Chemical Society Books, p. 224-243.
- Kimball, B.A., McKnight, D.M., Broshears, R.E., and Bencala, K.E.,
- Effect of instream pH modification on aluminum, in Kharaka, Y. K. and
Maest, A. S., eds., Water-Rock Interaction, Volume 1, Low Temperature Environments:
Rotterdam, A.A. Balkema, p. 393-396.
- Kimball, B.A., McKnight, D.M., Wetherbee, G.A., and Harnish, R.A.,
- Mechanisms of iron photo reduction in a metal-rich, acidic stream (St.
Kevin, Gulch, Colorado, U.S.A.): Chemical Geology, v. 96, p. 227-239.
- McKnight, D.M., 1988,
- Metal-tolerant algae in St. Kevin Gulch, Colorado, in, Mallard, G.E.,
ed., U.S. Geological Survey Toxic Substances Hydrology Program surface-water
contamination-Proceedings of the technical meeting, Denver, Colorado, February
2-4, 1987: U.S. Geological Survey Open-File Report 87-764, p. 113-117.
- McKnight, D.M., Kimball, B.A., and Bencala, K.E., 1988,
- Iron photoreduction and oxidation in an acidic mountain stream: Science,
v. 240, p. 637-640.
- Moore, J.N. and Luoma, S.N., 1990,
- Hazardous wastes from large-scale extraction: Environmental Science
& Technology, v. 24, p. 1278-1285.
- Ranville, J.F., 1992,
- Factors influencing the electrophoretic mobility of suspended sediments
in acid mine drainage: Golden, Colorado School of Mines, unpublished Ph.D.
dissertation T-4213, 270 p.
- Runkel, R.L., and Broshears, R.E., 1992,
- One-dimensional transport with inflow and storage (OTIS)-A solute transport
model for small streams: Boulder, Colo., Center for Advanced Decision Support
for Water and Environmental Systems, 85 p.
- Runkel, R.L. and Chapra, S.C., 1993,
- An efficient numerical solution of the transient storage equations
for solute transport in small streams: Water Resources Research v. 29,
- Runkel, R.L., Bencala, K.E., and Broshears, R.E., 1996,
- An equilibrium-based simulation model for reactive solute transport
in small streams, in Morganwalp, D.W., and Aronson, D.A., ed., U.S. Geological
Survey Toxic Substances Hydrology Program-Proceedings of the technical
meeting, Colorado Springs, Colorado, September 20-24, 1993: U.S. Geological
Survey Water-Resources Investigations Report 94-4015.
- Smith, K.S., Ranville, J.F., and Macalady, D.L., 1991,
- Predictive modeling of copper, cadmium, and zinc partitioning between
streamwater and bed sediment from a stream receiving acid mine drainage,
St. Kevin Gulch, Colorado, in Mallard, G.E., and Aronson, D.A., eds., U.S.
Geological Survey Toxic Substances Hydrology Program-Proceedings of the
technical meeting, Monterey, California, March 15, 1991: U.S. Geological
Survey Water-Resources Investigations Report 91-4034, p. 380-386.
- Stream Solute Workshop, 1990,
- Concepts and methods for assessing solute dynamics in stream ecosystems:
Journal of the North American Benthological Society, v. 9, p. 95-119.
- Tate, C.M., McKnight, D.M., and Spaulding, S.A., 1991,
- Phosphate uptake by algae in a stream contaminated by acid mine drainage,
St. Kevin Gulch, Leadville, Colorado, in Mallard, G.E., and Aronson, D.A.,
eds., U.S. Geological Survey Toxic Substances Hydrology Program-Proceedings
of the technical meeting, Monterey, California, March 11-15, 1991: U.S.
Geological Survey Water-Resources Investigations Report 91-4034, p. 387-391.
- Walton-Day, K., 1992,
- Hydrology and geochemistry of a natural wetland affected by acid mine
drainage affected by acid mine drainage St. Kevin Gulch, Lake County, Colorado:
Golden, Colorado School of Mines, unpublished Ph.D. dissertation T-4033,
- Zellweger, G.W. and Maura, W.S., 1991,
- Calculation of conservative-tracer and flume-discharge measurements
on a small mountain stream, in Mallard, G.E., and Aronson, D.A., eds.,
U.S. Geological Survey Toxic Substances Hydrology Program-Proceedings of
the technical meeting, Monterey, California, March 11-15, 1991: U.S. Geological
Survey Water-Resources Investigations Report 91-4034, p. 434-437.