EPA 540/R-09/002
September 2008
Mine Waste Technology Program
Activity III, Project 34
Bioremediation of Pit Lakes -
Gilt Edge Mine
By:
Brian Park
MSB Technology Applications, Inc.
Mike Mansfield Advanced Technology Center
Butte, Montana 59702
Under Contract No. DE-AC09-96EW96405
Through EPA IAG No. DW89938870-01-0
Diana Bless, EPA Project Officer
Sustainable Technology Division
National Risk Management Research Laboratory
Cincinnati, Ohio 45268
This study was conducted in cooperation with
U.S. Department of Energy
Environmental Management Consolidated Business Center
Cincinnati, Ohio 45202
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
TM
Geochemical And
Redox-Mediated Bio-Transformation
for
In Situ Pit Lake Remediation Of
Acid- And Metal-Toxic Mine Drainage
2001-2006 Treatability Study
Report of Results
ANCHOR HILL PIT LAKE
GILT EDGE MINE NPL SITE, SOUTH DAKOTA
-------
Disclaimer
This publication is a report of work conducted under the Mine Waste Technology Program that was
funded by the Environmental Protection Agency and managed by the Department of Energy under the
authority of an Interagency Agreement.
Because the Mine Waste Technology Program participated in EPA's Quality Assurance Program, the
project plans, laboratory sampling and analyses, and final report of all projects were reviewed to ensure
adherence to the data quality objectives. The views expressed in this document are solely those of the
performing organization. The views and opinions of authors expressed herein do not necessarily state or
reflect those of the United States Government or any agency thereof.
Reference herein to any specific commercial product, process, or service by trade name, trademark,
manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or
favoring by the United States Government or any agency thereof or its contractors or subcontractors.
Neither the United States Government nor any agency thereof, nor any of their employees, nor any of
their contractors, subcontractors, or their employees, makes any warranty, express or implied, or assumes
any legal liability or responsibility for the accuracy, completeness, or any third party's use or the results
of such use of any information, apparatus, product, or process disclosed, or represents that its use would
not infringe privately owned rights.
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's
land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to
formulate and implement actions leading to a compatible balance between human activities and the ability
of natural systems to support and nurture life. To meet this mandate, EPA's research program is
providing data and technical support for solving environmental problems today and building a science
knowledge base necessary to manage our ecological resources wisely, understand how pollutants affect
our health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center for investigation
of technological and management approaches for preventing and reducing risks from pollution that
threaten human health and the environment. The focus of the Laboratory's research program is on
methods and their cost-effectiveness for prevention and control of pollution to air, land, water, and
subsurface resources; protection of water quality in public water systems; remediation of contaminated
sites, sediments, and groundwater; prevention and control of indoor air pollution; and restoration of
ecosystems. NRMRL collaborates with both public and private sector partners to foster technologies that
reduce the cost of compliance and to anticipate emerging problems. NRMRL's research provides
solutions to environmental problems by developing and promoting technologies that protect and improve
the environment; advancing scientific and engineering information to support regulatory and policy
decisions; and providing the technical support and information transfer to ensure implementation of
environmental regulations and strategies at the national, state, and community levels.
This project was conducted under the Mine Waste Technology Program. It was funded by the EPA and
administered by the U.S. Department of Energy (DOE) in cooperation with various offices and
laboratories of the DOE and its contractors. It is made available at www.epa.gov/minewastetechnology
by EPA's Office of Research and Development to assist the user community and to link potential users
with the researchers.
Sally Gutierrez, Director
National Risk Management Research Laboratory
in
-------
Abstract
The U.S. Environmental Protection Agency (EPA) Region 8 Superfund Office and the EPA National Risk
Management Research Laboratory (NRMRL) Mine Waste Technology Program (MWTP) conducted a
field-scale treatability study demonstrating an in situ bio/geochemical treatment technology for
decontaminating acid/metal-toxic water within the Anchor Hill Pit lake at the Gilt Edge Mine Superfund
site near Deadwood, South Dakota. The purpose of the project, carried out between March 2001 and May
2006, was to develop performance data of the treatment approach for potential use in long-term water
treatment/management activities at the Gilt Edge site, as well as other similar sites. The treatment process
was applied to approximately 72 million gallons of acidic water, with high concentrations of metals
(including iron, aluminum, arsenic, selenium, copper, cadmium, and zinc), sulfate, and nitrate, and the pH
was approximately 3.
The treatment process involved pit neutralization, then application of nutrients to stimulate biological
activity. The treatment process was successful and approximately 40 million gallons of treated water that
met the State of South Dakota's strict surface water discharge standards were discharged from the Anchor
Hill Pit during the demonstration. All project objectives were met, and considerable experience and
insight was gained into how operational aspects of such a remediation technique would have to be
designed for future efforts.
IV
-------
Contents
Page
Disclaimer ii
Foreword iii
Abstract iv
Contents v
Figures vi
Tables vi
Acronyms and Abbreviations vii
Acknowledgments viii
Executive Summary E-l
1. Introduction 1
1.1 Background 1
1.2 Treatment Description 1
2. Project Chronology 3
3. Implementation 6
3.1 Sampling 6
3.2 Treatment Methods/Materials 7
3.2.1 Lime Addition 7
3.2.2 Nutrient Addition-May 2001 7
3.2.3 Nutrient Addition- September 2001 8
3.2.4 Nutrient and Caustic Addition - April 2002 8
3.2.5 Nutrient, Caustic, and Wood Chip Addition - September 2002 8
3.2.6 Fertilizer Addition 9
3.2.7 Discharge of Deep Zone Water, 2004 9
3.2.8 Evaluation of Use of Anchor Hill Pit for Ongoing Site Water Treatment, 2004.. 9
3.2.9 Evaluation of Likelihood of Lake Turnover, 2004 10
3.2.10 Discharge of Surface Water, 2005 10
3.2.11 Discharge of Surface Water, 2006 11
4. Results/Discussion 15
5. Lessons Learned/Quality Assurance Activities 21
5.1 Neutralization 21
5.2 Nutrient Addition 22
5.3 Monitoring 22
5.4 Discharge 23
5.5 Quality Assurance Activities 23
5.5.1 Field Data Review 23
5.5.2 Laboratory Data 23
6. Conclusions/Recommendations 25
7. References 26
-------
Contents (Cont'd)
Page
Appendix A: Technical Systems Review Response from MSB to EPA A-l
Appendix B: Evaluation of the Use of Anchor Hill Pit for Ongoing Site Water Treatment B-l
Appendix C: Evaluation of Options for Eliminating Dissolved Sulfide, and Subsequent Addition of
Concentrated Hydrogen Peroxide C-l
Appendix D: Evaluation of Likelihood of Water Column Turnover D-l
Figures
1-1. 300-acre Gilt Edge Mine site (left); Anchor Hill Pit lake (center & right) 2
3 -1. Photograph indicating the location of sample buoys in the Anchor Hill Pit 12
3-2. Camera used for video camera observations (left), gas bubbles within suspended bags of chips
(center), and sludge at bottom of pit (right) 12
3-3. Photograph showing Neutra-Mill docking location 12
3-4. Adding molasses (left) and methanol (right) to the pit 13
3 -5. Photograph showing white froth of gas bubbles within the floating wood chips 13
3-6. Photograph showing in situ hydrogen peroxide addition 13
3-7. Photograph showing evidence of excess sulfur in the shallow pit water after hydrogen peroxide
addition 14
4-1. pH versus depth priorto pit lake neutralization 16
4-2. pH profiles with depth after initial lime neutralization 17
4-3. pH profile following additional efforts to neutralize the Anchor Hill Pit 18
4-4. Nitrate data at each sample location throughout the demonstration 18
4-5. Anchor Hill Pit nitrate/nitrite and dissolved copper, cadmium, and zinc concentrations over the
project duration 19
Tables
1-1. Initial Pit Water Composition and ARARs for the Gilt Edge Mine Site 2
2-1. Chronology of Project Execution 3
4-1. Water Quality Improvement Over the Course of the Project 20
VI
-------
ARARs
ARD
BOD
BOR
CLP
DOE
EPA
GWS
IAG
MSB
MWTP
NRMRL
ORP
pH
QA
QAPP
RMBT
SC
SRB
TSS
-,TM
Acronyms and Abbreviations
applicable or relevant and appropriate requirements
acid rock drainage
biochemical oxygen demand
Bureau of Reclamation
Contract Laboratory Program
U.S. Department of Energy
U.S. Environmental Protection Agency
Green World Science, Inc.
Interagency Agreement Number
MSE Technology Applications, Inc.
Mine Waste Technology Program
National Risk Management Research Laboratory
oxidation-reduction potential
negative log of hydrogen ion concentration
quality assurance
quality assurance project plan
Redox-Mediated Biotransformation
specific conductance
sulfate-reducing bacteria
total suspended solids
vn
-------
Acknowledgments
This document was prepared by MSB Technology Applications, Inc. (MSB) for the U.S. Environmental
Protection Agency's (EPA) Mine Waste Technology Program (MWTP) and the U.S. Department of
Energy's (DOE) Environmental Management Consolidated Business Center. For this project, Ms. Diana
Bless was EPA's MWTP Project Officer, while Ms. Norma Lewis was EPA's MWTP Project Manager.
Mr. Gene Ashby was DOE's Technical Program Officer. Roger Wilmoth was instrumental in accepting
this project from EPA Region 8. Ken Wangerud was the site's Remedial Project Manager for EPA
Region 8. Ms. Helen Joyce was MSB's MWTP Program Manager and Brian Park was MSB's Project
Manager. Additional project-specific contributors that deserve acknowledgement include: Steve
Fundingsland, Jim Jonas, Paul Hight, and Marko Adzic (COM), Jim Harrington (Shepherd-Miller), and
Joe Harrington (Green World Science).
Vlll
-------
Executive Summary
The U.S. Environmental Protection Agency (EPA) Region 8 Superfund Office and the EPA National Risk
Management Research Laboratory (NRMRL) Mine Waste Technology Program (MWTP) conducted a
field-scale treatability study demonstrating an in situ bio/geo-chemical treatment technology for
decontaminating acid/metal-toxic water within the Anchor Hill Pit lake at the Gilt Edge Mine Superfund
site near Deadwood, South Dakota. The purpose of the project, carried out between March 2001 and May
2006, was to develop performance data on the treatment approach for potential use in long-term water
treatment/management activities at the Gilt Edge site, as well as other similar sites. The treatment process
was applied to approximately 72 million gallons of acidic water, with high concentrations of metals
(including iron, aluminum, arsenic, selenium, copper, cadmium, and zinc), sulfate, and nitrate, and a pH
of approximately 3.
The original objectives for the treatability study were to:
- successfully establish anaerobic and chemical-reducing conditions to reduce nitrate- and metal-
sulfate contaminants; and
- reduce toxic-metal concentrations.
As these objectives were successfully achieved, and the effectiveness of the treatment was evident, an
additional objective sought was to:
- achieve discharge standards and release "clean" and "non-toxic" water from Anchor Hill Pit lake.
All of these objectives were met, and considerable experience and insight were gained into how
operational aspects of such a remediation technique would have to be designed for future efforts.
The first step of the planned two-stage treatment consisted of using a Neutra-Mill (essentially a floating
lime slaker developed by Earth Systems, Pty. of Australia), to neutralize the pit water pH to
approximately 7 using lime (March 2001-May 2001). Following a short stabilization period, during
which the pH "settled" to a value of approximately 5.0, a patented process for in situ treatment, using an
organic formulation of molasses, methanol, and proprietary ingredients was implemented as the second
treatment step in May 2001 by Green World Science, Inc. The purpose of this step was to induce
reducing conditions, and stimulate bacterial activity for nitrate, selenium, and sulfate reduction to create a
stable system and improve water quality.
In 2002, the project team concluded that the pH 5.0 condition was inhibiting biological activity so the pH
was raised to 6.0 and wood-chips were added as substrate for bacterial growth. Robust bacterial activity
rapidly proceeded. By the summer of 2003, denitrification was complete and sulfate reduction was well
under way. The dissolved form of metals which form sulfide precipitates (e.g., copper, cadmium, zinc)
had decreased dramatically upon the onset of sulfate reduction. Due to the slow-settling nature of these
metal sulfide particles, the total metals values of those metals in collected samples were significantly
higher. In general, water in the pit met applicable South Dakota Ambient Water Quality Standards
(SDAWQS) with the exception of undisassociated hydrogen sulfide (H2S) (i.e., dissolved H2S as opposed
to HS") and biochemical oxygen demand (BOD); the elevated BOD was due to the presence of remaining
organic carbon as well as the elevated dissolved sulfide. In addition, the Anchor Hill Pit had become
meromictic, with a chemocline at a depth of 20 to 30 feet. During the year leading to the summer of
2004, the pit water column became more strongly stratified, with the surface zone being aerated and
ES-1
-------
meeting all applicable water quality standards, and the deep zone having increased dissolved H2S (-50
mg/L) due to continuing sulfate reduction.
While discharge of surface-waters could have begun, it was recognized that the strongly-reducing
conditions in the deeper zone might have attributes worth maintaining for additional acid rock drainage
(ARD) loading and treatment. It seemed important to better understand the deep-water chemistry and
treatment conditions. Because the strongly-reducing deep water condition was believed to be the result of
overdosing the carbon-nutrient compounds, it was surmised that the most effective metal reduction and
removal could be accomplished by maintaining this reaction-zone with strong reducing conditions. The
surface water was thought to be more vulnerable to perturbations, such as contaminated surface runoff
into the pit, which might easily result in metals such as cadmium or zinc increasing above discharge limit,
with no possibility of decontamination by sulfide precipitation - as would occur in the deeper zone. The
deeper zone was, in general, considered to be more stable and controllable; consequently, additional
ARD-loading for treatment might be more efficient because reducing conditions already existed, making
it unnecessary to add additional carbon to consume dissolved oxygen and establish anoxic conditions. It
was envisioned that in the future, contaminated water from the Gilt Edge Mine site might be injected,
along with nutrients, into the pit below the stable chemocline, with the surface layer simply serving as the
"protective" layer over what would be the "treatment zone". Certainly if this approach were pursued, the
relative densities of the deep zone water and the contaminated water added for treatment would have to be
considered. The decision was made to focus on discharging water from the deep zone. The elevated H2S
present in the deep water posed health and safety concerns, which were addressed and managed.
Mitigation of the deep water chemistry was attempted by pumping the deep zone waters through an air-
sparging unit, followed by a shallow holding-lagoon to complete BOD reduction and solids settling.
Since the residual H2S oxidized to elemental sulfur and formed colloidal particles that were very slow to
settle, attainment of regulated values of total suspended solids (TSS) was difficult. Interestingly, metals
were not remobilized by this sulfide oxidation, and approximately 150,000 gallons of water was
successfully discharged in two separate batches in October 2004.
Even though discharge-quality water was achieved, it was apparent that the time and process phases,
along with the area requirements for "polishing-lagoons," made this approach impractical. The notion of
operationalizing a treatment system that could be in-loading ARD to a deep reaction zone was deemed to
be fraught with too many uncertainties to warrant further consideration. In addition, the team concluded
that 1) any operationalized treatment process should assume that high-H2S conditions could likely occur
during each treatment "batch", 2) that a method for rapidly mitigating such a condition be sought, and 3)
attention should be turned to eliminating the high-H2S deep-water concentrations in a "single-step"
approach of in situ oxidation. The addition of various oxidants was considered, and hydrogen peroxide
was selected as the preferred method to eliminate the dissolved H2S. Thirty-five thousand gallons of
hydrogen peroxide was released into the pit below the chemocline in August 2005. The reaction was
successful, and as had been observed in the oxidation-lagoons, the lake experienced a milky-gray
coloration as elemental sulfur formed and began to settle.
In late-fall 2005 and spring 2006, large-volume skimming and decanting of surface-waters meeting water-
quality standards was carried out, resulting in the release of approximately 40-million gallons of treated
water into Strawberry Creek, which runs adjacent to the Gild Edge Mine site. Sampling had shown that
approximately 25-million additional gallons of dischargeable water was present in the treated pit lake,
when very large spring runoff and storm events required an emergency transfer of ARD-inventory from
the Gilt Edge Mine site into the Anchor Hill pit, functionally ending the treatability test in late
spring 2006.
ES-2
-------
1. Introduction
1.1 Background
This document is the final report for Mine Waste
Technology Program (MWTP), Activity III,
Project 34, Bioremediation of Pit Lakes (Gilt Edge
Mine). The MWTP is a program funded by the
U.S. Environmental Protection Agency (EPA) and
jointly administered by EPA and the U.S.
Department of Energy through an Interagency
Agreement (IAG). MSE Technology
Applications, Inc. (MSE) is the principal
contractor for the MWTP.
The EPA Region 8 Superfund office and the EPA
National Risk Management Research Laboratory
(NRMRL) MWTP, through MSE, conducted a
field-scale technology demonstration of an in situ
treatment of the Anchor Hill Pit lake at the Gilt
Edge Mine Superfund site near Deadwood, South
Dakota from March of 2001 through May 2006.
The project goal was to develop performance data
of the treatment approach for potential application
in long-term water treatment/management
activities at the Gilt Edge site, as well as potential
application at other similar sites. EPA's interest in
the in situ treatment process was to determine if
"semi-passive" treatment of acid rock drainage
(ARD) was possible in a pit lake, thereby avoiding
some portion of the costs of operating a
conventional water treatment plant.
In addition to summarizing the execution of the
project over an approximately five-year period,
this report also summarizes what was learned over
the course of the project, and provides a
description of how a similar effort might be
conducted in light of the lessons learned during the
execution of this project. Figure 1-1 (left) shows
an aerial photograph of the Gilt Edge Mine site,
Figure 1-1 (center) shows an aerial photograph of
the Anchor Hill Pit lake, and Figure 1-1 (right)
shows a contoured drawing of the Anchor Hill
Pit lake.
Prior to treatment, the Anchor Hill Pit contained
approximately 72 million gallons of acidic water,
with elevated metals, sulfate, selenium, and nitrate
content, and a pH of approximately 3. Table 1-1
presents the initial water composition for samples
collected in March 2001 prior to pit neutralization.
The table also presents the applicable or relevant
and appropriate requirements (ARARs) for the
Gilt Edge Mine site that would have to be
achieved to discharge water to Strawberry Creek.
1.2 Treatment Description
The treatment approach was originally envisioned
to consist of two steps that would occur within the
pit lake: (1) neutralize the pit water to near-
neutral pH using lime (CaO), applied using a
Neutra-Mill (essentially a floating lime slaker,
developed by Earth Systems, Pty. of Australia);
and (2) utilize Redox-Mediated Biotransformation
(RMB™) technology, developed and patented by
Green World Science, Inc. (GWS), to create
reducing conditions; stimulate bacterial activity
for nitrate, selenium, and sulfate reduction;
improve water quality; and create a stable system.
The RMB™ process involved addition of nutrients
to the pit, including methanol, animal feed-grade
molasses, and phosphoric acid, with the goal of
stimulating indigenous bacterial activity to first
reduce nitrate to nitrogen gas, and subsequently
reduce selenium and sulfate. Bacterial reduction
of selenium to its elemental state removes it from
solution, and reduction of sulfate produces sulfide,
which forms metal sulfide particles with copper,
cadmium, nickel, lead, and zinc. These metals are
reduced to dischargeable concentrations at neutral
pH via precipitation as metal sulfides. The metal
sulfides were intended to settle to the bottom of
the pit, where a permanent anoxic zone would be
maintained, ensuring their long-term stability.
-------
Figure 1-1. 300-acre Gilt Edge Mine site (left); Anchor Hill Pit lake (center & right).
Table 1-1. Initial Pit Water Composition and ARARs for the Gilt Edge Mine Site
Parameter
Unionized Ammonia as N
Dissolved Oxygen
PH
Total Suspended Solids
(TSS)
Acidity as CaCO3
Chloride
Nitrate/Nitrite as N
Sulfate
Contaminant
Dissolved Aluminum
Dissolved Antimony
Dissolved Arsenic
Dissolved Cadmium
Dissolved Calcium
Dissolved Chromium III
Dissolved Chromium VI
Dissolved Copper
Dissolved Iron
Dissolved Lead
Dissolved Magnesium
Dissolved Manganese
Dissolved Mercury
Dissolved Selenium
Dissolved Silver
Dissolved Zinc
Cyanide (weak-acid
dissociable)
Average Concentration +
Standard Deviation (n=4 unless
otherwise noted) (|ig/L unless
otherwise noted)
No data available
9.0+0.90 (n=3 shallow samples)
3.1SU+0.07(n=29)
6+1 .2 mg/L
11 15+1 10 mg/L
37.3+3.3 mg/L
85. 5+15 mg/L
3260+30 mg/L
224,000+19,000
17+0.5
73+3
576+25
506,000+27,000
No data available
<500
43,300+2500
15,600+
31+2.3
196,000+6800
27,100+680
0.1
26+1.6
<1.3
14,100+400
No data available
South Dakota ARARs Criteria
<0.02 mg/L
>5.0 mg/L
>6.5-<8.8 S.U.
<10mg/L
N/A
N/A
<50 mg/L
N/A
Acute Aquatic Life Value (|ig/L)
N/A
N/A
360
3
N/A
1708
15
63
N/A
281
N/A
N/A
2.1
20
37.4
370
22
Special Conditions
30-day average
30-day average
24-hr composite
30-day average
Hardness*
Hardness*
Hardness*
Hardness*
Hardness*
Hardness*
The discharge limits are calculated by formulas incorporating hardness. The values shown are for the maximum allowable
hardness value of 400 mg/L as calcium carbonate.
-------
2. Project Chronology
A project chronology, shown in Table 2-1,
presents an overall view of how the project was
executed. More detailed explanations of selected
elements of the project are presented in other
sections of this report.
Table 2-1. Chronology of Project Execution
Date
Action Taken
July 2000 The project was selected for execution by EPA-NRMRL's MWTP in collaboration with EPA
Region 8. A project kickoff meeting was held at the Gilt Edge Mine including personnel from EPA
Region 8, CDM Federal Programs, Inc., the U.S. Bureau of Reclamation, EPA NRMRL, Green
World Science, Inc., Shepherd-Miller, Inc., South Dakota Department of the Environment and
Natural Resources (SDDENR), MSE, and the U.S. Department of Energy. Responsibilities were
assigned as follows: MSE would have a sole-source subcontract with Shepherd-Miller (North
American licensee for the Neutra-Mill technology) to neutralize the pit with the Neutra-Mill and a
sole-source subcontract with GWS to implement its RMB technology by dosing nutrients after lime
neutralization. MSE would also prepare a quality assurance project plan (QAPP) for the project and
participate in data collection and evaluation activities. EPA Region 8 would cover analytical costs,
including data validation for the project and provide the lime for the pit lake neutralization. CDM
would provide quality assurance (QA) oversight and facilitate on-site support along with the U.S.
Bureau of Reclamation (BOR). Note that at that point in time, the site was operated by the BOR.
September 2000 Shepherd-Miller personnel visited the site to obtain vertical profiles of physical measurements and
collect water for lime titrations, with the goal of establishing water depths for future monitoring, and
estimating lime requirements. Initially, the pit only contained 35 to 40 million gallons. Additional
water was transferred to the pit in the fall of 2000, including high-nitrate water from the heap leach
pad, until the final water volume was about 72 million gallons. The purpose of this was to provide
higher nitrate concentrations to challenge the technology, as well as to simply provide more water
for treatment.
February 2001 The project QAPP was finalized by MSE after receiving approval from EPA.
March 2001 Initial samples and profiles vs. depth were obtained, continuous monitoring equipment was installed,
and lime neutralization was initiated. Various problems were encountered in delivering lime to the
Neutra-Mill, as well as Neutra-Mill operation. These are discussed in more detail later in this report.
May 2001 Lime neutralization was completed around May 10, based on pH profiles indicating near neutral pH
through the bulk of the water column, with higher pH values near the bottom. Nutrient dosage was
planned for two weeks later. Immediately prior to adding nutrients, pH profiles were obtained.
These indicated that the pH of the water column had dropped to approximately 5 over the previous
two weeks. This is discussed in more detail below. Since trucks containing the nutrients were due
to arrive imminently, it was decided to proceed with nutrient dosing, which was completed in late
May.
June 2001 Continuous monitoring equipment indicated that dissolved oxygen was consumed within several
weeks of nutrient addition.
August 2001 EPA-NRMRL performed a technical systems audit on August 29, 2001. There were four findings
related to QAPP deviations, timely receipt of data so that previous sampling event data was available
for review, anomalies noted for field pH measurements, and the questionable nature of data gathered
from the continuous monitoring probes. All findings, observations, and technical comments were
addressed by MSE in a response to EPA. This response memo is contained in Appendix A.
September 2001 Additional nutrients were added to the pit. This was justified by GWS by the presence of much
higher nitrate than originally expected due to the addition of high-nitrate heap leach pad water.
More nitrate to be reduced requires more carbon to provide the necessary electrons.
-------
Date
Action Taken
October 2001 - Monitoring of the pit conditions continued. Nitrate reduction was proceeding at a rate much slower
September 2002 than expected. This was attributed to a combination of lower than optimum pH along with the
presence of elevated aluminum in solution.
April 2002 Additional molasses was added to the pit, along with sodium hydroxide to raise the pH of the water
to near neutral.
September 2002 A concerted effort was made to place the pit into optimum conditions for biological activity going
into the winter of 2002-2003. These efforts included raising the pH to neutral with sodium
hydroxide, as well as adding more nutrients. In addition, wood chips were added with the intention
that they would become waterlogged and sink, and thereby provide a suitable substrate for bacteria
to attach to and grow on.
October 2002 Foam was observed on the water surface, likely due to nitrogen and carbon dioxide gases produced
from bacterial activity. Samples taken indicated a significant drop in nitrate concentration.
April 2003 Samples and profiles taken through the winter of 2002-2003 showed rapid progress in nitrate
reduction. Physical evidence of sulfate reduction was observed after ice-off in April 2003, by the
presence of black solids and hydrogen sulfide gas aromas whenever the surface water was disturbed.
April 2003 - Continued monitoring confirmed the progression of sulfate reduction, as evidenced by a dramatic
April 2004 drop in dissolved concentrations of metals that form sulfide precipitates (e.g., copper, cadmium,
zinc); the presence of excess sulfide; very low oxidation-reduction potential (ORP) levels; and the
smell of hydrogen sulfide gas associated with samples. The pit water column had become strongly
stratified, with aerated water floating on top of strongly reducing, high-sulfide water below.
April 2004 Discussions occurred relating to potential discharge of water from the pit. These discussions
focused on how the pit might be used as part of long-term water treatment scenarios at the site.
Options considered were directly discharging surface water, or filtering and aerating deeper water.
It was decided to focus on discharging deeper water, since it was thought that the deep zone could be
better used for long-term treatment. The rationale for this is discussed later in this report.
May-July 2004 Preliminary filtration tests were performed to evaluate the required filtration of deep water, and a
filtration apparatus using bag filters was assembled. A shallow aeration/settling pond was created by
lining an existing pond. Initial transfer of small amounts of water produced issues with hydrogen
sulfide gas management, along with issues related to total suspended solids. The excess sulfide in
the water oxidized rapidly to elemental sulfur, which was very difficult to settle or filter. This is
discussed further below.
July 2004 Whole effluent toxicity tests of Anchor Hill Pit surface water, as well as water from the shallow
aeration/settling pond, showed the water was not toxic.
July-September In order to evaluate potential use of the high-sulfide, anoxic zone in the pit for future water
2004 treatment, two buckets were suspended in the pit. Both buckets contained four gallons of deep
Anchor Hill Pit water and one gallon of Surge Pond water, which was neutral pH, containing
elevated nitrate and a small amount of metals. One bucket had nutrients (methanol, molasses,
phosphoric acid) added, while the other did not. When these buckets were retrieved after six weeks,
it was found that the nitrate initially in the bucket with nutrients added had reduced to ammonia.
This was surprising and had very significant ramifications for use of the deep water zone for water
treatment.
August-October The initial 100,000-gallon batch of deep water was aerated and successfully discharged after all
2004 discharge requirements were met.
-------
Date
Action Taken
November 2004 An evaluation was performed of the stability of the stratification in the Anchor Hill pit, indicating
that turnover was very unlikely. A turnover event would be potentially dangerous due to the
presence of excess dissolved hydrogen sulfide gas in the deep zone. At approximately the same
time, the surface of the pit turned a different color, and a sewage-type smell was produced. This
indicated partial vertical mixing to a slightly deeper depth than mixing had previously occurred. The
likelihood of a turnover event was evaluated by determination of the Lake and Wedderburn
numbers, utilizing historical wind speed information from the site along with the estimated vertical
density gradient in the water column. Results of this analysis indicated that no vertical mixing
would be expected under average recorded wind speeds of 10 miles per hour (mph). Under
sustained wind speeds at the maximum recorded value of 35 mph, no hypolimnion mixing would be
expected, but there would be potential for mixing of the metalimnion with the surface zone. A
sustained wind speed of 63 mph would be required to produce conditions potentially leading to a
turnover of the overall water column. This was considered very unlikely, and no further steps were
deemed necessary to address the issue of elevated H2S in the deep zone; however, reducing the H2S
levels in the deep zone became an important goal late in the project.
February 2005 About 4 million gallons of surface water from under the ice on the pit was discharged. This took
several attempts before a configuration was established that did not draw water from below the
chemocline.
May 2005 Since it had been established that utilizing the reducing, anoxic conditions below the chemocline for
long-term water treatment was not going to be viable, discussions were undertaken regarding how to
best address the high sulfide levels in the deep water in preparation for project completion.
Evaluations were performed of using ferrous iron salts, ferric iron salts, and oxidants such as
hydrogen peroxide or bleach to eliminate the high sulfide levels. It was ultimately decided to utilize
concentrated hydrogen peroxide due to lower cost and ease of handling.
July-August About 15 million gallons of surface water meeting applicable South Dakota water quality standards
2005 was discharged.
August 2005 Concentrated hydrogen peroxide was added to the pit to eliminate excess sulfide by oxidizing it to
elemental sulfur. Much of the peroxide initially went directly to the bottom of the pit, leaving a zone
of untreated water between a depth of about 10 feet down to about 45 feet. Over the ensuing
months, mixing and reactions slowly occurred.
October 2005 Similar to the previous year, the color of the water surface changed and a sewage-type smell was
produced. Once again this probably indicated vertical mixing to a deeper depth than previously
encountered.
March 2006 Analytical data indicated that the pit water no longer contained any sulfide, indicating that hydrogen
peroxide treatment had been successful.
April 2006 An additional 15 million gallons of surface water was discharged from the pit. By this time in the
project, over 40 million gallons of water had been successfully discharged from the pit.
May 2006 Due to high amounts of runoff late in the winter at the Gilt Edge site, ARD was transferred from the
Sunday Pit to the Anchor Hill Pit. This effectively ended the monitoring phase of the project.
-------
3. Implementation
The following describes how elements of the
project were physically conducted and provides
more details about implementation of each step of
the project.
3.1 Sampling
Sample points were established at two locations
along the long dimension of the pit. These were
designated NE for northeast and SW for
southwest, since the longest dimension of the pit
runs from southwest to northeast. Orange marine
buoys marking the locations were attached with
plastic-coated cable to buckets filled with
concrete. Based on original profiles taken in
September 2000 by Shepherd-Miller personnel, a
thermocline could be expected to extend down to
the 20- to 25-foot depth in summer. In an attempt
to provide depth and lateral coverage without
driving up analytical costs excessively, sample
points were established at depths of 5 feet and 35
feet at the northeast location (thereafter designated
NE05 and NE35) and at depths of 20 feet and 60
feet at the southwest location (thereafter
designated SW20 and SW60). Generally
throughout the project, samples were taken at
these locations. Occasionally, samples were
obtained at additional depths to help answer
questions existing at those times. In addition,
vertical profiles of temperature, pH, ORP,
dissolved oxygen, and specific conductance (SC)
were obtained every five vertical feet at the NE
and SW locations using either a Hydrolab or YSI
sonde. After a history was established showing
the NE and SW profiles were generally identical,
profiles were typically taken at only the SW
location since the pit was deeper at that end.
Sediment samples were also obtained on several
occasions at the NE and SW locations.
Sampling was usually performed by attaching a
tube to the sonde, lowering the sonde to the proper
depth and location, and using a peristaltic pump to
collect water from that depth. The pump was
operated for an extended period prior to collecting
samples to allow purging of the tube. Several
sampling events were performed using a
Kemmerer sampler, in which an open tube is
dropped to the proper depth, and a "messenger"
sent down the rope to trip the sampler, capturing
the tube volume at that depth. Sediment samples
from the pit bottom were generally obtained using
an Ekman grab sampler.
Early in the project, "Troll" data-collecting
instruments, manufactured by In Situ, Inc., were
placed in the pit at the NE and SW locations.
These turned out to require extensive maintenance
to ensure trustworthy values were obtained,
particularly after addition of nutrients to the pit
and the subsequent initiation of biological activity.
The instruments became coated with organic
material that looked somewhat like algae. Since
on-site labor was not available to undertake this
maintenance, those instruments were only used
sporadically the first year, and eventually removed
and used elsewhere at the Gilt Edge site by CDM
Federal Programs.
Safety issues associated with sampling while the
pit was frozen over had to be addressed. For work
on the ice during neutralization (March-April
2001), a "walkway" was constructed using a series
of floats tied together, extending from shore to the
NE sample point, and from there to the SW sample
point. Each float was essentially a 4-inch thick, 4-
foot by 8-foot piece of closed cell foam
sandwiched by two 4-foot by 8-foot sheets of
plywood. This walkway had a cable running
down the middle, and personnel were required to
clip in a harness to this cable while using the
walkway. This was quite cumbersome since, due
to the construction of the system, personnel had to
unclip and re-clip to the cable when stepping from
one float to the next. This system was used for the
first two winters. Subsequently, personnel had to
drag one float with them out to the holes cut in the
ice for sampling purposes. Ultimately, CDM
imposed safety procedures that required a certain
minimum ice thickness for work on the ice.
The project highlighted the need for innovative
monitoring methods to better understand the
-------
progress of each reaction phase. Analysis and
interpretation of traditional laboratory data proved
to be an inadequate means of evaluating the time-
sensitive reactions that were proceeding in real
time. In March 2004, a video camera was used to
explore the Anchor Hill Pit to provide visual
evidence of what was occurring in the pit. The
video images captured by the camera (shown in
Figure 3-2, left) helped the project team determine
that biological processes were underway as
evidenced by the gas bubbles seen in the
suspended bags of wood chips (Figure 3-2, center)
and also allowed the team to view the sediments at
the bottom of the pit and also estimate the depth
the these sediments (Figure 3-2, right).
3.2 Treatment Methods/Materials
During the project, several different materials
were added to the pit lake during the treatment
process. The following sections describe the
amount of each material added and the methods
used to add the materials to the pit.
3.2.1 Lime A ddition
Lime addition was accomplished with a Neutra-
Mill, developed by Earth Systems, Pty. of
Australia. A photograph of the Neutra-Mill is
depicted in Figure 3-3. Shepherd-Miller was the
North American licensee of that technology at that
time. Since lime addition ended up being
performed essentially under winter conditions, this
turned out to be quite difficult. Initial plans by
Shepherd-Miller were to have dump trucks deliver
lime to the work site, but the truck drivers were
unwilling to back their trucks down the haul road
into the pit under icy conditions. For this reason, a
lime silo was rented and placed in an accessible
location on the rim of the pit above the Neutra-
Mill docking location. The silo had an auger
system integral to it, and a configuration was
assembled in which the auger would transfer lime
to a 6-inch pipe, and the lime would flow through
the pipe down the pit wall to the working
elevation, and then across the haul road to the
Neutra-Mill. Note that this was pebble quicklime,
and this system did not work. The coarse lime did
travel down the pit wall, but could not flow across
the relatively level haul road. Ultimately, a
conveyor was rented to transfer the pebble
quicklime across the haul road to the Neutra-Mill.
Once the problem of delivering lime to the Neutra-
Mill was solved, the Neutra-Mill itself presented
numerous operational problems. A common
problem was that the rotating drum would become
out of balance, which would require a shutdown
until personnel could redistribute the load so the
drum could function, until the next shutdown.
It was anticipated that the Neutra-Mill would be
capable of delivering, grinding, and disseminating
three to five tons of lime per hour to the pit lake.
The throughput was well below this anticipated
level. In an attempt to achieve a higher
throughput, the project switched to fine hydrated
lime delivered in bulk bags. These bulk bags were
transported by forklift down the haul road to the
Neutra-Mill, which was then fitted with a small
hopper and auger. The bulk bag was lifted above
the hopper, and cut open, spilling the lime into the
hopper. Water was needed to "sluice" the lime
from shore to the Neutra-Mill. Throughput never
exceeded 1.5 tons per hour.
The Neutra-Mill mixed the lime into the water
immediately below the platform on which the
Neutra-Mill sits. In an effort to enhance mixing
into the pit lake, a portable pump capable of
transferring approximately 400 gallons per minute
(gpm) was utilized. The pump inlet drew
water/lime slurry from beneath the platform, and
the discharge was through a 4-inch hose with the
exit placed in the southwest section of the pit. A
total of 292 tons of lime had been added to the pit
when neutralization was considered complete.
3.2.2 Nutrient Addition - May 2001
Nutrient addition was performed by Green World
Science in May 2001. This consisted of
offloading six truckloads of animal feed-grade
molasses and one truckload of methanol into the
pit water. These were not blended into the water
in a complicated manner, they were simply
offloaded through large hoses manned by the truck
drivers. Refer to Figure 3-4 (left) for a photograph
of molasses addition and Figure 3-4 (right) for a
-------
photograph of methanol addition. Due to the high
specific gravity of the molasses (1.5 to 1.6), the
molasses likely sank immediately to the bottom of
the pit. The methanol could be seen spreading out
on top of the water surface, and likely mixed in
well with the water.
3.2.3 Nutrient Addition - September 2001
Two truckloads of molasses were added to the pit
in the same manner as before - offloaded to the pit
surface by a hose.
3.2.4 Nutrient and Caustic Addition — April
2002
One truckload of molasses was added to the pit by
offloading through a hose onto the ice surface (the
pit was still frozen) at the SW end of the pit.
Upon the ice melting, the molasses then mixed
into the pit water column.
One truckload of 50% sodium hydroxide (caustic)
was also added because pH values were still not
optimum for bacterial activity despite the first pit
lake neutralization with lime that was discussed
previously. The caustic solution was added by
gravity flow through a pipe out to the middle of
the pit, on top of the ice. The solution was quite
warm; typically 50% sodium hydroxide is
delivered at 120 °F to ensure adequate flowability.
This solution melted a hole in the ice, and upon
contacting the cold water under the ice, the caustic
then probably sank to the bottom of the pit.
Anecdotal discussions with others indicated that
warm 50% caustic forms "globs" when contacting
cold water and does not mix well. It is possible
that the caustic sank to the bottom of the pit and
never reacted with the water, and it is also possible
that the caustic reacted very slowly with the water.
No direct evidence of a beneficial pH increase was
ever noticed in the vertical profiles after this
addition of caustic.
3.2.5 Nutrient, Caustic, and Wood Chip
Addition - September 2002
Through the spring-summer of 2002 (due to the
continuing absence of indicators of robust bio-
reactions) the test-team was concerned that
"something was missing" for onset of strong
reducing conditions. Debate focused on three
contributing factors:
• Dissolved-aluminum concentrations high
enough to possibly cause adverse toxicity to
bacteria at the cell membrane, inhibiting
metabolism, growth and reproduction;
• Insufficient "substrate" for bacterial
colonization; or,
• continuation of suppressed pH conditions, not
conducive to robust biology for the sulfate-
reducing bacteria (SRB).
The project team decided that the test should
incorporate changes to optimize the pit conditions
for all of the above factors, including:
• Adding caustic to raise the pH to -6.5, thereby
assuring the precipitation of dissolved
aluminum, thus minimizing the potential for
continuing cellular toxicity from aluminum;
• Adding wood chips (by floating woodchips,
by dropping woodchip-filled netting-bags to
the sediment interface at the lake-bottom, and
by installing suspended vertical columns of
woodchip-filled netting-tubes through the
water column—spanning the various lake-
layers).
These actions were carried out over the fall of
2002 and winter-spring of 2002-03.
Two truckloads of molasses, one truckload of
methanol, and two truckloads of 50% caustic were
added to the pit. These were approached much
differently than previous efforts. A pumping
system was set up to draw near-surface water from
the NE end of the pit, and discharge it at the
surface near the middle of the pit, with the goal of
inducing some better lateral mixing than would
normally be expected. The pumping rate was
approximately 800 gpm. The nutrients and caustic
were slowly metered into the pumping loop to
ensure they were well mixed into the water when
it was discharged to the pit. This avoided the poor
-------
mixing observed previously with the caustic
addition in April 2002, and provided better mixing
of the nutrients into the water as well.
Lastly, 96 tons of wood chips were added to the
pit by offloading dump trucks of wood chips into
the SW portion of the pit. The purpose of the
wood chips was to provide bacteria with a
substrate to attach to and also provide a long-term
carbon source for the bacteria. It is unclear if
these wood chips were required or whether the
reactions would have proceeded without their
addition, but these wood chips floated for a few
weeks before becoming waterlogged and sinking.
Physical indications of nitrate reduction were
evident nearly immediately; biological "slimes"
formed within the floating chips, and within 10-14
days a white froth of gas bubbles appeared and the
pit-surface was "fizzing." Figure 3-5 is a
photograph showing a white froth of gas bubbles
within floating woodchips.
3.2.6 Fertilizer Addition
Monitoring indicated that nitrate reduction was
complete by early 2003 and that sulfate reduction
was underway by summer 2003. Fertilizer was
added to the pit in August 2003 in an attempt to
stimulate an algae bloom. This was part of the
overall approach put forth by Shepherd-Miller and
Green World Science at project inception. The
idea was to stimulate algal blooms, so that when
the algae dies and sinks, it can serve as an
additional nutrient source for the SRB in the
deeper water. The fertilizer added was granular
triple super phosphate with the formula
Ca(H2PO4)2 (17% to 23% P; 44% to 52% P2O5). It
was added by slowly driving a boat around the pit
water surface and adding scoopfuls of fertilizer
into the surface water. The dosage was sufficient
to provide 1 mg/L of phosphorus to the top 3 feet
of the water column. An algal bloom was never
produced, and it is unclear if this fertilizer addition
had any impact on the overall treatment scheme.
Once near optimum conditions for treatment were
established, the reducing conditions did occur and
the bulk of water in the pit did improve from a
water quality standpoint. At this point, the project
team began to focus on treatment options to
further upgrade the water in the Anchor Hill Pit to
meet the very strict discharge requirements for the
site. These were very involved efforts, and only a
summary will be presented in the body of this
report, with a more complete description presented
in Appendix B.
3.2.7 Discharge of Deep Zone Water, 2004
Discharge of deep zone water was attempted
during the summer and fall of 2004. Bench-scale
tests focused on developing a filtration and
aeration process to meet discharge requirements.
Filtration would be required to remove residual
suspended metal sulfides, and aeration would be
needed to increase dissolved oxygen and return
biochemical oxygen demand (BOD) to
dischargeable levels. Safety issues associated with
hydrogen sulfide gas release arose and were
solved. Problems with TSS arose due to oxidation
of excess sulfide to elemental sulfur during the
aeration step. Approximately 100,000 gallons
were ultimately discharged by the further
processing treatment of the deep zone pit water.
The difficulties with filtration, hydrogen sulfide,
and the very fine sulfur solids were significant,
and pursuing discharge in this manner was not
practical. Further information about this phase of
the project can be found in Appendix C.
3.2.8 Evaluation of Use of Anchor Hill Pit
for Ongoing Site Water Treatment, 2004
In parallel with the efforts focused on discharging
deep Anchor Hill Pit water described previously,
attention was also focused on ongoing water
treatment options. It was envisioned that water
from the site could be injected with required
nutrients under the existing chemocline, taking
advantage of existing anoxic conditions and
elevated sulfide, resulting in rapid denitrification
followed by sulfate reduction as seen previously in
the pit. As part of the evaluation of injecting
water for treatment below the chemocline, bucket
tests were performed to gain further information
about what results could be expected. Water from
the Surge Pond at the site was selected for
treatment, since it contained elevated nitrate (120
mg/L as N), was relatively neutral in pH, and had
-------
small amounts of dissolved metals present.
Approximately 1 gallon of Surge Pond water
along with 4 gallons of deep Anchor Hill Pit water
was placed in each bucket. Nutrients sufficient to
completely reduce the nitrate were added to one of
the buckets, and the other bucket received no
nutrient addition. The buckets were placed at the
60-foot depth in the pit in July 2004, to both
ensure no oxygen passed through the bucket wall,
and that representative temperatures were
maintained. The buckets were retrieved six weeks
later. It was anticipated that some or all nitrate
would be reduced, and that some or all of the
existing sulfide would be oxidized. The results
indicated that sulfide was oxidized, but more
significantly, nitrate was apparently reduced to
ammonia. This was surprising, but upon
investigation, it was found that nitrate
ammonification is a known process occurring
under extreme reducing conditions. Hydrogen
sulfide can serve as an electron donor, along with
other compounds. It is surmised that a different
type of nitrate-reducing bacteria were predominant
in the deep water, relative to the oxic conditions
existing at the start of the project; however, this
was never verified. Judging by the lengthy time
(almost two years) since the completion of
denitrification, this was in retrospect not
surprising. However, the reduction of nitrate to
ammonia has very significant implications for the
Gilt Edge Mine site, since nitrate is present in
varying amounts in essentially all waters existing
on-site. Discharge limits for the site are 50 mg/L
as N for nitrate, and approximately 3-5 mg/L as N
for ammonia, depending on temperature and pH.
These results caused the project team to abandon
the concept of feeding site waters (with nutrients)
below the chemocline for treatment, since nitrate
is present in all site waters, and the risk of
excessive ammonia production was considered too
great. Abandoning the addition of site waters
below the chemocline meant that use of the
treatment process would have to be accomplished
on a batch basis, and the deep zone could not be
used as the primary reactor for ongoing water
treatment at the Gilt Edge site.
Based on the risk of nitrate ammonification
described above, it was surmised that any future
water treatment within the Anchor Hill Pit would
have to be performed on a batch basis (i.e., the pit
would be loaded with so much water, neutralized,
and dosed with sufficient nutrients to denitrify the
nitrate present and achieve a suitable amount of
sulfate reduction to produce dischargeable
dissolved metals concentrations). After allowing
some settling time for the metal sulfide
precipitates, the water would need aeration to
eliminate any remaining sulfide and BOD, and to
raise the dissolved oxygen, possibly along with
some clarification. Aerating the water, and
settling the sulfur to produce dischargeable water,
it is not practical to perform for very large
volumes of water so other ways to oxidize the
excess sulfide would be employed.
3.2.9 Evaluation of Likelihood of Lake
Turnover, 2004
Concerns about the potential for a turnover-mixing
event by the meromictic (lake layers that do not
mix) lake was evaluated in October/November
2004. This was accomplished by evaluation of the
Lake and Wedderburn numbers, utilizing historical
wind speed information from the site along with
the estimated vertical density gradient in the water
column. Results were that no vertical mixing
would be expected under average recorded wind
speeds of 10 mph. Under sustained wind speeds at
the maximum recorded value of 35 mph, no
mixing would be expected, but there would be
potential for mixing of the metalimnion (layer of
rapid temperature change) with the surface zone.
A sustained wind speed of 63 mph would be
required to produce conditions potentially leading
to a turnover of the overall water column. This
was considered very unlikely, and no further steps
were deemed necessary to address the issue of
elevated H2S in the deep zone; however, reducing
the H2S levels in the deep zone did become a
project objective. A more complete description of
this effort is provided in Appendix B.
3.2.10 Discharge of Surf ace Water, 2005
Having determined that (a) processing of the
deeper sulfide-laden water for discharge is
10
-------
difficult and likely not economically viable; (b)
use of the deep zone for ongoing water treatment
is not feasible due to the risk of excessive
ammonia production; and (c) use of the treatment
technology in the Anchor Hill Pit would have to
be done on a batch basis, it was decided to begin
discharge of surface water. This reverted back to
the original concept of placing a batch volume of
water in the pit, neutralize as necessary, add
nutrients, and let the biological treatment
processes proceed. Surface water would be
removed at some rate consistent with sulfur
settling through the water column, and enough
water would be left in the pit to provide a
reasonable cover for the less-stable, higher sulfide
at the bottom of the pit. Excess sulfide would
probably be present in all batches, due to the need
for excess sulfide to drive dissolved copper,
cadmium, and zinc down to very low
dischargeable levels. However, as the chemocline
would come closer to the surface as water is
discharged, remaining sulfide would be oxidized
to sulfur by interaction with the aerated surface
zone. The issue would then be whether the sulfur
residues would settle fast enough to accommodate
desired discharge rates as water levels drop, and
the thermo/chemocline readjusts. Developing this
approach began over the winter of 2004-2005.
In March 2005, while the pit was still frozen over,
approximately 4.3 million gallons (equating to
approximately 1 vertical meter of water column)
were removed from just under the ice surface, and
successfully discharged. This was encouraging,
and plans were made to discharge additional
surface water and eliminate excess sulfide in the
summer of 2005. The addition of various oxidants
was considered to oxidize the excess sulfide to
sulfur. Hydrogen peroxide was selected as the
preferred method to eliminate the dissolved H2S.
Thirty-five thousand gallons of concentrated
hydrogen peroxide were released into the pit
(Figure 3-6) in August 2005. The reaction was
successful, and as had been observed in the
oxidation-lagoons, the lake experienced a milky-
gray coloration as elemental sulfur formed and
began to settle (Figure 3-7).
Approximately 10 million gallons of water
meeting discharge standards was successfully
released from the surface of the pit between mid-
July and mid-August 2005.
3.2.11 Discharge of Surface Water, 2006
Analytical data indicated that the pit no longer
contained any sulfide, indicating that the oxidation
with hydrogen peroxide had been successful. In
April 2006, an additional 15 million gallons of
surface water was discharged from the pit. Since
the project began, over 40 million gallons of water
had been successfully treated and discharged from
the Anchor Hill Pit.
11
-------
Figure 3-1. Photograph indicating the location of sample buoys in the Anchor Hill Pit.
Figure 3-2. Camera used for video camera observations (left), gas bubbles within suspended bags of chips (center), and
sludge at bottom of pit (right).
'-*
J m
y%$i£,
~?i^-^k5
Figure 3-3. Photograph showing Neutra-Mill docking location.
12
-------
Figure 3-4. Adding molasses (left) and methanol (right) to the pit.
Figure 3-5. Photograph showing white froth of
gas bubbles within the floating wood chips.
Figure 3-6. Photograph showing in situ hydrogen
peroxide addition.
13
-------
Figure 3-7. Photograph showing evidence
of excess sulfur in the shallow pit water
after hydrogen peroxide addition.
14
-------
4. Results/Discussion
At the beginning of this project, the Anchor Hill
Pit contained about 72 million gallons of acidic
water, containing high levels of dissolved metals,
selenium, nitrate, and sulfate. Table 4-1, which
illustrates how water quality improved over the
course of the project, presents selected data.
Figure 4-1 presents the pH data prior to pit
neutralization.
Neutralization of the pit occurred between March
and May 2001 using the Neutra-Mill and lime.
After a several-week stabilization period following
neutralization, the RMB™ process was initiated in
May 2001 with the addition of carbon nutrients.
Additional sodium hydroxide additions were made
to the pit in response to a significant drop in pH
(from approximately 7 to 5) observed during the
post-neutralization stabilization period. Figure 4-2
shows the pH profiles with depth after lime
neutralization. Even after the attempt to neutralize
the pit lake, the bulk of the water was still at a pH
of approximately 5.
The RMB™ treatment proceeded slower than
expected, presumably attributed to starting the
treatment at a lower-than-optimum pH. Along
with pH conditions deemed lower than ideal for
bacterial growth, aluminum dissolved back into
solution at concentrations (-40 mg/L) that may
have adversely affected bacteria metabolism.
Denitrification proceeded slowly through the
remainder of 2001 and the bulk of 2002. In
September 2002 a concerted effort was made to
raise the pit pH to near 7 in order to provide a
better pH environment for the bacteria and to
precipitate aluminum from solution. Figure 4-3
shows the pH profile following additional efforts
to neutralize the pit. This effort was successful -
the pH was increased to near neutral, and
aluminum concentrations decreased to less
troublesome levels.
Wood chips were also added to provide a substrate
for bacterial growth, but the effect of the wood
chips is unknown. By October 2002, nitrate
showed a significant drop, and microbial counts
increased.
Nitrate concentrations continued to decrease
through the winter of 2002-2003, with non-
detectable levels achieved by March 2003. A plot
of nitrate/nitrite concentrations at each sample
location versus time is presented in Figure 4-4.
Note that nitrate reduction occurred more quickly
in the deep portions of the pit compared to the
shallower portions of the pit until the pH was
adjusted to near neutral in September 2002 and
nitrate reduction proceeded rapidly when more
optimum conditions for biological activity were
attained in the pit.
Thermodynamically, significant sulfate reduction
was not anticipated to occur until denitrification
was complete; and sulfate reduction was expected
to begin as soon as a population of SRB could be
developed. In mid-April 2003, field personnel
observed indications that sulfate reduction was
occurring. Whenever the surface was disturbed,
they noted black precipitates welling to the surface
and hydrogen sulfide gas odors.
Sampling through the summer of 2003 confirmed
the initiation and progress of sulfate reduction.
Average dissolved copper, cadmium, and zinc
concentrations showed a striking decrease. Figure
4-5 plots these parameters along with nitrate,
versus time, showing the dramatic decrease in
these metals (due to the onset of sulfate reduction)
immediately following denitrification. The pit
water also showed an increase in alkalinity in the
deeper water to approximately 400 mg/L (as
CaCO3), providing further evidence that biological
nitrate and sulfate reduction was occurring, since
bicarbonate is a byproduct of both the nitrate and
sulfate reduction reactions. Data through April
2006 is plotted. During May 2006, the test
effectively terminated with addition of ARD to the
Anchor Hill Pit when precipitation events caused
water management issues at the Gilt Edge site,
which prompted the addition of site waters to the
pit lake.
15
-------
Water quality in the Anchor Hill Pit continued to
improve until it approached discharge standards.
Both deep pit water and surface water were
successfully discharged from the pit. Discharge of
the water was only possible after overcoming
many technical challenges; however, the most
significant result of the project was that over 40
million gallons of water was successfully
discharged from the Anchor Hill Pit lake. In this
sense, the project was successful, but there were
several lessons learned during the project that
would be taken into account if a second batch of
water were processed in the Anchor Hill Pit or if a
similar project were undertaken at another pit lake.
10
20
30 -
*? 40
£
'S.
a 50
60
70 -
80 -
90 -
i
:
:
i
A
A
:
t
i
1
i
Nb bu
3/26/0
C\A/ n.
3/26/0
oy
1
1
6 8
pH
10
12
Figure 4-1. pH versus depth prior to pit lake neutralization.
16
-------
(c) 3rd Quarter
10
20
30
£40
70
80
90
- NE Buoy
8/29/01
- SW Buoy
8/29/01
2 46 8 10 12
pH
Figure 4-2. pH profiles with depth after initial lime neutralization.
17
-------
(d) 4th Quarter
£
m
9D
on
•in
50 -
60
70 -
80
qn
1
F^l • ~
4
N\
i
NE
1 n
_ •_ sv
10
Buoy
^24/02
i/Buoy
^24/02
— •— SW Buoy
11/8/02
SW Buoy
11/13/02
•**•
^
2 4 6 8 10 12
pH
Figure 4-3. pH profile following additional efforts to neutralize the Anchor Hill Pit.
*
Date
CO
Figure 4-4. Nitrate data at each sample location throughout the demonstration.
18
-------
100
0.01
0.001
c
NJ
•o
o
"3
o
I
"o
(0
in
o
Figure 4-5. Anchor Hill Pit nitrate/nitrite and dissolved copper, cadmium, and zinc concentrations over the project
duration. (Note: values averaged over all sampling locations.)
19
-------
Table 4-1. Water Quality Improvement Over the Course of the Project
Parameter
May 2001
January 2003
February 2004
April or May 2005
to
o
Dissolved
Aluminum
Dissolved
Cadmium
Dissolved
Copper
Dissolved
Iron
Dissolved
Zinc
Alkalinity
Nitrate/
Nitrite
Sulfate
Sulfide
NE05 NE35 SW20 SW60
49.7 45.2 41.8 40.3
0.315 0.349 0.35 0.345
16.3 16.6 16.4 15.9
0.337 0.0867 0.035 0.0479
7.19 7.65 7.61 7.49
<5 <5 <5 <5
68.7 73.2 79.3 71.6
2540 2700 2230 2510
<5 <1 <1 <1
NE05 NE35 SW20 SW60
3.38 3.43 4.09 3.51
0.203 0.206 0.198 0.184
0.871 1.47 3.3 2.64
1.03 1.27 0.976 0.858
3.38 3.43 4.09 3.51
134 134 176 186
0.4 0.2 3.8 7.21
ND ND ND ND
80 100 81 94
NE05 NE35 SW20 SW60
0.127 0.017 0.012 0.019
0.0038 0.005 0.005 0.005
0.014 0.008 0.005 0.004
0.03 0.048 0.046 0.034
0.127 0.017 0.012 0.019
106 366 378 404
0.14 0.88 0.14 0.5
1900 2400 2100 2300
4.5 9 8.7 11
NE05 NE35 SW20 SW60
0.02 0.27 0.02 0.01
0.001 0.001 0.001 0.001
0.01 0.01 0.01 0.01
0.04 0.05 0.18 0.08
0.02 0.27 0.02 0.01
98 372 418 452
0.19 0.9 0.19 0.96
ND ND ND ND
1.9 6 6.5 7.3
-------
5. Lessons Learned/Quality Assurance Activities
The following sections are lessons learned over
the course of the project that may have value to
others attempting similar work in the future. Also
included in this section is a discussion of QA
activities undertaken throughout the project.
5.1 Neutralization
• If possible, physical profiles should be
gathered frequently during times of the year
with rapid temperature change (spring and
fall) or with potential acidic run-off to more
confidently understand the trends observed.
Furthermore, regular physical profiles should
be gathered for at least one year for an existing
pit lake prior to initiating treatment such as
that presented here, for comparison purposes.
In addition, for pit lakes with uncertain
groundwater inflow or flow through, the use
of a bromide background indicator would be
useful. However, to be useful it must be well
mixed into the lake, which is easier said than
done. Ideally, it would be added a year or
more in advance to allow time for dispersion.
• The Neutra-Mill was not an efficient way to
neutralize this pit lake due to the relatively
low throughput and the relatively high
operator attention required. In fairness,
personnel at Earth Systems take the position
that the system was not properly configured
and operated by Shepherd-Miller. The fact
remains, however, that it was difficult to
transfer the lime to the Neutra-Mill without
"slurrying" it to the Neutra-Mill, and it did
require constant vigilance in operation. It is
probably quite suitable for smaller bodies of
water where the desire is simply to "get in, do
it, get out". For a larger body of water, where
a larger effort is inherently required, a reliable,
less labor-intensive approach would be
desirable. One possibility would be to set up a
mixed tank at an accessible location near the
lakeshore. Bulk bags of lime could be added
directly to the mixed tank, which would be
constantly agitating, with makeup water being
delivered via a pump immediately adjacent in
the water body. A second pump could transfer
lime slurry from the tank to the water body
through a flexible hose, which could be moved
occasionally around the lake to facilitate
mixing.
In April of 2002, one truckload of sodium
hydroxide was offloaded on top of the iced-
over pit lake, with the idea that as the ice
melted, the caustic would mix in with the
water. In reality, concentrated caustic is
delivered at elevated temperature, on the order
of 120 °F. This caustic was delivered to the
ice surface, where it promptly melted a hole in
the ice and probably sank to the pit floor.
Concentrated caustic does not mix well in cold
water; it simply forms "globs". This type of
neutralization approach should not be
attempted in a similar, future effort.
A more effective methodology was used in
September of 2002, when the concentrated
caustic was metered into a pumping loop.
Water was drawn from near the shore at
approximately 800 gpm, concentrated caustic
was metered into it at a low rate of around 1
gpm, and the mixture was delivered to the
water surface in the middle of the pit. This
was the approach taken in effectively
neutralizing the pit prior to the more rapid
commencement of biological activity.
The choice of lime versus caustic for in situ
neutralization depends, in part, on the initial
sulfate concentration and the desired final
sulfate concentration. Certainly, liquid caustic
has some advantages in ease of handling, but
will not effect the sulfate concentration as lime
can through the production of a gypsum
precipitate (CaSO4»2H2O). With the high
initial sulfate concentration present in the
Anchor Hill Pit, the use of lime was
advantageous in producing lower total
dissolved solids and sulfate values. In the case
of a hypothetical pit lake with sulfate
21
-------
concentrations well below the gypsum
saturation point, there may be no advantage in
using lime.
• Neutralization should not be conducted in the
winter. Due to project-specific circumstances,
neutralization was performed in the winter, but
it was extremely difficult. If a pit lake is
located in a climate where it freezes, then
neutralization should be conducted in the fall,
so nutrients can then be added just prior to
freezing over of the lake.
• If neutralizing to a target pH is attempted,
adequate time must be allowed for reactions to
complete, and the need for additional reagent
dosages are likely. Overdosing of the reagent
should be considered, as well as the partial use
of a reagent providing buffering capacity, such
as soda ash.
5.2 Nutrient Addition
• Nutrient dosages should be made based on a
recent analysis of the water to be treated, and
should be conducted either after the lake
freezes or immediately before ice forms (if the
climate is such that ice will form).
• Reduction reactions would likely have
occurred much faster had neutral pH been
maintained from the beginning.
• Nutrient dosage must be estimated carefully so
as to not overdose, avoiding excessive sulfide
production.
• A good microbial community analysis on the
initial water would be helpful to determine the
kinds of bacteria that are present and whether
bioaugmentation may be necessary. The
addition of inoculating bacteria to the pit
should be considered. It was not performed on
this project because a significant philosophical
aspect was that only native bacteria would be
stimulated and utilized for treatment. In other
applications it should receive serious
consideration since it might result in more
rapid treatment.
• The wood chips were a relatively cheap
material to add to the pit lake to provide
substrate for bacterial growth, but it is not
clear if wood chips were required for the
reactions to proceed.
• If fertilizer is added in an attempt to stimulate
algae growth, a liquid rather than solid form
should be utilized. In addition, careful thought
should be given to the makeup of the fertilizer
and its interaction with the surface water. In
this case, only granular triple super phosphate
[Ca(H2PO4)2] was added. It is likely that a
small amount of nitrogen should also have
been added, and further, it is likely that with
the high amounts of dissolved calcium already
present in the water, the calcium phosphate
could not readily dissolve.
5.3 Monitoring
• In situ physical measurements using a
datasonde-type apparatus are favored over the
use of a flow cell and peristaltic pump. This is
because changes can occur between the deeper
zones and the water surface. On one sampling
occasion (October 2001) a pump and flow cell
were utilized. The temperature of the water in
situ warmed up as it was pumped through
warmer water to the surface, and was
obviously inaccurate. In addition, noticeable
degassing of (probably) carbon dioxide was
observed in the pump tubing at the water
surface. This would tend to result in a pH
increase.
• If unexpected dissolved oxygen measurements
are encountered, the presence of other
dissolved gases (i.e., sulfur dioxide, nitrous
oxide, nitric oxide, chlorine, etc.) should be
considered and, if necessary, evaluated.
• The use of charge balance calculations as a
check of data quality and validity should be
utilized as necessary.
22
-------
• If biological treatment is being undertaken,
appropriate analyses (e.g., SRB counts, total
direct counts) should be performed prior to
treatment and regularly thereafter.
5.4 Discharge
Extraction and oxidation of deep water is not the
best way to discharge water. Taking water off the
oxygenated surface zone avoids the TSS problem
caused by oxidation of sulfide to elemental sulfur.
Over 20% of the water initially present in the pit
was successfully discharged this way.
5.5 Quality Assurance Activities
At the beginning of the project, responsibilities
were assigned to various members of the project
team. Quality assurance activities were assigned
to EPA Region 8 and their contractor CDM. The
following sections discuss data quality issues that
were identified during the project by CDM and
other members of the project team, the impact of
these issues on data quality, and how these issues
were addressed.
5.5.1 Field Data Review
Field data (pH, ORP, temperature, dissolved
oxygen) was collected by Shepherd-Miller, MSE,
and CDM during regular sampling events.
Continuous monitoring probes were also installed
in the Anchor Hill Pit to give real-time
information. As treatment proceeded, field data
was reviewed primarily to identify trends. Data
anomalies were identified for the continuous
measurements when these results were compared
to field data from sampling events. It was decided
that it was difficult, if not impossible, to maintain
valid calibration of these probes, so this data was
only used by the project for identifying trends and
was not used to determine if project objectives had
been met.
There were also occasional issues identified with
the field data collected at scheduled sampling
events. The pH data collected by MSE using the
YSI from the August 2001 sample event should
also be considered estimated because after
measurements were collected, all buffers measured
0.4 pH units above the accepted value for the
buffer. After this event, the YSI was calibrated at
each location prior to collecting vertical profiles,
thus not relying on one calibration for the entire
sampling event.
5.5.2 Laboratory Data
EPA Region 8 began submitting samples to the
EPA Contract Laboratory Program (CLP). Over
the course of the project, CDM reviewed field and
laboratory data as it was generated and prior to
release to MSE. CDM did validate the data and
indicate through data qualification if any data had
data quality issues. The data review process at
CDM did cause problems, as the data was not
available for review in a timely fashion so that
technical adjustments could be made.
Upon review of the data from 2001-2002, MSE
personnel noted that the charge balances were not
within tolerable limits for the samples collected.
To correct this, EPA Region 8 gave CDM
approval to send samples to non-CLP labs (Energy
in Rapid City, South Dakota and Mid-Continent in
Rapid City, South Dakota). CDM also reviewed
and validated this data prior to release to MSE.
The charge balance issue was corrected and
reporting limits improved after sending samples to
the non-CLP laboratories.
Also, as a result of the data anomalies reported by
MSE, EPA-NRMRL performed a technical
systems audit on August 29, 2001. There were
four findings related to QAPP deviations, timely
receipt of data so that previous sampling event
data is available for review, anomalies noted for
field pH measurements, and the questionable
nature of data gathered from the continuous
monitoring probes. All findings, observations, and
technical comments were addressed by MSE in a
response to EPA. This response memo is
contained in Appendix A. As mentioned
previously, CLP laboratories being used at the
project outset were not providing very defensible
data because charge balances were not within
tolerable limits.
The only relevant laboratory data that was deemed
unusable by CDM included total lead data from
23
-------
the SW60 sample location collected on 8/29/01.
Because lead was already below the discharge
criteria, this did not impact project objectives.
The most critical sampling events were those that
would determine if water met the ARARs and
could be discharged from the Anchor Hill Pit. As
summarized in Appendix B, much work was
performed to determine the best way to obtain
water that met the discharge standards. Data
relating to discharge requirements was intensely
reviewed by the project team and the State of
South Dakota to ensure that the data used to make
discharge decisions was defensible. COM did not
report any data quality issues for these samples.
24
-------
6. Conclusions/Recommendations
There is clearly the potential to utilize the pit lake
for treating site waters in batch mode. The
approach of neutralizing acidic water in the late
summer/fall followed by adding nutrients after an
extended stabilization time, will likely yield
dischargeable water quality at the surface about
1.5 years later (two winters and one summer).
Water can then be discharged off the surface. This
would probably result in significant cost savings
(20% to 40%) over the water treatment plant, but
there are many considerations, as discussed in the
previous section, to ensure that the technology is
applied appropriately.
With the successful discharge of Anchor Hill Pit
surface water in 2005 and 2006, it is possible that
significantly more surface water could be
discharged in a like manner. Water would be
discharged from the surface zone (top 5 to 10
feet), and then the new surface water would be
allowed to equilibrate with the atmosphere for
some period of time. If necessary, the dissolved
oxygen and BOD of that water would become
acceptable for discharge, along with settling of any
precipitates present, be they ferrous sulfide or
elemental sulfur. Once the surface water met
applicable standards, it could be similarly
discharged.
With the knowledge gained during the project, if a
new batch of water were ready for treatment, it
would be accomplished as follows:
• Good initial characterization over a year to
determine seasonal effects on pit lake water
quality;
• Bench-scale tests on actual water at
appropriate conditions (similar to bucket tests
performed in the pit) to assist in determining
proper dosages of the various reagents and
nutrients;
• Faster and more efficient means of pit lake
neutralization;
• Careful dosage of nutrients in the late fall so
that biological reactions could occur over the
winter;
• Detailed microbial community analysis to
determine nutrient requirements and need for
bioaugmentation;
• Consideration of adding bacterial inoculum to
speed the treatment process, particularly if
initial biological characterization indicated
that critical bacteria types were not present in
the initial water; and
• Discharge of water from the surface water
zone incrementally as treatment proceeds and
surface water meets discharge standards.
All original project objectives were eventually
met, and the project was a success. Also,
considerable experience and insight was gained
into how operational aspects of such a remediation
technique would have to be designed for a future
application of this technology. In late-fall 2005
and spring 2006, large-volume skimming and
decanting of surface-waters meeting water-quality
standards was carried out, resulting in the
successful release of approximately 40-million
gallons into Strawberry Creek over the course of
the project.
25
-------
7. References
Lewis, N.M., K.W. Wangerud, B.T. Park, S.D. Lewis, N.M., K.W. Wangerud, B.T. Park, S.D.
Fundingsland, J.P. Jonas, "Status of in situ Fundingsland, J.P. Jonas, "Anchor Hill pit lake
treatment of Anchor Hill Pit Lake, Gilt Edge Mine two-stage in situ treatment, Gilt Edge Mine
Superfund site, South Dakota", in Proceedings for superfund site, South Dakota, USA", in
ICARD 2003: Sixth International Conference on Proceedings for Tailings and Mine Waste '03, pp.
Acid Rock Drainage, July 14-17, 2003, Cairns, 345-358. Swets and Zeitlinger B.V., Lisse, The
Queensland, Australia, The Australian Institute of Netherlands, ISBN: 90 5809 593 02, 2003.
Mining and Metallurgy (AussIM), ISBN: 1-
875776-98-2,2003.
26
-------
Appendix A
Technical Systems Review Response from MSB to EPA
-------
MSE Technology Applications, Inc.
P.O. Box 4078
Butte, MT 59702
(406)494-7100
FAX (406) 494-7230
September 24,2001 01MSE-926
Norma Lewis
U. S. Environmental Protection Agency
National Risk Management Research Laboratory
26 W, Martin Luther King Drive
Cincinnati, Ohio 45268
Subject: Audit Response for Technical Systems Audit (TSA) of Mine Waste Technology Program,
Activity HI, Project 34, Bioremediation of Pit Lakes (Gilt Edge Mine)
Dear Norma:
Enclosed is the audit response for the assessment of implementation of the QAPP prepared by MSE for the
above project. The response was drafted to address findings and observations made by the auditor during the
assessment. The observation/finding appears in regular type while the response appears in italics.
I would like to compliment John Nicklas of SAIC for his assistance during the audit. He made many good
suggestions and provided assistance when he did not have to. As examples, he helped us evaluate the pH drift
issue, get the samples taped up and ready to go, and the next morning accompanied Shane Parrow and me to
the site quite early (after a long previous day) so we could go out in the raft to install the two southwest In-Situ
probes, perform a trial run with the sediment sampler, and obtain extra water for pH-drift evaluation purposes.
In addition, I think he had good ideas on some of the analytical problems the Project has been experiencing.
If you have any questions, please contact me at (406) 494-7415.
Brian T. Park, P.E.
Senior Process Engineer
Mine Waste Technology Program
Enclosure
cc: Lauren Drees, EPA
Roger Wilmoth, EPA
Gene Ashby, DOE-WETO
Creighton J. Barry, MSE-TA
Mary Ann Harrington-Baker, MSE-TA
A-l
-------
1. INTRODUCTION
A Technical Systems Audit (ISA) of the quarterly sampling activities at the Anchor Hill Pit,
located at the Gilt Edge Mine, was conducted August 29,2001 by John H. Nicklas of Science
Applications International Corporation (SAIC). (SAIC is a subcontractor to Neptune & Co., QA
support contractor for EPA.) The audit was conducted using Mine Waste Technology Program
Quality Assurance Project Plan for Bioremediation of Pit Lakes (Gilt Edge Mine), Revision 0
dated February, 2001. The following sections represent the response to the 'findings/observations
identified during the assessment.
GENERAL OA/OC ISSUES
Finding #1
QAPP Deviations. Current field activities are not always consistent with the
requirements in the approved QAPP. These items include items such as the collection of
field measurements with a YSI (as opposed to the QAPP required Hydrolab) and the
collection of waste samples with a Kemmerer sampler (as opposed to the QAPP required
weighted bottle sampler). In addition, some activities that deviate from the QAPP are not
being evaluated for their impact on project objectives.
A letter dated August 24, 2001 from Brian Park to Helen Joyce (MSB QA Officer)
itemized 13 recommended changes to the QAPP. This letter was available to the auditor
during the audit. Both critical and noncritical items were addressed in the letter.
However, these deviations had not been submitted to EPA project personnel for review
and approval.
One example of a change which may impact project objectives is that the developer has
apparently commented that since the pH of the pit water decreased from pH 7.0 to
approximately pH 5, the length of time required to achieve project objectives would
therefore increase. However, the August 24 letter proposes a decrease in the length of
time of the project due to budgetary constraints. Although separate issues, they need to
be evaluated together.
Another example of a deviation identified in the August 24 letter is the required addition
of the organic media through a hole in the ice and at a depth of 50 feet below the surface.
The organic was actually added after the ice thawed and added to the surface of the pit.
Recommended Corrective Action.
All deviations to QAPP requirements need to be documented. The contents of the August
24 letter should be reviewed by the appropriate personnel for completeness, evaluated for
impacts to project objectives, and a determination made regarding the acceptance or
A-2
-------
rejection of each item. The proposed changes, along with any additional QAPP changes,
should be incorporated into an addendum to the QAPP. A formal review and approval
process should be used.
An addendum to the QAPP will be developed. No deviations have impacted achieving the
project objectives. It must be realized that, due to the decision to try to neutralize the Pit
and dose the organic prior to the ice melting (with increased costs associated with work
on the ice and severe working conditions), and due to cost overruns on Shepherd-Miller's
part, causing MSE to absorb additional work scope, budget issues must be considered in
any decisions related to the Project. This is particularly critical in relation to project
duration and number of sampling events.
Findtng_#2
Review of Previous Sampling Event Data. Results from previous sampling events have
not all been reviewed for impact upon the current sampling event (i.e., the results from
the May sampling event have not been validated and approved for use). Additionally, it
appears as though all of the data may not yet have been received from the laboratories. It
has been determined that the results from the ICP metals analyses from the pre-
neutralization sampling event (conducted in March) show significant matrix interferences
and cannot be used to support project objectives. The results from the May sampling
event are not yet available to determine if they are capable of meeting project objectives.
If the same analytical protocols are used during this sampling event, it is possible that
results may not support project objectives.
Recommended Corrective Action. The results from previous sampling events should
be evaluated (prior to each sampling event) to determine if analytical methodology and
the requested analyses are sufficient to meet project objectives. The turnaround time for
the receipt and validation of analytical data should be reduced so the results are available
in a timely manner to make necessary project decisions.
MSE concurs with the recommended corrective action. However, EPA Region VIII and
their contractor, COM Federal Services, are responsible for producing analytical data
for the Project. As such, MSE is not in control of that aspect of the Project and cannot be
held responsible for poor analytical turnaround time. At this point MSE still has not
received non-CLP data from the May 2001 sampling event and only very recently
received supporting QA information for the CLP data from that sampling event. Note
this is now 3.5 months since that sampling event. This has been discussed with COM.
Apparently COM has begun sending samples to laboratories other than Energy Lab in
Rapid City due to their own problems with poor turnaround, COM had no
recommendations for achieving faster turnaround, other than to say that the bottleneck
appears to be the generating/formatting/validation of supporting QA information (at
COM for CLP data, at Energy Lab in for non-CLP data). Their suggestion was that data
A-3
-------
with no supporting QA would be available much faster, and could easily be obtained for
review prior to future sampling events. This suggestion is somewhat dangerous since,
without substantiating QA information, there would be no way to know whether the data
was good. If EPA Region VIII cannot be convinced to assign priority to the Anchor Hill
Pit project samples, CDM's suggestion may be the Project's only alternative.
Finding #3
Field pH Measurements. A YSI is being used to collect the field measurements. Prior
to collecting field pH measurements, the YSI was calibrated using pH 4.0 and pH 7.0,
and then a pH 5.0 solution was measured. The YSI appeared to calibrate correctly, and
the pH 5.0 solution measured pH 5.0 when checked. During the collection of field
measurements, it was noted by the field team that the pH measurements at the sampling
locations appeared anomalous. Upon returning to shore, the pH 4.0, 5.0, and 7.0
solutions were measured. All standards were reading 0.4 pH units high (pH 4.4, 5.4, and
7.4 respectively).
Recommended Corrective Action. At the time of the audit, field personnel determined
that the pH measurements during this sampling event are not critical to the determination
of the achievement of ARARs. The pHs were recorded as measured with appropriate
notation, and the water samples collected from the sampling points were measured with
the Orion pH meter.
Future pH measurements will be critical for evaluating project objectives and the
attainment of ARARs. The cause of the anomalous field pH measurements should be
determined, and a method to obtain accurate pH measurements must be determined.
Some water was collected during the sampling event to begin investigations. The other
field measurements (DO, ORP, specific conductance, and temperature) should be
evaluated to determine if they are impacted.
Extra water was obtained and brought back to MSE to investigate this further. Note that
the exact cause of the pH drift may not be determinable; a likely suspect is the amount of
organic material in the water and the associated "sliminess" noticed when sampling, but
this is not a given. Note also that the issue relates to level of accuracy. Based on the
probe calibration change, it is likely the measured pH values were high, and this is
consistent with what was measured with the Orion meter (Orion measured 4.68 for
SW20, 5.01 for SW65, 4.44 for NE05, and 4.80 for NE35, while the YSI gave 5.17, 5.42,
4.89, and 5.2 7, respectively). The ARAR minimum pH value is 6.0, so even with reduced
accuracy the pH was significantly out of compliance. Better accuracy would only be
needed if the pH climbs to approximately 6.0, which, if it happens at all, will be a very
slow process. Therefore this minor measurement drift is not a crisis. MSE will
experiment with the YSI probe and the Orion probe on the extra water brought back, and
if a reason for pH drift is established, appropriate corrective action will be taken. Some
A-4
-------
possibilities are to clean and recalibrate the YSIprobe between profiles; drop the
Kemmerer sampler to the proper depths, obtain samples, and measure them with the
Orion probe; etc. At a minimum, at the next sampling event, pH values of the four water
samples will be measured immediately in the raft with the Orion probe for comparison
with the YSI probe. The other measurements, calibrated prior to obtaining the profiles,
showed good agreement with calibration solutions after taking the profiles; this would
seem to indicate that they were not impacted.
Finding #4
Continuous Monitoring Probes. Due to the water conditions and the frequency of
calibration, the information gathered from the continuous monitoring probes is potentially
questionable. The probe at the Northeast location at a depth of 5 feet (NE 5) has failed.
The probe at the Northeast location at a depth of 35 feet (NE 35) was not calibrated
during this site visit.
Recommended Corrective Action. A decision needs to be made as to whether the
continuous monitoring probes are providing some critical or only noncritical
measurements. The purpose and necessity of these probes needs to be determined. If the
information is required, then the information gathered should be evaluated to determine
the accuracy (see finding #4). The continuous monitoring probes are identified within the
QAPP. The NE 5 probe should be repaired and replaced if required. The probes should
be calibrated quarterly or at another frequency determined to be necessary.
It is MSB's opinion that the data produced by the continuous monitoring probes are not
worth the added cost of maintaining their calibration. Realistically, to be trustworthy
they would have to be recalibrated every few weeks, and this is not possible with the
existing budget. From the beginning of the project the probes were considered a "nice-
to-have" (even by Shepherd-Miller), which would show physical variables changing with
time, but were not critical to the project goals. The only critical measurements that the
probes were originally intended to make were pH and dissolved oxygen, related to draft
South Dakota ARARs. In retrospect, it was not a good idea to trust the continuous
monitoring probes for this; the vertical profiles generated at each sampling event are
much more trustworthy since they are more representative (i.e., vertical profiles rather
than four discrete points) and are made with an instrument calibrated just prior to taking
the measurements.
Observation #1
Adherence to QAPP SOPs. Specific steps identified in the SOPs attached to the QAPP
are not being followed. Two examples are the collection of water samples using a
weighted sample bottle and decontamination of sampling equipment. The
decontamination procedure requires the cleaning and air drying of all sampling
A-5
-------
equipment that comes into contact with the samples. However, the cleaning of a
Kemmerer sampler (followed by air drying) between the 20' and 65' depths adds no value,
since the sampler is passed through the same water to reach the 65' depth.
Recommended Corrective Action. Modifications to SOPs should be documented and
distributed to project personnel to ensure consistent procedures are used throughout the
study.
The SOPs in the original QAPP were obtained from COM, and were a carryover from
the QAPP prepared by CDMfor the Gilt Edge Multicell project. MSE will develop new
SOPs as part of the QAPP addendum referred to in Finding #1 above. The new SOPs
will reflect actual practices. Note that no problems with actual practices were identified
in the audit.
ADDITIONAL TECHNICAL COMMENTS
Additional Technical Comment #1
Field Surveillances or Audits. The QAPP assigns the responsibility for determining the
need and conducting surveillances and audits to the CDM Federal QA Manager. No
surveillances or audits have been conducted by CDM, nor are any known to be scheduled.
Since sampling efforts are now being performed by MSE, responsibilities should be
reassigned.
Recommendation. EPA and MSE responsibilities for surveillances and audits should be
discussed in a QAPP addendum to ensure that these activities are performed when
deemed necessary.
The QAPP addendum referred to in Finding #1 above will reflect actual responsibilities.
Additional Technical Comment #2
Miscellaneous. The following were discussed during the audit.
• An ultra-fine point Sharpie® should be utilized for field logbooks, especially
when collecting water samples.
• A contact telephone number should be included on the chain-of-custody forms.
This telephone number is particularly critical for samples with short holding
times, and should be to a phone where someone is available during the short
holding times.
An ultra-fine point Sharpie® will be used in future sampling activities. A contact
telephone number will be included on future chain-of-custody forms.
A-6
-------
Appendix B
Evaluation of the Use of Anchor Hill Pit for Ongoing Site Water Treatment
-------
Evaluation of Use of Anchor Hill Pit for Ongoing Site Water Treatment
It was envisioned that water from the site could be injected with required nutrients under the existing
chemocline, taking advantage of existing anoxic conditions and elevated sulfide, and resulting in rapid
denitrification followed by sulfate reduction as seen previously in the pit and become an integral part of
the water treatment strategy for the Gilt Edge Mine site. As part of the evaluation of injecting water for
treatment below the chemocline, bucket tests were performed to gain further information about what
results could be expected. Water from the Surge Pond at the site was selected for treatment, since it
contained elevated nitrate (120 mg/L as N), was relatively neutral in pH, and had small amounts of
dissolved metals present. Approximately 1 gallon of Surge Pond water along with 4 gallons of deep
Anchor Hill Pit water was placed in each bucket. Nutrients sufficient to completely reduce the nitrate
were added to one of the buckets, the other bucket received no nutrient addition. The buckets were placed
at the 60-foot depth in the pit in July 2004, to ensure that no oxygen passed through the bucket wall, and
that representative temperatures were maintained. The buckets were retrieved six weeks later. It was
anticipated that some or all nitrate would be reduced, and that some or all of the existing sulfide would be
oxidized. The results indicated that sulfide was oxidized, but more significantly, nitrate was apparently
reduced to ammonia. This was surprising, but upon investigation, it was found that nitrate
ammonification is a known process occurring under extreme reducing conditions. Hydrogen sulfide can
serve as an electron donor, along with other compounds. It appeared that a different type of nitrate-
reducing bacteria were predominant in the deep water, relative to the oxic conditions existing at the start
of the project. Judging by the lengthy time (almost two years) since the completion of denitrification, this
is in retrospect not surprising. However, the reduction of nitrate to ammonia had very significant
implications for the Gilt Edge mine site, since nitrate is present in varying amounts in essentially all
waters existing on-site. Discharge limits for the site are 50 mg/L as N for nitrate, and approximately 3-5
mg/L as N for ammonia, depending on temperature and pH. These results caused the project team to
abandon the concept of feeding site waters (with nutrients) below the chemocline for treatment, since
nitrate is present in all site waters, and the risk of excessive ammonia production was considered too
great. Abandoning the addition of site waters below the chemocline meant that use of the treatment
process would have to be accomplished on a batch basis, and the deep zone could not be used as the
primary reactor for ongoing water treatment at the Gilt Edge site.
Treatment & Discharge Options: In Situ and Ex-Situ Process Considerations
Treatment of the entire pit-lake water body obviously did not occur as envisioned at the outset. Except
for the surface layer, the fine suspended metal-sulfide precipitates in the water did not readily coagulate
and completely settle. The team's attention and efforts in 2004 therefore focused on dealing with the fine
residual suspended solids in the water column, as well as chemistry issues associated with deep zone
waters.
Over the course of treatment, the pit lake had become meromictic (i.e., permanently stratified) with the
upper layer separated from the deeper zone by a strong density gradient (or chemocline). The team
recognized that the strong redox condition of the deeper zone is where the most robust kinetics for
"decontamination" occurred, and that the overlying surface layer served to protect and maintain the
thermodynamic conditions of the deep zone. Even though the upper surface-waters were "clean", the
team initially felt that discharge of the surface layer water might compromise the stability and integrity of
the multi-layered pit-lake reactor. Accordingly, the team felt that the deep zone (containing the largest
volume of pit-waters) should be targeted as the primary extraction zone—recognizing that the issues
related to fine residual suspended solids and the inherent chemistry of the deep-waters (residual carbon,
B-l
-------
excess sulfide, and dissolved non-metal constituents) would have to be dealt with. The team realized that
an additional stage of treatment would be necessary to accomplish deep zone water discharge.
Ex-Situ Deep Zone Water Processing
The team developed an approach for filtering the water while pumping from the pit, followed by aeration
in a shallow lagoon. Filtering would remove the fine suspended solids, while surface-exposure and
aeration would remove dissolved hydrogen sulfide, bio-reduce organic constituents to decrease
biochemical oxygen demand (BOD), and increase dissolved oxygen, thereby achieving dischargeable
water. Planning began for filtering of the water, and also for construction of a 0.1-acre, 5-feet deep,
150,000-gallon lined lagoon to serve as an aeration and oxidation cell.
Filter testing. In late-January 2004, filtration tests were performed on water from the deep zone in which
water was filtered through 0.45-, 1-, 5-, 10-, and 20-micron filter cartridges and the filtrates subjected to
total metals analysis. The purpose of this was to determine approximate filtration requirements for
sufficiently removing the residual fine metal sulfide precipitates from solution when pumping from the
deep zone. Results indicated that filtration at a level of 5-microns or less would be needed. While
discharge metals standards are based on dissolved analytical values, a conservative approach focusing on
total metals values was followed, since it was thought that suspended metal-sulfide particles might
oxidize and re-dissolve upon prolonged exposure to air.
Field-filters and initial pump-runs. A filtration configuration consisting of an initial 25-micron bag filter
followed by two 1-micron bag filters in parallel was selected as an inexpensive, simple way to filter initial
batches of water. These bag filters were #2-size (7-inch diameter by 32-inches long), contained in
standard, off-the-shelf filter housings. The water fed to the filtration system was to be pumped from the
50-foot depth by a Godwin HL80M pump. The pump-filtration setup and the lagoon are shown in Figure
B-l. Due to anticipated off-gassing of carbon dioxide and hydrogen sulfide gases released by pumping
water from the deep zone to the surface, tests were performed at increasing scale to assure worker health
and safety. An initial pumping test of a small batch of 200 gallons confirmed the formation and release of
small amounts of hydrogen sulfide gas. As a result, the Site Health & Safety Plan was modified for
increased instrumentation and monitoring capability, self-contained breathing apparatuses (SCBAs) were
obtained for emergency contingency in the event of extreme off-gassing, and a weather station was set up
in the vicinity of the aeration lagoon to assure that pumping into the lagoon was only done under
sufficient wind conditions to assure rapid dispersion of H2S.
Figure B-l. (left) Godwin pump and filter-set configuration in the pit; (right) aeration/settling lagoon
(ASL) with initial 2,000 gallon run
B-2
-------
Scale-Up Operations and Results. After addressing the anticipated hydrogen sulfide gas concerns, a
2000-gallon pumping test was conducted in late June 2004. Initial pumping using the Godwin pump
produced erratic, surging flow rates, likely due to the anticipated off-gassing of carbon dioxide and
hydrogen sulfide gases resulting from pressure-differentials when pumping water from the deep zone to
the surface. A subsequent 20,000-gallon pumping test, also in late-June, at a lower pump speed showed
much better pump behavior, with a flow of approximately 350 gpm. Filtration performance was not as
effective as desired, with some suspended solids remaining visible in the filtrate in the lagoon. Following
both pumping tests, field analyses of sulfide and turbidity were performed in the ensuing days. In both
cases, the sulfide level decreased and the turbidity increased. This is believed to be caused by the
formation of elemental sulfur precipitates resulting from the oxidation of residual dissolved sulfide. The
implication of this was that the aeration lagoon would have to doubly serve as a settling basin upon
completion of aeration/oxidation.
Initial Full-Scale Process Discharge. In mid-July an additional 100,000 gallons were pumped from the
deep zone through the filters to the ASL. Once again the water initially showed a decrease in sulfide and
increase in turbidity from the oxidation of sulfide to elemental sulfur, as shown in Figure B-2. The time
required for completion of this process was longer than the previous batches due to (a) the increased
volume and depth of the water in the lagoon, (b) conservative operations in delaying active aeration to
allow slower release of small amounts of residual H2S gas, and (c) to gain de-gassing experience. Active
aeration initially consisted of recirculation of lagoon water with small submersible pumps (too time
consuming), followed by use of the Godwin pump at approximately 500 gpm. This recirculation occurred
for about five days, at which time the pumps were shut off to observe settling behavior of the suspended
sulfur and the anticipated decay in BOD. Measurements were made regularly during the aeration process
with a multi-parameter probe for pH, temperature, dissolved oxygen, and oxidation-reduction potential at
the water surface and at the bottom of the lagoon. In addition, the lagoon was sampled regularly for
BOD, total organic carbon, as well as daily measurement of dissolved sulfide and turbidity. Laboratory
results for BOD and field results for dissolved oxygen at the top and bottom of the ASL are presented in
Figure B-3.
10 T
D)
E
"5
CO
Dissolved Sulfide (mg/L)
Turbidity (NTU)
0.001
0
7/12/04 7/17/04 7/22/04 7/27/04 8/1/04 8/6/04 8/11/04 8/16/04 8/21/04
Time
Figure B-2. Dissolved sulfide and turbidity in ASL following 100,000-gallon addition
B-3
-------
7/17/2004 7/22/2004 7/27/2004 8/1/2004 8/6/2004 8/11/2004 8/16/2004 8/21/2004
Time
Figure B-3. Biochemical oxygen demand (BOD) and dissolved oxygen in ASL following
100,000-gallon addition
Upon observing the expected decreases in dissolved sulfide, turbidity, and BOD in the several weeks
following the final water transfer, it became apparent that the water in the ASL was approaching
dischargeable quality. To investigate this, several sampling events occurred to provide further data. The
first event was on August 3, 2004 with the analytical results presented in Table B-l.
This dataset was encouraging since all metal values reported were within discharge standards. A further
sampling event occurred on August 5 assessing discharge parameters at three different depths within the
ASL. (Note: The opportunity was also taken to collect a sample from the surface of the Anchor Hill Pit
for comparison.) The results for both of these samples are shown in Table B-2 below.
B-4
-------
Table B-l. August 3, 2004 ASL Metals Results
Sample was collected from 0-2 feet
All results are from Energy Laboratories, Rapid City, SD
All units are mg/L
Analyte
Aluminum
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
Potassium
Selenium
Silver
Sodium
Titanium
Vanadium
Zinc
ASL 8/3/04
Total
0.58
<0.01
0.04
0.03
<0.005
<0.001
570
<0.01
0.02
0.03
0.3
<0.01
110
17
<0.0002
0.01
28
<0.005
<0.005
360
<0.005
<0.01
0.04
Dissolved
<0.05
<0.01
0.05
0.09
<0.005
<0.001
590
<0.01
0.02
0.02
<0.02
<0.01
110
16
0.0002
0.01
30
<0.005
<0.005
370
<0.005
<0.01
0.07
SD AWQC
Acute
0.36
0.017
0.015
0.063
0.281
0.0021
4.569
0.02
0.037
0.37
Chronic
0.19
0.003
0.01
0.037
0.011
0.000012
0.508
0.005
0.338
B-5
-------
Table B-2. August 5, 2004 ASL and Anchor Hill Pit Surface Water Compared to Strawberry Creek
Receiving Standards
All results are from Mid-Continent Laboratory, Rapid City, SD
All units are mg/L
pH
Conductivity at
25 °C
Temperature
Dissolved oxygen
Total alkalinity as
CaCO3
Total Dissolved
Solids
Suspended Solids
Sodium
adsorption ratio
un-ionized
ammonia nitrogen
asN
(all assuming
22°CandpH8.0)
Nitrates as N
Undisassociated
tiydrogen sulfide
Anchor Hill
Pit Lake
1-2 ft
7.88
2,460
Not Analyzed
Not Analyzed
54.4
2246
13
2.5
<0.0022
<0.0022
0.068
<0.4 (total)
ASL
0-1 ft
7.97
3,550
1-2 ft
7.96
3,560
2-3 ft
8.00
3,560
16 - 27 °C
4.7 mg/L (8/9/04)
285
3428
11
3.5
0.055
(total
NH3-N
1.25)
0.055
<0.05
<0.4
(total)
285
3303
15
3.5
0.057
(total
NH3-N
1.29)
0.057
<0.05
<0.4
(total)
279
3454
22
3.6
0.055
(total
NH3-N
1.26)
0.055
<0.05
<0.4
(total)
Std.
>6.5-<8.8
<2,500
<4,375
<75
>5.0
<750
<1313
<2500
<4,375
<10
<17.5
<10
<0.02
<0.035
(1.75 times
the
applicable
criterion)
<50
<88
<0.002
units
umhos/c
m
umhos/c
m
°F
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
See section
74:5 l:01:07~no change
in receiving water
greater than 0.5 units
30-day average
daily maximum
See section
74:51:01:31-notemp
change over spawning
beds, <4 °F change
30-day average
daily maximum
30-day average
daily maximum
24-hr composited
sample
Grab sample maximum
[Na+]/sqrt(([Ca+2] +
[Mg+2])/2) all in meq/L
30-day average
daily maximum
30-day average
daily maximum
B-6
-------
BOD5
Fecal Coliform
(May 1 -Sept 30)
TPH
Oil and Grease
Awaiting
Results
Not Analyzed
Not Analyzed
Not Analyzed
<10
<17.5
<1,000
<2,000
mg/L
mg/L
/100ml
/100ml
mg/L
mg/L
24-hr composited
sample
Grab sample maximum
geometric mean based
on a minimum of 5
samples obtained
during separate 24-hour
)eriods for any 30-day
)eriod, and they may
not exceed this value in
more than 20% of the
samples examined in
this same 30-day period
in any one sample
See section
74:51:01:10-Cannot
impart a visible film or
sheen on the surface of
the water or adjacent
shoreline
See section
74:51:01:10»Cannot
impart a visible film or
sheen on the surface of
the water or adjacent
shoreline
Dissolved Concentrations
Parameter
Acute Chronic
Arsenic (Mg/L)
Cadmium (Mg/L)
Chromium (III)
(Mg/L)
Chromium (VI)
(Mg/L)
Copper (Mg/L)
Cyanide (weak acid
dissociable) (Mg/L)
Lead (Mg/L)
Mercury (Mg/L)
Nickel (ng/L)
Selenium (Mg/L)
Silver (ng/L)
Zinc (Mg/L)
<5
2
<1 (total)
16
<10
<1
<0.2
17
12
<1
119
23
<1
<1 (total)
15
<10
<1
<0.2
23
22
<1
56
23
<1
<1
(total)
15
<10
<1
<0.2
23
23
<1
79
24
<1
<1
(total)
16
<10
<1
<0.2
25
22
<1
95
360
17
1708
15
63
22
281
2.1
4569
20
37.4
370
190
3
554
10
37
5.2
10.9
0.012
508
5
—
338
based on dissolved cone (for
acute) and total recoverable for
chronic
B-7
-------
Calcium, Dissolved
(mg/L)
Magnesium,
Dissolved (mg/L)
Sodium, Dissolved
(mg/L)
Potassium,
Dissolved (mg/L)
319
69.2
188
12.2
510
107
333
25.9
518
107
334
26
497
105
335
25.8
Hardness (in mg/L
CaCO3)
use 25 mg/L as a
minimum
use 400 mg/L as a
maximum
Note: For cadmium, chromium, copper, lead, nickel, silver, and zinc, the discharge limits are calculated by
formulas incorporating hardness. The values shown are for the maximum allowable hardness value of 400
mg/L as calcium carbonate.
This dataset showed little variation in metals data vertically, but also identified a potential problem.
Selenium values reported by Mid-Continent Laboratory were typically 0.020 to 0.025 mg/L, as compared
to nondetectable values at 0.005 mg/L produced by Energy Laboratories. This is significant because the
chronic discharge limit is 0.005 mg/L and the acute limit is 0.020 mg/L.
These samples had ammonia-nitrogen concentrations of 1.25-1.30 mg/L, suggesting that the breakdown
of nitrogen- and carbon-organics manifested itself as increased ammonia-nitrogen.
With these further encouraging results, it was decided to obtain a sample that would be fully
representative of ASL discharge water. It was apparent that dissolved oxygen would need to be boosted
from the approximate 4 mg/L present to above the discharge requirement of 5 mg/L. A decant skimmer
was set up on the ASL to take water from the upper six-inches, and pump the water through a "riffle run"
of corrugated PVC pipe to further aerate the water and increase the dissolved oxygen. The riffle run is
shown in Figure B-4. Approximately 2,000 gallons of ASL water was pumped through this riffle run into
a tank, with a sample taken from the tank for a comprehensive set of analyses representing the "final
discharge" waters. For ease of discussion, the team has coined an acronym for the entire post-Anchor
Hill Pit deep-water treatment process—the SOX process. This acronym represents the chemical- and
process-operations involved with the removal of residual concentrations of sulfur (S) and organics (O)
and the addition of oxygen (X). The comprehensive results for the SOX-process waters are presented in
Table B-3 below.
B-8
-------
Figure B-4. "Riffle run "for aerating ASL water prior to
discharge; the discharge line connects to the fitting on lower
tank, connecting with a tributary clean water-diversion pipe
near the base of slope.
B-9
-------
Table B-3. August 11, 2004 SOX Process Results Compared to Strawberry Creek Discharge Standards
PH
Conductivity at 25 °C
Temperature
Dissolved oxygen
Total alkalinity as
CaCOS
Total Dissolved Solids
Suspended Solids
Sodium adsorption ratio
Un-ionized ammonia
nitrogen as N
Nitrates as N
Undisassociated
hydrogen sulfide
BODS
Fecal Coliform (May 1-
Sept 30)
SOX |Std.
8.04
3,680
22°C
6.7
271
3551
13
3.6
0.055 (pH 8.0,
22°C) (total NH3-
N1.26mg/L)
0.055
<0.05
<0.05 (total)
<3
Not Analyzed
>6.5-<8.8
<2,500
<4,375
<75
>5.0
<750
<1313
<2500
<4,375
<10
<17.5
<10
<0.02
<0.035(1.75
times the
applicable
criterion)
<50
<88
<0.002
<10
<17.5
<1,000
<2,000
units
umhos/c
m
umhos/c
m
°F
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
/lOOmL
/lOOmL
See section 74:51:01:07— no change
in receiving water greater than 0.5
units
30-day average
daily maximum
See section 74:51:01:31— no temp
change over spawning beds, <4oF
change
30-day average
daily maximum
30-day average
daily maximum
24-hr composited sample
Grab sample maximum
[Na+]/sqrt(([Ca+2] + [Mg+2])/2) all
in meq/L
30-day average
daily maximum
30-day average
daily maximum
24-hr composite sample
Grab sample maximum
geometric mean based on a
minimum of 5 samples obtained
during separate 24-hour periods for
any 30-day period, and they may
not exceed this value in more than
20% of the samples examined in
this same 30-day period
in any one sample
B-10
-------
TPH
Oil and Grease
Not Analyzed
Not Analyzed
<10
<10
mg/L
mg/L
See section 74:51:01:10— Cannot
impart a visible film or sheen on the
surface of the water or adjacent
shoreline
See section 74:51:01:10-Cannot
impart a visible film or sheen on the
surface of the water or adjacent
shoreline
Parameter
Dissolved Concentrations
Acute Chronic
Arsenic (ng/L)
Cadmium (ng/L)
Chromium (III) (ng/L)
Chromium (VI) (jig/L)
Copper (ng/L)
Cyanide (weak acid
dissociable) (ng/L)
Lead (ng/L)
Mercury (ng/L)
Nickel (ng/L)
Selenium (ng/L)
Silver (ng/L)
Zinc (ng/L)
23
<1
<1 (total)
16
<10
<1
<0.2
18
11
<1
<50
360
17
1708
15
63
22
281
2.1
4569
20
37.4
370
190
3
554
10
37
5.2
10.9
0.012
508
5
—
338
Calcium, Dissolved
(mg/L)
Magnesium, Dissolved
(mg/L)
Sodium, Dissolved
(mg/L)
Potassium, Dissolved
(mg/L)
526
108
345
NA
based on dissolved cone (for acute)
and total recoverable for chronic
Hardness (in mg/L [actual 1758
CaCO3) mg/L CaCO3]
use 25 mg/L as a
minimum
use 400 mg/L as a
maximum
Note: For cadmium, chromium, copper, lead, nickel, silver, and zinc, the discharge limits are calculated by
formulas incorporating hardness. The values shown are for the maximum allowable hardness value of 400
mg/L as calcium carbonate.
B-ll
-------
This SOX process sample showed that the water was in compliance with standards with minor
exceptions. Conductivity and total dissolved solids are above the 30-day average limit, similar to the
current situation with the Gilt Edge water treatment plant, and which is pursuant to an interim-waiver for
total dissolved solids (EPA Interim-ROD, November 2001). Un-ionized ammonia is above both the limit
of 0.02 mg/L for 30-day average and 0.035 mg/L for daily maximum. Selenium is reported at 0.011
mg/L, between the chronic value of 0.005 mg/L and the acute value of 0.020 mg/L. BOD results for the
August 11 SOX-process sample were pending, although previous ASL analyses showed BOD to be 5
mg/L, below the 10 mg/L limit.
The continuing elevated un-ionized ammonia levels were somewhat unexpected, since previous analyses
in the Anchor Hill Pit had indicated that ammonia levels were within discharge levels. Extended aeration
would likely take care of this problem; however, in the interest of timely discharge operations, repeat
processing is not preferred. Alternatively, the pH of the water in the ASL could be lowered to
approximately 7.5 in order to meet the un-ionized ammonia nitrogen requirement.; this is the preferred
option in order to facilitate discharge. Hydrochloric or muriatic acid would be used to lower the pH of
the ASL from 8.0 to 7.5 through the addition of a small amount of concentrated acid, thereby achieving
the un-ionized ammonia-nitrogen limit.
Biotoxicity Testing Results. Samples of the SOX process water collected August 11 were received
August 12 at the ASCI Laboratory in Duluth, MN to test for acute toxicity to both juvenile fathead
minnows and Ceriodaphnia dubia. Based on the control organism performance and reference toxicity
results, the effluent exposures met minimum performance requirements specified by EPA and produced
valid toxicity results. The SOX-process sample collected did not cause sufficient lethality to allow
calculation of a 48-hour (Ceriodaphnia) LC50=>100% (i.e., there waslOO% organism survival). The
sample also did not cause sufficient lethality to calculate a 96-hour median lethal concentration for the
fathead minnows (i.e., there was 100% organism survival).
To check the toxicity of the surface layer in the Anchor Hill Pit Lake, a surface water sample (2-foot
depth) from the pit lake was collected at the same time and likewise sent to ASCI for identical testing.
48-hour Ceriodaphnia results were 100% survival and 96-hour fathead minnow results were 95% survival
(reported as no observable adverse effect). The toxicity results indicated that neither the deep zone water
treated with the SOX process, nor the surface water from the Anchor Hill Pit posed toxicity issues for
receiving waters.
Successful Discharge of Water from Below Chemocline, Summer/Fall 2004
Utilizing the results from the SOX process testing, a larger discharge batch from below the chemocline
was targeted. While the surface water met discharge requirements, it was recognized that the strongly-
reducing conditions in the deeper zone might have attributes worth maintaining for future use. While the
strongly-reducing deep water condition was believed to be the result of overdosing the nutrients, and
could be better managed if implemented in the future, most effective metal reduction and removal could
be accomplished by maintaining a condition with strong reducing potential. The surface water was
thought to be more vulnerable to disruptions such as contaminated surface runoff into the pit, which
might easily result in metals such as cadmium or zinc increasing above their discharge limit, with no
possibility of treatment by sulfide precipitation as in the deeper zone. The deeper zone was in general
considered to be more stable and controllable. In addition, the overall treatment process might be more
efficient since reducing conditions already existed, and carbon would not be needed to consume dissolved
B-12
-------
oxygen and establish anoxic conditions. It was envisioned that in the future, contaminated water from the
site might be injected, along with nutrients, into the pit below the chemocline, with the surface layer
simply serving as the "protective" layer over what could be considered the "treatment zone". Certainly if
this approach were pursued, the relative densities of the deep zone water and the contaminated water
added for treatment would have to be considered. The decision was made to focus on discharging water
from the deep zone. The elevated H2S present in the deep water posed health and safety issues, which
were addressed and managed. Mitigation of the deep water chemistry was attempted by air sparging
followed by the use of a lagoon to complete BOD reduction. In addition, the settling of solids as
previously discussed, is important since the H2S was oxidized to elemental sulfur, forming colloidal
particles that were very slow to settle. Interestingly, metals were not remobilized by this sulfide
oxidation, and approximately 150,000 gallons of water pumped from the deep zone was successfully
discharged in October 2004.
B-13
-------
Appendix C
Evaluation of Options for Eliminating Dissolved Sulfide, and Subsequent Addition of Concentrated
Hydrogen Peroxide
-------
Evaluation of Options for Eliminating Dissolved Sulfide, and Subsequent Addition of
Concentrated Hydrogen Peroxide
Following the implementation of the neutralization and RMB™ steps, the Anchor Hill Pit contained about
67 mg/L of excess sulfide in the deep zone below the chemocline (i.e., below a depth of about 30 feet).
The presence of this excess sulfide was due to overdosing of carbon nutrients (methanol and animal feed-
grade molasses) early in the project. Reducing conditions were established, and subsequently,
denitrification and sulfate reduction occurred. Sulfate reduction continued to occur beyond what would
have been desired strictly from a water quality standpoint because the sulfide produced was in excess
stoichiometrically compared to the metals levels that would form sulfides. Sulfate reduction continued
due to the presence of required reactants (carbon, other nutrients, and sulfate) along with suitable
environmental conditions.
The presence of excess sulfide posed several problems: 1) potential health and safety concerns associated
with possible release of hydrogen sulfide gas if the pit water column were to "turn over" and mix
vertically; 2) health and safety concerns associated with release of hydrogen sulfide gas when handling
the deep water for potential discharge; and 3) problems associated with generation of suspended solids
from oxidation of the excess sulfide to elemental sulfur when handling the deep water for discharge.
These risks led the project team to consider possible approaches for reducing or eliminating the excess
sulfide. Consuming the excess sulfide via addition of metals-laden site waters was ruled out due to fears
of nitrate ammonification occurring. All site waters contain appreciable nitrate (>30 mg/L as N), and
bucket tests performed in 2004 in which deep Anchor Hill Pit water was mixed with Surge Pond water
did show evidence of nitrate being reduced to ammonia rather than to nitrogen gas. Site discharge
standards for ammonia are much lower than those for nitrate, and the only viable process for removing
ammonia is to oxidize it back to nitrate via biological nitrification. This would raise other batch treatment
process complications, and therefore this option was not considered further.
CDM personnel performed bench-scale titration tests in May 2005. These titration tests evaluated the use
of ferric chloride (FeCl3), ferrous chloride (FeQ2»4H2O), and hydrogen peroxide (H2O2) to eliminate
excess sulfide. Ferric chloride would oxidize a portion of the sulfide to elemental sulfur while the ferric
iron would be reduced to ferrous. Subsequently the ferrous iron would remove the remaining sulfide
from solution as a ferrous sulfide precipitate. Ferrous chloride would remove the excess sulfide as a
ferrous sulfide precipitate. Hydrogen peroxide would remove the excess sulfide by oxidizing it to
elemental sulfur, similar to oxidizing it by aeration in the SOX Process in the summer and fall of 2004.
Results were much as expected. All three reagents were very effective, essentially quantitative, in
removing sulfide. There had been some interest in what effect the iron salts might have on the pH of the
mixture, since the ferric chloride solution (10% strength) was found to have a pH of 0.9 and the ferrous
chloride solution (10% strength) had a pH of 2.1. At a stoichiometric ratio of over 200% to remove the
67 mg/L sulfide present, both resulted in a pH drop to about 6.5 from a starting pH of 7.0; the drop in pH
was limited by the strong buffering capacity in the deep Anchor Hill Pit water (alkalinity ~450 mg/L as
CaCO3). The addition of excess hydrogen peroxide did not show dissolution of the suspended metal
sulfide precipitates; this is consistent with what was seen in the SOX Pond in the summer and fall of
2004.
Having gained confidence that any of these three reagents should be capable of successfully reducing or
eliminating excess sulfide in the deep water, the issue then became the determination of the best path to
take. The following table summarizes some advantages and disadvantages associated with each material.
C-l
-------
Table C-l. Comparison of options to oxidize excess sulfide in the Anchor Hill Pit lake.
Strength
Quantity Needed
(assuming
46,000,000 gallons
[current volume
minus top 20 feet], 67
mg/L sulfide, 100%
stoichiometric
dosage)
Cost (same
assumptions as for
quantity)
Effectiveness for
Sulfide Removal
Handling Issues
Specific Gravity
Viscosity (water at
room temperature is 1
cp, olive oil is about
100 cp)
Notes
FeCl2»4H2O
25%
308 wet tons
(14 truckloads)
~$59K
High
Low pH (10% strength
was 2.9, 25% strength
probably less)
1.28 (25% strength)
(could not find)
If dosed in excess,
could provide "sink"
for future sulfide
generated
FeCl3
33%
127 wet tons
(6 truckloads)
~$23K
High
Very low pH( 10%
strength was 0.9,
33% strength would
be less)
1.33 (33% strength)
-12 cp
H2O2
50%
26 wet tons
(1.2 truckloads)
~$16K
High
pH 4.5 (50% strength),
strong oxidizer
1.2 (50% strength)
~1.6 cp
1 . Could be an issue
with rapid
decomposition
generating O2 gas
bubbles, could strip
H2S or destabilize
water column. This
could be avoided by
adding H2O2 slowly.
2. Could be somewhat
overdosed to provide
some dissolved
oxygen (e.g., 3-5
mg/L) to inhibit
further sulfate
reduction.
Several other items considered were:
C-2
-------
• There are probably sulfide precipitates on the wall of the pit, as the water level in the pit is
pumped down, and there was a possibility that these could oxidize and degrade the surface water
quality.
• Addition of hydrogen peroxide essentially performs the SOX Process in the pit rather than the
aeration pond.
• If hydrogen peroxide were added, the elemental sulfur sludge generated would be very stable, and
would cover existing metal sulfide sludge in the pit.
• If a small amount of sulfate reduction continues due to the small amount of carbon present near/in
sediments, additional dosage may be needed. This may not be needed if another batch of water is
to be treated.
• Some added dosage may end up being needed simply due to inefficiencies of mixing into a large,
still body of water.
It was decided to add sufficient 50%-strength hydrogen peroxide to the pit to stoichiometrically oxidize
the sulfide present. This selection was made based on the fact that hydrogen peroxide would be the
cheapest, it would be relatively easy to handle, and it would chemically mimic the reactions observed
during the aeration and discharge of sulfide-laden deep water in summer 2004 (i.e., it would result in
production of elemental sulfur and should not remobilize metal sulfides).
On August 17-18, 2005, approximately 3900 gallons of 50% hydrogen peroxide by weight were added to
the pit in a total of twelve 325-gallon totes. The twelve totes were added approximately four at a time in
three locations: the southwest end; the approximate middle of the pit; and approximately midway between
those two locations. They were added by simply draining each tote down 2-inch diameter pipe to a 10-
foot deep pipe at the point of delivery. Initial totes were drained very rapidly, while later totes were
throttled to drain down in approximately 20 minutes. There were occasional disruptions seen at the water
surface, resulting from the hydrogen peroxide pushing air ahead of it in the empty pipe between
offloading totes, and from apparently vigorous reactions occurring in the deeper zone.
Initial results indicated that the hydrogen peroxide sank through the bulk of the water column and only
oxidized sulfide below about the 45-foot depth, with excess hydrogen peroxide remaining below that
depth. Subsequent mixing of deep- and mid-level waters by pumping from a shallower depth in one
portion of the lake to a deeper depth in another portion of the lake was attempted and appeared to be
successful, and as of October 2005 the excess sulfide in the deep water was significantly reduced to
between 10 and 15 mg/L. Profiles of sulfide concentrations with depth are depicted in Figure C-l. By
March of 2006, excess sulfide had been successfully oxidized (see Figure C-2).
C-3
-------
3 30
CO
^ 40
50
10 30 40 50
Sulfide (mg/L)
Figure C-l. Change in sulfide with depth after H2O2 dosage.
120
100
l6-Aug-05
8-Sep-05
6-Oct-05
11 -Jan-06
Date
Figure C-2. Sulfide concentrations with time at sampling locations.
C-4
-------
Appendix D
Evaluation of Likelihood of Water Column Turnover
-------
MEMORANDUM (VIA E-MAIL)
TO: Ken Wangerud - U.S. Environmental Protection Agency (U.S. EPA)
Richard T. Wilkins - U.S. EPA
Mark Lawrensen - South Dakota Department of Environmental and Natural Resources
(SD DENR)
FROM: Steven D. Fundingsland - COM Federal Programs Corporation (CDM)
Marko E. Adzic - CDM
CC: Brian Park - MSB Technology Applications, Inc. (MSB)
DATE: November 9, 2004
RE: Gilt Edge Mine Superfund Site - Anchor Hill Pit Lake
Gentlemen,
As you know in the spring of 2001 and through the EPA's National Risk Management Research
Laboratory (NRMRL) Mine Waste Technology Program and the EPA Region VIII Superfund office, a
field scale demonstration project demonstrating a two-step in-situ treatment of acidic and metal laden
waters was carried out within the Anchor Hill Pit Lake at the Gilt Edge Mine Superfund site. By
February 2004, waters within the top 2.5 ft of Anchor Hill (AH) Pit Lake achieved dischargeable
standards (i.e., South Dakota Ambient Water Quality Criteria) and contaminants of concern (COCs)
were generally decreased by more than 99%. Given the success of the project the U.S. EPA and SD
DENR have expressed the possibility of releasing dischargeable waters from AH Pit Lake into
Strawberry Creek. Prior to conducting such activities however, CDM evaluated the potential for deep
mixing to occur and the subsequent increase in COCs and off-gassing of hydrogen sulfide.
Since the spring of 2003 a distinct horizontal stratification within AH Pit Lake has been observed. This
stratification is supported by field recorded measurements (e.g., temperature, conductivity, oxidation-
reduction potential [ORP] and pH) collected throughout the water column, see attached. In general, the
observed horizontal stratification within AH Pit Lake is driven by a combination of thermal and
chemical gradients. As indicated by the attached data with the exception of the fall and winter months
(i.e., November to January) where temperature differences within the water column are minimized
and/or reversed, a distinct thermocline and chemocline are observed. As a result, the horizontal
stratification within AH Pit Lake is driven by both thermal and chemical gradients. Since horizontal
stratification within AH Pit Lake has been established there is no evidence that turnover and/or deep
water mixing has occurred. This is supported by the continuous negative ORP environment (i.e.,
reducing conditions) within deeper waters (i.e., depths >20 ft below surface) of AH Pit Lake. As
demonstrated by Robertson and Imberger (1994), dissolved oxygen (DO) concentrations within lakes
are strongly dependent upon the degree of deep mixing. Therefore, should deep mixing of AH Pit Lake
waters have historically occurred an increase in ORP values (i.e., more positive) would have been
observed. As confirmed by the historical ORP profiles however, this has not occurred since AH Pit
Lake has become horizontally stratified. Nevertheless to better understand lake-hydrodynamics within
D-l
-------
the AH Pit Lake, first order calculations and quantitative indicators such as the Wedderburn and Lake
Numbers were calculated (Imberger, 2001; Hamblin et al, 1999; and Robertson and Imberger, 1994).
The Wedderburn Number (W) evaluates the potential for metalimnetic water to uplift into the
epilimnion; while the Lake Number (LN) evaluates the potential for deeper hypolimnetic water to uplift.
One of the required input parameters in calculating the aforementioned quantitative lake indicators is
water density. The presence of horizontal stratification within AH Pit Lake confirms a definite
difference in water density throughout the water column (i.e., denser waters are near the bottom). These
density differences are the result of thermal and chemical differences within the water column.
Thomann and Mueller (1987) have put forth the following relationship that calculates water density as a
function of temperature and salinity:
p = \ + {(1(T3[(28.14 - 0.07357 - 0.004697') + (0.802 - 0.002T)(S - 35)]}
(Equation 1)
Where: p = Water density as a function of temperature and salinity (g/cm3)
T = Water temperature (°C)
S = Salinity (%o)
As indicated within Equation 1 a measure of salinity is required. Typically and to date, salinity has not
been analyzed at the Gilt Edge Mine Superfund site. As a result, salinity levels were estimated using the
relationship between conductivity and total dissolved solids (TDS). Eighty data points (i.e., sampling
events) were identified where both conductivity and TDS measurements were recorded within AH Pit
Lake over a period of time from October 26, 2000 to July 8, 2004. To better understand the relationship
between conductivity and TDS within AH Pit Lake, a scatter plot of the aforementioned data set was
generated, a best-fit relationship determined, and a correlation coefficient calculated using Excel. A
summary of this data is presented within Figure 1.0.
-------
Gilt Edge Mine - Anchor Hill Pit Lake
Total Dissolved Solids as a FYinction of Conductivity
y = 1.0572x- 397.79
R2 = 0.8269
n = 80; Correlation = 0.91
&*
<*
^^^»
«»
*c^
s^*k •
•
:^
«£j£¥
^>
•
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000
Conductivity (mS/crn)
Figure 1.0. Relationship between Measured Total Dissolved Solids and Conductivity within Anchor Hill Pit
Lake.
As indicated within Figure 1.0 there is a positive correlation between conductivity and TDS (i.e., r =
0.91); and with an r2 value of 0.83 a linear relationship was identified as being the best-fit. Using the
linear relationship noted within Figure 1.0, TDS levels were calculated for historic profile sampling
events within AH Pit Lake. Subsequent salinity levels were determined using the following relationship
(Thomann and Mueller, 1987):
S = •
TDS
' /perthousand
(Equation 2)
Where: S = Salinity (%o)
TDS = Total dissolved solids (mg/L)
Corresponding water densities were then calculated for each sampling event and density profiles as a
function of depth generated.
In general, there appears to be three distinct layers present within AH Pit Lake. They include: the
epilimnion (0-5 ft), the metalimnion (5-20 ft) and the hypolimnion (>20 ft). Average water densities for
the aforementioned layers were calculated to be approximately 1,001.2 kg/m3, 1,002.3 kg/m3, and
1003.3 kg/m3, respectively. The overall average density was calculated to be approximately 1,003
kg/m3.
-------
Since September 2004, using hydrometer measurements water densities have been recorded at depth
intervals of 3 ft and 25 ft below surface within AH Pit Lake. As indicated by Figure 2.0, for water
samples collected at depths of 3 ft below surface, a negative correlation between water temperature and
density (i.e., as water temperature decreases, water density increases) has been observed. On the other
hand, water temperature does not appear to any have measurable effect on density at a depth of 25 ft. At
this depth 'salinity' would appear to dominate water density effects. These results support the historical
observations of a diminishing thermocline but a persistent chemocline during the winter months, see
attached field water quality profiles.
QQQ
i nnn
i nm
OJ3
M
% 1 003 -
fD
Q
2 i nr>4
S i,UU4
&
1 006 -
1 007 -
Water Density at Varying Depths Within Anchor Hill Pit Lake
Note: Data points collected on the same date
color coordinated. Please note that all diamon
data points aie samples collected at 25 ft; whil
circular data points were collected at 3 ft belov
surface.
0
^
V
lave been
shaped
eall
*
,
• Depth = 3 ft
• Depth = 25 ft
24 6 8 10 12
Temperature (degrees C)
14
16
Figure 2.0. Hydrometer Density Measurements Recorded within Anchor Hill Pit Lake (September 24,2004
to November 5,2004).
Using the aforementioned calculated average water densities, Wedderburn and Lake Numbers were
determined. The Wedderburn Number can be used to examine the extent of upwelling of metalimnetic
water (Imberger, 2001). It is the ratio of the restoring moment about the center of volume of a lake to
the disturbance moment for two-layer stratification. Mathematically it is expressed as follows
(Imberger, 2001):
W =
xL
(Equation 3)
Where:
W = Wedderburn Number (unit less)
-------
h = Depth to thermocline (m)
M* = Water shear velocity (m/s)
L = Fetch length (m), and
The LN is a quantitative index of the dynamic stability of the water column and is defined as the ratio of
the moments about the water body's center of volume and of the stabilizing force of gravity (resulting
from the density stratification) to the destabilizing forces from wind, cooling, inflow, outflow, and
artificial destratification devices (Robertson and Imberger, 1994). In general, a LN of one indicates that
the wind energy is just sufficient to deflect the thermocline, a LN « 1 means that lake stratification is
weak with respect to wind stress and strong seiching would occur on the surface and turbulent mixing
within the hypolimnion is expected. A LN » 1 typically means lake stratification is strong and
dominates forces introduced by the wind. Little to no seiching and/or turbulent mixing within the
hypolimnion is expected (Imberger, 2001; and Robertson and Imberger, 1994).
Assuming that wind is the dominating force for mixing (i.e., inflow, outflow, and any artificial
destratification devices have minimal destabilizing force) the LN can be calculated as follows (Imberger,
2001; Hamblin etal, 1999; and Robertson and Imberger, 1994):
LN = - (Equation 4)
Where: zg = Center of volume (m)
z0 = Center of mass (m)
M = Total mass of water (kg)
ZT = Height to thermocline (m)
z = Total depth (m)
A = Lake surface area (m2)
Po = Average water density (kg/m3)
M* = Water shear velocity (m/s)
g = Acceleration due to gravity (9.81 m/s2)
It should be noted that the center of volume (zg) and center of mass (z0) were calculated as follows:
\z-Adz \m
= - - - (Equation 5)
3.2808/r
Where: zg = Center of volume (m)
dz = Incremental depth (i.e., 0.5 ft)
z = Depth (ft)
A = Incremental surface area (ft2)
x - (Equation 6)
^p(z)-A(z) 3.2808/r
-------
Where: z0 = Center of mass (m)
p(z) = Water density as a function of depth (kg/m3)
z = Depth (m)
A(z) = Surface area as a function of depth (m2)
Meteorological data (i.e., wind speed) collected from the on-site weather station as maintained and
operated by the State of South Dakota was reviewed for the purpose of this evaluation. Based on data
collected from September 2, 2003 to August 31, 2004 the average wind speed recorded at the Gilt Edge
Mine Superfund site during that time was approximately 10 mph (miles per hour). Maximum wind
speeds of approximately 35 mph were recorded during that same period of time. A plot of the historical
wind speeds recorded at the site is provided within Figure 3.0. It should be noted that the average wind
gust recorded at the Gilt Edge Mine was approximately 15 mph. Using Equations 3 and 4, a sensitivity
analysis with respect to wind speed was conducted for AH Pit Lake.
Under average wind conditions of 10 mph and the existing water level (i.e., water depth = 89.5 ft) no
seiching or turbulent mixing within AH Pit Lake is anticipated. Wedderburn and Lake numbers of 12
and 39 were respectively determined. As a result, under typical conditions AH Pit Lake would appear
to be a very stable lake and no seiching and/or mixing would be expected. This is confirmed based on
field data collected to date. Should a sustained wind speed of approximately 35 mph (i.e., site recorded
maximum) be observed, there is the potential for seiching to occur (W = 1) while no turbulent mixing
within the hypolimnion is anticipated (LN = 3).
Gilt Edge Mne Site: Anchor Mil Pit Lake
Plot of Average Hourly Wind S peed as a Function of Time
•a
VI
•a
Time
Figure 3.0. Average Wind Speeds Recorded at the Gilt Edge Mine Superfund Site from September 2003 to
September 2004 (data recorded by the State of South Dakota).
-------
To assess the potential for deep mixing to occur within AH Pit Lake a wind speed sensitivity analysis
was conducted. Based on our findings, turbulent mixing of the hypolimnion is not expected until a
sustained wind speed of approximately 63 mph was observed over the lake surface. It is important to
note that the only variable within the sensitivity analysis was wind speed. All other variables such as
fetch, water depth and depth to thermocline were held constant. It should be noted that the calculated 63
mph wind speed is the required minimum to generate turbulent mixing within the hypolimnion, and that
it does not preclude mixing will occur but rather that it is possible. As demonstrated by Robertson and
Imberger (1994), even during weakly stratified periods (i.e., low lake stability), deep mixing only occurs
when the wind force is sufficiently strong to cause LN values to drop below one.
Similar calculations were conducted assuming that 5 ft of water was discharged (i.e., approximately 7.1
million gallons) from AH Pit Lake. Should such an activity be conducted it is anticipated that AH Pit
Lake would become a two-layered stratified lake as opposed to the existing three-layered; and that a
reduced wind speed would be required to stimulate turbulent mixing within the deeper sections of the
lake (i.e., approximately 31 mph). Recall that the maximum recorded on-site velocities were
approximately 35 mph. A summary of our findings is presented within Figure 4.0, with a detailed
summary of calculations attached.
Gilt Edge Mine: Anchor Hill Pit Lake - Turnover Potential
Wedderburn and Lake Numbers as a Function of Wind Speed
1,000
Mixing
Note: Average wind speed as calculated using data
collected from the State of South Dakota on-site
weather station from September 2003 to August 2004
is approximately 10.1 mi/hr.
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
Wind Speed (mi/hr)
Figure 4.0. Wedderburn and Lake Numbers as a Function of Wind Speed within Anchor Hill Pit Lake.
-------
Based on first-order lake hydrodynamic numbers lake turnover is highly unlikely to occur within AH Pit
Lake. However, in the unlikely event that deep water mixing and/or turnover could occur, we also
evaluated the resulting effects for the potential to release hydrogen sulfide (H2S) due to off-gassing (i.e.,
volatization).
Volatization rate of hydrogen sulfide was calculated using the following relationship as defined by
Watts (1998):
„ (MKAy.(VP-P}~\ v . ^
Q = - - - L \xXH „ (Equation 7)
^ HS
Where: Q = Volatization rate (g/s)
M = Molecular weight of H2S (i.e., 34.086 g/mol)
K = Mass transfer coefficient per area (m/s)
A = Area of AH Pit Lake (i.e., 17, 855 m2 @ current water elevation)
VP = Vapor pressure (atm)
P = Partial pressure of H2S in atmosphere (i.e., 0 at time = 0)
XH2s = Mole fraction of H2S in solution @ 55 mg/L
R = Ideal gas constant (i.e., 8.21 x 10~5 m3-atm/mol-K)
T = Temperature (i.e., 8°C)
The mass transfer coefficient and vapor pressure for H2S were estimated using the following
relationships (Watts, 1998):
(Equation 8)
Where: KI = Mass transfer coefficient for H2S
K2 = Mass transfer coefficient of water (i.e., 0.83 cm/s)
M2 = Molecular weight of water (i.e., 18.02 g/mol)
M! = Molecular weight of H2S (i.e., 34.086 g/mol)
VP = HxS (Equation 9)
Where: H = Henry's constant (i.e., 8.56 x 10"3 atnvnvVmol)
S = Solubility of H2S (i.e., 398 g/100 g)
The average ambient air temperature of 8°C was calculated from hourly data collected from the on-site
meteorological station during the summer months of 2004. Data collected during the winter months
were not considered for two reasons: 1) they would adversely bias the resulting average, and 2) due to
the presence of ice cover turnover and/or lake mixing would not result from disturbing forces such as
wind.
Applying Equation 7 to the AH Pit Lake, it was determined that the maximum H2S concentration that
could off-gas should a turnover event occur would be approximately 65 ppm. A summary of detailed
-------
calculations as generated using Excel are attached. The resulting maximum H2S concentration that
could evolve from AH Pit Lake in the unlikely event of a complete turnover is significantly less than the
potentially fatal limit of 250 ppm (MSDS, Canadian Centre for Occupational Health and Safety).
We trust that the above information is sufficient at this time and look forward to hearing from you on
this matter in the near future.
MEA/mea
Attachments (1)
-------
REFERENCES
Hamblin, P.P., C.L. Stevens and G.A. Lawrence. (1999). "Simulation of Vertical Transport in Mining
Pit Lake". Journal of Hydraulic Engineering. 125(10), 1029-1038.
Imberger, Jorg. (2001). "Characterizing the Dynamical Regimes of a Lake". 1-23.
Robertson, Dale M. and Jorg Imberger. (1994). "Lake Number, a Quantitative Indicator of Mixing Used
to Estimate Changes In Dissolved Oxygen". Int. Revue ges. Hydrobiol. 79(2), 159-176.
Thomann, Robert V. and John A. Mueller. (1987). "Principles of Surface Water Quality Modeling and
Control". Harper Collins Publisher. Chapter 4.
Watts, Richard J. (1998). "Hazardous Wastes: Sources, Pathways, Receptors". John Wiley & Sons, Inc.
Chapter 7.
------- |