Office of Solid Waste and EPA-542-R-12-003
Emergency Response July 2012
(5102G) www.epa.gov/superfund/remedytech
www.clu-in.org/optimization
Optimization Review
Black Butte Mine Superfund Site
Lane County, Oregon
-------
OPTIMIZATION REVIEW
BLACK BUTTE MINE SUPERFUND SITE
LANE COUNTY, OREGON
Report of the Optimization Review
Site Visit Conducted at the Black Butte Mine Superfund Site on
January 10,2012
July 13, 2012
-------
EXECUTIVE SUMMARY
The purpose of this optimization review was to evaluate site conditions and identify optimal approaches
for conducting a remedial investigation (RI) of the Black Butte Mine (BBM) Superfund Site. The review
was conducted using U.S. Environmental Protection Agency (EPA) optimization review methods.
This document focuses on the fate and transport of mercury and other trace metals at and downstream of
the BBM Superfund Site as a means to focus and streamline the sequence of RI activities. It is expected
that this report may form the basis for additional systematic project planning among the optimization
review team, project technical team, and stakeholders to develop, review, and finalize Rl-specific work
planning and implementation documents.
EPA's Office of Solid Waste and Emergency Response (OSWER) and the Office of Superfund
Remediation and Technology Innovation (OSRTI) define optimization as follows:
"Efforts at any phase of the removal or remedial response to identify and implement
actions that improve the action's effectiveness and cost-efficiency. Such actions may also
improve the remedy's protectiveness and long-term implementation which may facilitate
progress towards site completion. To identify these opportunities, regions may use a
systematic site review by a team of independent technical experts, apply techniques or
principles from Green Remediation or Triad, or apply some other approach to identify
opportunities for greater efficiency and effectiveness. Contractors, states, tribes, the
public, and PRPs are also encouraged to put forth opportunities for the Agency to
consider. "
Optimization reviews include a "systematic site review," whereby the site as a whole is often considered.
However, optimization can focus on a specific aspect of a given cleanup phase (or a particular operable
unit [OU]), with other phases and site areas considered to the degree that they affect the focus of the
optimization effort. For optimization reviews conducted before a Record of Decision (ROD) is issued, the
focus is on developing the conceptual site model (CSM) by leveraging existing data and exploring
potentially applicable sampling and analysis tools and strategies that facilitate a comprehensive
systematic planning process.
The recommendations in this report are intended to help the site team identify opportunities for an
optimized RI approach. Where noted in this report, further analysis of a recommendation may be needed
before the recommendation can be implemented. The recommendations are based on an independent
evaluation and represent the opinions of the optimization review team. These recommendations do not
constitute requirements for future action, but rather are provided for consideration by the Region and
other site stakeholders. While the recommendations may provide some details to consider during
implementation, the recommendations are not meant to replace other, more comprehensive, planning
documents such as work plans, sampling plans, and quality assurance project plans (QAPP).
Site-Specific Background
The BBM Superfund Site (the site) is located in Lane County, Oregon, approximately 35 miles southeast
of Eugene and approximately 10 miles upstream from the Cottage Grove Reservoir (CGR) (Figure 1).
ES-1
-------
Mercury mining and processing operations were active at the site from the late 1890s to the late 1960s.
The site has been identified as a significant contributor of mercury to sediment and fish tissue in CGR.
EPA Region 10 (Region 10) entered the site on the National Priorities List (NPL) in 2010.
Located on the north face of Black Butte, the mine area is drained by Dennis Creek, Garoutte Creek, and
Furnace Creek, which border the north, west, and south sides of the site (see Figure 2). Both Dennis
Creek and Furnace Creek are tributaries to Garoutte Creek which, after joining Big River to form the
Coastal Fork Willamette (CFW) River, flows to the CGR approximately 10 miles downstream. The
principal site features include collapsed and open mine adits, the Main Tailings Pile located adjacent to
Dennis Creek, the Old Ore Furnace Area, the New Furnace Area, and the Furnace Creek Tailings Area
(Ecology and Environment 2006) (Figure 2).
The CGR was constructed in 1942 as a U.S. Army Corps of Engineers (USAGE) flood-control reservoir.
Reservoir levels are decreased annually between the end of September and November. The low pool level
is maintained until early February, when management practice requires that water levels begin increasing
to the maximum pool level, which is attained in mid-May. At full pool, the reservoir area is 1,158 acres;
at low pool, the water area is reduced to approximately 25 percent of the maximum acreage.
The site was identified as a potential source for mercury contamination in CGR by the Mercury Working
Group of the Oregon Department of Environmental Quality (ODEQ) Water Quality Division during an
evaluation of Oregon's lakes. This evaluation focused on the analysis of mercury in fish tissue samples
collected in 1993 and 1994. In 1994, the site was referred to ODEQ's Site Assessment Section (SAS)
staff for review. The SAS recommended further site investigation as a medium priority (ODEQ 1996).
Pursuant to a cooperative agreement with EPA Region 10, ODEQ conducted a preliminary assessment
(PA) of the site (ODEQ 1996). EPA Region 10 completed a site inspection in 1998 and removal
assessment investigation in 2006. EPA Region 10 conducted a removal action (RA) at the site in 2007
(EPA Region 10 2008). The RA consisted of soil and tailings characterization, excavation, relocation of
mine wastes that exceeded site-specific criteria, and placement of mine wastes that exceeded site-specific
screening criteria in a repository constructed on site.
In 2009, EPA Region 10 completed a Hazard Ranking System (HRS) evaluation for the site (Ecology and
Environment 2009). Results of the overland flow/flood component of the evaluation were sufficient to
qualify the site for inclusion on the NPL. Other pathways (groundwater, groundwater to surface water,
and soil) were excluded from the HRS evaluation because their effect on the overall ranking score was
negligible. As a result of the HRS evaluation, the BBM Site was added to the NPL on March 4, 2010.
Currently, the EPA (with ODEQ participation) is in the planning stages for a RI for the BBM Site. The
optimization review documented in this report was conducted to ensure that the RI work plan addresses
all of the potentially significant sources of mercury contamination observed in surface water, sediments,
and biota immediately downstream of the BBM Site and farther downstream within the CGR. A factor
that was important in deciding to optimize the RI planning process for the site is the large size of the
affected area, which includes the actual mine site, downstream rivers, and the CGR. In addition, the
complex fate and transport mechanisms involving transformations between various mercury compounds
over the 10-mile transport distance also contributed to the need for this optimization review. The seasonal
variation of CGR water levels for flood control adds additional complexity. The main goal of this review
was to lay the foundation for the design of an RI that effectively and efficiently characterizes the nature
and extent and evaluates risks to human health and ecological receptors for all major sources of mercury
contamination occurring at the BBM Site and in the downstream surface water features, including the
CGR.
Summary of Methods
ES-2
-------
The methods used in the optimization review included:
• A site visit by the optimization review team, conducted on January 10, 2012,
• Literature reviews to examine previously constructed conceptual site models (CSMs) for mercury
sourcing and transport,
• Construction of a revised CSM that considers all potential ongoing sources of mercury
contamination, including those associated with the BBM Site and other sources that could be
associated with historical sediments in the CGR, and
• Construction of a sampling strategy with decision logic aimed at testing the hypotheses inherent
in the revised CSM.
Summary of Conceptual Site Model
For the purpose of this optimization review, the project technical team is employing the CSM
project life cycle concept further described in the EPA document "Environmental Cleanup Best
Management Practices: Effective Use of the Project Life Cycle Conceptual Site Model" (EPA
2011), available at http://www.epa.gov/tio/download/remed/csm-life-cvcle-fact-sheet-final.pdf.
This preliminary CSM and limited existing data indicate that the Furnace Creek Tailings Area may be an
important source of contemporary loading of mercury and potentially other metals to downstream surface
water features, including CGR. It is important to note that more than a century of BBM mercury inputs
have historically loaded downstream surface water, including the CFW River and subsequently the CGR
through its 60-year existence. Based on CGR sediment coring results, Curtis (2003) reports that sediment
mercury concentrations were up to three times greater in the early 1970s relative to more recently
measured concentrations. As a result, historical mercury loading likely exceeded contemporary loading by
several orders of magnitude. The CSM recognizes, therefore, that significant mercury concentrations may
exist in CGR as a result of historical loading.
The preliminary CSM further hypothesizes that an important source for contemporary loading from the
BBM Site to downstream surface water features is suspended particles with elevated concentrations of
mercury (and other metals). This suspended particulate mercury is generated by mechanical erosion of
BBM Site tailings from the Furnace Creek Tailings Area and, potentially, the Main Tailings Pile. Some
particulate mercury may also be converted to a dissolved form in surface water transport. At the BBM
Site, much of the mercury contained within these suspended particles may exist in a less bio-available
form (mercuric sulfide [HgS]), which may be converted to more soluble forms during transport or after
deposition. Mercury enters the CGR in both particulate and dissolved phases. Some of the dissolved
mercury is converted to methylmercury (MeHg) by bacteria inhabiting anoxic environments (the
sediments and potentially the anoxic portion of the water column). Over time, some of the particulate
mercury that settles in the sediments may be converted to dissolved mercury and become available for
methylation. Once methylated, mercury can bioaccumulate in the food chain resulting in unacceptably
elevated concentrations in sport fish.
Summary of Findings
The summary below lists findings identified by the optimization review team as significant to optimizing
the RI approach. Findings are presented first for the BBM Site and vicinity, followed by the findings for
CGR.
Key findings related to the BBM Site and vicinity include:
ES-3
-------
• Consistent with the preliminary CSM focus on Furnace Creek, very steep terrain and evidence of
active tailings erosion and mobilization were observed adjacent to Furnace Creek during the site
visit. Site data indicate tailings from the Furnace Creek Tailings Area contain more elevated
mercury concentrations than the tailings from other portions of the site and that the mercury in the
Furnace Creek tailings is typically in a more bio-available form (Ecology and Environment
2006).
• A post-RA surface water loading assessment (Thorns 2008) suggests that the transport of
suspended solids containing mercury appears to be the primary mode of mercury transport from
the site. Based on one sampling campaign during non-storm conditions, the assessment estimates
that Furnace Creek could contribute between 50 to 75 percent of the mercury load in the CFW
River. Re-calculation of this value by the optimization review team suggests a lower contribution
(26 to 59 percent); however, it still represents a potentially significant source.
• Although the tailings thickness exceeds 10 feet over much of the site, direct evaluation of
potential impacts from the tailings leachate on groundwater quality is likely infeasible because of
the site's fractured bedrock geology and the occurrence of the water table within the bedrock.
• As a result of historical airborne deposition of elemental mercury during mining and ore
processing operations, the surrounding hillsides and the non-mine portions of Black Butte may
serve as significant sources of mercury loading to surface water. Curtis (2004) determined that
from 44 to 87 percent of mercury in off-site hillside soil was complexed with organic matter.
Mercury complexed with organic matter is more readily converted to methylmercury.
• Historical data indicate the presence of potential mercury impacts in surface water sediments
from the site downstream to CGR. The contribution of historical mercury present in surface water
sediments versus the flux of new fmed-grained material with elevated mercury from BBM is not
well understood.
• The pH of the groundwater discharging to two of the mine adits visited during the site visit was in
the neutral range, suggesting the general absence of acid mine drainage impacts at the site.
Key findings that relate to CGR include:
• Deposition of atmospheric mercury attributable to various industrial and mining sources (such as
coal-fired power plants) world-wide (also referred to as the global mercury pool) likely
contributes a small but unknown fraction of the total dissolved mercury burden of the CGR.
Given that one of the sources of mercury to CGR is deposition from the global mercury pool,
mercury reductions in fish tissue may be limited to some baseline level that reflects this ongoing
source.
• Methyl mercury generation generally requires the presence of three constituents: dissolved
mercury, microbial labile organic carbon, and sulfate. Uncertainties exist regarding the factors
that control the availability of these constituents and subsequently the methylation process. The
seasonal changes of the water level in the reservoir may result in the cycling of sulfide to sulfate,
thus perpetuating the availability of sulfate and potentially the methylation process.
• Sediment cores from CGR indicate that elevated mercury concentrations are present in older,
legacy sediments. Exposure of these sediments to surface water erosion during the yearly low
water level period may be an important ongoing source of mercury to the reservoir.
ES-4
-------
• Review of the available total mercury, dissolved mercury, and fish tissue mercury concentration
data for CGR suggests that even if only a small fraction of the total mercury is present in
dissolved phase, sufficient mercury methylation will occur to result in elevated mercury in fish
tissue. Based on existing data, calculations by the project team indicate that the percentage of
total mercury that is methylated in CGR water is only 6 percent. In sediments, the percentage is
only 0.1 percent. These low levels are apparently sufficient to support methylation.
Summary of Recommendations
Recommendations for the BBM Site portion of the RI include:
• To improve the understanding of dominant sources of total mercury, dissolved mercury, and
methylmercury release from the BBM Site, quarterly surface water sampling under storm and
non-storm conditions is recommended along with the collection of concurrent sediment and
groundwater samples. Discharge measurements of the site streams should accompany the
sampling to support the determination of mercury loading estimates.
• Site data regarding the assessment of potential impacts of tailings leachate on site groundwater
quality consist only of a limited number of laboratory-based leaching procedure samples; the
procedure had a high detection limit (above 440 nanograms per liter [ng/L]). The majority of the
analytical results for these samples were below this detection limit. However, since the typical
mercury background concentration is less than 200 ng/L, the potential impacts of the tailings on
groundwater are unknown. The collection of vadose zone groundwater samples from the site
tailings areas is necessary, but may be not be readily accomplished because the water table at the
site occurs in fractured bedrock. Therefore, the optimization review team recommends other
groundwater data could be used. If the presence of a historical tailings disposal area on the
Garoutte Creek floodplain can be verified, the optimization review team recommends that
groundwater samples be collected from the saturated sediments (if any exist) beneath the tailings.
In addition, the potential presence of groundwater under saturated conditions in tailings adjacent
to Dennis and Furnace Creeks should be evaluated. If groundwater is determined to exist in these
tailings, this groundwater should also be sampled.
• If BBM environmental media and Furnace Creek tailings in particular are not found to be major
contributors to mercury and trace metal contamination in Garoutte Creek, it is recommended that
the project team consider increased sediment sampling in Garoutte Creek and sediment sampling
in the CFW River to assess the significance of these potential sources for downstream loading of
total mercury, dissolved mercury, and methylmercury.
• It is recommended that a Demonstration of Method Applicability (DMA) analysis be conducted
for X-ray fluorescence (XRF) and Lumex field-based metals analysis. Results of this analysis can
be used to assess confidence in RA characterization results and to assess the utility of field-based
methods for metals analyses during the RI. The DMA could include site soil and hillside
sampling and generate preliminary data to optimize sampling for the RI's human health and
ecological risk assessments.
Recommendations for the CGR RI include:
• Development of the data necessary to understand the source of methylmercury in CGR fish tissue
requires investigation of the major sources of mercury mass influx to the reservoir (in addition to
ES-5
-------
the current contribution from BBM) and of the factors controlling the availability of the rate-
limiting constituents (dissolved mercury, labile organic carbon, and sulfate). In light of technical,
administrative, funding, and schedule challenges, it may prove beneficial for Region 10 to
consider conducting activities at BBM and CGR as separate OUs.
• A major objective to consider for the CGR RI is establishing baseline data for (1) fish tissue
mercury concentration levels in the CGR and (2) the influx rate of mercury (total, dissolved, and
methylated) to the reservoir. This baseline data can be used to assess the effects of any mitigation
efforts at the BBM Site or in the CGR.
• It is recommended that various CGR environmental media be sampled to enable a preliminary
assessment of the factors controlling methylmercury generation. Sampling may include the
collection of quarterly or semiannual surface water, sediment, and sediment pore water. Specific
objectives of this sampling would include (1) confirming the existence of sulfate cycling in the
high pool sediments and (2) assessing the significance of internal loading through the erosion and
mobilization of elevated mercury concentration legacy sediments during low pool conditions.
ES-6
-------
NOTICE
Work described herein was performed by Tetra Tech EMI for the U.S. Environmental Protection Agency
(EPA). Work conducted by Tetra Tech EM Inc., including preparation of this report, was performed
under Work Assignment 2-58 of EPA contract EP-W-07-078. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
-------
PREFACE
This report was prepared as part of a national strategy to expand Superfund optimization practices from
remedial investigation to site completion implemented by the United States Environmental Protection
Agency (EPA) Office of Superfund Remediation and Technology Innovation (OSRTI). The project
contacts are as follows:
Organization
Key Contact
Contact Information
EPA Office of Superfund
Remediation and Technology
Innovation
(OSRTI)
Stephen Dyment
EPA
Technology Innovation and Field Services
Division
1200 Pennsylvania Ave., NW (5203P)
Washington, DC 20460
dyment.stephen(o),epa. gov
Phone: 703-603-9903
Tetra Tech EM Inc. (EMI)
(Contractor to EPA)
Jody Edwards, P.O.
Tetra Tech EMI
1881 Campus Commons Drive, Suite 200
Reston,VA 20191
j ody. edwards@tetratech. com
phone: 802-288-9485
Tetra Tech EMI
(Contractor to EPA)
Mark Shupe, P.O.
Tetra Tech EMI
1881 Campus Commons Drive, Suite 200
Reston,VA 20191
mark, shupeigjtetratech. com
phone: 703-390-0653
-------
LIST OF ACRONYMS
BMP
CFW
CGR
CSM
DMA
DO
DOC
DPT
EMMA
DU
EPA
FS
GIS
Hg
HgP
HgR
HgT
HRS
ICS
kg/yr
LASAR
MeHg
mg/kg
mg/L
ng
ng/L
NGVD
NOAA
NPL
ODEQ
ORD
ORP
OSRTI
OSWER
OU
PA
P&T
PEL
PRG
QA
Best management practice
Coastal Fork of the Willamette
Cottage Grove Reservoir
Conceptual site model
Demonstration of methods applicability
Dissolved oxygen
Dissolved organic carbon
Direct-push technology
End Member Mixing Analysis
Decision unit
United States Environmental Protection Agency
Feasibility study
Geographic information system
Mercury
Suspended (particulate-bound) phase mercury
Reactive mercury
Total mercury
Hazard Ranking System
Incremental composite sampling
Kilograms per year
Laboratory Analytical Storage and Retrieval
Methylmercury
Milligrams per kilogram
Milligrams per liter
Nanogram
Nanograms per liter
National Geodetic Vertical Datum of 1929
National Oceanic and Atmospheric Administration
National Priorities List
Oregon Department of Environmental Quality
Office of Research and Development
Oxidation-reduction potential
Office of Superfund Remediation and Technology Innovation
Office of Solid Waste and Emergency Response
Operable unit
Preliminary assessment
Pump and treat
Probable effects level
Preliminary Remediation Goal
Quality assurance
in
-------
QAPP
QC
RA
RI
ROD
RSE
SAS
sc
SPLP
SLV
SPP
SQuiRTs
SSE
TAL
TEL
TMDL
TOC
TSS
UCL
UNEP
USAGE
USGS
XRF
Quality assurance project plan
Quality control
Remedial Action
Remedial investigation
Record of Decision
Remediation System Evaluation
Site Assessment Section
Specific conductance
Synthetic Precipitation Leaching Procedure
Screening level value
Systematic project planning
Screening Quick Reference Tables
Selective sequential extraction
Target analyte list
Threshold effects level
Total Maximum Daily Load
Total organic carbon
Total suspended solids
Upper confidence limit
United Nations Environmental Programme
U.S. Army Corps of Engineers
U.S. Geological Survey
X-Ray fluorescence
IV
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY ES-1
PREFACE ii
LIST OF ACRONYMS iii
1.0 INTRODUCTION 1
1.1 PURPOSE 1
1.2 REVIEW TEAM COMPOSITION 2
1.3 DOCUMENTS REVIEWED 3
1.4 QUALITY ASSURANCE 4
1.5 PERSONS CONTACTED 5
2.0 SITE BACKGROUND 6
2.1 LOCATION AND PRINCIPAL SITE FEATURES 6
2.2 SITE HISTORY 6
2.2.1 HISTORICAL LAND USE AND OPERATIONS 6
2.2.2 CHRONOLOGY OF ENFORCEMENT AND REMEDIAL ACTIVITIES 7
2.3 POTENTIAL HUMAN AND ECOLOGICAL RECEPTORS 7
2.4 EXISTING DATA AND INFORMATION 8
2.4.1 BBM SITE EXISTING DATA 8
2.4.2 CGR EXISTING DATA 11
3.0 DESCRIPTION OF PLANNED OR EXISTING REMEDIES 15
4.0 CONCEPTUAL SITE MODEL 16
4.1 CSM COMPONENTS FOR BBM 16
4.2 CSM COMPONENTS FOR CGR 17
4.3 DATA GAP IDENTIFICATION 18
4.3.1 SITE DATA GAPS 18
4.3.2 CGRDATAGAPS 20
4.4 IMPLICATIONS FOR REMEDIAL STRATEGY 21
4.4.1 RECOMMENDATIONS FOR RI IMPLEMENTATION AT THE BBM SITE 21
4.4.2 RECOMMENDATIONS FOR RI IMPLEMENTATION AT THE CGR 32
5.0 FINDINGS 38
6.0 RECOMMENDATIONS 41
6.1 COMPARISON OF RECOMMENDATIONS TO TRADITIONAL OPTIMIZATION
Focus AREAS 42
7.0 REFERENCES 44
-------
List of Tables
Table 1. Screening Levels Referenced in Previous Investigations
Table 2. Summary of Proposed Sampling
List of Figures
Figure 1. Surface Water Features in General Vicinity of Black Butte Mine
Figure 2. Black Butte Mine Site Area
Figure 3. Preliminary Pathway Receptor Network Diagram
Figure 4. Schematic Representation of Conceptual Site Model
Figure 5. Surface Water Sampling Locations at Black Butte Mine
Figure 6. Data Evaluation Logic for Black Butte Mine Surface Water Sampling Task
Figure 7. Data Evaluation Logic for Black Butte Mine Sediment Sampling Task
Figure 8. Proposed Groundwater Sampling Locations at Black Butte Mine
Figure 9. Data Evaluation Logic for Black Butte Mine Vadose Zone Groundwater Sampling Task
Figure 10. Data Evaluation Logic for Black Butte Mine Groundwater Sampling Task
Figure 11. Data Evaluation Logic for Black Butte Mine Tailings Sampling Task
Figure 1 la. Proposed Area for Hill Slope Sampling
Figure 12. Data Evaluation Logic for Black Butte Mine Soil Sampling
Figure 12a. Proposed Sampling Locations for CGR
Figure 12b. Petite Ponar Dredge Grab Sampler
Figure 12c. Example Tool for Performing Pore Water Sampling in Soft Sediments
Figure 13. Evaluation of CGR Internal Loading - Low Pool Sediments
Figure 14. Evaluation of Potential of Internal Loading - High Pool Sediments
Figure 15. Data Evaluation Logic for Cottage Grove Reservoir Sampling
Attachments
Attachment A. Site Visit Photo Log
Attachment B. Description of Incremental Composite Sampling
VI
-------
1.0 INTRODUCTION
1.1 PURPOSE
The Black Butte Mine (BBM) Superfund Site is located in Lane County, Oregon, approximately 35 miles
southeast of Eugene and 10 miles upstream from the Cottage Grove Reservoir (CGR). Mercury mining
and processing operations were active at the site from the late 1890s to the late 1960s. The site has been
identified as a significant contributor of mercury to sediment and fish tissue in CGR (Park and Curtis
1997). Region 10 included the site on the National Priorities List (NPL) in 2010.
The purpose of this optimization review was to evaluate site conditions and identify optimal approaches
for conducting a remedial investigation (RI) of the BBM Site. The U.S. Environmental Protection
Agency's (EPA's) emphasis on the optimization of site investigation projects such as Black Butte is
rooted in an on-going program of evaluating operating remedies at Fund-lead sites. During fiscal years
2000 and 2001, independent Remediation System Evaluations (RSEs) were conducted at 20 operating
pump and treat (P&T) sites (those sites with P&T systems funded and managed under Superfund by the
EPA, other federal agencies, and by the states). In light of the opportunities for system optimization that
arose from those RSEs, the EPA Office of Superfund Remediation and Technology Innovation (OSRTI)
has incorporated RSEs into a larger post-construction complete strategy for Fund-lead remedies as
documented in Office of Solid Waste and Emergency Response (OSWER) Directive No. 9283.1-25,
Action Plan for Ground Water Remedy Optimization. Concurrently, the EPA developed and applied the
Triad Approach and related best management practices (BMPs) to optimize site characterization
strategies, methods and technologies. The Triad Approach and related BMPs include the increased use of
conceptual site models (CSMs) as the basis to identify project data gaps and focus on addressing data
gaps when developing site characterization objectives and work plans. The EPA has expanded the reach
of optimization to encompass reviews of projects at the investigation stage (such as for the BBM Site).
EPA OSWER and OSRTI define optimization as follows:
"Efforts at any phase of the removal or remedial response to identify and implement
actions that improve the action's effectiveness and cost-efficiency. Such actions may also
improve the remedy's protectiveness and long-term implementation which may facilitate
progress towards site completion. To identify these opportunities, regions may use a
systematic site review by a team of independent technical experts, apply techniques or
principles from Green Remediation or Triad, or apply some other approach to identify
opportunities for greater efficiency and effectiveness. Contractors, states, tribes, the
public, and PRPs are also encouraged to put forth opportunities for the Agency to
consider. "
Optimization reviews include a "systematic site review," whereby the site as a whole is often considered.
However, optimization can focus on a specific aspect of a given cleanup phase (or a particular operable
unit [OU]), with other phases and site areas considered to the degree that they affect the focus of the
optimization effort. For optimization reviews conducted before a Record of Decision (ROD) is issued, the
focus is on developing the conceptual site model (CSM) by leveraging existing data and exploring
potentially applicable sampling and analysis tools and strategies that facilitate a comprehensive
systematic planning process.
-------
A strong interest in sustainability has also developed in the private sector and within federal, state, and
municipal governments. Consistent with this interest, OSRTI has developed a methodology (EPA 2012)
for environmental footprint evaluation (http://www.cluin.org/greenremediation/methodologv/index.cfm).
and now routinely considers green remediation and environmental footprint reduction during optimization
reviews.
For a site in the investigation stage, the optimization review process includes reviewing site documents,
potentially visiting the site for 1 day, and compiling a report that includes recommendations for design
and execution of a comprehensive, efficient, and cost-effective investigation strategy.
The recommendations in this report are intended to help the site team identify opportunities for an
optimized RI approach. Where noted in this report, further analysis of a recommendation may be needed
before the recommendation can be implemented. The recommendations are based on an independent
evaluation and represent the opinions of the optimization review team. These recommendations do not
constitute requirements for future action, but rather are provided for consideration by the Region and
other site stakeholders. While the recommendations may provide some details to consider during
implementation, the recommendations are not meant to replace other, more comprehensive, planning
documents such as work plans, sampling plans, and quality assurance project plans (QAPP).
The national optimization strategy includes a system for tracking consideration and implementation of
optimization recommendations and includes a provision for follow-up technical assistance from the
optimization team as mutually agreed on by the site management team and EPA OSRTI.
The optimization review and site technical teams participated in a site visit and early systematic planning
from January 9 to 11, 2012. This optimization review report provides findings and recommendations
resulting from review of site documentation and data in conjunction with the site visit and systematic
planning efforts. Suggestions provided for sample numbers, collection and analytical methods, locations,
and other parameters may be adjusted to meet project specific schedule, budget, and logistical
considerations.
This document addresses the fate and transport of mercury and other trace metals at the BBM Site and the
CGR as a means to focus and streamline the sequence of RI activities. It is recognized that sampling for
metals (including mercury) and other parameters may be necessary to assess total risk and that sampling
to assess exposure routes and areas for human and ecological risk assessment are integral components of
any RI. Where appropriate and timely, suggestions address these considerations; however, it is expected
that this report will form the basis for additional systematic planning among the optimization review
team, project technical team, and stakeholders to develop, review, and finalize RI specific work planning
and implementation documents.
1.2 REVIEW TEAM COMPOSITION
The optimization review team consisted of the following individuals:
Name
Steve Dyment
Mark Shupe
J. Edward Surbrugg
Affiliation
EPA OSRTI
Terra Tech EMI
Terra Tech EMI
Phone
703-603-9903
703-390-0653
406-441-3269
Email
dyment. stephen@epa. gov
mark. shupe@tetratech. com
edward.surbrugg@tetratech.com
-------
1.3 DOCUMENTS REVIEWED
Key documents that provided the significant basis for the formulation of preliminary CSM components
include:
• Preliminary Assessment of Black Butte Mine (Anderson 1996): This document states the theory
that tailings pile erosion and particulate transport in surface water are the primary mode for off-
site mercury migration.
• Mercury Distribution in Sediments and Bioaccumulation by Fish in Two Oregon Reservoirs:
Point Source and Nonpoint Source Impacted Streams (Park and Curtis 1997): This document
compares mercury loading in CGR with the neighboring Dorena Reservoir, a watershed where
there are no known mercury mines. The results of the comparison suggest that the BBM site is a
significant point source of mercury contamination to CGR.
• Black Butte Mine Site Inspection Report (Ecology and Environment 1998): This report
documents the first comprehensive environmental investigation of the site. Soil, tailings,
sediments, surface water, and groundwater samples were collected and analyzed via conventional
laboratory methods for mercury and selected target analyte list (TAL) metals. Elevated (relative
to background) concentrations of mercury and sporadic elevated concentrations of some trace
metals were measured in soil, tailings, and sediments. Mercury was not detected above a
detection level of 200 nanograms per liter (ng/L) in any of the groundwater or surface water
samples, or in a sample collected from one of the BBM adits; however, these samples exceeded
background concentrations for some trace metals (antimony, chromium III, cobalt, copper,
manganese, and nickel exceeded background groundwater level concentrations).
• Final Report - Reconnaissance Soil Sampling at the Black Butte Mine (Curtis 2004): This
document describes an investigation to characterize mercury concentrations and speciation in
soils at the site and off-site on adjacent hillsides and ridge tops. Maximum off-site concentrations
are less than the Region 9 Preliminary Remediation Goal (PRG) of 23 milligrams per kilogram
(mg/kg). Selective sequential extraction (SSE) analyses of the off-site samples indicated that less
than 20 percent of the mercury was present in relatively insoluble mercuric sulfide forms and 44
to 87 percent of the mercury was complexed with organic matter which is more readily converted
to methylmercury.
• Mercury Levels and Relationships in Water, Sediment, and Fish Tissue in the Willamette Basin,
Oregon (Hope and Rubin, 2205): Sampling of surface water mercury concentrations in the CFW
River are discussed. Seasonal sampling upstream of the CGR shows that on average, 56 percent
of the total mercury (HgT) load was transported in the dissolved phase.
• Black Butte Mine, Removal Assessment Report (Ecology and Environment 2006): The
investigation documented in this report was conducted to generate a comprehensive site data set
to support a removal action (RA) to eliminate significant mercury releases from the site. This
investigation included X-ray fluorescence (XRF) and Lumex field-based analyses of mercury and
arsenic in soil, tailings, and sediment samples and conventional laboratory analysis of a subset of
these samples. In addition, synthetic precipitation leaching procedure (SPLP), methylmercury,
and SSE analyses were performed for a subset of these samples. Results indicate that the mercury
present in these materials generally occurs in insoluble forms that are not readily methylated.
• Final Removal Action Report for Black Butte Mine, Cottage Grove, Oregon (EPA 2008): This
report documents the 2007 RA, which consisted of the characterization of soils and tailings and
regrading, excavation, and on-site disposal of soils and tailings that exceeded site-specific
screening criteria. Excavated tailings that exceeded site-specific screening criteria were placed in
an on-site repository. Characterization sampling along Furnace Creek revealed the presence of an
extensive volume of tailings with mercury concentrations that exceeded the RA-defined action
-------
level for tailings adjacent to surface water. The Furnace Creek tailings were left in place as a
result of funding restrictions.
• Mercury Loading Assessment Results, Black Butte Mine (Thorns 2008): This post-RA surface
water loading assessment suggests that the transport of suspended solids (fine silt and clay size
particles) containing mercury persists as the primary mode of mercury transport to the Coastal
Fork Willamette (CFW) River and CGR. Based on one sampling campaign during baseflow
conditions, the assessment estimates that Furnace Creek could contribute between 50 and 75
percent of the mercury load in the CFW River. Re-calculation of this value by the optimization
review team suggests the contribution may be lower (26 to 59 percent); however, Furnace Creek
still represents a potentially significant source.
A complete list of information sources reviewed for the site is provided in Section 7.0. In addition,
historical aerial photographs, topographic maps, and geographic information system (GIS) shapefiles
were obtained and evaluated during this review.
1.4 QUALITY ASSURANCE
This optimization review uses existing environmental data to interpret the CSM, evaluate principal study
questions, identify data gaps, and support systematic planning to make recommendations for streamlined
sequencing, sampling, and analytical strategies. The quality of the existing data was evaluated by the
optimization review team before the data were used for these purposes. The evaluation for data quality
includes a brief review of how the data were collected and managed (where practical, the site QAPP is
considered), the consistency of the data with other site data, and the intended use of the data in the
optimization review. Data that were of suspect quality were either not used as part of the optimization
review or were used with the quality concerns noted. Where appropriate, this report provides
recommendations to improve data quality.
The key data source documents reviewed for this evaluation include Ecology and Environment (1998),
Curtis (2003), Curtis (2004), Ecology and Environment (2006), EPA (2008), and Thorns (2008). The data
from Ecology and Environment (1998 and 2006) and the Lumex mercury data from EPA (2008) were
subjected to validation in accordance with EPA (1990) and EPA (2004) guidance. The investigation
documented in Thorns (2008) included the collection of quality assurance /quality control (QA/QC)
samples of surface water and sediment. The data for this investigation, however, were not validated.
Information regarding validation of the data reported in Curtis (2003) and Curtis (2004) is unavailable in
these documents. Data quality issues regarding the Curtis (2004) data, however, were anecdotally
conveyed by the Oregon Department of Environmental Quality (ODEQ) to the project team. These issues
included sample extraction and analyses at separate laboratories and the lack of associated documentation.
Thorns (2008) reports mercury concentration results of total mercury in surface water for a field sample
(25.6 micrograms per liter (ug/L)) and a field duplicate (14.1 ug/L) collected in Furnace Creek
downstream from the Furnace Creek Tailings Area. Both samples exceed the total mercury measured in
other surface water samples collected during that investigation by more than 3 orders of magnitude. The
elevated result for both the field sample and its duplicate suggests that the two samples most likely reflect
the variability inherent in a grab sample.
Qualitative interpretation of these data indicates that the mercury level in Furnace Creek is inordinately
elevated relative to the levels in other surface water features in the BBM vicinity. The elevated Furnace
Creek total mercury concentrations are consistent with the close proximity of the stream to the old
furnace-derived tailings which, as discussed in the next section, contain elevated, more leachable forms of
mercury relative to other tailings at the site.
-------
1.5 PERSONS CONTACTED
The optimization review team met with stakeholders on January 9 through 11, 2012, at the ODEQ offices
in Eugene, Oregon. In addition to the optimization review team, the following persons were present for
the stakeholders meeting and include members of the project technical team:
Name
Richard Muza
Chris Eckley
Kira Lynch
Bernie Zavala
Kay Morrison
David Reisman
Bryn Thorns
Don Hanson
Affiliation
EPA Region 10 (remedial project
manager)
EPA Region 10
EPA Region 10
EPA Region 10
EPA Region 10
EPA Office of Research and
Development (ORD)
ODEQ
ODEQ
Email Address
muza.richard@epa. gov
ecklev.chris@epa. gov
lynch. kira@epa. gov
zavala.bernie@epa.gov
morrison.kay@epa. gov
reisman.david@epa.gov
thoms.bryn@deq.state.or.us
hanson.don@deq.state.or.us
On January 10, 2012, the individuals listed above (with the exception of Kay Morrison) and the
optimization review team toured the site.
-------
2.0 SITE BACKGROUND
2.1 LOCATION AND PRINCIPAL SITE FEATURES
BBM is located in Lane County Oregon in the CFW River Basin, approximately 10 miles upstream from
the CGR (see Figure 1). Located on the north face of Black Butte, the mine area is drained by Dennis
Creek, Garoutte Creek, and Furnace Creek which border the north, west, and south sides of the site (see
Figure 2). Both Dennis Creek and Furnace Creek are tributaries to Garoutte Creek which, after it joins the
CFW River, flows to the CGR. Principal site features include collapsed and open mine adits (entrances),
the Main Tailings Pile located adjacent to Dennis Creek, the Old Ore Furnace Area, the New Furnace
Area, and the Furnace Creek Tailings Area (Ecology and Environment 2006) (see Figure 2).
The CGR was constructed in 1942 as a U.S. Army Corps of Engineers (USAGE) flood-control reservoir.
The reservoir, which was created by constructing a dam on the CFW River, receives drainage from a
watershed with an area of approximately 99 square miles. Reservoir levels are decreased annually
between the end of September and November. The levels are decreased from the maximum pool level of
791 feet, National Geodetic Vertical Datum of 1929 (NGVD) until the low pool level of 750 feet, NGVD
is attained. The low pool level is maintained until early February, when management practice requires
that water levels begin increasing to the maximum pool level, which is attained in mid-May. At full pool,
the reservoir area is 1,158 acres; at low pool, the water area is reduced to approximately 25 percent of this
size.
2.2 SITE HISTORY
This section describes (1) historic land use and operations conducted at the site and (2) the chronology of
enforcement and remedial activities.
2.2.1 HISTORICAL LAND USE AND OPERATIONS
Before it was developed as a mercury mine, the site was undeveloped forest land. The mercury deposit of
concern at the site occurs primarily as cinnabar, a mercuric sulfide mineral. Elemental mercury was
produced on the site by heating crushed ore in a furnace where the elemental mercury was volatilized and
subsequently condensed and collected for bottling.
BBM was initially developed by S.P. Garoutte in 1890 with installation of a Scott-Hutner furnace to
process 40 tons/day of mercury ore. The Quicksilver Mining Company took ownership of the property in
1897 and expanded the mine operation and furnace capacity. BBM ceased operations in 1909 because of
declining mercury prices. A New York-based company (identity unknown) resumed mercury production
in 1916 with a redesigned Scott furnace. However, production was again suspended in 1919 because of
falling mercury prices. With installation of at least one rotary furnace during the period from 1927 to
1942, a new operator (Quicksilver Syndicate) increased mercury ore processing to 150 tons/day. Some
reprocessing of tailings from earlier operations also occurred during this time. Declining mercury prices
again resulted in mine closure in 1943. In 1956 and 1957, the mine was leased by Mercury & Chemicals
Corporation of New York. Information regarding ownership and production is unavailable in the years
from 1957 to the final abandonment of the mine in the late 1960s (Region 10 2008). The site was
purchased by Land and Timber Company in 1990 and is currently used for timber production and wildlife
habitat.
-------
During its peak operating years, BBM was the second largest mercury mine in Oregon. From 1900 to
1957, a total of 16,904 flasks of elemental mercury were produced at the mine (one flask equals 76
pounds) (Region 10 2008).
2.2.2 CHRONOLOGY OF ENFORCEMENT AND REMEDIAL ACTIVITIES
No enforcement actions were identified from the documents reviewed.
With regard to remedial activities, the site was identified as a potential source for mercury contamination
in CGR by the Mercury Working Group of the ODEQ Water Quality Division during an evaluation of
Oregon's lakes. This evaluation was focused on the analysis of mercury in fish tissue samples collected in
1993 and 1994. In 1994, the site was referred to ODEQ's Site Assessment Section (SAS) staff for review.
The SAS recommended further site investigation as a medium priority (ODEQ 1996). Pursuant to a
cooperative agreement with EPA Region 10, ODEQ conducted a preliminary assessment (PA) of the site
(ODEQ 1996). Region 10 completed site inspections in 1998 and removal assessment investigations in
2006. An RA was conducted at the site by Region 10 in 2007 (Region 10 2008), which consisted of soil
and tailings characterization, excavation, and relocation of mine wastes that exceeded site-specific action
levels; mine wastes that exceeded site-specific action levels were placed in a repository constructed on
site.
In 2009, Region 10 completed a Hazard Ranking System (HRS) evaluation for the site (Ecology and
Environment 2009). Results of the overland flow/flood component of the evaluation were sufficient to
qualify the site for inclusion on the NPL. Other pathways (groundwater, groundwater to surface water,
and soil) were excluded from the HRS evaluation because their effect on the overall ranking score was
negligible. As a result of the HRS evaluation, BBM was added to the NPL on March 4, 2010.
2.3 POTENTIAL HUMAN AND ECOLOGICAL RECEPTORS
Ecological hazards resulting from elevated mercury relate to the capability of the mercury to chemically
transition to its methylated form. Once present in the environment, methylmercury accumulates in fish
and other predators through their consumption of organisms lower in the food chain. Similarly, human
health exposure risks may be posed by the ingestion of fish, water fowl, or plants containing
methylmercury or the ingestion of mercury in surface water and groundwater.
The RI is expected to evaluate human health and ecological risks associated with a variety of media
potentially contaminated from BBM. The optimization review team considered the CSM in development
of the preliminary pathway receptor network diagram provided in Figure 3. The primary media and
receptors at the site include:
• Groundwater: potential contaminated discharge to surface water, human consumption of supply
well water;
• Soil: contact/ingestion by casual trespassers or forest workers. Ecological receptors include plant
uptake and birds and mammals that may ingest or contact soil;
• Surface water: fish, waterfowl contamination; human consumption;
• Sediment: fish, waterfowl contamination; human consumption; and
• Air: potential ground surface-to-air flux of elemental mercury.
-------
2.4 EXISTING DATA AND INFORMATION
This section presents information available from existing site documents. Interpretations included in this
section were generally extracted from the cited documents. Particular attention was paid to CSM elements
and conclusions that may warrant consideration during the Rl/feasibility study (FS). Existing data are
summarized sequentially for the BBM Site and CGR.
2.4.1 BBM SITE EXISTING DATA
The following sections present the existing data for the BBM Site. The sections describe sources of
contamination at the BBM Site, the geology and hydrogeology of the site, and existing analytical
characterization data for soil, groundwater, surface water, and sediment.
2.4.1.1 BBM SITE SOURCES OF CONTAMINATION
Known sources of contamination at the site are the tailings contained in the Main Tailings Pile and the
Furnace Creek Tailings Area (Figure 2). Available data suggest that mercury is not readily mobilized to
groundwater (Ecology and Environment 2006). Analytical results for samples collected from water
draining from mine adits were non-detect for mercury. The detection level associated with these analyses,
however, were relatively elevated (1,400 nanogram/liter [ng/L]) compared with the Garoutte Creek
background concentration of less than 0.5 ng/L). In addition, although synthetic leaching testing results
obtained from the tailings indicate a low propensity of the tailings to leach mercury, the testing was
conducted at an elevated detection level (440 ng/L) compared to the background mercury concentration in
groundwater (less than 200 ng/L) (Ecology and Environment 2006). As a result, the limited available data
show only that the tailings do not leach mercury to groundwater at elevated concentration levels in excess
of 3 times background values. Based on the available information, the relative proportions of dissolved
and suspended load mercury transport from the site are not known; however, limited seasonal sampling
from the CFW River upstream of the CGR suggests that suspended and dissolved mercury are roughly
equal contributors to the total mercury load. Since mercury is a dynamic contaminant, the relative
proportions of dissolved mercury and suspended mercury being exported from the mine site may differ
from what is being measured downstream and may vary in response to stormflow conditions. Dissolved
and suspended mercury samples collected from the CGR indicate that, during low pool periods, the
suspended concentration is actually higher than CFW River values, suggesting the importance of particle
entrainment as exposed sediments are eroded during low pool conditions (as shown in Figure 4). During
high pool conditions in the summer, mercury associated with suspended load sediments is transported
from the BBM and deposited in CGR.
2.4.1.2 BBM SITE GEOLOGY AND HYDROGEOLOGY
The following sections summarize site geology and hydrogeology and the environmental sampling results
from previous investigations. To evaluate the sampling results, measured constituent concentrations in
soil, tailings, sediment, surface water, and groundwater are compared with relevant screening levels.
Table 1 provides a combined listing of the screening levels referenced in this report.
-------
The site is located on a steep hillside underlain by thin soils directly underlain by bedrock. Accumulations
of recent alluvium occur in Garoutte Creek Valley, located at the foot of Black Butte. Bedrock consists of
hydrothermally altered andesitic lavas, silicic ash tuff, and volcanic breccias. Locally, these deposits have
been intruded by basalt and andesite. The bedrock is faulted and fractured. Black Butte fault is exposed
on the summit of Black Butte; the mercury ore deposit is believed to have formed along this fault (Derkey
1973).
The principal aquifer units in the vicinity of the site include (1) the bedrock aquifer, which occurs in the
fractured bedrock units underlying the hillside, and (2) the overlying floodplain aquifer, which occurs in
alluvium, where present, in the Dennis Creek, Furnace, and Garoutte Creek valleys. The two aquifers are
likely hydraulically connected, with the water table likely occurring in Garoutte Creek floodplain
sediments at the foot of Black Butte and in the shallow bedrock underlying the mine site (Anderson
1996).
2.4.1.3 BBM SITE SOIL CONTAMINATION
The materials present at ground surface at the site include native soils, waste rock or overburden, and
processed mine tailings. Only small amounts of waste rock or overburden exist on the site because most
of the mining occurred underground and within the ore-bearing materials. Previous characterization
results for native soil and waste/rock tailings are discussed in the following sections.
2.4.1.3.1 NATIVE SOIL
Mercury concentrations in native soil have been characterized on hillsides in the general vicinity of the
site (Curtis 2004) and on site at the New Furnace Area (Ecology and Environment 1998). Samples
collected near the tops of surrounding ridges facing the mine and analyzed via a fixed-based laboratory
contained up to 8 mg/kg total mercury. SSE conducted on these samples indicated that less than 20
percent of this mercury was found in the relatively insoluble mercury sulfide form and 44 to 87 percent of
the mercury was complexed with organic matter; the latter is more labile (Curtis 2004). Soil samples
collected from the New Furnace Area had elevated concentrations of mercury, arsenic and other metals
including chromium, copper, nickel, and vanadium (Ecology and Environment 1998). During the RA,
soils from the New Furnace Area were excavated, consolidated, and placed in an on-site capped
repository (EPA 2008).
2.4.1.3.2 WASTE ROCK/TAILINGS
Waste rock/tailings at the site are classified into two groups: those produced from the older furnaces
(referred to collectively as the Old Furnace), which operated prior to introduction of the New Furnace in
1927; and those produced as a result of the New Furnace operations. New Furnace tailings were derived
from ore processed in the rotary kiln furnace operated at the site from 1927 to 1942. The New Furnace
extracted a greater fraction of mercury from ore compared to the Old Furnace. As a result, New Furnace
tailings contain lower levels of mercury that are potentially leachable (Ecology and Environment 2006).
New Furnace tailings were disposed in the Main Tailings Pile. Old Furnace tailings were disposed in and
along Furnace Creek, referred to as the Furnace Creek Tailings Area. Samples collected from the two
locations were subjected to various analyses; results are discussed below.
Main Tailings Pile. Mercury concentrations in the Main Tailings Pile were generally less than 100 mg/kg
(Thorns 2008). Results from SPLP analysis of three tailings samples from the Main Tailings Pile and New
Furnace Area were non-detect to low concentration (ranging from 0.00149 to less than 0.002 milligrams
-------
per liter [mg/L] compared with the Oregon Leachate Reference Concentration of 0.2 mg/L), suggesting
that the dissolution of mercury from these materials is not a significant contributor to groundwater and
surface water (Ecology and Environment 2006). The noted detection levels (2,000 and 1,490 ng/L) are
elevated, however, relative to the observed background mercury concentrations in Garoutte Creek (below
0.5 ng/L); the tailings may, therefore, still leach mercury at lower but significant levels. The SPLP mimics
leaching caused by contact of the tailings with precipitation. Methylmercury analytical results for the
tailing pile equaled the ODEQ Level II soil screening level value (SLV) for plants of 0.0002 mg/kg
(Ecology and Environment 2006). SSE analyses indicated that mercury in the tailings generally exists as
relatively insoluble elemental mercury and mercuric sulfide species (Ecology and Environment 2006)
and, thus, exhibits a low potential for leaching.
Samples were analyzed for selected trace metals (Ecology and Environment 1998). All samples exceeded
the Region 9 PRG for arsenic (0.39 mg/kg); subsequent field-based XRF analysis confirmed elevated
arsenic concentrations.
Furnace Creek Tailings Area. Mercury concentrations in the Furnace Creek Tailings Area range from
approximately 400 to 2,000 mg/kg as determined via XRF (EPA 2008). Although SPLP analysis results
for three waste rock tailings samples from the Old Furnace Area were non-detect for mercury, the
detection level for the analyses was elevated (440 ng/L) (Ecology and Environment 2006). As a result, the
available data show only that the tailings may not leach mercury to groundwater at significantly elevated
concentrations (higher than 440 ng/L); the tailings may act as a source of mercury to groundwater and
surface water. Two Old Furnace Area tailings samples were analyzed for methylmercury; both exceeded
the screening level (0.0002 mg/kg) with the maximum concentration exceeding by a factor of 6 (Ecology
and Environment 2006). Results of SSE analysis indicate that the mercury species present in this area
have a higher solubility than in the Main Tailings Pile and an increased potential to form methylmercury
(Ecology and Environment 2006).
Samples analyzed for arsenic showed elevated concentrations (47 to 131 mg/kg) (Ecology and
Environment 2006). To date, these tailings have not been analyzed for metals other than mercury and
arsenic.
2.4.1.4 BBM SITE GROUNDWATER CONTAMINATION
Historical groundwater sampling at the site is limited to 11 locations, including an on-site spring, an on-
site well, a former mine adit, seven off-site wells, and an off-site spring (Ecology and Environment 1998).
All analyses were performed with a detection level of 200 ng/L. Mercury was not detected in any of the
11 samples, including the sample from the mine adit. Later sampling from a BBM adit confirmed a below
detection level result for mercury (Ecology and Environment 2006) (but the detection limit for the later
analysis was higher [1,400 ng/L]).
Analytical results for groundwater samples also exhibited elevated concentrations of one or more of the
following metals: arsenic, chromium, copper, lead, nickel, and selenium relative to a background well
(Ecology and Environment 1998).
2.4.1.5 BBM SITE SURFACE WATER CONTAMINATION
Furnace Creek and Dennis Creek surface water samples (analyzed for total metals) collected downstream
from the site exhibited elevated levels of mercury (based on fixed-based laboratory results). Samples
collected during the post-RA sampling event snowed that the downstream concentration exceeded
background by a factor of 570 for Furnace Creek and by a factor of 1.9 for Dennis Creek. In the
immediate vicinity of the site, downstream concentrations in Garoutte Creek were greater than
10
-------
background (Thorns 2008). The Garoutte Creek background total mercury concentration (0.5 ng/L) was
measured in a sample collected upstream from BBM (Thorns 2008). Mercury concentrations in all
downstream samples from Furnace Creek (current and historical) exceed the National Oceanic and
Atmospheric Administration (NOAA) Screening Quick Reference Tables (SQuiRT) threshold effects
level (TEL) and probable effects level (PEL) benchmark screening levels (see Table 1). Concentrations in
samples collected from Dennis Creek and Garoutte Creek were less than the SQuiRT TEL (Ecology and
Environment 2006). Samples from Dennis Creek and Garoutte Creek collected downstream from the site
were also analyzed for metals other than mercury (Ecology and Environment 1998) and concentrations
generally exceeded background. Analytical results for samples collected from Dennis Creek exceeded
benchmark screening levels for antimony, chromium, copper, lead, and nickel. Samples collected from
Garoutte Creek exceeded benchmark screening levels for lead and nickel and are similar in magnitude to
concentrations detected in samples collected from Dennis Creek.
2.4.1.6 BBM SITE SEDIMENTS
The following discussion of sediment characterization results summarizes the sediment quality as
characterized by fixed-base laboratory analysis.
Sediments samples collected from Furnace Creek and Dennis Creek downstream from the site
consistently exhibit elevated levels of mercury. Samples collected during the post-RA sampling event
showed that the downstream concentration exceeded background by a factor of 79 for Furnace Creek and
by a factor of 10 for Dennis Creek. In the immediate vicinity of the site, downstream sediment
concentrations in Garoutte Creek were elevated relative to background (Thorns 2008). In general,
concentrations in all three creeks exceed the NOAA SQuiRT TEL and PEL benchmark screening levels.
A Furnace Creek sediment sample collected downstream from the Old Furnace Area exhibited an
elevated methylmercury concentration (0.0127 J mg/kg), which is two orders of magnitude higher than
the ODEQ Level II soil screening level for plants (0.0002 mg/kg) (Ecology and Environment 2006). The
methylmercury concentration for a Garoutte Creek sediment sample collected downstream from the
confluence of Dennis Creek and Garoutte Creek was less than the ODEQ Level II soil screening level.
SSE analysis results for the above noted Furnace Creek and Garoutte Creek sediment samples indicated
that the majority of the mercury exists in relatively insoluble forms that are not readily methylated
(converted to methylmercury) (Ecology and Environment 2006).
Concentrations of metals other than mercury in downstream sediment samples are similar to background
levels. Arsenic, chromium, copper, nickel, and zinc exceeded benchmark screening levels in both
upstream and downstream samples (Ecology and Environment 1998).
2.4.2 CGR EXISTING DATA
Contamination sources and the existing analytical characterization data for the CGR are discussed in the
following sections. Primary investigations used to develop the sections below include Curtis and Allen-
Gil (1994), Curtis and Park (1996), and Curtis (2003). The sampled media include sediment, surface
water, and fish tissue.
2.4.2.1 CGR CONTAMINATION SOURCES
Total mercury is sourced to CGR as suspended load (particulate mercury [HgP], meaning mercury
associated with suspended sediment particles) and as dissolved load, including dissolved inorganic and
organic (methylated) species. Elevated mercury in the tissue of the fish that inhabit CGR may originate
from several sources, both internal and external to the reservoir. To bioaccumulate, mercury must be
present as methylmercury which is formed through the life processes of bacteria. Given appropriately
11
-------
reducing conditions and sufficient supplies of dissolved mercury, sulfate, and labile organic carbon, the
bacteria convert dissolved inorganic mercury into methylmercury. In addition, methylmercury, under
certain conditions, may revert back to dissolved inorganic species through a process called demethylation.
2.4.2.1.1 EXTERNAL SOURCES
The primary external sources of mercury to CGR include suspended and dissolved mercury from BBM
and the Garoutte Creek watershed at large, suspended and dissolved mercury from the CGR watershed at
large, and deposition of atmospheric mercury attributable to various industrial and mining sources (such
as coal-fired power plants) world-wide (also referred to as the global mercury pool).
BBM and the Garoutte Creek Watershed. The contributions from the BBM Site and the Garoutte
Creek Watershed at large will be assessed in the BBM portion of the RI. CGR receives drainage from a
watershed with an area of approximately 99 square miles. By comparison, the area of the Garoutte Creek
Watershed, is approximately 17 square miles, or approximately 20 percent of the parent CGR Watershed.
As noted previously, Thorns (2008) estimated that the Furnace Creek (a tributary to Garoutte Creek)
contributes up to 50 to 75 percent of the mercury load in the CFW River. Re-calculation of this value by
the project team suggests the contribution may be lower (26 to 59 percent); however, it still represents a
potentially significant source. If actual loading approaches the upper end of the loading estimates, the 20
percent of the CGR Watershed drained by Garoutte Creek contributes a disproportionately large fraction
of the total mercury load to the CGR.
CGR Watershed at Large. CGR is situated within the Black Butte-Elkhead Mercury District (Thorns
2008) and, as such, some amount of mercury is naturally released to the environment through weathering
of mineralized bedrock and soil. Park and Curtis (1997) indicate that, for the Dorena Reservoir located
approximately 10 miles northeast of CGR but also within the mercury-mineralized district, mercury levels
in sediments and fish tissue were elevated, but were approximately 2/3 lower than the levels observed in
CGR.
Global Mercury Pool. Natural sources of mercury include volcanoes, geothermal activity, wildfires, and
the weathering of rocks and soils. As a result of global industrialization, anthropogenic sources have
increased the atmospheric mercury load. A 2005 global inventory of mercury emissions (United Nations
Environmental Programme [UNEP] 2008) estimates that anthropogenic loading is in the same range as
the loading from natural sources (1,930 compared with 1,600 metric tons per year). UNEP estimates that
atmospheric gaseous elemental mercury concentrations globally range from 1.1 to 4 nanograms per cubic
meter (ng/m3).
2.4.2.1.2 INTERNAL SOURCES
Internal sources of mercury to the CGR contribute mercury to the water column and sediments within the
reservoir. A potentially significant internal source is CFW River erosion and mobilization of historical,
mercury-contaminated sediments during low pool conditions. Evidence for this internal loading source
derives from analysis of mercury in sediment cores from the CGR and observed conditions during the site
visit.
Three sediment cores have been collected from CGR, one in 1995 and two in 2002 (Curtis 2003). All
cores were collected from the deepest portion of the reservoir and did not penetrate the full thickness of
sediment. Based on lead isotope and cesium analysis, a mean sedimentation rate of 0.37 inch/year is
estimated for CGR. Mercury concentrations were measured in the three cores in approximately 4.0-inch
intervals. The results indicate that, with the exception of a spike in approximately 1998, mercury
concentrations in the sediments deposited over the period from 1980 through 2002 were relatively stable
12
-------
(ranging from 0.5 to 1.0 mg/kg), suggesting that mercury input has remained relatively constant. The
maximum mercury concentrations (2.0 to 3.5 mg/kg) in the cores were measured in sediments deposited
in the early 1970s.
Conditions observed during the January 10, 2012, site visit suggest that the historical mercury-
contaminated sediments are likely subjected to erosion and mobilization during the low pool conditions.
As shown in the photos taken during the site visit stop at the Wilson Creek Boat Ramp (dry at the time of
the visit) (Appendix A), the CFW River was observed flowing in a narrow channel floored by bedrock.
The channel was incised into lake-bottom sediments that appear to consist predominantly of fine sand,
silt, and clay. The sediments appeared to range from 3 to 5 feet thick. Based on observations from the
boat ramp, the river appeared to be actively eroding and mobilizing the lake-bottom sediments. Once
mobilized, the sediments would be transported by the river to the low pool portion of the reservoir, where
they would be deposited. Through the process of erosion, transport, and deposition, any mercury
contained in the historical sediments could be converted to dissolved form and subjected to methylation.
2.4.2.2 CGR SEDIMENT
Sediment characterization results from the Curtis and Allen-Gil (1994), Park and Curtis (1997), and
Curtis (2003) investigations are summarized below.
Curtis and Allen-Gil (1994) measured mercury concentrations in CGR sediment in September 1989, June
1990, September 1990, and September 1992. The September events correspond to the end of the high
pool season, while the June event represents conditions approximately 1 month after the high pool
conditions became established. The same two locations were sampled for each event, both located in
shallow water: one near the dam, and the other midway between the CRW River entry point to the
reservoir and the dam. From the available information in the report, however, it is unclear to which of the
sampling points the reported data specifically applied. Notwithstanding this uncertainty, the average
concentration for the two locations ranged from a high of 0.53 mg/kg for the September 1989 event to a
low of 0.06 mg/kg for the June 1990 event.
In comparing CGR with the neighboring Dorena Reservoir, Park and Curtis (1997) collected six sediment
samples along the long axis of the CGR from the CFW River entry point to a point just upstream from the
dam. The sampling was performed in March 1994, presumably under low pool conditions. Mercury
concentrations ranged from 1.11 mg/kg for the point upstream from the dam to 0.18 mg/kg at a location
near Wilson Creek, in the high pool portion of the reservoir. In general, the most elevated concentrations
were observed in the three low pool area samples, which increased in concentration toward the dam (0.68,
1.03, and 1.11 mg/kg). The concentration for the CFW River entry point sample (0.83 mg/kg) was in the
same range as the three low pool area samples.
In an event that essentially served to resample five of six Park and Curtis (1997) locations, Curtis (2003)
collected sediment samples distributed along the long axis of CGR. Sampling was performed between
July and September 2002, presumably under high pool conditions. The results confirmed the sampling
conducted by Park and Curtis (2003). Specifically, mercury concentrations increased from a low of
0.68 mg/kg at the upstream end of the reservoir in the high pool portion of the reservoir to a high of 3.6
mg/kg in the low pool area.
2.4.2.3 CGR SURFACE WATER
CGR surface water mercury concentrations were evaluated by ODEQ for the Total Maximum Daily Load
(TMDL) assessment of the Willamette Basin (ODEQ 2006) and through sampling events conducted in
2002 and 2003. For the Willamette Basin TMDL, ODEQ (2006) estimates that the ambient mercury load
13
-------
supplied by the CGR watershed is 3.13 kilograms per year (kg/yr) and that the average annual water
column mercury concentration is 0.92 ng/L.
Based on data available in the ODEQ online Laboratory Analytical Storage and Retrieval (LASAR)
database (http://deql2.deq.state.or.us/lasar2/). ODEQ measured total suspended solids (TSS), total
mercury, and dissolved mercury concentrations in four monitoring events spanning the period from mid-
fall 2002 to late spring 2003. Measureable TSS concentrations, ranging from 5.3 to 8.2 mg/L, were
observed for the three events corresponding to low pool conditions (October, December, and March). TSS
was non-detect for the one high pool condition event in mid-June. The occurrence of measurable TSS
concentrations during low pool conditions likely reflects erosion of the lake-bottom sediments exposed
during low water level periods. Total mercury concentrations are elevated for the low pool sampling
events in comparison to the mid-June event. The average low pool total mercury concentration was 6.5
ng/L compared with 1 ng/L for the mid-June sample. Dissolved mercury concentrations appeared less
dependent on pool level. The low pool samples averaged 1.7 ng/L compared with 1 ng/L for the mid-June
sample.
2.4.2.4 CGR FISH TISSUE
CGR was the first body of water in Oregon to be placed under a public health advisory as a result of
elevated mercury concentrations in fish tissue. In 1979, the Lane County Health Department, in
consultation with the Oregon Health Division, issued a public health advisory suggesting safe
consumption limits for fish caught from the reservoir. CGR fish tissue testing performed by ODEQ in
2003 indicated an average mercury concentration in fish tissue of 0.53 micrograms per gram (ug/g).
Mercury levels measured by ODEQ in bass from CGR ranged up to 1.6 ug/g. The Oregon Department of
Human Services generally issues mercury advisories for fish from a water body if average tissue
concentrations exceed 0.35 ug/g (http://www.oregon.gov/DHS/news/2004news/2004-0422.shtml).
Curtis and Allen-Gil (1994), Park and Curtis (1997), and Curtis (2003) also provide fish tissue sampling
results for CGR. With the exception of one sample, tissue samples from 12 bass bioassays from 1990 and
1992 sampling events conducted by Curtis and Allen-Gil (1994) exceeded the health advisory level
(0.35 ug/g). For the 1990 sampling event, concentrations in bass tissue ranged from 0.22 to 1.79 ug/g and
averaged 0.86 ug/g. For the 1992 event, concentrations ranged from 0.37 to 0.74 and averaged 0.51 ug/g.
From fish samples obtained in June 1993, September 1994, July 1995, and November 1995, Park and
Curtis (1997) obtained concentrations in bass tissue ranging from approximately 0.25 to 0.87 ug/g.
Results of their investigation indicated that mercury concentrations in bass tissue increase with fish age
and that concentrations varied and trended differently for each sampling event. Curtis (2003) measured
mercury concentrations in bass tissue in August and September 2002. Concentrations ranged from 0.86 to
1.2 ug/g and averaged 1.1 ug/g. Curtis (2003) also measured mercury concentrations in benthic
invertebrate tissue and determined that average mercury concentrations were more than an order of
magnitude less than the average observed for bass tissue. These results suggest increased bioaccumulation
of mercury associated with a higher position in the food chain of CGR.
14
-------
3.0 DESCRIPTION OF PLANNED OR EXISTING REMEDIES
An RA was implemented at the BBM Site in 2007 to address the potential for erosion of tailings and
associated release of mercury via surface water transport of tailings and soil particles to downstream
surface water bodies (EPA 2008). The RA consisted of characterization of soils and tailings. Soils that
exceeded site specific action levels were excavated, and placed in an on-site repository (Figure 2).
The RA's principal objectives were to stabilize the Main Tailings Pile slope adjacent to Dennis Creek,
stabilize the Old Ore Furnace tailings area near Furnace Creek, and cap site areas with potential to
contribute mercury to site surface water and sediments. Near Dennis Creek, the Main Tailings Pile slope
was reduced through regrading to minimize tailings spillage into the creek. The tailings generated by this
re-grading were either placed in the on-site repository or were used as capping material in other portions
of the site, as discussed below.
Three action levels, based on the EPA Region 9 PRG for mercury in residential soil (23 mg/kg, dermal
contact), SSE analysis results, and mercury background sediment concentrations (Ecology and
Environment 2006) were used in the RA (Table 1). If field mercury analysis (via XRF or Lumex)
indicated concentrations at a location exceeded the applicable action level, the area was either capped or
the material exhibiting the elevated concentrations was excavated and placed in the on-site repository.
The three action levels included the following:
• The EPA Region 9 PRG value (23 mg/kg) was used for areas where sequential extraction
analyses indicated that mercury was present in more soluble forms; the Old Furnace area was
evaluated using this action level.
• For areas where the SSE analysis indicated that the mercury was generally present only in the less
soluble mercuric sulfide form, an action level of 115 mg/kg was applied; the Main Tailings Pile
and the New Furnace Area (SSE indicated that mercury was present in low solubility forms in
both areas) were evaluated using this action level.
• An action level of 10 mg/kg (three times site background mercury concentration in sediment) was
used for site areas where tailings were susceptible to erosion and particulate transport in surface
water. This action level was applied to the Main Tailings Pile in close proximity to Dennis Creek,
the Old Ore Furnace Area, and to delineate tailings located in the Furnace Creek Tailings Area.
Because the SPLP results from the Main Tailings Pile suggested the tailings in this pile were not leaching
mercury at elevated concentrations in groundwater (above 440 ng/L) and since the SSE results indicated
that the mercury in Main Tailings Pile tailings generally occurs in relatively insoluble forms, these
tailings were used as "clean" fill in the RA. Main Tailings Pile tailings were, therefore, used to cap the
New Furnace Area and Old Ore Furnace Area.
The RA also included an assessment of Furnace Creek; the results indicated the presence of a larger than
anticipated volume of tailings that, based on SSE analysis results, contained mercury in more leachable
forms. As a result of funding limitations and because the tailings did not present an imminent risk to
human health and the environment, however, the tailings located in the Furnace Creek Tailings Area were
left in place (EPA 2008). The Furnace Creek Tailings Area may therefore still represent a potentially
significant source of mercury contamination to Garoutte Creek, CFW, and CGR
15
-------
4.0 CONCEPTUAL SITE MODEL
This section discusses the optimization review team's interpretation of historical information, existing
characterization, and RA data to explain how historical events and site characteristics have led to current
conditions. Section 4.1 provides a summary of the optimization review team's interpretation of the
preliminary CSM, and subsequent sections provide additional detail to help support that interpretation.
Identified data gaps are discussed in Section 4.4, while findings and recommendations associated with
sequencing RI activities are provided in Sections 5 and 6.
Currently, the preliminary CSM includes two key elements, (1) the release and transport of mercury from
the BBM Site and (2) the mercury methylation processes in the CGR. These elements are presented in
Sections 4.1 and 4.2. It is also recognized that other components of the CSM such as mercury on
neighboring slopes and storage/transport/methylation processes within the CFW may be important factors
to consider. To the extent possible these CSM elements are included in considerations and
recommendations for optimizing RI efforts discussed in Section 4.4 of this report.
The site and CGR CSM elements collectively reflect the current best available understanding of how
mercury is released, transported, and converted to methylmercury. The CSM elements include only those
processes that are believed to be dominant and controlling for mercury fate and transport. For example,
although methylation may be occurring upstream from CGR, the CSM assumes that the dominant
methylation processes responsible for elevated mercury in fish tissue occur within the CGR itself.
Similarly, although mercury could originate from other sources at the site, such as contributions from
groundwater or by overland flow from adjacent hillsides, the Furnace Creek Tailings Area is assumed to
be the dominant mercury source. Only the most plausible processes were included in the CSM. Other
processes will be evaluated and incorporated into the CSM as necessary through RI sampling and data
evaluation. Processes not represented in the CSM will be evaluated in the RI through the data evaluation
logic discussed in Section 4.4. Through the data evaluation process, the CSM will evolve and mature
from the current preliminary form into the baseline CSM for the site systematic project planning (SPP)
efforts.
The elements of the CSM for the site and the CGR are consistent with previous models developed and
presented in the site documents reviewed, including:
• The proposal by the U.S. Geological Survey (USGS) to USAGE for estimation of the mercury
budget for CGR (USGS 2011);
• The ODEQ mercury loading analysis conducted in 2007 (Thorns 2008);
• The ODEQ mercury loading estimate for the Willamette River Basin TMDL Project (Hope 2003)
4.1 CSM COMPONENTS FOR BBM
In an ODEQ preliminary assessment of the BBM Site, Andersen (1996) indicates that surface water is the
most significant migration pathway for mercury from the site. He suggests that "surface transport of soil
and sediment (from the BBM Site) has resulted in widespread contamination distribution downstream of
the mine site in Dennis Creek, the Coast Fork of the Willamette River, and in CGR." In accordance with
the information obtained from document review and from discussions with Region 10 and ODEQ
personnel, this CSM was been refined as follows:
16
-------
• The BBM Site, specifically the Furnace Creek Tailings Area, may represent a significant source
of mercury contamination to Furnace, Garoutte Creeks, the CFW River, and the CGR. Potentially
important site sources of suspended and dissolved mercury contamination include physical
erosion of tailings piles and mine soils, surface water transport of the eroded tailings and soil
particles, and deposition of the sediments in downstream surface water features. Leaching from
existing tailings or dissolution of mercury, originating as suspended mercury, also likely occurs
along the transport pathway.
The preliminary CSM is consistent with the results of the surface water and sediment sampling and
analysis conducted by Thorns (2008). As discussed in previous sections, based on one sampling event of
flow representative of non-storm conditions (Thorns 2008), it was estimated that Furnace Creek could
contribute between 50 and 75 percent of the mercury load in the CFW River. Re-calculation of this value
by the optimization review team suggests that the contribution may be lower (26 to 59 percent); however,
Furnace Creek still represents a potentially significant source. Figure 4 is a schematic illustration of these
processes.
A primary motivation for addressing mercury contamination from BBM is to minimize or eliminate the
currently existing conditions that led to elevated mercury concentrations in the tissue of the fish that
inhabit the CGR and the watershed in general. As discussed further in Section 5.0 (Findings), however,
there are significant challenges associated with accurately determining the relative role of the BBM Site
in causing the elevated concentrations of mercury in fish tissue.
Key elements of the CSM describing dominant processes responsible for the release and transport of
mercury form the BBM Site and supported by site investigation documents include:
• The Furnace Creek Tailings Area is a key source for elevated mercury concentrations in
downstream surface water features, including CGR.
• Fine grained, mercury-laden tailings particles are generated from mechanical erosion and are
transported via surface water flow into Furnace Creek.
• Mercury in sediments present in Furnace Creek bed may also be remobilized and resuspended
during storm flow conditions.
• A portion of the eroded particles contributes to the suspended load in the surface water flows in
Furnace Creek, Garoutte Creek, CFW River, and CGR.
• Similar processes in Dennis Creek and historical impacts in Garoutte Creek along with
background watershed contributions result in lower contributions of mercury to CFW and CGR;
and
• Mercury is delivered to CGR in both dissolved mercury and suspended mercury phases consisting
of both inorganic and methylated forms. Geochemical conditions in CGR bottom waters and
sediments are favorable for bacteria that can convert dissolved mercury to methylmercury, which
is then available for uptake into the food chain. Some of the suspended mercury in the CGR may
be converted to dissolved mercury over time.
4.2 CSM COMPONENTS FOR CGR
The CSM element describing the dominant processes responsible for the occurrence of methylmercury in
the CGR includes the following components:
• Methylmercury in CGR can originate from internal and external production. Internal methylation
is influenced by the seasonal changes in CGR water levels. Existing management practices set the
17
-------
water level at a high point during the spring and summer months and at a low level during the fall
and winter.
• The current CSM and limited existing data indicate that the Furnace Creek Tailings Area is likely
an important source of contemporary loading of mercury and potentially other metals to
downstream surface water features, including CGR. It is important to note that more than a
century of inputs have historically loaded the CFW and subsequently the CGR through its 60-year
existence. Based on CGR sediment coring results, Curtis (2003) reports that sediment mercury
concentrations were up to three times greater in the early 1970s as compared with more recent
concentrations. The CSM recognizes, therefore, that significant mercury concentrations exist in
CGR as a result of historical loading.
• A potential internal source of mercury to the CGR is CFW River erosion of historical, elevated
concentration lake bottom sediments exposed during low pool conditions. The historical
sediments thus mobilized are carried via suspended load to the low pool portion of the CGR,
where some of the sediment is deposited.
• Mercury methylation occurs primarily in the uppermost sediment layers and the anoxic section of
the water column in the reservoir; deeper sediments are effectively isolated from the methylation
process.
• Mercury is supplied to the uppermost sediments via transport of suspended tailings and particles
in incoming surface water or from the re-erosion of previously deposited sediments exposed
during low pool conditions.
• Sedimentation occurs within the low pool portion of CGR during the late fall and winter season
and throughout CGR during the high pool conditions (spring and summer season).
• Geochemical conditions at the bottom of CGR are favorable for conversion of the mercury
contained in the tailings to methylmercury. The methylmercury levels in CGR sediments are a
small percentage of the total inorganic mercury load, indicating that an understanding of the small
sub-fraction available for methylation remains an uncertainty in understanding the mercury
dynamics in CGR.
• Mercury methylation requires anoxic conditions, the presence of inorganic mercury in a dissolved
bioavailable form, a microbial labile organic carbon source, and sulfate. Understanding which of
these key variables is controlling methylmercury production in CGR can provide important
information necessary to consider reservoir management strategies as a means to control
methylation.
• Methylation activity is enhanced in the sediments that are re-submerged after exposure to the
atmosphere during low pool conditions. During atmospheric exposure, the sulfide is oxidized to
sulfate, allowing the resumption of methylation during high pool conditions.
4.3 IDENTIFICATION OF DATA GAPS
The CSM is the primary tool to identify significant data gaps in the existing site information. Data gaps
identified from the site and CGR CSM elements, and considerations for filling those data gaps, are
discussed in the following sections.
4.3.1 SITE DATA GAPS
Data gaps associated with the site exist with respect to the occurrence and variation of mercury in creeks
near the site and in the downstream surface water features. Data gaps also exist with respect to the
capacity of the tailings and mercury-contaminated mine soils to leach dissolved mercury or other metals
to groundwater and surface water, to the accuracy and reliability of field-based methods (XRF and
Lumex) for measuring metals concentrations in site media, and to the potential presence of soil
18
-------
contamination in the vicinity of the tailings disposal areas. Additional details regarding each of these data
gaps follow.
Surface Water Flow and Quality. A significant data gap is the need to gain an understanding of the
chemical and physical form of mercury as it is released from the site and the role that stream flow
conditions play. To address this data gap, data are needed regarding mercury concentrations (measured by
chemical species and form [total versus dissolved]) as well as concentrations for other ancillary water
quality parameters in surface water in Dennis Creek, Furnace Creek, and Garoutte Creek. At one station
on each stream, sampling should be performed during periods of baseflow as well as periods with storm
water runoff (the stormflow hydrograph for a given station and stream). Ideally, this sampling should
include up to two baseflow and stormflow events representing the range of annual antecedent moisture
and flow conditions. In addition, data are needed at locations up and downstream from the BBM Site to
evaluate the seasonal variation of mercury (speciation and form) and ancillary parameter concentrations
during both storm and non-storm conditions. Data generated from these events could be used to confirm
the results of the 2008 post-RA surface water loading assessment (Thorns 2008) and to verify whether the
bulk of the loading occurs during storm flow events.
Tailings Leaching Capacity and Groundwater Quality. Additional data are needed regarding the
overall capacity of the site tailings and mine soils to leach mercury and other metals into groundwater and
the role, if any, of groundwater in the off-site migration of mercury. Also unknown is the contribution to
the site groundwater mercury flux from transient groundwater flow, defined as vadose zone groundwater
recharged from a specific precipitation event and in transit to the water table. Data are needed regarding
mercury and trace metal concentrations in groundwater beneath/immediately downgradient from the Main
Tailings Pile and the Furnace Creek Tailings Area. These data are necessary to evaluate the significance
of groundwater as a pathway for off-site mercury migration.
Demonstration of Method Applicability (DMA) for Field-Based Soil Analyses. Additional data are
needed regarding the correlation of mercury and metals concentrations measured using field-based
methods (XRF and Lumex) and concentrations measured via fixed-based laboratory methods. In
accordance with BMPs, a DMA consisting of the collection of samples to be homogenized and split for
field-based and fixed-base laboratory analysis from selected site areas (including the Old Ore Furnace
Capped Area, New Furnace Capped Area, and adjacent hillsides) and analysis of mercury and metals
could be conducted. The results of these paired analyses could be used to establish the correlation
between field-based and fixed-based laboratory analyses and assess the general applicability of the field-
based methods for RI activities. In addition, a DMA data can provide preliminary information to optimize
analytical and sampling strategies to address human health and ecological risk assessment needs.
Information on conducting DMAs and establishing relationships for collaborative analytical methods is
provided in EPA (2008a), available from the following web address:
http://www.brownfieldstsc.org/pdfs/Demonstrations of Methods Applicability.pdf.
Furnace Creek Tailings Trace Metal Concentrations. A data gap exists regarding trace metal
concentrations in the tailings disposed in the Furnace Creek Tailings Area. These tailings are believed to
have been generated as a result of Old Ore Furnace operations. Trace metal concentrations should be
measured in samples of tailings from the Old Ore Furnace Area and from the Furnace Creek Tailings
Area. An additional data gap exists regarding confirmation of the extent of the Furnace Creek Tailings
Area, as does generation of a site tailings data set sufficient to meet requirements for risk assessment.
General Site Soil. A data gap exists regarding trace metal concentrations in soil underlying the Main
Tailings Pile, the tailings in the vicinity of the Old Ore Furnace, and the site in general. Mercury and other
trace metal concentrations should be measured in site soil samples. In addition, a data gap exists regarding
confirmation sampling of the results obtained by Curtis (2004) for methylmercury on hillsides adjacent to
19
-------
Black Butte. Generation of a site soil data set sufficient to meet requirements for risk assessment is also a
data gap.
Sediment. Additional sediment data are needed to establish baseline concentrations for sediments in
Garoutte Creek near BBM and to assess the potential for stream sediments to contribute to suspended
loading of mercury and other metals in surface water. After baseline conditions in Garoutte Creek have
been established, more comprehensive sediment characterization may be necessary to satisfy the data
requirements for risk assessment. Similarly, if important for remedial design, the team may consider
downstream sediment sampling in Garoutte Creek or CFW River to understand potential sinks and
mercury transformation processes as mercury moves to CGR.
4.3.2 CGR DATA GAPS
CGR water levels are seasonally managed for flood control, conservation storage, and water release to
downstream areas. Low water levels are maintained in the winter months, while water levels are held at
higher levels during the summer. Significant data gaps exist in the understanding of the origin of the
elevated total mercury concentrations measured in the tissues of fish inhabiting the CGR. Fundamental to
understanding fish tissue concentrations is an accurate characterization of the overall mercury budget for
the reservoir. Specifically, the relative importance of internally versus externally generated
methylmercury is unknown. In addition, data gaps exist regarding the methylation processes operating
internally within the CGR. For example, the data needed to compare the significance of methylation in the
low pool sediments with methylation in the wetland areas exposed during low pool but inundated during
high pool are unavailable (Figure 4). Uncertainty also exists regarding the time horizon needed to assess
the beneficial impact on fish tissue concentrations from any actions taken to reduce the methylmercury
flux from sources internal or external to the reservoir.
Given the above data gaps in the existing CGR characterization, several fundamental data gaps may be
addressed during the RI. The resulting data will support assessment of the benefits associated with
changes in reservoir management practices or future remedial actions at the BBM Site that have the
potential to reduce future mercury concentrations in fish tissue.
Total Mercury Concentrations for Range of Trophic Levels. Data are needed regarding how total
tissue mercury concentrations vary with food web trophic level. Total mercury concentrations in biota
should increase with trophic position. Organisms such as snails or other invertebrates at the lower bound
of the food chain should have low total mercury concentrations, while the tissue of predatory fish at the
upper bound should have elevated total mercury concentrations. The trophic sampling may potentially
support risk assessment needs and is expected to support the identification of appropriate species for
continued tissue sampling as a means to evaluate potential reductions of mercury in tissues achieved
through source mitigation or reservoir management measures. Where appropriate, fish species identified
for trophic level sampling will focus on CGR species potentially supporting recreational or sustenance
consumption.
Baseline Total Mercury Concentrations in Upper Trophic Level Fish. In the event that the trophic-
level sampling results indicate that mercury concentrations in fish tissues are unrelated to trophic position,
baseline total mercury concentrations should be measured in upper level fish. As indicated above, this
sampling may potentially support risk assessment and would be intended to provide the basis for
determining mercury reductions in fish tissues achieved by potential source mitigation or CGR cleanup
measures.
Mercury In-Flow/Out-Flow Budget. To date, a mercury budget has not been established for CGR;
however, the USGS is collecting the data needed to establish a mercury budget for CGR (USGS 2011).
20
-------
This analysis will reflect existing (baseline) conditions for the watershed. If source mitigation measures
are implemented at BBM, this sampling can be extended to document any mercury loading reductions
achieved.
4.4 IMPLICATIONS FOR REMEDIAL STRATEGY
Suggested RI sampling strategies and associated data evaluation logic developed by the optimization
review team are presented in this section. Table 2 summarizes the proposed number of samples, sampling
approach, and analytes for each CSM element (the BBM Site and the CGR) and potentially affected
environmental medium. A key objective of this optimization review effort is source identification and
characterization; the media critical to this objective include surface water, sediment, shallow
groundwater, and native soil. Although considered noncritical for source characterization, other media
(for example, CFW River surface water) will be critical for the human health and ecologic risk
evaluations that will be based on the data collected in the RI. General recommendations are provided in
Table 2 for consideration in evaluating these media.
It should be noted that the recommended sampling approaches and numbers of samples provided are
preliminary - that is, they should be considered to be a starting point for planning the RI. In addition,
qualitative judgments govern each decision point (such as, "elevated" sampling result). The project
technical team may determine appropriate quantitative judgments for these decision points. Final
sampling design and specification of quantitative decision point parameters will require input from all
stakeholders (including Region 10 human health and ecologic risk assessment staff) involved in the RI.
The focus of this optimization review was the identification and sequencing of sampling approaches and
associated decision logic to guide future RI efforts at the BBM Site. Data collection approaches and
evaluation strategies are also presented for CGR, but with the goal of establishing a baseline data set that
can be used to assess the effects of source mitigation or reservoir management measures and that can be
used to help assess potential candidate management practices that could improve conditions in the
reservoir.
4.4.1 RECOMMENDATIONS FOR RI IMPLEMENTATION AT THE BBM SITE
Sampling to test and confirm the site CSM should include the collection and analysis of:
• Baseflow and storm event surface water samples from the creeks in the immediate vicinity of the
site (Dennis, Furnace, and Garoutte).
• Groundwater samples from saturated unconsolidated material underlying on-site tailings
(sampling conducted concurrent with the storm event surface water sampling);
• Precipitation (rate and concentration);
• Sediment samples from these creeks;
• Native soils potentially contaminated by site operations; and
• Tailings from the Furnace Creek Tailings Area and the Main Tailings Pile.
Sampling locations, rationale, and data evaluation logic are presented in the following sections for each of
these media.
4.4.1.1 SEQUENCING OF CHARACTERIZATION BY MEDIA
The recommended RI characterization sequence consists of three phases as follows:
21
-------
• Phase 1 includes the initial groundwater characterization and the storm event sampling. These
tasks are prerequisites for the seasonal groundwater and sediment sampling and for establishing
the appropriate timing for the seasonal storm surface water grab sampling. In addition, the Phase
1 tailings and soil field analyses and associated fixed-base laboratory analysis may also be used
for conducting the DMA for XRF and Lumex field analyses methods.
• Phase 2 consists of concurrent seasonal surface water, sediment, and groundwater grab sampling
during storm and non-storm stream flow events. Phase 2 would also include baseline soil and
tailings sampling and analysis, and the DMA for XRF and Lumex field analysis. These baseline
soil samples would be collected from the site soils and adjacent hillsides as well as the tailings
from the Furnace Creek Tailings Area. Evaluation of Phase 2 results will guide the design of
Phase 3 sampling.
• If the CSM is supported by the sampling from Phase 2, Phase 3 sampling will consist of detailed
characterization of the Furnace Creek Tailings Area. If the Phase 2 sampling suggests that a
source other than the Furnace Creek Tailings Area is a significant contributor of mercury and
other metals contamination to the environment, the focus of the sampling may be adjusted
accordingly (for example, to focus on hillside soil).
4.4.1.2 SURFACE WATER CHARACTERIZATION
Surface water characterization is proposed to address the data gaps identified in the existing surface water
characterization for the site (see Section 4.4.1). The primary objective of the surface water sampling
would be to evaluate the significance of suspended mercury and dissolved mercury transport from BBM
to area creeks. The data collected can help quantify annual loading from Dennis and Furnace Creeks to
Garoutte Creek, help identify areas of mercury inputs originating on the BBM property, and provide
pathway information on mercury transport and transformation to and within these creeks. This sampling
and the associated data evaluation may establish the potential existence of other significant mercury
sources at, or in the immediate vicinity of, the site. The 2008 surface water mercury loading estimate
completed by ODEQ (ODEQ 2008) identified the Furnace Creek Tailings Area as a significant
contributor of mercury in the CFW River and potentially CGR. The surface water sampling approach
proposed in this section is based in part on the recommendations developed from that study.
4.4.1.2.1 SAMPLING APPROACH
The optimization review team recommends collection of surface water samples from Garoutte Creek,
Furnace Creek, and Dennis Creek at the following eight locations (see Figure 5):
• Garoutte Creek (Station 1) - upstream of the confluence with Furnace Creek,
• Garoutte Creek (Station 2) - downstream of the confluence of Furnace Creek in a location
upstream of the confluence with Dennis Creek,
• Garoutte Creek (Station 3) - downstream of the confluence of Dennis Creek,
• Garoutte Creek (Station 4) - upstream of where Garoutte Creek and Big River merge to form the
CFW River
• Furnace Creek (Station 5) - upstream of the site,
• Furnace Creek (Station 6) - downstream of the Furnace Creek Tailings Area,
• Dennis Creek (Station 7) - upstream of the site, and
• Dennis Creek (Station 8) - downstream of the Main Tailings Pile.
22
-------
Station 1 will serve as a reference location for water quality upstream from the site. Sampling that occurs
quarterly for 1 year is recommended. During each season (quarter), one sampling event occurring during
storm flow conditions and one sampling event conducted during normal and low flow conditions is
recommended. The proposed eight sampling locations and the eight events (two per quarter for 1 year)
will generate 64 annual samples, plus appropriate QC samples.
Samples should be analyzed for total mercury, dissolved mercury, methylmercury (total), methylmercury
(dissolved), reactive mercury (total), reactive mercury (dissolved), and target analyte list (TAL) metals
(total), as well as total suspended solids (TSS), pH, dissolved organic carbon (DOC), and common ions.
This combination of analytes will support the estimation of mercury and trace metal concentrations in
dissolved and suspended phases and will provide insight on the speciation of the suspended and dissolved
phases. Results of the reactive mercury analysis will indicate the relative proportions of a sample in which
the mercury is present in less reactive species (such as mercuric sulfide) versus the amount present in
more reactive, mobile species (such as organically complexed mercury). Reactive mercury analyses will
help in understanding the hillside mercury loading (likely organically complexed) versus tailings-sourced
loading (likely dominated by mercuric sulfide species). These results will be significant to understanding
mercury sourcing, particularly if collection of samples of vadose zone groundwater in the hillslope and
tailings areas is difficult to achieve.
Furnace Creek and Dennis Creek discharge rates should be continuously monitored using a water level
transducer and a weir structure. Garoutte Creek discharge can be gauged using direct measurement
methods or can be estimated using USGS gauging data, if available. Consistent with the stream sampling
method used by the USGS in the ongoing project to evaluate the mercury flux in CGR, the depth/width
composite sampling method can be evaluated for use in this project during preparation of the RI work
plan.
To evaluate the influence of storm events on mercury transport (including the "first flush"), preliminary
sampling and gauging can be conducted at one station on each creek (Furnace, Dennis, and Garoutte).
Ideally, this preliminary sampling and gauging would be conducted during two storm events (preferably
with differing antecedent conditions) before surface water grab sampling is initiated for the first season.
During first flush, mercury loads may be disproportionately elevated. Determination of first flush timing
for each creek before seasonal grab samples are collected will enable the project team to optimize timing
of the storm flow grab sampling to coincide with the period of peak loads. To characterize the mercury
flux response to precipitation events for each creek, creek discharge, total mercury, dissolved mercury,
methylmercury (total), methylmercury (dissolved), reactive mercury (total) (analyzed for a sub-set of
samples), and TAL metals (total phase-only), as well as TSS, pH, DOC, and common ions are
recommended. This sampling and data collection should be conducted at a regular time interval sufficient
to accurately characterize creek discharge and chemistry through the period of each storm. To account for
atmospheric mercury contributions in project data evaluations, precipitation amount and chemistry
samples can be collected and analyzed during each precipitation event for most of the same analytes as
the stream samples. Ideally, the precipitation amount would be continuously monitored throughout the
year to help with the interpretation of the stream discharge data.
During the stormflow hydrograph sampling task, coincident sampling of the eight vadose zone
piezometers installed in accordance with Section 4.4.1.4.1 can be conducted at a frequency similar to
collection of stream samples. The vadose zone groundwater sample analysis can include dissolved
mercury, methylmercury (dissolved), reactive mercury (dissolved), TAL metals (dissolved), pH, DOC,
and common ions.
Sampling would be initialized based on the observed trend in stream discharge for each storm event. After
the event begins, stream discharge will begin to increase after some lag time depending on the intensity
23
-------
and proximity of the rain event. Chemical sampling should begin when the discharge rate begins to
increase in response to the storm. The specific intervals of sampling can be identified at a later point
based on changes in discharge from the first flush event. Sampling and flow measurements would end
during the declining phase of a station's response to the given storm.
4.4.1.2.2 DATA EVALUATION LOGIC
Figure 6 shows the suggested data evaluation logic for surface water sampling at the BBM Site. As
indicated, the surface water data would be evaluated using five decision points denoted by diamonds on
the figure. The logic associated with each decision point is discussed below:
• Is the downstream Garoutte Creek mercury flux greater than the upstream Garoutte Creek
flux at BBM? A "yes" result is consistent with the CSM (the mercury flux from Furnace Creek
is reflected by an increase in the downstream Garoutte Creek flux). A "no" result is inconsistent
with the CSM and suggests that Furnace and Dennis Creeks may not be conveying significant
mercury flux to Garoutte Creek. Given a "no" result, a discussion with the technical team and
Region 10 management would be necessary to determine the appropriate path forward for the RI.
Is the Furnace Creek mercury flux large compared to the Garoutte Creek mercury flux? A
"yes" result is consistent with the CSM. A "no" result is inconsistent with the CSM. A potential
alternative source is mercury contained in the runoff from hillsides in the vicinity that may have
been historically contaminated by airborne mercury emissions from the site's former ore
processing operations. Another potential mercury source is from seepage of contaminated
groundwater to surface water. Results from the groundwater sampling task will assist in the
evaluation of a "no" result at this decision point. Given a "no" result, the path forward would be
to complete the surface water data evaluation process and then proceed to the soil data evaluation
logic diagram (Figure 12) to design an appropriate soil sampling strategy. The comparison of the
Furnace and Garoutte Creek mercury fluxes will likely include multiple decision criteria since the
flow in Furnace Creek is a fraction of the flow in Garoutte Creek. Two criteria that may be
applicable include a straight comparison of the mercury fluxes for the two creeks and a
comparison of the downstream versus the upstream Furnace Creek flux. For example, if the
downstream Furnace Creek flux exceeds the upstream flux by at least 3-fold, then the Furnace
Creek would be considered a significant mercury source. Similarly, if no significant Garoutte
Creek concentration change is observed between the Furnace Creek downstream sample and the
Dennis Creek downstream sample the likelihood of Dennis Creek being a major source of surface
water mercury flux to Garoutte Creek is low.
• Are suspended load mercury concentrations elevated relative to dissolved load
concentrations? A "yes" result is consistent with the CSM (the dominant source of total
mercury in downstream surface water is from the mechanical erosion of fine tailings particles
from the Furnace Creek Tailings Area). A "no" result is inconsistent with the CSM and suggests
that the mercury flux resulting from groundwater discharge to surface water is significant relative
to the mechanical erosion of tailings at the Furnace Creek Tailings Area. Similar to the previous
decision point, the results from the groundwater sampling task will assist in this evaluation. Given
a "no" at this decision point, the path forward is to complete the groundwater data evaluation and
proceed to the soil data evaluation logic diagram (Figure 12) to design an appropriate soil
sampling strategy.
• Do suspended particle mercury concentrations in downstream Garoutte Creek approximate
sediment concentrations in Furnace Creek? The primary input to this decision point is
24
-------
determined by comparing the Garoutte Creek suspended particle mercury concentration (derived
using the Garoutte Creek suspended load mercury and TSS concentrations measured just
downstream of Furnace Creek but upstream of Dennis Creek) to the Furnace Creek mercury
concentration in fine sediment. If the Garoutte Creek suspended mercury concentrations
approximate the fine sediment concentrations in Furnace Creek, the CSM is supported. (This
result provides evidence that the downstream Garoutte Creek suspended load concentrations
likely originated from Furnace Creek.) If the Garoutte Creek suspended particle concentrations
are significantly dissimilar to the Furnace Creek fine sediment mercury concentrations and more
closely resemble Garoutte Creek fine sediment concentrations, internal loading within Garoutte
Creek is likely occurring, a result that is counter to the CSM. Given a "no" result at this decision
point, the path forward is to complete the surface water data evaluation process and proceed to
the sediment data evaluation logic diagram (Figure 7) to design an appropriate sediment sampling
strategy.
How do Furnace Creek mercury speciation results compare with Garoutte Creek speciation
results? The primary input to this decision point is determined by estimating the concentration
of suspended particulate-bound, non-mobile mercury for the Garoutte and Furnace Creek
samples. This estimate is calculated as follows:
Suspended particulate-bound non-mobile Hg =
[HgT-HgD] - [MeHg(total)-MeHg(dissolved] - [HgR(total)-HgR(dissolved)]
Where -
HgT: total mercury
HgD: dissolved mercury
MeHg(total): total methylmercury
MeHg(dissolved): dissolved methylmercury
HgR(total): total reactive mercury
HgR(dissolved): dissolved reactive mercury
A good correlation between the downstream Furnace Creek sample (Figure 5, Station 6) and the
nearest downstream Garoutte Creek station (Figure 5, Station 2) supports the CSM (evidence
exists that the mercury observed in Garoutte Creek is significantly sourced to the Furnace Creek
Tailings Area), whereas a poor correlation is unsupportive (this result suggests an alternative
source exists such as mercury sourced to runoff from adjacent hillsides or from groundwater
discharge). If the observed correlation is poor, the path forward is to proceed to the soil data
evaluation logic diagram (Figure 12) to design an appropriate soil sampling strategy.
After the surface water and groundwater sampling tasks have been completed (discussed later in this
section), it is recommended that the combined data set be subjected to the End Member Mixing Analysis
(EMMA) data analysis technique (Gary and others 2011). The EMMA is recommended as a check on the
data evaluation results obtained from the decision logic described in this section.
EMMA assumes that creek water is a mixture of waters supplied by distinct components of the watershed,
each with distinct concentrations of naturally occurring ions. The EMMA uses observed surface water
geochemistry to trace the contributions of these watershed components to total creek flow. The EMMA
will use the common ion and general chemistry constituent concentrations measured in the surface water
and groundwater samples.
25
-------
4.4.1.3 SEDIMENT CHARACTERIZATION
The objectives of sediment characterization are: (1) to provide total mercury concentration data for
comparison with the calculated suspended mercury concentrations (to assess potential internal loading in
Garoutte Creek); (2) to provide general characterization data regarding the temporal and spatial variability
of total mercury and methylmercury in sediments in the vicinity of the BBM Site; and (3) to provide
information that can be used to support risk assessment in accordance with risk assessor-defined data
needs.
4.4.1.3.1 SAMPLING APPROACH
The optimization review team recommends collection and analysis of sediment samples from each of the
surface water sampling locations during each of the eight surface water sampling events (see Figure 5).
Samples collected using incremental composite sampling (ICS) methods (Appendix B) are recommended
to control short scale heterogeneity (large differences in concentration in close spatial proximity). These
samples can be biased toward finer grain sizes that are potentially more readily mobilized during storm
flow conditions, and be analyzed for total mercury, methylmercury, TOC, TAL metals, and grain size.
The sediment data will be used in combination with the surface water data to assess the possibility that the
surface water suspended mercury load is significantly influenced by mobilized historical creek sediments
versus from erosion and mobilization of fine tailings particles from the Furnace Creek Tailings Area.
4.4.1.3.2 DATA EVALUATION
Figure 7 shows the data evaluation logic for sediment sampling. Evaluation of the sediment data should
proceed once all surface water and sediment sampling has been completed. As shown on Figure 7, the
sediment and surface water data can be evaluated in combination as follows:
• Are the suspended sediment mercury concentrations in downstream Garoutte Creek similar
to the sediment concentrations in Furnace Creek? The calculation of suspended mercury
concentration for each station was discussed in the surface water sampling data logic (Section
6.2). A "yes" result is consistent with the CSM (evidence exists that Furnace Creek is the
dominant source of suspended mercury in Garoutte Creek). If the Garoutte Creek suspended
mercury concentration more closely resembles the mercury concentration in Garoutte Creek
sediment (a "no" result), suspended mercury in Garoutte Creek is likely the result of internal
loading (remobilization of Garoutte Creek bed load sediments). After consultation with Region
10, the development of a comprehensive sediment characterization plan may be required to
address internal loading within Garoutte Creek.
4.4.1.4 GROUNDWATER CHARACTERIZATION
Consistent with the data gaps identified in the CSM (Section 4.4.1), the objectives of the recommended
groundwater characterization task are (1) to directly assess the potential for tailings to leach mercury and
other metals to groundwater, and (2) to provide groundwater characterization results to support the
evaluation of surface water sampling data. It is suggested that the task be conducted in two phases. Phase
1 consists of initial soil and groundwater characterization sampling of the transient (vadose) and phreatic
(saturated) zones and installation and sampling of temporary monitoring wells. Phase 2 would then
consist of collection of seasonal groundwater grab samples coinciding with surface water sampling
events.
26
-------
4.4.1.4.1 SAMPLING APPROACH
Vadose Zone Groundwater. Up to nine piezometers are recommended at the BBM Site and on the
opposing hillside on the opposite side of Garoutte Creek from BBM to evaluate transient groundwater
flow during storm and non-storm events. It is recommended that a direct-push technology (DPT) drilling
approach be used as the method for installing the vadose zone piezometers. However, a mini sonic or
other drilling platform may be appropriate if geologic conditions adverse to DPT drilling are encountered.
The nine suggested locations, eight of which are shown on Figure 2, include:
• Two locations in the Main Tailings Pile adjacent to Dennis Creek,
• Two locations in the Furnace Creek Tailings Area adjacent to Furnace Creek,
• Two locations in the Main Tailings Pile at the approximate ridge crest separating the Dennis and
Furnace Creek drainages,
• Two locations on the hillside opposite Garoutte Creek from BBM, and
• One background location (not shown on Figure 2), unaffected by the site.
The one background sampling location (for installation of up to three piezometer wells) should be defined
in consultation with the project team before the groundwater sampling tasks begin and with regard to
appropriate security and access considerations.
Up to two piezometers are recommended at each location, the first installed with the base of the screen
interval coinciding with the bedrock surface and the second screened in a shallower zone selected based
on field conditions (such as evidence of perched groundwater conditions). In the absence of any evidence
of perched groundwater, only one piezometer may be installed. Groundwater samples will be collected (if
sufficient sample volume can be obtained) from the piezometers during each of the surface water grab
sampling events. During the piezometer well installation task, use of drive-point or hand augered soil
borings will be evaluated to assess, to the extent possible, the potential that saturated tailings exist at the
Main Tailings Pile adjacent to Dennis Creek and the Furnace Creek Tailings Area adjacent to Furnace
Creek. If the presence of saturated tailings is identified at either location, an additional piezometer well is
recommended at that location.
Vadose zone groundwater samples analyzed for dissolved mercury, methylmercury (dissolved), reactive
mercury (dissolved), TAL metals (dissolved), as well as pH, DOC and common ions are recommended.
Tailings samples are recommended for collection during the advancement of each piezometer borehole.
Sampling is recommended on a 3-foot sampling interval for mercury and other metals analyses by XRF
and Lumex. A percentage (10-20 percent is recommended) of these samples, representative of the range
of observed field concentrations, may also be submitted for fixed-base laboratory analysis of total
mercury and TAL metals.
Saturated Zone Groundwater. Saturated groundwater can be characterized through installation of eight
temporary monitoring wells in the Garoutte Creek floodplain located down slope from the BBM. In
addition, three staff gauges situated in close proximity to the monitoring wells can be installed in the
creek. It is recommended that a rotary sonic DPT drilling approach (for example, Geoprobe Model 8140
or equivalent) be used as the method for installing the wells. Prior to groundwater sampling, the area
should be cleared of vegetation and surface soils should be mapped by visual inspection. Soil boring
installation is recommended, with first priority given to any identified tailings areas. A subset of borings
will also be installed in non-tailings areas. The lithology of soil borings should be logged and soils
sampled and analyzed for total mercury and metals analyses via XRF and Lumex. A percentage (10-20
percent is recommended), representative of the range of observed field concentrations, may also be
27
-------
submitted for fixed-base laboratory analysis of total mercury and TAL metals. Representative samples
(up to three) of tailings and of the unconsolidated sediments underlying the floodplain are also
recommended for grain size analysis.
If the presence of tailings is confirmed, a minimum of eight locations are recommended for drive point
ground-water and soil sampling from the tailings areas with an equal number of these samples collected
from the non-tailings areas. Temporary monitoring wells, sufficiently durable to withstand multiple
sampling events over a 1 year period, can be installed at eight of the drive point soil and groundwater
sampling locations. Up to three locations are recommended from the immediate vicinity of the confluence
of Dennis Creek and Garoutte Creek (and, if possible Furnace Creek and Garoutte Creek); the remaining
samples should be collected from the general floodplain area. Figure 8 shows recommended preliminary
locations for groundwater sample collection. The locations shown may be modified to address access and
drilling logistics. Each drive point soil and groundwater sampling location, the temporary monitoring
wells, and the three staff gauges should be surveyed for Oregon state plane coordinates (to an accuracy of
0.1 foot). Ground surface elevations for drive point soil and groundwater sampling locations should be
surveyed to an accuracy of 0.01 foot relative to NGVD; the reference elevation of each staff gauge and
the top of casing elevation (relative to NGVD) for each monitoring well should be surveyed to an
accuracy of 0.001 foot.
Background Garoutte Creek floodplain groundwater quality can be characterized by sampling two to
three locations on the Garoutte Creek floodplain upstream from, and unaffected by, the BBM. Before the
groundwater sampling tasks begin, the background well locations should be defined in consultation with
the project team and with consideration given to the availability of appropriate security and property
access requirements. Up to three background wells are recommended. The wells would be sampled
regularly along with the other wells and piezometers.
Groundwater samples collected from the eight temporary monitoring wells on a quarterly basis are
recommended. To the extent possible, sampling should be timed to coincide with the seasonal surface
water grab sampling as a means to conserve resources and limit mobilizations. The recommended analyte
list for the unconsolidated-material, saturated-zone groundwater samples includes dissolved mercury,
methylmercury (dissolved), reactive mercury (dissolved), and TAL metals (dissolved), as well as pH,
DOC, and common ions.
4.4.1.4.2 DATA EVALUATION
Vadose Zone Groundwater. Figure 9 shows the data evaluation logic for vadose zone piezometer
installation and the review of vadose zone groundwater sampling and analytical results. Evaluation of the
transient groundwater data should proceed concurrently with the surface water data evaluation. As shown
on Figure 9, evaluation of the groundwater data includes two decision points:
• During baseflow conditions, is evidence for perched groundwater observed in the soil core?
Soil cores can be obtained during the installation of the piezometers at each monitoring location.
If evidence of perched conditions is present in at least one of the cores retrieved during drilling,
installation of one piezometer is recommended such that its screen interval monitors the perched
zone, and the other piezometer installed such that the base of its screen interval coincides with the
bedrock surface. If no evidence of perched conditions is observed, only one piezometer is
recommended.
• During stormflow and non-stormflow conditions, is there evidence of vadose zone saturated
flow and/or overland flow and are vadose zone concentrations elevated? Measurement of the
water level and groundwater chemistry within piezometers is recommended during stormflow,
and, if sufficient water is present for sampling, non-stormflow conditions. Understanding the
28
-------
hydrology of the mine site during precipitation events is critical for identifying how contaminants
may be transported to the area's streams. Understanding the variable source area of saturation
near Dennis and Furnace Creeks during storm events is necessary to characterize the area over
which mechanical erosion of the tailings may be occurring. The concentration data from the
vadose zone groundwater samples will be compared with background. If mercury concentrations
are similar to background, this result would support the CSM. Elevated concentrations suggest
potential vadose zone mercury loading to surface water. If mercury loading is confirmed,
additional vadose zone characterization sampling, designed in consultation with Region 10, may
be necessary to estimate mercury and other metals mass loading to Furnace, Garoutte, and Dennis
Creeks.
Saturated Zone Groundwater. Figure 10 shows the recommended data evaluation logic for temporary
monitoring well installation and the review of groundwater sampling results. Evaluation of the
groundwater data should proceed concurrently with the surface water data evaluation. As shown on
Figure 10, evaluation of the groundwater data includes three decision points:
• Presence of tailings confirmed? After surface mapping and drive point soil sampling of the
Garoutte Creek floodplain, the first decision point seeks confirmation regarding the presence of
tailings. If tailings are present, groundwater sampling may partially focus on the tailings areas. If
tailings are absent, groundwater sampling should focus on the floodplain areas in close proximity
to the Dennis and Furnace Creek valleys. Placement of sampling locations in these areas assumes
that a greater bedrock fracture density is present and thus an increased likelihood exists that the
groundwater samples from these areas may capture potential groundwater impacts from the BBM
Site.
• What levels of mercury concentration are detected? After the initial groundwater
characterization and collection of the seasonal groundwater grab samples, the total mercury
concentration in groundwater should be evaluated relative to the background level. If the mercury
concentration in the floodplain groundwater samples is similar to background, the CSM is
confirmed (the tailings areas are not significantly contaminating groundwater). If the
concentrations are elevated, the groundwater to surface water mercury flux should be calculated.
• Is the total mercury flux in groundwater elevated compared with the Garoutte Creek total
mercury flux? A "no" result is consistent with the CSM (mechanical erosion of tailings from the
Furnace Creek tailings area is the dominant source of mercury loading to Garoutte Creek). A
"yes" result is inconsistent with the CSM, as it suggests that mercury contamination in
groundwater is a major contributor to mercury loading in Garoutte Creek. If the mass flux is
elevated relative to the mercury flux in Garoutte Creek, the groundwater flux may be considered a
significant contributor the mercury flux in Garoutte Creek. Given this result, RI data collection
and subsequently FS evaluations may need to consider groundwater source mitigation measures.
As a result of the challenges associated with characterizing and identifying effective remedial
approaches in fractured bedrock terrain such as exists at the BBM Site, Region 10 risk
management assessment and decisions would be necessary to identify the appropriate path
forward, given this outcome.
4.4.1.5 TAILINGS CHARACTERIZATION
It is recommended that an initial tailings characterization task be performed during the Phase 1
groundwater and surface water characterization sampling events. After the Phase 1 data have been
evaluated and the importance of tailings to identified impacts in surface water and groundwater have been
29
-------
considered, higher density sampling of tailings may be considered, particularly in support of identified
risk assessment data needs. The objectives of the initial tailings investigation are:
• To establish the relative strength of the correlation between XRF and Lumex field-based metals
analysis results with fixed-base laboratory analytical results, and
• To better characterize the thickness and areal extent of tailings in the Furnace Creek Tailings
Area.
As data requirements for human health and ecological risk assessment are considered, tailings sampling to
address these objectives may be combined or included as a second phase of tailings investigation.
4.4.1.5.1 SAMPLING APPROACH
The recommended tailings sampling approach includes a focused DMA, followed by sampling and
analysis for broader site characterization sampling. The DMA can be conducted to establish the relative
strength of the correlation between XRF and Lumex field metals analysis with fixed-base laboratory
analyses on a set of paired samples. The samples evaluated in the DMA should include tailings and native
soil samples across a range of expected concentrations (based on existing data). Data for the DMA can be
generated from the tailings and soil sampling components of the groundwater sampling tasks (see Section
4.4.1.4). The project team may choose to focus DMA-related sampling on one or the other of these two
media as determined by the data evaluation logic discussed in the next section.
Tailings can be investigated using an adaptive approach in which initial sampling locations for field
analyses are selected before field sampling begins and follow-up field sampling locations are selected
based on real-time analysis results to target uncertainties or anomalies. The initial tailings sampling
locations should be distributed along specific transects so that a broad characterization (including both
elevated and low concentrations) of spatial patterns is established for the site. To further address short
scale heterogeneity (large differences in concentration in close spatial proximity) at transect points, use of
ICS (Appendix A) for fixed-base laboratory analysis or XRF/Lumex field analysis may be performed in a
grid configuration around selected transect points.
Alternatively, the combined Furnace Creek and Main Tailings Areas (approximately 27 acres, Figure 2)
may be subdivided into decision units (DUs) and ICS conducted on each DU to satisfy general
characterization needs and to generate data potentially appropriate for risk assessment purposes.
Appropriate DU delineation is critical to the ICS approach. DUs should be defined via the systematic
planning process such that risk characterization objectives are achieved with the optimal number of
required samples. ICS samples analyzed for fixed base analyses of total mercury, methylmercury, TAL
metals, and grain size are recommended.
4.4.1.5.2 DATA EVALUATION
Figure 11 shows the data evaluation logic for tailings characterization. At the decision point, the degree to
which the CSM is confirmed by the Phase 2 surface water, sediment, and groundwater data
characterization is assessed. If the CSM is supported by this characterization, the DMA and subsequent
site characterization can focus on tailings, with soils characterization as a secondary focus. Specifically,
the characterization priority should delineate tailings in the Furnace Creek Tailings Area. If the data
evaluation does not support the CSM, the DMA and subsequent site characterization activities can focus
on soil, with a secondary focus on tailings characterization. Specifically, the characterization priority will
be the identification and delineation of contaminated soil areas that are potentially a significant source for
the release of mercury from the site.
30
-------
4.4.1.6 SOIL INVESTIGATION
It is recommended that the soil characterization task be performed following the evaluation of the data
generated by the Phase 2 surface water, sediment, groundwater, and tailings characterization tasks.
Similar to the tailings characterization, the focus and objective of the soil characterization will depend on
how closely the CSM is supported by the data from the other media.
4.4.1.6.1 SAMPLING APPROACH
At a minimum, surface soil samples collected for XRF, LUMEX and fixed-base laboratory analysis in
sufficient quantities to meet human health and ecological risk characterization requirements are
recommended. Additional sampling may be necessary to characterize potential alternative sources of
mercury contamination once data from surface water, groundwater, and sediment are assessed.
Soil sampling can be conducted using an adaptive approach in which initial sampling locations for field
analyses are selected before field sampling begins and subsequent field sampling locations use real-time
analysis to target uncertainties or anomalies. The initial soil sampling locations can be distributed along
specific transects so that broad characterization (including both elevated and low concentrations) of
spatial patterns is established for the site. To further address short-scale heterogeneity at transect points,
use of ICS (Appendix A) for fixed-base laboratory analysis or XRF/Lumex grids around transect points
should be considered.
Alternatively, broad application of the ICS sampling approach may satisfy general characterization needs
and generate data potentially appropriate for risk assessment purposes. For the ICS approach, the BBM
Site vicinity (Figure 11 a) may be defined based on topography and potential for airborne deposition of
elemental mercury that may have occurred during ore processing operations. Curtis (2004) collected soil
samples from the hillsides adjacent to Black Butte for a soil sampling event that encompassed a several
square mile area centered on the BBM Site and determined that mercury concentrations, although below
the EPA Region 9 screening level of 23 mg/kg, were comparably more elevated on the hillsides facing
Black Butte than facing the opposite direction. The larger of the two areas shown on Figure lla includes
the adjacent hillsides in the general Black Butte vicinity, while the smaller includes the hillsides
immediately adjacent to the site. Given the closer proximity to the airborne mercury source, the smaller
area (825 acres) may warrant smaller DUs compared with the larger area (2,900 acres). Appropriate DU
delineation is critical to the ICS approach. DUs should be defined via the systematic planning process
such that risk characterization objectives are achieved with the optimal number of required samples. ICS
soil samples should be analyzed via a fixed base laboratory for total mercury, methylmercury, and TAL
metals.
Soil column profile sampling may be considered as an approach for delineating OU boundaries for the
BBM Site. The BBM was situated in an area in which the local geology is naturally enriched in mercury.
Other zones of mercury mineralization likely exist on the adjoining hillsides. Soil sampling can be
performed with the aim of distinguishing between mercury sources (natural geologic versus attributable to
BBM emissions). One potential approach to meet this objective would be to collect soil samples using a
hand-held soil corer. Mercury concentration data from the surface samples and samples from the base of
each core could distinguish between areas with only elevated surface mercury concentrations (attributed
to atmospheric inputs) versus areas that also, or exclusively, have elevated subsurface concentrations
reflecting geologic sources from weathered bedrock. Identifying the zone of contamination attributable to
the BBM will assist with delineating the boundaries of the OU containing the mine site.
4.4.1.6.2 DATA EVALUATION
31
-------
Figure 12 shows the data evaluation logic for soil sampling. Given confirmation of the CSM (that the
Furnace Creek Tailings Area is the dominant source for off-site mercury migration), the primary objective
of the soil characterization task will be to meet the sampling requirements of the human health and
ecologic risk assessments. If the CSM is not supported, the soil sampling may, in addition to meeting the
requirements of the risk assessments, also characterize potential alternative sources of mercury
contamination, including the soils at the site and on the adjacent hillsides.
4.4.2 RECOMMENDATIONS FOR RI IMPLEMENTATION AT THE CGR
The objectives of the recommended RI characterization activities at CGR are to (1) generate
characterization data from various site media (sediment, sediment pore water, surface water) to enable
better understand factors controlling the production of methylmercury; (2) establish baseline levels for the
representative biota populations; (3) implement ongoing monitoring of the mercury concentrations in the
populations should source mitigation measures be implemented; and (4) define the conditions for which
follow-up detailed evaluations of various CGR media are appropriate.
4.4.2.1 CGR ENVIRONMENTAL MEDIA CHARACTERIZATION
The collection of water, sediment, and sediment pore water data are necessary to develop an
understanding of the factors controlling the production of methylmercury in the CGR. Mercury
methylation likely occurs in at least two subareas of the CGR:
• The anoxic water column and deep bottom sediments in the low pool portion of the reservoir, and
• The sediment/wetland areas submerged only during high pool conditions.
Figure 12a shows the proposed sampling locations for the CGR investigation. Sampling approaches for
each medium are discussed in the following sections followed by integrated data evaluation logic for all
media.
4.4.2.1.1 WATER SAMPLING
It is recommended that CGR water samples be collected and analyzed quarterly for 1 year. Samples may
be collected in January and March to reflect low pool and end-of-low pool conditions. Similarly, samples
collected and analyzed in July and September would correspond to high pool and end-of-high pool
conditions. Methylation is expected to occur in the anoxic, basal layer of water (or hypolimnion) in a
thermally stratified lake such as CGR. The surficial, oxygen-enriched layer of water is known as the
epilimnion. Accordingly, before the samples are collected, field parameter profiling, including standard
field parameters (oxidation-reduction potential [ORP], pH, dissolved oxygen [DO], temperature, and
specific conductance [SC]) are recommended to identify the most reducing depth horizon. A water
sample collected at each sampling location from both the epilimnion and the hypolimnion layers and
analyzed for the parameters indicated below is recommended.
Suggested sample locations include three samples in the low pool portion of CGR and three samples in
the portion of the lake that is inundated under high pool conditions. The geospatial coordinates for each
sampling point should be identified before the first round of sampling. Samples for all four quarterly
sampling events should be collected from a consistent set of locations. During low pool, the three samples
reserved for the high pool portion of the CGR should be collected from the CFW River channel flowing
through or incised in the lake bottom sediments exposed during low pool. The samples should be spaced
such that one is located immediately upstream from the entry point of the river into the exposed lake
32
-------
bottom sediment area, one from mid-way between the first location and the entry point of the river into
the low pool, and the third from immediately upstream of the river's entry point into the low pool. The
CFW River lake-bottom channel samples will provide an indication of surface water total mercury
loading resulting from erosion of the legacy lake bottom sediments.
All samples are recommended for anlaysis of total mercury, dissolved mercury, methylmercury (total),
reactive mercury (total), DOC, major ions (including sulfate) and TSS, as well as the above noted
standard field parameters.
4.4.2.1.2 SEDIMENT SAMPLING
A two-phased sediment sampling approach consisting of a high pool and a low pool sampling event is
proposed. High pool sediments are sediments exposed to the atmosphere during low pool conditions,
while low pool sediments are those from the portion of the CGR that is perpetually inundated.
High Pool Sediment Sampling. High pool sampling is recommended to consist of four sampling events
over a period of 1 year. The timing of each event should correspond to the shift from high to low pool and
low back to high pool. The objectives of the sampling are to assess sulfide and sulfate cycling as a
function of pool level and to obtain data regarding the timing and significance of methylation processes in
the high pool sediments. The first sampling event should be performed within 1 week after low pool
conditions have been established. The second event is recommended to be performed approximately 1
month after the first event. Likewise, the third event would be performed within 1 week after high pool is
established, and the fourth performed 1 month after the third event.
Sampling is recommended at eight locations, evenly distributed across the high pool sediment area. The
geospatial coordinates for each sampling point should be identified before the first sampling event.
Samples for all four events should be collected from a consistent set of locations Samples can be
collected from the surface to a depth of 2 to 4 inches using a stainless steel spoon (low pool time) or a
petite Ponar dredge sampler (high pool time). Figure 12b provides a description of the petite Ponar
dredge sampler.
The samples recommended for analysis include total mercury, methylmercury, reactive mercury, total
organic carbon (TOC), sulfate, and sulfide.
Low Pool Sediment Sampling. Low pool sampling is recommended to consist of two events performed
over a period of 1 year. One sampling event should be performed 1 month after low pool is established
and the other performed 1 month after water levels are reset at high pool. An objective of the sampling is
to obtain preliminary data describing the timing and significance of methylation in the low pool
sediments.
Sampling is recommended at eight locations, evenly distributed across the low pool area. The geospatial
coordinates for each sampling point should be determined before the first sampling event. Samples for
both events should be collected from a consistent set of locations. Samples can be collected from the
surface to a depth of 2 to 4 inches using a petite Ponar dredge sampler (high pool time).
The samples should be analyzed for the same parameters specified above for the high pool sediment
samples.
4.4.2.1.3 SEDIMENT PORE WATER SAMPLING
33
-------
Pore water samples collected from the top few inches of sediment will provide constituent concentration
data from the shallow sediment zone, which is prime habitat for methylating bacteria. Similar to the
collection of sediment samples, a two-phased sediment pore water sampling approach consisting of a high
pool and a low pool sampling event is proposed. In situ pore water samples may be collected by pushing a
slotted stainless-steel drive point into the sediment to a depth of approximately 2 inches below the lake
bottom. A circular, stainless steel flange welded to the drive point can be used to control depth of
penetration and to restrict the entry of surface water. Figure 12c shows an example sediment pore water
sampling tool. Other methods may also be identified as appropriate for pore water collection.
High Pool Sediment Pore Water Sampling. The recommended high pool sampling approach consists of
two events timed to coincide with the high pool conditions. The objectives of the sampling are to assess
sulfide and sulfate cycling as a function of pool level and to obtain data regarding the timing and
significance of methylation processes in the high pool sediments. The first sampling event is
recommended within 1 week after high pool conditions are established. The second event is
recommended approximately 1 month after the first event.
Sampling is recommended for the eight locations used to collect the high pool sediment samples. The
recommended analyte list includes total mercury, dissolved mercury, methylmercury, reactive mercury,
DOC, sulfate, and sulfide as well as the above noted standard field parameters.
Low Pool Sediment Pore Water Sampling. Low pool sampling is recommended to consist of two
events performed over a period of 1 year and during the same event as the low pool sediment sampling
task discussed above. As such, one event will be performed 1 month after low and high pool conditions
are established. An objective of the sampling is to obtain preliminary data describing the timing and
significance of methylation in the low pool sediments.
Sampling is recommended for the eight locations used to collect the high pool sediment samples. The
recommended analyte list includes dissolved mercury, methylmercury (dissolved), reactive mercury
(dissolved), and DOC.
4.4.2.1.4 ENVIRONMENTAL MEDIA DATA EVALUATION
The proposed sampling approach is intended to provide the basis for evaluating the sources of dissolved
mercury and methylmercury to CGR and the locations where methylation processes are active in the
water body. The data generated across the three media included in the CGR characterization can be
evaluated using differing logic and objectives. The approaches described in this section include the
evaluation of methylation in the low pool sediments and water column and in the high pool sediments.
Low Pool Sediment Evaluation. Figure 13 shows the logic for evaluating methylation in the low pool
sediments and water column. With surface water, low pool sediment, and low pool sediment pore water
data as inputs, the diagram includes three decision points. The following discussion pertains to sediment
and sediment pore water samples collected from sampling locations defined at low pool. As such, note
that "methylmercury concentrations during high pool period" refers to sampling results obtained from the
low-pool-defined sampling points under high pool conditions.
• Are methylmercury concentrations elevated? The surface water, sediment, and sediment pore
water methylmercury concentrations are compared with an appropriate background level or
published standard (such as SQuiRT). If methylmercury concentrations are not elevated in these
media, the methylation in the low pool sediments is unsubstantiated, given the available data set.
If methylmercury concentrations are elevated, methylation is likely occurring in the low pool
sediments and water column.
34
-------
• Are methylmercury concentrations measured during the high pool period elevated relative
to methylmercury concentrations measured during the low pool period? Given that water
column stratification is likely more dominant under the high pool summer months relative to the
low pool winter months, anoxic conditions and, therefore, methylation processes are expected to
be most active under high pool conditions. Therefore, a "yes," at this decision point suggests that
methylation is most actively occurring during high pool.
• Is total mercury elevated in the downstream portion of the CFW River lake bottom channel
relative to the CFW River entry point to CGR? A "yes" suggests that the CFW lake bottom
channel is actively eroding and mobilizing elevated total mercury concentration sediments before
it discharges to low pool. A "no" indicates that the eroded sediment is not significantly increasing
the total mercury load in the channel.
High Pool Sediment Evaluation. Figure 14 shows the logic for evaluating methylation in the high pool
sediments. With surface water, high pool sediment, and high pool sediment pore water data as inputs, the
diagram includes three decision points.
• Are methylmercury concentrations in sediments and sediment pore water elevated at the
end time relative to the start time of high pool? A "yes" indicates that methylation processes
are active in the high pool. At the start of high pool, sulfate concentrations should approximate
concentrations in CFW River and methylmercury concentrations should be low. After an
extended period (1 month), anoxic conditions should exist in the sediments and methylmercury
concentrations will likely show an increase. A "no" indicates that active methylation processes in
the high pool sediments are unsubstantiated.
• Is methylmercury elevated in the downstream portion of the CFW River lake bottom
channel relative to the CFW River entry point to CGR? A "yes" indicates that some high
pool sediments exposed during low pool conditions are anoxic and contribute methylmercury to
the CFW River before it discharges to the low pool or that active methylation is occurring in the
channel itself.
• Do sulfate concentrations in high pool sediment/sediment pore water increase from the start
time to the end time of low pool? A "yes" indicates that after exposure of the high pool
sediments, sulfide is oxidized to sulfate thus generating a necessary compound for the occurrence
of mercury methylation and thus providing evidence for sulfate cycling in the high pool
sediments.
4.4.2.2 CGR BIOTA AND MERCURY INFLUX CHARACTERIZATION
In addition to environmental media characterization, it is also recommended that baseline total mercury
concentrations be characterized in indicator fish species and the baseline mercury influx (total, dissolved,
and methylated) (see Figure 15). Baselining these parameters will provide levels that can be used to
compare future analytical results to gauge the effectiveness of any source mitigation or reservoir
managment measures that have been implemented at the BBM and CGR. Once source mitigation
measures are implemented, harvesting, and analysis of an appropriate fish species and determination of
the mercury influx on an on-going basis is recommended to evaluate any potential reductions achieved.
Continuation of the influx measurements associated with the current USGS investigation may be an
appropriate approach for monitoring CGR mercury loading on an ongoing basis.
35
-------
4.4.2.2.1 DATA EVALUATION
Figure 15 shows the data evaluation logic for the recommended environmental sampling at CGR. An
explanation of the various decision points follows:
• Are fish tissue total mercury concentrations greater than established fish consumption
advisories for the area? This decision applies following the collection of fish tissue from a
range of trophic levels. A sufficient number of fish (as determined by human health and
ecological risk assessment needs) can be harvested over a range of size classes. From this data, a
regression can be developed of fish mercury concentration versus fish length. Numerous studies
have shown that fish mercury concentrations increase with fish length. Assuming that mercury
concentration is correlated with fish length, the mercury concentration of typical size classes of
fish that humans consume from the lake can be identified and evaluated against fish consumption
guidelines.
• Are aquatic tissue total mercury concentration levels related to trophic position? The
anticipated condition is that mercury concentrations in fish tissue increase with trophic level. If
this condition is verified, evidence exists that methylmercury is entering the reservoir food web
through the base level. To evaluate effectiveness of any remedial actions taken at BBM or CGR, a
baseline can be established for low trophic level species, which should respond earlier than higher
level species. However, in recognition of public and human health concerns, a mercury level
baseline may also be established in high trophic level sport fish. If mercury concentration levels
in fish tissue are unrelated to trophic level, this result suggests fish uptake mercury by an
undefined process or that trophic level sampling results are unrepresentative of the actual
conditions. Given this result, ongoing fish tissue monitoring should proceed using a species
selected based on professional judgment.
• Does the CGR mercury mass balance suggest a downward trend in resident total and
methylmercury mass in CGR? At this decision point, it is assumed that source mitigation
measures have been implemented (either at the site or in the CGR) and that the CGR mercury
influx monitoring is ongoing. If the mercury influx monitoring indicates that the mercury influx is
trending downward, the potential exists that mercury concentrations in fish tissue are also
trending downward and the potential for rescinding the consumption advisory can be considered.
If the methylmercury mass is stable or increasing, the development of a CGR characterization
plan should be considered, with the initial focus on evaluating methylation processes in the
sediments and the anoxic water column.
• Do the aquatic tissue concentration trends and the CGR mass balance results merit
consideration of the planning of a CGR environmental media investigation? Given that
sufficient aquatic tissue and mercury mass balance data have been collected for meaningful trend
analysis, this decision point seeks to determine whether to continue fish tissue and mercury influx
sampling or initiate planning for a more intensive reservoir characterization effort that would
include sediment and other media. If downward trends are observed in the mercury
concentrations in fish tissue and influx data, but additional sampling is needed to confirm these
trends, then sampling aquatic tissue and CGR mercury mass fluxes should continue. If trends are
stable or increasing, then, in addition to the continuation of the aquatic tissue and CGR mercury
mass balance sampling, additional reservoir characterization to evaluate the factors controlling
methylation can be considered.
36
-------
4.4.2.3 CGR SAMPLING SEQUENCING
CGR sampling is proposed to occur in two phases that can be conducted independent of the BBM Site
investigations discussed in Section 4.4.1. However, conducting the CGR investigations coincident with
the BBM Site investigations is recommended so that all data are contemporaneous, thus facilitating
potential co-analysis and preserving project resources. The first CGR sampling phase includes one-time
sampling associated with the investigation of environmental media to evaluate the factors controlling the
production of methylmercury (Section 4.4.2.1 sampling tasks). The second phase includes recurring
sampling associated with establishment of baseline mercury concentrations in biota and surface water
inflow to the CGR (Section 4.4.2.2 sampling tasks).
Section 4.4.2.1 One-Time Sampling Tasks. The environmental media sampling tasks discussed in
Section 4.4.2.1 are timed to coincide with the establishment of high and low pool levels. It is
recommended that all sampling tasks proposed in Section 4.4.2.1 be performed in the same calendar year.
The proposed surface water sampling should occur on a different schedule than sediment and sediment
pore water sample collection. Surface water sampling should occur in January and March for low pool
conditions and in July and September for high pool conditions. The proposed sediment and sediment pore
water sampling should occur within 1 week of a change in pool level (estimated as the end of October for
high pool and the end of March for low pool) and 1 month after the first event for the given change. To
facilitate data comparability, these two media should be sampled together in each sampling event, with
sediment pore water sampling first, followed by sediment sampling.
Section 4.4.2.2 Recurring Sampling Tasks. It is recommended that sampling to establish baseline
concentrations in biota and surface water inflow to the CGR proceed at the earliest opportunity in the RI.
These tasks are recurring, with no specific end time specified in this review. To facilitate comparability
with potential future CGR sampling events, it is further recommended that the Section 4.4.2.1
environmental media sampling tasks be timed to occur after the baseline sampling begins.
37
-------
5.0 FINDINGS
The findings in this section are the combined interpretations of the optimization review team based on
historical information and data review, a site visit conducted on January 10, 2012, and SPP efforts
conducted with team members January 9 and 11, 2012. These findings are not intended to imply a
deficiency in the any of the previous characterization work, or the RA performed, but are offered as
constructive, forward looking suggestions in the best interest of Region 10, the public, ODEQ, and other
stakeholders. These observations also have the unique benefit of being formulated based on the collection
of additional data after the RA.
The mercury contamination concerns associated with the site include human and ecologic exposure to
mercury in soil and tailings at the site and the off-site migration of mercury with potential to
bioaccumulate in the tissues of fish inhabiting the downstream surface water features, including CGR.
Findings viewed by the optimization review team as significant to defining the optimal approach for
conducting the RI are presented first for the BBM Site and vicinity, followed by the findings for CGR.
These findings are provided in addition to the data gaps identified in for the BBM Site and CGR in
Section 4.3.
Key findings related to the BBM Site and vicinity include:
• During the site visit, the optimization review team noted very steep terrain (see photograph log
prepared by Tetra Tech, 2012, Appendix A), evidence of flood and mechanical erosion events,
and the presence of significant tailings in Furnace Creek. Historical data (EPA 2008) also indicate
the presence of higher concentrations (EPA 2008) and more bioavailable forms (Ecology and
Environment 2006) of mercury occurring in this drainage.
• A post-RA surface water loading assessment (Thorns 2008) suggests that the transport of
suspended solids containing mercury appears to be the primary mode of mercury transport from
the site. Based on one sampling campaign during non-storm conditions, the assessment estimates
that Furnace Creek could contribute between 50 and 75 percent of the mercury load in the CFW
River. Re-calculation of this value by the optimization review team suggests that contribution
may be lower (26 to 59 percent); however, it still represents a potentially significant source.
• Although the available data indicate that the mercury present in site tailings generally occurs in
insoluble forms that are not readily leached and methylated, these conclusions are based on a
relatively small number of samples (six or fewer, depending on the analysis) with detection limits
that are several orders of magnitude above environmentally relevant concentrations. Since nearly
all of the tailings are underlain by bedrock, collection of a groundwater sample beneath the
tailings at the site may be problematic for achieving this objective because of the challenges
associated with drilling in bedrock and the uncertainties regarding groundwater flow patterns in
fractured bedrock.
• During the site visit, the caretaker of the site and a former BBM worker (Mr. Michael Pooler)
identified a portion of the Garoutte Creek floodplain where tailings were historically stockpiled.
Groundwater sampling beneath and adjacent to tailings and at locations where groundwater may
enter surface water features may provide justification for removal of the groundwater medium
from further consideration in the RI. If groundwater sampling indicates leaching is occurring at
38
-------
concentrations and fluxes of concern, consideration of additional groundwater characterization
options may be warranted.
• SSE analysis of the soil samples collected from the ridge tops and hillsides in the vicinity of the
site indicated that less than 20 percent of the mercury contained in these samples was present in
relatively insoluble mercuric sulfide forms, and 44 to 87 percent of mercury was complexed with
organic matter. The organic matter-complexed forms are more readily converted to
methylmercury. These sampling results suggest that, in addition to contributions from the site,
soil erosion and surface water transport and groundwater discharge from nearby hillsides
(potentially previously contaminated by site mining operations through the deposition of
elemental mercury) may also be a source of mercury to surface water.
• Historical data indicate the presence of potential mercury impacts in surface water sediments
from the site downstream to CGR. The mercury contribution of this material present in surface
water body sediments versus the flux of new fined grained material with elevated mercury from
BBM is not well understood.
• The pH of the groundwater discharging to two of the mine adits visited during the site visit was in
the neutral range, suggesting the general absence of acid mine drainage impacts at the site.
Key findings that relate to CGR include:
• Surface water reservoirs in areas without mercury mining are known to contain fish with elevated
mercury levels in their tissue. Atmospheric deposition from the global mercury pool is believed to
be the source of this mercury. This source is likely responsible for some of the CGR mercury
burden. Neighboring Dorena Reservoir, with no known mercury mines in its watershed, contains
fish with elevated mercury in their tissue, although at lower concentration levels compared with
CGR. Given that one of the sources of mercury to CGR is deposition from the global mercury
pool, reductions in mercury concentrations in fish tissue may be limited to some baseline level
that reflects this ongoing source. Whereas controlling atmospheric sources is well beyond the
scope of this project, management actions occurring locally within the CGR watershed (for
example, forestry operations) and within CGR itself (such as changes in water level) can be
important in affecting the amount of atmospheric mercury that accumulates in fish tissue.
• Analysis of mercury transformation processes in CGR requires a detailed evaluation of all
mercury complexes and rate limiting constituents (organic carbon and sulfate). Conclusions from
such an undertaking would require significant extrapolation and inferences from a limited spatial
and temporal data set. In addition, any mass balance determination will be subject to uncertainties
regarding atmospheric deposition, watershed contributions, and internal methylation and
demethylation processes operating within the CGR.
• The mercury profiles in the available sediment cores from CGR indicate that significantly
elevated mercury concentrations are present in the sediments deposited up to 40 years ago. As
observed during the site visit, sediment exposed in the shallow portions of the reservoir during
low pool periods is actively being eroded by the CFW River and deposited in the low pool. The
eroded sediment includes the sediment with elevated mercury concentrations deposited decades
ago. The remobilization of mercury by CFW River erosion of older, legacy sediments exposed
during low pool may be an important ongoing source of mercury to the reservoir.
39
-------
• Direct determination of the contribution of mercury from the site to CGR would require the
quantification of mercury fluxes to CGR and the collection of mercury speciation data to define
the key mercury methylation processes that occur in the reservoir. In addition, any mass balance
determination will be subject to uncertainties regarding the significance of atmospheric
deposition or other watershed contributions. Development of a detailed mercury mass balance for
the reservoir and definition of the important methylation processes may require time and
resources beyond the scope of the current RI.
• Although uncertainties exist regarding the factors controlling the net mercury methylation rate in
the CGR, methylmercury generation generally requires the presence of three constituents:
mercury in a bioavailable form, microbial labile organic carbon, and sulfate. Methylating
bacterial processes typically involve the reduction of sulfate to sulfide. Once all sulfate has been
converted to sulfide, or the supply of mercury in a bioavailable form or microbial labile organic
carbon is exhausted, the bacteria become dormant and methylation ceases. Assuming relatively
abundant organic carbon and mercury, sulfate availability may be the likely rate limiting
constituent for the methylation process. Organic carbon and bioavailable mercury may also play a
role in limiting methylation. If data collection indicates sulfate is the rate limiting factor, a
potential approach for limiting mercury methylation processes in CGR is to permanently increase
the reservoir's operating level. It is recognized that USAGE would allow this action only if a
proper balance of other management priorities for the CGR can be achieved. By increasing the
reservoir's operating level, sulfate concentrations (and as a result methylation rates) may be
reduced because sulfide would not be recycled and fresh sulfate inputs would be limited only to
those from atmospheric and watershed inputs. Perhaps more importantly, permanently raising the
reservoir level would essentially eliminate the erosion and remobilization of historical sediments
with elevated mercury concentrations that has been ongoing over the years during low pool
conditions.
• Review of the available data for mercury concentration in fish tissue for CGR suggests that even
if only a small fraction of the total mercury is present in dissolved phase, sufficient mercury
methylation will occur to result in elevated mercury in fish tissue. Based on existing data,
calculations by the project team indicate that the percentage of total mercury that is methylated in
CGR water is only 6 percent. In sediments, the percentage is only 0.1 percent. These low levels
are apparently sufficient to support methylation.
40
-------
6.0 RECOMMENDATIONS
The purpose of this optimization review was to evaluate site conditions and identify optimal approaches
for conducting an RI of the site. The recommended sampling approach and data evaluation objectives
were presented in Section 4.3. This section summarizes the key recommendations reflected in the
proposed media characterization approaches, first for the BBM Site followed by the CGR. Note that while
the recommendations provide some details to consider during RI work plan preparation, they are not
intended to replace the RI work plan or other more comprehensive planning documents.
Recommendations for the BBM portion of the RI include:
• A major objective to consider for the BBM RI is an improvement of the understanding of the
mercury flux (total, dissolved, methylated) from BBM Site environmental media to Furnace,
Dennis, and Garoutte Creeks and to evaluate the mercury flux from Garoutte Creek to
downstream surface water features including CFW River and CGR. Consistent with this
objective, quarterly analysis for mercury and metals during storm and non-storm events with
coincident measurement of storm and non-storm stream flow discharge is recommended. This
data will provide the foundation for determining the important pathways for the release of
mercury from the site and quantify the site contributions to the downstream mercury load for each
of the three site creeks. Installation of weirs or use of direct measurement techniques for gauging
flow in Furnace, Dennis, and Garoutte Creeks should be considered.
• Limited, existing data suggest that site groundwater concentrations are not altered by mercury and
other metals leaching at elevated concentrations from site tailings. To understand whether
leaching is occurring at lower (but still environmentally relevant) levels, groundwater samples
should be collected from saturated native alluvial sediments underlying site tailings. Since nearly
all of the tailings are underlain by bedrock and the water table occurs within the bedrock,
collection of a groundwater sample beneath the tailings piles at the site is complicated by the
practical challenges that exist in accurately sampling fractured bedrock groundwater. During the
site visit, a portion of the Garoutte Creek floodplain adjacent to BBM was identified as a potential
location for historical tailings storage/disposal. Assuming that the floodplain is underlain by
unconsolidated materials, this potential tailings area provides an opportunity for assessing
possible impacts to groundwater from tailings leachate. Based on the ground elevation relative to
Garoutte Creek and the relatively broad floodplain in the vicinity, the water table likely occurs in
unconsolidated material and should be easily accessible using a drive point sampling approach.
The presence or absence of tailings in the area could not be confirmed during the site visit
because of the thick vegetation.
• If BBM environmental media, and Furnace Creek tailings in particular, are not found to provide
major contributions to the introduction of new mercury and trace metal contamination in Garoutte
Creek, the project team may consider increased sediment sampling in Garoutte Creek and
sediment sampling in CFW River to further assess the contribution of historical sediments to
methylmercury in surface water and CGR fish tissue. If appropriate, the additional sediment
sampling and analysis may be combined with human health or ecological risk exposure
assessments.
41
-------
• A DMA analysis is recommended for XRF and Lumex field-based metals analysis. Results of this
analysis can be used to assess confidence in RA characterization results and the utility of field-
based methods for metals analysis during the RJ. Similarly, the results can be used to establish
correlations between methods necessary to provide appropriate confidence in field screening tools
and develop field based action levels for these tools. The resulting action levels will provide
high confidence in clean/dirty decisions or can indicate where the collection of collaborative
laboratory data would be most beneficial.
Recommendations for the CGR RI:
• Development of the data necessary to understand the source of methylmercury in CGR fish tissue
requires investigation of the major sources of mercury mass influx to the reservoir (in addition to
the current contribution from BBM) and of the factors controlling the availability of the rate-
limiting constituents (dissolved mercury, organic carbon, and sulfate). The annual cycling of the
CGR water level between low and high pool and the potential release of mercury through CFW
River erosion of legacy sediments with elevated mercury concentration during low pool will
complicate the investigation effort. In light of the technical, administrative, funding, and schedule
challenges, it may prove beneficial for Region 10 to consider conducting activities at BBM and
CGR as separate OUs.
• A major objective to consider for the CGR RI is the establishment of baseline mercury
concentration levels in fish tissues and of the influx of mercury (total, dissolved, and methylated)
to the reservoir. It is recommended that mercury in fish tissue be monitored on an annual basis
and that both game species and species at the base of the food web be included. The collection of
fish tissue and mercury influx data will provide the basis for assessing the effects of any
mitigation efforts at the BBM Site or in CGR itself.
• Consideration should be given to the generation of analytical data from the various CGR
environmental media to enable a preliminary assessment of the factors controlling methylmercury
generation. These efforts may include the collection and analysis of quarterly or semiannual
surface water, sediment, and sediment pore water samples. Specific objectives of this sampling
would include acquiring evidence to confirm the existence of sulfate cycling in the high pool
sediments and assessing potential temporal variation in the methylation process.
6.1 COMPARISON OF RECOMMENDATIONS TO TRADITIONAL OPTIMIZATION
Focus AREAS
As discussed in Section 1.0, optimization review recommendations have traditionally been provided to
maximize protectiveness, cost-effectiveness, technical merit, and closure efficiency while minimizing the
environmental footprint of sites with planned or operating remedies. For sites that are in the RI phase
(such as BBM), potential or likely remedy options are presently not well understood. The goal for
optimizing sites in this phase is to provide a framework for planning an optimal RI focusing on CSM
refinement, sequencing of activities to identify contaminants and pathways of greatest concern, and
collection of data for risk assessment.
To the extent practical, this section compares the recommendations with each of the traditional
optimization focus areas.
• Protectiveness. While not specific to remedy protectiveness, the recommendations provided in
this document are based on refinement of the CSM to provide a basis for designing an effective
42
-------
RI. RI goals are to determine site risks, and as applicable, support the evaluation and selection of
an appropriately protective remedy. Recommended sampling and sequencing are provided to
identify dominant controls on the release and transport of mercury and metals from the site to
nearby surface water bodies, including CGR. Recommendations for sequencing and applying an
effective characterization for surface water, groundwater, sediment, tailings, and soil are provided
as a means to offer an accurate identification of fate and transport issues necessary for the
selection and design of appropriately protective remedies. The data collection framework and
accompanying decision logic enable the collection of important human health and ecological risk
data. The logic seeks to ensure that all potentially specific site pathways are considered.
• Cost-effectiveness. The recommended framework maximizes the use and value of data and other
results from previous site investigations and removal actions to form a CSM for both the site and
the CGR. The recommended sampling approach uses prioritized sampling results to address
critical data gaps and provides the ability, as necessary, to react dynamically to site conditions
identified during initial surface water, groundwater, and sediment sampling. The scale of hillside
soil sampling, site soil, and tailings sampling can be optimized based on estimated contributions
of these media to contaminant flux in surface water features. Optimization supports improved
cost effectiveness of sampling. The recommended sampling approach also seeks to establish
baseline conditions in the CGR, while defining the requisite conditions for when a more intensive
investigation of CGR may be appropriate.
• Technical merit. The recommendations establish an adaptive framework for the investigation.
As a result, the potential for expenditure of time and resources on non-critical portions of the site
or specific constituents should be minimized. In addition, in accordance with investigation BMPs,
sampling logistics, schedule and locations can be optimized to maximize resources and limit site
mobilizations. For example, groundwater seasonal grab sampling can coincide with planned
storm and non-storm seasonal surface water and sediment sampling. Similarly, soil sampling
locations can be assessed and refined in the field based on real-time field analysis, such as XRF
measurements. Use of real-time measurement technologies such as XRF and Lumex can
beneficially increase data density while optimizing sampling for ecological and human health risk
assessments.
• Site closure. The recommendations define an RI framework for accurately identifying the key
factors controlling the release of site constituents and, thus, may lead to the effective design of
appropriate mitigation measures and efficient site closure. Similarly, timing, milestones, budget,
and logistics may make it administratively attractive to separate activities at the site and CGR into
multiple OUs.
• Environmental footprint reduction. Traditional footprint considerations for optimization
remedy reviews focus on energy use, water use, and other factors that may significantly influence
the project footprint. For investigation stage optimization reviews, footprint reduction should
focus on use of energy efficient and low emission equipment, minimizing investigation-derived
waste, and use of field and mobile laboratory services. Recommendations for the site and CGR
are focused on closing data gaps in the understanding of the release and transport of site
constituents and in the needs for assessing site risks. A fact sheet describing best practices for
consideration of green remediation principles for investigation activities can be found at
http://www.clu-in.org/greenremediation/docs/GR Fact Sheet SI 12-31-2009.pdf
43
-------
7.0 REFERENCES
Anderson, Keith, April 1, 1996, Oregon Dept. of Environmental Quality, Preliminary Assessment, Black
Butte Mine (Reference 17).
Blakey. N., 2008. Standard Operating Procedure for Obtaining Freshwater Sediment Samples,
Washington State Department of Ecology, Environmental Assessment Program, V. 1.0, March
2008.
Gary, R.H., Dowd, J.F., and Peters N.E., 2011. Determining Watershed Flow Pathways Using
Geochemistry and Timing. Proceedings of the 2011 Georgia Water Resources Conference, April
11 - 13, 2001, University of Georgia.
Curtis, L.R., 2003. Final Report Sources and Chronology of Mercury Contamination in Cottage Grove
Reservoir. Prepared for the U.S. Army Corps of Engineers, May 20, 2003 (Reference 19).
Curtis, L.R, 2004. Final Report Reconnaissance Soil Sampling at the Black Butte Mine, Department of
Environmental and Molecular Toxicology, Oregon State University, prepared for the Oregon
Department of Environmental Quality, August 9, 2004 (Reference 24).
Derkey, R.E., 1973. Geology of Black Butte Mine, Lane County, Oregon. Master's thesis, University of
Montana (Reference 9).
Ecology and Environment, Inc., 1998. Black Butte Mine Site Inspection Report, TDD: 98-04-0004,
prepared for the U.S. Environmental Protection Agency Region 10 (Reference 4).
Ecology and Environment, Inc., 2006. Black Butte Mine Removal Assessment Report, Lane County,
Oregon, TDD: 06-01-0005, prepared for the U.S. Environmental Protection Agency, Region 10
(Reference 13).
Ecology and Environment, Inc. 2009. Hazard Ranking System Document, Black Butte Mine. Prepared for
U.S. Environmental Protection Agency Region 10, September 2009.
Hope, B., 2003. Willamette River Basin Total Maximum Daily Load Project, Estimates of Mercury Mass
Loads and Sources in the Willamette River Basin, Draft Final, Oregon Department of
Environmental Quality Land Quality Division, August 6, 2003.
Hope, B.K. and Rubin, J.R., 2005. Mercury Levels and Relationships in Water, Sediment, and Fish Tissue
in the Willamette Basin, Oregon. Environmental Contaminant Toxicology, April 2005, 48(3):
367 - 80.
Curtis, L.R. and Allen-Gil, S., 1994. Mercury Dynamics and Methylmercury Accumulation by Fish in
Three Oregon Reservoirs. Prepared for the Oregon Department of Environmental Quality.
Prepared by Department of Fisheries and Wildlife, Oregon State University, March, 1994.
Oregon Department of Environmental Quality, 2006. Willamette Basin TMDL Report, September 29,
2006.
44
-------
Park J.G. and Curtis L.R., 1997. Mercury Distribution in Sediments and Bioaccumulation by Fish in Two
Oregon Reservoirs: Point-Source and Nonpoint-Source Impacted Streams, Department of
Environmental Health and Toxicology Program, Oregon State University, respectively, July,
1997 (Reference 23).
Tetra Tech, 2012. PowerPoint file "Black_Butte_Mine_Photo_Log_Shupe.pptx" available on the U.S.
EPA Environmental Science Connector.
Thorns, Bryn, R.G., August 21, 2008, WR Cleanup Program, State of Oregon, Department of
Environmental Quality, memorandum to Max Rosenberg, R.G., WR Cleanup Manager regarding
Black Butte Mine Mercury Loading Assessment Results (Reference 20).
United Nations Environmental Programme Chemicals Branch, 2008. The Global Atmospheric Mercury
Assessment: Sources, Emissions, and Transport
(http://www.unep.org/hazardoussubstances/Mercury/tabid/434/language/en-US/Default.aspx),
December 2008.
U.S. Environmental Protection Agency, 1990. Quality Assurance/Quality Control Guidance for Removal
Activities, Sampling QA/QC Plan and Data Validation Procedures, Interim Final, EPA/540/G-
90/004, OSWER Directive 9360.4-01.
U.S. Environmental Protection Agency, 2004. USEPA Contract Laboratory Program National Functional
Guidelines for Inorganic Data Review, OSWER 9240.1-45, EPA 540-R-04-004.
U.S. Environmental Protection Agency Region 10, 2008. Final Removal Action Report for Black Butte
Mine, Cottage Grove, Oregon (Reference 5).
U.S. Environmental Protection Agency, 2008a. Demonstrations of Method Applicability under a Triad
Approach for Site Assessment and Cleanup - Technology Bulletin, August, 2008, EPA 542-F-06-
005.
U.S. Environmental Protection Agency, 2009. Green Remediation Best Management Practices: Site
Investigation, Office of Solid Waste and Emergency Response, December 2009, EPA 542-F-09-
004.
U.S. Environmental Protection Agency, 2011. Environmental Cleanup Best Management Practices:
Effective Use of the Project Life Cycle Conceptual Site Model, Office of Solid Waste and
Emergency Response, July 2011, EPA 542-F-l 1-011.
U.S. Environmental Protection Agency, 2012, Methodology for Understanding and Reducing a Project's
Environmental Footprint, Office of Solid Waste and Emergency Response, February 2012, EPA
542-R-12-002.
U.S. Geological Survey, 2011. Data Program for Collecting Water-Quality and Hydrologic Data for
Estimating a Mercury Budget for Cottage Grove Reservoir, Statement of Work submitted to
USACE Portland District, September 1, 2011.
45
-------
TABLES
-------
Table 1 - Screening Levels Referenced in Previous Investigations
PART 1 - GENERIC MEDIA-SPECIFIC1
Waste Rock/Tailings/Soil
Environmental Protection Agency Region 9 PRGs
Environmental Protection Agency Region 9 PRGs
Oregon DEQ Maximum Allowable Soil Concentrations
Oregon DEQ Maximum Allowable Soil Concentrations
Residential Soil
Industrial Soil
Residential Soil
Industrial Soil
mg/kg
mg/kg
mg/kg
mg/kg
Sediment
National Oceanic and Atmospheric Administration SQuiRT - TEL
National Oceanic and Atmospheric Administration SQuiRT - PEL
Oregon DEQ Level II Screening Level Values - Plants
Oregon DEQ Level II Screening Level Values - Invertebrates
Oregon DEQ Level II Screening Level Values - Birds
Oregon DEQ Level II Screening Level Values - Mammals
Sediment
Sediment
Soil
Soil
Soil
Soil
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Surface Water
EPA National Recommended Water Quality Criteria (Freshwater CMC)
EPA National Recommended Water Quality Criteria (Freshwater CCC)
Oregon DEQ Level II Screening Level Values - Aquatic
Oregon DEQ Level II Screening Level Values - Birds
Oregon DEQ Level II Screening Level Values - Mammals
Water
Water
Water
Water
Water
Mg/L
Mg/L
Mg/L
Mg/L
Mg/L
Leachate
Oregon DEQ Leachate Reference Concentration
Leachate
mg/L
Waste Rock/Tailings/Soil
Environmental Protection Agency Region 9 PRGs
Environmental Protection Agency Region 9 PRGs
Oregon DEQ Level II Screening Level Values - Plants
Oregon DEQ Level II Screening Level Values - Invertebrates
Oregon DEQ Level II Screening Level Values - Birds
Oregon DEQ Level II Screening Level Values - Mammals
Residential Soil
Industrial Soil
Soil
Soil
Soil
Soil
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Total Total
Mercury Arsenic
23 0.39
310 1.6
80 0.4
600 3
Total Mercury
0.174
0.486
0.3
0.1
1.5
73
Total Mercury
1.4
0.77
0.77
3,300
10,000
Mercury
0.2
Methylmercury
6.1
6.2
0.0002
—
0.025
4
PART 2 - SITE-SPECIFIC ACTION LEVELS USED DURING THE REMEDIAL ACTION
BLACK BUTTE MINE REMOVAL ASSESSMENT2
Waste Rock/Tailings/Soil
Environmental Protection Agency Region 9 PRGs
Oregon DEQ Maximum Allowable Soil Concentrations
Oregon DEQ Maximum Allowable Soil Concentrations
Area
Old Ore Furnace
Area
New Furnace
Area/Main
Tailings Pile
Dennis Creek
Unit Total Mercury
mg/kg 23
mg/kg 115
mg/kg 1 0
Key:
PRGs = Preliminary Remediation Goals
DEQ = Department of Environmental Quality
mg/kg = milligrams per kilogram
SQuiRT = Screening Quick Reference Tables
PEL = probable effects level
TEL = threshold effects level
CMC = Criteria Maximum Concentration
CCC = Criterion Continuous Concentration
ug/L = micrograms per liter
Ecology and Environment, Inc., 2006
2 Region 10, 2008
-------
Table 2 - Summary of Proposed Sampling
CSM
Element
BBM
BBM
BBM
BBM
BBM
BBM
Media
Surface Water,
vadose zone
ground-water, and
precipitation
Dennis, Furnace,
Garoutte Creek
Sediment
Groundwater (and
tailings during
piezometer
installation)
Vadose Zone
(Unconsolidated
Material)
Groundwater
Saturated Zone
(Unconsolidated
Material)
Groundwater
Saturated Zone
(Bedrock)
Tailings
Report
Section
4.4.1.2
4.4.1.3
4.4.1.4
4.4.1.4
Not
Evaluated1
4.4.1.5
Proposed Sampling Approach Summary
(Fixed-Base Laboratory)
Phase 1
(Storm event hydrograph Sampling)
Estimate 30 total samples, 3 locations; 2 storm
events; grab; plus precipitation
Phase 1
Vadose zone groundwater sampling associated with
storm event hydrograph sampling
24 total samples, 8 locations
Phase 2
64 total samples: 8 locations; quarterly - storm/non-
storm; grab
64 total samples: 8 locations; quarterly - coinciding
with surface water sampling events; combination:
incremental composite
Phase 1
Tailings
Estimate 8 samples, drive point grab
Phase 2
Groundwater
72 total samples: 9 locations; quarterly - coincide
with surface water sampling events; grab (vadose
zone piezometers)
Phase 1
Groundwater
16 total samples: 16 drive point locations; grab
Phase 1
Soil/Tailings
Estimate 16 samples, drive point grab
Phase 2
Groundwater
36 total samples: 9 locations; quarterly - coincide
with surface water sampling events; grab (temporary
monitoring wells)
Potential RFI task: resample the 1 1 bedrock
monitoring wells that were sampled for TAL metals
by Ecology and Environment (1998)
See Phase 1 vadose zone (tailings sampling) and
Phase I saturated zone (soil/tailings sampling),
Section 4. 4. 1.4
ICS sample count dependent on number of DUs and
other ICS parameters that require stakeholder input
Proposed Analytes
HgT, HgD, MeHg (total), MeHg (dissolved), HgR
(total), Tal metals (total), DOC, TSS, pH, common
ions
HgD, MeHg (dissolved), HgR (dissolved), TAL
metals (dissolved), pH, DOC, common ions
HgT, HgD, MeHg (total), MeHg (dissolved), HgR
(total), HgR (dissolved), TSS, pH, DOC and
common ions
HgT, MeHg, TAL metals, TOC, grain size
HgT and TAL metals
HgD, MeHg (dissolved), HgR (dissolved), TAL
metals (dissolved), DOC, pH, and common ions
HgD, MeHg (dissolved), HgR (dissolved), DOC,
pH, and common ions
HgT and TAL metals
HgD, MeHg (dissolved), HgR (dissolved), DOC,
pH, and common ions
HgD, MeHg (dissolved), HgR (dissolved), DOC,
pH, and common ions
HgT and TAL metals
HgT, MeHg TOC, TAL metals, grain size
-------
Table 2 - Summary of Proposed Sampling (Continued)
CSM
Element
BBM
BBM
CFW
River
CFW
River
CFW
River
CGR
CGR
CGR
CGR
CGR
Media
Hill Slope Soil
Garoutte Creek Fish
Surface Water
Sediment
Fish
Surface Water
Sediment
Sediment Pore
Water
Fish
Surface Water
Report
Section
4.4.1.6
Not
Evaluated
Not
Evaluated
Not
Evaluated
Not
Evaluated
4.4.2.1.1
4.4.2.1.2
4.4.2.1.3
4.4.2.2
4.4.2.2
Proposed Sampling Approach Summary
(Fixed-Base Laboratory)
ICS sample count dependent on number of DUs and
other ICS parameters that require stakeholder input
Approximately 10 fish per species representing a
range in size classes
Sufficient number of samples to quantify variability;
grab. Include background location from Big River.
Sufficient number of samples to quantify variability;
ICS. Include background location from Big River.
Sufficient number of samples to quantify variance
and seasonable variability. Include background
sampling from Big River.
48 total samples : 6 locations; 2 sample depths per
location (epilimnion and hypolimnion) quarterly -
low pool/high pool; grab
High Pool
32 total samples, 8 locations, 4 sampling events, grab
Low Pool
16 total samples, 8 locations, 2 sampling events, grab
High Pool
16 total samples, 8 locations, 2 sampling events, grab
Low Pool
16 total samples, 8 locations, 2 sampling events, grab
8 sport fish /year
8 lower trophic /year
Annual Inflow Monitoring
40 total samples, quarterly for estimated 10 years
Proposed Analytes
HgT, MeHg, TOC TAL metals
HgT, plus carbon and nitrogen stable isotopes
HgT, MeHg (total), HgR (total), , TSS, DOC, and
common ions
HgT, MeHg, TAL metals, grain size
HgT plus carbon and nitrogen stable isotopes
HgT, HgD, MeHg (total), HgR (total), DOC,
common ions (including sulfate), sulfide, pH and
TSS
HgT, MeHg, sulfate, sulfide, TOC
HgT, MeHg, sulfate, sulfide, TOC
HgD, MeHg (dissolved), HgR (dissolved), DOC,
sulfate, sulfide, pH,
HgD, MeHg (dissolved), HgR (dissolved), DOC,
sulfate, sulfide, pH,
HgT, plus carbon and nitrogen stable isotopes
HgT, MeHg (total), TSS, DOC
1. Sampling for this media was not evaluated in this review.
Key:
BBM = Black Butte Mine
CGR = Cottage Grove Reservoir
DOC = Dissolved organic content
HgD = Dissolved phase mercury
HgT = Total mercury
HgR = Reactive mercury
MeHg = Methyl mercury
TAL = Target analyte list
TOC = Total organic carbon
TSS = Total suspended solids
-------
FIGURES
-------
FIGURE 1: SURFACE WATER FEATURES SURROUNDING
BLACK BUTTE MINE SITE
-------
xPooler
Residence
Dennis Creek
Regraded Area
Main
Tailings
Approximate
L'6cation,\Old'*dre
Furnac^Area
Soil
Repository
New
Furnace
Area
Old Ore Furnace
Capped Area
New Furnace
Capped Area
Legend
^ Proposed Vadose Zone
Groundwater Monitoring Piezometer
^^™ Areas of Operation During the RA
Main Tailings Pile
Furnace Creek Tailings Area
Roads
Rail
Creeks
100-Year Floodplain (FEMA)
Structures
Note: Extent of Main
Tailings Pile and
Furnace Creek tailings
area based on
estimates provided by
ODEQ.
260
520
780
FIGURE 2: BLACK BUTTE MINE SITE AREA
-------
Figure 3: Preliminary Pathway Receptor Network Diagram
Primary Source
Primary Release
Mechanism
Secondary Source
(Affected Media)
Secondary
Release
Mechanism
Pathway
Exposure Route
Receptors
Human
reational
s
0£
idential
,
Ingestion
Inhalation
Dermal
•
®
®
•
®
®
•
®
®
•
®
®
Direct
Contact
Ingestion
Inhalation
Dermal
®
®
•
®
®
•
®
®
•
®
®
®
Infiltration,
percolation
Groundwater
w
Ingestion
Inhalation
Dermal
®
®
®
O
®
®
®
®
®
O
®
®
Mechanical
erosion via
wind
±
Airborne Dust
Ingestion
Inhalation
O
O
6?)
O
O
6?)
O
O
6?)
®
®
6?)
mercury
>
™
Aquatic
Life [1]
.
^
Ingestion
Inhalation
•
®
(x)
^^
•
®
6f>
^^
•
®
(X)
^^
•
®
6f>
^^
Plant uptake
Ingestion
Inhalation
Dermal
U
®
®
U
®
®
U
®
®
09
®
®
Legend
[1] Assumes fish for human consumption; others for general aqualtic life
• Completed pathway
O Possible complete pathway (data required)
® Incomplete pathway
-------
Furnace Creek
Tailings Area
Wetland Areas
^(Exposed\DurirTqiliow.flool
Black Butte Mine Site
• Storm-flow-induced, mechanical
erosion of tailing particles from
the Furnace CreekTailings Area
Cottage Grove Reservoir
• Low energy surface water in the reservoir
results in deposition of tailings particles
• Potential anoxic conditions result
in the formation of MeHg
Cottage Grove Reservoir Wetland
Exposed Low Pool
• Active erosion of previously deposited BBM
tailings with elevated Hg concentration
• Sulfide converted to sulfate during low pool
and available for generation of MeHg during
anoxic high pool conditions
CFW River and Garoutte Creek
• Relatively high energy surface water flow in
Garoutte Creek and CFW River keeps fine
mercury-bearing tailings particles in suspension.
Another portion of the load exists as dissolved
phase mercury.
Legend:
I ^. ''I Dominant source area for surface water total
mercury contamination
•<—O Transported, suspended sediment
MeHg Methylated Mercury
FIGURE 4
SCHEMATIC REPRESENTATION OF
CONCEPTUAL SITE MODEL
TETRA TECH EM INC.
-------
¥.-
^
Alton
• j •..
Stream Sampling Location.
"~ TV
FIGURE 5: SURFACE WATER SAMPLING LOCATIONS AT
BLACK BUTTE MINE SITE
-------
Surface water samples will be collected for one year from Garoutte
Creek, Furnace Creek, and Dennis Creek at a total of eight locations.
Sampling will occur seasonally at high and low flow conditions for a
total of 64 samples (eight at each location for the year). Analyses will
include HgT, HgD, MeHg (total), MeHg (dissolved), HgR (total), HgR
(dissolved), TAL metals (total), TSS, and common ions. This
combination of analytes will allow the estimation of the amount of
mercury in dissolved and suspended phases and will provide insight on
the speciation of the suspended phase. Stream discharge will also be
measured at each sampling station during each sampling event. Furnace
Creek and Dennis Creek discharge will be measured by installing a weir
structure and monitoring the water level using a transducer. Garoutte
Creek discharge will be measured by direct gauging or estimated using
available gauging data. Sediment samples will be collected at each
station during each surface water sampling event and evaluated using
separate logic.
Comparable
CSM Supported
\ Enter Tailings Process /
Yes"
How do Furnace
//Creek Hg speciation results^
to Garroute Creek/
\ speciation results?/
Yes
/Do suspended particle Hg/
concentrations in downstream
Garoutte Creek approximate
^sediment concentrations in Furnace/
Creek?
Yes
/Are suspended/
/load Hg concentrations
elevated relative to
dissolved load
/concentrations?/
No
Revise CSM.
• Bed load sediment
contribution (internal
loading)?
Measure total and
dissolved Hg
concentrations in
surface water
seasonally and in
storm/non-storm flow
conditions.
Yes
Is downstream
Garoutte Creek Hg flux >
\upstream Garoutte Creek/
Hg flux?
"No
/Is Furnace Creek/
Hg flux large compared />
to Garoutte Creek Hg ,
flux?
"No
Revise CSM.
Groundwater mass
loading important?
Other onsite process?
Not comparable
Revise CSM.
Offsite hillside
contribution
• Other onsite
process?
Revise CSM.
• Off-site, upstream
source?
• Obtain Region 10 input
regarding the path forward
for the investigation.
Revise CSM.
• Hillside
contribution?
Enter Sediment Logic \
' Diagram after completing
Surface Water Logic
Diagram
Review groundwater conclusions
Enter Soil Logic Diagram after
completing Surface Water Logic
Diagram
Figure 6. Data Evaluation Logic for Black Butte Mine Surface Water Sampling Task
-------
Sediment samples will be collected during each of the eight surface water
sampling events. Samples will be biased toward finer grain sizes that could
potentially be mobilized during storm flow conditions and will be analyzed for
total and monomethyl mercury. The sediment data will be used in concert
with the surface water data to evaluate the potential that surface water
suspended Hg load is the result of mobilized creek sediments and not from
erosion and mobilization of fine tailings particles from the Furnace Creek
Tailings Area. Sediment samples will be analyzed for HgT, MeHg, TOC,
TAL metals, and grain size.
Complete sediment and surface water
sampling tasks (measure total and dissolved
Hg concentrations in sediment and surface
water seasonally and in storm/non-storm flow
conditions).
CSM Supported
With the exception of Furnace Creek,
bedload sediments in Dennis Creek and
Garoutte Creek are not significant
contributors to the suspended Hg load.
Yes
\
/Are the suspended sediment\
mercury concentrations in down -
stream Garoutte Creek similar to
the sediment concentrations in
Furnace Creek?
No
Revise CSM
Bedload sediments in Garoutte Creek and
Denis Creek are significant contributors to
the suspended Hg loads in these creeks.
Obtain Region 1 0 input regarding the
appropriate path forward for the
investigation (e.g., development of a
sediments characterization plan).
Figure 7. Data Evaluation Logic for Black Butte Mine Sediment Sampling Task
-------
Proposed
groundwater
sampling
location
Possible
tailings
Fill area
Garoutte
Creek
floodplain
Location, Old Ore
Figure 8. Proposed Groundwater Sampling Locations
-------
At 9 locations (defined in text
at right), collect soil core
from ground surface to the
top of bedrock
Yes
Is evidence for perched groundwater
observed in the core?
Install a piezometer with the base
of the screen interval at the
bedrock surface.
Install a second piezometer
screened in the perched
groundwater zone
No
Install 1 piezometer with the base
of the screen interval at the
bedrock surface
Sample each peizometer for total
metals (total and methyl Hg),
dissolved (total and methyl Hg),
and common ions. Collect
samples to coincide with the
groundwater and surface water
grab sampling events.
Using direct push methods, nine vadose zone piezometers will be installed to collect
vadose groundwater samples from the hill slopes at the BBM Site, from the hillslope
opposite Garoutte Creek from the BBM Site, and a background location. The
piezometers will be installed at 2 locations in the Main Tailings Pile upslope from
Dennis Creek, 2 locations in the Furnace Creek Tailings Area upslope from Furnace
Creek, 2 locations along the approximate ridge crest that forms the drainage divide
between Dennis and Furnace Creeks, 2 locations on the hill slope on the opposite side of
Garoutte Creek from BBM, and at a background location, up-gradient and unimpacted by
BBM. Two piezometers will be installed at each location, the first installed with the base
of the screen interval coinciding with the bedrock surface and the second screened in a
shallower zone determined based on field conditions (e.g. evidence of perched
groundwater conditions). In the absence of any evidence of perched groundwater, only
one peizometer will be installed. Groundwater samples will be collected (if sufficient
sample volume can be obtained) from the piezometers during Phase I stormflow
hydrograph sampling and seasonally to coincide with the surface water grab sampling
events. The samples will be analyzed for HgD, MeHg (dissolved), HgR (dissolved), TAL
metals (dissolved), pH, DOC, and common ions. Low to non-detect mercury
concentrations in the vadose zone groundwater samples support the CSM. Conversely,
elevated mercury in hillside vadose zone groundwater suggests groundwater loading to
surface water may potentially be significant, a result counter to the CSM. Additional
vadose zone characterization sampling, designed in consultation with Region 10, will be
necessary to estimate mercury and other metals mass loading to Furnace, Garoutte, and
Dennis Creeks.
' Evidence of vadose zone ^
saturated flow & are
concentrations of Hg and other
metal elevated?
No
CSM is supported
I
Yes
Groundwater loading to
surface water may potentially
be significant, a condition
counter to the CSM and
requiring evaluation with
regard to the saturated zone
groundwater monitoring
results.
Figure 9. Data Evaluation Logic for Black Butte Mine Vadose Zone Groundwater Sampling Task
-------
The rotary sonic drilling method will be used to collect groundwater samples from
beneath the Garoutte Creek floodplain at the base of the Site. Prior to sampling,
tailings areas will be mapped by visual inspection. Soil borings will be installed
in both tailings and non-tailings areas. The soil borings will be lithologically
logged and sampled for mercury and other metals analyses via XRF and
laboratory analyses. A minimum of eight groundwater samples will be collected
from the tailings areas identified. In addition, a minimum of eight samples will
also be collected from non-tailings areas. Up to three groundwater samples will
be collected from the immediate vicinity of the confluence of Dennis Creek and
Garoutte Creek (and, if possible Furnace Creek and Garoutte Creek); the
remaining samples will be collected from the general floodplain area. Eight
temporary monitoring wells will be installed. Groundwater samples will be
collected from these wells seasonally to coincide with the seasonal surface water
grab sampling. The samples will be analyzed for HgD, MeHg (dissolved), HgR
(dissolved), TAL metals (dissolved), pH, DOC, and common ions. Low to non-
detect mercury concentrations in the groundwater samples from the Garoutte
Creek floodplain support the PCSM. However, if the mercury concentrations in
the samples are elevated, additional data collection (hydraulic conductivity testing
and gradient determination) will be conducted to determine the groundwater
mercury mass flux to Garoutte Creek. If the mass flux is elevated relative to the
Garoutte Creek mercury flux, the groundwater flux will be considered a
significant contributor the Garoutte Creek mercury flux. Given this result, Region
10 risk management assessment/decisions will be necessary to determine the
appropriate path forward.
-Yes-
Measure total metals (total and methyl Hg), dissolved
metals (total and methyl Hg), and common ion
concentrations in at least eight groundwater samples
collected from the portion of the Garoutte Creek floodplain
visually contaminated with tailings and collect at least eight
groundwater samples from non-tailings areas. At least
three samples should be collected near the confluence of
Garoutte and Dennis and (if possible) Furnace Creeks.
Install eight temporary monitoring wells for continued
seasonal monitoring.
Map surface soil material on the Garoutte
Creek floodplain. Visually differentiate
between areas potentially underlain by
tailings from areas underlain by native
soils.
From tailings and non-tailings areas,
install direct push soil borings, prepare
boring logs, and collect soil samples for
XRF & Lumex analysis and conventional
laboratory analyses of Hg and other site
metals of interest.
Presence of tailings
confirmed?
Measure total metals (total and methyl Hg), dissolved
metals (total and methyl Hg), and common ion
concentrations in at least three groundwater samples
collected from Garoutte Creek flloodplain near
confluence with Dennis and (if possible) Furnace
Creek; collect the remaining samples from the general
floodplain area below the Site. Install eight temporary
monitoring wells for continued seasonal groundwater
monitoring.
No
Collect additional data required to estimate
Hg mass flux from groundwater to surface
water (hydraulic conductivity, hydraulic
gradient).
«— Elevated-
-1/Vhat levels of Hg"
concentrations are
detected?
Is Hg flux in
groundwater elevated
.compared to Garoutte
\Creek Hg flux?,/'
I
Yes
Low to
below detection
Revise CSM to account for
groundwater contribution to surface
water Hg flux. Given this result,
obtain Region 10 input regarding the
appropriate path forward.
CSM Supported
Figure 10. Data Evaluation Logic for Black Butte Mine Groundwater Sampling Task
-------
If the results from the surface water sampling task support the CSM, a
Demonstration of Methods Applicability (DMA) will be conducted to establish
the relative strength of the correlation between XRF and Lumex field-based
metals analyses with laboratory analyses on a set of paired samples of
primarily tailings with a secondary focus on native soil. Characterization
sampling will then target the Furnace Creek Tailings Area. If the surface
water sampling task results are unsupportive of the CSM, a DMA will also be
conducted, but will focus on establishing the correlation of XRF and Lumex
field-based analyses with laboratory analyses of site native soils with a
secondary focus on tailings. Characterization sampling will then target native
soils at the Site and on the surrounding hillsides. Tailings samples will be
analyzed for HgT, MeHg, TAL metals, and grain size.
Complete surface water and sediment
sampling tasks (measure total and dissolved
Hg concentrations in surface water and
sediment seasonally and in storm/non-storm
flow conditions).
Yes
CSM
Confirmed?
No
CSM Confirmed.
Conduct a DMA to assess accuracy and
representativeness of XRF and Lumex
field-based metals technologies. Focus
the DMA on tailings analyses with
secondary consideration of native soil.
Conduct a DMA to assess accuracy and
representativeness of XRF and Lumex
field-based metals technologies. Focus the
DMA on site soils analyses with secondary
consideration of tailings.
Conduct tailings characterization to
determine the extent and metals
concentration levels for the tailings
disposed of in the Furnace Creek Tailings
Area.
Conduct soil/tailings characterization
focusing on hillside soils.
Figure 11. Data Evaluation Logic for Black Butte Mine Tailings Sampling Task
-------
:/t f ,
CottageGrove
Legend
• • • •
J B B BI Proposed High Density Sampling Area
Proposed Low Density Sampling Area
r^
-
Black
Butted
FIGURE 11A: PROPOSED AREA FOR
HILL SLOPE SOIL SAMPLING
-------
The objective and focus of the soil sampling task will be defined based on the results
of the surface water sampling task. Specifically, if surface water sampling results are
supportive of the CSM (e.g., Furnace Creek mercury flux is a significant greater than
the Garoutte Creek mercury flux, suspended load mercury concentrations are elevated
relative to dissolved load mercury, down-gradient speciation of suspended mercury
correlates with the down-gradient Furnace Creek suspended mercury species, and the
groundwater mercury flux is negligible), soil sampling for the RI will be conducted to
satisfy risk assessment/characterization objectives. If the surface water sampling
results are unsupportive of the CSM, evidence exists that the main source of mercury
loading to surface water is from non-tailings soil runoff or the groundwater mercury
flux to surface water. Given this situation, in addition to sampling to support risk
assessment, soil sampling for the RI will also focus on source-characterization of soils
underlying the site and the hillsides in the site vicinity. Soil samples will be analyzed
for HgT, MeHg, and TAL metals.
Complete surface water and
sediment sampling tasks (measure
total and dissolved Hg
concentrations in sediment and
surface water seasonally and in
storm/non-storm flow conditions)
/Furnace Creek Hg\
flux is greater than
Garoutte Creek Hg flux
and
\
Suspended load Hg concentrations in
down-gradient samples are elevated relative to
dissolved load Hg concentrations
and
Down-gradient Garoutte Creek speciation
results are correlated with Furnace
Creek speciation results
and
Groundwater mass
\ flux is negligible /
Yes
CSM Confirmed
Soil Characterization will consist of the
collection of a limited number of samples
(<20) located based on a random
sampling-within block approach. Samples
should be analyzed for Hg and other site
metals-of-interest via conventional
laboratory analyses.
No
Revise CSM
Evidence exists that main source of Hg
contamination is from non-tailings soil
runoff/groundwater contribution. As as
result, a soil sampling plan that focuses on
characterization of soil-bound Hg in
adjacent hillside soils should be developed.
Figure 12. Data Evaluation Logic for Black Butte Mine Soil Sampling
-------
CFW River
High pool
Low pool surface water
High pool surface water
High pool sediment and sediment pore water
Low pool sediment and sediment pore water
Low pool lake bed channel,
CFW River (estimated)
Low pool
2000ft
CFW River
Figure 12a Proposed Sampling Locations for Cottage Grove Reservoir
-------
Petite Ponar Grab Sampler Operation (Blakley, 2008). The petite Ponar grab sampler is equipped with a pair of weighted, tapered jaws
that are held open by a catch bar held in place by a spring-loaded pin. The sampler is triggered by impact with the bottom, which relieves
the weight on the catch bar, allowing the spring-loaded pin to eject. The upper side of the jaws is covered with a fine mesh screen that
allows water to flow through the jaws during descent. This reduces the bow wave created by the sampler and disturbance of the sediment
surface. After the sampler is retrieved, the mesh screen can be removed to gain access to the sediment sample.
Open position for sample collection
Closed position for sample retrieval
Figure 12b. Petite Ponar Dredge Grab Sampler
-------
TOP VIEW
DRIVE POINT
SAMPLER DETAIL
Figure 12c. Example tool for performing pore water sampling in soft sediments
-------
Sample surface water, low pool
sediments, and low pool
sediment pore water
Surface water samples, low pool sediment, and low pool sediment pore water samples will be
collected from CGR. Two sampling rounds for sediment and sediment pore water sampling
will be conducted (one for high and one for low pool). Surface water sampling will be
conducted quarterly from both the high pool and the low pool portions of the CGR. During low
pool, surface water samples will be collected from the CFW River channel incised into exposed
CGR sediments. Sediment will be analyzed for HgT, MeHg, HgR, TOC, sulfate, and sulfide.
Surface water will be analyzed for HgT, HgD, MeHg (total), HgR (total), DOC, common ions
(including sulfate), TSS, and standard field parameters (pH, temperature, ORP, DO, and
specific conductance). Sediment pore water will be analyzed for HgD, MeHg (dissolved), HgR
(dissolved), DOC, common ions (including sulfate), sulfide, and standard field parameters.
Are sediment and sediment
pore water methyl mercury
„ concentrations elevated? ,
Yes
Is methyl mercury during high^
pool elevated relative to low
pool period?
Yes
Evidence exists
that methyl
mercury
production rate
may be related to
season/pool level
Is total mercury elevated in the
downstream portion of the CFW River
lake bottom channel relative to the
CFW River entry point to CGR
Yes
No
No
CFW River
channel in lake
bottom sediments
is a source of total
mercury to the low
pool
Eroded lake bottom
sediments
unsubstantiated as a
significant source of
total mercury
Figure 13. Evaluation of CGR Internal Loading - Low Pool Sediments
-------
Sample surface water, high
pool Sediments, and high pool
sediment pore water
Surface water samples, high pool sediment, and high pool sediment pore water
samples will be collected from CGR. Four sampling rounds will be conducted.
Sampling of high pool sediments and sediment pore water and sediment pore water
will occur within one week of the establishment of high pool and after a period of
one month of high pool water levels. Similarly, sampling of low pool sediments and
sediment pore water will occur within one week of the establishment of low pool
and after a period of one month of low pool water levels. Sediment will be analyzed
for HgT, MeHg, HgR, TOC, sulfate, and sulfide. Surface water will be analyzed for
HgT, HgD, MeHg (total), HgR (total), DOC, common ions (including sulfate), TSS,
and standard field parameters (pH, temperature, ORP, DO, and specific
conductance). Sediment pore water will be analyzed for HgD, MeHg (dissolved),
HgR (dissolved), DOC, common ions (including sulfate), sulfide, and standard field
parameters.
/Are methyl mercury\
/concentrations in sediments\
and sediment pore water
elevated at the end time
\compared to the start time/
of high pool?
No
Is methyl mercury
elevated in the downstream portion
of the CFW River lake bottom
channel relative to the CFW River
entry point to CFR?
Yes
Do sulfate concentrations
increase from the start time
\to the end time of low pool?,.
Methylation in high
pool sediments is
unsubstantiated
No
Methylation in high
pool sediments is
unsubstantiated
No
Additional
sampling needed
to characterize
sulfate cycling
process
Figure 14. Evaluation of Potential of Internal Loading- High Pool Sediments
-------
Conduct sufficient
sampling of
species from a
range of tropic
levels and size
classes sufficient
to document with a
high degree of
confidence the
presence or
absence of
correlation
between trophic
level and Hg tissue
concentration
-\
/Are fish tissue HgT
concentrations above
/
established fish
consumption
advisories?
No
No
Are aquatic tissue
HgT concentrations
related to trophic
position?
Yes
Yes
Fish obtain most
Hg through
undefined process
or check for
characterization
error
Use best
professional
judgment to select
species to
evaluate THg
tissue levels
No
MeHg entering at
the base of the
food web
Use low trophic
position species to
evaluate THg fish
tissue levels
Yes
/"Do trends/concentration levels
indicate need for planning a
, sediments investigation? ,
Continue annual
tissue sampling
And
USGSCGR Hg
budget sampling to
assess potential net
Hg reduction
achieved through
BBM source
mitigation measures
No
Does CGR Hg mass
balance suggest
downward trend in
MeHg in CGR?
Continue aquatic
tissue sampling
and Hg mass
balance sampling
Develop CGR
characterization
sampling plan with
initial focus on
assessing
potential benefits
of permanently
increasing CGR
water level
Yes
/Have THg fish tissue^
concentrations
declined below
established
consumption
advisories?
\/
Yes
Fish consumption
advisory can be
rescinded
Figure 15. Data Evaluation Logic for Cottage Grove Reservoir Sampling
-------
ATTACHMENT A
-------
Attachment A - Site Visit Photo Log
Site Visit Occurred on January 10, 2012
Photo: 1
Description:
Bulletin Board at Cottage
Grove Reservoir.
HEALTH & SAFETY
ATTENTION
HIGH MERCURY I.K\ KI.S IN FISH
1 i'
• ' ' ••idrwi »nd
• -• nc mu*t M mk (ram \lic efftdcr. hallhy
beyond cfalldbenine Bge. Hnd iHher bcalihy *Juf
Ituo MwH-aunce mralprrmuulh An X-num.c meal »
The HUTCC of the mercury «ppc»n li> he fmm ihe natural geology o(
rhc rrtinir. uiti hutluOi: mining pnclicn within the ^aimiicj
Photo: 2
Description:
Cottage Grove Reservoir -
At main parking lot.
Looking down stream
toward dam. Note mud flat
and low water level.
A-l
-------
Photo: 3
Description:
Cottage Grove Reservoir -
View from boat ramp
adjacent to parking lot near
dam. Note mud flat and
low water level.
Photo: 4
Description:
Cottage Grove Reservoir -
View from boat ramp
adjacent to parking lot near
dam. Close-up of mud flat;
note fine grain size.
A-2
-------
Photo: 5
Description:
Cottage Grove Reservoir -
Looking upstream from
boat ramp adjacent to
parking lot near dam. Note
mud flat and low water
level.
Photo: 6
Description:
Cottage Grove Reservoir -
Looking toward dam from
boat ramp adjacent to main
parking lot. Note mud flat
and low water level.
A-3
-------
Photo: 7
Description:
Garoutte Creek - Looking
upstream from bridge for
entrance to the BBM site.
Note stream velocity and
distribution of sediment
grain sizes.
Photo: 8
Description:
Garroute Creek flooplain
& base of Black Butte
slope (in distance) as
viewed from the bridge to
Site caretaker's residence.
Note uneven/elevated
hummocky surface of
floodplain. Caretaker (a
former mine worker)
indicates that mine tailings
were stockpiled
historically in this area.
A-4
-------
Photo: 9
Description:
Garoutte Creek floodplain.
Note uneven/elevated
hummocky surface of
floodplain. Site caretaker
indicates that mine tailings
were stockpiled
historically in this area.
Photo: 10
Description:
View downslope toward
Dennis Creek from the
Main Tailings Pile. Note
excessive vegetation
(Scotch Broom) and steep
slope. Trees to the right are
growing in tailings.
A-5
-------
Photo: 11
Description:
View downslope toward
Dennis Creek from the
Main Tailings Pile. Note
excessive vegetation
(Scotch Broom) and steep
slope. Trees shown are
growing in tailings.
Photo: 12
Description:
Surface of Main Tailings
Pile. Note range in sizes
(gravel to fines).
A-6
-------
Photo: 13
Description:
Spring and associated
water storage tank located
below Main Tailings Pile.
Photo: 14
Description:
Looking east downstream
direction in Furnace Creek
Valley.
A-7
-------
Photo: 15
Description:
Furnace Creek bank. Note
pile up of sediment on
upstream side of this tree;
suggests active mass
wasting/mobilized
sediment during flood
conditions.
Photo: 16
Description:
On-site supply well
located upslope of the
Main Tailings Pile. Well
appeared to be actively
flowing at a low rate
(seeping).
A-8
-------
Photo: 17
Description:
On-site supply well
located upslope of the
Main Tailings Pile. Casing
cap has two pipes
protruding from it. Orange
discoloration marks
seeping water.
Photo: 18
Description:
Tipple-type superstructure
associated with the New
Furnace & Rotary Kiln.
A-9
-------
Photo: 19
Description:
Structures at the base of
the tipple-type
superstructure associated
with the New Furnace &
Rotary Kiln.
Photo: 20
Description:
Rotary kiln at the base of
the tipple-type
superstructure associated
with the New Furnace.
A-10
-------
Photo: 21
Description:
Downslope view toward
the exit point of the rotary
kiln associated with the
New Furnace.
Photo: 22
Description:
View of the rotary kiln and
mill. Note Hg vapor
capture device installed at
left side of building.
A-ll
-------
Photo: 23
Description:
First (lower) adit
encountered. Field pH
measured in the exterior
pool (result: approx. 8.0).
Photo: 24
Description:
First (lower) adit
encountered.
A-12
-------
Photo: 25
Description:
First (lower) adit
encountered.
Photo: 26
Description:
Second (upper) adit
encountered; possibly the
"404" adit.
A-13
-------
Photo: 27
Description:
Second (upper) adit
encountered; possibly the
"404" adit. View to the
right.
Photo: 28
Description:
Abandoned equipment
near Old Furnace.
A-14
-------
Photo: 29
Description:
Potential tailings on the
Garoutte Creek floodplain;
observed while searching
for down gradient
confluence of Graroutte
and Furnace creeks.
Photo: 30
Description:
At CG Reservoir, Wilson
Creek boat ramp. CFW
River flowing on bedrock,
incised in lake bottom
sediments. Estimated
sediment thickness is 3 - 5
ft.
A-15
-------
Photo: 31
Description:
At CG Reservoir, Wilson
Creek boat ramp. CFW
River flowing through lake
bottom sediments.
Photo: 32
Description:
At CG Reservoir, Wilson
Creek boat ramp. CFW
River flowing through
sediments deposited on
bottom.
A-16
-------
Photo: 33
Description:
At CG Reservoir, Wilson
Creek boat ramp. CFW
River flowing through
sediments deposited on
bottom.
Photo: 34
Description:
At CG Reservoir, Wilson
Creek boat ramp. CFW
River flowing through
sediments deposited on
bottom. Note sediment
thickness (estimate 3-5
ft).
A-17
-------
Photo: 35
Description:
At CG Reservoir, Wilson
Creek boat ramp. CFW
River flowing through
sediments deposited on
bottom. Note sediment
thickness (estimate 3-5
ft).
A-18
-------
ATTACHMENT B
-------
Incremental Composite Sampling Methodology:
Source: ITRC. 2012. Incremental Sampling Methodology. February.
The incremental composite sampling (ICS) methodology is a composite sampling approach that
statistically reduces data variability associated with discrete sampling and provides mean
concentrations of contaminants within a specified area or volume of soil referred to as a decision
unit (DU). The mean concentrations are used for comparison to regulatory threshold values and
action levels, or are used for risk assessment calculations. Conventionally, discrete samples have
been collected to estimate average contaminant concentrations, but use of ICS has been
increasing.
Using an ICS sampling approach, soil increments of equal mass are collected from multiple, un-
biased locations across a defined DU. The sampling locations within the DU must be evenly
distributed to ensure representativeness. The soil increments are mixed together and
homogenized to produce one uniform ICS sample. A sub-sample is collected from the
homogenized ICS soil sample and sent to a laboratory for analysis. The analytical results for the
sample are referred to as the average or mean concentrations of the DU. Generally, the collection
of three replicate ICS samples is recommended so that reliability of the sampling methodology
can be assessed.
In comparison to more conventional sampling approaches involving the collection of discrete
samples, results from ICS applications have shown concentration data to be more consistent, less
variable, and more reproducible. The use of an ICS sampling methodology is also more likely to
provide a better representation of the DU and is more effective in identifying heterogeneous
contamination. ICS replicate samples generally exhibit a normal data distribution as opposed to
the positively skewed data distribution often observed from discrete samples.
The use of an ICS approach requires the use of systematic planning. Elements of the planning
required include: establishing a conceptual site model (CSM), defining data quality objectives
(DQOs), defining suitable DU locations and size, and developing an ICS sampling protocol that
states the number and size of increments that will be sampled for each DU. Typically, an
incremental composite sample is comprised of 30 to 100 increments.
Cost savings associated with the use of ICS methodology arise from the reduced number of
samples that are sent for laboratory analysis compared to the discrete sampling approach. It
should be noted, however, that cost per analysis is higher for ICS due to the additional
processing required before conducting the analytical procedures.
-------
Incremental Composite Sampling Methodology:
Source: ITRC. 2012. Incremental Sampling Methodology. February.
The incremental composite sampling (ICS) methodology is a composite sampling approach that
statistically reduces data variability associated with discrete sampling and provides mean
concentrations of contaminants within a specified area or volume of soil referred to as a decision
unit (DU). The mean concentrations are used for comparison to regulatory threshold values and
action levels, or are used for risk assessment calculations. Conventionally, discrete samples have
been collected to estimate average contaminant concentrations, but use of ICS has been
increasing.
Using an ICS sampling approach, soil increments of equal mass are collected from multiple, un-
biased locations across a defined DU. The sampling locations within the DU must be evenly
distributed to ensure representativeness. The soil increments are mixed together and
homogenized to produce one uniform ICS sample. A sub-sample is collected from the
homogenized ICS soil sample and sent to a laboratory for analysis. The analytical results for the
sample are referred to as the average or mean concentrations of the DU. Generally, the collection
of three replicate ICS samples is recommended so that reliability of the sampling methodology
can be assessed.
In comparison to more conventional sampling approaches involving the collection of discrete
samples, results from ICS applications have shown concentration data to be more consistent, less
variable, and more reproducible. The use of an ICS sampling methodology is also more likely to
provide a better representation of the DU and is more effective in identifying heterogeneous
contamination. ICS replicate samples generally exhibit a normal data distribution as opposed to
the positively skewed data distribution often observed from discrete samples.
The use of an ICS approach requires the use of systematic planning. Elements of the planning
required include: establishing a conceptual site model (CSM), defining data quality objectives
(DQOs), defining suitable DU locations and size, and developing an ICS sampling protocol that
states the number and size of increments that will be sampled for each DU. Typically, an
incremental composite sample is comprised of 30 to 100 increments.
Cost savings associated with the use of ICS methodology arise from the reduced number of
samples that are sent for laboratory analysis compared to the discrete sampling approach. It
should be noted, however, that cost per analysis is higher for ICS due to the additional
processing required before conducting the analytical procedures.
------- |