United States Office of
Environmental Protection Emergency and
Agency Remedial Response
EPA/ROD/R08-90/030
September 1990
&EPA Superfund
Record of Decision:
-------
110272.101
REPORT DOCUMENTATION /1. REPORTNO. 12.
, PAGE EPA/ROD/R08-90/030
3. Recipienfa AcceMion No.
.&. TIle Ind ..,...
SUPERFUND RECORD OF DECISION
Silver Bow Creek, MT
First Remedial Action
7. Aulhllf(a)
5. Report Dllte
09/28/90
6.
8. l'erf0nnlng Organization Rapt. No.
e. PIIrfonning Org8lnlzatlon Nama and Add....
10. ProjectfTa8kJWort Unit No.
11. Contr8ct(C) Of Grant(O) No.
(C)
..
(0)
12. Spoll80rlng OrganIzatIon Nama 8nd Addre..
U.S. Environmental Protection
401 M Street, S.W.
Washington, D.C. 20460
13. Type 0' Raport . PerIod CoMl'8d
Agency
800/000
..
14.
'.
15. Supplem8ntaryNo...
'.
11. Abahct (limit: 2110 _Ida)
The Silver Bow Creek site is a mining and processing area in the Upper Clark Fork River
Basin, Deer Lodge County, Southwestern Montana. This Record of Decision (ROD) documents
the selected interim remedial action for one of eleven operable units for the site, the
Warm Springs Ponds operable unit, which covers approximately 2,500 acres just aboye the
h~ginning of the Clark Fork River. Several onsite creeks (e.g., Warm Springs, Silver
"w, Mill, Willow) and a stream bypass (Mill-Willow Bypass) serve as principal
~adwaters to Clark Fork River. Three settling ponds (i.e., Warm Springs Ponds), an
area between the northern most pond and the Clark Fork River's beginning point, and a
series of wildlife ponds are located in close proximity to the streams. Contamination
at the site is the result of over 100 years of mining and process operations in the
area. Mining, milling, and smelting wastes were dumped directly into Silver Bow Creek
and transported downstream to the Clark Fork River with final deposition downstream as
far as 130 miles. The three settling ponds, which cover over four square miles, were
built to allow the wastes from mining, milling, and smelting operations that were
deposited in Silver Bow Creek to settle out before discharge to the Clark Fork River.
An estimated 19 million cubic yards of tailings and heavy metal-contaminated sediment
17. Doc:un8nt AnaIyaIa .. D88cr1pto1'8
Record of Decision - Silver Bow Creek, MT
First Remedial Action
Contaminated Media: soil, sediment, gw, sw
Key Contaminants: metals (arsenic, lead)
b. """""'Op8n-&Id8cI T81'1'118
c. COSATI R8IdIGroup
t!- .AYIIIIabIIty St8Iem8nt
18. SecurIty CI- (1h1a Report)
None
20. SecurIty CI- (1hia "'118)
Nnnp
21. No. of PalIN
250
22. PrIce
(See ANSI-Z38.18)
s.. m.1rUI:fJ- on ".-
. )
(Flll'm8fly NTI~)
D8paI1mant 0' Collllll8l'ce
-------
EPA/ROD/R08-90/030
Silver Bow Creek, MT
First Remedial Action
stract (Continued)
and sludges have collected in the ponds. An estimated 3 million cubic yards of
contaminated tailings remain upstream of the ponds, along' the banks of Silver Bow Creek.
Principal threats from the site include the possibility of pond berm failure due to
flood and earthquake damage that could release millions of cubic yards of tailings and
sediment to the river. Furthermore, the creeks are contaminated with dissolved metals,
and exposed soil and tailings are contaminated with elevated levels of several metals.
The primary contaminants of concern affecting the soil, sediment, ground water, and
surface water are metals including arsenic, cadmium, copper, lead, and zin~.
The selected remedial action for this site includes raising and strengthening all pond
berms; increasing the capacity of settling Pond 3 to receive and treat (using metals
precipitation) flows up to a 100-year flood level, and constructing new inlet and
hydraulic structures to prevent debris "from plugging the settling Pond 3 inlet;..
upgrading the treatment capability of Ponds 2 and 3 to treat all. flows up to the"
100-year peak discharge, and constructing spillways for routing excess flood water into
the bypass channel; flooding (wet-closure) all dry portions of settling Pond 2;
reconstructing the Mill-Willow Bypass channel, and removing all remaining tailings and
contaminated soil from the bypass, followed by consolidating these with dry tailings and
contaminated soil within the dry portion of settling Pond 1 and settling Pond 3, capping
and revegetating the closure areas; dewatering wet portions of settling Pond 1, covering
the area with a RCRA-type cap and revegetating; constructing interception trenches to
collect contaminated ground water in and below settling Pond 1, then pumping the water
to settling Pond 3 for treatment; establishing surface and ground water quality
nitoring systems; and implementing institutional controls and site restrictions.
cisions concerning remediation of contaminated soil, tailings and ground water in the
area below settling Pond I will be made within one year, pending evaluation of various
wet- and dry-closure alternatives and public review. Until these decisions are made,
soil cleanup levels cannot be determined. Three chemical- and location-specific ARARs
pertaining to water quality standards and potential solid waste disposal requirements
will be waived in this remedy. The present worth cost for this remedial action is
$57,416,000 which includes an estimated annual O&M cost of $379,000 for 5 years.
PERFORMANCE STANDARDS OR GOALS: All exposed tailings and contaminated soil in the
Mill-Willow Bypass have already been removed and placed in the closure area behind the
settling Pond 3 berm. Final soil cleanup levels will be set within one year of this
ROD. Ground water pumping and discharge for treatment in the pond system will comply
with State standards, with the exception of arsenic (0.02 mg/l) and lead (0.05 mg/l) .
State surface water concentrations of arsenic and mercury require an ARAR waiver based
on technical impracticability and the fact that this is an interim remedy. The
replacement criteria are arsenic 0.02 mg/l and mercury 0.0002 mg/l at the beginning of
-------
RECORD OF DECISION
Silver Bow Creek/Butte Area NPL Site
Warm Springs Ponds Operable Unit
Upper Clark Fork River Basin, Montana..
United States Environmental Protection Agency
-------
4.0
5.0
6.0
TABLE OF CONTENTS
Page
PART I: THE DECLARATION
Statement of Basis and Purpose. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. 1-1
Assessment of the Site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . .. 1-2
Description of the Remedy. . . . . . '.' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~~.. 1-2
Declaration
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . .. 1-6
PART II: THE DECISION SUMMARY
1.0
SITE NAME, LOCATION AND DESCRIPTION
. . . . . . . . . . . . . . . . . . .. 2-1
2.0
ENFORCEMENT ACTIVITIES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2-4
3.0
IDGHLIGHTS OF COMMUNITY PARTICIPATION. . . . . . . . . . . . . . .. 2-5
3.1 Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2-5
3.2 Public Perception of its Involvement at
Warm Springs Ponds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2-6
3.3 Public Input Regarding Proposed Remedial Action. . . . . . . . . . . . .. 2-7
SUMMARY OF SITE CHARACTERISrICS
4.1
4.2
4.3
Surface Hydrolog}' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2..8
Groundwater Hydrology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2-8
Nature and Extent of Contamination. . . . . . . . . . . . . . . . . . . . . . . .. 2-9
APPUCABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS 2-20
5.1 Chemical-Specific ARARs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-21
5.2 weation-Specific ARARs ................................. 2-22
5.3 Action-Specific ARAR.s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-22
SUMMARY OF HUMAN HEALTII AND ENVIRONMENTAL RISKS. . 2-23
6.1 Human. Health Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23
6.2 Summary of Toxicity Assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-24
6.3 Summary of Exposure Assessment. . . . . . . . . . . . . . . . . ~ . . . . . . . . . 2-25
6.4 Risk Characterization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-25
. 6.5' Environmental Risks. . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-27
-------
7.0
8.0
9.0
. 10.0
TABLE OF CONTENTS (continued)
Page
PROBLEM DEFINITION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-30
7.1 Pond Bottom Sediments. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 2-30
7.2 Surface Water. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . 2-32
7.3 Problems Upstream. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . 2-37
DESCRIPTION OF ALTERNATIVES................. '....'... .'...2-39
'8.1 Alternative 1 ........................................... 2-43
8.2 Alternative 2 ........".....'.........".', ~ . ... . . . . . . . . . . 2-46
8.3 Alternative 3 ............' ~ . . . .. . . . . . . . . . .'. . . . . . . . . . . . . . 2.:48 '.
8.4 Alternative 3 + 3A ....................................... 2-50
8.5 Alternative 4 .......................................... 2-56
8.6 Alternative 5 .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-57
8.7 Alternative 6 ($55,100,000) ................................ 2.:60
8.8 Alternative 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-61
Comparative Analysis of Alternatives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-63
9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8
9.9
Overall Protection of Human Health and Environment. . . . . . . . . . . 2-64
Compliance with ARARs . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . 2-68
Long-Term Effectiveness and Permanence. . . . . . . . . . . . . . . . . . . . . 2-69
Reduction of Toxicity, Mobility, and Volume
Through Treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-75
Short-Term Effectiveness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-76
Implementability ...... . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-79
Cost. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-81
State Acceptance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-81
Community Acceptance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-83
The Selected :Remedy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-85 .
10.1 Overall Protection of Human Health and the
Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-87
Compliance with ARARs . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . ~ . 2-88
Long-Term Effectiveness and Permanence. . . . . . . . . . . . . . . . . . . . . 2-89
Reduction of Toxicity, Mobility, and Volume,
Through Treatment. . . . . '. . . . . . . . .' . . . . . . . . . . . . . . . . . . . . . . . . 2-90
Short-Term Effectiveness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-91
Implementability .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-92
Community and State Acceptance. . . . . . . . . . . . . . . . . . . . . . .'. . . . 2-92
Summary of the Preferred Alternative. . . . . . . . . . . . . . . . . . . . . . . . 2-92
10.1
10.3
10.4
10.5
10.6
10.7
-------
TABLE OF CONTENTS (continued)
Pa oe
~
11.0 Statutory Determinations. . . . . . . ~ . . . . . . . . . . . . ... . . . . .. : . . . . . . . . . . . 2-94
11.1 Protection of Human Healtb and the Environment. . . . . . . . . . . . . . 2-94
11.2 Compliance with Applicable or Relevant and
Appropriate Requirements (ARARs) . . . . . . . . . . . . . . . . . . . . . . . . . 2-94
11.3 Cost-Effectiveness....................................... 2-96
11.4 Utilization of Permanent Solutions and Alternative
Treatment Technologies to the Maximum Extent . .
Practicable. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 2-96
12.0 Documentation of Significant Changes to Components of . .
. the Selected Remedy.. . . ~ . . . . ~. . . . . . . . . . . . . . . . . . . . . . ; . . . . . . . . . . . 2-97
12.1 The Originally Preferred Remedy. . .. . . . . . . . . . . . . . . . . . . . . . . . 2-97
12.2 Significant Differences Between the Originally
Preferred Alternative and Selected Remedy. . . . . . . . . . . . . . . . . . . . 2.;99
12.3 Reasons for Significant Changes. . . . . . . . . . . . . . . . . . . . . . . . . . . 2-100
ATTACHMENT TO PART II
APPUCABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS,
STANDARDS, CONTROLS, CRITERIA, OR LIMITATIONS FOR 1HE SILVER BOW
CREEK/BUTTE AREA SUPERFUND SITE - ORIGINAL PORTION - WARM
SPRINGS PONDS OPERABLE UNIT
LIST OF TABLES
Table 1
Summary of Areas and Volumes of Contaminated
Media. . . . . . . . . . . . . . . . . . . . '. . . . . . . . . . .'. . . . . . . . . . . . . . . . . 2-13
Table 2
Ground Water Quality Data Summary Warm Springs
Ponds Operable Unit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-19
Table 3
Relationship of Site Problems to Remedial
Action Objectivrs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-40
Table 4
Assembled Alternatives for Warm Springs Ponds
Operable Unit Feasibility Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-41
Table 5
Alternative Design Compliance Summary for
Federal and Montana ARARs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-70
Table 6
-------
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
LIST OF FIGURES
Components of the Warm Springs Ponds and Mill-Willow
Bypass Removal Action . . . . . . . . . . . . . .' ~ . . . . .': . . . . . . . . . .. 1-8
Silver Bow Creek Site Schematic. . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
Conceptual Model of Contaminant Migration Pathways. . . . . . . . '.' . 2-11
Decrease and Copper Concentrations Through the Pond System.. ~ . . 2-16 .
Location of Ground Water Monitoring Wells. . . . '.' . ~. . . . . . . . . . . 2-18
-------
1.0
2.0
3.0
4.0
PART III: mE RESPONSIVENESS SUMMARY
1.0
PART A - PUBLIC COMMENTS.
RESPONSES TO PUBUC COMMENTS, AN OVERVIEW. . . . . . . . . . .. 1-1
2.0
RESPONSES TO PUBUC COMMENTS. . . . . . . . . . . . . . . . . . . . . . . . . .. 1-4
2.1 General Comments. . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . .. 1-4
2.2 Site Characterization and Problem Description. . . . . . . . . . . . . . ~. . 1-20
2.3 ARARs and Cleanup Standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-31
2.4 Identification and Evaluation of Alternatives. . . . . . . . ... . . . . . ... . . 1-38
ATTACHMENT TO III-A
PUBLIC COMMENT CROSS-REFERENCE
PART B - RESPONSES TO ARCO COMMENTS
INmODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1-1
RESPONSES TO ARCO COMMENTS, CHAPTER 2.0 SITE DESCRIPTION
AND CONTAMINATION CHARACfERISTICS .................... 2-1
2.1 Site Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2-1
2.2 Site Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2-1
2.3 Nature and Extent of Contamination. . . . . . . . . . . . . . . . . . . . . . . .. 2-5
2.4 Previous Studies of Remedial Actions. . . . . . . . . . . . . . . . . . . . . . . . 2-19
RESPONSES TO ARCO COMMENTS, CHAPTER 3.0 SUMMARY OF mE
APPLICABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS
AND THE PUBLIC HEALm AND ENVIRONMENTAL ASSESSMENT. 3-1
3.1 Applicable or Relevant Appropriate Requirements. . . . . . . . . . . . .. 3-1
3.2 Public Health and Environmental Assessment Summary. . . . . . . . . . . 3-12
3.3 Risk-Based Cleanup Goals Development. . . . . . . . . . . . . . . . . . . . . . 3-26
RESPONSES TO ARCO COMMENTS,
CHAPTER 4.0
PROBLEM
D EFIN'ITI ON. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 4-1
4.1
4.2
Environmental and Human Health Problems. . . . . . . . . . . . . . . . . .. 4-1
Relation to the Environmental Concerns for
-------
6.0
7.0
8.0
9.0
TABLE OF CONTENTS (continued)
Paoe
-=-==
5.0
RESPONSES TO ARCO COMMENTS, CHAPTER 5.0 IDENTIFYING
REMEDIAL OBJECTIvEs AND GENERAL RESPONSE ACTIONS. . .. 5-1
5.1 Statutory and Regulatory Requirements. . . . . . . . . . . . . . . . . . . . . .. 5-1
5.2 Remedial Action Objectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5-1
5.3 General Response Actions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " ," 5-4
5.4 Institutional Controls . ,~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " ~ .. 5-5
5.5 Areas and Volumes of Contaminated Media. . . . . . . . . . . . . . . . . .. 5-6
RESPONSES TO ARCO COMMENTS, CHAPTER 6.0 IDEl'fTIFYING .
A RANGE OF MEDIA-SPECm:C ACTIONS. . . . . . . . . . . . . . . . . . . . . .. ,6-1
'6.1 Alternatives Identification Process. . . . . . . . . . . . . . . . . . . . ~ . . . . .. 6-1
6.2 Range of Media-Specific Actions. . . . . . . . . . . . . . . . . . . . . . . . . . .. 6-7
RESPONSES TO ARCO COMMENTS, CHAPTER 7.0 DETAILED
DEVELOPMENT OF THE MEDIA-SPECIFIC ACTIONS. . . . . . . . . . . .. 7-1
7.1 Media-Specific Action 1: Stabilize Pond Berms to
Withstand Floods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '. . .. 7-2
7.2 Media-Specific Action 2: Stabilize/Solidify
Pond Sediments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7-8
7.3
Media-Specific Action 3: Stabilize Pond
Berms to Withstand A Maximum Credible Earthquake. . . . . . . . . . . 7-13
Media-Specific Action 4: Remove and Dispose of.
Contaminated Surface Soils from Along the
7.4
Mill-Willow Bypass. . . .' . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . 7-15
7.5 Media-Specific Action 5: Improve the Pond
Treatment System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-16
7.6 Media-Specific Action 6: Construction an Upstream
Flood Impoundment/Settling Basin. . . . . . . . . . . . . . . . . . . . . . . . . . 7-36
7.7 'Media-Specific Actions 7, 8, 9, and 10: Isolate
Tailings Deposits and Contaminated Soils. . . . . . . . . . . . . . . . . . . . . 7-45
RESPONSES TO ARCO COMMENTS, CHAPTER 8.0, ASSEMBLY
AND ANALYSIS OF ALTERNATIVES. . . . . . . . . . . . . . . . . . . . . . . . . .. 8-1
8.1 Development of Alternatives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '. 8-1
8.2 Detailed Analysis of Alternatives. . . . . . . . . . . . . . . . . . . . . . . . . . .. 8-2
8.3 Individual Analysis of Alternatives. . . . . . . . . . . . . . . . . . . . . . . . . .. 8-2
8.4 Comparative Analysis of Alternatives. . . . . . . . . . . . . . . . . . . . . . . . . 8-26
-------
APPENDIX A
APPENDIX B
APPENDIX C
APPENDIX D
TABLE OF CONTENTS (cantin"ued)
LIST OF APPENDICES
RESPONSES TO ARCO COMMENTS, APPENDIX A PUBLIC
HEALTH AND ENVIRONMENTAL ASSESSMENT
RESPONSES TO ARCO COMMENTS APPENDIX B,
DETERMINATION OF ARARs
RESPONSES TO ARCO COMMENTS, APPENDIX C,
INITIAL SCREENING OF REMEDIAL TECHNOLOGIES
AND PROCESS OPTIONS".
RESPONSES TO ARCO COMMENTS, APPENDIX D,
-------
RECORD OF DECISION
PART I: TIlE DECLARATION
Silver Bow Creek/Butte Area NPL Site
Warm Springs Ponds Operable Unit
Upper Clark Fork River Basin, Montana
United States Environmental Protection Agency
September 1990
-------
RECORD OF DECISION
PART I: THE DECLARATION
Silver Bow Creek/Butte Area NFL Site
Warm Springs Ponds Operable Unit
Upper Clark Fork River Basin, Montana
STATEMENT OF BASIS AND PURPOSE
This decision document presents the selected interim remedial action for the Warm
Springs Ponds, an operable unit of the Silver Bow Creek/Butte Area NPL Site (original
portion), in the Upper Clark Fork River Basin of southwestern Montana. The selected
remedial action was developed in accordance with the Comprehensive Environmental
Response, Compensation, and Liability Act of 1980 (CERClA), as. amended by the
Superfund Amendments and Reauthorization Act of 1986 (SARA), 42 USC Sec. 9601, ~
~. and, to the extent practicable, the National Contingency Plan (NCP), 40 CFR
Part 300. This decision is based on the administrative record for the site.1
All determinations reached in the Record of Decision were made in consultation with the.
Montana Department of Health and Environmental Sciences (MDHES), which conducted
the Remedial Investigations and Feasibility Study for this operable unit and participated
fully in the development of this Record of Decision.
The administrative record index and copies of key aite documents are available for public revi..
at the Hissoula Public Library. the Montana tech Library on Wast Park Street in Butto and other
information repositories in the Clark Fork Basin. The complete admini.trative record may be
reviewed at the office. of the U.S. EPA. 301 South Park, Federal Buildina. Balana. HT.
-------
ASSESSMENT OF THE SITE
Actual and threatened releases of hazardous substances from this site, if not addressed by
implementing the response action selected in this Record of Decision, may present an
imminent and substantial endangerment to public health, welfare, or the environment.
DESCRIPTION OF THE REMEDY
The Warm Springs Ponds Operable Unit is
-------
.
floods, and increase the storage capacity of Pond 3 to receive and treat
flows up to the 10o-year flood;
.
Construct new inlet and hydraulic structures to prevent debris from
plugging the Pond 3 inlet and to safely route flows in excess of the 100-
year flood around the ponds; .
.
Comprehensively upgrade the treatment capability of Ponds 2 and 3 to
fully treat all flows up t.o 3,300'._cfs (100-year peak discharge) and.
construct spillways for routing excess flood water into the bypass chann.el;
.
Remove all remaining tailings and contaminated soils from' the Mill-
Willow Bypass, consolidate them over existing dry tailings and
contaminated soils within the Pond 1 and Pond 3 berms and provide
adequate cover material which will be revegetated;
.
Reconstruct the Mill-Willow Bypass channel and armor the north-south
berms of all ponds to safely route flows up to 70,000 cubic feet per
second (one-half of the estimated probable maximum flood);
.
Flood (wet-close) all dry portions of Pond 2;
.
Construct interception trenches to collect contaminated ground water in
and below Pond 1 and pump the water to Pond 3 for treatment;
Dewater wet portions of Pond 1 and cover and revegetate (dry-close) all
areas within the Pond 1 berms;
-------
.
Establish surface and ground water quality monitoring systems and
. perform all other activities necessary to assure compliance with all
applicable or relevant and appropriate requirements;
.
Implement institutional controls to prevent future residential
development, to prevent swimming; and to prevent consumption of fish by
humans; and
.
Defer, for not more than. one year after the. effective date of this.
document, decisions concerning the remediation of contaminated.soils,
tailings, and ground water in the area below Pond 1, pending evaluation
of various wet- and dry-closure alternatives and a public review.
Although the majority of known tailings and contaminated sediments and soils deposits
within this operable unit will be remediated by actions specified in this Record of
Decision, a final soil cleanup level is not selected. A decision regarding a final soil
cleanup level, which affects primarily the area below Pond 1, but also the Mill-Willow
Bypass and all dry ponions of the ponds, will be made within one year of the effective
date of this document. In addition, the final decision concerning the ultimate disposition
of Ponds 2 and 3 must be deferred until upstream sources are cleaned up and the two
ponds cease to be ne~ded as treatment ponds. Each of these decisions will be subjected to
separate public reviews, during which a range of alternatives will be examined and public
input solicited.
The s~lected remedy presented in this Record of Decision attempts to permanently
remediate the principal threats posed by contamination at the site. The remedy will
reduce or eliminate most of the human health and environmental threats present at this
operable unit, but the remedy is an interim measure for the reasons stated below. Future
records of decision, or other decision documents, will direct cleanup actions at the other
operable units and NPL sites that affect Silver Bow Creek and the Warm Springs ponds.
-------
Until those source areas are cleaned up, the effectiveness and permanence of this remedy
cannot be fully or finally determined.
One component of the selected remedy presented in this Record of Decision departs
significantly from the preferred remedy, as originally identified and evaluated in the
feasibility study and described in the proposed plan. Whereas the feasibility s~dy and
proposed plan recommended construction of an upstream sediment settling basin, and as. .
a consequence, discontinuance of Pond 2 as a treatment pond, the selected remedy
presented herein calls for storage and treatment o~ flood flows (up to the tOO-year event) .
in Pond 3 and retention of Pond 2 as a treatment pond.
The rationale for this significant change is as follows:
2. .
1.
There was considerable public opposition to the proposed upstream
settling basin. Residents of the Deer Lodge Valley were concerned about
economic and environmental impacts that might have been caused by the
impoundment.
Upon examination of an alternative proposal presented by the potentially
responsible party, the Atlantic Richfield Company (ARCO), specifically to
store and treat flows up to the tOO-year flood within Pond 3, the EPA and
State concluded that that is an acceptable alternative to the concept of an
upstream settling basin. In fact, treatment of dissolved metals in flood
waters would not have been a feature of the upstream settling basin;
however, such treatment will be possible once the selected remedy is in
place. This revised component of the selected remedy offers the
additional advantage of keeping contaminants within the existing
boundaries of the operable unit.
-------
While this departure represents a significant change to the preferred remedy identified in
the proposed plan, it was developed through constructive dialogue with the public and
ARCO. The overall remedial objectives, as evaluated in the feasibility study and described
in the proposed plan, remain unchanged.
As a result of the dialogue with the public and ARCO, which followed a series of public
meetings concerning the proposed plan, the Mill-Willow Bypass Removal Ac~ion was.
initiated. On July 3, 1990, the EP A and ARCO entered into. agreement through an
Administrative Order on Consent to unde~take expedited action on the tailings and
contaminated soils along the. Mill-Willow. Bypass. In the process of developing a wo.r:k
plan for this removal action, many state and federal agencies, ARCa, and the public have
cooperated to assure that the extensive excavation, consolidation and disposal of tailings
and contaminated soils, and raising, widening, and armoring of the north-south pond berms
are completed in a manner consistent with the overall remedy. At the time of signing of
this document, the removal action is proceeding well and invaluable experience has been
gained concerning site conditions, which will facilitate followup work prescribed in this
Record of Decision.
DECLARATION
The selected remedy. is protective of human health and the environment; attains and
complies with federal and state requirements that are applicable or relevant and
appropriate for this remedial action except where waivers, as noted, have been applied;
and is cost-effective. The remedy utilizes permanent solutions and treatment a1t~rnatives
that reduce the toxicity, mobility, or volume as a principal element to the maximum. extent
practicable for this operable unit. The use of treatment alternatives to address the human
health and environmental threats posed by the pond bottom sediments, exposed tailings,
and contaminated soils was determined not to be practicable because of the extensive
volume of material present on the site and the. absence of available technologies to
effectively treat the contaminants.
-------
.' Because this remedy will result in hazardous substances remaining onsite, a review will be
conducted within five years after commencement of remedial action to ensure that the
remedy continues to provide adequate protection of human health and the environment.
Additionally, the remedy selected by this Record of Decision will be subject to a separate
public review once work at the other NPL sites that affect this operable unit is completed.
Signature:
J&Z:'
Regional Administrator (Region VIII)
U.S. Environmental Protection Agency
~ ff',
Date
-------
RECORD OF DECISION
PART II: THE DECISION SUMMARY
'.'
Silver Bow Creek/Butte Area NPL Site - -
Warm Springs Ponds Operable Unit
Upper Clark Fork River :Basin, Montana-
United States Environmental Protection Agency
September 1990
-------
RECORD OF DECISION
PART II: THE DECISION SUMMARY
1.0 SITE NAME, LOCATION AND DESCRIPTION
The Warm Springs Ponds Operable Unit is part of the Silver Bow Creek/Butte Area NPL
Site. The ponds are located at the downstream end of Silver Bow Creek, just. above the
confluence of the Mill-Willow Bypass and Warm Springs Creek.. That confluence is the
defined beginning point of the Clark Fork River. '!be Warm Springs Ponds Operable Unit.
is the first operable unit of the Silver Bow Creek/Butte Area NPL Site for which a remedy
has been selected and a Record of Decision entered.
The Clark Fork River Basin, which includes the Silver Bow Creek/Butte Area Site, is one.
of the largest geographic areas in the nation being addressed under Superfund. The site.
has been impacted by over 100 years of mining and processing operations in the Butte and
Anaconda areas.
,
,
Mining began with the discovery of gold in 1864 on Silver Bow Creek. By 1884, the Butte
area contained over 300 combined copper and silver mines, at least nine 'silver mines, and
at least eight smelters. Many of these mines, mills, and smelters were owned and operated
by the Anaconda Minerals Company or its predecessors. Mining and smelting continued
until 1982, when the Atlantic-Richfield Company, the successor corporation to the
Anaconda Minerals Company, closed the Berkeley Pit in Butte. Mining and milling has
since resumed, with the takeover of operations by Montana Resources, Inc., and others, in
1986.
Over the years, the mining and related activities have resulted in extensive soil, water, and
air contamination within the Clark Fork River Basin, including Silver Bow Creek and the
Warm Springs Ponds. Contamination of Silver Bow Creek occurred from the outset of
. .
mining activities. Mining, milling, and smelting wastes were dumped directly into Silver
-------
Bow Creek and transported downstream to the Clark Fork River. Substantial deposits of
these wastes have been found along the 120-130 miles of river below the Warm Springs
Ponds, as far downstream as the Milltown Reservoir near Missou~a. Approximately six
million cubic yards of wastes from Butte, Anaconda, and Silver Bow Creek lie within the
Milltown Reservoir, a separate NPL site.
. .
In 1911, the Anaconda Copper Mining Company built its first treatment pond. near the
community of Warm Springs to settle out wastes from Silver Bow. Creek before the water
reached the Clark Fork River. This is n~w known as Warm .Springs Pond 1. Warm..
Springs Ponds 2 and 3 wer~. constructed in approximately 1916 and 1959, respectively,. as
additional settling capacity was needed (see Figure 1). The ponds now cover an area of
approximately 4 square miles. Over the past 80 years, an estimated 19 million cubic yards
of tailings and heavy metal-contaminated sediments and sludges have collected in the
ponds. The volume of wastes present could cover the playing area of 100 football fields
90 feet deep.
,
I.
Mining wastes are no longer released directly into Silver Bow Creek, but tailings deposits
along the creek banks continue to erode and travel down the creek, particularly during
periods of above-average flows and floods. It is estimated that approximately tpree million
cubic yards of contaminated tailings are still present along the banks of Silver Bow Creek.
Through dissolution, the tailings and sediments cause the water ,flowing in Silver Bow
Creek to be contaminated with dissolved metals. Copper and zinc concentrations are
particularly high. Other metals found to be elevated in Silver Bow Creek include arsenic,
9 .
cadmium, lead, iron, aluminum, and manganese. -
The Warm Springs Ponds are still used to contain entrained sediments and treat the
contaminated water flowing down Silver Bow Creek before it reaches the Clark Fork
River. The ponds operate by settling out tailings particles arid other solids and by
reducing the concentrations of the dissolved metals.
-------
The berms containing the Warm Springs Ponds are susceptible to. flood and earthquake
damage. Their failure potentially could release millions of cubic yards of the tailings and
sediments into the Clark Fork River. Because this could cause considerable environmental
damage downstream of the ponds, the EP A and the State of Montana identified the ponds
as the first operable unit of the original Silver Bow Creek Site to be cleaned up.
The Warm Springs Ponds Operable Unit also presents two ,other significant environmental
and human health concerns:
.
The surface waters of 3.11 three creeks (Mill, Willow, and Silver Bow)"that
enter the operable unit are contaminated with dissolved metals. The
surface water quality standards adopted under the Montana Water
Quality Act are frequently exceeded for copper and zinc within the area.
.
Large are,as of surface contamination, comprised of tailings and
contaminated soils, are present within the boundaries of the Warm
Springs Ponds Operable Unit. The tailings and contaminated soils, which
include previously submerged pond bottom sediments that are now
exposed, contain elevated levels of several metals and are either void of
vegetation or sparsely vegetated. These tailings and contaminated soils
subject humans to risks from exposure. Copper and zinc, which are
significant contaminants in the tailings, are also suspected of causing
several fishkills observed in the Mill-Willow Bypass and the Clark Fork
River.
-------
2.0 ENFORCEMENT ACTMTIES
In August 1967, the Anaconda Minerals Company received an ord.er from the Montana
Water Quality Board, requiring steps to be taken to prevent the introduction of heavy
metal salts into the Clark Fork River from the Warm Springs Ponds. In response to this
order, water from below Pond 1 was pumped back into Pond 1 for further treatment.
Additionally, in response to a fishkiU in July 1989, ARCO (Anaconda Mineral (:ompany's
successor) agreed to isolate streamside tailings deposits by constructing berms between the
tailings and the Clark Fork River. Finall~, the EPA, in July 1990, ordered ARCO to.
remove all tailings and soils. contaminated with heavy metals from the Mill-Willow ByPass.
This work is ongoing and is expected to be completed by late 1990.
Tbe Phase I Remedial Investigation Report of the entire Silver Bow Creek Site was
released in 1987.1 The Phase II Remedial Investigation Report, which concentrated solely
on the Warm Springs Ponds Operable Unit, was completed in May 1989.2 The remedial
investigations focused on the nature and extent of contamination within the operable unit.
The feasibility study incorporating the information obtained during the remedial
investigations, was released for public comment on October 26, 1989.3 The feasibility
study developed and evaluated a range of remedial alternatives for c1e.anup of the
operable unit.
Hultitech, 1987.
Phase I Ramedial Investisation Report.
2
CH2H HILL, 1989.
Phase II Ramedial Investisation Report.
3
. Hontana Department of Health and Environmental Sciences, 1989. Feasibility Study for the Warm
Sprinss Ponds Operable Unit. Volume I, Report; Volume II, Appendixes.
-------
3.0 HIGHLIGHTS OF COMMUNl1Y PARTICIPATION
3.1
BACKGROUND
Community involvement in the Silver Bow Creek Superfund Site activities began early in
the project. The initial community relations plan, in 1983, designated the Butte-Silver Bow
County Health Department as the focal point for community relations and included the
formation of a citizens advisory committee. That committee was active in the selection of
a contractor for the initial Phase I remedi~- investigations of Silver Bow Creek.
Late in 1985, EPA conducted an assessment of the Site Community Relations Plan. The
assessment recommended several improvements to the plan, including installation of a -toll-
free telephone number, preparation of fact sheets and updates, and an increase in the
number of informal public meetings or briefings. Most of these improvements were in
place by 1987.
Information repositories, containing key site studies, indexes and reports, are presently
maintained at the following locations: Montana State Library in Helena, Montana
Historical Society in Helena, University of Montana Library in Missoula, Missoula Public
Library, National Park Service Main Office in Deer Lodge, Hearst Free Library in
Anaconda, Montana Tech Library in Butte, Butte Public Library, and Montana State
University Library in Bozeman. The complete administrative record is maintained atthe
EPA's offices in Helena.
The Phase IT remedial investigation, followed by a feasibility study, began at !be Warm
Springs Ponds Operable Unit in 1986 and continued through 1989. During that time,
MDHES and EPA staff provided information about the Warm Springs ponds activities at
public meetings and through fact sheets and progress reports. These reports were
distributed to people on a mailing list (271 individuals in 1987 and 800 individuals in 1990)
in November 1986, November 1987, May 1988, July 1988, August 1988, October 1988, Jun~
-------
1989, September 1989, and May 1990. Special interest groups that indicated concern '
about the site included the Clark Fork Coalition, Butte Chapter of Trout Unlimited,
Skyline Sportsmen of Anaconda, the Deer Lodge Chapter of Trout Unlimited, George
Grant Chapter of Trout Unlimited, Anaconda Sportsmen;s Club, Pintlar Audubon, and
Upper Clark Fork Chapter of Trout Unlimited.
The Warm Springs. Ponds Feasibili~ Study and Proposed Plan were released tor public
review in October 1989. The MDHES held public' informational' meetings in Butte,
Anaconda, and Missoula during October a~d formal public hearings in the same cities in ,
December. The public comment period for the Feasibility Study and Proposed Plah ,was
open from October 1989 until the end of January 1990.
3.2
PUBUC PERCEPTION OF ITS INVOLVEMENT AT WARM SPRINGS PONDS
The EP A and MDHES received 162 comment letters and 40 people presented testimony
at the public hearings. Most comments indicated dissatisfaction with the ievel of publi
involvement in the Superfund process at the Warm Springs Ponds. The EPA and MDHES
are striving to involve more fully all interested parties and other agencies in future
activities at the Warm Springs Ponds and at other sites in the Clark Fork Basin.
Public involvement in the Mill-Willow ,Bypass Removal Action is an example of the effort
to involve the public early in Superfund activities. A public scoping meeting on the Mill-
Willow BYPass Removal Action was held in February of 1990. The agencies held five
public meetings in February and May of 1990 to gather input from the general public on
the removal activities and other 'actions planned by the agencies and ARCO. Coordination
meetings involving local government officials, representatives of interested state agencies,
and public interest groups were held in preparation for the summer's removal action. The
agencies will continue similar efforts to involve the public in the Superfund process.
-------
3.3
PUBLIC INPUT REGARDING PROPOSED REMEDIAL ACTION
The remedy selected in this Record of Decision was. developed, .to a large extent, to
address comments and recommendations provided by ARCO and the general public
during the public comment period. Several key revisions were made to the original
preferred altema.tive:
There was considerable public opposition to the construction cmd use of an. upstream
settling basin to catch and control flood fl~ws on Silver Bow Creek. This. element has .
been dropped in favor of a major upgraae of Ponds 2 and 3 to store and treat flood flows.
This upgrade includes substantial changes to the berms, as well as new intake structures
and a new lime addition facility.
There was overwhelming support for expediting the removal of tailings from the Mill-
Willow Bypass in an effort to prevent any future fishkills in the upper Clark Fork River.
This work has already been started as part of the Mill-Willow Bypass Removal Order
signed in July 1990. The majority of the removal is expected to be completed by the end
of 1990.
There was considerable support for the protection of the pond berms to the full maximum
credible earthquake .and at least h~f of the probable maximum flood.. The original
preferred remedy would have used full earthquake protection, but less than 0.5 probable
maximum flood protection for berms on Ponds 1 and 2. The agencies have agreed that 0.5
protection for all the ponds is appropriate, so the selected remedy now provides for full
maximum credible earthquake and 0.5 probable maximum flood protection for all ponds.
-------
4.0 SUMMARY OF SITE CHARACl'ERISTICS
4.1
SURFACE HYDROLOGY
The Warm Springs Ponds include, the primary hydrologic features within the operable unit.
They cover an area of approximately 2,500 acres (about 4 square miles). Three creeks
from the south and the west flow through the operable unit (see Figure 2). Silver Bow
Creek; the longest of the three creeks, flows from the south and. enters Pond 3 near the
southern end of the operable unit. Mill an~ Willow creeks from. the west and south flow. .
into the Mill-Willow Bypass, a diversion ditch, which routes the comparatively .l~ss
. .
contaminated water in these two creeks around the ponds and to the Clark Fork River.
Water flowing out of Pond 3 goes primarily into Pond 2, with a smaller volume being used
to maintain several wildlife ponds located between Ponds 2 and 3 (see Figure 1). The
effluent from Pond 2 flows into the Mill-Willow Bypass, as a regulated point-source
discharge, and then down the bypass to the Clark Fork River. The average flows in the
three creeks are 73 cubic feet per second (cfs) for Silver Bow Creek, and 27 cfs for
combined Mill and Willow creeks.
The average flow of 100 cfs in the lower portion of the Mill-Willow Bypass is joined by the
average flow of approximately 47 cfs in Warm Springs Creek at the northern end of the
operable unit to form the Clark Fork River. Warm Springs Creek is also contaminated,
possibly due to milling and smelting activities in the Anaconda area, west of the operable
unit.
4.2 GROUNDWATER HYDROLOGY
The shallow ground water system in the Warm Springs Operable Unit is complex, owing to
the heterogeneity of the near surface geology in the area. The site is in a ground water
discharge area for the upper Deer Lodge Valley, typified by shallow ground water tables
-------
and swamps. The presence of the pond system affects shallow ground water elevations
"and ground water movement within the site.
Shallow aquifers occur along present-day stream channels but do not extend laterally
throughout the site. Deeper aquifers are associated with Tertiary-age valley fill and thick
deposits of glaciofluvial material. These aquifers generally exhibit moderate to low
permeabilities and are probably connected on a regional scale, although fine-grained inter-.
beds tend to confine the deeper aquifers locally.
The uppermost. aquifer at tbe site is a "10- to IS-feet-thick sand and gravel unit, whiCh is
typically present approximately 10 feet below ground surface. This sand and gravel aquifer
appears to be present throughout most of the site. Ground water movement through the
site is generally south to north, although a significant component of ground water enters
from the Opportunity Ponds area to the southwest. (See Figure 2).
No domestic well is located within the Warm Springs Ponds Operable Unit. Several are
located east of the pond system within a mile of the operable unit, but these wells are
completed in bedrock aquifers that do not appear to be affected by the pond system. The
"town of Warm Springs derives its water from supply wells constructed in unconsolidated
Tertiary deposits, from depths of approximately 200 feet. These wells appear to be
supplied with water derived from ground water resources west of and hydraulically isolated
from the Warm Springs Ponds.
4.3 NATURE AND EXTENT OF CONTAMINATION
Sediments, surface water, soils, and ground water are all affected by contaminants in the
Warm Springs Ponds Operable Unit. A schematic that shows the contaminated areas and
the migration pathways is presented as Figure 3. Four contaminated media have been
identified for the operable unit: pond bottom sediments, surface water, tailings deposits
and contaminated soils, and ground water. The media are discussed in the following
-------
.'
~ TYPICAL
. ~ AGRICULTURAL
~->-i.~~ LANDS
':..:. ~
UPPER .
CLARK FORK
RIVER
@
NOT TO SCALE
PRO~TANA
SITE ..
I
KEY MAP
+--
~
POINT SOURCES
NON-POINT SOURCES
GROUND WATER DISCHARGE
AND SURFACE RUNOFF'
fiGURE 2
SILVER BOW CREEK
SITE SCHEMATIC
-------
fLt'IIS IN WILLlIW CREEK AND
( ",- - ~" ""..-..
OPPORTUNITY PONOS
@
r
/
\ I / "
. : . /
-\ /1>'
if"
/
" TAIlMGS AlONG MIU..wn.LOW BYPASS;
IlETAWC SAlTS WICK TO THE SURfACE
ANDAAE DISSOlVEO INTO THE BYPASS CHANNEL
DURING SUUIIER THUNDEASTORIotS . . PROBABlE
CAUSE Of FISH IQUS IN THE UPPER
ClARK FORI< RIVER
BYPASS
UPPER pH
SHACK; LI!CE
ADDITIOII 10
fAClLlTAH METALS
PRECIPITATION
HleH fLOI/S IN SILVER
BOI/ CREEK DIVERTED
UIITREATED DIRECTLT
INTO MILL-WILLOW ITPASS
MILL'WILLOW BYPASS
ACTS AS SINK fOR
COLLECTlNe DEGRADED
GRIUIIIIIATER EMANATlNe
fROII OPPOIITUIII TT POIIOS
AND WARM SPRINeS POIIOS
.'
-- -=--==-
~
.'
SAND AOUIFER
:~
. CONTAMINATED GROUHD IIATER EXIIS
POlIO SYSTEM; INTERCEPTED BT
CLARK fORK RIVER
FIGURE 3
CONCEPTUAL MODEL OF
CONTAMINANT MIGRATION PATHWAYS
-------
sections. Table 1 presents a breakdown of the areas and volumes for each of the four
media.
4.3.1 Sediments. Tailings. and Contaminated Soils
Two of the media--the pond bottom sediments, and the tailings deposits and contaminated
soils--contain the majority of the contaminants in the Warm Springs Ponds Operaple Unit.
These >materials are typically fine to coarse sand and generally contain'metals associated
with the sulfide ore body present near Butte~, Pond bottom sediments are also comprised"
of precipitated hydroxides an9 oxyhydroxides resulting principally from the addition of liIrle
to treat the water entering the pond system and from biologically mediated precipitation.
The exposed (unsubmerged) sediments, tailings deposits and contaminated soils cover an
area, of approximately 634 acres within the Warm Springs Ponds Operable Unit.
Thicknesses of these deposits range from less than 1 inch to several feet. The submerged
sediments in Ponds 1, 2, 3, and the wildlife ponds cover an area of approximately
1,227 acres and range in thickness from less than 1 foot to over 20 feet. (See Table 1.)
4.3.2 Surface Water
J
The data obtained during the remedial investigation characterize the surface water for
near-average flow rates. Few data are available to characterize the surface water quality
during higher flows because of drier-than-normal conditions in the area experienced during
the remedial investigation. No opportunity was available during the sampling period to
collect flow and contamination data during one of the high runoff events that cause inflows
to be diverted around the pond system.
Surface water samples were collected at 25 sampling points in and adjacent to the Warm
Springs Ponds Operable Unit during Phase I and Phase II remedial investigations. The
Phase I remedial investigation showed that metals are being removed from the Silver Bow
-------
TABLE 1
SUMMARY OF AREAS AND VOLUMES OF CONTAMINATED MEDIA
Volume
Area
(acra)
(acre-Ccet)
(cubic yards)
Pond Bottom Sediments
Pond 1
Exposed Sediments
Vegetated/Submerged Sediments
59 455 734,000
225 1.340 2.156.000
284 1,795 2,890,000
155 800 1,300,000
347 2.230 3.590.000
502 3,030 4,890,000
66S 6,903 11,180,000
1,451 11,755 18,960,000
Pond 2
Exposed Sediments
Vegetated/Submerged Sediments
Pond 3
Submerged Sediments
Total Pond Bottom Sediments
Surface Water
Silver Bow Creeka
Mill and Willow Creeksb
Tailin2S Deoosits and Contaminated Soil
Mill-Willow Bypassc
Exposed failings
Vegetated Tailings & Contaminated Soil
Area Above Pond 3
Exposed Tailings
Vegetated Tailings &: Contaminated Soil
21 47 75,800
33 J!Q 130.000
54 127 205,800
22 56 90,300
268 700 1.130.000
290 756 1,220,300
17 48 77,400
59 ~ 397.000
76 294 474,400
180
Area Below Pond 1
Exposed Tailings
Vegetated Tailings &: Contaminated Soil
Ground waterd
Area of contaminated aquifer beneath & downgradient of Pond 1
bRow ranges from 28.112 ds (73 ds average). Data cOllection from Marcb 1985 to August 1985
Row ranges from 3-87 ds (21 ds average). Data collected from December 1984 to August 1985.
clnscn Mill-Willow Bypass tailings and contaminated soils are being removed by an expedited action schedule for completion in November
d 1990.
Exceedences of primary maximum contaminant levels for arsenic and cadmium.
-------
.Creek flow by the current pond treatment system. Inflow loads of total copper and total
zinc were reduced by over 90 percent by the time the water left the pond system during
the summer months and by 50 to 70 percent during winter months. Although metals
concentrations are reduced' in the pond system, Montana's chronic ambient water quality
standards for copper, lead, and zinc were occasionally exceeded in the water leaving the
pond system, particularly in winter months. Ambient standards for cadmium and iron
,were also frequently exceeded during the sampling events.
Four 24-hour, or diurnal, sampling ei>isod~s were completed within the Warm Springs..
Ponds system during the Phase II remedial investigation to gain a better understanding of
changes in water quality over I-day periods and on a seasonal basis. These sampling
episodes were completed in September 1987 and in January, April, and July 1988.
Hourly data from the diurnal sampling studies have been compiled.4
The data for the 24-hour sampling episodes indicate the following:
. pH varied by up to 2.2 units throughout the day at all stations sampled.
. Total metals concentrations decreased 50 to 90 percent between pond system
inflow and outflows.
. Dissolved metals concentrations for copper and zinc were generally 20 to
50 percent higher in the winter at all sampling stations in the pond system.
Higher dissolved metals concentrations in the winter correlate directly with lower
pH values measured during winter sampling events.
4
CB2H BILL, 1989.
Ph... II Rem.dial Inv..tisation Data Summary.
2 - 14
-------
The pond system reduced metals concentrations at the outflows from the system during the
four diurnal sampling events, frequently to levels below both chronic and acute aquatic -
standards. Figure 4 shows an example of this phenomenon recorded during one of the
diurnal sampling events.
Removal of metals in the ponds i.s accomplished by physical, biological, and chemical
processes. Physical reduction of met~-bearing solids occurs through simple sediIIlentation.
Increases in pH, which are partly due to the addition of lime and partly due to
photosynthesis, can precipitate metals as a result of changing p1etals solubilities. Yet
another important metals removal mechanism may be the precipitation of calcite',and
coprecipitation of metals and phosphorus, which follow the photosynthetic removal of
carbon dioxide and a compensating shift in the bicarbonate buffering system.s Direct ,up-
take or absorption of metals by algae and aquatic macrophytes is also probable. Addition
of lime to the Silver Bow Creek inflow during the winter months also contributes to
precipitating metal contaminants when the amount of sunlight to support photosynthesis is
reduced.
Several fishkills have occurred in the Mill-Willow Bypass and in the upper Clark Fork
River, with the most recent known episode being in July 1989. Analysis of fish tissue by
Montana Department of Fish, Wildlife, and Parks from- one event in the summer of 1986
revealed acute copper poisoning as the cause of the fish mortality. Although MDFWP did
not determine the source of metals responsible for the killings, that source most likely
consists of tailings material along the Mill-Willow Bypass.
5
Wetzel. R.G., 1975. L1mnoloav.
Philadelphia: W.B. Saunders Company.
-------
POND 2
OUTLET:
FIGURE 4
DECREASE IN
COPPER CONCENTRATIONS
THROUGH. THE POND SYSTEM
WARM SPRINGS PONDS
400
Ambient Water Quality Standards
Based on Average Hardness 0' 160 mg/I
300
.....
m
~
" ACID SOL C PPER
"~,I
,
,
,
,
,
,
,
"
"
"
"
z
o
~
cc
a:
~
z
.w
U
Z
o
u
200
100
o
55-28 PS-12
POND 3
INLET
POND 3
POND 2
. INLET
POND 2
NOTE:
BASED ON PHASE II RI,
24 HOUR SAMPLlNG--SEPTE!v1BER 1987.
SOURCE: PHASE II RI REPORT; CH2M HILL 1989
ACUTE
STANDARDS
CHRONIC
-------
4.3.3 Ground Water
Ground water quality data were generated through sampling of 19 monitoring wells on two
occasions (January and May, 1988) Figure 5 shows the locations of the monitoring wells at
the site. Table 2 summarizes ground water quality data for these monitoring wells.
Ground water beneath Ponds 2 and 3 may be contaminated also. Wells were not installed
to determine the quality of the ground water beneath those two ponds. (}iven the
hydrogeology of the site, contaminated ground water under the ponds' would flow north
and be detected at the northern end of the pond system.
With one exception, all detected exceedences of the primary maximum contaminant levels
for metals (arsenic and cadmium) were north of the Pond 1 berm. Ground water quality
"
downgradient of Pond 1 is generally of poorest quality immediately north of the berm;
most metal contaminants decrease to the north, or downgradient of the pond system.
Concentrations of most metals also decrease with depth.
Highest concentrations of metals are generally associated with the shallow sand and gravel
aquifer in the area immediately below the Pond 1 berm. Calculations of ground water
discharge from the area below Pond 1 into the Clark Fork River indicate that the ground
water system contributes very little flow to the river because of the relatively low
permeability and low gradient of the shallow aquifer. Under average conditions, the flow
in the Clark Fork River is approximately 137 cfs, while the ground water discharge to the
river is approximately 1.0 cfs. Nevertheless, th~ exceedences of the maximum contaminant
levels for arsenic and cadmium in the ground water constitute a violation of the drinking
water standards.
-------
. .~. J'" .., .._~ .-..- .- ~'; . --' '(,--'"
f.' . ~ . .. _.\ ".~"". ".,--_.. ~.-.. - -~~.'" :;:"-,- r-'ij \ .)5~~...\
'. , . / .,~~. ~,..' T... .'.. ".' - '\.1.<.., \, "'" ':J':'-~L ~'-Y'(?" ),. .'...:; , . . -)1' ..t~\J(:~
. f' .-...'tf(.,~.~sr ; j'" ,,,,"., -, . \ . "..:. .--" " ." \; \ ".; I . \ --,.' o.'''/i~'.' ...~,~,\\I
~-,«;JJr -"0' . 1',., . \ "" -,.. I . J. 'I . \rp-'4"~
,.-' JJ.... :"j '....{ ,...... .:..-~,,;:. -"""", z\ .;':-,\. .-.~~~ "'...,-.;1 . ~ .f,; !\~0~'~'~1'
. -' -:. -- \-.. t.. I~' . '", ,'- .;/~..... ~ ~.' +. / ".' '-. ...," " ,.J-~/I " j ..'""': (;? 1: J ?~'i-~;~:.~.
< Ld .' "'. "" J 4" \ " ., I, ..;,...' .,,,....,) J,""" : \ i~ ?£i.J!..''\l!
~. ... " !...,... I '..- .," /" -., ,1...__).', .-....-} :.-..~~. ,1'(- r" ......."."t:ii: ~ /'.- ~. ,"; ~' -/ ,~\ ", 1;~i~~7'
--....; ~. / .- . .,'''..' --_.__.~ ~1"-'1.~'"" ----- '''".If ! ,:-<.-~~
. ,,,' :'- n : .,i....' I I.~':'~:~: ~.~ ~;. f~")G:'\r: ~;\ "'-F:~ ~..:~,-"",;,'~,\J/t:.~:;..~~~
. ' I.../!",-' l I l' - '.- ". .-,/ ~. :;,t, ,I .. ~.'--"'I, .1 ,~-........~,:" ','''!......" ,,'
, " ',; u- >v/::'" {" "':._,v/~;>:~,;~:/?fi~ri~"f0,t,~.~~
. ",,' .' . )to ~ '~". ,:, :, :J~..:: ~ I , ..1,{ ! ~ -....;,... . .' >;..'Q..12' \ ",,!;;;..,..-'
- - ) /.;'- \ '~)~:;.' ~'ik.~J.'" ell ~.'W.; ,.;;rr"'," .QW.o2..;.'r-oo(:,..'~ \ \ jl'::,;",-
- ,,,~,"', /' ~' -./.;. i ~ . '~... :><..'.~~'~;:;' --,' . ,-;..~$. 'f..- -0~-- 0,,"'.'0..18 f""j'7 \; j:,\;,'~,
- 1#'; - \..', - .../ :', ~ ' - '~'h.:.,,-:-._.~~J.,.r\ -;:~~-"1' : .-.. :1,~:>~~/// ~., '"'--,w:.o1'~ :.:.,{:~ i\.) t{-~
.;. ,L-.. r-:-'--- _h_..U.- "'~,-~. \,", \<1, #.,t.l.b;-""" ,..~~~, /') ',' )"'.,IQW"4" \il).;
. \\ J'" ": t-...n . : ,..' ';',-l~-'~~ {~~{' ~:>.,: iJJJ;'/ . ,eQ~:10 ,~,/ t~~J;'~',~ ,'.:'( :\SS
\ I .. ., ~ \!; "(0' .t1<:~-'~:J~," /~~" . 'f ~OW'1H.'1.':I;!:,'i~
. ':. -'. I,~,' ."--(~~) :' ",,'!!'J~~Y~~,1}.~ J./~ .."',~~,.:", ...,>...::~~~~:/f.:k~.
If-''-- . _.-I~ ( , .' ''!: .,A> ,-""."-if' .. '" -:.. ",- r < ~.,.".....,.:.. r, ",
.! \ ,~,"'I" lf~~v/ ~~J;{",~!:,~;,)t( ;' ii:,. ~'i;~.:.:. 'lIoW'" '.c'~::::~~'(L\i~
.~ ,,: , r.... J ,0';;', , 'I 'iJ' 'ow... : \ ,'" "1'-.::), : "
:~~: .2"t '--k)":S;:;?C}?;".p:"ll~;\" ~; -\, "?)!:::l
,. ~. " . I . ¥~ ." ,,.-,, . ,'''', :.-!! """.' ".\ ". '-'.,'~ '\ '.' , .,.~.. 'e=--'
-..- ",.'/.' ~:;" .y~-. 'r~ -. ! -v". :-.~~ -- ~!,:>~'-'~ j~~l #..~.\ C;:'..,-;\, , .:!'! .. ."''';~ . \'~,,;:~":r'~
-.' '.., ~ -... . -"" '-,. "t"'" -~-::'1'" r'.,., .' 'I" ~- \ ',',. ". " ..~!'I'K '.\'''~-
:--"\ (.;:~',~~.i.~~,:~j~:1..~;~~~ :-'! .:' ~~~T.:~T~"~;:'~_~h'\~':' "~'_~'~I' '~.::" W r~~~:\ ,u,>~::::::.:;~~,j!:" :/:.,;"~." .,~, ) .f?!,:;~:.
"~~ \ ,~--"1';,, "~' ..-'" . --,1,~ L",.~". ,. ,. ',' '. " '''''~'',...
""..:.CJt';':'" . . I ~...--,~:::::\.>,.,.~.\;:~.., "-~~~~,Y"f~(r? (' ..,\L:\,:~..;~, !,' k~~~;;~\~~~l\ )' / " ,.'< .> > <~; : ~"~)I{.'{:;.~
'8' ';y~'i- '\' l-.~". ,"-,") '< ~'J! ,1./ ' J.4}f ~( ,'>;:: , ih1\$
. . ,/, ) j ,,\"--' ,,~ ~._,. .~. it I (,,, ;r"~"" "
~~<2~:, '-;: -,.' ". 11-:'...-, '\'\ ';'- )i/'~j > >1',;1 . JJ';}t~~J!~S\!~:~~'~~:
\ I ." "/ "~\i " \ ,.,'-.. I it ff.";;~~"'=:"'~
l,~ "/' ," ' , '" ~" T" ,,,:. J1t ' .;'., ""'"i '~'..',,,"'.~- -.~
.~
'.,',
:-:"'-.7."
,"..:, "
@
LEGEND
. . PHASIf I AI MONITORING WELL
. PHASE. AI MONITORING WELL
. PHAse. PUMPING WELL
FIGURE 5
LOCATION OF
GROUNDWATER MONITORING WELLS
WARM SPRINGS PONDS
o
.
2000
.
BCALE"'FEET
"
~
-------
TABLE 2
GROUND WA11!R QUAUIY DATA SUMMARy
WARM SPRINGS PONDS OPERABLE UNIT
Muimum Contaminant
M.uim MiDim,!m (a) ~~(a) Number Lcve.JSal (MOOtaDa Ground
Panmc:tcr CooccDtra:i:.(a) CooccDtraboll. of Samples a er Rcgulaboas)
Upgradicat MoaitoriD& Welk
nic 6.8 2.6 4.3 8 SOb
dmium 7.0 <5.0 3.4 8 lOb
Copper 9.7 6.1 5.8 8 1,
-------
5.0 APPLICABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS
Applicable or relevant and appropriate requirements (ARARs) are a basic standard by
which all aspects of contaminant cleanup are measured. . 'Compliance with ARARs or
invocation of an appropriate ARAR waiver, is required by Section 121 (d) of CERCLA.
The feasibility study evaluated potential compliance of the developed remedial alternatives
with federal and Montana ARr\Rs. Compliance with ARARs is a threshhold
determination for selection of a remedy. 40 CFR ~ 300,430(f)(i)(A). ..
'.
The discussion of ARARs ir1 this section is a general discussion, which highlights the major
ARARs for the remedial action. A full list of all ARARs and compliance points, as well as
information to be considered ('TECs"), and other relevant legal requirements, is conta~ned
in the attachment to Part II: The Decision Summary. The basis for EPA's selection of
the ARARs is given in the feasibility study and Part TII, Responsiveness Summary.
ARARsare divided into three categories: chemical-specific, location-specific, and action-
specific. Chemical-specific ARARs include laws and regulations that set human health- or
environmentally-based numerical values governing materials having certain chemical or
physical characteristics. These values set the acceptable concentrations of chemicals that
may be found in, or released to, the environment. Location-specific ARARs restrict
contaminant concentrations or cleanup activities due to the site's geographic or physical
. .
location. Action-specific ARARs are based on actions taken during contaminant cleanup.
Section 121(d)(4)-of CERCLA, 42 V.S.C. ~ 9621(d)(4), provides for the waiver of ARARs
if certain criteria are met. This Record of Decision waives two ARARs for surface water--
arsenic and mercury--and establishes replacement numeric limitations for those standards
waived. The waivers are based on technical impracticability from an engineering
perspective, as permitted under section 121(d)(4)(c) of CERCLA, 42 V.S.C.
-------
~ 962l(d)(4)(c). The replacement criteria will remain fully protective of human health and
the environment. The replacement criteria are:
Mercury: 0.0002 mg/l
Arsenic: 0~02 mg/l
There is uncertainty over whether creation of permanent disposal facilities within Ponds 1
and 3 and the Pond 2 and 3impouI;ldments in place is in compliance with a relevant and
appropriate requirement from the State's Solid Waste Disposal Regulations, which
prohibits disposal of solid waste within th~, 100-year floodplain." EP A believes that the, ,
waste units will be outsid~ of the floodplain when the Pond berms are raised' '~nd
strengthened to specified standards. Even if the, water within the, ponds is considered part
of the floodplain, the disposal units are probably outside of the 100-year flood pool of the
water within the Ponds. To the extent the areas within the pond berms are considered to'
be within the 100-year flood plain, EPA waives the Solid Waste Disposal ARAR pursuant
to section l2l(d)(4)(c), as technologically infeasible from an engineering perspective and
pursuant to section l2l(d)(4)(A), as an interim action.
Additionally, if it is later determined that the area within the Pond berms is within the
100-year floodplain, then a waiver of the state's solid waste disposal regulations,
prohibiting disposal within the lOO-year floodplain, is invoked, on the same bases as above.
5.1
CHEMICAL-SPECIFIC ARARs
The most significant state and federal chemical-specific ARARs consist of standards
protecting the. quality of surface and ground water resources for human health and
environmental purposes. Surface water ARARs include ambient water concentration
limits to protect both aquatic life and public health, point source discharge standards for
discharges from the pond system, and drinking water standards. Ground water ARARs
include only drinking water standards. The contaminants of concern at the site are
arsenic, cadmium, copper, iron"lead, silver, selenium, mercury, aluminum, and zinc.
-------
5.2
LOCATION-SPECIFIC ARARs
Important location specific ARARs include cleanup activity restri~tions to protect and
minimize impacts on historically significant features and endangered species.
5.3
ACTION-SPECIFIC ARARs
Action-specific ARARs pertinent to the Warm Springs Ponds Operable Unit include
regulations concerning dam safety in event of floods and- earthquakes, ,hazardous waste
management and land reclamation for mining areas.
Dam safety regulations address berm design and modification for the existing treat~ent
system. Hazardous waste management ARARs include requirements fqr contaminant
disposal. Reclamation ARARs require proper grading, backfilling, subsidence
stabilization, water control, revegetation and other measures needed in surface mining
areas to eliminate damage from soil erosion, subsidence, landslides, water pollution, and
hazards dangerous to life and property.
-------
SECfION 6.0
SUMMARY OF HUMAN HEALTH AND E1'MRONMENTAL RISKS
A public health and environmental risk assessment was conducted by the Montana
Department of Health and Environmental Sciences to identify and characterize the actual
and potential threats to human health and the environment posed by contaminants present
at the Warm Springs Ponds Operable Unit. Carcinogenic and noncarcinogepic human
health effects were characterized, as were significant environmental effects. With respect
to both human health and the environment.! endangerment was established.
6.1
HUMAN HEALTH RISKS
The EP A has determined that the Warm Springs Ponds Operable Unit poses the following
actual or potential endangerment to-human health:
. Workers at the ponds face an increased risk of cancer estimated to be 2 x 104,
or two excess cancers in 10,000 individuals exposed for a lifetime, due to
incidental ingestion of arsenic iri the contaminated soils, sediments and tailings.
Recreationists (hunters, fishermen, bird watchers) also face increased cancer risk
from exposure to arsenic.
. Workers and recreationists face additional cancer and noncancer health risks due
to ingestion of lead and o~her hazardous substances in the contaminated soils,
sediments, and tailings.
,
. Current residents adjacent to the ponds face actual or potential risks from
contaminated soils, sediments, and tailings becoming wind-borne. If homes were
to be built within the operable unit boundaries, residents would also face risks
greater than the levels noted above.-
-------
. The contaminated ground water below Pond 1 poses a potential threat to users
of the groundwater.
. The berms protecting the ponds fail to meet current dam safety standards. Their
failure due to a flood or earthquake could result in catastrophic consequences,
including loss of life.
The baseline risk assessment establishes current and poteniial threats' 'to human health.
40 CFR ~ 300.430( d)( 4).
The NCP states that the goal of a Superfund cleanup should be reduction of risk to
acceptable ranges, if ARARs do not exist or are not sufficiently protective. The poipt of
departure, or target risk range, is 1 x 10-0 for cancer risk and levels that do not create
adverse effect, incorporating a margin of safety, for systemic toxicants. 40 CFR ~
300.430( e )(2)(i)(A)(2).
The preamble to the NCP states that the 1 x 10-0 risk range should be the goal of any
cleanup, unless revision to a lesser protective level is appropriate for site specific reasons.
55 FR 8715-8717. Risks should not exceed 1 x 104.
6.2
SUMMARY OF TOXICITY ASSESSMENT
Arsenic, a known carcinogen, is present at this operable unit. Samples of exposed tailings
and contaminated soils contained a maximum arsenic concentration of 597 mg/kg and an
average of 349 mg/kg arsenic. Lead, a hazardous substance that is both a suspected
carcinogen and toxic noncarcinogen, is also present at elevated concentrations (maximum
of 1000 mg/kg and average of approximately 490 mg/kg). Risks from lead were not
quantified in the risk assessment, but the presence of lead risks is noted. In addition to its
suspected carcinogenic effects, lead is known to damage the central nervous system anJ
cause other serious health effects. The EP A believes there is no safe threshhold for l~aJ
-------
.intake.
Other hazardous substances, such as cadmium, are also present at elevated
concentrations.
6.3
SUMMARY OF EXPOSURE ASSESSMENT
In addition to serving as an active water treatment system for contaminants transported by
Silver Bow Creek, the Warm Springs Ponds and surrounding area also function as. a
wildlife management area. Since two employees of the Mont(Ula Department of Fish,
Wildlife and Parks work within the ope~able unit, managing.. the wildlife area, their..
occupational exposure was evaluated. A recreational exposure scenario was also evaluated
. .
because hunters and fishermen are often present at the ponds. The risk to current
residents was evaluated because several homes are located near the operable. .unit
boundary.
As required by EP A policy, the risk assessment also examined risks. under a future
residential scenario. Because the operable unit is comprised almost entirely of the ponds
and associated wetlands, EP A considers it unlikely that homes will be built within its
boundaries. To ensure that future residential development does not occur, the Record of
Decision requires implementation of institutional controls. The remedy then focuses on
active measures to address the occupational, recreational, and environmental threats.
The current human exposure routes are summarized on Figure 6 for each exposure
scenario. The principal component of human health risk comes from incidental ingestion
of arsenic during occupational activity.
6.4
RISK CHARACfERIZATION .
The risk assessment evaluated risks from carcinogenic elementS such as arsenic, lead, and
cadmium, and risks from numerous noncarcinogenic elements such as copper, iron, lead.
and .zinc. The human health risks from noncarcinogens are evaluated based on their
-------
Groundwater
~ ...
.0 Q)
Qi >
m a:
- .:!It.
II) ...
m &
a. .:!It.
>- ...
m CIS
B C3
.S 0
Q) ;;
~ ~
.c c
u 0
.!!} D..
C
Surface Water
c c
.Q 0
- ~
~ e-
O! 0
C II)
- .D
n; <
C Iii
Q) e
1) ...
.- Q)
~ c
Recreationists
~ II)
CIS C
- 0
::J -
U -
~ ~
D.. ~
- CIS
o ~
O! .9-
~ g
s:: ...
c7j D..
Sediment
C C
o 0
.~ ~
:g e-
O! 0
.E ~
S -
C CIS
~ ~
~ ~
---..--.-..--
Channell
Pond Bottom
j!)
C
CIS
C
~ .~
5 ~
o .Q
- I.L
o }
-g
!
Jt
Floodplain!
Exposed Areas
Bioconcentrationl
Uptake
U
:.:: II)
S ~
Q) 0
~ u:
- .r:.
o O!
C .-
o J:
~.' ~
~ II)
:c 19
~. II)
o ~
:g 0
a: .
ta .... Air
po
-
C
Q)
e
C
.~ -
II)
- ::J
C
W C
1) C
C Q)
~ ~
-.J
C
Q)
e
'6
~
'0
Q) C
0
e ~
CIS nJ
~ n;
.c
C
II) -
u:
---.-..-...-........--....-...
,:,
Adjacent
Residents I
810
-
o
C
o
:0:
II)
Q)
O!
.E
. Workers
(Occupational)
FIGURE 6
POTENTIAL CURRENT HUMAN
PATHWAYS OF EXPOSURE
-------
. hazard index. If the combined chemical hazard index is. greater than one (based on a .
detailed calculation presented in the risk assessment), then an unacceptable risk is present.
Although some risks due to noncarcinogens were found, the hazard index was in all cases
less than one. As indicated previously, lead was not quarititatively evaluated in the risk
assessment. . However, the EP A believes there is no safe threshhold for lead intake.
Although copper and zinc do not present a risk to human health, they do pose significant
risks to the environment, especiallYH to aquatic organisms.
The maximum excess lifetime cancer risk due to arsenic ~.xp()sure (arsenic is . the. .
cont.aminant of primary co~cem) for work~rs at the ponds is estimated to be 2 x 10:4, or
two excess cancers in every 10,000 exposed individuals. This estimated risk is based on
exposure to maximum measured concentrations of arsenic in exposed tailings.. and
contaminated soils present at the Warm Springs Ponds, but excluding the . Mill-Willow
Bypass.
Because of difficulties in developing risk-based cleanup levels for the occupational and
recreational scenarios, EP A has elected to delay selection of. a specific health-based soil
cleanup action level. The EP A will continue to examine appropriate methods for
calculating specific soil cleanup levels for this operable unit. Nevertheless, EPA is
confident that the risk assessment has demonstrated actual and potential risks posed by
conditions at this operable unit to justify the Record of Decision requirements. The next
section, concerning environmental risks, explains how the human health risks will be
reduced by mitigation of the environmental risks.
6.5
ENVIRONMENTAL RISKS
The EPA has determined that the Warm Springs Ponds pose the following actual or
potential endangerment to the environment.
-------
. Periodic fish kills have occurred due to salts of copper and zinc washing from.
tailings deposits into the Clark Fork River during thunderstorms. Contaminate
soils, sediments, and tailings also pose an unquantifiable .chronic risk to aquatic
life and wildlife, both within the b01:1ndaries of the. operable unit and in the rive!
downstream.
. Water quality criteria for. the protection of aquatic life have been e~ceeded by
water discharged from the ponds, and by water routed around the ponds without
treatment.
.
The berms protecting the contaminated pond water and sediments fail to meet
current dam safety standards. Their failure due to.floods or earthquakes .could
result in catastrophic environmental consequences in the Clark Fork River.
Although this Record of Decision does not require a specific soil cleanup action level,
. EP A is confident that the risk assessment has sufficiently demonstrated the actual an
potential environmental risks posed by conditions at the Warm Springs Ponds to justify the -
cleanup requirements.
The actions required by this Record of Decision are necessary and appropriate to address
the risks described above, even though an exact quantification of acceptable risk levels was
not determined. The actions required will reduce or eliminate the principal risks. This
statement is based on the knowledge that several components of the selected remedy
require excavation or covering of exposed tailings, sediments, and contaminated soils. For
example, drying and covering Pond 1 will retard or stop the ground water contamination
which currently exists, and increasing the operational level of Pond 2 will flood areas of
contaminated soils, sediments, and tailings, thereby reducing exposure by direct contact to
those areas.
-------
6.6
FUTURE RISK ASSESSMENT ACTIONS
The determination of a final soil cleanup action level, which will be necessary for
contaminated areas deferred by this action, and appropriate' measures to remediate those
areas, will be made within one year of the effective date of this document.
-------
7.0 PROBLEM DEFINITION
Eight environmental and human health concerns were identified for which the feasibility
study developed remedial objectives and alternatives for remedial action. The eight
. . .
problems are based on the results of the remedial investigations, the applicable or relevant
and appropriate requirements (ARARs) analysis, and the public health and environmental
risk assessment.
The eight human health and environmen~~ problems are describ.ed in terms of four
contaminated media: (a) pqnd bottom"sediments, (b) surface water, (c) tailings deposits
and contaminated soils, and (d) groundwater. The contaminated media are discussed
below in terms of the problems each medium presents to the Warm Springs Ponds
Operable Unit.
7.1
POND BOTTOM SEDIMENTS
Dam Stability During Floods. Montana's dam safety rules control the minimum level of
flood protection for the design of dams within the State of Montana. The dams at Warm
Springs Ponds are classified as high hazard dams for which the State's dam safety rules
require the ponds' outlet structures to pass varying fractions of a probable maximum flood.
As the volume of w~ter stored increases, the fraction becomes greater, t~ a maximum of
one-half. The pond berms, as currently constructed, would likely fail during a moderate to
major flood. In the event of partial or catastrophic dam failure during such a flood, the
contaminated pond bottom sediments could cause incalculable damage to the Clark Fork
River.
Dam Stability During Earthquakes. The Warm Springs Ponds are located within or very
. .
near the northern section of the Intermountain Seismic Belt, which is a zone of major
-------
. earthquake activity within the North American tectonic plate.6 At least 230 earthquakes
with magnitudes greater than 4.0 have occurred at epicenters within 187 miles of the
Warm Springs-Butte area during the last 107 years of recorded earthquakes.'
The ground-shaking that occurs during an earthquake can cause berms that are not
adequately designed or constructed to flow somewhat like a liquid, causing them to slump
and release the water and semisolids behind them. Earthquakes can also cause sloshing of
the water in a pond, creating great waves that overflow and erode herms, often causing
berm failure. A review of the limited information available on the construction of the
Warm Springs Ponds berms shows that they are not strong enough to withstand' even
moderate earthquakes.
The Montana dam safety rules require that if a dam is in a region subject to earthquakes,
the dam must be design~d to withstand the most severe earthquake that can be reasonably
anticipated. This design earthquake is known as the maximum credible earthquake.
A review of available information regarding the embankment materials confirms that the
east-west and north-south berms are likely to fail in a moderate-to-severe earthquake. The
likelihood of failure appears to be greater than previously reported. This was determined
by a preliminary stability evaluation performed for this study, which indicated that the
downstream slopes of the berms have potential to fail at accelerations from 0.05 to 0.07 g
(g is the standard symbol for the acceleration of gravity). For comparison, in 1981, the
International Engineering Company determined that the acceleration at Warm Springs
Ponds during a maximum credible earthquake could be as high as 0.23 g. These
preliminary conclusions will b~ investigated further and confirmed during the remedial
design phase.
6
International EDlineerine Company (IECO> , 1981. Geotechnical and Bvdroloaic Studies. Warm
Sprinas Tailinas Ponds. Anaconda. Hantana. Prapared tor Anaconda Copper Company, Donver, Colorado.
7
I!lli! .
-------
"
.Failure of the upstream slopes (faces) of the berms was not examined in this study because.
information on the materials and construction of the upstream slopes was not available.
Duri'ng the remedial design investigation, the potential for upstream slope failure also will
be investigated.
Failure of the berms during an earthquake could result in at least partial release of the
. ,
contents of the ponds. The sludges and tailings in the ponds are sufficiently liquid that.
they could migrate a considerable distance if released from the ponds. . Although it has not
been determined that the tailings in Warm Springs P()nds could. also exPerience sponta-.
neous liquefaction, this is a process that has been observed in seismically induced fail~~es
of other tailings ponds. If this process did occur, the tailings and sludges could flow for
miles, contaminating the Clark Fork River downstream.
7.2
SURFACE WATER
Fishkills in the Mill-Willow Bypass. There have been five documented fish kills associated
with the Mill-Willow Bypass since 1983. They occurred on August 9, 1983; August 2, 1984;
July 3, 1987; May 27, 1988; and July 13, 1989, and are documented in Montana
Department of Fish, Wildlife, and Parks memoranda for these years.
All five fishkills followed a similar pattern. They were associated with locally intense
thunderstorms in the Warm Springs Ponds area, usually after extended dry periods. The
flshkills started in the Mill-Willow Bypass and extended down the Clark Fork River for
various distances. In the 1984 event, over a thousand dead fish were observed in a 15-to
20-mile stretch of the Clark Fork River. In July 1989, over 5,000 dead fish were reported.
The fish kills have been linked to high concentrations of copper in the water; zinc
concentrations and low pH levels may also be factors in fish mortality.
The available data indicate rapid elevation and dissipation of the metals concentrations
during storm events, which implies that they are derived from a readily avail~ble source of
-------
highly soluble compounds, Le., metal salts. A source of such salts has been identified
along the Mill-Willow Bypass. During extended dry periods, salts of copper and zinc form
by surface oxidation or the evaporation of soil moisture on the tailings deposits that exist
along the bypass. There are approximately 21 acres of tailings deposits along the bypass.
The copper salts are clearly visible on the tailings deposits during warmer months as
green- and blue-colored surface deposits.
The postulated mechanism for the fishkills is that the rain water dissolves the metal salts
and washes them into the bypass, resulting in metal concentrations high enough to cause,
mortality. Elevated levels ()f metals detected in the gills of dead fish suggest that the' ~sh
were exposed to acute levels of metals.
While transient phenomena such as the observed fishkills are difficult to study and even
more difficult to model, the evidence available at this time points to the visible salts on the
tailings deposits as the primary cause of the fishkills. .'
Metal Loads in the Stream Flows. Silver Bow Creek, and to a lesser degree Mill and
Willow creeks, are all contaminated with detectable levels of heavy metals; primarily
copper, arsenic, lead, and zinc. For example, in the Phase I remedial investigation, on the
average, the inflow to the ponds, the discharge from Pond 2, and the combined flows of
Mill and Willow creeks exceeded Montana's chronic water quality standard for copper in
effect during the Phase I remedial investigation. The standard was exceeded in 100, 70,
and 60 percent of the samples for those three sampling points, respectively.
The Mill-Willow Bypass was constructed to route the comparatively cleaner Mill and
Willow creeks flows around the ponds and to the Clark Fork River without mixing with
the comparatively more contaminated Silver Bow Creek flow. However, recent data
indicate that, although Mill and Willow creeks are cleaner than Silver Bow Creek, they
still contribute a portion of the total amount of metals reaching the operable unit (arsenic-
-34 percent, copper--6 percent, cadmium--3 percent, lead--3 percent, zinc--4 percent).
-------
..The pond system treats contaminated water by combinations of physical, chemical and
biological process. Physical settling of suspended solids occurs simply because the flow
velocities in the ponds are very low compared to the velocities in the creek channel. The
removal of dissolved metals occurs in part because of photosynthetically-induced chemical
precipitation, and uptake of metals by, and subsequent settling of, aquatic plants. The
effectiveness of the ponds is enhanced by the addition of lime to precipitate metals during
colder months when the amount of light available for photosynthesis and biologi.cal activity
is diminished.
Without the treatment in the pond system, the Montana chronic water quality stand~rds
for the protection of aquatic life would be far more frequently exceeded at the Pond 2
outlet immediately upstream of the beginning of the Clark Fork River. For example,
available information indicates the standard for copper (12 ug/l for a calcium carbonate
hardness of 100 mg/l), would be exceeded more than 75 percent of the time. Even though
the pond system currently treats Silver Bow, Mill, and Willow creeks, the water quality
standards for several contaminants are often exceeded, particularly in winter months. The
,
dissolved metals in the three creeks ultimately contribute to the chronic exposure by fish
downstream.
Tailings in the Mill-Willow Bypass. The total amount of identifiable surficial tailings in
the Mill-Willow Bypass has been estimated. at 79,000 cubic yards. This includes
76,000 cubic yards of exposed tailings deposits and 3,000 cubic yards of tailings with
vegetation cover. The primary source of these tailings is Silver Bow Creek. On numerous
occasions over the past 20 years, the inlet structure of Pond 3 has been plugged by flood
debris. This has caused Silver Bow Creek to enter the Mill-Willow Bypass and deposit its
sediment load-much of it in the form of tailings--along the banks of the bypass channel.
These tailings have been further eroded and transported out of the bypass and into the
Clark Fork River particularly during high flow conditions. Once deposited in and along
-------
the bariks of the Clark Fork River, these contaminated tailings add to the problems that
already exist there and thus contribute to adverse effects on aquatic organisms.
The Mill-Willow Bypass Removal Action, being conducted under an Administrative Order
on Consent and scheduled for completion during late fall of 1990, will remove tailings and
contaminated soils from the uppermost four miles of the bypass channel. The remaining
portion of the bypass channel (approximately one-half mile), to its confluence with Warm .
Springs Creek, will be cleaned up as part of the overall remedial actioi'f for Warm Springs
Ponds. All work required by the removal order is part of the overall remedy described.
herein and thus enforceable under this' Record of Decision.
Transport of Upstream Tailings to the Clark Fork River. The Warm Springs Ponds. are
27 river miles from Butte, where most of the mining-related activities occurred that led to
the contamination at the Warm Springs Ponds. Silver Bow Creek is contaminated along
most of those 27 miles, with several large deposits of tailings interspersed with many
smaller deposits. There are also much smaller deposits of tailings along Mill and Willow
creeks.
Altogether, some 3 million cubic yards of streamside tailings are estimated to exist
upstream of the Warm Springs Ponds.8 These tailings are eroded by normal and above
normal flows in the creeks; however, high flows move larger quantities of these tailings. A
recent flood study estimated that a 100-year flood on Silver Bow Creek would deliver
100,000 cubic yards (one football field 47 feet deep) of sediments to the Warm Springs
Ponds.9 These sediments would consist of both natural sediments and tailings.
8
Hydrometric., 1983. Summit and Deer Lod~e Valley. Lon~-Term Environmental Rehabilitation
Study. Butte-Anacon~a. Montana. Volume VII. Warm Sgrin~s Ponds. Prepared for the Anaconda Hineral.
Company, Butte, Hontana.
9
CB2H BILL, 1988. Silver Bow Creek Flood Hodelina Study. Prepared for State of Hontana
Department o~ Bealth and Environmental Science., Belena, Hontana.
-------
. Tailings Deposits and Contaminated Soils
In addition to the areas of tailings deposits around the pond system, there are soils that
contain varying concentrations of metals or are mixed with tailings. In addition, there are
. .
areas of pond bottom sediments that were historically submerged in Ponds 1 and 2, but
which are now exposed. The total area of tailings deposits and contaminated soils is esti-
mated to be approximately 420 acres with a corresponding volume of 1.9 ~illon cubic
yards. (See Table 1)
The primary pathways idelltified for potential human exposure to these contaminant~ are
direct (skin) contact, inhalation of dust from the surface, and incidental ingestion of con-
taminated soil and sediment. In addition, these contaminants may present environmental
threats, through adverse effects on fish and wildlife within the pond system.
Ground Water
Exceedences of primary maximum contaminant levels for cadmium and arsenic were
detected in one well located within Pond 1 and in several wells downgradient of Pond 1.
The affected wells downgradient of Pond 1 are completed in the shallow sand and gravel
aquifer. These exceedences could pose a threat to users of the aquifer, either currently or
in the future, and to aquatic organisms in the Clark Fork River.
The known area of primary maximum contaminant level exceedences in groundwater
caused by the pond system is in and below Pond 1 and is estimated to cover 180 acres.
There are likely two primary reasons why the area of contamination is not ~ore extensive.
Most significantly, the pond bottom sediments (tailings and sludges) form a low
permeability layer on the bottoms of the ponds, particularly in Ponds 2 and 3. Thus, the
contaminated water in the ponds and in the sediments does not readily leak into and
contaminate the groundwater to the degree that it otherwise would. Additionally, upward
gradients in the aquifer north of Pond 1, and the interception of the groundwater in tha
-------
area by the Clark Fork River, have kept the groundwater contamination in the area of the'
ponds from spreading very far north.
7.3
PROBLEMS UPSTREAM
In addition to the eight human health and environmental problems described above, the
relationships among those problems and the remainder of the' Silver Bow Creek/Butte
Area" NPL Site problems upstream of the Warm Springs Ponds are also important. Most
significantly, the upstream areas are the sources of the very larg~ volume of contaminated'
water flowing into and through the operable unit. 'The ponds, while not currently capable
>. '.
of providing totally adequate treatment of the contaminated flows in Mill, Willow, and
Silver Bow creeks, are nonetheless an important treatment system. They provide
significant protection of the Clark Fork River from the continuous flow of contamination
currently coming from upstream areas.
The levels of contamination in Mill, Willow, and Silver Bow creeks will likely be reduced
by future cleanup actions taken upstream of the Warm Springs Ponds. But, until that
time, the pond system will be needed to treat the flows and thereby improve the water
quality in the three creeks. This is an important factor in determining the types of alterna-
tives that can be developed for the Warm Springs Ponds Operable Unit. Alternatives that
would eliminate or substantially alter the existing pond treatment would have to include
alternative treatment capacity for the contaminated surface water if an equivalent level of
aquatic protection is desired.
In spite of the environmental problems, the Warm Springs Ponds have become a major
nesting and resting place for abundant waterfowl in the upper Clark Fork River. Brown
and rainbow trout also inhabit the. wildlife ponds and Ponds 2 and 3. The ponds are an
important sport fishing and hunting spot, attracting sportsmen from all parts of the United
States. Trout are caught frequently in the range of 8-12 pounds. These points are
noteworthy in light of the long term plans for improving the ponds' ability to support fish
-------
and wildlife. The selected remedy, which includes provisions for improving water quality,
increasing wetlands areas, eliminating exposed tailings, and improving the configuration of
the bypass channel, is not only a Superfund cleanup proposal but it is also a major fish and
wildlife habitat enhancement proposal.
-------
8.0 DESCRIPTION OF ALTERNATIVES
Objectives for remediation of the Warm Springs Ponds Operable Unit were identified as
part of the feasibility study. These objectives were develo'ped from the identification of
environmental and human health problems, utilizing ARARs and site-specific human
health and environmental protectiveness standards identified through the public health and
environmental assessment. The remedial action objectives are listed in Table 3."
Following the identification of the remediation objectives, potent.~al ~emedial technologies
~nd process options were identified ..and" evaluated for use at the site. All of. the
. .
" ".
technologies and process options were initially screened to eliminate those that were
unrelated to the problems at the site or that were technically infeasible for use at the. site.
The retained technologies and process options were evaluated a second time based on
effectiveness, implementability, and cost to further reduce the list of potential technologies. .
The technologies remaining following the second screening were combined to form media-
specific actions addressing the remedial objectives identified for each of the media. The
media-specific actions were developed to the conceptual design level in the feasibility
study.
Six comprehensive remedial action alternatives were assembled in the feasibility study by
combining one or more media-specific actions for each of tbe affected media into an
overall remediation package. The action alternatives were assembled from tbe 16 media-
specific actions developed in the feasibility study. In addition, a "no-action" alternative was
added to the range of alternatives and evaluated with the action alternatives as required"by
the National Contingency Plan. The seven alternatives developed in the feasibility study
for evaluation cover a range of possible combinations (Table 4). Also included in Table
for comparison is Alternative 3 + 3A, the selected remedy.
-------
TABLE 3
RELATIONSHIP OF SITE PROBLEMS TO REMEDIAL ACfION OBJECfIVES
Problems
Pond Bottom Sediments
Pond integrity--floods
Pond integrity--earthquakes
Surface Water
Fishkills
Metal loads in the flows of Mill, Willow,
and Silver Bow Creeks
Erosion of tailings in the Mill-Willow
Bypass into the Clark Fork River
Transport of tailings from upstream
reaches of Silver Bow Creek to the Clark
Fork River during floods and other high
flow events .
Tailings Deposits and Contaminated~
Human and environmental exposure to
surface contamination
Groundwater
Contaminated groundwater in the Pond 1
area
. Objectives
Prevent the release of the pond sediments
from design floods and earthquakes
.. Meet ambient water quality standards for
aquatic life at the identified compliance.
point.
Prevent ingestion above maximum
contaminant levels and established.
reference doses for copper, iron, zinc, and
cadmium. Also prevent ingestion of water
containing arsenic in concentrations that
would cause an excess cancer risk greater
than 10-4 to 10.7
Reduce the potential for tailings in the
Mill-Willow Bypass to reach the Clark
Fork River
Reduce the potential for tailings in
upstream areas of Silver Bow Creek to
reach the Clark Fork River
Reduce the potential for human exposure
to exposed tailings and other surface
contamination to satisfy acceptable intake
criteria .
Reduce the metals contamination in the
groundwater downgradient of the ponds
to achieve compliance with maximum
. contaminant levels
-------
TABLE 4
ASSEMBLED AllmNATlVES FOR WARM SPRINGS PONDS OPERABLE UNIT FEASlBIUTY STUDY
MEDIA POND BOTTOM SEDIMENTS SURFACE WAlm GROUNDWAlm SOIL
Sediments Sediments Erosion Transport of Contaminated Tailings Deposits &
!!! Floods Earthauakes Flshkills Dissolved Metals of Tallinas Upstream Tailinas Groundwater Contaminated Solis
Cease current Cease current Remove tailings Construct a new Remove tailings Construct an Install trench drains Excavate all
operation of the operation of the deposits and more effective deposits and upstream flood In and below Pond material above the
pond system and pond system and contaminated solis treatment pond contaminated solis Impoundment of 1. Pump to new action level; II
solidify sludges and solidify sludges and from bypass; from the bypass; 8,000 acre-feet treatment pond disposal In Pond 2
sediments In all sediments In aI dispose of in Pond dispose of In Pond Inlet for treatment or 3 prior to
three ponds three ponds' 1 and solidify 2 or 3, and solidify solidifying the
ponds
2 Protect ponds Protect ponds Remove tailings Comprehensive Remove tailings : Construct an Install trench drains Excavate all
against a PMF against an MCE deposits and upgrade of the deposits and upstream flood in and below pond exposed areas
contaminated solis Pond 3 treatment contaminated solis impoundment of 1. Pump to Pond 3 above the action
from the bypass; system from the bypass; 8,000 acre-feet Inlet for treatment. level;8 offslte
dispose of in an dispose of In an disposal; cap and
offslte TSDF. offsite TSDF revegetate Pond 1
and flood Pond 2.
3 Protect ponds Protect ponds Remove tailings Comprehensive Remove tailings Construct an Install trench drains Excavate all
against fractions of against an MCE deposits and upgrade of the deposits and . upstream settling In and below Pond exposed areas
. a PMF.b contaminated solis Pond 3 treatment contaminated solis basin of 2,000 acre- 1. Pump to Pond 3 above the action
from the bypass; system from the bypass; feet Inlet for treatment. level;1I dlsposalln
dispose of In Pond dispose offn Pond Pond 1; cap and
1 prior to capping. 1 prior to capping revegetated Pond
1. Flood Pond 2.
3+3A Protect an three Protect ponds Remove tailings Upgrade of the Remove tailings Raise the Pond 3 Install trench drains Cover and
ponds against a 0.5 against an MCE deposits and pond treatment deposits and berms so that It can In and below Pond revegetate Pond 1;
PMF contaminated ~olls system contaminated soils provide flood 1. Pump to Pond 3 flood tailings
from the bypass; from the bypass; detention and Inlet for treatment deposits above
dispose orin Pond dispose of In Pond settling up to Ii 10()- Pond 2.
1 or pond 3 prior to 1 or Pond 3 prior to year capacity Oood.
covering and covering and
revegetatlng. revegetatmg.
-------
TABLE 4 (conooued)
ASSEMBLED ALTERNATIVES FOR WARM SPRINGS PONDS OPERABLE UNIT FEASIBIlITY sruOy
MEDIA
POND BOTTOM SEDIMENTS
SURFACE WATER
SOIL
Ait
4
Sediments
~
Protect an ponds
against fractions of
a PMF.b
Comprehensive
upgrade of the
Pond 3 treatment
system.
Sediments
Earthquakes
Protect ponds
against an MCE.
Dissolved Metals
Erosion
of Tallinas
Remove tailings
deposits and
contaminated solis
from the bypass;
dispose of in Pond
1 prior to capping.
Flshkllls
Remove tailings
deposits and
contaminated solis
from the bypass;
dispose of In Pond
1 prior to capping.
Transport of
Upstream Tallinas
Construct an
upstream settling
basin of 2.000 acre-
feet. .
GROUNDWATER
Contaminated
Groundwater
Install trench drains
In and below Pond
1. Pump to Pond
3 Inlet for
treatment.
5 Protect ponds Protect ponds Remove tailings Partial upgrade of Remove tailings Construct an Install trench drains
against fractions of against an MCE. deposits and the treatment deposits and upstream settling below Pond 1; treat
a PMCF.b contaminated solis system; replace contaminated solis basin 01 2,000 acre- In on site wetlands.
from the bypass; fuse plug; Improve from the bypass; leet.
dispose of In Pond trashrack and lime dispose of In Pond
1 prior to capping. addition. 1 prior to capping.
6 Protect ponds Protect ponds Remove tailings Partial upgrade of Remove tailings No action. Install trench drains
against fractions of against an MCE. deposits and the treatment deposits and below Pond. 1 ; treat
a PMF.b contaminated solis system; replace contaminated solis In onslte wetlands..
from the bypass; fuse plug; improve from the bypass;
dispose of In Pond trashrack and lime dispose 01 In Pond
1 prior to capping. addition. 1 prior to capping.
7
No action.
No action.
No action.
No action.
No action.
No action.
No action.
Tailings Deposits &
Contaminated 80lls
Cap and revegelate
all material above
the action level.
Excavate tailings
deposits and
contaminated solis
In bypass; dispose
In Pond 1. Flood
Pond 2.
Cap an~ revegetate
all exposed areas
above the action
level. Excavate
tailings deposits
and contaminated
solis In bypass;
dispose In Pond 1.
Flood all applicable
areas;a excavate
tailings deposits
and contaminated
solis in bypassand
dispose In Pond 1;
cap and revegetate
Pond 1.
No action.
All areas containing contaminated solis and tailings for which the spec/lied technologyls appropriate or applicable. For example, excavation Is appropriate above Pond 3. below Pond I, and along the bypass;
it Is not applicable for the tailings In Ponds 1 and 2. Preliminary action levels are 250 ppm for arsenic and 750 ppm for lead. .
Alternative would protect Pond 3 to a 0.5 PMF, Pond 2 10 a 0.3 PMF, and Pond 1 to 0.2 PMF.
PMF. Probahle Maximum Flood.
MCr MaXImum Credible Earthquake
-------
" During the development of the feasibility study report, ARCQ developed its own proposed
plan (called Alternative 3A). It incorporated many of the features of the agencies'
Alternative 3, but was significantly different in tWo ~ajor respects~"" Because the ARCQ
alternative had certain useful features, and it was clear that "a combination of the features
of the agencies' Alternative 3 and ARCQ's Alternative 3A could be developed as an
effective alternative for remediation of the site, a cQmbined alternative, called Alternative
3+3A in this Record of Decision, has been developed. It is the selected remedy for the
operable unit. Table 4 lists the alternatives and describes the specific actions that e:;lch
includes. Each of the alternatives, including 3 + 3A, is described" ~eparately below.
8.1 ALTERNATIVE 1 ($1,191,500,000)
The components of Alternative 1 include solidifying all onsite contaminated soils, tailings,
sediments, and sludges to protect against a probable maximum flood (PMF) and a maxi- "
mum credible earthquake (MCE); constructing a new treatment pond for surface water
treatment and an upstream flood impoundment to capture flood flows for additional
treatment; and installing a groundwater interception trench to capture and then treat
contaminated groundwater as it migrates from the ponds.
The current inability of the three existing ponds to withstand floods and earthquakes
would be addressed by using an in situ solidification process to stabilize the pond bottom
sludges and sediments. This would minimize the risk of pond failure due to an earthquake
or flood event. In addition, contaminated soils and exposed tailings that exceed an action
level of 250 ppm for arsenic and 750 ppm for lead would be excavated and disposed of in
the existing ponds prior to solidification.
This alternative would effectively limit the toxicity and mobility of tailings to acceptable
concentration levels and greatly reduce the potential for future human or animal contact
with harmful contaminants.
-------
Alternative 1 would also improve surface water quality with the construction of a new
p'ond treatment system: A new treatment pond would be. constructed to replace the
existing, now solidified, pond system. The new pond would be capable of capturing and
treating flows up to 600 d's. This is the flow the current pond system is capable of
treating.
. .
In addition, an upstream flood impoundment (8,000 acre-feet) would be co~tructed to
provide settling and treatment of flows on Silver Bow Creek up. to the volume of a 100-
year flood or the maximum flow rate of th~. maximum flood of 4.000 cfs. Currently, flood.. '.
flows on Silver Bow Creek that exceed 600 cfs (the design limit of the Pond 3' "iI1let
structure) are routed around the ponds, untreated. A flow of 600 cfs on Silver Bow Creek
represents a 2- to 3-year return flood.
The goal of the upstream impoundment is to' prevent large quantities of sediments and
dissolved metals from bypassing the pond system and flowing into the Clark Fork River.
The impoundment would serve two functions. First, it would serve as a. conventional
sedimentation basin; as the influent velocity. slowed in the impoundment, the sediment
being transported by the flow would settle out. Second, the impoundment would have the
storage capacity to contain flows up to the 100-year flood. The water cC?uld then be
metered to the ponds for treatment of dissolved metals. Floods exceeding 4,000 cfs would
be routed around the impoundment to protect it from damage caused by scouring.
Contaminated groundwater moving from the operable unit would be collected from in and
below Pond 1 through the installation of an open groundwater trench. The collected
groundwater would t~en be pumped to the inlet of the new pond for treatment. This
would reduce the discharge of contaminated groundwater into the Clark Fork River and
enable the aquifer to be used for drinking water and other beneficial uses.
Alternative 1 is one of two alternatives expected to exceed at least one ARAR. Wherea~
Montana's dam safety standards require protection of the existing Ponds 1,2, and 3 to G.:!.
-------
.0.3, and 0.5 PMF, respectively, the in situ stabilization process would provide protection of
all three ponds against the full PMF. Alternative 1 is expected to meet all other ARARS
with one exception; surface water standards for arsenic and mercury for protection of
public health from ingestion of contaminated water and fish are technically impracticable
to meet using. this or any other remedial alternative.
The actions proposed in Alternative .1, however, would have a substantial advers~ affect on.
existing wetlands. Over 1,200 acres of wetlands and open habitat for birds, fish, and
mammals would be destroyed.
A potential adverse affect on an identified cultural resource within the area also exists. A
concrete arch bridge located within the dry portion of Pond 2 has been determined to be
eligible for inclusion in the National Register of Historic Places. Consultation with the
State Historic Preservation Office would be necessary to minimize potential impacts to the
bridge prior to commencing any remediation activities. Consultation with that office will
be necessary with the remaining alternatives as well.
Certain institutional controls would be required for Alternative 1 and all the other
remedial alternatives, as well. Institutional controls are generally defmed as legal
mechanisms that prevent or limit human access and exposure to the contamination and are
used to enhance the effectiveness of a given remedial alternative. Upon solidification and
closure of the ponqs, the local zoning or land use authority and the EP A Regional
Administrator must be notified. of the type, location, and quantity of waste disposed of in
each pond. A notation or deed to the facility property must be recorded in accordance
with State law to notify any potential purchaser that the land has been used to manage
hazardous waste. Finally, the prohibition against consumption of any fish caught within
the pond system must be continued.
With the appropriate design, ~onstruction, and maintenance, Alternative 1. should reliably
red~ce hU,man health and environmental risks. Because of the enormous volume of pond
-------
'sludges (19 million cubic yards), Alternative 1 would take approximately 17 years to'
complete. Full risk reduction would not occur until that point. The estimated present
worth cost for this alternative is $1,191,500,000. This present worth cost includes both
capital costs and annual operations and maintenance costs. All future costs are reduced to
present worth costs to allow remedial action alternatives to be compared on a relatively
equivalent basis.
8.2 ~TERNATIVE 2 ($241,500,000)
"
Alternative 2 is the most comprehenSive of the alternatives that retain the current 'pond
treatment system. Its components include protecting the pond system against a probable
maximum flood and the maximum credible earthquake; excavating and disposing offsite all,
contaminated soils and tailings within the Mill~ Willow Bypass, Pond 3, and below Pond 1;
capping Pond 1; flooding exposed tailings and contaminated soils within Pond 2; and
upgrading the treatment system in' Pond 3. It also includes two of the components of
Alternative 1: constructing an upstream flood impoundment and installing groundwate
interception trenches.
Pond stability would be achieved by protecting all three ponds against both a full probable
maximum flood (PMF) and maximum credible earthquake (MCE). Thus, maximum
protection is provided against release of the pond bottom sediments. While some damage
to the pond berms could still occur under extreme conditions, there would be minimal risk
of losing the pond bottom sediments during an earthquake or flood event.
All exposed tailings and contaminated soils along the Mill-Willow Bypass, and all exposed
tailings and contaminated soils within Pond 3 and below Pond 1 that exceed an action
level of 250 ppm arsenic and 750 ppm lead, would be removed and disposed of at an
offsite RCRA disposal facility. The closest treatment, storage, and disposal facility able to'
accept the waste is near Boise, Idaho, approximately 480 miles from the site. Exposed
tailings and contiuninated soils within Pond 2 would be flooded, and Pond 1 would be
-------
. 3~
5.
6.
7.
.capped with a RCRA-compliant cap. All excavated areas and Pond 1 would subsequently
be revegetated.
..
Alternative 2 would improve surface water quality by completely upgrading the current
pond treatment system. The improvements would include the following seven measures:
1.
Diverting Mill and Willow Cr.eeks into Pond 3 for treatment
2.
Modifying the inlet structure to Pond 3 by adding a trash..rack and overflow weir, .
and relocating the fuse plug
Channelizing Silver Bow Creek within the dry areas of Pond 3 to minimize. the
interaction of Silver Bow Creek with exposed tailings and controlling the direction
of flow
4.
Improving the lime addition system to enhance metals precipitation
Adding a berm across Pond 3 to help prevent short-circuiting of flow and thereby
increase settlement of solids
Constructing a new effluent structure in Pond 3 to minimize scouring and
resuspension of pond sediments
Wet closing Pond 2 since the sludge storage capacity of the pond has been
exhausted
As in Alternative 1, an upstream flood impoundment (8,000 acre-feet) would be
. constructed to provide settling and treatment of flows on Silver Bow Creek up to the 100-
year flood (4,000 cfs).
-------
.' Contaminated groundwater would be collected through interception trenches below both'
Pond 1 and Pond 2 berms. The groundwater would then be pumped to the inlet of Pond 3
for treatment.
Alternative 2 is one of two alternatives expected to exceed at least one ARAR. Whereas
Montana's dam safety standards require protection of the existing Ponds 1, 2, and 3 to 0.2,
0.3, and 0.5 PMF, respectively, Alternative 2 stabilizes all pond berms against aOfull PMF.
This ,alternative is expected to attain aquatic water quality standards for surface water
. . .
(except for arsenic and mercury, as desc~bed in Alternative 1), IJlaximum contaminant.
levets for groundwater, and selected RCRA closure requirements for Pond 1.
All of the components of Alternative 2 should reliably reduce the human health and
environmental risks at the site, if properly designed, operated, and maintained. The
actions proposed may result in adverse effects to wetlands, endangered species, or
historical resources. The estimated present worth cost for this alternative is $241,500,000.
It is estimated that the remediation measures will take 5 years to complete.
8.3 ALTERNATIVE 3 ($71,100,000)
Alternative 3, identified by MDHES and EP A in the feasibility study and the proposed
plan as the preferred alternative, is similar to Alternative 2 in that is includes protecting
the ponds against an maximum credible . earthquake completely upgrading the pond
treatment system, capping Pond 1 and flooding Pond 2, and installing ground water
interception trenches. It is different from Alternative 2 in that it requires protection of the
ponds to a fraction of the probable maximum flood instead of the full probable maximum
flood; it includes excavation of exposed tailings and contaminated soils with subsequent
disposal. in Pond 1 instead of offsite disposal; and it includes the smaller upstream settling
basin in lieu of a large upstream impoundment. Only the new components are discussed
below.
-------
. Pond stability in this alternative is achieved by protecting Pond 1 against a 0.2 PMF, Pond
2 against a 0.3 PMF, and Pond 3 against a 0.5 PMF. These are the standards that are
required by Montana's dam safety regulations for high hazard dams such as those at the
Warm Spring Ponds.
In Alternative 3, all exposed tailings and contaminated soils in the Mill- Willow Bypass,
within Pond 3, and below Pond lthat exceed an action level of 250 ppm az:senic and
750 ppm lead would be excavated and disposed of in Pond 1. Pond"l would be closed
with a RCRA-compliant cap as described in Alternative 1.
Consolidating excavated material into Pond 1 under a RCRA-compliant cap would
effectively isolate the material from direct contact and effectively limit the mobility of the
material. It would also effectively consolidate all material which exceeds the cleanup
criteria within a smaller area. As long as the cap is properly maintained, the material"
would be safe from release because of erosion of the cap.
. '.
The final difference between Alternatives 2 and 3 is that Alternative 3 includes the
construction of a smaller upstream settling basin (2,000 acre-feet). During flood flows on
Silver Bow Creek greater than 600 cfs, surface water would pass through the upstream
settling basin. The settling basin would be similar to the upstream impoundment with two
exceptions. First, the storage capacity. would be much lower (2,000 acre-feet versus
8,000 acre-feet). Second, the amount of water that would receive full treatment for both
suspended solids and dissolved metals would be less.
During flood flows between 600 and 4,000 cfs, all surface water from Silver Bow Creek
would pass through the upstream settling basin. Full treatment would be provided for
floods that do not completely fill and then overflow the 2,000 acre-foot settling basin.
Suspended solids would settle within the basin and the captured water would then be
released slowly from the basin for treatment of dissolved metals in Pond 3. Floods tha t
exceed the storing capacity of the settling basin, however, would be only partially treated.
-------
Up to 80 percent of the suspended solids would continue to be settled out within the
basin, but only flows up to 600 cfs (the inlet capacity of Pond 3) would then be treated in
the ponds for dissolved metals. The remainder of.the flows discharg~d over the spillway of
the settling basin would be routed around Pond 3 and flow down the bypass without
treatment of dissolved metals.
The actions proposed in Alternativ~ 3 are expected to result in compliance wit~ all State
and federal ARARs. These include Montana's dam safety standards, aquatic water quality
standards (with the exception of arsenic an~ mercury, as previously described), maximum..
contaminant levels, and sele.cted RCRA closure requirements.
The actions proposed for Alternative 3 are technically feasible and are expected to reliably
reduce the environmental and human health risks at the site. The actions proposed may
result in adverse effects to wetlands, endangered species, or historical resources. The
estimated present worth cost is $71,100,000. It is estimated that the remediation measure
. identified will take 5 years to complete. .
f'
8.4 ALTERNATIVE 3+3A $(57,416,000)
Alternative 3 + 3A, identified by the EP A and MDHES as the selected remedy, is a
synthesis of Alternative 3 and ARCQ's Alternative 3A Alternative 3 + 3A was developed
following consultation with the public and ARCa to address concerns about some of the
aspects of Alternative 3 as presented in the feasibility study. Alternative 3 + 3A includes
many of the features of Alternative 3, including protecting. the pond berms against the
maximum credible earthquake and fractions of the probable maximum flood, upgrading
the treatment system, removing Mill-Willow tailings, covering and revegetating Pond 1. .
and ins'taIling ground water interception trenches. It is different from Alternative 3 in that
storage of flood flows would be within Pond 3 rather than in an upstream impoundment:
the bypass channel would be realigned in places; Pond 2 would be improved and retained
-------
as a treatment unit; and disposal of contaminated soils would be within the dry areas of
either Ponds 1 or 3. The primary features of Alternative 3 + 3A are discussed below.
"
Pond stability would be achieved by altering all pond berms so that they would be stable
during the maximum credible earthquake. This would be accomplished by flattening the
downstream slopes or adding toe berms for stability. Additionally, the upstream faces of
the berms would be analyzed during the remedial design phase to insure their stability ,
during the maximum credible earthquake. All north-south berms along the Mill-Willow
Bypass would be raised and strengthened to ,protect against failure ,duling flood flows up. to "
70,000 cfs, which is one-half the peak flow rate of a probable maximum flood. The slopes
of the berms along the bypass would be protected against scour by constructing soil-cement
armoring for the entire length of the bypass.
"
,
The tailings and contaminated soils along the Mill-Willow Bypass would be excavated and
disposed of at two locations: within Pond 1 prior to covering and within a dry area of
Pond 3 near the Pond 3 berm (see Figure 1). This excavation and disposal was begun
during the summer of 1990 as part of the Removal Action. The remainder of the
excavation and disposal will be performed as part of the Remedial Action covered by this
Record of Decision. The disposal areas in Pond 1 and in Pond 3 will ultimately be
covered with lime and soil barriers, then revegetated with native species. The amount of
contained materials to be disposed of at each location will be determined based upon the
economics of haul distances.
The measures to upgrade Ponds 2 and 3 for this alternative would serve two primary
purposes: 1) storage of flood flows up to the 100-year event and 2) improvement of the
treatment processes to achieve the water quality standards at the point of discharge. The
main features include:
. Raise Pond 2 and 3 embankments to increase storage capacities within those
ponds and enable storage and treatment of the 100-year flood event in Pond 3.
-------
The total storage capacity of Pond 3 would be increased to 13,000 acre-feet.
The operating volume of Pond 2 would be increased to 2,200 acre-feet to
increase retention time and improve treatment.
. Modify and replace hydraulic structures. The intake structure to Pond 3 would
be completely replaced with a larger, more efficient structure capable of passing
flows up to 3,300 cfs (the estimated peak flow of the 100-year flood event).
Flows exceeding that amount would be routed to .the Mill-Willow Bypass
Channel using a combination of ~n overflow spillway and .a fuse plug dike. The. .
intake structure would be de.signed to minimize plugging through use of a. tr~sh
rack. At the maximum water surface elevations anticipated during a major
flood, the intake structure would be capable of passing no more than 4,000 cfs
into the ponds.
The two decant outlets on Pond 3 would be raised and modified to provide
controlled releases into Pond 2, not to exceed 200 cfs. Additional outflows are.r.;
required to avoid exceeding the allowable storage volume in Pond 3 during the
100-year flood. Outflows in excess of 200 cfs would be routed directly into the
Mill-Willow Bypass channel via a pipe from the west decant tower. The out~et
pipe to the bypass would be capable of discharges up to 500 cfs. The discharge
to the bypass would be through an energy-dissipation structure to avoid excessive
erosion. The outlet structure in Pond 2 would be raised and modified to
accommodate the water level increase.
. Construct emergency spillways in the Pond 2 and Pond 3 berms. In Pond 2, the
spillway would be designed to allow passing up to 12,500 cfs from a flood in the
eastern hills, which is one-half the probable maximum flood of that drainage
area above Pond 2. In Pond 3, the entire volume expected during a flood of
one-half the probable maximum flood, from the eastern hills, can be contained
within the upgraded storage capacity of Pond 3. However, as noted above, the
-------
inlet structure to Pond 3 can pass as much as 4,000 cis during a major flood in
Silver Bow Creek. Thus, the spillway in Pond 3 must be capable of passing 4,000
cfs directly into the bypass channel to avoid overtopping the berms during a
major flood in Silver Bow Creek. The emergency spillways would be constructed
in the western embankments of Ponds 2 and 3 and would be constructed using
soil-cement similar to the soil-cement used to armor the embankments slopes.
-" . Upgrade lime treatment facilities and water quality - controls. A new lime
addition facility would be install~_d at the intake structur.e. to Pond 3. The ~ew. .
facility would add hydrated -lime to the Silver Bow Creek ~uent at a>rate
sufficient to raise and maintain pH levels at a minimum of 9.0. The treatment
facility would be designed to handle both normal floWs and flood flows up to the
lOO-year event. Pond 3 would provide sufficient retention time to allow metals
to react and form insoluble hydroxide precipitates. Pond 2 would provide
greater volume and retention time for final settling and clarifying of the Pond 3
effluent before discharging.
The contaminated ground water would be addressed using the same facilities as described
for Alternatives 2 and 3. The ground water would be collected in interception drains
below and within or adjacent to Pond 1. The ground water would then be pumped back to
the inlet of Pond 3 for treatment.
Both surface and ground water quality monitoring would be needed. The existing ponded
water in the eastern portion of Pond 1 would be pumped out and Pond 1 would then be
dry-closed. The tailings and contaminated soils in Pond 1 would be protected from direct
exposure by covering with lime and soil barriers, followed by revegetation with native
species. The dry-closed Pond 1 would be protected from floods .in the eastern hills
through construction of a channel around the east side of the pond, discharging below
Pond 1 to the Clark Fork River. The channel would be designed to safely pass a flood
-------
. from the eastern hills of 8,500 cis, which is one~half the p'robable maximum flood of that
drainage area above Pond 1.
The exposed tailings above Pond 2 would be flooded as a result of the increased water
elevation and volume of this pond. The exposed tailings deposits above Pond 3 will not be
addressed as part of this alternative. The area above Pond 3 will be part of the active
receiving pond, with floods up to the 100-year flood being rou~ed into the pond. During
these events, additional tailings and sediments will be deposite(j in Pond 3. Hence, the
removal or capping of the exposed tailing~ in this area will be. addressed at the 5-year.
review and at the time of the final closure of the pond system.
Institutional controls to prevent future residential development would be implemented.
Deed notices and recording the locations of Ponds 1, 2 and 3 and all disposal areas would
be required. Specific contractual provisions with the State may be required. Further
development of Deer Lodge County's zoning scheme will be required. Institutional
controls to prevent s :vimming and consumption of fish by humans is necessary. All othe
activities needed to comply with the final ARARs, Attachment to Part 2 would also be
required.
The actions proposed in Alternative 3 + 3A are expected to result in compliance with state
and federal ARARs. These include Montana's dam safety standards, aquatic water quality
standards (with the exception of the standards for arsenic and mercury, which will be
waived as previously described), and maximum contaminant levels.
The actions proposed for Alternative 3 + 3A are technically feasible and are expected to
reliably reduce the environmental and human health risks at the site. The actions
proposed may result in adverse effects to. wetlands, endangered species, or historical
resources. The estimated present worth cost is $ 57,416,000. It is estimated that
remediation will take 3 to 5 years to complete.
-------
8.4.1 COMPONENT UPGRADE.
A component upgrade of the treatment system in Pond 3 may be necessary in the event
that the remedies proposed in Alternative 3 + 3A for handling flood flows are not as
effective as currently anticipated. The purpose of the component upgrade would be to
address the potential for resuspension of bottom sediments in Pond 3.
. . .
The pond bottom sediments of concern are the very fine grained .~ettled materials that are
'. ..
essentially composed of flows and sludges resulting from the existing (and proposed)
treatment processes. These flows and sludges exist as a sludge blanket on the bottom of
Pond 3. They may be subject to resuspension during high. winds or high flows. . The
amount and the effects of resuspension cannot be determined using existing modeling
techniques.
Accordingly, tests will be performed on the pond bottom sediments to determine their
impact on aquatic life. The tests would be performed in two phases:
. Phase 1 will include a bioassay survey of the pond bottom sediments. A model
will be constructed assuming various levels of resuspension of these materials.
Waters containing these levels of resuspended materials will then be used in a
series of bioassays. Standard EP A-approved test species of biota (including fish
and macroinv~rtebrates), or native biota if possible, will be subjected to acute
. and chronic bioassays using waters containing the materials to determine the
effects on their ability to survive. In conjunction with the bioassays, a full
spectrum of chemical analyses will be performed on the waters contaimng the
resuspended materials. The bioassay survey will be completed prior to
September 30, 1991.
-------
. Phase 2 will be performed only if bioassay results indicate that there are advers
affects on the biota as a result of pond bottom material releases. Phase 2 would
incorporate field scale resuspension testing using in situ techniques to determine
the parameters necessary to develop resuspension modeling. Once these
parameters have been defined, the Pond 3 system would be modeled to
determine the extent of resuspension during high flows or high winds. . .
<-
If the Phase 1 or Phase 2 investigations indicate that resuspension of pond bottom
sediments will result in adverse effects to..human. health or the. enVironment, addi~~onal .
meaSures would be. required as part of the selected remedy. These measures would
include:
. A separate study amendment to identify and analyze additional remedial
measures to address the resuspension of pond bottom sediments.
. Construction of the selected additional remedial measures.
8.5 ALTERNATIVE 4 ($77,000,000)
Alternative 4 contains many of the same components as Alternative 3. These include
protecting the pond system against a full maximum credible earthquake and a fraction of
the probable maximum flood, capping. Pond 1, completely upgrading -the pond treatment
system, constructing an upstream settling basin, and installing ground water interception
trenches. The only difference between this alternative and Alternative 3 is that this
alternative provides for capping exposed tailings and contaminated soils in place instead of
excavating and consolidating them in Pond 1 prior to capping.
In Alternative 4, the only areas of exposed tailings and contaminated soils that would n«)!
be . capped in place would be those along the Mill-willow Bypass and within Pond ...
-------
"Material from the Bypass would be excavated and placed into Pond 1 prior to capping.
The areas of exposed tailings and decontaminated soils within Pond 2 would be flooded.
All other areas that exceed an action level of 250 ppm for arsenic and 750 ppm for lead
would be capped in place. The capping would involve covering these areas with a 6-inch
layer of agricultural lime to help prevent metals migration and then covering the area with
18 inches of topsoil. Capping the contaminated soils and exposed tailings in place with an
18-inch cap would effectively reduce the mobility of the material but would ,not be as
effective or permanent in containing the wastes and minimizing the exPosures as removal,
consolidation, and placement under a RC~-compliant cap as specified in Alternative 3. , ,
, .
Fertilizer, soil amendments, and seed would be spread as necessary over the are'a, to
establish stable vegetative cover in accordance with State reclamation requirements.
The actions proposed in Alternative 4 are expected to result in compliance with all state
and federal ARARs. These include Montana DNRC dam safety requirements, aquatic
water quality standards (with the exception of arsenic and mercury, as previously
described), maximum contaminant levels. RCRA compliant closure requirements (Pond
1), and State reclamation standard (exposed tailings and contaminated soils).
All of the components of Alternative 4 are technically feasible, and with appropriate
design, construction, operation and maintenance, would reliably reduce the human health
and environmental risks at the site. The actions proposed in Alternative 4 may have an
adverse effect on wetlands, endangered species, or historical resources. It is estimated that
implementation of this alternative will take 5 years at a total present worth cost of
$77,000,000.
8.6 ALTERNATIVE 5 ($66,3000,000)
Alternative 5 is similar to Alternative 4 in all aspects except two. First, Alternative 5
includes a partial upgrade to the treatment system instead of the complete upgrade
-------
.provided in Alternatives 2, 3, and 4. Second, Alternative 5 provides for treatment 'of .
contaminated ground water in an onsite wetland treatment system instead of in Pond 3.
. .
The partial upgrade of the pond treatment system would include the following four
measures:
1.
Diverting Mill and Willow Creeks into Pond 3 for treatment
2.
Modifying the inlet to Pond 3 by 3:~ding a trash rack and an overflow weir and. .
relocating the fuse plug
3.
Improving the lime treatment system
4.
Retaining the existing effluent structures in Pond 3 and keeping Pond 2 in service
This less comprehensive upgrade to the pond system would provide some improved
treatment to surface waters, but not to the extent necessary to effectively treat flows up to .
600 cfs as provided in Alternatives 2, 3, and 4. Consistent treatment would be provided
for flows only up to approximately 210 cfs. This flow rate is based upon calculations that
determine the maximum flow rate that could be directed from Pond 3 to Pond 2 while still
providing acceptable metals removal in Pond 2 and preventing the resuspension of pond
bottom sediments. Since the effluent structure that directs the flow from Pond 3 to Pond.
2 will not be modified, Pond 2 remains as an active treatment cell in the pond system and
becomes a limiting factor with regard to the volume of flow that can be treated in the
ponds.
Because of the limited capacity of Pond 2, flows greater than 210 cfs would be directed
around the pond system, without treatment for dissolved metals. This will result in
violations of aquatic water quality standards during above average flows. (The average
flow of surface waters through the operable unit is approximately 90 cfs). Also, because
-------
. the effective treatment capacity of Pond 2 is nearly exhausted due to the volume to
sediments accumulated in the Pond, keeping Pond 2 in the treatment system provides an
~pportunity for sediments to be resuspended during periods of high winds. The future life
of Pond 2, and therefore the future life of the treatment system, would be limited to an
estimated 15 years.
The ground water contamination problem would be addressed by constructing ~ wetlands.
treatment system below Pond 1. Contaminated ground water wou~d .be collected by an
open ground water trench and pumped up to the entrance of the wetlands for treatment.. .
The. area available for the establishment of a wetlands treatment system is approximately
100 acres.
Two separate treatment cells would be constructed within the wetlands to enhance the
metals removal efficiencies. The cells would operate in series, with effluent water from
the first cell discharging. into the second cell. Treated water from the second cell would be
discharged to the Clark Fork River.
Due to plant uptake of toxic metals and vegetation die-off, periodic removal of organic
matter from the wetlands area would be necessary. However, with periodic cleaning and
proper maintenance, the wetlands could be expected to remain viable for at least the life
of the treatment system. Treating contaminated ground water in an onsite wetlands should
result in compliance with groundwater standards. However, wetlands treatment is not.
expected to be as consistently reliable as the pond treatment system proposed in
Alternatives 2, 3, and 4.
Alternative 5 is expected to result in compliance with most but not all ARARs. By
providing only a partial upgrade to the pond treatment system, exceedences of aquatic
water quality standards can be expected. Compliance with maximum contaminant levels
for ground water is expected to be achieved, but not with the consistency expected with
-------
,alternatives 1 through 4. Certain RCRA closure requirements for Pond 1 and reclamation
standards are expected to be achieved.
The actions proposed in Alternative 5 may have an adverse effect on wetlands, endangered
species, or cultural resources. It is estimated that this alternative will take 5 years to
implement at a present worth cost of $66,3000,000.
8.7 ALTER~ATIVE 6 ($55,100,000)
The components of Alternative 6 are a' combination of many of the components fou'rid, in
Alternatives 1 through 5. Alternative 6 includes protecting the pond system against a full
maximum credible earthquake and a fraction of the probable 'maximum flood, excavating
tailings and decontaminated soils within the Mill. Willow Bypass and disposing of them in
Pond 1, partially upgrading the pond treatment system, and collecting and treating
contaminated ground water in an onsite wetlands treatment system.
The unique features of Alternative 6 are that: 1) it does not include' the installation of an
upstream impoundment or settling basin; and 2) the action proposed for isolating the
contaminated soils and tailings within the site includes flooding wherever pc;>ssible. Only
the two unique features are discussed below.
This alternative does not address the transportation of contaminated soils and tailings from
upstream sources expect for flows less than 210 cfs. As discussed in Alternative 5, this'
flow rate is a limitation of Pond 2 and the partial upgrade of the pond treatment system.
Therefore, flows greater than 210 cfs on Silver Bow Creek would bypass the pond system
since no upstream impoundment or settling basin would be present to detain larger flows
and thus enhance settlement of solids and treatment of metals. In addition, over the long
'term, deposition of upstream sources may lead to recontamination of the Mill. Willow
Bypass.
-------
. In Alternative 6, exposed tailings and contaminated soils below Pond 1 and within Ponds
2 and 3 would be isolated by flooding the areas and maintaining a constant water level.
The flooding of tailings and contaminated soils below Pond 1 would be accomplished
through the construction of the wetlands treatment system.
The exposed tailings and contaminated soils within Pond 2 would be flooded. A small
berm would be designed to cross p'ond 2, running east to west in order to fa~ilitate the
flooding of the higher southern end of the pond. A small amount' of water would be
discharged from Pond 3 to Pond 2 in order to keep the newly bermeci area wet. Discharge
" .
from Pond 2 would flow di~ectly into the Mill-Willow Bypass.
The actions proposed for Alternative 6 should result in compliance for most but not all
ARARs identified. Because only a partial upgrade to the pond treatment system will be
realized, and an upstream impoundment or settling basin will not be constructed,
compliance with aquatic water quality standards will only be met at flows less than 210 cfs
on silver Bow Creek. Compliance with maximum contaminant levels for ground water is
expected to be achieved but not with the consistency expected with Alternatives 1 through
4. Certain RCRA closure standards and State reclamation standards are expected to be
met.
The actions proposed in Alternative 6 may result in adverse effects to wetlands,
endangered species, or cultural resources at the site. It is estimated that Alternative 6 will
take 5 years to implement at a total present worth cost of $55,100,000.
8.8 ALTERNATIVE 7
Alternative 7 is the no-action alternative required by the National Contingency Plan. It is
used as a baseline alternative against which to judge the other alternatives. As the name
implies, this alternative does not include any remediation activities. Current activities at
the site being carried out by the agencies would diminish substantially. The only activities
-------
,assumed to continue are those being carried out by the owner (e.g., lime addition to the
influent during winter months and general maintenance of the site). Accordingly, there
would be no reduction in risk or increase in protectiveness of human health and the
environment. As a result of the continued occurrence of a number of processes onsite, the
risks to human health and the environment would increase over time if left unmitigated.
Major fishkills will continue to occur on a periodic basis. Catastrophic failure of the ponds
could occur in a moderate earthquake or a moderate flood (probably less than ,a tOO-year
event).
-------
9.0. COMPARATIVE ANALYSIS OF ALTERNATIVES
The alternatives described in the previous section, with the exception of Alternative 3 + 3A,
were based on certain standards and criteria which have since been reevaluated. Those
standards are the probable maximum flood standard for pond berms (now 0.5 for all
berms), the RCRA standards for an impermeable cap (no longer required), and the action
levels' for lead and arsenic deferred and substituted by other criteria. These changes
require only minor adjustments in the actual cleanup actions. "In performing the
comparative analysis of their various a~~ernatives,. EPA and. MDHES assessed the. .
alternatives with the revisions. as described. The comparative analysis which follows
assumes these changes would be incorporated into the. alternatives. Cost estimates given
would not change significantly due to these changes.
CERCLA requires EP A to examine several factors when selecting a remedy. EP A has
identified nine evaluation criteria to be examined. 40 CFR ~ 300.515(e)(9)(iii): ~
300.515(f)(1)(i).
Two of the criteria are threshhold criteria--the remedy must be protective of human health
and the environment and must comply or result in compliance with ~s, unless a
specific ARAR is waived.
Five of the criteria are primary balancing, criteria--Iong-term effectiveness and
permanence; reduction of toxicity, mobility, or volume through treatment; short-term
effectiveness; implementability; and cost.
The two remaining criteria are modifying criteria--state and community acceptance.
This section of the ROD analyzes the various alternatives against each of these criteria
and weighs the advantages and disadvantages of each alternative relative to the oth~ r
alternatives.
-------
The evaluation is presented using the nine evaluation criteria as headings. Under each
heading the alternatives are discussed approximately in order of decreasing performance
for that criterion.
9.1 OVERALL PROTECI10N OF HUMAN HEALTH AND ENVIRONMENT
As previously explained, two elements of the operable unit cleanup are deferred to a later
date, "as permitted by 40 CPR ~ 300.515(f)(5)(iii)(D). Final action levels for contaminants
in soils, sediments, and tailings will be se!ectedwithin one year." This will determine"
additional cleanup requirements, espe'Cially for the area below Pond 1. Final cJeanup
, ,
decisions for the final disposition of Ponds 2 and 3, after the need for treatment of the
incoming water is no longer necessary, cannot be made until after the effectiveness of
upstream cleanup actions is known. Therefore, this analysis addresses protectiveness
within the scope of this interim -remedy only, and does not address the deferred actions
described above.
Each of the first six alternatives (including Alternative 3 and 3A) addresses the eight
human health and enVironmental concerns identified at the site to varying degrees of,
protectiveness. Alternative 6 leaves one of the concerns unaddressed--the transport of
tailings down the bypass during flood flows in excess of 210 cfs. Other alternatives do not
, consider containment and treatment of the 100-year flood flows. Alternative 7 is the no-
action alternative; it would not alter the site, and it does not address any of the identified
concerns.
Overall, Alternative 1 is the most protective of the alternatives, because it alone ,contains
measures to treat the pond bottom sediments, tailings deposits, and contaminated soils to
permanently reduce their mobility. However, because in situ solidification is still a
developing ,technology" its feasibility would have to be further explored during the
pre design or design stage. It would destroy important fish and wildlife habitat and
necessitate treatment ponds elsewhere in the Clark .Fork River Basin. The remaining
-------
-- action alternatives (Alternatives 2-6) include measures to stabilize the pond berms to limit -
the mobility of the sediments and sludges by improving their existing containment.
Alternatives 2 through 6 also include measures to contain the tailip.gs and contaminated
soils exposed at the surface throughout the operable unit either onsite (Alternatives 3-6) or
offsite (Alternative 2).
Of the alternatives that do not eJiminate the existing pond system (Altern(ltives 2-7),
Alternative 2 offers the most protection against pond failure by protecting the pond berms
. against the maximum expectable forces-~_a probable maximum f100d or a maximum - -
credible earthquake. Al~ernatives 3-- through 6 protect the pond berms against - - an
maximum credible earthquake and fractions of. a probable maximum flood. Of these,
Alternative 3 + 3A would be the most protective because it would upgrade all the ponds to
withstand a 0.5 PMF, whereas the other alternatives would upgrade Ponds 1, 2 and 3 to
withstand a 0.2, 0.3, and 0.5 PMF, respectively. In addition, Alternative 3+3A is most
effective in preserving and enhancing wetlands, and containing the 100-year flood without
creating new contaminated areas.
~
The no-action alternative would be the least protective by leaving the berms in their
present unstable state.
Containment measures for tailings deposits and contaminated soils are proposed for
Alternatives 2 through 6. In general, the containment measures are not expected to be as
protective or as permanent as the solidification action proposed for Alternative 1. This is
because the tailings and soils would still exist in a form that could, in theory, be disturbed
by severe weather or other forces, though the probability of dispersal of the contaminants
would be very low for most of these alternatives. Alternative 2 includes offsite disposal of
the contaminated material. This would remove the direct threat from the site, but it
would also introduce new risks and the liability associated with the offsite disposal of
untreated material. Alternative 3 would consolidate the material onsite under a RCRA-
compliant cap. This could effectively contain the material without introducing the
-------
. .additional liabilities and risks of offsite disposal. Alternatives 3 + 3A would cap'
contaminants within Pond 1 and dry portions of Pond 3; it would also flood the tailings
deposits and contaminated soils in Pond 2. All of the contaminated. materials would be in
flood protected areas. The exposed tailings within Pond 3 would not be remediated at the
present time because this will be an active area of the pond. Floods up to the 100-year
flood will be routed into Pond 3 resulting in occasional flooding of some of these tailings.
. .
Alternatives 4 and 5 would cap in place all of the materi3.I possible. Less protective caps
would be used, and the lack of consolidation would increase maintenance costs and the
potential for cap failure. Alternative 6, pooding, is the least .protective of the action.
alternatives. Flooding these materials would limit direct contact but may incre~se
mobility. Alternative 7 does not address the risks associated with the tailings deposits and
contaminated soils.
The surface water and ground water actions included as parts of Alternatives 1 and 2 and
3 + 3A would provide the most effective and most comprehensive treatment for surface
water and ground water of the seven alternatives. Surface water treatment in an upgrade
pond system would be provided for all flows up to the volume of a 100-year flood
(13,000 acre-feet) or until the maximum flow of 3,300 cfs had been reached. With
appropriate design and operation, water quality ambient and point sour~e discharge
standards should be met for nearly all flows up to the 100-year flood event. Alternatives 3,
4, 5, and 6 include. various levels of pond treatment, decreasing in the degree of effec-
tiveness. Alternatives 3 and 4 would provide suspended solids treatment for flood flows
between 600 and 4,000 cfs. Except for 2,000 acre-feet stored in the settling baSin, flows
above 600 cfs would not be treated for dissolved metals. Alternative 5 would upgrade the
pond system, but it would only accept flows up to 210 cfs for dissolved metals tr~atment.
Again, except for 2,000 acre-feet stored in the settling basin, flows between 210 and
4,000 cfs would receive treatment for suspended solids only. The lack of dissolved metals
treatment for above-average flows would lead to increased violations of water qual i ty
standards for flows out of the operable unit. Because of the decreases in levels or volume:
of treatment, water quality standards would be violated with greater frequency for eac~
-------
.. decrease in the upgrade of the treatment system. Alternatives 6 and 7 would likely.
experience regular violations during above average flows.
Trench drains for ground water collection and treatment are included as part of all seven
action alternatives. Alternatives 1, 2, 3, and 4 include trench drains both in and below
Pond 1. Alternative 3 + 3A includes an interception trench below Pond 1 and further
examination of the need for additional trenches, such as in Pond 1. Alternative.s 5 and 6
only,call for a single drain below Pond 1. A'single drain would effectiveiy limit the offsite
migration of contaminated groundwater but would not aid atte~pts to dewater Pond 1.
Alternatives 1 through 4 would treat the groundwater in the pond system. Alternatives 5
and 6 include the addition of onsite wetlands for treatment. The wetlands system would
decrease pumping requirements but could increase treatment operation and maintenance
requirements and lead to further contamination of the soil and ground water at the lower
end of the site.
,-
In general, permanence of the remedial actions is greatest for the more comprehensive
alternatives. The solidification of pond bottom sediments is the only alternative that
would permanently limit the mobility of the pond bottom sediments. Actions to stabilize
the pond berms (Alternatives 2 through 6) would protect the sediments as long as they are
maintained but would .not permanently affect the sediments themselves.
Surface water treatment would continue for as long as the ponds are functional. Under
current conditions, the estimated life of Pond 3 is approximately 10 to 25 years. The new
treatment pond (Alternative 1) could be constructed with an estimated life of up to 100
years. The increased pond volumes established with Alternative 3 + 3A would significantly
increase the estimated life Pond 3. However, any increase in estimated life would result in
decreased storage capacity for flood flows.
The permanence of efforts to cleanup the bypass would depend on efforts to keep the
. bypass free of future deposition. Alternatives 1 and 2, and 3 + 3A offer the greatest
-------
effectiveness, containing and treating all flows up to the 10o-year event. For Alternatives
3, 4, and 5, flows up to the 100-year event would have up to 80 percent of the suspended'
solids load removed.
The permanence of soils and tailings remediation is greatest for Alternative I because all
contaminated material would be solidified in the ponds. Alternatives 2 and 3 would
contain the material in several consolidated units that would reduce m~tenance.
requirements and aid permanence. The permanence of capping iri':'place or flooding
(Alternatives 4 through 6) is much more dependent on the continued maintenance of the. .
cap or cover.
9.2 COMPUANCE WITH ARARs
All of the alternatives would comply with most of the ARARs and replacement standards,
except for surface water ARARs. All of the alternatives would control non-point source
contamination from the Bypass, and contribute to overall surface water ARAR
compliance. Alternative 1, 2, and 3 + 3A would result in compliance with point source
discharge ARARs for surface water in normal conditions, and would ensure compliance up
to 100 year flood flows, by trapping those flows in full treatment systems. Alternatives 3
and 4 would result in compliance with point source discharge ARARs, but would fully
capture only limited pood events, and would not achieve ARARs compliance for surface
water during other flood events. Alternatives 5 and 6 would not achieve compliance with
point source discharge ARARs during certain times, and would not achieve compliance
with surface water ARARs during certain flood events. (Alternative 7, the no action
alternative, would not achieve compliance with any of the identified ARARs.)
Alternatives 5 and 6 would not achieve compliance with Floodplain Management
requirements. . Alternatives 1, 2, 3 + 3A, and 4 would comply with these ARARs.
-------
, Table 5 summarizes each alternative's compliance with federal and Montana ARARs. A
complete list of MARs is found in Attachment 1.
9.3 LONG-TERM EFFECTIVENESS AND PERMANENCE
The residual risks that would remain after implementation of each of the alternatives
increase from Alternative 1 throl1gh the no-action alternative (Alternative, ,7), which
involves the greatest residual risk. Alternative 1 would result in the'least residual risk
becal1se the measures it includes to eliminate or contain the risks"are'more comprehensive
- .
than any of the other alte.rnatives. This is primarily the case in regards to the. p~nd
bottom sedimeuts and the tailings and contaminated soils. Alternative 1 is the only
alternative to include treatment (solidification) of these materials.
Although Alternative 2 would not reduce the residual risk to the same risk level as
Alternative 1, it would protect the ponds from the threat of failure to a greater extent than
the remaining alternatives"~ and it would remove the majority of the contaminated soils and
tailings from the site. The residual risk in Alternative 2 results from the presence of the
untreated pond bottom sediments onsite.
Alternatives 3 through 6, in turn, contain a slightly higher level of residual risk from the
pond bottom sediments because the pond berms would be protected against only a fraction
of a PMF, rather than a full PMF (Alternative 2). However, the probability of a
catastrophic failure of the pond berms during a flood would still be small because the
likelihood of even a 0.2 PMF is quite small. (No specific return intervals are associated
with probable maximum floods, though their probability of occurrence is only once in sev-
eral thousand years.) Alternative 7, which would not further stabilize the pond berms.
would carry the greatest residual risk of flood damage to the ponds. The extent of
environmental damage that would result from a pond failure would also increase over time
with Alternative 7 because of the continued deposition of sediments within the pond.
-------
TABl£ S
ALTERNATIVE DESIGN
COMPUANCE SUMMARV FOR FEDERAL AND MONTANA AMAs
Montana SlIItuloryl
ARAR RegaAatory
5cAJ)ect SectIon Dc~ of P\"o\IIsIon AIL 1 AIL 2 AIL 3 All 3t3A AIL" AILS All 6 All 7
Surface Water ARM18.20.822 Amblenl surface water 'quality Ves. coutd treat Ves could Ireal the Ves. Ves Yes. Noc Noc No.
standards lor Montana's C-2 the 100 year flood 100 year flood
cl85slftcallon
ARM16.20.622 Weste treatment requtremenls Ves, could treat Ves. could treat Ves. Ves Ves. Noc No' No.
lor point aource dlschargea 10 the 100 year nood the 100 year nood
receMng watera
ARM16.20.203 Maximum contaminant levels for Ves. coutd treat Yes, coutd treat Ves. Vea Ves. No' No' No.
inorganic chemicals In potenllal the 100 year nood the 100 year flood
community water systems
40 CFS 122-125 MPOES monitorIng requtremenls Ves Ves Yes Yes Ves Ves Yes No
40 CFR 12S.13O(b) Besl Management P\"actlce Ves Yes Ves Yes Ves Ves Ves No
40 CFT Parts 230 & Dredge and ffll requirements Ves Yes Ves Ves Ves Ves Ves NA'
231
:
Air ARM16.8.815 Montana air standard for lead Yes Ves Ves Yes Ves Yes Yes NA
ARM16.8.818 Monlana partlcutate air standard Yes Ves Ves Yes Ves Yes Ves NA
ARM16.2.1401 (4) Montana PM-10 standard Yes Ves Ves : Yes Ves Yes Yes NA
40 CFR Part 50 National primary and secondary Yes Ves Ves Ves Ves Yes Ves No
ambient air quality standards
40 CFR Part 61 Nadonal emission standards for Yes Ves Ves Ves Ves Ves Yes No
hazardous air pollutants
Otound weier ARM16.20.1003 Montana Clan II ground water Yes Vas Ves Ves Ves Ves Ves No
quaRty standards
floodplain ARM36.15.606 Criteria for flood conlrol works Ves Ves Ves Ves Ves No No :No
~gement allowed within designated
floodways
-------
TABU: 5 (cor1tJoocd)
AlTERNAll-": DESIGN
COMPlIANCE SUMMARY R>R FEDERAL AND MONTANA AMRs
Montana Statutoryl
AnAR ReglMtory
WlJoct SectIon ~on of PrOlllslon AIt. 1 Alt.2 Alt.3 AIt. 3.3A Alt.4 Alt.5 Alt.6 Nt.. 7
ARM36.2.404 EvaluatIon criteria for Natural Yes Yes Yes Yes Yes Yes Yes No
Streambed and land Preservation Act
(1975) standards and guidelines
40 CFR 6(App. A) Floodplains Management Y"s Yes Yes Yes Yes Yes Yes Yes
Exec. Order 11.9118
40 CFR 6 (App. A) Protection of Wetlands Ves Ves Yes Yes Yes Yes Ves Ves
Exec. Order 11.990
Dam Salely ARM38.14.501 Design criteria lor high hazard dams Yes Yes Yes Yes Yes Yes Yes NA
ARM38.14.502 High hazard dam Innow design criteria Exceeds AMR Exceeds AMR Yes Exceeds Yes Yes NA NA
AMR
Reclamation ARM26.4.505 Require adequate cover over waste Ves Yes Yes: Yes Ves Yes Yes No
materials to prevent salt mIgration.
adverse plant ellects. and water
pollution :
ARM26.4.520 Crllerla for dIsposal olson materials Yes Yes Yes Yes Yes Yes Yes Yes
(tailings and contamInated solis)
ARM28.4.633 Require that allsurlace drainage Irom Yes Yes Yes Yes Yes Yes Yes No
disturbed areas be treated by the
best available technology
ARM26.4.642 Permanent Impoundment construction Yes Yes Yes Ves Yes Yes NA NA
and operation requtrements
ARM26.4.711 Revegetation actions must establish a Ves Yes Yes Yes Yes Ves Yes NA
diverse. ellectlve. and permanent
cover cepeble olsell-regeneratlon and
plant succession
Hazardous Wast. ARM18.4.701. Incorporation ollederal regulatory Yes Ves Yes Yes Yes Yes Yes No
ARM18.4.703 requirements that establish the
standards lor permitted hazardous
waste management lacillties
Solid Waste ARM18.14.505 Siting criteria lor solid waste dillposal Yes Yes Yes Yes Yes Yes Ves No
slles
40 CFR Part 257 Crlterta lor sond waste classlncetlon Ves Yes Yes Yes Yea. Yes Yes No
and disposal
-------
TABLE 5 (continued)
ALTERNAnve DeSIGN
OOMPUANCE SUMMARY FOR FEDERAL AND MONTANA ARARs
Montana Statutory,
ARAR Reptory
WI)ect SecIIan ~ of PrOlllslon AIt. 1 Alt.2 Alt.3 AIt. 3+3A Alt.4 Alt.5 Alt.6 Alt.7
OccupatloneJ Salet)' 29 use 85HJ78 Occupational Salet)' and Health Act Ves Ves Ves Ves Ves Ves Ves No
Hlstorfc 18 use 489 Archaeological and Historic Ves Ves Ves Ves Ves Ves Ves Vea
~eservatlon 40 CFR 8.301 (c) Preservation Act
18 use 470 NatloneJ Historic Preservation Act Ves Ves Ves Ves Ves Ves Ves Ves
40 CFR 8.301 (b)
38 CFR Part 800
18 use 4810467 Hlstorfc Sites, BWldlngs, and Ves Ves Ves Ves Ves Ves Ves Ves
40 CFR 8.301(a) Antiquities Act
18 US 1531.1568 Fish and WUdllle Coordination Act Ves Ves Ves Ves Ves Ves Ves No
40 CFR 8.302(g)
Natural Resources 18 use 1531.1543 Endangered Species Act Ves Ves Ves: Ves Ves Ves Ves No
50 CFR Paris 17 & 402
40 CFR 6.302(h)
:
40 CFR 6(App.A) Protection 01 Wetlands Ves Ves Ves Ves Ves Ves Ves No
ARAR compllancelndlcatll that the requlremente etated wID be met with the exception that ambient Wlter quaDt)' standards lor arsenic and mercury wiD not be met under an)' alternative and wID require a waiver based on technicallmpracticabWt)'.
811 discussed In Chapter 8. .
Alternative. 3 end 4 could trllt pond In-II0Wi up to 6000 cIs. ThIs magnitude olllow Is expected to be exceeded once every 2 )'ears.
AIternativII 5 and 6 could treat pond In-IIoWI up to 210 cis. ThIs magnitude olllow IseKpected to be exceeded once each )'ear.
Cunnt operating proceduree rllult In tha Ixcled8nce ol_ter quallt)' &tandards. about 40 percent 01 the time.
NA . not appDcable.
-------
Alternatives 1 and 2, and 3 + 3A would carry the least residual risk resulting from surface
water and groundwater contamination. Both alternatives. include measures to treat
basically all flows less than a 100-year flood passing through or frO.tI1 the system. Surface
water flows up to a 100-year flood flow would be detained and treated for suspended
solids and dissolved metals. Only flows greater than 100-year flood flow would pass
through the system untreated.
The 'residual risks associated with contaminated surface water increaSe with each of the
remaining alternatives. Alternatives 3 and .~, althougbtreating most flows in an upgraded
treatment system, include. only. suspended solids treatment for flows between 600. a~d
4,000 cfs that exceed 2,000 acre-feet. Alternatives 5 and 6 retain the current pond system
with a few modifications and would only allow dissolved metals treatment in the ponds for
flows up to 210 cfs because of the capacity limitations of Pond 2. For Alternative 5, flows
between 210 and 4,000 cfs would be treated for suspended solids in the upstream settling
basin. Up to 2,000 acre~feet of the flow would be retained and could be metered slowly
into the ponds for dissolved. metals treatment, if required. The modification of the current
pond system in Alternatives 5 and 6 also would not address the problem of potential short-
circuiting in the ponds. Short-circuiting decreases the effectiveness of the ponds.
Alternative 6, which does not include any treatment for flows greater than the current
capacity of the treatment system (210 cfs), may eventually result in the recontamination of
the bypass and the area below Pond 1.
Water quality standards for ambient and point source discharges should be met for all
flows up to at least lOO-year flood for Alternatives 1, 2, and 3 + 3A. Alternatives 3 and 4
may exceed these water quality standards during flows above 600 cfs. Alternatives 5, 6,
and 7 would experience more frequent violations and violations at lower flows than
Alternatives 1 through 4.
Both of the groundwater treatment actions should be able to meet the appropriate
treatment standards in the long term. However, pumping the collected groundwater to
-------
"Pond 3 for treatment (Alternatives 1 through 4) would require less startup effort and less
long-term operation and maintenance. The wetland system (Alternatives 5 and 6) would
require periodic maintenance and/or replacement of vegetCition, and more intensive care
during winter months than the pumping option. . The addition of the wetlands would also
result in the eventual recontamination of soils, sediments, and groundwater in the
treatment area, which may require future remediation during system closure.
. The operation and maintenance requirements for all of the action alternatives would be
fairly constant, because most of the requirements \Vould be related to the water treatment-
systems. Alternatives 1 and. 2 would re"quire somewhat greater operation and mainten~nce
than Alternatives 3, 4, 5, and 6 with respect to the requirements of the upstream flood
impoundment. Alternatives 3, 4, and 5 would have less substantial operation.. and
maintenance requirements associated with the upstream settling basin. Alternative 3 + 3A
would not involve operation and maintenance of a separate flood detention basin.
Alternative 6 would require the periodic removal of tailings from the bypass.
Alternatives 5 and 6 include a wetlands treatment system that would require some
operation and maintenance. Alternatives 3 through 5 include onsite capping of
contaminated material. Maintenance of the cap(s) will also increase the operation and
maintenance requirements. Alternative 3, which would consolidate the contaminated
material under a single RCRA-compliant cap, would have less maintenance requirements
" .
than the alternatives that would cap the material in place. Flooding the contaminants
(Alternative 6) would potentially have the greatest maintenance costs while offering the
least protection.
Monitoring requirements would basically be the same for all alternatives, limited to
ensuring conformance with surface and groundwater standards. No monitoring above the
current MPDES monitoring requirements would be added with Alternative 7, no-action.
-------
9.4
REDUCTION OF TOXICITY, MOBIUTY, AND VOLUME THROUGH
TREATMENT
Alternative 1 is the only alternative to use treatment in the remediation of pond bottom
sediments and contaminated soils and tailings. This treatment would reduce the mobility
of these materials by solidifying the sediments in place. It would also decrease the
potential for leaching metals from the sediment. A drawback to the in situ solidification
process is that it would substantially increase the total volume of the pond bottom
material. Approximately 2 cubic yards of ~olidification agents would have to be added to"
every 1 yard of sediments' treated in the wet portions of the ponds, thus tripling the vblume
of the sediments in these areas.
The remaining action alternatives would contain but would not treat the pond bottom
sediments. The containment actions would reduce the sediments' mobility to a lesser
extent than Alternative 1 because they stabilize the containment structures but not the
material itself. The containment actions would not affect the toxicity or volume of the
material. Alternative 2 includes the most stringent of the containment actions, stabilizing
the existing pond berms against the largest expectable forces, a probable maximum flood
and an maximum credible earthquake. The remaining action alternatives would limit the
mobility of the, pond bottom material during events up to a full maximum credible
earthquake and a fraction of a probable maximum flood. The no-action alternative would
not aff~ct the toxicity, mobility, or volume of the pond bottom material.
Remediation options proposed for the tailings and contaminated soils also vary in their
effectiveness in limiting the future mobility of the material. Through offsite disposal of a
majority of these materials, Alternative 2 removes the threat of remobilization at the site.
although the material would continue to exist in an untreated state at a separate si te.
Alternative 3 offers the best onsite reduction in mobility through consolidation followed by
containment under a RCRA-equivalent cap. Alternative 3 + 3A incorporates tailings and
contaminated soils disposal in Ponds 1 and 3 under a protective soil cover, revegetatcl!
-------
with native grasses. Alternatives 4 and 5 would also reduce the mobility of the material
although not to the same extent as Alternative 3. Alternative 3 + 3A and 6 would not
greatly reduce the mobility of the contaminated tailings and soils that would be flooded.
although they would reduce the threat of direct contact.
All of the action alternatives include treatment to reduce the toxicity of the surf~ce water
to some degree. The alternatives differ in the amount and level of treatment. Alter-
natives 1 and 2, and 3 + 3A include treatment to reduce the toxicity of contaminated water
for all flow conditions up to a 100-year floo.d. Flo~s above the lOO-year flood flow would. .
bypass the system untreated.
The remaining alternatives reduce the toxicity of the surface water to a lesser extent.". For
Alternatives 3 and 4, only flows below 600 cis and 2,000 acre-feet of flows above 600 cis .
would be treated in the pond system for suspended solids and dissolved metals. Flows
between 600 and 4,000 cis would be treated for suspended solids only in the upstream
impoundment. Alternatives 5 and 6 retain the present pond system with a few
modifications to improve treatment. Alternative 5 includes a settling basin to contain up
to 2,000 acre-feet and treat flows between 210 and 4,000 cis for suspended solids. Flows
above 210 cis would not be treated for dissolved metals. Alternative 6 does not include
any treatment for flows greater than the capacity of the current pond treatment, 210
system.
9.5 SHORT-TERM EFFECTIVENESS
Most of the components of the action alternatives would take 2 to 3 years to c~mplete.
The components are similar for the most part, varying primarily in size or layout. The
solidification of the pond bottom sediments is an exception to this. Alternative 1 would
require approximately 17 years to complete because of the large volume of soils and sedi-
ments to be solidified. Although the stability of the sediments would increase during the
-------
. solidification process, it would still take a substantially longer time to reach complete
protection from Alternative 1 than from any of the other alternatives.
The wetlands treatment system included as part of Alternatives 5 and 6 may need up to
5 years startup time to reach the design objectives of the system. This time is needed to
establish plant species in the system in order to realize effective treatment. None of the
other treatment components included with the alternatives would require all extended
startup period, though optimizing operation of a modified or new pond treatment system
may require a full year or more of operati~nal experience.
Alternative 2 would have substantial iinpacts on the area, by causing trucks carrying
contaminated soils to travel on public roads.
None of the action alternatives are expected to substantially affect the community of
Warm Springs during remediation. Local releases of metal-contaminated tailings or dust
would likely occur during construction work carried out in contaminated areas, but such
releases would be minimized by dust control techniques and would not be expected to
affect the community. There is also the potential for short-term violations of tbe water
quality standards at the compliance point as a result of remediation work in or adjacent to
the bypass and stream beds. Those violations would differ somewhat between alternatives
and could be minimized through use of sedimentation barriers and sedimentation ponds.
Construction contractors would need protection against dermal and respiratory exposure to
the tailings while working in contaminated areas. Dermal tbreats could be controlled
using long-sleeve protective clothing, and inhalation threats could be controlled using
appropriate dust or face masks. Health risks to operation and maintenance workers would
be substantially less than for such workers under the existing conditions (see the PHEA,
Appendix A of the Warm Springs Ponds Operable Unit Feasibility Study). These risks
would be similar for all alternatives.
-------
-Planning for all remediation activities would have to consider potential impacts to a pair
of bald eagles, which are protected under the Endangered Species Act, tJ:lat have previ-
ously nested within the operable unit. The eagles are not currently nesting within the
operable unit but they continue to use the ponds as a food source during the summer
months. Only Alternative 1 would substantially affect this food source. If the eagles
return to nesting in the area surrounding the ponds, steps would be required to minimize
any impact resulting from construction. This would be done during project planning in the.
design and construction phase of remediation. With attention to the necessary controls,
adverse impacts to. the eagles can likely be.. avoided. This would be true for all seven of. .
the action alternatives.
Environmental impacts to the operable unit would be greatest for Alternative 1 because of
the in situ solidification process proposed for the existing ponds. Solidification of the
existing ponds would alter several hundred acres of land that is currently wetland wildlife
habitat. Prior to solidification, the ponds would be drained; about 17 years later, following
solidification and covering with soil and vegetation, they would be left as dry, vegetated
terrestrial habitat. Some of the lost open-water habitat would be replaced by the new
treatment pond, which' would be constructed upstream of the present ponds. . The new
flood impoundment pond, although not typically containing water, would permanently
remove approximately 1,000 acres of rangeland from use, bringing the total acreage
affected by this alternative to approximately 2,250 acres.
The remainder of the alternatives would not significantly affect the environment in and
around the pond system on a long-term basis, except for the loss of wetlands in Pond 1
and the effects of the upstream impoundments. The flood impoundment, as discussed
above, would affect approximately 1,000 acres. The smaller settling basin (Alternatives 3
through 5) would affect approximately 500 acres. Alternatives 2 through 6 (except 3+3A)
would affect the local environment during implementation as a result of construction
activities. These alternatives would affect the surrounding wildlife habitat with increased
noise and .dust levels. Some habitat would also be temporarily destroyed as a result (Ii
-------
. .necessary earthwork. These impacts would likely be short-lived and the areas returned to .
their preconstruction condition fairly quickly.
9.6 IMPLEMENT ABILITY
For the most part, there is not a great deal of difference in the implementability of the
seven action alternatives. Most of the components proposed as part of the alteI1latives are
well-developed technologies, used to some extent in either the hazardous waste, water, or
standard civil engineering disciplines~ 1?e technical feasibility .Of these components..
appears to be good. The exceptions are the two innovative components included as.part
of a number of the alternatives: in situ solidification, and wetlands treatment for metals
removal.
--\
The technical feasibility of in situ pond bottom solidification (Alternative 1) is not known
for certain at this time. It has been used with success to solidify marshlands for foundation
stabilization in Japan, but it has not been used extensively on hazardous waste sites.
Consequently, it has a greater risk of implementation difficulties and failure than any of
the other media-specific actions proposed for the pond bottom sediments. If it fails to
adequately solidify the pond bottoms, for whatever reason, additional stabilization of the
pond berms (as in Alternatives 2 through 6) would be necessary.
Wetlands treatment (Alternatives 5 and 6) has been used with some success for removing
metals loadings from acid mine drainage, and its technical feasibility is somewhat more
defined than in situ solidification. However, because effective treatment relies on the
development of a resilient living ecosystem in the wetland, the implementation of an
. .
effective wetland could prove difficult and/or time consuming. The effectiveness of the
wetlands system also depends to some extent on the weather. A large winter buildup of
ice could result in severe short-circuiting in the wetland, decreasing the observed removal
effectiveness. The technical feasibility of the other groundwater treatment component.
which relies on treatment in the pond system (Alternatives 1-4), is greater.
-------
. The technical feasibility of the remainder of the components would be about equal
Protecting the pond berms against a fraction of a PMF (Alternatives 3-6) would be more
feasible than protecting the berms from a full PMF (Alternative 2) . simply because of the
magnitude of the project. The same holds true for the upstream settling basin (Alterna-
tives 3, 4 and 5) versus the upstream flood impoundment (Alternatives 1 and 2). Because
the settling basin would be smaller and would require fewer materials, its overall feasibility
would be' greater.
.. , .
Froman administrative feasibility standpo~~t, all of the alternatives. are about eCJ.ual.. All.
eight alternatives (no-actioll alternative- included) would require compliance with disch,,:rge
standards for water from the treatment system into the Clark Fork River. The discharge
standards are more likely to be met for Alternatives 1 through 4 because they include a
more comprehensive upgrading of the treatment system. They are not likely to be met
with sufficient regularity under Alternatives 5 through ,7. Alternative 2 would require
obtaining permits for off-site disposal.
,-.
Alternatives 1 through 5 (except Alternative 3 + 3A) would require the acquisition of 500
to 1,000 acres of rangeland for construction of the settling basin or the upstream flood'
impoundment. Because less land is needed for the smaller settling basin, Alternatives 3, 4
and 5 might be easier to implement.
The offsite disposal option, proposed for the majority of the tailings deposits. and
contaminated soils as part of Alternative 2, would be more difficult to implement than the
remainder of the contaminated soils options. Required permits for off-site disposal would
have to be obtained. The interstate transport of up to 1.5 million cubic yards of untreated
waste would be administratively undesirable from both a transportation and disposal point
of view~ The' onsite disposal options (Alternatives 1 and 3-6) would likely be easier to
implement.
-------
An apparent lack of locally available rip rap would favor the alternatives that require.
smaller amounts of that material (e.g., Alternatives 3-6 over Alternatives 1 and 2).
However, Alternative 3 + 3A utilizes soils cement which incorporates on-site materials.
This would be significant especially if the material would need to be quarried specifically
for implementation. Other materials and equipment would be readily available for
construction. The in situ solidification units for Alternative 1 would require up to
9 months for fabrication, but this could be incorporated into the scheduled implementation
without causing unforeseen delay.
9.7 COST
The cost comparisons are straightforward. Comparing present worth costs, Alternatiy~ 1 is
most expensive and Alternative 6 is the least expensive of the action alternatives. The
. long implementation schedule more strongly affects the present worth cost for.
Alternative 1 than do the implementation schedules of the other alternatives, which are
shorter. The costs of the action alternatives are listed in Table 6, both with and without
present worth considerations.
9.8 STATE ACCEPTANCE
The State of Montana, acting through the Department of Health and Environmental
. .
. Sciences, generally agrees with this Record of Decision. The State has withheld
concurrence on this Record of Decision until EP A selects cleanup action levels and
determines appropriate measures for the control of soils, sediments and tailings above
those levels which are not addressed by this action.
The State agrees with the final ARARs list. The State is particularly concerned that the
point source discharge from the Warm Springs Ponds remain as a regulated discharge
subject to the MPDES permit requirement.
-------
TABLE 6
OOST ES11MATES R)R THE ACT10N Al1ERNAll~
Cost CompaII8tIIII Altemdlv81 Altllmllllve2 Alternative 3 Altemstlve 3.3A Altem&tIYe 4 Alternative 5 Alternative 8
Estimated Construction 11,665,000,000.00 1250,000,000.00 160.100.000.00 45,700,000.00 165.500,000.00 $56,000,000.00 $46,100.000.00
Cost
EngIneering Design 4,000,000.00 4,000,000.00 4,200.000.00 3,500.000.00 4,600.000.00 3,900,000.00 3.200,000.00
Services DurIng 7,000,000.00 4,000.000.00 4,200,000.00 3,500,000.00 4,600,000.00 3.900,000.00 3,200,000.00
Construction
Administrative & Legal 3.000,000.00 3.000,000.00 3.000.000.00 2.800.000.00 3,300.000.00 2,800.000.00 2,300.000.00
Tot81 Project Coat 1.679.000.000.00 261,000.000.00 71,500,000.00 55.500,000.00 78.000.000.00 66.~.000.00 54.800.000.00
Operation and 283.000.00 301,000.00 300.000.00 379.000.00 293.000.00 284.000.00 270.000.00
Maintenance
(Vearty Costs)
TotIII Pr-c Wor1h Cost 11,191.500,000.00 1241,500,000.00 171,100,000.00 157,416,000 177,000,000.00 $66.300.000.00 $55,100.000.00
-------
. 9.9 COMMUNITY ACCEPTANCE
Community reaction to the proposed plan was vigorous and widespread across
communities in the Clark Fork River Basin. A full response to comments from the
community and from ARCO is contained in the responsiveness summary attached to this
Record of Decision.
Generally, the comments from the community fell into these categories:
. The process for. .selecting a. remedy should provide for additional and earlier
community involvement. In response EP A extended the public comment period
for the proposed plan to 90 days, and has held several public meetings; and
meetings with interested groups over the last year. EP A has initiated several
activities at other operable units within the' Clark Fork River Basin to increase
community involvement in the Superfund process at earlier stages.
EP A will also include community involvement in the process which will address
further remedy action selection at the Warm Springs Ponds.
. There was widespread opposition to the creation of an additional upstream
settling basin, particularly by residents of Opportunity and Anaconda. In
response, EP A has reevaluated the preferred alternative described in the
proposed plan, and has selected alternative 3 + 3A, which provides for flood
storage within existing Pond 3.
. Many comments stressed the need for flood protection and treatment of flood
flows before water from the operable unit enters the Clark Fork River. In
response, EPA has reexamined the berm-strengthening ARAR, and determined
that a standard of one-half of the probable maximum flood is necessary for all
. berms within the operable unit. In addition, EPA has, selected a remedy whidl
-------
. will fully treat flows entering the system up to the tOO-year flood flows before
that water is released into the Clark Fork River.
. Many comments. stressed the need for permanent remedies which utilized
treatment of hazardous substances found at the site. In response, EP A carefully
examined the issue and concluded that the extensive berm strengthening and
contaminant cover requirements of the selected remedy provide a p~rmanent
remedy for the site. The EP A notes that floods of Mill. and Willow creeks may
not be allowed to enter the pond~ after the berms are. strengthened, and that.. .
Silver Bow Creek floods will enter the ponds in a secure and controlled mariner
which will prevent releases of contaminants.
Treatment options and off-site disposal options are not technically feasible at this time, or
present negative aspects such as destruction of wetlands or excessive traffic, and are
extremely expensive in relation to the benefits gamed. Superfund remedies are required to
be cost effective.
The EP A will continue to examine treatment options carefully at other operable units in
the Clark Fork River Basin. The EP A also notes that the final determination for Ponds 2
and 3 will be made at a later date, when sources of contamination from upstream have
been cleaned up and the ponds are no longer needed as treatment facilities.
-------
10.0 THE SELECfED REMEDY
After evaluating alternatives with respect to each other and the nine required criteria, the
EPA and MDHES have identified Alternative 3+3A as the selected remedy for this Warm
Springs Ponds Operable Unit Record of Decision. The agencies have determined that
Alternative 3 + 3A is the most effective of the alternatives evaluated, offers the greatest
potential for being a permanent remedy, is supported by the public and is cost . effective.
The selected remedy is an interim cleanup measure that provides the highest degree of
certainty that it will be successful and pe~anent, The final measure of these qualities. .
awaits additional actions at this operable unit and cleanup decisions upstream,' 'The
components of Alternative 3 + 3A are as follows:
. Allow the ponds to remain in place; Ponds 2 and 3 will continue to function as
treatment ponds until upstream sources of contamination are cleaned up;
. Raise and strengthen all pond berms according to specified criteria, which will
protect against dam failure in the event of major earthquakes or floods, and
increase storage, capacity of Pond 3 to receive and treat flows up to the lOO-year
flood;
. Construct new inlet and hydraulic structures to prevent debris from plugging the
Pond 3 inlet and to safely route flows in excess of the lOO-year flood around the
ponds;
. Comprehensively upgrade the treatment capability of Ponds 2 and 3 to fully treat
all flows up to 3,300 cubic feet per second (1 DO-year peak discharge) and
construct spillways for routing excess flood water into the bypass channel;
. Remove all remaining tailings and contaminated soils from the Mill-Willow
Bypass, consolidate them over existing dry tailings and contaminated soils within
-------
the Pond 1 and Pond 3 berms, and provide adequate cover material which will
be revegetated.
. Reconstruct the Mill-Willow Bypass channel and armor the north-south berms of
all ponds to safely route flows up to 70,000 cubic feet per second (one-half of the
estimated probable maximum flood);
. Flood (wet-close) all dry portions of Pond 2;
. . Construct interception trenches to collect contaminated ground water hi and
below Pond 1 and pump the water to Pond 3 for treatment;
. Dewater wet portions of Pond 1 and cover and revegetate (dry-close) all areas
within the Pond 1 berms;
. Establish surface and ground water quality monitoring systems and perform all
other activities necessary to assure compliance with all applicable or relevant and
appropriate requirements;
. Implement institutional controls to prevent future residential development, to
prevent swimming, and to prevent consumption of fish by humans; and
. Defer, for not more than one year after the effective date of this document,
decisions concerning the remediation of contaminated soils, tailings, and ground
water in the area below Pond 1, pending evaluation of various wet- and dry-
closure alternatives and a public review.
-------
10.1 OVERALL PROTECTION OF HUMAN HEALTH AND TIlE ENVIRONMENT .
Alternative 3 + 3A reduces or eliminates those risks to human health and the environment
at this operable unit which are within the scope of this Record of Decision.
Pond stability is addressed by protecting the ponds against both the maximum credible
earthquake and one-half of the probable maximum flood. Only under extreIIle flooding
conditions would the stability of the pond berms be in question. Protection of the pond
berms would continue for as long as the b~rms are properly maintained and repaired.
Alternative. 3 + 3A will improve surface water quality by completely upgrading the existing
pond treatment system to provide treatment for all flows up to 3,300 cfs or 13,000. acre-
feet (the estimated peak flow and volume, respectively, of the 100-year flood), by removing
tailings from along the bypass, and by raising the berms for Ponds 2 and 3. Pond 2 will be
increased in volume, reducing the problem of resuspension of sediments during high winds
and flow rates. Wi~h the upgraded treatment system, most of the surface water quality
violations, which now occur, should be avoidable.
All flood flows up to 3,300 cfs will be routed through the pond system, whic~ will result in
removal of the majority of the suspended particles. Thus, Alternative 3 + 3A will
substantially reduce the potential for future recontamination of the bypass by settled
tailings and reduce the continued transport of tailings into the Clark Fork River.
Contaminated ground water moving from the site will be collected from Pond 1 and below.
It will then be pumped to the Pond 3 inlet for treatment in the pond system. This will
reduce the discharge of metals loading into the Clark Fork River and should enable
compliance with primary MCLs for groundwater at the selected compliance point.
Dry-closure of Pond 1, which includes dewatering, covering and revegetation of tailin~s
and contaminated soils, will effectively isolate them from direct contact and limit their
-------
mobility. The cap will provide a barrier against ingestion, inhalation, and runoff. As long
as the cap is maintained, the material will be safe from releas~s due to erosion of the cap.
Capping the material in Pond 1 will not, however,reduce the toxicity, volume, or
persistence of the material.
Flooding the tailings deposits and contaminated soils in the dry portions of Pond 2 will
reduce the potential for exposures to these materials, although the exposed tailings above
Pond>3 will not be addressed until final closure of the ponds.
All of the components of Alternative 3 + 3A are technically feasible. With the appropriate
design, construction, operation, and maintenance, the components of Alterative 3 + 3A will
reliably reduce the risks for which they are proposed. Any increased risks to. the
surrounding environment and community during implementation .can be kept to a
minimum with appropriate containment and construction safety measures.
10.2 COMPLIANCE WITH ARARS
Within the boundaries defmed below, Alternative 3 + 3A should result in compliance with
all the state and federal ARARs identified for the Warm Springs Ponds Operable Unit.
The actions proposed for Alternative 3 + 3A meet the Montana ARARs for protecting the
pond system against one-half of the probable maximum flood and the maximum credible
earthquake. Providing flood detention within the pond system and the upgrading of the
pond treatment system should result in effluent compliance with ambient and point source
discharge surface water quality standards for all flows up to 3,300 cfs, the estimated peak
. .
flow of the 100-year flood. .
Alternative 3 + 3A should comply with Montana ground water standards, and would satisfy
Montana's' requirements for floodplain management. Excavating and moving tailings
deposits and contaminated soils from the Mill-Willow bypass to dry portions of Pond 3
-------
prior to capping will comply with state and federal siting criteria for solid and hazardous
waste disposal and can be done so as to selected RCRA requirements for closure of a
hazardous waste management facility.
Flooding the tailings deposits and contaminated soils in Pond 2 will reduce the risks
identified in the public heath and environmental assessment. and meet the remedial
objectives established for the operable unit. A complete summary of this alternative's
compliance with the state and federal ARARs is presented in Table 5:.
The actions proposed in Alternative 3 + 3A could result in adverse effects on wetlaJ.1ds,
endangered species, and historical resources. To mitigate these potential impacts, addi-
tional consultation with state and federal resource agencies will be required during
implementation of this interim remedy.
10.3 LONG-TERM EFFECTIVENESS AND PERMANENCE
=-
Alternative 3 + 3A addresses all of the identified risks at the site by using measures
intended to limit or remove the risks. The primary risk at the site, the release of 19 million
cubic yards of metal-contaminated tailings located in the treatment ponds, will be
addressed by protecting the pond berms against. failure due to a full maximum credible
earthquake and one-half of the probable maximum flood. This will address the threat of
pond failure in all but extreme cases and this aspect of the re~edy is permanent and
effective over the long term.
Residual risks will result from the continued existence of the 19 million cubic yards of
sediments and sludge in the pond system although those risks will be reduced to a low
level. The sediments in the ponds may still be released to the. environment in either
dissolved or suspended form under extreme conditions. This includes not only the sludges
presently in the ponds, but also all the excavated soils and tailings from around the site
that will be placed in Pond 1 or Pond 3 prior to capping.
-------
,Because the material beneath the cap will be untreated, maintenance and periodic
inspection of the Pond 1 and Pond 3 caps will be necessary. ¥aintenance ~ctivities will be
directed at preventing erosion or deterioration of the cap. Periodic inspection and
maintenance of the stabilized pond berms will also be necessary to ensure continued
protection.
Continued maintenance of the ground water interception and pump system will be
necessary. By drying Pond 1, rather than wet closing it, this oper;;ition maintenance period
is expected .to be shorter and less complicated.
10.4
REDUCTION OF TOXICITY, MOBIliTY, AND VOLUME, THROUGH
TREATMENT
Protecting the pond berms against failure due to a maximum credible earthquake and one-
half of the probable maximum flood will substantially reduce the potential mobility of the.
sludges in the ponds, although this will not affect their volume, persistence, or toxicity.
Only in the extreme case of flooding above the design floods for Alternative 3 + 3A would
the current containment of the pond bottom sediments be affected.
. Onsite disposal of excavated soils and tailings within Ponds 1 and 3, designed to meet
selected RCRA. requirements will reduce the mobility of those materials and will prevent
direct human contact. The materials will be taken from currently exposed areas along the
bypass. The tailings deposits and contaminated soils in Pond 2 will be covered by water,
but the naturally deposited tailings above Pond 3 will remain exposed until final closure of
the ponds. The pond bottom sediments in Ponds 2 and 3 will remain covered by water bu t
will not be completely immobilized against wind and water action.
The toxicity, volume, persistence, and propensity to bioaccumulate of the pond bottom
sediments and the tailings deposits and contaminated soils will not be altered by the pr~.
-------
. posed actions of Alternative 3 + 3A . To date, no feasible technology exists which will
provide effective treatment of wastes present at this operable unit.
The flood detention capacity will address the threat resulting from floods, the transport of .
. .
tailings through the system and into the Clark Fork basin, by ensuring that all flows
through the reach, up to the 100-year flood, are at least treated for suspended solids. .
Flows in excess of 3,300 cfs will bypass the pond system and !lot be treated for either
suspended or dissolved metals. This may l~ad to short-term water quality exceedences in..
the effluent from the operable unit during high flows,but it will probably not have alQng-
term impact on the operable unit.
10.5 SHORT-TERM EFFECTIVENESS
It will take approximately two construction seasons to protect the pond berms against a
maximum credible earthquake and one half of the probable maximum flood. During that
time, the protectiveness provided for the downstream community of Warm Springs and the
environment will decrease. The overall level of protectiveness of Alternative 3 + 3A will
not be attained until construction is completed. The potential for increased risk to either
the community or the environment during the berm stabilization process is limited.
Remediation contractors may be at risk from direct contact and inhalation of contaminants
during foundation excavation and associated tasks. These risks will be controlled by using
protective equipment as necessary.
Risks to the remediation contractors will be limited to standard construction risks
associated with similar projects. The diversion and inlet structures will be constructed in
the contaminated stream channel. Precautions will be required to avoid excessive
additional contamination of the creek flows during construction of these features.
-------
10.6 IMPLEMENTABILITY
Protecting the pond berms against a maximum credible earthquake and one-half of the
probable maximum flood is feasible. The current uncertainties involve the existence of
suitable foundation material downstream of the toes of the existing berms, the nature of
the upstream slopes, and the actual value of the maximum credible e~hquake.
Preliminary investigations indicate that the base material beneath the surface soils will be .
, . . .
adequate, but this will have to be verified during the design phase. The materials and
construction of the upstream slopes, and the maxilImm' credible ~arthquake also will have. .
to be' determined during the design phase.
A revised Montana Pollution Discharge Elimination System' permit to discharge water
from the treatment system into the Clark Fork River will be required and the discharge
standards are expected to be met.
10.7
COMMUNITY AND STATE ACCEPTANCE
This remedy was designed to meet the community concerns expressed during the comment
period. The State has been actively involved with the development of this alternative, and
generally agrees with its selection.
10.8 SUMMARY OF TIlE PREFERRED ALTERNATIVE
In summary, the preferred alternative satisfies the statutory preference for remedies that
employ treatment as a principal element to reduce the toxicity, mobility, or volume of
contamination at the site to the maximum extent practicable. With the comprehensive
upgrade of the. current pond treatment system, both surface and ground water will be
treated and their toxicity.will be reduced.
-------
. The preferred alternative will attain federal and State requirements that are applicable or
relevant and appropriate for the site with minor exceptions: . State ambient water concen-
trations of toxic or deleterious substances to protect public health from ingestion of .
contaminated water and fish for arsenic and mercury require a waiver based upon
technical impractability and upon the fact that this is an interim remedy. The arsenic
standard for water and fish ingestion is 2.2 nanograms per liter and the mercury standard
is 144 nanograms per liter. It is not technically feasible to treat water to tho~e levels at
this time. In addition, arsenic cannot be detected at 2.2 nanograIIlS per' liter with sampling
and detection methods currently availabl~. Because it is not possible to treat or to. '.
determine compliance with. these standards, and because this remedy is an interim cleanup
action, these requirements are waived. In addition, should the areas within the pond
berms be considered within the 100-year floodplain, requirements prohibiting disposal of
solid waste within the floodplain are hereby waived.
Based upon the information available at this time, the State and EP A believe that the
selected remedy will be protective of human health and the environment, will comply with
federal and State ARARs, will be cost-effective, and will utilize permanent solutions and
treatment technologies to the maximum extent possible., recognizing the scope of this
interim cleanup action.
-------
11.0 STATUTORY DETERMINATIONS
Under their legal authorities, the EPA and MDHEShave the primary responsibility at
Superfund sites to undertake remedial actions that achieve protection of human health and
the environment. In addition, Section 121 of CERCLA establishes that, the selected
remedial action must comply with applicable or relevant and appropriate environmental
standards established under federal.and State environmental laws unless a statut9ry waiver
is justified. The selected remedy also must be cost-effectiv~ and. utilize permanent
solutions and alternative treatment techno~?gies or resource rec-overy technologies to the. .
maximum extent practical.. The stafute also includes a preference for remedies. .that
employ treatment that permanently and significantly reduce the volume, toxicity, or
mobility of hazardous wastes as their principal element. The following sections discuss
how the selected remedy meets these statutory requirements.
11.1 PROTECTION OF HUMAN HEALTH AND THE ENVIRONMENT
The selected remedies for the various contaminant sources are protective of human health
and the environment, within the scope of this interim action. They will meet the ARARs
identified for the operable unit and reduce the risks identified in the PHEA to acceptable
levels.
11.2 COMPLIANCE WITH APPUCABLE OR RELEVANT AND APPROPRIATE
REQUIREMENTS (ARARs)
Federal and State applicable or relevant and appropriate requirements have been
determined. The selected remedy will comply with most applicable or relevant and
appropriate requirements. However, three cbemical- and location-specific ARARs
pertaining to water quality standards and potential solid waSte disposal requirements will
be waived.
-------
11.2.1 Waivers and Promu~ Stand.ID:Qs
Federal law recognizes there may be instances in which ARARs cannot be met with
respect to remedial actions onsite. . It, therefore, identified six circumstances under which
ARARs may be waived. However, other statutory requirements--specifically, the
requirement that remedies be protective of human health and the environment--cannot be
waived. Waivers occur as the exception, not the rule. Waivers are appropriate .if:
. The remedial action selected is an interim remedy and. only part of a total..
remedial action that will attain .ARARs.
. Compliance with ARARs at the site would result in greater risk to human health
and the environment than alternative options.
. Compliance with ARARs is technically impracticable, from an engineering
perspective.
. The remedial actions selected will attain an equivalent standard of performance,
although ARARs are not met.
. With respect to State ARARs, the State has not consistently applied ARARs in
similar circumstances at other remedial actions within the State.
. In the case of fund-financed remedial actions, financial restrictions within the
Superfund program require fund-balancing such that satisfactions of ARARs at
the site must give way to a greater need for protection of public health and
welfare and the environment at other sites.
The feasibility study, which provides a detailed analysis of the remedial action alternatives,
identifies how each alternative complies with ARARs. If an ARAR would not be satisfied,
-------
then a waiver may be required, based on the interim nature of this action and the
technical infeasibility of meeting those standards. See section 4.0 of the ROD and the
ARARs list. There is the possibility that the area within Ponds 2 and 3 to the 100 year
flood flow event may be considered part of the 100 year flood plain. If so, the ARAR
prohibiting disposal of solid waste within the 100 year flood plain is waived on the same
basis.
11.3 COST-EFFECTIVENESS
The selected remedial alternatives are cost-effective options for cleanup of the Warm
Springs Ponds. This determination is based on the cost and overall effectiveness of the
selected remedy when viewed in light of the cost and overall effectiveness of other
alternatives.
11.4
UTILIZATION
TREATMENT
PRACTICABLE
OF PERMANENT SOLUTIONS AND ALTERNATIVE
TECHNOLOGIES TO THE MAXIMUM EXTENT
The selected remedy satisfies the statutory preference for utilization of permanent
solutions and alternative treatment technologies to the maximum extent practicable.
Treatment of contaminated water is an element of the selected alternative.
Implementation of the selected alternative. Other forms of treatment are not yet
technically feasible or practicable at this time. Implementation of the selected alternative
will decrease the concentrations of contamination sources.
-------
.12.0 DOCUMENTATION OF SIGNIFICANT CHANGES TO COMPONENTS OF THE
SELECfED REMEDY
Section 117(b) of CERCLA requires documentation and explanation of any significant
change from the preferred alternative originally presented in the Proposed Plan. The
remedy selected in this Record of Decision does, in fact, reflect significant changes to the
originally preferred alternative. Therefore, in accordance with specific requir~ments of
Superfund guidance (OWSER Directive 9335.3-02)10, the originally preferred alternative
will be identified, the significant changes.. described, . and the reasons for the changes.
explained.
12.1
THE ORIGINALLY PREFERRED REMEDY
The Wann Springs Ponds Proposed Plan (October, 1989)11 described six cieanup
alternatives. The preferred alternative, Alternative 3, may be summarized as follows:
. Allow the ponds to remain in place; Pond 3 would continue to function as a
treatment pond;
. Raise and strengthen all three pond berms to protect against dam failure in the
event of major earthquakes or floods;
. Construct new inlet and hydraulic structures to prevent debris from plugging the
Pond 3 inlet and to safely route flows in excess of 600 cfs around the ponds.
'0 OSWER Directive 933S.3-C2, November, 1989, EPA!S4fJ!G-89-007, Interim rlDal Guidance on Preparing Superfund
Decision Documents
" Montana Department of Health and Environmental Sciences and United States Environmental Protection Agency, 1989. Warm
Sprin~ Ponds Proposed Plan, Silver Bow Creek Superfund Site Report.
-------
. Construct an upstream sediment settling basin capable of storing up to 2,000
acre feet of flood waters, with hydraulic structures to. meter the water into Pond
3 for treatment;
.. Comprehensively upgrade the treatment capability of Pond 3, including
construction of a berm' across the pond to prevent flows from short-cir~iting.
"
>. Remove all tailings and contaminated soils in the Mill-Willow Bypass and
consolidate them over existing dry..tailing~ and soils behind the Pond 1 berm;'
. Reconstruct the Mill-Willow Bypass channel and armor the north-south berms of
all three ponds to withstand fractions of the probable maximum flood (0.2; 0.3
and' 0.5 PMF for Ponds 1, 2 and 3, respectively);
. Flood (wet-close) or excavate, consolidate and cap (dry-close) all exposed tailings
and contaminated soils with arsenic or metals concentrations exceeding the
prescribed health-based action levels;
. Construct ground water interception trenches within the below Pond 1 to prevent
contaminated ground water from entering the Clark Fork River and pump the
collected water up to Pond 3 for treatment; and
. Excavate tailings and contaminated soils below the Pond 1 berm (largely within
the original Silver Bow channel), consolidate and cap them behind the Pond 1
berm, and dry-close Pond 1.
-------
.12.2
SIGNIFICANT DIFFERENCES BETWEEN THE ORIGINALLY PREFERRED
ALTERNATIVE AND.SELECfED REMEDY
The remedy selected in this Record of Decision differs from the originally preferred
alternative in the following respects:
. The upstream sediment settling basin will not be constructed. Instead, flood
flows up to the 100-year event will be routed into Pond 3;
. Pond 2 will be retained as 3: treatment pond, as opposed to simply being wet-
closed; .
. The berms of all three ponds will be raised, strengthened, and their north-south
aspects armored, to withstand one-half of the estimated probable maximum flood
(70,000 cis), as opposed to less protective fractions of the PMF for Ponds 1 and
2 (0.2 and 0.3 PMF, respectively);
. Comprehensive upgrading of the treatment capability of Pond 3 will not include
construction of a berm across the pond. Instead, bioassay tests will be conducted
to evaluate the effect of resuspended bottom sediments on aquatic species. If
effects are observed, measures other than a berm can be incorporated as a
component upgrade.
. Two aspects of the decision are deferred. A final cleanup level for soil
contaminants will be selected at a later date. Once this decision is made,
additional cleanup of soils, sediments, and tailings may be required, especially
below Pond 1. The decision on final disposition of Ponds 2 and 3 is also
deferred, until upstream cleanup decisions are made and there is no longer a
need for use of the ponds as water treatment facilities.
-------
.12.3
REASONS FOR SIGNIFICANT CHANGES
The most prominent difference betWeen the originally preferred remedy and the selected
remedy is the elimination of the upstream sediment settling basin. In fact, that change is
perhaps the only significant difference betWeen the tWo cleanup approaches; the other
differences summarized above are consequences of the decision to route floods. up to the
100-year event through the pond system. Their expected performance in relation to their
predecessor components and in relation to the nin~ criteria specified by the NCP is
thoroughly evaluated and described in Section 8.0, comparative analysis of the alternatives. .
The rationale for eliminating the upstream sediment settling basin is explained in the
Declaration (page 1-5) and documented in the Responsiveness Summary. Brief1y~ the
changes were made in response to public opposition to another contaminated pond in the
vicinity of the Opportunity Tailings Ponds or the town of Opportunity.
Additionally, an alternative proposal presented by the potentially responsible party,
ARCO, was determined by the EP A and State to be an acceptable remedy for storage and
. .
treatment of flows up to the 100-year flood. It obviates the need for the upstream
impoundment and it offers the additional advantages of improved treatm~nt of dissolved
metals in flood waters and keeping contaminants within the existing boundaries of the
operable unit. This detailed proposal is part of the administrative record for the site, and
is referenced in the Proposed Plan. The EP A has determined these changes are
significant; however, a revised Proposed Plan or renewed public comment period is not
required. Guidance (OSWER Directive 9335.3-02)12 states:
"If the significant change to a component of the alternative could have been reasonably
anticipated by the public, the lead agency need only document the significant change in the
Decision Summary". In this instance, a majority of the public requested the change, fully
120SWER Directive 933S.J..02, N~mber. 1989, EPAfS40fG-89fOO7
-------
aware that the elimination of the upstream impoundment would necessitate routing of
flood flows into the pond system.
The decision to defer certain aspects of the cleanup does not significantly change those
aspects of cleanup which are selected in this Record of Decision. Therefore, there is no
need to submit a revise~ Proposed Plan to address this decision.
. .
Finally, it should be recognized that these changes are the product of a constructive
dialogue with both the public and the ..potentially responsible . party, which retains..
. .
ownership of the ponds and has exterisive experience in operating them as an effective
water treatment facility. The selected remedy formulated and evaluated in this Record of
Decision blends the remedial action plans of the regulatory agencies and the potentially
responsible party, and it is supported by the public.
-------
AITACHMENT TO PART II
APPUCABLE OR RELEVANT AND APPROPRIATE
REQUIREMENTS, STANDARDS, CONTROLS, CRITERIA, OR UMITATIONS
FOR THE SILVER BOW CREEK/BUTfE AREA SUPERFUND ,SITE -
ORIGINAL PORTION - WARM SPRINGS PONDS OPERABLE UNIT
Section 121(d) of CERCLA, 42 U.S.C. Section 9621(d), certain provisions of the current
National Contingency Plan (the NCP), 40 CFR Part 300 (1990);-and guidance and policy
issued by the Environmental Protection Ag~ncy (EP A) require that remedial actions taken,
pursuant to Superfund authority shall comply with substantive provisions of applicable or
relevant and appropriate "standards, requirements, criteria, or limitations from State
environmental and facility siting laws, and from federal environmental laws at the
completion of the remedial action, and/or during the implementation of the remedial a,ction,
unless a waiver is granted. These requirements are threshold standards that any selected
remedy must meet. The Feasibility Study for the Warm Springs Ponds operable unit
proposed a set of such requirements, and gave justification for identifying the proposed'
requirements. After consideration of public comments on the proposed requirements, and
further review of applicable guidance and standards, including the NCP, the following is the
fmallist of ARARs for the Warm Springs Ponds operable unit Record of Decision. '
Each ARAR or group of related ARARs identified here is followed by a specific statutory
or regulatory citation, a classification describing whether the ARAR is a federal or State
requirement and whether the ARAR is applicable or relevant and appropriate, and a
compliance description which addresses how and when compliance with the ARAR will be
measured (some ARARs will govern the conduct of the implementation of the remedial
action, some will govern the measure of success of the remedial action, and some will do
both). Contaminant specific ARARs are followed by a' description of the point of
compliance, which describes where compliance with the ARAR will be'measured.
Also contained in this list are policies, guidance or other sources of information which are
"to be considered" during the selection and implementation of the ROD. Although not
enforceable requirements, these documents are important sources of information which EP A
and the State of Montana Department of Health and Environmental Sciences (MDHES)
referred to during selection of the remedy, especially in regard to the evaluation of public
health and environmental risks; or which will be referred to as appropriate during evaluation
and approval of various activities during the ROD implementation.
,Finally, this list contains other legal provisions or requirements which should be complied
with during the implementation of this ROD.
-------
. Responses to comments on the proposed ARARs and further discussion of EP A's basis for
selecting these ARARs is co~tained in the responsiveness summary attached to this ROD.
The portions of the Feasibility Study (FS) which address ARARs (primarily Chapter 3 and
Appendix B) and the ARARs section of the responsiveness .summary, and applicable EP A
guidance, policy, regulation; and statutory authority, form the basis for the final selection of
ARARs contained in this list. . ..
ARARs are divided into contaminant specific, location specific, and action specific
requirements, as described in the new NCP and EP A guidance. Each category contalns both
federal and State ARARs. For contaminant specific ARARs, ARARs are listed .according
to the appropriate media.
Contaminant specific ARARs include those. laws and regulations' governing the release to.
the environment of materials possessing certain chemical or physical characteristics or
containing specific chemical compounds. Contaminant specific ARARs generally set health
or risk based numerical values or methodologies which, when applied to site-specific
conditions, result in the establishment of numerical values. . These values establish the
acceptable amount or concentration of a chemical that may be found in, or discharged to,
the ambient environment. Location specific ARARs are restrictions placed on the
concentration of hazardous substances or the conduct of cleanup activities because they are
in specific locations. Location specific ARARs relate to the geographic or physical position
of the site, rather than to the nature of the site contaminants. Action specific ARARs are
usually technology or activity based requirements or limitations on actions taken with respect
to hazardous substances.
For action specific ARARs, certain provisions pertain to the entire cleanup action and are
so indicated. Other ARARs pertain to specific portions of the cleanup, and are so
indicated.
Many requirements listed here are promulgated as identical or near identical requirements
in both federal and State law, usually pursuant to delegated environmental programs
administered by EPA and the States, such as the requirements of the federal Clean Water
Act and the Montana Water Quality Act. Tbe preamble to the new NCP states that such
a situation results in citation to the State provision as the more stringent standard, but
treatment of the provision as a federal requirement.
-------
].
Contaminant Specific
1.
Groundwater
,A Arsenic . 0.02 milligrams per liter (mg/l)
B. Cadmium 0.010 mg/l ,
C. Chromium 0.050 mg/l
'.'
D. Lead 0.050 mg/l
E. Mercury 0.0002 mg/l.
. ,
F. Nitrate (as N) 10.000 mg/l
G. Selenium 0.010 mg/l
H. Silver 0.05 mg/l
POINT OF COMPLIANCE: The standards identified in 1.1.A - H. above must be
complied with at the down gradient edge of the ground water capture trench, located belo J'
Pond 1, if the waste below Pond 1 is excavated. If the ROD is amended to include
construction of a wetland at Pond 1 rather than dry closure, the point of compliance would
be at the downgradient edge of the wetland and/or capture trench. No mixing zone will be
used in determining compliance.
CITATION: The standards identified .in I.1.A - H. above are promulgated at
Administrative Rules of Montana (ARM) Sections 16.20.1003, .203, .204, .206, and .207, and
are known as Maximum Contaminate Limits (MCLs). Classification of the contaminated
shallow aquifer within the Warm Springs Ponds operable unit as a Class II aquifer suitable
for future drinking water or domestic use was done pursuant to ARM sS~ 16.20.1002. Such
a determination is consistent with the classification criteria found in EP A's "Guidance on
Remedial Actions for Contaminated Ground Water at Superfund Sites", December, 1988,
- (OSWER Directive # 9283.1-2). These standards were promulgated pursuant to the'
authority given in the Montana Public Water Supplies Act and the Groundwater Pollution
Control Act. Corresponding federal citations for the federally authorized and delegated
program are the Safe Drinking Water Act, 42 U.S.C. sS~ss~ 300£, et seQ., and at
40 CFR sS~ss~ 141.11- 16 (MCLs). The standards for arsenic, cadmium, lead, and silver are
also found at 40 CFR sS~ 264.94, pursuant to the Resource Conversation and Recovery Act,
42 U.S.C. sS~ss~ 6901 et seQ. .
-------
. CLASSIFICATION: Federal, relevant. and appropriate
COMPUANCE: At the conclusion of the remedial action and continuing thereafter.
I. Contamination of ground water is prohibited. Ground water wells must be constructed
and maintained so as to prevent waste, contamination, or pollution of ground water.
POINT OF COMPUANCE: At the location of any ground water well located on-site.
CITATION: Promulgated at MCA sS~ 85-2-505, a provision of the Montana Water Use Act.
Only. those provisions of section 505 described above are identified as applicable to this
action. . .
CLASSIFICATION: State, applicable "
COMPLIANCE: During the implementation of the remedial action, construction of any
ground water wells must comply with this standard. .
J. Non-degradation. Any ground water whose existing quality is higher than the established
ground water quality standards for its classification must be maintained at that high quality,
unless it has been affirmatively demonstrated that such a change is justifiable. and will not
preclude present or anticipated use of such waters. Compliance with the standards
identified in I.1.A - H. above will achieve compliance with this standard.
POINT OF COMPUANCE: Because compliance with the MCL standards will achieve
compliance with this standard, the standard must be complied with at the down gradient
edge of the ground water capture trench, located below Pond 1.
CITATION: Promulgated at ARM sS~ 16.20.10il
CLASSIFICATION: State, applicable.
COMPLIANCE: At the conclusion of the remedial action and continuing thereafter.
2. Air
A Lead - No person shall cause or contribute to concentrations of lead in the ambient air
which exceed 1.5 micrograms per cubic meter (mg/ cm) of air, measured over a 90-
day average.
POINT OF COMPUANCE: Within the confines of the Warm Springs Ponds operable unit,
where human exposure is probable.
-------
.CITATION: Promulgated at ARM sS~ 16.8.818 as part of a federally approved State
Implementation Plan (SIP), pursuant to the Clean Air Act of Montana, MCA 75-2-101 ~
~. Corresponding federal regulations are found at 40 CFR sS~ 50.12, promulgated
pursuant to section 109 of the Clean Air Act, 42 U.S.C sS~. 7409. . .
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action, and at the conclusion
of the action and thereafter. Compliance shall be measured in accordance with .methods
described in 40 CFR Part 50, and corresponding State provisions.
B. Particulate matter that is 10 microns in diameter or smaller (PM - 10) - No person shall
cause or contribute to concentrations of PM - 10 in the ambient air which exceed:
- 150 micrograms per cubic meter of air, 24 hour average, no more than one expeCted
exceedence per calendar year.
- 50 micrograms per cubic meter of air, annual average.
POINT OF COMPLIANCE: Within the confines of the Warm Springs Ponds operable unit,
where human exposure is probable.
CITATION: Promulgated at ARM sS~ 16.8.821 as part of a federally approved SIP,
pursuant to the Clean Air Act of Montana, MCA 75-2-101 et seQ.. Corresponding federal
regulations are found at 40 CFR sS~ 50.6, promulgated pursuant to section 109 of the Clean
Air Act, 42 U.S.C. sS~ 7409.
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action, and at the conclusion
of the action and thereafter. Compliance shall be measured in accordance with methods
described in 40 CFR Part 50, and corresponding State provisions.
C. Airborne particulate matter - Construction must not be undertaken unless reasonable
precautions are taken to control emissions of airborne particulate matter.
POINT OF COMPLIANCE: At the construction activity.
CITATION: Promulgated at ARM 16.8.1401(4), pursuant to the Clean Air Act of Montana.
MCA 75-2-101 et seQ.. These regulations were promulgated pursuant to an approved State
Implementation Plan pursuant to section 110 of the Clean Air Act, 42 U.S.C. ss~ 7410.
CLASSIFICATION: Federal, applicable.
-------
-COMPLIANCE: During tbe implementation of the remedial action. Compliance shall be
measured in accordance with methods described in 40 CPR Part 50, and corresponding State
provisions. .
D. Opacity - Emissions of airborne particulate matter from .any stationary source shall not
exhibit any opacity of 20 percent or greater averaged over six consecutive minutes.
POINT OF COMPUANCE: At the source of emission.
CITATION: Promulgated at ARM 16.8.1401(4), pursuant to the Clean Air Act of Montana,
MCA 75-2-101 et seQ.. These regulations were promulgated pursuant to an approved State
Implementation Plan pursuant to section 110 of the Clean Air Act, 42 V.S.C. sS~ 7410.
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action. Compliance shall be
measured in accordance with methods described in 40 CFR Part 50, and corresponding State.
provisions. .'
E. Road dust suppression - Construction activity must employ reasonable measures to .
control road dust.
POINT OF COMPLIANCE: At the construction activity.
CITATION: Promulgated at ARM 16.8.1401(3), pursuant to the Clean Air Act of Montana,
MCA 75-2-101 et seQ.. These regulations were promulgated pursuant to an approved State
Implementation Plan pursuant to section 110 of the Clean Air Act, 42 U.S.C. sS~ 7410.
CLASSIFICATION: Federal, applicable.
COMPUANCE:D1,lring the implementation of the remedial action.
F. Settled particulate matter - No person shall cause or contribute to concentrations of
particulate matter in the ambient air such that the mass of settled particulate matter exceeds
10 grams per square meter, 30-day average.
POINT OF COMPLIANCE: Within the confmes of the Warm Springs Ponds operable unit,
where human exposure is probable.'
CITATION: Promulgated at ARM sS~ss 16.8.818, pursuant to the Clean Air Act of
Montana, MCA 75-2-101 et seQ..
CLASSIFICATION: State, applicable.
-------
'COMPUANCE: During the implementation of the remedial action, and at the conclusio
and thereafter.
G. Occupational Health and Safety Standards. No worker. shall be exposed to:
. Arsenic
0.5 micrograms per cubic meter
(mg/m3)
10.0 ug/m3
1.0 mg/m3
0.15 mg/m3
5.0 mg/m3
0.2 mg/m3
0.01 mgfm3
'. 0.2 mg/m3, 8 hour time weigh~ed
average
0.1 mg/m3 acceptable ceiling
250 millions of particulates ..per
cubic foot of air
10 mg/m3
15 mppcf
5 mg/m3
50 mppcf
15 mg/m3
Inorganic Arsenic
Copper dusts
. . Lead
Manganese
Selenium compounds
Silver
CadIIlium dust
Mercury
Silica-crystalline quartz
Inert or nuisance dust
. Total dust
POINT OF COMPUANCE: At the worker.
CITATION: Promulgated at 29 CPR sS~ss~ 1910.1000, 1910.1018(c), and 1910.1025(c),
pursuant to the Occupational ~afety and Health Act, 29 V.S.C. sS~ss~ 651 - 678, except for
those standards marked with a ., which are promulgated at ARM sS~ 16.42.102, pursuant
to the Occupational Health of Montana, MCA sS~ 50-70-113.
CLASSIFICATION: Federal, applicable, except for. standards, which are State, relevant
and appropriate.
COMPUANCE: During implementation of the remedial action.
H. Generators of air pollution must achieve and maintain such levels of air quality as
will protect human health and safety, to the greatest extent practicable.
Point of compliance: within the Warm Springs Ponds operable unit, where human exposure
is pro~able.
-------
. Citation: MCA ~ 75-2-102, pursuant to the Clean Air Act of Montana.
Classification: State, applicable
Compliance: During implementation of the remedial action and thereafter. Compliance
with the numeric standards listed will achieve compliance with this standard.
3. Surface Water - Ambient.
Chronic (mg/l) Acute (mg/l)
-.'
A Arsenic (III) 0.36 0.19
B. Arsenic (V) 0.85 0.048
C. Arsenic (Total) 2.2
D. Cadmium .0.0039* 0.0011 *
E. Copper 0.018* 0.012*
F. Iron 1.0
G. Lead 0.082* 0.0032 *
H. Mercury 144.0 ng/l 0.000012
1. Selenium 0.28 0.036.
J. Silver 0.0041 * 0.00012 *
K. Zinc 0.12* 0.11*
* indicates an assumption of 100 mg/l hardness. H the average hardness can be
demonstrated to occur at a different level within the Ponds for this compliance point, or
within the receding stream, for the ambient water compliance point. These standards will
be adjusted appropriately.
L Dissolved oxygen concentrations may not be reduced below 7.0 mg/l from October 1
through June 1, nor below 6.0 mg/l from June 2 through September 30.
M. Induced variation of pH within the range of 6.5 to 9.0 must be less than 0.5 pH unit.
Natural pH outside this range must be maintained without change. Natural pH above 7.0
must be maintained above 7.0 .
N. The maximum allowable increase above naturally occurring turbidity is 10 nephelometric
turbidity units except for short-term construction or hydraulic projects, game fish population
restoration, as permitted in ARM ss~ 16.20.633.
O. A 1 degree F maximum increase above naturally occurring water temperature is allowed
within the range of 32 degrees to 66 degrees F; within the naturally occurring range of 66
degrees F to 66.5 degrees F, no discharge is allowed which will cause the water temperature
to exceed 67 degrees F; and where the naturally occurring water temperature is 66.5 degrees
F or greater~ the maximum allowable increase in water temperature is. 0.5 degrees F. a 2
-------
,degree F-per-hour maximum decrease below naturally occurring water temperature is
allowed when the water temperature is above 55 degrees F, and a 2 degree F maximum
decrease below naturally occurring water temperature is allowed within the range of 55
degrees F to 32 degrees F.
P. No increases are allowed above naturally occurring concentrations of sediment, settleable
solids, oils, or floating solids which will or are likely to create a nuisance or render the
waters harmful, detrimental, or injurious to public health, recreation, safety, welfare,
livestock, wild animals, birds, or other wildlife. '
Q. True color must not be increased more than 5 units above naturally occurring color.
POINT OF COMPLIANCE: These standa,ds must be met at the beginning of the Clark, ,
Fork River, that is just above the confluence of Warm Springs Creek and just past the' Mill-
Willow Bypass.' , ,
CITATION: These standards are promulgated at ARM sS~ 16.20.622(2), pursuant t() the
Montana Water Quality Act, MCA sS~ss~ 75-5-101 et seq.. These standards are based upon
the designation of the upper reaches of the Clark Fork River as a C-2 class river in ARM
ss~ 16.20.604(1)(e), as further described in ARM ss~ 16.20.622, where designated uses are
described. These standards were developed by EP A pursuant to section 304 of the Clean
Water Act, 42 U.S.c. ss~ 1314, and are published in the "Gold Book", Water OuaHty Criteria
for Water 1986, EPA 44/5-86-001 (May 1, 1986). The State has enacted them as applicable"
requirements pursuant to delegated authority found section 303 of the Clean Water Act, 42
U.S.C. ss~ 1313.
CLASSIFICATION: Federal, applicable (by virtue of being promulgated State water quality
standards which are directly applicable to the river bodies).
COMPLIANCE: Upon the completion of the remedial action, and thereafter.
ARAR WAIVER: The standards for arsenic (total) and mercury described above cannot
be achieved using currently available technology. Pursuant to section 121(d)(4)(A) and(C),
42 V.S.C. ss~ 9621(d)(4)(A)and (C), EPA is waiving compliance with these standards. The
standards would be replaced by the following:
Arsenic (total) 0.02 mg/l
Mercury 0.2 ug/l
The standards identified for arsenic and mercury are below detection limits and cannot be
achieved using curren1;:- available technology, and are waived due to the interim nature of
this action. The repla~':, lent standards are based on the detection limits for mercury, and
the' non-degradation standard for arsenic.
-------
. S. The State's non-degradation requirements require that sources of pollution do not
degradate existing high quality water. Compliance with the specific criteria identified above
will achieve compliance with this provision. . ..
POINT OF COMPLIANCE: These standards must be met at the bevnning of the Clark
Fork River, that is just above the confluence of Warm Springs Creek and just past the Mill- .
Willow Bypass.
CITATION: Promulgated at ARM sS~ 16.20.702, pursuant to the Montana Water. Quality
Act, MCA sS~ 75-5-303. ..
ClASSIFICATION: State, relevant and appropriate.
..
COMPLIANCE: Upon the completion of the remedial action, and thereafter.
4. Point Source Discharge
A Because the discharge from Pond 2 which is expected to remain after completion of the
remedial action will enter the Clark Fork River, the water quality standards identified above
in I. 3. A - K., including the ARAR waiver replacement standards, are identified as the
appropriate numeric limitations for the point source at Pond 2 which will remain after
completion of the remedial action.
POINT OF COMPLIANCE: At the point of discharge. These numeric standards must be
met when stream flows equal or exceed the minimum consecutive 7-day average flow, which
may be expected to occur on the average of once in ten years. Special note - Because this
discharge is a pre-existing permitted discharge, which has been subject to a MPDES water
quality permit for several years, continuing the application and permit process would ensure
continuity within the State's program. Although not required by CERCIA, permit for this
discharge must continue to be applied for (as part of the remedial design process) and
received. The permit should be consistent with the standards stated in section 1.4. of this
ARARs list, and any other standards determined to be applicable through the permit
process.
CITATION: Promulgated at ARM.ss~ 16.20.622, pursuant to the Montana Water Quality
Act, MCA sS~ 75-5-101 et seQ.. Point source standards are required in section 402(a) of the
Clean Water Act, 42 V.S.C. sS~ 1342(a).
CLASSIFICATION: Federal, applicable.
COMPUANCE: At the conclusion of the remedial action and thereafter. No mixing zone
will be applied to measure compliance with these requirements.
-------
r)
. B. A maximum pH standard of 9.5, as contained in the current MPDES permit for th
pond discharges, is identified as applicable to the point source discharge from Pond 2 whic
will remain after this remedial action. " "
POINT OF COMPLIANCE: At the point of discharge. These numeric standards must be
met when stream flows equal or exceed the minimum consecutive 7-day average flow, which
may be expected to occur on the average of once in ten years.
CITATION: Promulgated at ARM sS~ 16.20.622, pursuant to the Montana Water"Quality
Act, MCA sS~ 75-5-101 et seQ.. Point source standards are required in section 402(a) of the
Clean Water Act, 42 V.S.C. sS~ 1342(a).
ClASSIFICATION: Federal, applicable. ..
COMPLIANCE: At the conclusion of the remedial action and thereafter, as provided in
the water quality permit which will be required for the point source discharge.
C. Monitoring and best management practices described at
40 CPR sS~ 440.104 are required for point source discharge from Pond 2 which wilI"remain
after this remedial action (See action ARARs).
ClASSIFICATION: Federal, applicable.
COMPLIANCE: At the conclusion of the remedial action and thereafter, as provided in the
water quality permit which will be required for the point source discharge.
n. Location Specific
1. Structures such as parks and wildlife management areas are permitted within floodplains.
CITATION: Promulgated at MCA sS~ 75-5-402, as part of the Floodplain and Floodway
Management Act.
CLASSIFICATION: State, applicable.
COMPLIANCE: During the implementation of the remedial action and thereafter.
2. Water conversation and flood control projects, including projects for conversation,
recreation and wildlife protection, streamflow stabilization, and pollutant abatement are
permitted. These may include dikes, embankments, impounding reservoirs, and other
watercourse improvements.
CITATION: Promulgated at MCA ssss~~ 75-5-1101 and 1102, as part of the Floodplain and
Floodway Management Act.
-------
.' CLASSIFICATION: State, relevant and appropriate.
COMPUANCE: During the implementation of the remedial action and thereafter. Only
those substantive provisions described above are identified as ARARs for this action.
3. Flood control works are permitted if they are, protective to the 100 year flood frequency
flow.
CITATION: ARM sS~ 36.15.606, pursuant to the Floodplain and Floodway Management
Act.
>.
CLASSIFICATION: State, relevant and appropriate.
COMPUANCE: Compliance with these provision has been achieved or will be achieved
through construction of the' Mill':'Willow Bypass and associated berm in accordance with the
approved Work Plan for that effort. Only those substantive provisions described above are
identified as ARARs for this action.
4. Wildlife management and natural areas are permitted and encouraged uses within a
floodplain.
CITATION: ARM sS~ .36.15.801, pursuant to the Floodplain and Floodway Management
Act.
CLASSIFICATION: State, applicable (substantive provisions only).
COMPUANCE: During the implementation of the remedial action and thereafter. Only
those substantive provisions described above are identified as ARARs for this action.
5. Soil erosion and sedimentation to Montana natural rivers musty be kept to a minimum.
CITATION: MCA sS~ 75-7-102, of the Natural Streambed and Land Preservation Act of
1975.
CLASSIFICATION: State, applicable (substantive provisi,ons only).
COMPLIANCE: During the implementation of the remedial action and thereafter. Only
those substantive provisions described above are identified as ARARs for this action.
6. The rainbow bridge within Pond 3 has been identified as eligible for the Register of
Historic Places. The selected remedy may have adverse effects on the bridge. Accordingly.
the following mitigation measures are required during the conduct of the remedial action.
-------
The bridge must be photographed and. recorded, according to state regulations. Additional
measures, such as establishments of a roadside display, may be identified during the
remedial design phase. . . .
CITATION: 40 CFR sS~ 6.301(b) and 40 CFR Part 800, pursuant to the National Historic
Preservation Act, 16 V.S.C. sS~ 470 et seQ.. .'
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action.
7. If significant scientific, prehistorical, historic, or archaeological data is found at the Warm
Springs Ponds operable unit, it must be pre~erved in an appropri'ate'manner. To date, no.
such data has been found at the operable unit. However, if such data is discovered,' this
ARAR.must be complied With. '.
CITATION: 40CFR sS~ 6.301(c), pursuant to the Archaeological and Historic Preserv~tion
Act, 16 V.S.C. sS~ 469. .
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action.
8. A pair of bald eagles has been identified as nesting near the Warm Springs Ponds
operable unit. The appropriate mitigative measures to be followed during the conduct of
the remedial action are:
Continued consultation with the V.S. Fish and Wildlife Service to determine mitigative
measures regarding on-site construction.
CITATION: 50 CFR Parts 17 and 402, 40 CFR sS~ 302(h), pursuant to the Endangered.
Species Act, 16 V.S.C. sS~ 1531.
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action.
9. Modification of the Mill-Willow Bypass and Silver Bow Creek must provide for adequate
protection of fish and wildlife resources. The specific requirements. for this ARAR were
incorporated into the Work Plan for the Mill-Willow Bypass removal.
CITATION: 40 CFR sS~ 6.302(g), pursuant to 16 V.S.C. sS~ss~ 661 et seQ..
CLASSIFICATION: Federal, applicable.
-------
COMPliANCE: Compliance with these provision has been achieved or will be achieved
through construction of the Mill-Willow Bypass and associated berm in accordance with the
approved Work Plan for that effort. Only those substantive provisions described above are
identified as ARARs for this action. ' '
10. The activities described in the ROD will minimize potential harm to or within the
floodplain and improves the natural and beneficial values of the floodplain.
CITATION: 40 CFR sS~ 6.302(b) and Executive Order on Floodplain Management, No.
11,988. ','
-,'
CLASSIFICATION: Federal, applicable.
..
COMPUANCE: During the implementation of the remedial action.
described in this ROD will comply with this ARAR.
The activities
11. The activities described in the ROD have avoided, to the extent possible, adverse
impacts to existing wetlands within the Warm Springs Ponds operable unit, and avoid
construction in wetlands if practicable.
CITATION: 40 CFR 6.302(a) and 40 CFR Part 6, Appendix A, Executive Order No.
11,990.
CLASSIFICATION: Federal, applicable.
COMPUANCE: During the implementation of the remedial action.
12. The Pond '1, 2 and 3 disposal facilities must be designed, constructed, operated, ,and
maintained to avoid washout, because they are located next to the 100 year flood plain for
Silver Bow, Mill, and Willow Creeks, and the Clark Fork River.
CITATION: ARM sS~ 16.44.702, pursuant to the Montana Hazardous Waste Management
Act. Corresponding federal regulations of this federally authorized and delegated program
are found at 40 CFR sS~ 264.18(a) and (b), pursuant to the Resource Conversation and
Recovery Act, as amended, 42 V.S.C. sS~ss~ 6901 et seq..
CLASSIFICATION: Federal, relevant and appropriate.
COMPLIANCE: At the completion of the remedial action and thereafter.
ill. Action Specific
1. General ARARs
-------
A During construction at the site, and afterwards, standards governing the protection of
.. occupational health and safety must be complied with. These include the establishment of
health and safety programs and practices for on-site workers, and the provision of protective
equipment, should conditions warrant. .
Full requirements are contained in the cited provisions.
CITATION: 29 CFR Part 1926, 20 CPR ss~ss~ 1910.12Q, and 1910.132, promulgated
pursuant to the Occupational Health and Safety Act, 29 V.S.C. ss~ss~ 651 - 678.
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action,. and thereafter if
jurisdictional requirements are met for conditions existing after the completion of the.
remedial action. . .
B. Every employer must provide a safe place of employment, provide safety devices and.
safeguards and use practices, and enSure that operations and processes. are reason~bly
adequate to render the place of employment safe. The employer must also do every other
thing reasonably necessary to protect he life and safety of its employees.
CITATION: MCA sS~ 50-71-201, of the Montana Safety Act.
CLASSIFICATION: State, applicable.
COMPLIANCE: During the implementation of the remedial and thereafter, if the
jurisdictional prerequisites of the statute are met.
C. Each employer must maintain at the work place a list of chemical names of each
chemical in the work place, and indicate the work area where the chemical is stored or
used.: If any Material Safety Data Sheets exist for the chemicals, they must be kept at the
work place. Employees must be informed of the chemicals at the work place, and trained
in the proper use of the chemicals.
CITATION: MCA ss~ss~ 50-78-202, 203, 204, and 307, promulgated pursuant to the
Employee and Community Hazardous Chemical Information Act.
CLASSIFICATION: State, applicable.
COMPLIANCE: During the implementation of the remedial and thereafter, if the
jurisdictional prerequisites of the statute are met.
-------
. 2. Cleanup and reconstruction of the Mill-Willow Bypass.
Part of the remedial action involves cleanup and reconstruction of the Mill-Willow Bypass.
ARARs for that action were previously identified as an attachment to the Administrative
Order on Consent governing that early action. Those ARARs are hereby incorporated by
reference. In addition, the following ARARs are identified for the continued cleanup and
reconstruction activities taking place in the Mill-Willow Bypass.
A At this time, no additional mitigative measures are identified to achieve compliance with
the dredge and fill requirements of the Clean Water Act. Consultation with the Corps of
Engineers is continuing, and requirements may be identified during the implementation of
the remedial action, including the cleanup and reconstruction of the Mill-Willow Bypass.
.. .
CITATION: 40 CFR Parts 230, 231 (substantive j>rovisions only), 33 CFR Parts 323 and
330 (substantive provisions only), pursuant to section 404 of the Clean Water Act, 42 U.S.c.
sS~ 1344,
CLASSIFICATION: Federal, applicable.
COMPLIANCE: During the implementation of the remedial action.
B. Reclaimed drainages must be designed to emphasize channel and floodplain dimensions
that will blend with the undisturbed drainage above and below the area to be reclaimed.
The channel must be restored to its natural habitat or characteristic pattern with a
geomorphically acceptable gradient. The drainage must safely pass through a 24-hour
precipitation event with a 100-year recurrence interval. Reclamation must provide for long-
term stability of the landscape, establishment or restoration of the stream to include a
diversity of aquatic habitats (generally a series of rimes and pools), ~d restoration
enhancements, or maintenance of natural riparian vegetation.
CITATION: ARM sS~ 26.4.634, promulgated pursuant to the Montana Strip and
Underground Mine Reclamation Act, MCA 82-4-101 et seq..
CLASSIFICATION: State, relevant and appropriate.
COMPLIANCE: At the conclusion of the remedial action.
C. Temporary diversion structures at the Bypass or on Silver Bow Creek must be
constructed to safely pass the peak run-off from a precipitation event with a 10-year, 24-hour
recurrence interval. Channel lining must be designed using standard engineering practices
such a riprap, to safely pass designed velocity. Free board must be no less than 0.3 feet.
CITATION: ARM sS~ 26.4.636, promulgated pursuant to the Montana Strip and
Underground Mine Reclamation Act, MCA 82-4-101 et seq..
-------
'CLASSIFICATION: State, relevant and appropriate.
COMPLIANCE: At the conclusion of the remedial action.
D. Disturbed areas which will remain above high flow levels along the Mill-Willow Bypass
must comply with general revegetation requirements described in the following section.
3. General revegetation requirements. The remedial action will involve excavation of some
contaminated material into disposal facilities, covering some contaminated areas with clean
soil, and creating two disposal facilities. The following requirements are ARARs for those
activities. These requirements are not ARARs for contaminated areas which will be flooded
or made into or maintained as a wetland. . -.
A The disposal units anc~ other revegetated areas will be capped with clean soil and
revegetated in an appropriate manner, consistent with the Timber Butte removal action and
its accompanying work plan.
CITATION: 30 CFR sS~ 816.111, promulgated pursuant to the Surface Mining Control and
Reclamation Act, 30 U.S.C. sS~ss~ 1201 - 1326.
CLASSIFICATION: Federal, relevant and appropriate.
COMPLIANCE: At the conclusion of the remedial action, and thereafter.
B. Revegetation of the disposal units, the excavated areas, and the covered, contaminated
areas must meet the substantive standards of the regulations cited below.
CITATION: ARM sS~ss~ 26.4.501, .501(a), .505, .520, .631, .633, .638, .644,.703, .711, .713,
.714, .716, .718, .719, .721, .724, .726, .727, .728, .729, .730, .751, and .761, all of which are
promulgated pursuant to the Montana Strip and Underground Mine Reclamation Act, MCA
sS~ss~ 82-4-101 et seQ-.. These standards provide the specific method to ensure compliance
with ~ 82-4-231 and 82-4-233 of the MSUMRC.
CLASSIFICATION: State, relevant and appropriate.
COMPLIANCE: At the completion of the remedial action, and thereafter.
4. Continued operation of Ponds 1 and 2.
Ponds 2 and 3 will continue to function as contaminant capture and treatment surface
impoundments until water quality standards and other ARARs are achieved upstream, and
. contaminated soils and waste are stabilized or removed from the floodplain. Until that
time, the operation of Ponds 2 and 3 must comply with the following ARARs:
-------
A The structural integrity of the Ponds must comply with the provisions cited below, to
prevent overtopping and other problems. The operation and maintenance should provide
for regular inspection and maintenance of the Ponds. .
CITATION: Certain provisions (only substantive provision which incorporate 40 CFR sS~ss~
264.221(£), (g), (h), and .226) of ARM sS~ss~ 161.44.701 - 703, which are promulgated
pursuant to the Montana Hazardous Waste Management Act. Corresponding federal
regulations for this authorized and delegated program are found at 40 CFR sS~ss~ 264.221(£),
(g), (h), and .226, which are promulgated pursuant to the Resource Conversation and
Recovery Act, as amended, 42 U.S;c. sS~ss~ 6901 et seQ.. .
CLASSIFICATION: Federal, relevant and appropriate.
. .
COMPLIANCE: After completion of the remedial action, and thereafter.
5. Berm Strength
The berms contained within the Warm Springs Ponds operable unit are considered dams
and/or reservoirs, pursuant to the Montana Dam Safety Act. Further, the dams have been
classified as high hazard dams, pursuant to ARM sS~ 36.14.202. All berms within the
operable unit must comply with the following ARARs.
A All dams and reservoirs which divert or store water must be constructed in a secure,
thorough, and substantial and safe manner.
CITATION: MCA ~~ 85-15-207 and 208, a provision of the Montana Dam Safety Act.
CLASSIFICATION: State, applicable.
COMPUANCE: At the conclusion of the remedial action and thereafter.
B. All high hazard cams must comply with the criteria given in the provision cited below,
including compliance with the Maximum Credible Earthquake standards.
CITATION: ARM Sl~ 36.14.501, which is promulgated pursuant to the Montana Dam
Safety Act.
CLASSIFICATION: State, applicable.
COMPUANCE: At the conclusion of the remedial action and thereafter.
C. All high hazard dams must be able to safely pass the flood calculated from the inflow
design flood. In this situation, all berms within the operable unit must be able to safely
manage 0.5 Probable Maximum Flood (PMF).
-------
CITATION: ARM sS~ 36.14.502, which is promulgated pursuant to the Montana Dam
Safety Act.
CLASSIFICATION: State, applicable.
COMPUANCE: At the conclusion of the remedial action and thereafter.
6. Closure and post closure care of the two disposal facilities.
Pond 1 and the upland disposal facility above Pond 3 (created as a result of the removal .
action at the Mill-Willow Bypass) will be used to permanently dispose of contaminated soils
and sediments, and tailings. These disposal facilities must be closed and cared for according
to the following ARARS: '." '. .
. .
A All waste of disposed within the facilities must be drained of free liquids, and stabilized
appropriately.
CITATION: Certain portio~ of ARM sS~ 16.44.702 (namely, that portion which
incorporates 40 CFR sS~ 264.228(a) which addresses the standard described above),
promulgated pursuant to the Montana Hazardous Waste Management Act. Corresponding
federal regulations for this authorized and delegated program are found at 40 CFR sS~ss~
264.228(a), which are promulgated pursuant to the Resource Conversation and Recovery
Act, as amended, 42 ').S.C. sS~ss~ 6901 et seQ..
CLASSIFICATION: Federal, relevant and appropriate.
COMPUANCE: At the completion of the remedial action and thereafter.
B. Closure must be done in such a manner as to minimize the need for further maintenance
and to control, minimize or eliminate, to the extent necessary to protect public health and
the environment, post-closure escape of hazardous wastes, hazardous constituents, leachate,
contaminated run-off or hazardous waste decomposition products to the ground water or
surface waters or to the atmosphere. This ARAR does not require an impermeable cap or
liners.
CITATION: Certain portions of ARM sS~ 16.44.702 (namely, that portion which
incorporates 40 CFR Sl~ 264.111 . which addresses the standard described above),
promulgated pursuant to the Montana Hazardous Waste Management Act. Corresponding
federal regulations for this authorized and delegated program are found at 40 CFR sS~ss~
264.111, which are promulgated pursuant to the Resource Conversation and Recovery Act.
as amended, 42 V.S.C. SI~SI~ 6901 et seQ..
CLASSIFICATION: Federal, relevant and appropriate.
-------
COMPLIANCE: At the completion of the remedial action and thereafter.
C. The disposal facility cover for each unit must function with minimum maintenance,
promote drainage and minimize erosion or abrasion of the. final cover, and accommodate
settling and subsidence. .
CITATION: Certain portions of ARM sS~ 16.44.702 (namely, that portion which
incorporates 40 CFR sS~ 264.228(b)(c) and (d) which address the standards described
above), promulgated pursuant to the Montana Hazardous Waste Management Act.
Corresponding federal regulations for this authorized and delegated program are found at
40 CFR sS~ss~ 264.228(b)(c) and (d), which are promulgated pursuaIit to the Resource
Conversation and Recovery Act, as amended, 42 V.S.C. sS~ss~ 6901 et seq..
CLASSIFICATION: Fede~al, relevant and appropriate.
COMPLIANCE: At the completion of the remedial ac~ion and thereafter.
D. The owner of the disposal facilities must submit to the local land use or zoning authority
a survey plat indicating the location and dimensions of waste disposed of in each unit.
Additionally, the owner must record a deed restriction, in accordance with State law, that'
will in perpetuity notify potential purchasers that the property has been used for waste
disposal and that its use is restricted.
"
CITATION: Certain portions of ARM sS~ 16.44.702 (namely, that portion which
incorporates 40 CFR sS~ 264.116 and 119 which address the standards described above),
promulgated pursuant to the Montana Hazardous Waste Management Act. Corresponding
federal regulations for this authorized and delegated program are found at 40 CFR sS~ss~
264.116 and .119, which are promulgated pursuant to the Resource Conversation and
Recovery Act, as amended, 42 V.S.C. sS~ss~ 6901 et seq..
CLASSIFICATION:. Federal, relevant and appropriate.
COMPLIANCE: At the completion of the remedial action and' thereafter.
E. A private party's solid waste may be disposed of on property belonging to the private
party, unless such disposal creates a nuisance or public health hazard.
CITATION: MCA sS~ 75-10-214, which is part of the Montana Solid Waste Management
Act.
CLASSIFICATION: State, applicable.
COMPLIANCE: At the conclusion of the remedial action or thereafter.
-------
, 'F. Solid waste must be disposed of outside of the 100 year flood plain, must be dispose
of in a manner which prevents pollution of the ground or surface water, contain adequate
drainage structures, and prevent run-off from entering disposal areas. Solid waste must be
transported to the area in such a manner as to prevent its discharge~ dumping, spillage, or
leaking. '
CITATION: Certain provisions of ARM sS~ss~ 16.14.505 and 523, as described above, which
are promulgated pursuant to the Montana Solid Waste Management Act. Corresponding
federal regulations are found' in specific portions of 40 CFR Part 257, whic,h was
promulgated pursuant to the Resource Conversation and Recovery Act, as amended, 42
u.S.€. sS~ss~ 6901 et seq.. " "
CLASSIFICATION: Federal, relevant and appropriate.
COMPUANCE: At the conclusion of the remedial action and thereafter.
WAIVER: Hazardous substances will be left within the Pond berms. EP A has examined
the Solid Waste Management Act, and believes the area within the berms, after berm
strengthening activities, is not within the 100 year flood plain of Silver Bow, Mill, or Willow
Creeks, or the Clark Fork River. Therefore, EPA believes that this action is in compliance
with this ARAR.
However, if it is determined that the materials are within the 100 year flood plain, a waiven/.
of this ARAR is appropriate, pursuant to section 121(D)(4)(A) of CERCLA, as this action
is an interim action.
7. Ground water monitoring.
The ongoing waste management units and waste disposal units at the site must be monitored
for compliance with ground water ARARs described in section 1.1. above. The monitoring
system for this site must comply with the following ARARs:
A The monitoring system must comply with the provision cited below, for detection of
those contaminants identified in section 1.1. above only. The monitoring system can treat
the collection of Ponds and disposal units as one consolidated unit.
CITATION: Certain portions of ARM sS~ 16.44.702 (namely, those portions which
incorporate 40 CFR s'~ 264.97), which is promulgated pursuant to the Montana Hazardous
Waste Management Act. Corresponding federal regulations for. this authorized and
delegated program are found at 40 CFR sS~ss 264.97, which is promulgated pursuant to the
Resource Conversation and Recovery Act, as amended, 42 U.S.C. sS~s'A 6901 et seq.. In
addition, compliance with this requirement will also achieve compliance with ARM A 16.20-
1016.
-------
"CLASSIFICATION: Federal, relevant and appropriate.
COMPLIANCE: Upon installation of the ground water monitoring system.
8. Operation of Pond 2.
Pond 2 will continue to be a point source discharge to the Clark Fork River. Numeric
standards for that discharge are identified in section 1.4. above. As previously stated,
because the discharge is a preexisting discharge, the operator of Pond 2 (ARCO) must
obtain an MPDES permit for the point source discharge from Pond 2. Tbat permit must
ensure compliance with the following ARARs, at a minimum. ""
A The discharge must be monitored in "compli~nce with the "provision cited below, to"
ensure compliance with th~ standards "identified in section 1.4 above.""
CITATION: ARM sS~ 26.20.904, promulgated pursuant to the Montana Water Quality Act.
Corresponding federal regulations for this authorized and delegated program are found at
40 CPR sS~ 122.41.
CLASSIFICATION: Federal, applicable.
COMPUANCE: At the conclusion of the remedial action and thereafter.
B. The Pond must be managed using Best Management Practices, to ensure compliance
with the standards identified in section 1.4. above.
CITATION: ARM ss~ 26.20.904, promulgated pursuant to the Montana Water Quality Act.
Corresponding federal regulations for this authorized and delegated program are found at
40 CPR sSfi 125.100.
CLASSIFICATION: Federal, applicable.
COMPUANCE: ~t the conclusion of the remedial action and thereafter.
C. Ponds such as the Warm Springs Ponds must be operated and maintained so as to
prevent pollution of surface waters above the numeric standards identified in section 1.4
above.
CITATION: ARM sSfi 16.20.633 & 75-6-112(2) &75-5-605, promulgated pursuant to the
Montana Water Quality Act.
CLASSIFICATION: State, applicable.
COMPUANCE: At the conclusion of the remedial action and thereafter.
" "
-------
POLICIES, GUIDANCE, ADVISORIES, CRITERIA OR OTHER INFORMATION TO B
. CONSIDERED
Identification of policies, guidance, advisories, criteria, or otller information which does not
rise to the level of ARARs by the lead agency is authorized in 40 CFR ~ 300.400(g)(3).
TBCs are to be used as appropriate in developing Superfund remedies. As the Preamble
to the final NCP states, TBCs may be useful in helping to determine what is protective at
a site, or how to carry out certain actions or requirements. 55 FR 8744-8745.
Accordingly, the following list is divided into those TBCs which were used by EPA and the
State-.in considering and evaluating human health and environmental risks posed by the site,
and those that will be used by EP A and the State as it continues' to implement or monitor
implementation of the ROD. . .
I.
TBC used in evaluating risks at the site.
Agency of Toxic Substance and Disease Registry (ATSDR). . 1988. Draft, toxicological
profile for lead. U.S. Public Health Service, Atlanta, GA.
EPA, 1986. Guidelines for the health risk assessment of chemical mixtures. Federal
Register 51(185):34014-34025.
EPA, 1986. Superfund public health evaluation manual. EPA 540/1-86/060, Office of .
Emergency and Remedial Response, Washington, D.C.
EP A, 1987. Final, Superfund exposure assessment manual. Office of Emergency and
Remedial Response, Washington, D.C.
EPA, 1988. Final, Superfund exposure assessment manual. Office of Emergency and
Remedial Response, Washington, D.C. . .
EP A,' 1988. Final, Superfund exposure assessment manual. Office of Emergency and
Remedial Response, Washington, D.C. (OSWER Dir. # 9285.5-1)
EP A, 1988. Integrated risk information system. Office of Research and Development,
Cincinnati, OH. .
EP A, '1989. Second quarter FY 89 health effects assessment summary tables.
Environmental Criteria and Assessment Office, OERR 9200.6-303-(89-1). Cincinnati,OH.
EPA, 1989. Regulating Lead: an update. AWWA J. 81(7): 24.
-------
. Epa, 1989. Evaluation of the potential carcinogenicity of lead and lead compounds in
support of reportable quantity adjustments pursuant to CERCLA section 102. EP A/600/8-
89/045A, Office of Health and Environmental Assessment, Washington, D.C.
EP A, September, 1989. Interim Guidance on Establishing Soil Lead Cleanup Levels at
Superfund Sites. OSWER Dir. #9355.4-02. .
Recommended Agency Policy on the Carcinogenicity Risk Associated with the Ingestion of
Inorganic Arsenic, June 21, 1988, Lee Thomas, EPA Administrator.
SpeQal Report on Ingested Inorganic Arsenic: Skin Cancer; Nutritional Essentially (EP A,
1988).. ..
Interim Final Guidance for Soil Ingestion Rates (EP A, 1989; OSWER Dir. # 9850.4).
Supplement to Interim Guidance on Establishing Soil Lead Cleanup Levels at Superfund
Sites, (EPA, 1990; OSWER Dir. #9355.4-02A)
Risk Assessment Guidance (EP A, 1990)
Interim Final Environmental Evaluation Manual, (EPA, 1990; OSWER Dir. # 9285.7-01);
otherwise known as the Risk Assessment Guidance for Superfund - Environmental
Evaluation Manual. .
. EP A's Proposed Drinking Water Standard for maximum Concentration Limits for Copper
and lead, 53 FR 31516 (August 18, 1988). .
. EPA's Proposed MCLG levels for cadmium, mercury, and selenium. 54 Fed. Reg. 22,062,
May 22, 1989).
. State of Montana's Ambient Air Guidelines for Non-Criteria Air Pollutants.
-These standards are not promulgated as of the date of this ROD, and therefore are not
ARARs. Because existing ARARs for these contaminants, taken from State water quality
standards or the State's Clean Air Act are protective of human health and the environment,
these standards were not chosen for this action. EP A and the State reserve the right to use
these standards at other sites Within the Clark Fork Basin or the State. of mQntana to
achieve full protection of human health and the environment.
II. TBCs to evaluate the conduct of the remedial action. These TBCs are not mandatory,
and will be used only as guidance as appropriate by EP A as it reviews the remedial design
and .remedial action activities.
EP A's RCRA Design Guidelines for Surface Impoundments.
-------
. EPA's RCRA Permit Writer's Guidance Manual for Hazardous Waste Land Treatment
Storage, and Disposal Facilities.
EP A's RCRA Technical Resource Document for Closure. of HaZardous Waste Surface
Impoundments.
EPA's NPDES Guidance Document on NPDES Best Management Practices (June 1981);
EP A's Guidance on Remedial Action for Contaminated Ground Water at Superfund Sites,
OSWER Dir. # 9283.1-2, December, 1988. . .
* *The State of Montana general inspection and reporting re.quirements for dam and
reservoir construction and mine revegetation found at MCA ~ 85-15-211, 85-15-213, ~5-15'; .
310,82-4-237, and 82-14-237; and AR~t ~ 36.14.601, .602, .603, 26.4.305, .307, .309, .320~.and
.1129. .
* *The State of Montana's general permit application requirements for dam construction and
mine revegetation, found at MCA ~ 85-15-212 and 84-4-222, and ARM ~36.14.301, .303,
.305, .305, .306, .308, .402 - 407, .503, 26.4.305, .307, .320, and 36.15.216, and .801, and
36.2.404, and 26.4.311 - 315.
"The State of Montana's bond assurance and liability protection provisions found at MCA
~ 82-4-223, and ARM ~ 36.14.309, .311, and 26.4.1102, .1119, and 1125.
"These requirements are part of the State's environmental laws, and are. administrative
requirements, as defined by the NCP. They therefore are not ARARs, and are not
mandatory requirements to be followed at the site during remedial action. Nevertheless,
EP A recognizes that such provisions may be useful to the State and EP A as it reviews and
approves of various deliverables during the remedial design and remedial action Superfund
process, such as remedial design plans, remedial action plans, operation and maintenance
plans, and financial assurance submittals.
OTHER POTENTIALLY RELEVANT LAWS
The ARARs process is the exclusive process for applying federal or State environmental or
siting laws to a Superfund cleanup. However, the State of montana has identified a non-
comprehensive list of other State laws which may impact the conduct of the remedial action.
Those laws are:
Noise levels for protection of on-site workers, found at ARM ~ 16.42.101.
Ground water well and monitoring well drillers must be licensed and registered as stated
in ARM ~.36.21.402, .403, .405, .406, .411, .701, and .703.
-------
. Ground water wells must be logged and reported to the Department of Natural Resources
Conservation, as statedin MCA ~ 85-2-516.
Water rights must be protected as stated in MCA ~ 85-2-301, 85-2-306, 85-2-311, 85-2-402,
. 75-7-104, 87-5-506, and ARM ~ 36.16.104 - .106, and 26.4.648.
-------
RECORD OF DECISION
PART III: THE RESPONSIVENESS SUMMARY
Silver Bow Creek/Butte Area NPL Site..
Warm Springs Ponds Operable Unit
Upper Clark Fork ~iver Basin, Montana
United States Environmental Protection Agency
September 1990
-------
PART A - PUBLIC COMMENTS
1.0
RESPONSES TO PUBLIC COMMENTS, AN OVERVIEW.
This Responsiveness Summary for the Warm Springs Ponds Operable Unit of the Silver
Bow Creek Site was prepared to document and respond to the issues and comments raised
by the public regarding the feasibility study (FS) and the Proposed Plan for the operable
unit.
A remedial investigation (RI) and a public health and environmental assessment (PHEA)
for the operable unit have been completed. The RI and the PHEA examined the human
health and environmental risks posed by the operable unit.
The FS developed a set of remedial alternatives representing a range of approaches to
protect human health and the environment from the risks identified in the RI and PHEA.
This range of alternatives was presented to the public by the release of the feasibility study
report. The Proposed Plan detailing the remedial approach favored by the Montana
Department of Health and Environmental Sciences (MDHES) and EP A, was released at
the same time as the FS. While the FS was being developed, ARCa prepared a
feasibility-level study of its own proposal, identified as Alternative 3A ARCa presented
this alternative to the agencies and the public at numerous meetings and public
presentations. The agencies have conducted a focused technical review of ARCO's
proposal (CH2M HTI.L 1990). That review is part of the administrative record.
The public comment period for the FS and Proposed Plan began in November 1989 with
the release of the two documents. The comment period was extended once and ran un t i I
the end of January 1990. Public reaction, as expressed at the public meetings and in
-------
The public comment period for the FS and Proposed Plan began in November 1989 with
the release of the two documents. The comment period was extended once and ran until
the end of January 1990. Public reaction, as eXpressed at the public meetings and in
written comments, included many questions about how the preferred alternative would
work, . concerns about the impacts of the remediation, suggestions for modifying the
Proposed Plan, and requests that the remediation begin as soon as possible. State and
local agencies also responded to the. Proposed Plan. Their comments included additional
,
questions and suggested modifications to the Proposed Plan.
Several concepts came up repeatedly in the public's comments. Many comment~rs
suggested that the contaminated sediments in the ponds (approximately 19 million cubic
yards) would have to be removed from the floodplain before the cleanup could be
considered a permanent remediation. Several commenters pointed out that one feature of .
the Proposed Plan, diverting the flows in Mill and Willow Creeks into the pond system for
treatment, would have negative impacts on the fisheries in those creeks and the upper
Clark Fork River. Many commenters expressed concerns about the possibility of .
constructing a settling basin in the location considered in the FS. Several commenters
expressed concerns that the Proposed Plan would not do enough to provide treatment for
the contaminated water in Silver Bow Creek, and several of these stated that the goal
should be to treat all flows in Silver Bow Creek, up to the flows of a 100-year flood, to
meet the aquatic criteria at all times. The selected remedy addresses all of these concerns
as addressed below.
ARCO submitted as comments on the FS its own Plan 3A and detailed technical
comments on the MDHES FS Report. ARCO's comments on the FS were extensive on
almost every section, and repeatedly argued for the alternate proposal they had submitted.
The Responsiveness Summary contains EPA's and MDHES' responses to comments
received from the public. Addressed are oral comments received at the public hearings.
written comments from individual citizens and private organizations and written comment,
-------
from various government entities. Because many of the comments addressed similar
issues, the comments were consolidated and summarized. A listing of each commenter is
included as Attachment ill-A to this Responsiveness Summary. Also included is a cross-
reference of individual conimenters against the summarized comments that are responded
to. By utilizing this cross-reference, each commenter should be able to locate his/her
comment and the agency response.
The'" comments received from ARCa were voluminous and comprehensive. They
addressed each individual section of the F~. point by point. Accordingly, the responses to .
ARCa's comments are separated from the public comments and the responses follow the
format of ARCa's comments.
All comments, whether the public's or ARca'S, were considered fully, and adjustments to
the Proposed Plan were made in response to the public comments and ARCa comments.
-------
.2.0
RESPONSES TO PUBLIC COMMENTS
2.1
General Comments
2.1.1 Overall Remediation Approach
The agencies received numerous comments and recommendations that. dealt with.
. .
the relationship between the Warm Springs Ponds (WSP) remediation and the
other Clark Fork River Superfund sites and operable unitS, and with the goal~ and.
objectives of WSP remediation. Several commenters (Letters 1, 56, 65, 91, 126, 127,
129, 144, 154, 157) made general requests that vigorous efforts be made toward
cleaning up the Clark Fork River. Three commenters (Letters 101, 108, 126) noted
that clear goals and objectives should be developed for the entire Clark Fork
Superfund site, that water quality and health risks should be controlling factors in
the goals, that the Warm Springs Ponds proposals must be evaluated with respect to
these overall goals, and that dividing the Clark Fork sites into subunits appears to
have fostered uneven progress. The same commenters added that the overall site
needs more coordination. Numerous other commenters (Letters 4, 45, 55, 57, 68,
84, 85, 86, 89, 92, 101, 105, 107, 108, 111, 119, 124, 139, 143, 151; Testimony A-2,
A-9, A-12, A-IS, A-16, B-6, B-I0, M-3, M-9, M-I0) also recommended that the
agencies begin cleanup activities at the sources upstream of the WSP.
Response: .MDHES and the U.S. Environmental Protection Agency (EPA) are
aggressively pursuing cleanup of the Clark Fork River Superfund sites. The overall
site cleanup is coordinated by EP A The agencies' strategy for the cleanup work is
described in the Clark Fork Superfund Master Plan. The most recent revision of
the Master Plan is scheduled for final release in October 1990. Because of the size
and extent of problems in the Clark Fork Basin, it was necessary that the site be
broken into smaller units for study.
-------
The agencies decided to put the Warm Springs Ponds cleanup ahead of the
upstream Sil~er Bow Creek operable units because of ~he potential for catastrophic
failure of the pond berms during floods or earthquakes. A catastrophic failure
would result in a release to the Clark Fork River of at least a portion of the 19
million cubic yards of tailings and sludges currently in the ponds. This type of
failure would result in significant damage to the Clark Fork River. This sequencing
will make it necessary to readdress the ultimate disposition of the ponds when
, upstream Silver Bow Creek is remediated, but iniiial action on the ponds could not
be delayed.
Cleanup at other parts of the Silver Bow Creek Site and at other sites in the Clark
Fork Basin is being moved along as fast as the Superfund process will allow.
Warm Springs Ponds is just one of the 25 operable units that require study and
cleanup in the Clark Fork Basin. These are all complex units that require action,
and it will take time to address all of them. Additional information on the
schedule for t~e site-wide cleanup is in the Clark Fork Superfund Master Plan.
One of the overall goals for cleanup of the Basin is to achieve water quality
standards for the surface-water bodies within the Basin. The selecteq remedy will
achieve this goal for water entering the Clark Fork River from Silver Bow, Mill,
and Willow Creeks, until upstream cleanup actions result in water quality standard
compliance in those streams.
One cominenter (Letter 151) stated that MDHES and EPA should recognize the
need for a phased response at the Warm Springs Ponds, in which the ponds can
serve on an interim basis as a treatment system for Silver Bow Creek' until the
. .
upstream sources are cleaned up. Once that is achieved, a final remedy could then
be chosen for Warm Springs Ponds.
-------
Response: The goals and objectives for the cleanup of the Warm Springs Ponds
are described in detail in Chapters 3 and 5 of the FS and in the ROD. Protection
of human health and meeting ambient water quality standards in the upper Clark
Fork River are two of the remedial objecnves.
The cleanup at Warm Springs Ponds is being phased in a manner consi~tent with
that requested by the commenter. Remediation is progressing rapidly with action at
", the Mill-Willow "Bypass under way during the 1990 construction season. Tailings
are being removed from the bypass, "and th(: western berms of the ponds are being""
" "
" strengthened to withstand the maximum credible earthquake (MCE) and design
floods as part of this action. The cleanup of the remainder of the remaining Warm
Springs Ponds area will follow the Mill-Willow Bypass removal.
"The ROD for this operable unit is an interim ROD. The selected remedies will
necessarily "remain in place until such time that the upstream contamination sources
and depositions along Silver Bow Creek are remediated and there is no longer a
need to treat Silver Bow Creek waters. At that time the ultimate disposition of the
Warm Springs Ponds will be determined.
One comment (Letter 129) suggested that the agencies keep working on increments
that show action at the site, such as the removal and renovation of the Mill-Willow
Bypass and the experimental reseeding of streamside tailings.
Response: Action will continue in these areas. As indicated above removal of the
tailings in the Mill-Willow Bypass and strengthening of the western berms in the
bypass are underway during this construction season." Work on developing methods
to revegetate streamside tailings is continuing under the Streamside Tailings and
Revegetation Study. It is the goal of MDHES and EP A to continue to move the
cleanup of these sites along as rapidly as possible.
-------
Several commenters (Letters 24, 27, 29, 41, 44, 55, 57; Testimony A-2, A-16) .
suggested that Superfund remediation efforts should be concerned more with
impacts on people than with impacts on fish and wildlife.
Response: Remediation strategies must address the impacts of site contamination
on people and impacts on the, environment. The remedial investigation and
feasibility study dealt with bqth. The selected remedy is thought to be protective of
both public health and the environment.
2.1.2 Remediation Schedule.
';
Many commenters (Letters 48, 56, 68, 70, 91, 96, 97, 98, 100, '109, 112, 114,,116,
118, 119, 122, 123, 125, 131, 132, 134, 135, 139, 142, 146, 147, 149, 151, 153, 154,
155, 159, 161; Testimony B-2, B-4, M-1, M-5, M-10) recommended that work should
start as soon as possible, and no later than the construction season of 1990. , One
commenter (Letter 139) recommended that the Mill-Willow Bypass receive
immediate attention to prevent a fishkill in 1990. One commenter (Letter 151)
recommended specifically that the upgrading of the pond treatment system, berm
stabilization, and removal of tailings from the Mill-Willow bypass be segregated
from development of flood control alternatives and that work should begin on the
upgrading, stabilization, and removal activities this construction season. The
commenter further recommended that this work should proceed during 1990 under
a unilateral order if negotiations with ARCO are unable to produce an acceptable
consent order, and that the stabilization and removal activities should utilize the
most conservative and protective design criteria. Two commenters (Letters 139,
154) thought that construction of improvements upstream of the ponds should begin
immediately.
Response: MDHES and EPA agree that as much work as possible should begin
during the 1990 construction season. A consent order was signed by EP A and
-------
ARca in July- to allow for removal of tailings from the Mill-Willow Bypass an
reinforcement of the western berms of Ponds 2 and 3 for earthquake and flood
protection. It was decided by the agencies that only that amount of work could be
reasonably completed by the end of the 1990 construction season. By the end of
this season, however, any threat of fishkills being caused by tailings in the Mill-
Willow Bypass will be virtually eliminated and the potential for catastrophic failure
of the Ponds due to floods and earthquakes will be substantially reduced.
The agencies intend to expedite and phase the remedial' design of this project in ,
order that construction actiVities can continue smoothly during the '1991
construction season. It is presently expected that the remaining berm
improvements for earthquake and flood protection and the inlet/outlet structure
and treatment improvements for Pond 3 will be undertaken at that time. The
specific schedule of the future remediation activities will be determined in the
remedial design. As discussed in detail. in the following section, the public will be
kept informed about all elements of the proposed remediation.
The schedule for remediation of contamination deposition along Silver Bow Creek
and in Butte will proceed as described in the Clark Fork Superfund Master Plan.
While the agencies agree that remediation upstream of the ponds should begin as
soon as possible, there remains the need to continue evaluation of alternative
remediation approaches before plans for upstream remediation can be finalized.
Those activities, which are proceeding, are outlined in the Clark Fork Superfund
Master Plan.
Another commenter (Letter 119) stated that design assumptions concerning the
prediction that the tailings along Silver Bow Creek and problems at the Berke ley
Pit will be cleaned up in 30 years appear to be unrealistic, since it took over 100
years to create the situation. The commenter further suggested that the agencie~
address the possibility of stretching out the cleanup in the operable unit up tl
-------
100 years, and providing flood protection for the ponds during the interim, based on
risks of failures and acceptable occurrences of floods routing around the ponds
without treatment. Another commenter (Letter 93) suggested that the agencies
concentrate on developing an adequate solution over a long period.
Response: There is no reason that it should take as long to clean, up the
, ,
contamination as was spent creating it. ' It is the goal of both MDHES aIld EP A to ,
<> have these sources of contamination remediated within the 30-year time frame.
With respect to the stretching out th~ Warm Springs Ponds cleanup over 100 years, "
the agencies believe that a more immediate approach to the hazards presented by ,
the ponds is necessary. Leaving the berms unprotected, or only partially protected
for several decades, while the upstream contaminants are remediated, is not an
acceptable approach.
2.1.3, Interim vs. Permanent Remedies
\
~
Numerous commenters (Letters 53, 64, 65~ 68, 69, 73, 74, 75, 78, 89, 92, 96, 97, 98,
100, 101, 106, 107, 108, 109, 112, 115, 116, 117, 118, 123, 124, 126, 128, 129, 130,
131, 133, 135, 136, 138, 140, 142, 143, 146, 147, 149, 151, 152, 159; Testimony A-15,
M-5, M-7, M-8, M-9, M-10, M-11, M-13, M-15) stated that both MDHES/EPA's
and ARCQ's proposals for remediation of the Warm Springs Ponds are interim
remedies and that, although these remedies are needed now, permanent cleanup up
of the sites between Butte and Warm Springs Ponds are needed before making a
final decision on Warm Springs Ponds.
Response: MDHES and EPA recognize that the current proposal for Warm
Springs Ponds is an interim remedy. The final decision on the cleanup of the ponds
2 and 3 will be delayed at least 5 years. At that time, the degree of cleanup on
. Silver Bow Creek will be assessed and, if the cleanup has progressed far enough,
alternatives for the final disposition of the ponds will be presented to the public.
-------
All studies to date have indicated that permanent treatment or total removal of the
pond sediments are not likely to be the most desirable permanent solution.
CERCLA requires EPA to rely on treatment of wastes to. reduce their toxicity,
mobility, and volume whenever practicable in order to achieve permanent reme-
dies. However, the RI/FS guidance document recognizes that permanent treatment
solutions may not be practicable for high volume waste sites such as mining sites.
The volumes of wastes can be so large that treatment of the wastes is not feasible.
". An in situ treatment-based alternative was developed in tbe FS to allow. the public
and the agencies to gauge the costs Qf treatJIlent-based approaches for the volumes. .
of wastes that exist at Warm Springs Ponds. The results indicate that even "for in
situ treatment, which is often less expensive and quicker than treatment approaches
requiring excavation, the time and costs involved are not reasonable. In short, the
prospects for a treatment-based permanent solution to the wastes in the ponds are
not good, even several years from now once the upstream areas have been
remediated. However, if at that time new treatment processes have become
I
available that offer some promise for treating the nearly 19 million cubic yards ot\
wastes at this operable unit, those options can be explored.
The future potential for moving the sediments to another location is also not
promising. EP A and ARCa are presently conducting a study to site a RCRA-
equivalent waste repository facility in the Anaconda area. However, the technical
difficulties of trying to remove 19 million cubic yards of contaminated materials and
safely transport them to a disposal facility is daunting. Such a repository facility for
these wastes would also require continuing operation, maintenance, and monitoring,
and would be a continuing threat to groundwater. In addition, such a facility would
be no more permanent than an upgraded pond system. It may well be that an
.
upgraded pond system is the safest, most cost-effective, and environmentally sound
permanent remediation of the site.
-------
Another commenter (Letter 119) thought that the remediation should be "for.
posterity." This means that the proposed structures should be. free from hydrologic
and geologic hazards and should be maintenance free for aUeast 1,000 years. The
commenter stated that if the sediments are left in the pond system, the criteria
. .
could not be met. He also noted that the measures proposed in the FS would
require periodic maintenance to ensure that they continue to provide adequate
protection.
Response: EP A and MDHES reco~nize that maintenance wijl be required on the. .
berms and their as~ociated flood protection. Budgetary costs for operation. and
maintenance shown in Chapter 8 of the Feasibility Study include allowances for
berm and flood protection maintenance. Maintenance of the various structures will
likely be the responsibility of ARCa. Requirements for maintenance will be
included in the Record of Decision, and will be more fully developed in the
remedial design/remedial action (RD /RA) . phase. It would be prohibitively
expensive to design and construct engineering structures that would last 1,000 years
without maintenance. Regardless of the final disposition of the Warm Springs
Ponds sediments, some maintenance will be required. Even if the sediments are re-
moved and disposed of outside the floodplain, some maintenance on items such as
. contaminant berms, liners, caps, landscaping, etc., will be required.
Three commenters (Letters 101, 108, 126) stated that provisions must be clearly
spelled out concerning what happens after remediation, if the standards are
violated, and who will pay the bills for any necessary additional remedial measures.
Six commenters (Letters 11, 20, 34, 39, 43, 86) asked about liability insurance to
require that the cleanup is done correctly. ane commenter (Testimony B-4) noted
that ARCa will remain liable even after a remedy is in place. Another commenter
(Testimony M-13) suggested that ARCa be required to establish a trust fund to
cover the costs of future improvements to the remedial action that may be required.
-------
Response: The provisions of CERCLA (the Superfund Act) are very specific i
terms of financial liability. In ihis case, ARCa is responsible for paying for all -
necessary remedial actions, now and in the future.
Mter the issuance of the Record of Decision, a monitoring plan will be developed.
The provisions of the monitoring and compliance plan will be very specifi~. in terms
of locations, parameters, types of analyses, standards to be met, and. reporting
~ requirements. In addition, penalties can be imposed by EP A for violations of the
compliance requirements.
, It should also be noted that the provisions of CERCLA require periodic reviews at
5-year intervals for remedial actions that leave wastes in place. These reviews will
examine in detail the performance of the' remediation in meeting the goals
established by the Record of Decision. If the goals are not being met, further
actions may be required in . the future and would be paid for by the responsible..-
party. In any enforcement action, the responsible party will be required to
demonstrate assurances of financial capability. Those assurances may include the
establishment of trust funds or bonds.
-------
2.1.4 Public Participation.
Numerous commenters (Letters 3, 12, 14, 15, 17, 19, 20, 23,26, 28, 34, 35, 36, 40,
..
49, 60, 76, 78, 81, 87, 89, 96, 100, 101, 105, 108, 110, 113, 118, 123, 126, 129, 131,
138, 139, 151, 159; Testimony A-2, A-5, A-6, A-7, A-8, A-12, A-14, A-15, B-4, B-8,
B-10, M-5, M;.7, M-8, M-I0, M-11, M-12, M-16) expressed a desire for better public
involvement arid greater cooperation among the agencies, local govenlIDental and.
.,. elected officials, and local citizens and public interest groups du.ring the Superfund
process. Some indicated that Anaco~da/Deer Lodge County should be involved in ..
the decision-making ..process. Many of these commenters formally requested. an
extension of the public comment period. Another commenter (Letter 151) stated
that the public participation process followed for the Warm Springs Ponds FS was
flawed because it did not give the public an opportunity to participate in the
scoping of alternatives, and recommended that the agencies use an approach similar
to that followed under the National Environmental Policy Act (NEPA). One
commenter (Letter 138) interpreted the CERClA guidelines to not allow public
agencies to disclose cleanup. alternatives to the public before they have been
approved and screened by EP A. One commenter (Letter 150) stated that the
agencies did a good job in keeping the public informed.
Response: A short description of the public involvement program for the Warm
Springs Ponds operable unit is provided in the ROD. Although the public
participation process followed for the WSP FS was in compliance with all
requirements of the National Contingency Plan (NCP), it is obvious that additional
efforts are needed to facilitate increased involvement of local citizens early in the
process. The agencies are striving to involve all appropriate parties and agencies in
future activities at Warm Springs Ponds and at other sites in the Clark Fork Basin.
-------
.-":.
Public involvement in the Mill-Willow Bypass Removal Action is representative Q
this effort. A scoping meeting on the Mill-Willow Bypass and other issues was held
on February 6, 1990, with participation by EPA, MDHES and other state agencies,
various city and county representatives, and public interest group representatives.
Public meetings were held at Fairmont and Missoula on February 27 and 28, 1990,
respectively, to gather input from the general public on the Mill-Willow Bypass
activities and other actions 'planned by the agencies and ARCO. ,Once the Mill-
Willow Bypass removal plans were more fully developed, three more public
meetings were held (in Anaconda, Deer Lodge, and' 'Missoula) in late May;,
, , ,
Numerous' coordination meetings involving local government officials,
representatives of interested state agencies (such as the Department of Fish,
Wildlife, and Parks and the MDHES Water Quality Bureau), and public interest
groups were held in preparation for this summer's removal action. Active efforts to.
involve the public will continue at Warm Springs Ponds and the other Clark Fork
Basin sites.
Although presentation of the alternatives and proposed plan are mandated by the
NCP to occur at the conclusion of the preparation of the Feasibility Study and
Proposed Plan, there is nothing in the guidelines to preclude public involvement at
an earlier stage. In fact, such involvement is encouraged. As discussed above, EP A
and MDHES are making strong efforts to increase' early public involvement.
Specific elements about how and when the public can be involved are made on a
site-by-site basis and are included in the site's Community Relations Plan. In all
cases, the public is involved when the Public Draft Feasibility Study and proposed
plan are completed, and public comment is taken and considered at tha~ time.
The public comment period for this project was extended for an additional one
month. The number of comments received is a good indication of the success of
extending the comment period.
-------
ARCO commented at the public hearings (Testimony A-3, B-3, M-6) that it had not.
been given the opportunity to be involved throughout the Warm Springs RIfFS
process. ather commenters (Letters 60, 87; Testimony B-8) stated that ARCa
should have been given the opportunity to participate.
Response: EP A and MDHES agree that responsible parties must be included in
the RIfFS process. ARCa has not been excluded from the CERCLA process at
,.. the Warm Springs Ponds. When CERCLA activities began. on the Silver Bow
Creek Site (Warm Springs Ponds are a part of this site), ARca was offered the
, .. . . .
opportunity to conduct the activities at the site. ARCa declined that offer. . As a
. . . . .
result, the agencies conducted the RIfFS studies at WSP. ARCa was given the
opportunity to comment on all studies conducted at the site and all documents
produced, and. has commented formally to the agencies on most of the site.
activities. Recently, ARCO has been more receptive to offers to conduct the
various Superfund activities under agency oversight and, in fact, is presently
conducting many studies and activities, under agency enforcement supervision, on
Clark Fork Superfund sites. The removal action at the Mill-Willow Bypass and the
berm improvements for earthquake and flood protection this summer are being
undertaken by ARCa.
One commenter (Letter 46) asked about the status of the proposal to bring an EP A
office to Butte?
Response: An EPA office has been set up in Butte and is located in the Butte.
Silver Bow City-County Building.
-------
2.1.5 MisceUaneous Genera! Comments.
Two commenters (Letters 52, 70) stated that all cOlltractors. and their employees
should be required to comply with 29 CFR Part 1910. Tbe same commenters
recommended that EP A and MDHES implement rules ensuring all contractors pay
the prevailing wages for work performed and that locally-trained pers?nnel be
employed to remediate the environmental hazards at the Warm Springs Ponds.
Response: 29 CFR Part 1910 specifi~~ req~ements for employee health and safety..
training and employee protection programs for work on hazardous waste sites. .It is
an Occupational Safety and Health Administration (OSHA) requirement. Since it
is expected that ARCO and its contractors will be conducting the remediation, .
ARCO will be required to have its employees and contractors meet this applicable
OSHA requirement.
If ARCO implements the remedy at Warm Springs Ponds, it will be up to ARCO
to decide which contractors will conduct and perform the work. ARCO's work
being performed under the Mill-Willow Bypass Removal this summer has utilized
local contractors and employees from local unions. If EP A implements the selected
remedy, then specific federal regulations regarding the hiring of contractors and
workers Will apply.
Another commenter (Letter 50) wondered why an out-of-state company (CH2M
HILL) was hired to drill the monitoring wells at Warm Springs Ponds for
approximately $1.4 million. Will they just leave the state after the work is done?
Wouldn't a Montana well driller be cheaper?
.Response: CH2M HILL was selected to conduct the remedial investigation and
feasibility studies (RIfFS) for the Silver Bow Creek Site, which includes the Warm
Springs Ponds, based on the results of the competitive procurement process
-------
conducted by MDHES. The cost for the work included many activities in addition
to drilling wells. These activities included sampling o~ soils, tailings, surface water
and groundwater, conducting treatability tests, developing a flood model, preparing
the remedial investigation, public health and endangerment assessment, and the FS
report.
Much of the work conducted as pari of this project was conductecl. by local
~ subcontractors. The driller used at Warm Springs Ponds. was O'Keefe Drilling of
Butte.
One commenter (Testimony A-B) stated that the FS needed to look at more
alternatives, and that the alternatives need to be evaluated in more detail.
Response: In response to the comments from the public and ARCO on the FS and
Proposed Plan, the agencies have evaluated and considered alternatives not
specifically addressed in detail in the FS. The selected remedy is a combination of
elements of several alternatives. The level of engineering detail in the FS is
consistent with that required to complete a full, feasibility-level evaluation of
alternatives. Detailed engineering analyses will be part of the remedial design
phase of this project.
Three commenters (Letters 46, 63, Testimony A-4) stated that too many studies had
been done already on the Silver Bow Creek and Clark Fork Basin Superfund sites,
and that these studies have been a waste of time and money.
Response: CERCLA requires that EP A investigate and develop remedial actions
that are protective of human health and the environment and are permanent and
cost effective to the extent possible. Responsible parties are required to pay for or
reimburse EPA for all cleanup investigations and actions. It would be
-------
.'
inappropriate to undertake large, multi-million dollar cleanup actions without
substantial detailed investigations to determine effective remediation approaches.
One comment (Letter 113) recommended Superfund funding be directed to
landowners for conservation efforts on their land. Another commenter (Letter 130)
thought that monies should be given to Dear Lodge County as compensation for
their lost economic potential: .
'.-
Response: Superfund monies. cannot be . used for payriieritsto landowne~s. . for' .
conservation efforts... Remedies for cleanup of existing contamination can' be
implemented on private, non-PRP lands, however. This work would be paid for but
either the party responsible for the contamination or by EP A Compensation for
lost economic potential of contaminated areas could not come from CERCLA
(Superfund) funds, but would have to be obtained in separate action from those
parties responsible for the contamination.
One comment (Letter 155) requested notice of the approved cleanup plan when it
was completed.
Response: The Record of Decision (ROD) states EP A's decision on the cleanup
methods to be implemented at Warm Springs Ponds. This has been prepared after
reviewing all public comments and reconsidering the various possible remediation
alternatives. A public notice will be issued regarding the availability of this ROD
and Responsiveness Summary to the public.
One cOII?JDenter (Letter 101) encouraged everyone to help the agencies to attract
and keep the quality of personnel and the commitment of resources needed to
move through the cleanup process.
. Response: MDHES and EP A agree.
-------
Three commenters (Letters 25, 26, 41) opposed the concept of removing topsoil
from pasture land for reclamation work at Warm Springs Ponds.
Response: The need for topsoil during the remediation is discussed in the FS.
Much of the topsoil required will largely come from areas within the operable unit.
It would not be reasonable to strip acres of pasture land of their topsoil to provide
the soil needed. Instead, poorer soils that can be amended to serve as, <;:over soil
", will be used wherever possible. Sources of suitable soils and' the amendments
necessary to make them work for the, ,intended purposes will be explored during the, ,
remedial design ph~~ of the project.
One commenter (Letter 48) expressed support for the ARCa berming project
downstream of the Warm Springs Ponds.
Response: The referenced berming work was done by ARCO under order from
MDHES in an effort to alleviate future fishkills in the upper Clark Fork River.
That work is being done downstr~am of the Warm Springs Ponds operable unit and
has therefore not been addressed in this FS or ROD.
One commenter (Letter 3) asked if the arsenic on Smelter Hill is being stored
temporarily or,permanently. Another commenter (Testimony A-I) suggested that
the agencies investigate beryllium sites in the Opportunity Ponds. Another
commenter (Letter 86) recommended keeping water in the Opportunity Ponds to
reduce dust.
Response: Cleanup studies at the Anaconda Smelter site, including the
Opportunity Ponds and Smelter Hill, are ongoing. No final cleanup decision has
been made to date. The problems of fugitive dust, groundwater contamination,
and beryllium disposal will be addressed as part of these, activities. While these
studies are ongoing, actions have been taken to reduce dust from the ponds. These
-------
activities have included putting a layer of limestone on the surface of the dry
portions of the ponds.
2.2
SITE CHARACfERlZATION AND PROBLEM DESCRIPTION
2.2.1 Groundwater
. .
-- One commenter (Letter 67) stated that the Superfund investigations of the Warm
Springs Ponds may underestimate.. the current amount of groundwater that..
discharges from the Warm Springs Ponds to the Mill-Willow Bypass and the Clark
Fork River and therefore minimize the importance of the contaminants this
groundwater contains. Based on the data presented in the Phase I and II Rem~dial
Investigation (RI) reports, groundwater contributes a substantial portion of the flow
and contaminant load to the river during low flow periods. Except for a trench,
which may intercept a fraction of the flow in one area, no remediation is planned.
Using data from the Phase I and Phase II RI reports, the commenter states that the
combined calculated groundwater inflow to the upper bypass from the east and west
is probably somewhat greater than 3.4 cfs . and that the inflow from the lower
bypass, though more difficult to estimate, may be close to 3.8 cfs. The total
groundwater discharge to the entire Mill-Willow Bypass could be as high as 7.5 cfs.
None of the Warm Springs Ponds studies recognize that magnitude of groundwater
discharge.
The comment also notes that contaminant loads in surface water increase though
the bypass and upper Clark Fork and that the most likely source of copper and zinc
contaminants is the groundwater plume downgradient of Pond 1. Groundwater in
this area has high sulfate and zinc conce~;trations.
-------
To summarize, contaminated groundwater emanating from the ponds discharges
metals and sulfate to the bypass and river and exacerbates the poor aquatic life
conditions in the river during low flow periods.
Response: Groundwater inflow rates were estimated for various reaches of the
Mill-Willow Bypass using both direct analytical calculations and empirical methods.
Because direct analytical calculations of groundwater inflow to the bYPass require.
, use of numerous assumptions, EP A and MDHES contend that inflow estimates
using empirical methods are more representative of site conditions.
Discharge was measured in the . upper Mill-Willow Bypass during July 1988, when
surface water in the Mill-Willow Bypass was diverted into the upper end of Pond
3. Measurements of flow below the diversion point were considered to be the most
accurate means of determining the rate of groundwater inflow to the bypass
channel, because, all water flowing in the bypass channel at the time was derived
from groundwater seepage. Discharge measurements completed at four locations
along the bypass below the point of diversion to midway along Pond 2 indicated the
total seepage rate to the Mill-Willow bypass was 2.57 cfs. The average gain in
surface flow between stations 55-18 and 55-25 during low flow measurements was
2.4 cfs. Based on these data, the combined rate of groundwater inflow to the
channel from the east and west was approximately 0.18 cfs per 1,000 feet of bypass
channel. Extrapolating these unit inflow rates to the entire bypass from 55-18 to
the northwest comer of. Pond 1 results in a total groundwater inflow rate to the
bypass of approximately 3.7 cfs. The agencies believe this inflow rate is con-
servative and is a much better characterization of site conditions than the 7.5 cfs
presented by the commenter.
The purpose in completing groundwater inflow calculations (both empirical and
analytical) for the Mill:- Willow Bypass was to provide reasonable estimates of inflo"
quantity for use during the F5 in evaluating construction of a groundwat~r
-------
interception drain along the entire reach of the bypass. Construction of this type of
interception drain was evaluated to determine the feasibility of: l) intercepting
groundwater emanating from the Warm Springs Ponds and the Opportunity Ponds
before the groundwater enters the Mill-Willow Bypass, and; 2) maintaining
groundwater levels at an elevation below the base of tailings located adjacent to the
. bypass.
-.' Subsequent data collected at the site indicated that groundwater inflow to the
bypass does not exceed maximum.. contaIDinant levels' or. Gold Book aquatic..
standards. Data to support this' conclusion were collected primarily in conjunction
with surface water sampling completed at three locations in the bypass channel
during July 1988, when Mill and Willow Creeks were diverted into Pond 3.
Samples collected at that time from sampling sites SS-l8C, SS-l8Cl, and SS-l8D
essentially represented groundwater seeping into the bypass channeL Analytical
results of those. samples indicated that all parameters analyzed were below Gold
Book standards. Because freshwater aquatic criteria are based on acid-soluble
analyses, use of dissolved cadmium concentrations ranging from 5.8 to 6.4 ~ g/l
measured at sampling stations SS-l8C, SS-l8Cl, and SS-l8D, in evaluating
. exceedances of chronic and acute water quality criteria is not appropriate. In
addition, concentrations of metals measured in samples obtained from monitoring
wells located adjacent to the Mill-Willow Bypass are less than Gold Book
standards. Because of these data, it became apparent that interception of
groundwater inflow to the bypass channel was unnecessary in meeting ARARs
established for the operable unit.
Groundwater inflow to the Clark Fork River between Pond 1 and Perkins Bridge
(SS-29) was not calculated or presented in the Phase II RI. The values cited by the
commenter as inflow to the bypass in this reach of the bypass (1.8 and 3.8 cfs) are
actually estimates of the groundwater flux in the shallow sand and gravel aquifer
beneath the Pond 1 berm. It is unknown what portion of the groundwater movi n ~
-------
beneath the Pond 1 berm actually surfaces in the Clark Fork. River south of Perkins
. .
Lane Bridge.
However, surface water quality data collected during the Phase I RI do not show a
meaSurable increase of copper and zinc at SS-29, as compared to upstream sites SS-
25, SS-28, and PS-12, even during low flow periods when the largest impacts from
groundwater inflows should be realized in. the stream. In addition, gr.oundwater
~ samples obtained from monitoring wells located downgradient .of Pond 1 near the
Clark Fork River exhibited copper and zinc concentrations well below that..
measured in the Clark Fork River.
The same commenter (Letter 67) notes that well completion logs for Wells WSP-
GW-17, 18S, 18D, 19S, and 19D and WSP-PW-01 are not in the Phase n RI, SQ a
complete analysis of the groundwater investigation could not be done.
Response: These logs were inadvertently omitted from the Phase n RI. Well
completion logs for these wells have been added to the agencies' response to
ARCQ's comments on the Phase n RI, which is part of the administrative record.
In addition, Letter 67 states that well development for most observations wells was
inadequate because the fmal water produced from the wells was not clear. Wells
completed in sands and gravels, such as those encountered in the area, can be
developed to produce clear water, but it can take longer than the 10 to 85 minutes
spent at each well. Measured trace-metal concentrations in turbid water samples
from wells that have been inadequately developed or purged may not be
representative of actual levels in groundwater.
Response: Well development following monitoring well installation using hand-lift
pumps, surge blocks, and bailers was the initial step in ensuring representative
formation water was obtained for laboratory analysis. Prior to obtaining a sampl~
-------
from each monitoring well during each sampling episode, water in the well was
evacuated until relatively clear, sand-free water was obtained. Well evacuation was
continued while the evacuated water was monitored .for field 'parameters including
temperature, specific conductivity, and pH. When measurements of these field
parameters were within 5 percent for three consecutive samples of the evacuation
water, the well was deemed ready for sampling. This process sometimes re~ulted in
an additional 1 to 1.5 hours of development time. for particularl~ turbid wells.
"
All samples collected for metals analysis were field-filtered with a O.45~ filter to. .
. .
remove any residual turbidity prior to preserving the samples with nitric acid.' This
procedure was consistent with the project Sampling and Analysis Plan and is
standard practice for preparing water samples for dissolved metals analysis.
The same commenter (Letter 67) describes the map of the extent of groundwater
contamination as incomplete because the boundaries of the plume either stop at the
boundaries of the operable unit or at the Mill-Willow Bypass and the Clark Fork
. River. The plume probably extends beyond these boundaries. These streams may
be a groundwater divide and, therefore, limit further migration of the plume as the
RI investiiations seem to assume, but this has not been documented. A complete
risk assessment cannot be done without knowledge of the full extent of the plume.
Response: The intent of Figure 2-19 was to provide the reader with a feel for the
. .
extent of groundwater contamination as related to iron, manganese, and sulfate
concentrations within and directly adjacent to the Warm Springs Ponds Operable
Unit. Separate studies of the Anaconda Smelter-Opportunity Ponds site and the
Clark Fork River are being conducted to characterize groundwater quality west of
the Mill-Willow Bypass and north of the Warm Springs Ponds, respectively.
Combining data from the three studies would likely indicate that the regional extent
of the. groundwater contaminant plume (associated with relatively high
concentrations of iron, manganese, and sulfate) is greater than that depicted on
-------
Figure 2-19. The risk assessment for the Warm Springs Ponds Operable Unit was
completed for the operable unit itself; future risk assessments will investigate other
components of the upper Clark Fork Basin CERCLA sites. .
Finally, Letter 67 notes that interconnections betWeen the shallow and deep
aquifers may be more significant than the RI reports indicate. Samples from nested
wells WSP-GW-7 and 15 along the bypass and WSP-GW-10 near Pond ~show that
, sulfate concentrations are higher in the deeper aquifer ~an ihe shallow aquifer.
Manganese is also present in the de~per aquifer's plume. .It appears that the plume..
in the deeper aquifer has not developed as fully as in the shallow aquifer, but 'given
sufficient time, it could.
/
Response: ARARs established for groundwater at the Warm Springs Ponds
Operable Unit include primary maximum contaminant levels. Sulfate and
manganese are not included in these standards and, as such, are not addressed in
remedial alte=-natives associated with the FS.
The occurrence of these parameters in the deeper aquifer in the vicinity of the
Mill-Willow Bypass and Pond 1 is consistent with the presence of ~elatively high
sulfate and manganese concentrations in groundwater systems in the vicinity of the
Opportunity :Ponds. This suggests that the distribution and occurrence of sulfate
and manganese in the groundwater environment is more regional and is probably a
result of multiple contaminant sources and pathways of contaminant movement.
2.2.2 Tailings and~ments
One commenter (Letter 72) states that the FS notes on page 2-30 that seve n
samples of tailings deposits were collected in the areas above Pond 3 and below
Pond 1 and tested for EP Toxicity. None of the samples failed the test. The
comment states that the fact that the samples did not fail the EP Toxicity test d()e~
-------
not adequately indicate their hazard to the environment. The comment suggests
that the samples should be tested by "bulk sediment and elutriate bioassay testing."
Response: The paragraph disCussed in this comment was not intended to imply
that the tailings and contaminated soils deposits do not. present a threat to the
environment simply because they do not fail the EP Toxicity test. It. .is fairly
.' standard in a remedial investigation to test various wastes by this test~ as one
.. indication of whether the hazardous waste management regulations should be
considered or followed in treating,. storing, or disposing' of the wastes. The',
'. .
.. indicated paragraph merely reports the results of testing the materials by this
standard test.
The two types of test suggested in the comment are not further described. It is not
clear what specific tests are being recommended or how the results from such tests
could be used to determine cleanup levels for tailings and soils. Numerous samples
of tailings have been analyzed for metals content. The resulting data have been
used in developing remediation alternatives for these materials.
One commenter (Letter 3) asked how the estim~te of 19 million cubic yards of
sediments in the ponds was made.
Response: The volume of pond bottom sediments (approximately 19 million cubic
yards), as presented in Table 2-2 of the FS, was calculated using data collected
during the Phase I Remedial Investigation and the Phase n Remedial investigation
at the Warm Springs Ponds Operable Unit. These data included bottom sediment
thickness information collected during bottom sediment sampling activities at about
45 locations within the three ponds and data developed to prepare a bathymetric
map of the pond bottom surface. The base of the pond bottom sediments was
defined as the contact between fine-grained sediments and native material. Native
-------
material at the Warm Springs Ponds was typically' either peat or coarse-grained
sand and gravel.
One commenter (Letter 154) stated that the annual sediment loads and the
sediment loads from smaller flood events in Warm Springs Creek are more
significant than the sediment loads from larger events and should be addressed in
the FS.
Response: The accumulation of sediments in the Warm ~prings Ponds was used to
estimate an average annual sediment loading to the ponds. The purpose of the
.. .
FLUVIAL-12 bedload sediment transport study was to determine potential erosion.
of bank and floodplain tailings sediments from Silver Bow Creek during .flood
events. Erosion of tailings was expected to be insignificant for ~ow ranges up to
nearly bankfull (estimated at about ?OO cfs or a 2- to 5-year flood event). Since
most of the bedload toxic sediment transport would only occur in major fl.ood
events, the treatment and containment system was designed considering the peak
flow and volume of the IOO-year flood.
Total sediment reaching the ponds, as. presented in this comment, is defined quite
differently. Total sediment includes both suspended load and bedload from all
sources and flow ranges. A large quantity of suspended load originates from
"natural" forest and rangeland erosion. This total load is what the commenter is
describing with the analysis of USGS flow and sediment data. Total sediment load
transport would be expected to greatly increase for flows above the 90th percentile
flow duration exceedance; since most of the annual flows in that range represent
overland spring snowmelt runoff from forest and rangeland. Peak flood flows, in
the 100year to lOO-year range, are _above the 99th percentile on a flow-duration
exceedance curve.
-------
2.2.3 .sw:face Waters
One commenter (Letter 72) points out the correlation. (noted in the FS) between
low pH values and higher dissolved metals levels in the streams. The commenter
notes that copper. toxicity varies with pH and. that the interaction be.tween
aluminum and low pH may pose a significant hazard to aquatic fauna at the
concentrations detected in the surface water at the site.
Response: EPA and MDHES agree that copper toxicity varies. The agencies feel
that alkalinity plays a somewhat more important role in the toxicity of copper,. as
evidenced by the dependence of the freshwater ambient water quality criterion to
alkalinity (alkalinity is generally considered equal to carbonate hardness), but not
pH. The copper ion is complexed by anions, which, in turn, affects the toxicity of
copper. At lower alkalinity, copper is generally more toxic.
A significant volume of recent literature has discussed the association of low pH
(below 5.2) and extreme aluminum toxicity. The agencies believe that if the
criterion of pH is met, the. toxicity of aluminum can be controlled.
Aluminum was analyzed in samples taken from three sampling stations during the
spring 1986 high flow event. The pH range during this sampling event was from a
low of 6.5 to a high of 9.5. The maximum concentration of dissolved aluminum at
each sampling station did not exceed the acute ambient water quality criterion of
760 #Jg/I. At the two sampling stations where only two samples were taken
(outflow from Pond 3 and Mill-Willow Bypass), the concentrations of dissolved
aluminum exceeded the chronic criterion of 87 IJ.g/I (averages of 97 and 139 IJ.g/l).
The chronic criterion was not exceeded at the sampling station with 12 sampling
events (average of 67 IJ.g/I .at the inflow to the ponds). Total aluminum
concentrations exceeded both acute and chronic criteria at the Mill-Willow Bypass
-------
sampling station and chronic criteria at the inflow to the ponds and within the
ponds.
The potential exists for a hazard to aquatic fauna from aluminum toxicity, as
several sampling events did detect aluminum at concentrations greater than the
acute criterion.
,.. Several commenters (Testimony M-1, M-2, M-3) expressed general concern over
fishkills in the Clark Fork River. O~e comment (Letter 138) noted that the July.
1989 fishkill was caused not only.by the tailings in the Mill-Willow Bypass, but also
by the streamside tailings downstream of the Warm Springs Ponds. Another
commenter (Letter 3) asked why fishkills occur if fish can live in the ponds. ARCO
(Testimony A-3, B-3, M-6) recognized the problem of fishkills, but stated that, on
the whole, fish and wildlife in the Warm Springs Ponds operable unit are healthy
and abundant.
Response: The agencies share everyone's concern over fishkills in the Clark Fork
River. The fishkills are thought to be the result of shock loadings of waters in the.
Mill-Willow Bypass and upper Clark Fork River with highly soluble metal salts
during summer thunderstorm events after extended dry spells. The problem exists
because of the exposed tailings slickens found within the bypass channel and along
the upper Cark Fork banks, not within the ponds. The fish in the ponds are not
subjected to the extremely high concentrations of metals that cause the fishkills.
For a full discussion of the causes of the fishkills, see Chapter 4 of the FS.
It is recognized that tailings downstream of the Warm Springs Ponds contributed to
I
the July 1989 fishkill that extended downstream to Deer Lodge. Temporary control
of the downstream tailings is being addressed through ARCO's benning project, but
long-term solutions will be developed in the Clark Fork River remedial
investigation and feasibility study. It is understood that actions at the Mill-Willow
-------
Bypass along the Warm Springs Ponds will not solve the fishkill problems for those
cases where the tailings below the ponds are the cause of the fishkills.
The agencies agree. that terrestrial and aquatic life. appear to be productive and
improving from past years. However, impacts to terrestrial organisms are difficult
to determine unless they are acute or cumulative. There remains the potential for
chronic effects on individuals organisms. Chronic aquatic life criteria ~ave been
and continue to be exceeded for selected contaminants.
2.2.4 Risk Assessment
Two commenters criticized the environmental risk assessment. One commcnter
(Letter 154) thought that the ecosystem analysis was casual, ad hoc, and without a
guiding plan, and that ecosystem techniques have not been applied. That same
commenter and others (Letters 101, 108, 126) added that downstream ecological
risks of toxic metal sediments in the Clark Fork River should be evaluated. Several
comments suggested that the agencies conduct an incremental risk assessment in
developing dam safety ARARs.
Response: A detailed ecosystem analysis has not been conducted at the Warm
Springs Ponds. At the time the investigations were conducted for Warm Springs
Ponds, an full ecosystem analysis was not a requirement in the CERCLA process.
The agencies do not agree with the commenter that the ecosystem analysis is
inadequate. Sampling of key receptors, including fish and waterfowl, has been
conducted. Additionally a survey of all literature on macroinvertebrates from Butte.
to Deer Lodge and an analysis of algae and vegetation were conducted. While the
sampling program at the Warm Springs Ponds was not designed to quantitatively
answer questions on the environmental health of the pond system, the data
obtained can and have been used to qualitatively determine risk to the ecosystem.
-------
Ecosystem assessments to determine the effects of contamination are difficult to
undertake and time consuming. It is often difficult to separate the natural interac-
tions and cycles in the environment from the effects ()f contamination, unless acute
effects can be seen. Acute effects are rarely seen and even when they are (e.g., the
Kesterson Reservoir natural selenium contamination problem) the interactions and
ripple-through effect on the ecosystem are mostly hypothesized and can ~arely be
shown through quantitative sampling as cause-and-effect.
The downstream ecosystem was not..ignored in the assessment .of ecological risk. '.
Only qualitative statements could be made as data with which to model the effects
of a massive release of tailings do not exit. The assessment was conducted
according to EP A guidelines available at the time the report was prepared to ineet
the requirements of CERCLA. While a more detailed risk assessment may be
necessary, it will be prepared in the future as part of other operable units.
Those comments regarding an incremental risk assessment for dam safety design
are addressed in the next section under Dam Safety ARARS.
2.3 ARARS AND CLEANUP STANDARDS
One commenter (Letter 111) thought that the wording in the Proposed Plan is
indefinite about meeting MDHES and federal ARARs and reducing risks.
Response: CERCLA requires any remedial action to protective of human health
. .
and the environment, and to comply with ARARs unless an appropriate waiver is
invoked. The selected remedy, documented in this ROD, meets these criteria. Any
language in the Proposed Plan that suggests any other interpretation was not
intended.
-------
2.3.1 Surface Water Ouality ARARs.
Numerous commenters (Letters 53, 64, 65, 66, 69, 73, 74, 75, 77, 89, 97, 98, 99, 101,
102, 104, 106, 107, 108, 109, 111, 112, 115, 116, 118~ .121, 122, 123, 126, 128, 131,
132, 133, 134, 135, 138, 139, 142, 143, 144, 145, 146, 147, 151, 153, 154, 155, 159,
162; Testimony A-15, M-1, M-3, M-4, M-5, M-7, M-8, M-9, M-10, M-11, M-13, M-
15) recommended requiring remediation of the ponds to result in Gold Book
'.'. (Federal Ambient Water Quality Criteria, A WQC) values being met in the Clark
Fork River and the Mill-Willow Bypass. Many of these copunenters stated that the
appropriate goal shOllld be to prevent all exceedances of Gold Book criteria up to
the 100-year flood. level until upstream sources are cleaned up, and then for all
flows after that. Several thought that Gold Book criteria should be met fOf all
floods and at all times.
Response: Although the term "Gold Book" was not used in Chapter 4 of the FS for
stating the goals for the operable unit, the water quality criteria cited are the Gold
Book A WQC. These criteria were adopted as standards by the State. Remediation
of the Warm Spring Ponds alone cannot guarantee that the standards will be met in
the Clark Fork River, even at the headwaters of the river. Warm Springs Creek
and the Clark Fork River itself are also contaminated, and will have to be cleaned
up before the water quality in the Clark Fork can be assured. However, the
. .
selected remedy will modify the pond system in order to achieve the Gold Book
standards under normal conditions for the water leaving the operable unit, and this
. will go a long way toward improving the water quality in the river.
The ambient water quality regulations do not address the concept of meeting the
water quality standards during major floods. On the contrary, the regulations allow
for periodic exceedances of the standards: as often as once in a 3-year period, the
4-day average can exceed the standards without being considered a violation.
-------
However, the selected remedy. will treat flows to the IOO-year flood and water
leaving the operable unit should meet these standards. .
It is important to note that the standards are probably less likely to be exceeded
during large floods than during small runoff events. The maximum concentrations
of metals in the bypass are seen when a short, but intense, thunderstorm rinses the
soluble metal salts off the tailings deposits and into the bypass with a .minimum
". amount of dilution. This is what has caused fishkills. Uirge floods are unlikely,
near their peak flows, to exceed the s~andards. The reason-Jor. this is simple: there
is too much water and too little readily available contaminants.
(
Metals levels that would be seen in the river, under all the possible flood scenarios
up to IOo-year floods, cannot be predicted with accuracy. Models to deal with a
contamination situation as complex as that along Silver Bow Creek do not exist.
Models would have to be developed, and considerable data would have to be
collected to calibrate the models. This would delay cleanup for years, and it is not
certain that a model satisfactory to all parties could be developed. Instead of
attempting-to model and understand the contamination of the creek water under
any and all flow conditions, the FS took the approach of identifying major
contributions to the contamination and then developing systems to treat a wide
range of flows. In this way, whatever the details of the runoff event and the
resulting contamination, the treatment system would be able to provide treatment
and protection for the Clark Fork River.
The water within the Ponds are not part of any river or creek, and are not covered
by the State's water quality standards. Therefore, those standards are not
applicable to the Ponds themselves. Nevertheless, the remedy must be protective of
human health and the environment, including the environment within the Ponds.
The ROD requires that exposed contaminants and tailings within the Ponds will be
-------
flooded or covered and revegetated. This will protect the ecosystem within the
Ponds, and support the fish and wildlife population which already exist there.
The agencies believe that the methods proposed in the ROD for treating flood
flows are adequate and reasonable and that such treatment would enable the water
quality at the compliance point to meet the Gold Book standards at nearly all
times.
<-
One commenter (Letter 67) recommended that the proposed. compliance point. for. .
.. I . . .
the Warm Springs Ponds Operable Unit be located near the current beginning
point of the Clark Fork River (Fig. 5-1, CH2M HILL, 1989b). As noted elsewhere
in these .comments and in the FS (Note 4, Fig. 6-1, CH2M HILL, 1989b),
contaminated groundwater discharges to the Clark Fork River. Therefore, it may
be advisable to relocate the compliance point far enough downstream to ensure
interception of the operable unit's entire current and future groundwater plume.
Response: The precise locations of the compliance points for groundwater and
surface water are described in the ROD. The compliance area illustrated on
Figure 5-1 is accurate. These compliance points will ensure protection of the
aquifer and the Clark Fork River.
With respect to the FS's proposed waiver of the ambient water quality standard for
mercury, one commenter (Letter 72) recommended an additional mechanism for
determining whether mercury is adversely affecting the environment within the
Warm Springs Ponds. The commenter suggested that the tissues of fish from Pond
3 be periodically analyzed for mercury and other heavy metals to determine if the
selected remedial action is reducing the threat that these substances pose to public
health and the environment.
-------
Response: Analyzing tissue from fish in Pond 3 would provide more information
on the degree of cleanup on Silver Bow Creek than Warm Springs Pqnds. The goal
of the improvement of the treatment in the ponds is to have the effluent of the
ponds meet ambient water quality standards, not the water in the pond system.
Tissue from fish downstream of the ponds could be analyzed, but it may be difficult
to isolate impacts of Warm Springs Ponds from other sources of contamination,
such as Warm Springs Creek. Analyzing tissue from fish in Pond 3 might provide
,.. useful information on the impacts to the fish, but it IIlay not provide specific
information on the success of the \y~rmSprings Ponds remediation.
One cori:nnenter (Letter 111) thought that the standards for arsenic and mercury
should be maintained at below detection levels.
Response: EP A believes that the waiver of the mercury and arsenic standards is
appropriate, given the detection limits for both contaminants and the inability of
current techn,110gy to achieve these standards. The replacement standards are still
very low, and are protective of human health and the environment, based upon
currently available information.
2.3.2 Dam Safety ARARs for Eart1J..Qwlke and....ElQQd..ProtectiQn.
Two commenters (Letters 119, 151) stated that MDHES should have applied the
MDNRC dam safety regulations based on the total volume of water and sediments
in the ponds and on the basis of treating the entire pond system as one pond.
Using the implied value of the stored contents of the total pond system (water and
sediments), the level of protection for the ponds should be 0.75 probable maximum
flood (PMF) for all 3 of the ponds. Numerous commenters (Letters 45, 79, 89, 91.
93, 97, 98, 99, 100, 101, 102, 104, 106, 108, 111, 114, 116, 118, 122, 123, 125, 12h,
131, 132, 134, 138, 142, 143, 145, 149, 151, 152, 153, 157, 161; Testimony A-1 ~)
added that the remedy should assure that all sediments remain contained in tt~~.
-------
..
ponds up to the maximum credible earthquake (MCE) and half of the probable
maximum flood (0.5 PMF) or greater. Other commenters (Letters 106, 116, 118,
119, 123, 135, 136, 140, 143, 145, 147, 149, 151, 159,161; Testimony A-15, M-3, M-
.5, M-11, M-13) suggested that the agencies perform a risk analysis to determine the
appropriate level of protection for the ponds. They further stated that all ponds
should be protected equally, and the protection should be conservative. '. .
....".
Response: EP A and MDHES have reconsidered the .use of varying levels of
protection for the ponds and . concluded . that it is more' reasonable, ~d' in' .
compliance with the applicable law, to provide all 3 ponds with the same level. of .
protection. Consequently, the selected alternative will include protection of all 3
. ponds to withstand a 0.5 PMF. The agencies do not believe that there is reason to
protect the ponds to a greater level than this. The specified level of earthquake
protection is to the MCE.
In order to conduct a quantitativehazardjrisk assessment for the pond system, a
model would need to be developed' that would simulate the various failure
. scenarios under different flood flows at the ponds. Additionally, a model to predict
the transport of the mobilized sediments and their deposition downstream on the
Clark Fork River would also have to be developed. It is unlikely that models could
be developed that would accurately predict the environmental, human health, and
economic damage caused by these events in the flood ranges of interest, 0.5 to 1.0
PMF.
In developing an appropriate level of protection, not only the damage caused by the
event, but also the probability of the event occurring, is important. Although no
specific frequency of occurrence is established for the PMF or fractions of the
PMF, the probability of flows greater than 70,000 cfs occurring on Silver Bow Creek
. .
is extremely small.
-------
One commenter (Letter 138) was not clear whether the design standards would
protect against flooding in Silver Bow Creek, Mill and Willow Creeks, or the three
drainages combined. .
Response: The design floods presented in the FS include flows from all three
drainages combined for areas below the confluence of Silver Bow Creek and Mill-
Willow Creeks. These are the flood flows on which the 0.5 PMF pr~tection is.
, based.
. 2.3.3
One commenter (Letter 72) recommended that action levels for copper, cadmium,
and zinc in contaminated soils and tailings need to be set based on aquatic life
exposure.
Response: It is recognized that action levels for copper, cadmium, and zinc would
be desirable for soils and tailings that lie within the Mill. Willow Bypass. It is very
difficult, however, to develop cleanup criteria for soils based on a direct
relationship between ambient water quality criteria and metal concentrations in
soils.
The mass of metals available from tailings and contaminated soils is dependent on
a number of factors including the mass of metals in the material, the metallic
compounds themselves, the rate of sulfide oxidation, the rate of transport of metals
to the surface of the tailings, the mass of tailings, and the time elapsed since the
previously accurilUlated surface salts were washed away during a precipitation
event. Additionally, the intensity of a precipitation' event and the surface water
flow rate Will impact the metals concentration in the surface water.
-------
.'
Because of the difficulty in establishing meaningful soil cleanup concentrations
based upon aquatic standards, a more direct approach was taken for the initial
removal action in the Mill-Willow Bypass. The depth of copper, cadmium, and zinc
migration from the tailings was determined by screening sampling, and was
confirmed to be consistently correlated with visual staining of contaminated soils.
The amount of tailings and contaminated soils to be removed or o.therwise
controlled was established based on visual identification, correlated with target
. .
> metal concentrations developed from evaluating the metal concentrations in the soil
profile, and subject to confirmation sampling and analyses. While the fmal.. '.
. confirmation sampling has not been completed, preliminary results indicate that this
approach has resulted in cleanup to background levels in the Bypass.
As stated in the ROD, a final action level for soils cleanup and accompanying.
additional cleanup measures will be determined at a later date. The action number
will be based on human health threats. EP A believes that such a number will also
provide adequate protection for the environment. The ultimate ch'eck on this will
be the requirement that surface water standards be met at the point of compliance.
2.4
IDENTIFICATION AND EVALUATION OF ALTERNATIVES
The majority of th'e comments received on the Feasibility Study dealt with the
identification, selection, and evaluation of alternative remediation approaches. To more
easily understand the comments themselves, and the Agencies' consideration of and
response to those comments, they have been grouped into 15 subject areas. Some of these
deal with remediation alternatives in general, much as the Media Specific Actions in the
FS were presented. Other comments relate to specific alternatives, either presented in the
FS or found elsewhere.
-------
2.4.1 Flood Modeling Studies.
Several comments questioned the estimates of various flood .events that have been
developed as part of the Silver Bow Creek Superfund studies. EPA and MDHES
prepared a Flood Modeling Study that utilized historical meteorological and
hydrological data and several computer models to estimate the inte~ity and
duration of various potential. flood events. ARCO countered with other. estimates
'. based on different assumptions and modeling approaches.
..
. One commenter (Le~ter 119) thought the methodology for calculating the PMF \Vas
inaccurate and imprecise. Another commenter (Testimony M-10) suggested that,
since MDHES and ARCO disagree on the magnitude of the various floods, a third
party, such as the United States Geological Survey .(USGS), should perform an
independent evaluation of the flood modeling.
Response: USGS was consulted for an independent evaluation of the flood
modeling. USGS evaluated the 100-year event and concluded that the peak
discharge for this event for Silver Bow Creek above the Mill-Willow confluence
should be 3,910 cubic feet per second (cis), which is closer to the 4,000 cfs
estimated by MDHES than the 3,30~ cis estimated by ARCO. The USGS also
stated that the assumptions and selection of model parameters used in the MDHES
modeling yielded results that are well calibrated for use in the upper Clark Fork
Basin.
USGS 3.Iso stated that both the MDHES model and the ARCO model may have
\ .
overestimated the volume of the 100-year flood. Both models predicted a total 5-
day volume of approximately 13,000 acre-feel The most important parameter of
the 100-year flood is the design volume of runoff because it governs the amount of
flow to be treated. For design purposes, the agencies have decided to use 3,300 cfs
. as the peak design flow for the Pond 3 intake structure and 13,000 acre-feet as the
-------
design volume for the Pond 3 upgrade. For more detail, please refer to the
responses to ARCO comments on Section 4.1.1 of the FS.
Although USGS didn't r~view the modeling of the PMF calculations, the same
hydrologic model used by MDHES for the 100-year event was also used to calculate
the probable maximum flood (pMF). Thus, the agencies believe that tbe PMF
modeling is appropriate and defensible. As discussed in the Flood Modeling Study'
prepared by CH2M HILL, the PMF depends on a number of assumptiom. They
include future climatic conditio~, precipita~ion event characteristics, antece~ent '.
'precipitation, ground' conditions, and hydraulic channel characteristics. Even
though it is impossible to substantiate these assumptions exactly, they can
reasonably be studied and estimates can be made from historic records. The Flood
Modeling Study used a calibrated precipitation vs. runoff model (HEC-1) to .
calculate flow values for various frequency floods. This model was calibrated using
existing recorded rainfall and corresponding runoff data. Publications exist that
present methods for calculating probable maximum precipitation (PMP) for a given
area. Since the Silver Bow Creek Drainage is located along the Continental Divide,
there are two publications which cover the drainage for computation of PMP.
These are published by National Oceanic and Atmospheric Administration and are
Hydrometeorological Report No. 43 and 55A Probable maximum precipitation, as
calculated using the above reports, was input to the calibrated HEC-1 model to
calculate runoff during a PMP. This calculation produced a range of possible
values of the PMF, 129,000 cfs to 201,000 cfs. The agencies have adopted a PMF
of 140,000 cfs for design purposes.
One commenter (Letter 78) stated that the FS should consider a simultaneous flood
on Warm Springs Creek and Silver Bow Creek, and its impact on Pond 1.
Response: The lOO-year flood, as calculated for the Cark Fork River downstream
of Warm Springs Creek in the flood modeling study, included the contribution of
-------
Warm Springs Creek. The flood modeling study used streamflow records recorded
within the entire Silver Bow Creek Basin and also those drainages upstream of the
gage o~ the Clark For.k River at Deer Lodge. Historical.floods were used as
calibration for the hydrologic model constructed for the entire drainage upstream of
Deer Lodge including Warm Springs Creek and Silver Bow Creek. Flood flows do
. occur simultaneously, but due to differing basin hydrology and flood-producing
mechanisms, the 100-year flood would likely not occur at the same tim~. on both
~ Warm Springs Creek and Silver Bow Creek.
.. .
One commenter (Let~er 47) was concerned that CHZM HILL could not provide. the
information at a public meeting that 27 percent of the annual flow in the Silver
Bow Creek watershed comes from Mill and Willow Creeks.
Response: The information on the annual flow from Mill and Willow Creeks was
determined as part of the Silver Bow Creek Flood Modeling Study conducted by
CH2M Hll.L The CH2M HILL representative at the Anaconda public meeting on
November 9, 1989, stated that, although he could not recall the exact figure, he
estimated that it was between 20 and 25 percent, but would need to check the
reports to respond with the precise value.
2.4.2
Several comments were concerned with the issue of how to effectively control and
treat the sediment-laden waters associated with flood events. Numerous
commenters (Letters 5, 53, 64, 66, 69, 72, 73, 74, 82, 92, 97, 98, 99, 101, 102, 104,
106, 107, 108, 109, 111, 112, 115, 116, 120, 121, 122, 126, 128, 131, 132, 133, 134,
135, 138, 139, 142, 143, 144, 145, 146, 147, 151, 153, 154, 155, 159, 161; Testimony
A-IS, M-l, M-3, M-7, M-8, M-9, M-IO, M-ll, M-13, M-15) recommended that some
or all of those Silver Bow Creek flows that are now diverted to the Mill-Willow
Bypass be controlled and treated because they are associated with erosive events
-------
.'
and contain higher than average loadings of contaminated sediments. Most
commenters suggested full treatment of up to the 100-y(:ar flood. Several suggested
treating even greater flood flows. Three commenters .(Letter 101, 108, 126) further
stated that water quality criteria violations should not be allowed, even during
floods, and that routing the flood flows through the pond system will not be
effective because the detention times during floods would be too short. The
comment supported the concept of an upstream impoundment to meter . flows into.
./ the pond system. .
. .
-. .
. Response: The high. flows mentioned are those associated with large floods. . At
present, the pond system is able to provide sedimentation and treatment for flows
" that are less than approximately 600cfs, the peak discharge of the 2- to 3:'year
flood. Under Alternative 3 in the FS (the State's Proposed Plan), the upstream
. impoundment would detain most of the flows associated with the 100-year flood
and remove the majority of the sediments in such flows. It appears that the flood
waters would n~t require further treatment in the ponds once they had passed
through the settling basin and would not pose a threat to the Clark Fork River.
Under the ROD's selected' remedy Ponds 2 and 3 improvements will be
implemented to enable the Pond system to hold and treat adequately the design
volume of the 100-year flood. For more discussion on the development of the
design criteria, please refer to the response to ARCO's Comment No.3 on the FS
Section 4.1.1.
One comment (Letter 104) recommended that a sediment survey be done and
correlated with historic flood amounts in order to project required storage capaci ty
of the Warm Springs Ponds to help determine their potential for use in controlling
flood events.
-------
Response: Qualitative and empirical regional data from other sediment surveys'
and studies of mass loading from streamside deposits ~d upland areas along with
sediment transport (sediment and bedload) are used to estimate reasonable ranges
of sediment inflow to the project area. Short-term monitoring and sediment surveys
are not considered representative of long-term trends due to the changes in
upstream sources (e.g., raw tailings and the Weed Concentrator flows are no longer
discharged directly to Silve~ Bow Creek). Monitoring' of sediments ,should be
.,. included in the monitoring plan (to be prepared during remedial design) to
evaluate the long-term project performance. and to "identify any additional"
maintenance requirements.
Another comrnenter (Letter 154) stated that an unkriown fraction of suspended
sediment will be carried through the treatment system under high flows and will
never settle out, and that treatability studies indicate that lime precipitation may
not be an effective treatment technique during flood events, given the large surface-
to-volume ratio and wind-generated mixing.
Response: The agencies acknowledge that turbid water may be released during
high flow conditions, and the removal efficiency of Pond 3 will likely be reduced
compared to normal flow conditions. However, with careful design and operation,
it will be possible to upgrade the current treatment system to provide adequate
treatment for all but extreme flows.
lime treatment was selected as the best available method for treating the large
volume of water entering the pond system. The conceptual design of Alternative 3
was based on providing treatment for 600 cfs, the design maximum flow into the
ponds. The lab-scale treatability studies identified in the commenter's text
evaluated only the removal efficiency provided by metal hydroxide precipitation.
Additional metals removal would be provided by bio-uptake, calcite coprecipitation,
-------
and the settling of suspended solids. In order to prevent adverse effects to aquati
life, the system's pH would be kept between 7.5 and 9.0.
One commenter (Letter 95) suggested that flood flows upstream of Pond 3 be
diverted into the Opportunity Ponds.
.
Response: The Warm Springs Ponds system is already in plac~ for the purpose of
.,. treating the waters of Silver Bow Creek. Utilizing the existing ponds is a much
more direct and cost-effective approach to handling Silver Bow Creek floods than-.
building a new system to divert the flood waters into the Opportunity Ponds..
Remediation of the Opportunity Ponds is being studied as part of the Anaconda
Smelter Superfund site.
2.4.3 Proposed {Jpstream Impound.Iillm1.
A large number of the public comments were opposed to the Agencies' preferred.
plan because of the upstream impoundment or settling basin that it included. The
opposition of most of the commenters (Letters 2, 3, 7, 8, 10, 11, 13, 14, 15, 16, 17,
18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36,37, 38, 39, 40,
41, 42, 43, 44, 47, 48, 49, 51, 55, 57, 58, 60, 76, 78, 80, 81, 83, 85, 86, 87, 105, 130,
148, 158; Testimony A-I, A-2, A-6, A-7, A-9, A-10, A-II, A-14, A-16, B-3, B-5, B-6,
B-9) was based on perceived impacts of the impoun~~nt on land, groundwater,
gardens, property values, county tax base, adjacent lands, public health, and
environmental aesthetics affecting tourism. One commenter (Letter 8) also thought
it would take too long to acquire the land for the impoundment.
. Response: EP A and MDHES agree that the settling basin would have some nega-
tive impacts at the location examined in the FS. In light of the overwhelming
public opposition to the upstream impoundment and careful review of the
possibility of treating major flood events in the Warm Springs Ponds system (after
-------
extensive capacity and treatment modifications), the selected remedy does not
include the upstream impoundment. H monitoring during the first years of
operation of the selected remedy reveals inadequacies. in the treatment and flood
handling capabilities of the pond system, then the ag.encies will have to reconsider
the need for an upstream impoundment.
One commenter asked for clarification regarding the location of the proposed 2,000
-;. acre-foot upstream impoundment.
Response: The upstream impoundiD.ent would have been located just south of
.. .
Montana Highwayl and west of Silver Bow Creek. Its location was shown on
Figures 7-10 and 7-16 of the FS.
One commenter (Letters 7, 49) sent two letters opposed to an 8,000-acre-foot
impoundment because that would be oversized for potential floods in this area.
Response: The Silver Bow Creek Flood Modeling Study (CH2M ~ November
30, 1989) was based on a comprehensive analysis of historical flood and
precipitation events for the Silver Bow Creek drainage area. As part of the study,
a precipitation-runoff computer mod~l was calibrated using recorded rainfall and
snowmelt zones vs. recorded streamflow events. Using this calibrated model and
statistics on precipitation/snowmelt frequency, the calibrated precipitation/runoff
model calculated the 100-year 5-day flood volume as 13,000 acre-feet for Silver Bow
Creek above the Mill-Willow Creek confluence. The 8,000 acre-feet size was
determined using an inflow-outflow mass balance analysis based on the storm
hydrograph.
Another comment (Letter 154) noted that a 2,000 acre-foot upstream impoundment
would not even contain the flood volumes from a 10-year event in Silver Bow
Creek. The commenter added that, as a settling basin, the impoundment's effec.
-------
tiveness would also be in question since it would not begin to divert Silver Bo
Creek flows until they reach 600 cis. At 600 cfs, Silver Bow Creek would already
be carrying a substantial sediment load. Because of. these factors, the commenter
suggested that Alternative 3 could not meet the water quality standards.
Response: E~A and MDHES agree that a 2,000 acre-foot impoundment would
likely not be able to totally retain a 10-yearevent in Silver Bow Creek. As noted in
. .
-:- Chapter 7 of the FS, the smaller impoundment was analyzed to determine the
efficiency of settling the entrained sedimellts as a comparison against the storage.
. .
mode of the 8,000 acre-foot impoundment. Preliminary estimates indicated that the
smaller impoundment was only ab~ut 5 percent less efficient as ~ settling basin than
the larger impoundment. The fact that either impoundment would not divert' flows
less than 600 cfs does not mean that these flows would go untreated. Flows below
600 cis would still be diverted into Pond 3 for treatment, including settling.
Preliminary calculations, however, indicate that the standards for dissolved metals
would likely be met with the original preferred alternative during flood events .
the 10- to 100-year range.
The selected remedy includes neither the 2,000 acre-feet or the 8,000 acre-feet
upstream settling basins. It will, however, use the upgraded existing pond system to
ret$ and treat the fulllOO-year flood flows.
Two commenters (Letters 5, 82) expressed concerns about the location of the
settling basin in the floodway, statmg that it should be protected to withstand a
100-year flood.
Response: The upstream settling basin, as developed in the FS, was to be
protected from a 05 PMF, a flood many times greater than a 100-year flood.
-------
Another commenter (Letter 72) suggested looking into the feasibility of providing.
lime treatment to runoff entering or exiting the upstream impoundment.
Response: This concept was explored during the iriitial phases of the FS, but was
not carried further because redundancy of treatment facilities was not determined
to be cost-effective. In addition, the upstream impoundments were prop9sed for
temporary storage and physical settling only. They were not designed to have the
detention times necessary for effective chemical treatment.
-'"
--
Numerous commenters (Letters-56, 64, 66,'68, 69, 73, 74, 75, 90, 99, 101, 107; 108,
p .
112, 113, 115, 116, 119, 120, 122, 125, 126, 128, 131, 134, 135, 138, 144, 145, 146,
150, 151, 154, 156, 160, 161; Testimony A-2, A-7, A-10, A-14, M-3, M-13) thought
that other upstream flood control dams in the upper drainages of the Clark Fork
basin should have been evaluated. during the FS. They suggested that flood control
dams would reduce the magnitude of major floods, would reduce the sediment
transport, and could be used to augment recreational opportunities in the area.
Response: EP A and MDHES did consider the possibility of' constructing flood
control dams on the tributaries of Silver Bow CI:eek during the FS. Preliminary
locations were identified on Silver Bow Creek, Browns Gulch, F1int Creek, Perdee
. Creek, Homestead Creek, Whitecraft Gulch, and German Gulch. There were
several reasons for not pursuing this concept further for the Warm Springs Ponds
FS.
.
The drainages are generally steep, which means that relatively high dams
would be required (generally 80 to 200 feet high), and that the storage
capacity gained by construction of the dams would not be great compared to
either' the expense of building them or the hazards they would represent.
-------
"'+
+
The structures would all be on-channel dams, unlike the off-channel
impoundments considered in the FS. Because they would be on-channe
. .
dams, each would have to be constructed with a .PMF or partial PMF
spillway for protection against failure during'inajor floods. These spillways
are very expensive structures, and would add considerably to the cost of
construction. .'
The number of dams that would have to be built to achieve the purpose
. .
(moderating flood flows o~. Silver Bow Creek) seemed unreasonable:.
Probably as II}any as six' dams would have to be built to partially cbntrol
flood flows on Silver Bow Creek. Maintaining and operating so many dams
would be a very expensive undertaking that would have to be funded and
monitored for the indefinite future. This is not.in keeping with the sense of
. permanence required for remedial actions under Superfund.
+
Some (\f the dams probably could not be built due to the adverse envi
,
ronmental and other impacts associated with their construction. For
instance, the possibility of constructing a dam on Silver Bow Creek, in the
canyon area near the confluence with German Gulch, would likely not be
feasible (at least for the limited purpose of flood moderation) because of the
. need to relocate two railroad lines and a power transmission line, and
because of the adverse environmental impacts that such a facility would
cause.
+
To gab?- the maximum flood-control capacity from these dams, they would.
have to be kept empty most of the. time. This would severely limit their
usefulness as water storage or recreational reservoirs.
-------
.
Some of the reservoirs would eventually collect contaminated sediments. .
This would be particularly true of a r~servoir. on Silver Bow Creek. The
sediments would have to be cleaned out of the res.ervoirs, or they would
become sources of acute releases of contaminants to the creek and possibly
cause fishkills, much as the tailings along the Mill-Willow Bypass currently
do.
',', One commenter (Letter 72) noted that the upstream impoundment would settle out
only the larger particle sizes in the s~diments carried by flood flows. It is suspected,
that the smaller particle sizes (which would not. settle out) carry proportion'ately
more meta,ls than the larger sediment sizes. The potential sediment toxicity to
aquatic organisms from these smaller sediment sizes '.should be evaluated more
thoroughly.
Response: The agencies agree that it would be desirable to evaluate the toxicity of
, flood sediments more' thoroughly. Unfortunately, it is difficult to determine the
toxicity without experiencing and sampling an actual flood event of sufficient
magnitude to yield meaningful results.
2.4.4
d Pr
Four commenters (Letters 62, 111, 119, 140) recommended that the hydrologic
standard for emergency and principal spillways, contained in the Dam Safety and
Administration Rules under Section 36.14.502, not be used as the sole criteria in
the selection of the design flood, due to the hazardous nature of the material stored
.in the dams. Numerous commenters (Letters 53, 56, 62, 64, 66, 68, 69, 73, 74, 75,
90, 91, 97, 101, 102, 104, 106, 108, 109, 114, 115, 116, 119, 121, 122, 126, 128, 131,
132, 134, 135, 136, 138, 142, 143, 145, 146, 149, 151, 152, 153, 157, 158, 159;
Testimony A-7, A-14, A-15, M-1, M-2, M-3, M-7, M-10, M-13, M-15) made general
statements that the pond protection should be conservatively designed. Several
-------
specifically stated that the level of protection should be 0.5 PMF or the full PMF.
Several of the commenters recommended that the study include a risk analysis of
possible dam failure and offer rationale for selection, of the design flood event.
Response: MDHES and EP A recognize that the Dam Safety and Administration
Rules represent a minimum level of protection required for the Warm ,Springs
Ponds. The level of protection required, based on these rules, c~nsiders 'that these
are high hazard dams with the potential for loss of life downstream due to flooding
in the event of a dam failure. ~e agencies. have decided "to not use varying levels "
, of protection, but rather to protect the entire set of pond berms along the major
flood route, the Mill-Willow Bypass, to a peak flood discharge of 70,000 cubic feet
per second (cfs). This level of protection is equivalent to protection from the 0.5
PMF for the entire pond system.
'.
While it might be desirable to conduct a full incremental risk assessment of
possible dam failure events in an attempt to determine the most cost-effective level'
of dam protection, the agencies feel that such a study would be too costly, take too
long, and would likely not provide definitive enough answers regarding expected
risk. It was felt that using a conservative number based on DNRC dam safety
requirements was the proper approach.
The selected remedy includes measures to protect the pond berms from failure
even in large floods, up to a 0.5 PMF. The ponds would not fail in floods up to
this level, and therefore would not add to the damage that would result from any
flood of less thari 05 PMF'. If a 1arger flood, such as a full PMF, did o~cur, and
damaged the ponds, the amount of water released by the failure of the ponds would
be small compared to the size of the flood. It's probable that the downstream
communities would be much more affected by the flood it$elf than by the failure of
the ponds.
-------
I .,
.'
One commenter (Letter 119) thought that twenty million dollars could be trimmed.
from the state's. preferred alternative by designing to substantially lower standards
than the fractional PMFs listed. The money saved could be applied to ultimate
. . .
removal .of the sediments from the floodplain. . .On the other hand, another
commenter (Letter 101) thought that the flood-protection measures needed to be
significantly more stringent than those proposed and should be based, on the
hazards and costs associated with catastrophic release of toxic materials. This
.. .
,.. commenter stated that the flood protection costs must be compared with removal to
a repository outside the floodplain.
Response: The ARARs for flood protection and earthquake stability were de-
termined based on the anticipated risks associated with catastrophic release of
sediments from the ponds. Even if it was known that the sediments would be
removed within a given time frame (say 30 years) the risk of catastrophic failure in
any given year would remain the same. Thus, designs would still have to be per-
formed to the level of protection as determined by the ARARs analysis.
The costs for providing adequate flood protection for the Warm Springs Ponds are
substantially less than the costs to remove the sediments from the floodplain. The
costs for raising the berms and armoring them against the desigriated floods are
estimated at about $13.5 million. The costs to remove and dispose of the pond
materials to a repository outside the floodplain would be approximately $400 to
$500 million, using conventional excavate-and-haul techniques.
One comment (Letter 160) thought the FS should have considered an option for
flood and earthquake protection that would entail driving piles (30 to 60 feet in
length) through the center of the berms on 18- to 24-inch centers. Then, if the
berms did give way, the sediments would be retained behind the piles.
-------
..
Response: This concept was not considered because it would be much mor
expensive than berm modifications (flattening the downstream slopes and riprap
armoring) in protecting against floods and earthquakes. It would not provide any
greater protection against flooding and,. without very special designs, might not
provide as much protection against earthquakes.
One comment (Letter 138). stated that the collection system designed to route
-;> eastside runoff around the pond system should be designe
-------
existing berms have downstream slopes of 2.25 to 1. Flattening the downstream .
slopes to 2.5 to 1 does not provide much additional strength.
Response: The original MDHES proposed plan included increasing berm heights
along the Mill-Willow Bypass. Figures 7-1 and 7-2 of the FS show the intent to add
that additional height. In addition, the ROD's selected remedy includes the ARCO
concept of treating major floQd events in the pond system. This altemati\,e requires
, substantial raising of all of the Pond 2 and 3 berms for increased flood storage and
treatment capacity. Depending upon location, the do~tream slopes of. the .
, existing berms vary from approximately 1.75:1 to more than 2.5:1. The seis~c
analysis was preliminary in nature and recommended a minimum of 2.5:1 for cost-
estimating purposes. A detailed seismic stability analysis will be performed during
the final design. The configuration and slope of the berm stabilization (to
withstand the MCE) will be optimized at that time based upon site-specific
information. DNRC dam safety engineers have reviewed the designs of the berms
along the Mill-Willow Bypass and have found them. to be appropriate for MCE
protection.
One commenter (Letter 138) stated that the siphons crossing under the Mill-Willow
Bypass from the Opportunity system should meet earthquake and flood protection
standards to prevent release of contaminants into the Cark Fork.
Response: The siphons from the Opportunity Ponds no longer carry substantial
flows, except during local rainfall events. However, during these events, contam-
ination due to a breaching of the siphon pipelines is possible.. The agencies agree
that the siphons should be evaluated for earthquake and flood stability. This
evaluation will be performeQ during remedial design.
-------
.'
2.4.6 Tailings Removal and.J)isposa1 Options.
,t
Numerous commenters (Letters 9, 56, 64, 66, 68, 69', 73, 74,}5, 78, 79, 90, 93, 97,
98, 99, 101, 102, 103,.106, 107, 108, 109, 114, 115, 116, 117, 119,.120, 121, 125, 126,
132, 133, 134, 135, 138, 140, 142, 143, 144, 145, 146, 149, 151, 152, 154, 155, 159,
161; Testimony A-IS, M-3, M-5, M-8, M-9, M-I0, M-ll, M-14, M-15) stated .that the
agencies should find a site outside the floodplain, such as Smelter' Hill in.
-;. Anaconda, for disposal of tailings and should reevaluate the alternative of removing
the Warm Springs Ponds tailings to..that disposal site. Many of the commenters,.
, thought that the contaminated materials currently contained in the Warm Springs
Ponds should be removed from the floodplain following permanent cleanup of
upstream sites, when the Warm Springs Ponds are no longer needed to treat Silver
Bow Creek waters.
Response: Because of the upstream contamination on Silver Bow Creek, the ponds
must remain in place to treat the creek to reduce contamination of the Clark Fork
River until the upstream sources are cleaned up. The current remedy for Warm
Springs Ponds is interim, and the ultimate disposition of the ponds will be
addressed as cleanup of Silver Bow Creek progresses.
Removal of aU contaminated materials from the Silver Bow Creek floodplain, with
disposal at a local repository, was considered during the screening of technologies
and process options in Chapter 6. It was screened from further consideration at
that time because of very high costs. The agencies' preliminary analysis indicated
that conventional excavation, transport, and disposal of the 19,000,000 cubic yards
of contaminated materials (using over-the-road transport vehicles) would cost
approximately $400 to $500 million. In addition, locating a permanent repository
for this volume of material would be difficult. For example, if the m~terials were
piled 30 feet deep, it would require a storage area of approximately 600 acres. It
-------
. .
may be difficult to find a suitable storage area this size within reasonable distance
of the Warm Springs Ponds thai would be acceptable to all parties concerned.
Three comments (Letters 58, 88, 124) recommend. that the mine wastes in the
ponds be removed and disposed of in the mined-out areas or in the Berkeley pit.
Two of the commenters (Letters 88, 124) thought that the agencies should have
considered an alternative which would use a slurry pipeline to pUmp the
-- contaminated sediments back to the Berkeley Pit near Butte. '.The contaminated
water now slowly flooding the Berke~ey Pit could be used ~ormakeup water for the. .
slurry pipeline. Thisoalternative would provide a permanent repository for the
. .
~ .
wastes, resolve the future problem of disposal of the Berkeley Pit waters, and allow
a central location for metallurgical or chemical removal of the heavy metals in .both
the Warm Springs wastes and Berkeley Pit waters.
Response: This is ~ potentially viable alternative for ultimate disposal of the Warm
Springs Ponds sediments. However, until the sources of contamination upstream of
the ponds are eliminated, the Warm Springs Ponds must remain in operation to
remove sediments and metals. It would not be advisable to begin the slurry
pumping operation until. the ponds are taken out of service as a treatment system.
The primary reason is that the slurry operation would require dredging the pond
bottom sediments, likely resulting in considerable resuspension of sediments. It
would be better to wait until the ponds no longer discharge to the Clark Fork River
before beginning this operation.
The costs for dredging and pumping the Warm Springs sludges to the Berkeley Pit
would be substantial. Very preliminary cost estimates indicate capital costS of $30
to $50 million with operation and maintenance costs of $2 to $3 million per year.
H a 12-inch pipeline were employed, it would require approximately 8 to 10 years
of around-the-clock operation to pump the 19 million cubic yards of material to the
Berkeley Pit.
-------
.'
Two important features of this concept should be pointed out.
.
The use of a second pipeline to allow use of Berkeley Pit waters as
makeup water for the slurry operation is not advisable. The Berkeley
Pit waters are very acidic, with pH of 1 to 2. Not only. is this type of
water highly corrosive to pumps and pipelines, but its use as makeup
water to slurry the pond sludges would likely result in redissolution of.
the metals. It is likely that the Berkeley ~it waters will eventually
have to be treated; th?S, the re-dissolved met~ would have to. be..
removed at that tiine.
.
The makeup water for the slurry would most likely come from Silver
Bow Creek in We vicinity of the Warm Springs Ponds. Because of
water rights issues, the water would have to be returned to Silver
Bow Creek at the Berkeley Pit after the slurry operation. This would
require a dewatering operation and likely a treatment plant for the
supernatant prior to discharge to Silver Bow Creek.
One comment (Letter 119) stated that removal of the sediments from the
floodplain to a nearby disposal site could be done economically ($3.00/cubic yard)
if done over a period of years utilizing an efficient transport system such as a slurry
pipeline. The construction of this. system should be deferred until after the
upstream contamination sources are cleaned up, since the Warm Springs Ponds will
be needed until then. However, if the money for the removal and disposal system
were invested today, the interest earned could substantially reduce the investment
required when the system is placed in operation.
Response: H it is decided in the future to remove the sediments from the ponds,
the agencies agree"that some form of slurry pipeline would likely be a more cost-
effective transport system than a conventional excavate and haul system. The pond
-------
,sediments can likely be dredged, slurried, and transported for costs similar to the
$3.00jcubic yard (1990 dollars) noted, depending upon the pipeline length.
However, substantial other costs would be incurred. There is some doubt as to
. , '
whether a feasible and acceptable repository can be located outside of a floodplain
within reasonable distance of the Warm Springs Ponds. H a repository can be
located, the costs to develop, construct, and then ultimately cap, close, and monitor
the repository would also haye to be considered.
EPA and MDHES agree that if the appropriate funds t~ develop these facilities
. ,
were invested today~ it would be likely that the interest earned would reduce ,the
investment required when the facilities are built. However, the responsibility for
funding the overall Silver Bow Creek remediation efforts rests with ARCo. and
other PRPs. The investment and financing decisions will be theirs.
Several comments, dealt with the agencies' proposal to leave the tailings in place
within the present Warm Springs Ponds system. One comment (Letter 79) was
opposed to disposal of contaminated materials in Pond 1 due to the potential for
contamination of the Clark Fork River.
Response: Pond 1 already contains almost 3 million cubic yards of tailings and
other contaminated materials. The Proposed Plan would add another 290,000 cubic
yards, or about an additional 10 percent over what is already there. The pond
would be drained and covered with a low permeability cap that would keep rain
and snow from penetrating into the wastes. This would Tesult in a 'large reduction
in the potential for Pond 1 to leach contamination into the groundwater and surface
water, as it currently does. The pond will also be stabilized to protect against
earthquake and flood events. This will result in a very secure and stable disposal
area.
-------
.'
The plans for cleaning up the Mill-Willow Bypass this summer include placing the
excavated materials into suitable dry areas of ~ond 3, thus reducing the amount of
material that will eventually be placed in Pond 1.
2.4.7 Tailings Reclamatjon and Reprocessing Options.
Numerous commenters (Letters 6, 64, 66, 68, 69, 71, 73, 74, ,75, 88, 99,,100, 102,
.:. 103, 107, 109, 115, 116, 117, 120, 123, 124, 128, 134, 140~ 143, 'i45, 146, 151, 155,
159, 161, 162; Testimony A-15, B-5,..M-1, M-3) thought the .agencies should have. ,
included alternatives, evaluating" additional treatment technologies for hazardous
wastes at the Warm Springs Ponds, including alternatives employing proven or
innovative technologies to extract and recycle minerals 'from wastes in the ponds.
Response: Media Specific Actions for treating the sediments from the ponds and
extracting minerals from them were included in the preliminary alternatives
screening of the FS. These alternatives were screened out early in the process
. '
because they were not found to be feasible, appropriate, or economical. Metals
recovery from the pond sediments would not be economically viable, would do little
to reduce the amount of wastes to be disposed of, and could pose ~onsiderable
threats to the environment. Metals recovery, even from normal ore, is expensive
and difficult, and poses numerous potential threats to the environment. The wastes
in the ponds contain much lower levels of metals than even the poorest quality
usable ores; they are the wastes left over after the metals have been removed.
Processing these wastes to extract the remaining metals, as a means to reduce their
toxicity, would not be cost-effective by today's standards. Much less expensive (bu t .
still very expenSive) treatment methods are available to reduce the environmental
threat posed by the pond bottom sediments, such as solidification. If technology in
the future allows for recovery of the metals from the sediments, recovery would be
reevaluated at that time.
-------
,-
!
Two comments (Letters 113, 139) stated that the agencies should continue to
. research, and implement where possible, revegetation approaches. One thought
that the Schafer and Associates pilot program for neutralizing the tailings holds
promise for curing the problem rather than just moving it.
Response: The pilot program described in the comment is based on technology
developed in an ongoing 3-y~ar program that has been conducted by :MDHES at
-.> the Silver Bow Creek site to develop innovative cleanup methods for the site. It is
applicable to exposed tailings deposits under certain cond~tions, but would not be
useful to treat the pond bottom sedimentS. It is also not by itself capable of
.' .
preventing release of the 19 million cubic yards of ~ediment in the ponds during
flood or earthquake events. It would therefore be necessary to incorporate IIleth-
ods other than neutralization and revegetation to reduce the risk of a loss of the
pond bottom sediments~
For the WSP operable unit, revegetation of exposed tailings areas and of the
disposal units will be undertaken. If the approaches described above have
application in this revegetation effort, they can be incorporated during the remedial
design phase.
2.4.8
Seyeral comments (Letter 138) deal with the capability of the ponds to provide
adequate treatment of suspended and dissolved metals to meet the desired water
quality criteria at the discharge point from Pond 2. One comment notes that high
levels of metals have been measured in the discharge, particularly in the winter and
spring. The same commenter, and another (Letter 115), also stated that
remediation of the ponds should address problems associated with the short
circuiting in the winter' and wind action during the ice-free period that are thought
to resuspend sediments and contribute to metals loading in the dark Fork River.
-------
Response: The agencies agree that the final pond configuration must be able to
pr~vide treatment that will meet the appropriate discharge requirements. The
agencies' original Proposed Plan contained elements, including the upstream
settling basin, improved liming facilities, and the construction of a berm across part
of Pond 3, to alleviate treatment problems. The selected remedy includes a more
comprehensive upgrade of both Ponds 2 and 3, and is thought to adequately
address concerns over pond retention time, sediment resuspension, lime addition,
> and treatment capabilities in general, without the .need" for the upstream
impoundment. . The ROD cont~ provisions to .continue to investigat~ .
resuspension issues., If it is shown tl1at resuspension may cause significant'-W~ter
quality violations, additional remedial measures will be required.
One commenter (Letter 72) recommended the consideration of a new pond
immediately south of Pond 3, stating that a new pond would reduce the extensive
modification required for the inlet structure; avoid the need to channelize Silver
Bow Creek within the dry areas of Pond 3; cover the contaminated soils and tailing
in the dry areas of Pond 3; and increase the pond capacity by 70 percent.
Response: The media-specific actions (FS Chapter 7) were developed to pro\'ide a
range of options to be combined into alternatives (Chapter 8). A new pond in the
dry areas within the Pond 3 berms was considered in Media-Specific Action 5C. In
the FS,- it was not included in the preferred alternative (Alternative 3), because it
would not be as cost-effective in dealing with the problems.
. The selected remedy includes major improvements to Ponds 2 and 3. This plan
includes elements that will accomplish all of the modifications included in the
comment. A new inlet structure will be built. Most of the existing exposed
contaminated soils and tailings in the dry areas of Pond 3 would either be capped
as part of the tailings' disposal area, or flooded by the new Pond 3 normal pool.
The pond capa~ty would be increased to handle and treat the 100-year flood event.
-------
One commenter (Letter 138) stated that the area in Pond 3 south of the proposed'
. -
new baffle is very shallow and that the baffle may not be effective at preventing
short circuiting during winter months.
Response: The proposed berm was intended to prevent short circuiting across the
main portion of the pond by forcing the water to travel through the berm. opening
on the eastern side of the. pond. This would extend the minimunlflow path
'. through the pond during all seasons. It should also be noted that the selected
remedy does not include this baffl~, but rather depend~ 0]1 increased operating.
capacity and greater retention -times to inSure that the discharge from the'pond
. .
. .
system meets the discharge criteria.
One commenter (Letter 154) would like to have seen treatability tests conducted on
site to more closely represent the area/volume ratio of the ponds, ambient weather
conditions, and used Silver Bow Creek water during high flow and low flow
periods. Additionally, other interactions between the Silver Bow Creek sediments
and the treatment process should have been characterized.
Response: The treatability study conducted in conjunction with the Phase II
Remedial Investigation was designed to evaluate the physical and chemical pa-
rameters that must be controlled to maximize metal removal in a pH-controlled
settling-type system. The water used in the "winter" tests was actual water from
Silver Bow Creek and did contain native suspended sediment from the creek. It
was recognized in the planning stages that in situ tests would provide additional
information on the treatment/settling mechanisms in the ponds; however, it was felt
that it would be more prudent to conduct these tests during the remedial design
phase if necessary.
-------
2.4.9 Groundwater Treatment AIternatim.
Several commenters (Letters 53, 60, 82, 101, 108, 12~, 133, 135, 147, 151, ~54, 155;
Testimony A-7, A-14, M-4) noted a general concern that the selected remediation
be protective of groundwater in the operable unit. One comment (Letter 138)
recommended that the final remedy should prevent contaminated groundw~ter from
causing surface water exceedances of Gold Book .standards.
~
Response: The shallow aquifer below Po~d 1 has been. contaminated over the.. .
. years by metals and other contaIninants from the pond system. The trench below
Pond 1 would be designed to cut off the source of contamination for this aquifer
(see MSA 11 and 12 in Chapter 7 of the FS). A portion of the existing
contaminated groundwater will continue to migrate downgradient toward the Clark
Fork River. However, measurements made during the Remedial Investigation
detected no impact to the water quality. of the Clark Fork River as a result of the
contaminated groundwater. Groundwater input into the Cark Fork River will not
be sufficient to cause exceedances of the Gold Book standards in the future.
Background information on the groundwater situation below Pond 1 can be found
in the response to comments on the Groundwater portion of the Site
Characterization and Problem Description Section of this Responsiveness
Summary.
One commenter (Letter 67) raised several concerns regarding. the . impact of the
Proposed Alternative on groundwater flows and contamination at the site and on .
the proposed method of groundwater treatment. The first set of comments dealt
with the proposed construction of a trench to intercept contaminated groundwater
downgradient of Pond 1. The design of the proposed trench raised the following
concerns:
-------
d.
e.
a.
The trench will probably not be deep enough to intercept all groundwater.
flow.
b.
A groundwater flow model was also developed to determine the
effectiveness of a trench and to estimate the flow rate into the trench. This
modeling effort has several deficiencies.
c.
If the open trench is not cleaned regularly, it will quickly begin .to fill with
vegetation, windblown dirt, and sloughed sidewall ~aterial. Obviously, if the
trench starts to fill, its effectiveness will be reduced.
The trench will intercept only that part of the .existing groundwater plume
near the trench. Some currently contaminated groundwater will continue to
flow toward the Mill-Willow Bypass and Clark Fork.
The trench may dewater tailings contained in Pond 1, thereby, causing
release of metals.
Response: Modeling efforts related to evaluating the feasibility of a groundwater
interception trench below Pond 1 were not intended to provide all the information
necessary for final design of the trench. Obviously, additional information will be
. .
necessary to address the variability of the aquifer along the e~tire length of the
Pond 1 berm, if this technology is incorporated into the final remediation at the
site. The calculations and f1
-------
..
quality data collected from dual-completed monitoring wells and paired monitorin
wells located north of Pond 1, groundwater quality improves markedly in wells
completed deeper than about 15 feet below ground surface.. Because the highest
metals concentrationS were measured in samples collected from wells completed in
the upper 15 feet of the aquifer, it is. presumed that most, if not all, groundwater
that exceeds federal primary drinking water standards (the ARAR for groun.dwater)
would be captured by an interception trench as presented. In addition,. depending
. Of on the vertical permeability of sediments below Pond 1, the interception trench is
expected to act as a groundwater dis~harge area where groundwater at some depth..
below the trench will migrate into the trench.
Groundwater flow in the area below Pond 1 was. modeled to: (1) estimate
conservative values of groundwater inflow to the trench; (2) estimate the effective
distance downgradient from the trench at which groundwater would be captured;
and (3) estimate the time it would take to capture degraded groundwater below the
trench. The initial model was executed using hydraulic conductivity values
. .
calculated from slug test data. The model was revised after a pumping test was
performed in a specially designed well located below the Pond 1 berm. Hydraulic
conductivity values derived from the pumping test data indicated hydraulic
conductivity values based on pumping test data were approximately twice as high in
the shallow sand and gravel aquifer as those derived from slug test data in the area
below Pond 1. Therefore, groundwater inflow rates to the interception trench were
adjusted to rates ranging from approximately 2.2 to 4.6 cfs.
Wells were used to simulate the trench because. this method was the most
applicable for the groundwater flow model used (prickett Lonnquist Aquifer
Simulation Model, PLASM). MDHES agrees that a more representative model
would include simulating the trench as a drain. Any future modeling activities are
expected to be completed to provide sufficient information to support remedial
. design of the interception trench. These modeling activities may include using the
-------
USGS MODFLOW groundwater flow model (or an equivalent model) that will
allow simulating'the interception trench as a drain. Simulating inflow to the trench
using a series of wells does not change the resultant opinion, offered regarding the
fe~ibility of such a system. ' '
A constant head boundary was used to simulate groundwater inflow from the
Tertiary hills to the east to p~ovide conservative estimates of groundwater inflow t
-------
b.
chemistry of this water would. be operative along itS flow path (e.g., dilution,
adsorption, dispersion).
Additionally, the commenter (Letter 67) expressed concern over tbe impact on
groundwater from dewatering presently immobilized tailings and pond bottom
sediments. hnportant questions posed included:
a.
What will be the effect of dewatering tailings and..contaminated sediments
that are currently immobilized in th~ reduced conditions of pond bottoms?.
. .
Oxidation of these materials could lead to large releases of metals. ..As
designed, the groundwater trench planned for the interior of Pond 1 will
dewater tailings. This action will allow oxidation and mobilization of the
reduced metals in sediments that have collected in Pond 1. As shown for.
the tailings in the Opportunity Ponds, the mobilization of metals can be sub-
stantial when metal-rich sediments change from reduced to oxidized
conditions. The extent of this mobilization should be quantified and its
effect understood. The trench below the Pond 1 berm may also dewater
contaminated sediments causing a similar mobilization of metals.
What will the groundwater plume in both the shallow and deep aquifers be
in the future? What contaminants will they contain? Will the aquifer's
neutralization and metal attenuation capacities be exhausted at some point?
These issues have not been addressed.
Response: Groundwater interception technologies presented in the. FS were
evaluated with respect to the potential for increasing metals mobility by changing
the metals source environment from reducing to oxidizing conditions. The primary
metals source areas of concern include Pond 1 and the area below Pond 1. The
proposed groundwater interception trenches would intercept seepage from Pond 2
into Pond 1 and seepage from Pond 1 into the area below Pond 1. All intercepted
-------
water will be pumped back to Pond 3 for treatment. This system was designed in
consideration of the possibility of metals releases due to changes in the geochemical
environment of the bottom sediments in Pond 1 caused by dewatering. The
. .
proposed system will effectively intercept and' treat metals-contaminated
groundwater within this portion of the operable unit and will allow for dry closure
of Pond 1. Therefore, definitive characterization of the geochemical fate of this
component of the area's groundwater system is unnecessary.
Several models of dewatered tailings and the potential m~tals production of these
areas are present both within the Warm Springs Ponds Operable Unit and. the
entire Silver Bow Creek CERCLA site. The best example of the long-term fate of
metals production in a dewatered tailings environment is the western portioQ of
Pond 1 and the area immediately below the Pond 1 berm in this area. The western
portion of Pond 1 has been dewatered for many years; water levels have dropped
below the base of the bottom sediments accumulated in the pond. Metals
concentrations in groundwater in the western portion of Pond 1 and in the area
below (downgradient of) this area are relatively low with no measured exceedances
of maximum contaminant levels. Other examples of dewatered tailings areas
exhibiting r'e1atively low metals concentrations in subjacent groundwater have been
identified in Ramsay Flats near Ramsay.
Dry closure of Pond 1 with a low permeability cap would s~rve to reduce vertical
infiltration of precipitation recharge to the underlying groundwater system. This
will further reduce the potential for metals migration vertically into the area's
groundwater system.
The anticipated extent of the metals plume, which exceeds maximum contaminant
levels in the area below Pond 1 following construction of the groundwater
interception trench, will likely not extend downgradient of the interception trench
afte~ the system reaches equilibrium. This assumes that the aquitard separating the
-------
upper sand and gravel aquifer and the underlying sand aquifer is relatively
consistent in the area and that the interception trench is capable of intercepting
most or all of the shallow groundwater system. Any contaminants that are not
intercepted by the trench will enter a relatively good quality groundwater
environment downgradient of the trench recharged with water from the foothills
east of the site and by the Mill-Willow Bypass to the west.
'. The issue the commenter raises about the neutralization and metal attenuation
capacities of the aquifer in the area .below Pond 1 is a moot. point given that the.
proposed groundwater interception trenches will hydraulically capture metals-
contaminated groundwater. EP A and MDHES believe that controlling the
hydraulics of the system will serve to control the chemistry.
The commenter (Letter 67) also was concerned over the impact of the P~oposed
Plan on the groundwater in the Mill-Willow Bypass area. The comment states that
consideration s~1Ould be given to the post-remediation groundwater gradient from
the Warm Springs Ponds to the Mill-Willow Bypass. If the water level in Pond 3 is
increased and the elevation of the bypass is lowered by excavation of contaminated
and borrow material, groundwater discharge to the bypass will likely increase.
Furthermore, if Mill and Willow Creeks are diverted into Pond 3, no surface water
(except during large flood events) would enter the bypass upstream of the new
Pond 3 outlet, and, therefore, the FS assumes that the upper bypass will be dry. It
is probable that the bypass will receive contaminated groundwater discharge from
the Warm Springs and Opportunity Ponds. With no surface water entering to dilute
the inflow from groundwater, water quality in the remediated upper bypass will
likely be considerably worse than it is now. .
Response: It is probable that increased groundwater inflow to the Mill-Willow
Bypass will be realized if the water elevation in Pond 3 is raised and if the bypass
. channel is excavated to a greater depth. Until such time as contamination sources
-------
in Mill and Willow Creeks are removed, it is likely that one or both of these. creeks
will typically be diverted into Pond 3 for treatment. If both streams are diverted
into the ponds, then groundwater inflow will be. the only flow source in the Mill-
Willow Bypass. This flow has been estimated to be. 1 to 5 cfs during the Bypass
removal this summer. The quality of this inflow, however, is not expected to be
poor. An opportunity presented itself during the Phase IT RI to empirically
determine the quantity and quality of groundwater inflow to the bypass. Water in
, the Mill-Willow Bypass was diverted into Pond 3 via the. northern channel
. connecting the Opportunity Pond discharges with Pon~ 3 during July, 1988.
Synoptic flow measurements were made in the Mill-Willow Bypass channel below
the point of diversion to the point where the Wildlife Ponds discharge into the
bypass channel. Samples of water flowing in this reach of the bypass were .also
collected in conjunction with flow measurements. Analytical results from these
samples . indicate that the water did not exceed either chronic or acute ambient
water quality criteria nor did it exceed any primary drinking water quality criteria.
Data from this synoptic flow measurement episode are contained in the Phase n RI
data summary report. Due to construction activities ongoing in the Bypass channel,
it is not possible to directly measure the typical quality of the groundwater inflow at
this time.
Another commenter (Testimony M-7) stated that it was unclear where the
contamination of the second-level aquifer is coming from, and asked for more
assurance that the location and source of contamination can be found and the
contamination cleaned up.
Response: The primary chemical contaminants in the deeper aquifer are sulfate
and manganese. These param~ters are not generally associated with the potential
sources at the Warm Springs Ponds. The occurrence of these parameters in the
deeper aquifer is consistent with the presence of relatively high concentrations of
sulfate and manganese in the groundwater in the vicinity of the Opportunity Ponds.
-------
This suggests that the sulfate and manganese occurrence is more regional in .nature
and is probably the result of multiple contaminant sources and pathways of
contaminant movement.
It should be noted that the ARARs established for groundwater at the Warm
Springs Ponds Operable Unit do not include manganese and sulfate since these
parameters are not included in the primary maximum contaminant levels
> established for drinking water. The sulfate and manganese contamination should
be addressed as part of the Anaconda Superfund site.
. 2.4.10 Costs of Alternatives.
Four commenters (Letters 58, 107, 111, 136) felt that cost seemed to be the driving
force in the selection of the preferred alternative, and recommended that the
agencies should select a conservatively protective remedy regardless of cost. Five
other commenters (Letters 101, 108, 126, 137, 154) thought that a full cost-benefit'
analysis should be performed to evaluate the alternatives. Several other
commenters (Letters 11, 26, 34, 43) supported ARCQ's plan because it was thought
to accomplish the desired remediation at substantially less cost.
Response: Cost is only one of the factors used in the selection of the preferred
alternative. The selection of the preferred alternative was made based on all nine
of the criteria required by Superfund. These criteria include: overall protection of
human health and the environment; compliance with ARARs (applicable or
relevant and appropriate requirements, i.e., laws that have a bearing on the
cleanup); long-term effectiveness and permanence; reduction .of toxicity, mobility,
and volume; short-term effectiveness; -implementability; cost; community
acceptance; and State acceptance. By using these criteria, the relative benefits of
each of the alternatives can be compared to the cost of each alternative. It is
-------
beyond the scope of the Feasibility Study to conduct a formal cost-benefit analysis
of the alternatives~
One commenter (Letter 137) suggested that the evaluation of alternatives needs to
take into account the costs of damage to natural resources caused by the
contamination.
,Response: The ROD and selected remedy are undertaken pursuant to Section 106
of CERCLA for the purpose of protection of public health and the environment. .
. Natural resources damage assessment issues are being dealt with under a separate
process by Federal and State natural resource trustees.
2.4.11 Mill-Willow Bypass Issues.
The Preferred Plan of the FS would divert Mill and Willow' Creeks into the pond
system for treatment. This raised concerns in several areas. One commenter
(Letter 46) stated that the preferred alternative would involve a 27 percent increase
in the capacity of Pond 3 and would, therefore, increase the potential for breaching
the pond. Several commenters (Letters 8, 46, 47, 48, 49, 51, 87, 115, 158;
Testimony A-2, A-7) expressed concern that diverting Mill and Willow Creeks into
the pond system would eliminate the fisheries on ~e upper portion of the creeks
and the upper Clark Fork River. Another comment (Letters 78, 138, 151;
Testimony M-I0, M-14) stated that the sources of the contaminants in Mill and
Willow Creeks should be identified and eliminated as a part of the Warm Springs
Ponds Operable Unit, and that until the sources are identified and eliminated, Mill
and Willow Creeks'should be routed into the pond system. One cOInmenter
(Testimony A-14) suggested possibly diverting Mill and Willow Creeks into the
bypass only during the high flow season, and letting them bypass the ponds at other
times.
-------
Response: EPA and MDHES are evaluating the need to route Mill and Willow
Creeks into the pond system. In conjunction with the Mill-Willow Bypass Removal
Action, ARCO has begun investigating the sources of contamination on Mill and
. .
Willow Creeks. It was hoped that discrete sources of contamination could be
identified and readily removed. Sampling to date this summer has not confirmed
that hope, although it bas been discovered that perhaps Mill Creek can be bypassed
witbout treatment through tbe ponds. The agencies would prefer that option over
> routing Mill and Willow Creeks into Pond 3. This would preserve fisheries habitat
in the Bypass and still meet water. qu~1ity ARARs for the site..
The diversion of Mill and Willow Creeks into the Warm Springs Ponds would
increase the average annual flow into the ponds. It would not, however, increase
tbe potential for breacbing the pond berms. The flow into tbe ponds would be
regulated by tbe capacity of the inlet structure. Flows above tbat capacity would be
routed into tbe Mill Willow Bypass and would not enter the ponds. Thus, there is
no increased potential for breaching the pond berms.
Numerous commenters (Letters 64, 66, 68, 69, 73, 74, 75, 87, 90, 98, 99, 107, 109,
117, 119, 120, 121, 128, 131, 132, 135, 139, 142, 143, 144, 145, 146, 151, 155, 157,
158, 159, 160, 161; Testimony A-7, A;'13, A-14, A-IS, M-5, M-I0) recommended
removal of tailings from the Mill-Willow Bypass during 1990 to prevent future
fishkills.
Response: The agencies agree. Removal of the Mill-Willow Bypass tailings is
currently underway under an Administrative Order on Consent signed by EP A and
ARCO in July 1990.
One commenter (Letter 72) recommends that the agencies consider buffering low
pH during storm events at several points in the operable unit to help prevent future
-------
,-
fishkills and reduce metals migration in the upper Clark Fork River. Buffering high
flows in the Mill-Willow Bypass could be a means to avoid future fishkills.
Response: Buffering pH is not a complete solution to either the fish1411 or the
metals migration problem. Low pH has little to do directly with the fishkill
problem., which is caused by dissolution of water-soluble metal salts from the
surface of tailings deposits. The salts are not dissolved by low pH water in the
, channel, but by rainwater; the salts are highly soluble even in neutral water.
Stopping the migration of metals contaminants to the Clark Fork River requires
. .
more than pH adjustments. The metals must be removed from the flows and'
deposited somewhere. That is the function of the pond treatment system., w}:lich
operates on the basis of physical settling, chemical treatment (pH adjustment), and
biological treatment to remove the metal contaminants from the flows.
Lime treatment at various' points in the operable unit is not thought to be
necessary. The flow management and treatment system included in the selected
remedy would be able to treat all. flows that require treatment, up to a lOO-year
flood. Lime treatment cannot be used as a quick fix at various stages in the flow
management and treatment process. Lime treatment relies on the slow formation
and settling of particles of insoluble metal hydroxides, and can only be successful in
a quiescent system with a long residence time, such as provided by Pond 3.
Avoiding future fishkills is an important goal. The agencies believe that the
ongoing Mill-Willow Bypass Removal and this 'ROD will adequately address the
fishkill problem without the need to buffer pH levels in the bypass. By removing
the tailings and contaminated soils from the bypass and isolating the bypass from
Silver Bow Creek flows except during large floods, the fishkill problem in the
bypass should be solved.
-------
Several comments (Letters 72, 95, 138) suggested that the' remedial measures,
especially along the MilI- Willow Bypass, be designed to include wildlife enhancing
features, such as improving the bypass habitat for trout spawning and rearing.
Response: The FS did not specifically address methods to improve the fisheries in
the Mill-Willow Bypass. . The primary goal of the cleanup in the bypass is to
eliminate the sources of contamination that cause the fishkills. Fish habitat
"improvements would be desirable and may be incorporated by ARCa in the final
design, either to address compliance. with ARARs or to address natural resource.
damage claims.
1.4.12 Wetlands and Wildlife Habitat.
Two commenters (Letters 63, 138) recommended that the agencies give greater
consideration to the need for and value of wetlands at the Warm Springs Ponds.
Another commenter (Letter 72) thought that the risks of developing wetlands below
Pond 1 outweighed the benefits because the wetlands would not remove heavy
metals, and in fact, might lead to recontamination of areas that will have been
remediated. In addition, the proposed periodic removal of contaminated
revegetation would disrupt the functioning of the wetland and lower its usefulness
to wildlife.
Response: EPA and MDHES do recognize the value of wetlands in the Warm
Springs Ponds area. The preferred alternative proposed by MDHES and EP A
would, on balance, create additional wetlands. Although some wetlands would be
eliminated in the Pond 1 area, additional wetlands would be created in Pond 2.
The development or removal of wetlands must be addressed in conjunction with
other concerns, such as prevention of groundwater contamination and protection of
human health. The removal of the wetlands in the Pond 1 area will help to reduce
. the groundwater contamination underneath and downgradient of Pond 1.
-------
As final design plans are prepared by ARCa for the remediation of Warm Springs
Ponds, EPA and MDHES will work with ARCa to identify potential opportunities
" "
for wetlands that are consistent with the requirements for the site cleanup. The
agencies and ARCa are considering a wetlands system for the area below Pond 1
that would be used to treat groundwater for metals removal. While metals do not
biodegrade, they would be assimilated by the aquatic vegetation in the wetland. By
periodically harvesting the v~getation, metals levels in the biomass can be kept
, below toxic levels. Discharge from the wetland would meet the appropriate
discharge standards for the operable unit.
The primary purpose for the wetlands would be to provide treatment for con-
taminated groundwater. Improvements to wildlife habitat would be a secondary
benefit. ance established, the wetlands would be expected to operate in a fashion
similar to the upper portion of Pond 2 with large areas of vegetation and wildlife
habitat.
Numerous commenters suggested that the remediation plans" include elements that
improve wildlife habitat at the Ponds. Several commenters (Letters 45, 47, 77, 78,
87, 95, 104, 114, 148, 158; Testimony A-13, B-2, B-7) stated that the ponds are an
excellent waterfowl and fisheries habitat and that this should be considered before
decisions are made about pond removal and dry capping. Another commenter
(Letter 139) recommended that the final remediation plan include provisions for
several "hog hole" size ponds to overwinter large fish similar to those in Ponds 2
and 3. Finally, one commenter (Letter 61) criticized the Proposed Plan because it
"does nothing for wildlife and fish."
Response: The CERClA remediation process does not allow funding specifically
for measures to enhance wildlife and fisheries habitat. The primary purposes of
proposed remedial actions are to provide long-term protection of public health and
the environment. To accomplish this, the selected remedy will reduce the risk of
-------
catastrophic failure of the pond system and improve water quality in the Clark Fork
through a variety of measures.
Only Pond 1 is being considered for dry closure. It is the primary source of the
groundwater contamination, and for this reason, needs to be dry closed. The
primary fisheries and waterfowl habitats are located in Ponds 2 and 3. The only
significant alteration to these habitats under the selected remedy would entail the
>flooding of the portions of Pond 2 that are currently dry. This flooding would
increase waterfowl habitat.
One commenter (Letter 130) thought that monies should be given to the
Department of Fish, Wildlife and Parks to purchase or "enhance wildlife and public
recreation areas.
Response: Compensation for contaminated areas could not come from CERCLA
funds (Superfund), but would have to be obtained in a separate action from those
parties responsible for the contamination. Compensation for impacts to natural
resources could also be addressed separately from Superfund under the natural
resource damage claims against those parties responsible for the environmental
damage.
2.4.13 Statements of Preference for Alternatives.
One commenter (Letter 150) stated a preference for Alternative 1, which included
in-place solidification of all tailings, contaminated sediments, and sludges, because
it could be considered a permanent remedy.
Response: Alternative 1 was included in the FS because the CERCLA statutes
require consideration of alternatives that utilize treatment to reduce toxicity,
mobility, and volume of contamination. Treatment alternatives are to be given
-------
preference as long as costs are not excessive. Alternative 1 was included to provide
a treatment alternative for comparison purposes.. The estimated costs of
Alternative 1, at over $1.6 billion, are considered excessive when compared against
its relative benefits (see Chapter 8 of the FS for the comparison of alternatives).
Several commenters (Letters 68, 79, 115, 131, 138, 153; Testimony M-2, M-4) stated
opposition to Alternative 3 .because they thought it inadequately dealt with the
.. possibility of downstream contamination due to floods. These commenters
supported Alternative 2, which included an 8,000-acre-foot. upstream impoundment
for flood control and treatment.. Letter 138 noted the need to implement a remedy
that will ensure attainment of EP A Gold Book criteria in the Clark Fork River up
to the 100-year flood event.
Response: By capturing nearly the entire volume of the 100-year flood in Pond 3
and providing treatment through liming and settling, the selected remedy is thought
to provide adequate protection from possible contamination of the Clark Fork
River due to flooding.
EP A and MDHES agree that achieving Gold Book standards up to the 100-year
flood is a desirable goal, and will result in compliance with ARARs. The primary
goal capturing the 100-year flood event is to limit sediment transport from Silver
Bow Creek through the Mill-Willow Bypass to the Clark Fork River. This goal is
to prevent recontamination of the bypass and to limit the future degradation of the
Clark Fork River by continued deposition of tailings.
In conjunction with opposition to the upstream impoundment compOnent of
Alternative 3, numerous commenters (Letters 8, 11, 18, 22, 26, 34, 35, 39, 43, 48,
54, 59, 60, 63, 71, 77, 87, 94, 148, 160; Testimony A-I, A-3, A-4, A-6, A-11, B-3, M-
6) expressed support. for ARCa Plan 3A or a similar approach that would include
many of the Plan 3A components, such as raising the Pond 3 dikes, adding
-------
improved intake structure.s to Pond 3, and improving wetlands, fisheries, and
wildlife habitat. One commenter (Letter 95) stated a preference for the ARCa
plan with minor modifications.
Response: EPA and MDHES have considered ARCO's Plan and have decided to
incorporate some of ARCO's suggestions into the final remedy. The ROD provides
a complete description of the new remedy, including those elements that come from
> ARCO's plan.
Several commenters. (Letters 78, 139) felt that the RIjFS should have studied ,an
alternative similar to ARCO's Plan, but noted that, since ARCO released their Plan
independently, there is no easy way to fairly compare costs, etc. For example, since
the ARCO Plan uses different assumptions than the MDHES Plan for IOO-year
flood, and neither includes Warm Springs Creek in their flood projections, it is
difficult to compare the relative merits of the approaches.
Response: Many of the components in ARCO's Plan were included in alternatives
developed in the RIjFS. However, the method of treating the lOO-year flood, as
proposed by ARCO, was not included in the RIjFS.
The fact that ARCO used different design assumptions does make it more difficult
to compare the two plans. The Flood Modeling Study used in the preparation of
the RI/FS did calculate the lOO-year flood on Warm Springs Creek. The flows
from Warm Springs Creek were not included as part of the flow at the inlet to the
pond system simply because they join the Mill-Willow Bypass below the pond
system.
A focused evaluation of ARCO's plan was conducted by EP A and MDHES and i (
rart of the administrative record. This analysis enabled the agencies to devise and
-------
. select the remedy described in the ROD, which combines portions of the original
Proposed Plan and ARCO's plan.
2.4.14 Monitoring and Long-term Maintenance.
Several commenters (Letter 101, 108, 126, 138) noted that the FS does not include
future monitoring plans, and expressed an interest in reviewing the monitoring plan
, when it is developed. Two commenters (Testimony M-9, M-10) recommended that
the agencies establish flow measure~ent stations upstream and downstream of the
ponds and that a comprehensive monitoring program be developed to gauge the
effectiveness of the proposed alternatives.
Response: Monitoring plans are generally not developed in the feasibility study.
These plans are normally developed during the remedial design phase after the
decision is made concerning which alternative will be implemented. The public will
have an opportunity to review and comment on the draft of the proposed
monitoring plans once they are developed.
The FS did consider the need for monitoring, and the operations and maintenance
cost estimates include allowances for such monitoring. EPA and MDHES agree
that flow measurement stations are desirable for future monitoring. Flow
measurement stations (including water quality monitoring) will likely be part of the
long-term monitoring program to be spelled out in the monitoring plan. It should
be noted that the USGS did maintain a gaging station (No. 12323750) on the Mill-
Willow Bypass just upstream from the confluence with Warm Springs Creek. This
gaging station was maintained from April 1972 through September 1979, and data
from this station were used in the preparation of the FS.
One comment (Letters 136, 146, 147, 148) asked who will be responsible for costs
associated with replacement or maintenance of the ponds in the future?
-------
Response: ARCO will probably be responsible for these costS.
2.4.15 Miscellaneous Comments Regarding Alternatives.
One commenter (Letter 154) thought it appeared that cataclysmic events had
received more attention in the FS than the slower erosional processes which "playa
larger role in the transport and enrichment of toxic metals in downstream
> environments."
Response: The FS addresses both the cataclysmic events and the slower year-to-
year processes that tend to add up to significant movement of contaminants. The
alternatives developed address both types of processes. The pond treatment system,
if upgraded and properly operated, will provide a valuable barrier to the erosional
processes that could eventually carry the majority of the tailings remaining along
Silver Bow Creek to the Clark Fork River.
-------
ATIACHMENT TO ill-A
PUBLIC COMMENT CROSS-REFERENCE
-------
z
0
.. ~ II) .. ..
:> ~ '" g ~ ~
~~ c c c o . ~ ~ c c ! -03 :>
o 0 0 ; c . . ~ 0
~~ ~.!! ~i ., . ii: :J . c ~ E E c '" c ~ ~ ..~ .. . . .
.. ~ .. c E ~ 01 0 C 0 "0 c.. ~ c .2 .. > u ~ ~ c
..." .-:> E:i a. "' ... ... II ~ .. >- a. c U. ,,'" ...~ ~11 WI .., ~~ II... ... II j" "'. ., c ~
",:I "'... "3 ~ OlE ~ >- :> -= ...! . c 01> E c c E ~;;; ... o. "3
. 0 II. u'O C .. ~~ E. .!!:i ce "'. ~:> .c. c 0 co!! :J .1 = E ~ ~~
z:l E ~ .t: E ~~ :> c.- ". .... 0'" ~ 0 :>~ .- c
E.c .. -=... t:~ ~ .. '" 0 " .. 0 o~ ~o :: E -= u :IL 0" " ~"' .- ~'.
. ~ 0 iiJ c j6 c;~ .-. 00 o. a.a. 0 0 ::~ Ij ..
~8 ~~ . u ~. ~ ".. Jlj 0- ~ '.. '.. t;~ 0 ~ .
cell) .!OIL ~IL "'C i; ",,11) II)U U:~ U:~ ;:)! iLII: ",II: ....ce ....a: ~.5 U ~m
-------
z
0
.. ~ I/) , It
C C :J ~ .., <; c ~
I/) c 0 " ~ ~ :: c ~ e;.3 :J
~ffi 0 0 ~ .!! .. :; c .. . .!! ) 0
~~ .~.! .. c iii c ii: :J . '" c ~ E E c '" c ~ ..~ .!! ' ... c:i .
> . . 'c E ~ 0 c 0 .. 0 M'ii" ..iii c It . C WI > u c
.- :J a. ~ "' . a. .!: U. ." .., :J.- .." ".. j WI ". .,. c 'C c ~
..,::1 ".. "" E:i Ii I:i " WI.. .. ~~ "' :J z ,,!S .. c a<; a> aE E c .... 0..
.. 0 u'iI c "'E u ~ ~ "Ii " ~ c E ~;; Ii
.... 'C E .. .. " 8;.: E ii WI c a: :: :J "" C 0 c~ :J ..l:a E ~ ~ ~
z::l E:" ~'5 u :J C .- ... . ~ 0 :J~
E.c .. ~ .. ~ 0 "''U t~ ~ .. -.. .., 0" 0.. .. 0 0 ~ t:~ '" E := IJ :0.. 0:1 WI ~" .- "c.5 u
~8 ~~ .. u ~" :i 'if. J!~ .." ~ ~d ~~ .-.. 00 0" a.a. 0 0 .::~ ~;. ~~ :: ~ ~j ~.. 0" ..
If 0.. 0 a::l 0- ~ ii:::i ii:~ ;)! ii:d: ~.5 "t;: 0 i
a:1/) So.. ..,c 1-1/) I/)U u ::im cd: ::i::i
~Q
.: N I'! .. oft .: N ..; .. .: N "; .: N ..; .. .; ,.; ,.: to .,; d - N ..; '!. oft
- - - -
26. Shlrlev & Duane Culwell . . . .
27. Mr. & Mra. Martin . .
Parent
28. Michael McGee . .
I 29. Don Hewitt . .
. 30. John SIewart .
31. DalAd McGee .
. 22. Georee Niland . .
33. Raymond Deakins & .
Larry & Donna SwoDe
34. Pele Reisenauer, Jr. . . . . .
35. Peler Relsenaur . . .
36. Robert Daniels . .
37. Gilbert Fischer .
38. SI6n Preshar .
39. Mr. & Mrs. Andy . . .
Anderwald
40. Judy Mlnnehan . .
41. Euaene Caluccl . . .
42. Maxine & DarreD Baker .
43. Millie Malhews . . . ..
44. Terrv :JDear . .
45. Jed Thomas . . . .
46. JOI' Roberts . . . .
47. MellAn SIokke . . . . . .
48. Robert Vine . . . . .
49. Nicki Leiss . . .
'j" , "',', .
--
-------
z
0
.. ~ III ..
III c c c " ~ III 0 ~ ~ ..
I 0 0 o . ~ ~ b c c :1 ii3 "
.!i ~~ .. ~ 0 . a: 4 :1 c c ~ . . 0 . 0
~III .~ .. c ~ j . E E c ... c ~ .~ .. . . .
>. t . .. c E /l '" 0 c ~ .0 .. c.. . c .!! .. > u c:i c
.... ._" E:i " ... ... . " .5 u. .... "." " ." :1.2 .... .... j" .... .. c 1: c ~
111::1 "'... . .. >- >- " . c
. 0 .. ~lJ c "'E u ~ .. ~ ~~ .... ...! ~! O'u 0> E c c E c- .. o. "3
z::l E ~ 'C E u " c- .. .... 8:.= E; .!!:i a:: b " ". c 0 c.!!l ~ 0 ,,~ :I 1::1 ~"" c ~ ~ ~
Es: . ~ ~~ .. ~ 0 :'='0 t:~ ~. III 0'" o. WI 0 t:~ "" E ::: U :IL O. ~... ... - c
"'8 ~: . u ~. :i '.. ~~ WI" 0( ~~ ~~ .-. oo .!!! .." 0 0 WI :: ~ .- ~i ..
a::111 .51L ~IL " a:c'l 0- ~ iL:i :d~ iLd: .0 ';0. ~~ .. c ;;~ 0 ~j >:. :i
~ 1110( ~III IIIU u.~ ...d: ~a:: ~- u :ill! ced:
t::z:
IIIU
.: N ..; .; wi .: N ..; .. .: N "; .: N ..; .. wi oD ,.: .0 .. d ~ N ..; '!. .;
I ~
51. F.R. Bennett. ADLRA . . .
52. Jim Keane. MT .
Operating Engineers
Tralnlno ProDfam
53. Missoula Co. Comm. . . . . . . .
54. SI~ve AnlonioU .
55. Robin Richards . .
56. Marv Beer . . . . . .
57. Robin Richards . .
;
I 58. Serae Myers . . .
.59. BIC'nda Patrick
.-.. .
00. Robert & Nlta Pertman . . .
61. Jack Atcheson .
-
62. La'lrence Siroky. DNRC . . .
63. Jack Jones . . .
64. Jean Peck. SAndpolnl . . . . . . . . .
10. CEntral Bus. Assn.
65. CItv 01 Sandnolnt. 10 . . .
66. Cllv of Kootenai. 10 . . . . . .. . . .
67. Jot! Moreland. USGS . . .
68. Frank Rives . . . . . . . . .
69. CII.. 01 Clark Fork. 10 . . . . . . . . .
70. James Murrv. AFL-CIO . .
71. Sam Worcester . . .
72. Kemper McMaster. . . . . . .. . .
USF & WS
...!..!....l:'~' ...'(.. '['I . . . . . . . . .
. . . . . . . . . .
..l'-. -
. . . . . . .
.
. -
-------
z
0
.. ~ I/) ...
1/) " " :J ~ '" g ~ ~ "
" 0 II ~ ~ " ,, ~ ~3 :J
~ffi 0 0 ~ 0 II :; " II II 0 ~ 0
;1: ~~ "" ., ii: :J II ,,~ E E " ", " ~ ..~ .... ; 3 ..
.! u > II .. 3 ~ .... " E ~ 01 0" 0 .. 0 ., "" ~" .!! " .~ ~ 3
..." '-:J E 5 a. "' ... .. II II " ,.., a. Z .!: U II ..'" .., :J.- 1fI~ 3.; "... ... II ~.. "'.. .."
",::II .. 0 "'... U'O . \j " OlE u ~ " ~ ~ ~ ,.., :J .... ...! II" ... ~ CTt; E" "E j;: .. II 0.. "3
z::ll E ~ ".. 'C E ~.~ U :J ".- .. II '" II 8:.: E G ,," a: 0'" :. :J .c: II Ee ,,~ ~ 0 ~i :J J.::J E ~ ~~
E.c: .. ~ .. ~ :':'0 t:~ ~ .. -.. '" 0.. ,,0 0 ~ t:~ :.: U :11. ~... .~.~
o .." « 36 ..- .-.. 0 0 .2 ~ 0.0. 0 0 .. ::~ 11- "
"'0 .. a. II u ~ II i u ~:; ~3 a:~ 0- ~ .. ~ ~~ ~~ 0 ~~ ~..
I:JU a:« a:1/) .511. ill. "'« c~ I/)U i&::! LLI- :)! i&:n: "'II. ~.5 I:J~ u :!m en: :!:1 i
t:x
I/)U
.: N ,.; ... on .: N ,.; ... .: N "; .: N ,.; .. on ..; ..: .; ai d ~ N ,.; '!. on
~ ~ ~
76. Al\I:.conda Chamber of . .
CommercO!
77. Terry Schoonen. Skyline . . .
SDortsme,l Assn.
78. MIc..ha~1 Grayson . . . . . . .
79. EdGar Butler . . . .
80. Pea Patterson .
. 81. James Davtson, . .
Anaconda Local
Develo']m~nt
82. Kenneth Bender . . .
83. Slaole'l Glovan, Jr. .
84. RoLin Richards .
85. Jck Dl-1fteld . .
<36. MIII,e Malhews . .
-
~7.An6conda~eerLodge . . . . .
tleclarr.atlon Advocales
.-.
86. Sle'/e Loken . .
89. Ber1ard Shaw . . . . . . . . . . . . . .
90. Freder!ck T ossbern . . . . . .
91. Wesley Wilson . . . . .
92. Rohert Berrlson . .
93. Bill Burnett . .
94. Paul & Dorolhy Gambee .
95. Boh Greene . . .
96. Allee & Brlnns Austin .
97. KathY :.hlenslaner . . . . . . . .
98. J. Geo.ae Hummel . . . . . . . . .
99. Ma, k Conr.ell . . . . . . . . .
-------
z
0
.. ~ Ct) ...
c: c: :> ~ ~ UI '0 C ~ to
,Ct) c: 0 : - i b c: .: (83 :>
~~ 0 0 0 c: . . ~ 0
:; ~ .~.! ~i ., . a: . :J . c: - E E c: " c: 0 - . - ... .. .
j ~ '" 0 c 0 . 0 '., c:.. ~ c .!! .. > u ,~~ c
. ... c E /2 .. 4'- ~~
'0. '-:> E:Ii Do UI '0 ... . .. ~?: Do .5 u. .'O 'U :> '., "'0 '0" ~.. '0. ., C ~
",::I . 0 '0'0 U '0 8 ~ c "'E u.. .. ~ ... :> '08 'O! . c O'u ",> E c c E j;; ." o. 3
. . E ii .!!:; a: ~:> '0. :J J. ~ c" --
z::l F ~ 'C E ~~ u :> c ,- .., .." 80:: 0'0 - 0 .c., c 0 coS - 0 :> - .. .
E.c ... to ~ 0 =="'0 t:- - . UI o. .. Do 0 - Sc; = E :.: U :IL O. to -'0 .- ].~
~~ to" ~ ~d C;~ .-., 0 0 .!! ~ 0 0 ::~ to
~8 . u -. :i i; '.. ~~ iX:l 0- ~ DoE 'ii. 'ii" O~ 0 ~~ -" i
a:Ct) SIL llL IIIU u:~ u:ot ...ot .. c: ~m Ciot ~~
UlC ...111 ...... ;:)- ...a: ...a: ....- U
t:X
IIIU ~ N vi
.: N ..; .. oft .: N ..; ..; .: N "; .: N ..; ..; vi .; ,.: .; cO - ..; -!.
- - - -
101. Missoula Water . . . . . . . . . . . . . .
Qualltv Advlsorv Group
102. John Roskellev . . . . . .
I 103. J:>hfl Wozniak .
I 104. Feter Forsvthe . . . .' . . .
. . .
105. Rob'n RIchards
106. RIchard & Edle Smith . . . . . . .
101. Jodie Canfield . . . . . . . . .' . .
108. Mls!.oula CIty.county . . . . . . . . . . . .
Board 01 Health
109. Marlt Story & Melissa . . . . . . . . .
Miller
110. Rob~rt Daniels .
111. Ron PIerce . . . . .
112. Carr & Sherrle Parker . . . . .
113. (;ene Murphy, Mile . . . .
Conservetlon 0151.
1 14. Joh~ Lenosta" . . . . . .
115. Johr. Fraley, Amer. . . . . . . . . . . .
Fisheries Society
1 i~. JlJmes Greene . . . . . . . . .
111. Cod" Brooks . . . .
..
118. Richard Faust, . . . . . . . . . . . . . .
Blnerr'Jot.Mlsslon Sierra
':Iub
119. Dsvl1 Kebler . . . . . . . . . .
120. JHmes Daub . . . . .
i21, OJUQlss Webber . . . .
1:',' ' . I'. .t'I,o' H.,tl . . . . . .
--
. . . . . . . . . . . . . .
..._-
. . . .
- . . -
. . . . .
-------
z
0
.. ~ III ....
1/1 c: c: c: :> ~ "" ~ c ~ ..
o 0 0 :1 ~ ~ C: :1 'C83 :>
~ffi ~ 0 .. c: .. . . 0
:~ ~ ~~ .. c: 0> c: 1t " :J .. c: ~ E E C: " c: ~ ~ ..~ .... .." ..
> .. .. . .... c: E ;2 '" 0 c: 0 .. 0 c:.. . c: .2 .. > u .5 ~ j
..." .-:> E:i a. .!! "' ... .... .. a. c: u" ,,'" ''' '''- ...1 ~2 .... ..." j" ..... .., c:
",2 "'''' .. 1; .. ~?: ... :> .: ...! .. c: o-t; .." ~ c:
.. 0 U'O c: "'E u ~ ... ~ E c: c: E c:- 0 ..
E;o. ".. 'c: E u "" .. ~ 8:;: E ii .!I:i It ..... ~ " ".. !~ c:.!! :J .1:1 .!!'" E ii ~ ~ ..
z2 :ii~ :> c: .- ... .. ~ 0 ~:
Es: .. ~ .. ~ 0 :;~ 't:~ ~ .. "" 0'" 0" ~ 0 O~ r ~ ::: U :IL ~"" g.~ u
~8 ~~ .." c ia a~ .-.. 00 .2 ~ .. a. 0 0 " ;.:~ .- ..
.. U ~.. ~ " ~i a::l 0- ; a.E "!! ~~ 'ii" 0 ~j ~..
Itfl) .51L ilL "'C 1-111 IIIU iL2 1&.1- :;)- iLd: "'IL 1-11: ~.5 CI':: U 2m cd: 22 ~
t::J:
IIIU
.: N ,.; .. vi .: N ,.; .. .: N "; .: N ,.; ... vi ..; ,.: ..; ai 0 ~ N ,.; '!. vi
~ ~ ~ ~
126. Cltv of Missoula . . . . . . . . . . . . . . .
127. Brian Jameson .
128. Bonner Boundary . . . . . . . . .
Board of Realtors
129. William Macareaor . . . .
130. Chris Marchlon . . . .
131. Frank RIves . . . . . . . . . . . . . .
,132. Dkk Baroni . . . . . . . .
133. Mllry Leonard . . . . .
134. Pa\Jt Roos. Big . . . . . . . . .
Blackfoot Trout Unlimited
135. Rlc.h Day, MT Wildlife . . . . . . . . . . .
Fed.
136. David Atkins . . . . .
137. Jonn Dulfteld .
138. Patrick Graham. . . . . . . . . . . . . . .
DFW&F"
139. 11m Sullivan . . . . . . .
140. Lynda Saul . . . .
141. ARCO Coal Company
(See Section 31
142. Janet Ellis, MT . . . . . . . .
Audobon Council
143. Arlene Dierker . . . . . . . . . .
144. Denise Moldroskl Mon . . . . . .
145. Kathleen Mclaughlin . . . . . . . . . .
146. Chris Kaufman, MEIC . . . . . . . . . .
147. Scon Luchessa . . . . . . .
148 Olr ~ SOlum . . . .
'49 5('I.r1 '/\,'8hl"). (hUll. . . . . . . . .
fl.,. d' .",
-------
z
0
.. ~ I/) -
I/) c: c: :0 ~ UI 0 C ~ ..
c: 0 D ~ i :' c :! -:: :0
~ffi 0 0 ~ o . i c D D .g ~ 0
:; ~ ~~ .. c ., j ii: :J .. '" c ~ E E c " c: c.. "~ ,g - . CO "'c ..
> . " UI - c E /J. Go 0 C ,," .g " 0 -1' :L; ~ c .. > u .~ ~ c
"" '-:0 E 5 Go " .. co . .. ... ~ U co :0'" D" "D j" ".. ., c: .!II
UI::I . 0 "" U'i] " Ij c "'E U ~ .. ~ ~ ~ "' :0 ,,! .. c ,,~ O"u E c c E c- "co '0.. ..
z::l E Ii .. '1: E ~l U :0 c.- "co .. co II:.:: E ii 11:i a: "CO ~ :0 IC co ~~ c~ ~ 0 :o~ :J .L:I .!II"" E ~ ~-
Er. .. ~ ='0 t:~ ~ .. UI 0" 0" ~ 0 o~ t:~ ::: u :0.. 0 co -" 1'~ U
. ~ 0 .." e( 3d ~~ .-. oo .!! ~ .. Go 0 0 .. :: ~ .- ..
~8 ,;;~ co U ~ co :i ,, ~.:: ~3 «.1 0- ~ GoE " 0 -;.. ~: ;:,~ 0 co- :: co
a:l/) Eo.. Ule( Cl)U iL::i 11..- ;)- iLd: ",d: .-a:: ~.5 u ::im ~i
-------
z
0
.. s CII .. ..
:> ~ III '0 ~ ~
CII c c c 0 : ~ ~ :> " C : .. II :>
~ffi 0 0 ~ 0 c .. .. ) 0
~~ .~.!! .. c .. .. ~ . :t .. c: ~ E E c: .,,, 0 ~ .~ .. .... ~:i ..
> .. . j ) .... " E II at 0" 0 . 0 .. c:.. )" .!! .. > U j
.... '-:> E:i '" III ... .... .. ... " U" .'" .. :>., ..;;; :J.~ "... ..." ~.. "'.. .. c: 'C c:
111::1 .. 0 "'... U '0 'i t; " at E u.. :I ~ ..!: ,.. :> 1.: ...! .." ... ~ au at> E c c: E j;: ... 0.. 'i
.... 'C E U ... 8:: E ii .!I:i a:: ..... :> :> 1:.. c: 0 c:.!! ~ 0 :t i ~ ,," ~~
z::l E Q. ii :> c: ,- ..... ~. .. .. U
EI: .... .. ~ 0 ='V t:~ ~ . III 0'" o. .. 0 o~ ~D := E == U :IL .. ~"' ..- 'c.5
~,8 ~~ .. U ~" :i '... J!~ i~ C ~d ~~ co!! ~ 0 0 .2! ...'" 0 0 ,... ,... .." 0 :::~ ~~ 1::" o. ..
a::CII .ElL IIIC ., ....'" CIIU iL~ ....... ::>~ iLct ...ct ....a:: ....a:: ;;,£ CJ~ U ~ID cct ~~ :i
1:%
CIIU ~ N ..; '!. ,,;
.: N ..; .. wi .: N ..; ... .: N "; .: N ..; ... ,,; ..; ..: ..; ai ~
~ ~ ~ ~
151. Peter Nielsen. Clark . . . . . . . . . . . . . .
Fork Coalition
152. Tom FRance. Natlonel . . . .
Wildlife Fed.
153. Land UndberQh . . . . . .
154. James O'Toole . . . . . . . . . . .
155. Sherr, Johnson . . . . . . .
156. Dan Ueland. .
Headwdtera At & D
157. Klrn Relneklf1Q. West . . . .
510....1 rout Unlimited
158. MO!lvIn Stokke . . . . . .
159. Dcn a. Rebecca . . . . . . . .
Holland
160. D.R. Stoecker . . . .
161. Claude RosenberQ. Jr. . . . . . . .
162. Judith Carnes .
--
- --
-------
z
0
. ~ 1/1 .... .
1/1 " " ,, :> . - 111 "0 C ~ :>
o i :> " :J '03
~ffi 0 0 - '!! . ; " . . 0 ~ 0
i~ ~~ ." ~ :! . ,,- E E " ... " - ..- .... .. .
> . . j .... " E ~ '" 0" 0 .. 0 .., ". ~" 51 ., >" .~ ~ "
..... -:> E:i 0. III ... ., II . o. .!: U. ..'" ", :>.- III 4' r!.~ ." ". ~., .... .." ~
1112 . 0 ...... " '0 Ii U " "'E ,,~ ., ~ ~?: ... :> .... ,,! ." ... ~ uti ",> E" " E j:: .. II o. .
.. 'C E .. . . 11:= E- ." II: :> :> '" OJ C 0 c~ :! J.:I c ~ --
z2 E ~ ~~ " :> ".- ... II - 0 :>-
E", . ~ . ~ =" t:- - .. -.. III 0" 0.. . 0 0 - t:- :: E :.:: u :G. 0 OJ -" ~ II .~.~ "
~~ 0 .., '" ~c3 d~ .-. 0 0 .2! 0.0. 0 0 ., ::~ .- ..
~8 ." -. :i 0 ~J: ,j;~ iic1 0- ~ .:let ~: .~. ;;~ 0 ~~ -.
11:1/1 .!:G. ~G. I/IU u:~ ::d~ u:et ~" ~CI) Clet ~~ i
~~ ILl- 1-11: 1-- U
I/IU g ... ..,;
,.: ... I'i .. .,; ": ... I'i .. ,.: N I'i ,.: N I'i ... .,; ... ..: .; .,; ~ I'i '!.
~
M-13. Unda Heldlna . . . . . . . .
M-14. JIM O'Toole . .
M-15. JIM Carlson . . . . . . .
M-16. 8a:TV Dutton .
"NACON~ HEAlING
A-1. f\IIckl Leiss . . .
"-2. Connie Ternes Daniels . . . . .
A-3. 8'" Williams . . . .
,,-4. AI Clark . .
A-5. F~ank 8eMetl .
A-6. Chuck Haellner . . .
A-7. Gene Vuckovtch . . . . . .
A-6. Mel Stokke . .
".g. Terr; Spear . . .
"-10. ~obert Daniels .
A-11. Mlck Fuller . .
A-12. 'led Menahan . .
"-13. .Jer-y Gallegher . .. .
A-14. Mike Gravson . . . . . . .
A-15. 3ruce Farllna . . . . . . . . . . . .
-------
NOTE:
PART B - RESPONSES TO ARCO COMMENTS
The re~ponse to ARCO comments will be formatted in the exact outline of
-------
CHAPTER 1.0
INTRODUCflON
No comments
-------
CHAPTER 2.0
RESPONSES TO ARCO COMMENTS, CHAPTER 2.0
SITE DESCRIPTION AND CONTAMINATION
CHARACTERISTICS
2.1 SITE BACKGROUND
Comment: This comment states that the responsibility for administering 'the Warm Springs
Ponds is inaccurately described in the FS, .because the FS fails to note that the Warm
Springs Ponds are a regulated facility under the Montana Pollutant Discharge Elimination.
System (MPDES).
Response: The intent of this section of the FS was to provide a general description of the
background and history of the Warm Springs Ponds. It is recognized that ARCO holds an
MPDES permit for discharges from Pond 2, and that the MDHES Water Quality Bureau
administers regulation of ARCq's compliance with the provisions of the MPDES permit.
The permit will continue to be required for the Pond 2 discharge.
2.2
SITE DESCRIPTION
2.2.1 .f.hysiogra.p~emograpby
No comments.
2.2.2 Climate
No comments.
2.2.3 Geology
No comments.
2.2.4 .fuills
-------
Comment: This comment states that soils within the Warm Springs. Ponds
Operable Unit bave not been affected by smelter wastes, waste rock, and
leacb pond deposits and that there is no information to indicate that waste
rock, leach pond deposits, or smelter wastes exist within the operable unit.
Response: Smelter wastes associated with historic discbarges from the
Opportunity Ponds into the Warm Springs Ponds are identifiable adjacent to
the two cbannels connecting tbe sites (SS-23 and SS-24," Figure 2-6 of the
FS). It is known that the Opportunity Ponds were used in conjunction with.
the waste streams. emanating from the Anaconda Smelter; bence," tbe
deposits present within the Warm Springs Ponds Operable Unit adjacent to
tbe channel connecting tbe two sites are presumed to be associated with
wastes derived from this source. Direct evidence of leacb pond deposits and
waste rock within the operable unit is lacking. The terms were used in a
generic sense to relay to the reader that tbe site has been affected by waste
material related to mining, milling, and smelting activities in the Butte and
Anaconda areas.
2.2.5 Surface Hydrology
Comment: This comment questions the value of 700 cfs as the capacity of
the Pond 3 inlet structure and states that the actual inJet structure capacity is
1,400 cfs. It also notes that plugging of the structure can cause washout of
the fuse plug at less than 1,400 cfs.
Response: Calculation of the inlet structure capacity (Silver Bow Creek
Flood Modeling Study, CH2M HILL, 1988) indicated a maximum flow rate
of approximately 900 cfs under ideal conditions. This estimate was
downrated to 700 cfs to account for plugging, age, and actual field
conditions. Plugging that causes premature washout of the fuse plug is noted
on page 7-26 of the FS.
-------
2.2.6Ground Water Hydrogeology"
No comments.
2.2.7 Land Use
1. Comment: This comment states that the description of land use
within the operable unit is incomplete because the principal land use
(water treatment and mining waste disposal) is not specifically
identified.
Response: As stated on page"2-5 of the FS, the Warm Springs Ponds
were originally developed to control the amount of sediment and
tailings carried into the Clark Fork River from Silver Bow Creek.
Page 2-6 of the FS indicates the ponds were not used as a water
treatment unit until 1967 when Pond 3 was converted from a
sedimentation pond into a treatment pond. Currently, a principal use
of the area is for water treatment. Other principal uses are
recreational and occupational uses, associated with fishing, hunting,
and wildlife management.
2.
Comment:
This comment states that the area and volume of
contaminated soils provided in Table 2-2 are not representative of
actual soils conditions below a depth of 1 inch, the maximum depth
investigated. The volumes for contaminated soils should be based on
contaminant action levels or a representative sampling methodology.
Response: The areas and volumes of tailings and contaminated soils
presented in Table 2-2 were not based on contaminant action levels
nor were they presented as such. The areas and volumes are values
estimatec;l through evaluation of data collected during the Phase II
-------
Remedial Investigation (RI) of the Warm Springs Ponds Operable
Unit.
As stated in the Phase II RI Data Summary Report and in the FS, calculation of
the areal extent of contaminated material in these areas was based on field
mapping of denuded areas characteristic of exposed tailing deposits, and through
evaluation of both X-ray fluorescence (XRF) spectrometer data and laboratory data
> for arsenic, copper, zinc, and iron in soil samples collected in conjunction with the
Rls. Samples from areas exhibiting arsenic, copper, zinc, or. iron greater than.
50 percent above the levels in' adjacent areas were classified as contaminated.
Hand-auger borings were made through these areas to determine thickness of the
deposits. It is not true that the "maximum depth investigated" was one inch.
Thicknesses of those areas determined to contain tailings and contaminated soils
were identified visually by lithologic color change; this break typically correlated
with decreases of 30 to 50 percent in XRF and laboratory data for copper, zinc,
arsenic, and iron.
A total of 115 sites were sampled in three portions of the Warm Springs Ponds
Operable Unit during the Phase II RI. Numerous samples were collected at these
sites to characterize materials chemistry with depth. Sampling methodologies used
during the RI were standardized and completed in accordance with the project
sampling and analysis plan. In addition, more than 50 hand-augered boreholes
were drilled in material adjacent to the Mill-Willow Bypass to aid in visually
determining thicknesses of tailing deposits and contaminated soils. These data were
incorporated into a field map showing the areal extent of various materials
identified in the bypass area.
This method of site characterization provided reasonable estimates of the volumes
of contaminated materials and provided a basis for cost estimates for the various
remedial alternatives. Detailed field characterization of the areas and volumes of
-------
materials that exceed contaminant action levels (to be established in conjunction'
with the ROD) will have to be completed during the design investigation phase.
2.3 NATURE AND EXTENT OF CONTAMINATION
1.
Comment: This comment states that Figure 2-1 should include identification
of the current MPDES compliance point at Pond 2.and the Wildlife Ponds
point of discharge into the Mill-Willow Bypass.
Response: The intent of Figure 2-1, as indicated by its title, was to illustrate
conceptually the pathways of contaminant migration within the Warm
Springs Ponds Operable Unit. The surface water media and Warm Spring
Ponds operation are described in detail in the surface water section of
Chapter 2 (beginning on page 2-35) and in Figure 2-3.
2.
Comment: This comment describes Figure 2-1 "as a diagram that is used to
show the areas of contamination and migration pathways associated with the
four contaminated media: (1) pond bottom sediments, (2) surface water,
(3) tailings deposits and contaminated soils, and (4) ground water." The
comment states that Figure 2-1 is incomplete because certain deposits of
tailings and contaminated soils are not shown in the figure.
Response: The intent of Figure 2-1 was to conceptually illustrate
contaminant migration pathways at the Warm Springs Ponds Operable Unit.
The figure was not intended or prepared to identify locations of all
contaminant source areas within the operable unit.
2.3.1 Sediments. Tailings. and Soils
-------
1.
Comment: The comment states that, based on the volume and area
indicated for contaminated soils present both adjacent to and
underlying the tailings deposits along. the Mill-Willow Bypass, the
average thickness of contaminated soils is approximately 2.4 feet.
The comment goes on to state that the estimate of extent and volume
of contaminated soils cannot be independently evaluated .lor the
following four reasons:
a.
Because analytical data for soils are combined with analytical.
. .
data for tailings deposits, it is not possible to determine which
analytical data were used to identify contaminated soils versus
tailings deposits and uncontaminated soils.
b.
The FS does not specifically define "contaminated soi1." The
criteria used are not provided in either the FS or the Phase II
RI.
c.
Analytical and XRF data results are limited to samples of soils
and tailings deposits collected only from depths ranging
between 0 and 1 inch, and cannot indicate soils conditions
below that depth.
d.
Figures are not provided in the FS and Phase II RI that
specifically identify the location of the 33 acres of
contaminated soils along the Mill-Willow Bypass.
Response: XRF and laboratory analyses of surface samples were used to
define the areal extent of tailing and contaminated soils. Hand-auger
borings were used to estimate associated thicknesses and, therefore, volume
of contaminated materials. The estimates of volumes of tailings and
-------
contaminated soils developed for the FS are suitable for the purposes of a
FS and provide a reasonable basis on which. remedial technologies and
associated costs can be evaluated. . Additional data on contaminated soils
and tailings may be necessary to refine area and volume estimates in support
of remedial design, once the preferred alternative for the site has been
selected and contaminant action levels have been established for site soils.
See the response to Section 2.2.7, Comment 2, for a more detailed discussion
regarding criteria used to define contaminated soil and estimate volumes and
acreages of contaminated so~ls within the operable unit. The field map
mentioned in the resporise to Comment 2 under Section 2.2.7 shows. tbe
areas of contamination at the bypass area.
2.
Comment: This comment states that XRF data are useful only as a general
indicator of metal concentrations in soil and tailings, and that, if XRF data
are. to be used for other purposes, rigorous Quality Assurance/Quality
Control (QA/QC) procedures should be used.
Response: XRF data collected at the Warm Springs Ponds Operable Unit
have been used only as a general indicator of metal concentra~ions in soils
and tailings. Quantitative soils and tailings data collected during the Phase
II RI at the Warm Springs Ponds included laboratory analyses of collected
samples and field measurements and observations. These data were used in
conjunction with the qualitative XRF data to characterize site conditions
relative to the needs of the FS process. Thus, the uses of XRF data in the
FS are appropriate to its level of accuracy. Additional quantitative data may
be necessary to support remedial design once the preferred alternative is
selected after contaminant action levels are established.
-------
3.
2.3.2 Surface Water
1.
Comment: This comment notes that the Silver Bow Creek 100-year
flood is sometimes listed as having a peak flow of 4,000 cfs and
sometimes as having a peak flow of 4,900 cfs.
Response: The 100-year flood on Silver Bow Creek alone was
calculated to be 4,000 cfs. The 100-year flQod on Silver Bow, Mill,
and Willow Creeks. cOqIbined was calculated' to be 4,900 cfs.
2.
Comment: This comment states that an independent analysis
performed by ARCa (1989a) using U.S. "Geological Survey (USGS)
methodology indicates that the peak flows are 3,300 cfS (Silver Bow
Creek alone) and 4,000 cfs (Silver Bow, Mill, and Willow Creeks
combined).
Response: The issues presented in this comment were discussed
during meetings with ARCa. Because of the uniqueness of
hydrological analysis, several approaches are possible to provide
quantitative results on flooding values for a basin. Because ARCa
used a different approach, the results of the two studies were
somewhat different. See the response to Section 4.1.1, Comment 3,
for additional detail.
Comment: This comment questions the value of 146,000 cfs for the
PMF on Silver Bow Creek. An independent flood analysis by ARCa
concludes that the best estimate for the Silver Bow Creek PMF is
approximately 80,000 cfs.
-------
Response: There are a number of reasons why ARCQ's calculations
pr(jduced a PMF value that is low compared to the 146,000 cfs
calculated by CH2M HILL. These reasons include an incorrect value
used by ESA for the peak one-hour rainfall amount and incorrect
assumptions of snow pack prior to the PMP. See the response to
Section 4.1.1, Comment 4, for a complete response to this comment.
4.
Comment: The value of 700 cfs cited as the capacity of the intake
structure to Pond 3 is incorrect (page 2-39). The actual capacity is
1,400 cfs, which is nearly equivalent to a 10-year flood event' for
Silver Bow Creek. Page 2-39 should be corrected to indicate that the
actual capacity of the intake structure is 1,400 cfs.
Response: Calculation of the inlet structure capacity (Silver Bow
Creek Flood Modeling Study, CH2M HILL, 1988) indicated a
maximum flow rate of approximately 900 cfs under ideal conditions.
This estimate was downrated to 700 cfs to account for plugging, age,
and actual field conditions.
5.
Comment: This comment states that the FS improperly applied
"chronic freshwater aquatic standards" and "primary drinking water
standards" to data collected in the ponds. The comment further
states that the surface water quality standards are not applicable
within the boundaries of the Warm Springs Ponds Operable Unit
(ARM 16.20.615), and that improper application of MDHES water
quality regulations has resulted in inappropriate and misleading
statements in the FS.
Response: The FS report cannot and does not attempt to "apply"
water quality standards to the data collected at the Warm Springs
-------
Ponds. Only the regulatory agencies can do that through the
permitting processes and through, the ROD process for the operable
unit. The discussion on page 2-57 of the FS was not intended to
imply that any water quality standards were violated at the Warm
Springs Ponds, but rather to give the reader a sense of the
contaminant levels in the ponds compared to the chronic freshwater
and primary drinking water standards.
6.
Comment: Paragraph ,3, page 2-57, indicates that flow in the Mill-
Willow Bypass exhibits an increase in hardness, sulfate, and 'zinc
concentrations between its head and its confluence with the Pond 2
discharge. The increase in concentrations of these parameters is
attributed solely to ground water inflow, based on synoptic flow
measurements and a specific conductance survey completed during
base flow conditions. The comment states that ground water inflow
to the bypass is not necessarily the only source and notes that the
increases in certain chemical constituents may be due, in part, to
direct contact of surface flows with sediments along the 3.5-mile
reach of the bypass.
Response: The sediments along the bypass were not considered a
potential major source for measured increases in hardness, sulfate,
and zinc concentrations. The statement in the FS should have been
that ground water is the primal)' source of increases in concentrations
of these parameters. Two types of data were used in reaching this
conclusion: sampling completed in the bypass during July 1988, when
all surface water in the bypass was being diverted into Pond 3; and
sampling of monitoring wells located adjacent to the Mill-Willow
Bypass.
-------
Samples of water in the bypass were collected below the point of
diversion of the surface water in the bypass into Pond 3 at Surface
Water Station SS-18C (Figure 2-:2, Phase n, RI Data Summary
Report). Water sampled at this site was ground water inflow to the
otherwise dry channel; the sampled water had flowed along the
bypass channel for only a short distance. The concentrations of total,
dissolved, and, acid soluble zinc in this sample were 112,' 67, and ,
200 p.g/l, respectively. In comparison, surface water in the Mill.
Willow Bypass, above. the point 9f diversiQn sampled 2 days later,
exhibited concentrations of total, dissolved, and acid soluble zinc, of
18, 13, and 77 J1. g/l, respectively. Likewise, the hardness
concentration in the sample collected at SS-18C, below the diversion,
was 644 mg/I as compared to a hardness concentration at SS-18 of
181 mg/I.
Ground water quality data collected during the Phase II RI at various
locations along the Mill-Willow Bypass indicate that zinc and sulfate
concentrations in ground water are considerably higher than
concentrations in surface water in the bypass, measured at its head.
For example, zinc concentrations in monitoring wells located adjacent
to the bypass ranged from 13 to 1,250 p. g/l as compared to average
total and dissolved zinc concentrations measured at S5-18 during the
Phase I RI of 45 and 28p.g/l, respectively. Sulfate concentrations in
monitoring wells located adjacent to the bypass ranged from 60 to
1,190 mg/I, as compared to an average concentration in surface water
at SS-18 of 53 mg/I.
The foregoing data, combined with an evaluation of the water table
map of the Warm Springs Ponds Operable Unit (Figure 2-12, FS),
strongly suggest that ground water inflow is the primary source of
-------
increases in hardness, zinc, and sulfate. Other sources may contribute
to the measured increases in parameter. specific concentrations, but
the magnitude of these contributions is probably relatively small.
7.
Comment: This comment states that data from the FS and the RI
indicate that upstream sources, other than dissolved metals from
tailings deposits, appear to have a substantial impact on pond system.
chemistry. The data show that the greatest variability over a 24-hour
period for non-metal parameters, including algae, occurs in the pond.
inflow. The greatest magnitude in algae, and in the nutrients which
feed algae, are found in Silver Bow Creek upstream of the ponds.
The investigation attributes these findings to varying inputs to. the
creek, including the Butte Sewage Treatment Plant and other
unspecified sources. In addition, large variability in copper and zinc
concentrations over a 24-hour period are found in the pond inflow,
indicating a varying source.
The comment states that two factors (metals loading and nutrient
supply) are especially critical in controlling Warm Springs Ponds
system performance for removal of metal contaminants, but they are
not addressed in the FS. The comment suggests that source control
restrictions on industries, municipalities, and other upstream point
source contributors may be necessary to assure that the Warm
Springs Ponds can maintain consistent compliance with the surface
water ARARs.
The comment concludes that "if the WSP system is managed as a
treatment unit for other upstream sources, ARARs which are
otherwise legal requirements for development of discharge limitations
for effluent from the WSP treatment system should be waived in light
-------
of the technical impracticability of treating surface waters exhibiting
highly variable quality characteristics. A waiver under such
circumstances is appropriate. Section 121 (4)(c) of CERClA. For
example, if the nutrient supply is not controlled then seasonal
excursions in pH (standard measure of the hydrogen ion
concentration) above 9.5 must be expected. The same holds true for
metals. If the WSP system is subjected to inflows exhibiting a wide
variation in metals concentrations, maintaining' consistent water
quality in the outflow ~ay be an impossible task."
Response: EP A and MDHES disagree with this comment in all of
the technical points it makes. None of the data from the RI indicate
that the changes in nutrient levels entering the pond system have ever
led to an upset in the pond system processes or a violation of a water
quality standard at the outlet. On the contrary, higher nutrient inputs
would likely only assist in the treatment occurring in the pond by
increasing biologic growth;
Pond 3 is a very large reservoir. Daily fluctuations in inputs, of either
metals or pH, have little or no measurable effect on the effluent
levels of metals or pH. The volume of the pond smoothes out such
small oscillations, primarily through mixing within the pond. (The
average retention time for Pond 3, with Silver Bow Creek flows only,
is 27 days.) This is why large fluctuations are seen only in the creek
upstream of the pond and not in the pond itself. To suggest that
"maintaining consistent water quality in the outflow may be an
impossible task" for the reasons given in this comment is
insupportable. In fact, ARCa has continually emphasized that its
alternative Plan 3A can adequately treat a Silver Bow Creek flood.
-------
8.
Comment: The FS should identify the fresh water aquatic standards
used to develop Tables 2-10 and 2-11.
Response: References for the standards used for Tables 2-10 and 2-
11 are presented at the bottom of each table.
2.3.3 Ground Water
1.
Comment: The last .paragraph of the Physical Characterization
subsection (page 2:'76) indicates that ground water discharge into. the
Mill-Willow Bypass includes 2.5 cfs from the Warm Springs Ponds or
east side, and 0.7 cfs from the west side; The FS should include a
detailed explanation of the methodology used to derive these
discharge rates. The methodology used is not presented in Appendix
A, or in either of the Phase I and Phase II RI reports. Without an
explanation of how these discharges were calculated, there is no way
to independently evaluate. the method used or the results. In
addition, an explanation should be included to clarify what synoptic
flow measurement method was used to enable the investigators to
confirm the contribution of ground water inflow from either side of
the bypass channel. Such measurements are best suited for estimat-
ing total inflow to a channel. They are not suitable for discriminating
between the inflow contribution from alternate sources of ground
water discharge. ~ independent evaluation of the reported synoptic
flow measurement results was not performed because information,
such as field notes, data and calculations, are not included in the FS,
Phase I RI or Phase II RI reports. All field notes, data and
calculations, which are the basis for such conclusory statements in the
FS, should be referenced and provided within the Appendix to the FS
report.
-------
Response: The rate of ground water inflow. to the Mill-Willow
Bypass waS estimated using empirical methods and direct analytical
calculations. Empirical methods provide more. reliable estimates of
ground water inflow to the bypass channel than analytical methods.
In fact, direct analytical calculations were performed by essentially
back-calculating to those inflow rates derived empirically.
Six synoptic flow measurements were completed in the Mill-Willow
Bypass channel on July 11, 1988, when surface water in the Mill-
Willow Bypass was diverted into Pond 3 at the North Opportunity
siphon channel. Because all flow in the bypass downstream of the
diversion was attributable to ground water inflow (with the exception
of flow from the Wildlife Ponds), the synoptic flow measurements
likely represent the most accurate estimate of ground water inflow to
the bypass channel.
Surface discharge was measured in the Mill-Willow Bypass at five
stations located approximately equidistant from one another
extending from the North Opportunity discharge channel to a station
located approximately midway along Pond 2 (SS-18E). An additional
discharge measurement was completed at the Wildlife Ponds
discharge. Results of these measurements indicated the total ground
water inflow to the bypass channel was approximately 2.57 cfs. (Dis-
charge measurement notes are available on request.)
Ground water inflow to the Mill-Willow Bypass from the west was
estimated by performing direct analytical calculations. Because
analytical calculations require specific data and numerous
assumptions, the empirical ground water inflow estimates were used
as a calibration tool. Direct analytical calculations using Darcy's
-------
Law for ground water flux in an unconfined. aquifer require the
following information:
.
. Hydraulic conductivity of unconsolidated material adjacent and
subjacent to the bypass channel
.
Ground water gradient to the bypass channel
.
Effective depth of ground water seepage.
Initially, direct analytical calculations were performed using hydraulic
conductivity values of approximately 55 feet per day (fpd) derived from .slug
test data of monitoring wells located adjacent to the bypass. The ground
water gradient to the bypass channel from the east was measured by
completing a level survey at five locations along the Mill-Willow Bypass
using pand elevations, static water levels in monitoring wells, and water
levels in the bypass channel. Ground water gradients measured to the Mill.
Willow Bypass from the east ranged from 3.7 percent (adjacent to the
Wildlife Ponds) to 1.1 percent (adjacent to Pond 2). The g~ound water
gradient south of Pond 3 near monitoring well WSP-GW-08 was measured
to slope from the Mill-Willow Bypass to Silver Bow Creek.
An average ground water gradient of 0.6 percent to the bypass from the west
was used for the entire area along the western side of the bypass channel;
this water is derived from the Opportunity Ponds area. The 0.6 percent
value was derived from water level data collected by TetraTech
(Geochemistry Report, July, 1986, Document Control No. TfB 160, FO)
during a ground water investigation at the Opportunity Ponds.
-------
Using ground water gradients measured at the five surveyed cross sections, a
hydraulic conductivity of 55 fpd, and an effective seepage depth below the
bypass channel of 10 feet (a depth at which all ground water seeps into the
channel), total ground water inflow to the Mill-Willow Bypass from the east
was estimated at 2.5 cfs. The rate of ground water inflow to the Mill-Willow
Bypass from the west was estimated as 0.7 cfs using the hydraulic
, ,
conductivity value of 55 fpd, an average ground water gradient of.
0.6 percent, and an effective seepage depth of 10 feet beiow the base of the
bypass channel.
Ground water inflow rates to the Mill-Willow Bypass were recalculated after
completion of a pumping test in monitoring well WSP-GW-07, located
adjacent to the bypass. Pumping test data collected at well WSP-GW-07
resulted in a hydraulic conductivity value of approximately 270 fpd. Using a
hydraulic conductivity of 270 fpd and the same assumptions used previously,
resulte.i ina total ground water inflow rate to the bypass channel of
approximately 15 cfs. Based on the synoptic flow measurements, a ground
water inflow rate of 15 cfs was determined to be excessive. Therefore, an
effective seepage depth below the bypass channel of 1 fOC?t was used,
resulting in similar groundwater inflow rates as those measured empirically.
Ground water inflow rates were calculated for the Mill-Willow Bypass to
provide data with which to evaluate the feasibility of constructing a ground
water interception trench along the bypass channel. Remedial alternatives
developed for the bypass channel do not include construction of this type of
interception trench, because it was determined that the quality of ground
water entering the Mill-Willow Bypass does not exceed Gold Book criteria
(see response to Public Comment 84.)
-------
2.
Comment: Page 2-87 of the FS indicates that ground water discharge
to the Mill-Willow Bypass results in a 30 percent increase in metal
loadings along the bypass during baseflow conditions (13 cfs). The
comment asks for an explanation of how the estimated increase in
metal loadings and baseflow value were determined and the data
used. The comment also asks whether the increase in metal loadings
includes potential increases resulting from surface water flow contact
with sediments along the channel.
Response: EP A and MDHES believe that ground water inflow to the
Mill- Willow Bypass is the primary source for increases in metals
concentrations in surface water in the bypass under base flow
condition. See response to Section 2.3.2, Comment 6, for more
detail.
The figure of 30 percent stated in the FS to describe the increase in
metals loading in the bypass due to ground water inflow was an
approximate value to describe metals data gathered during both the
Phase I and Phase n RIs. A summary of parameter-specific metals
load increases between the head and the mouth of the Mill-Willow
Bypass is contained in the Phase I RI (MultiTech, 1987; Appendix C)
in Table 3-19. This table was produced using data from low flow
sampling episodes to characterize the impact of ground water inflow
on the bypass. Additional data that indicate the magnitude of metals
load increase along the bypass from ground water inflow were
collected in conjunction with a seepage run conducted during the
Phase n RI in July 1988 when surface water in the bypass was
diverted into Pond 3. The 30 percent figure is conservatively low;
actual load increases along the bypass for several parameters (e.g.
sulfate and zinc) were greater than 100 percent.
-------
2.3.4 Air
No comments.
2.4
PREVIOUS STUDIES OF REMEDIAL ACTIONS
No comments.
-------
CHAPTER 3.0
RESPONSES TO ARCO COMMENTS, CHAPTER 3.0 SUMMARY OF THE
APPLICABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS
AND THE PUBLIC HEALTH AND ENVIRONMENTAL ASSESSMENT
3.1 APPUCABLE OR RELEVANT APPROPRIATE REQUIREMENTS
General Approach to ARARs
Attachment 1 to the Record of Decision contains the final list of applicable or relevant
and appropriate. cleanup standards, standards of control, and other substantive
requirements, criteria, or limitations (ARARs) for the Record of Decision, as well as a.list
of documents or other sources of information which are To Be Considered during the
remedy selection or during implementation of the remedy. EP A has identified the list
based upon the statutory provisions addressing ARARs found in CERCLA, particularly
section 121(d) of CERCLA, 42 V.S.C. sS~ 9621(d); the new National Contingency Plan, 40
CFR Part 300 (1990); the preambles to the proposed NCP and the fmal NCP, 53 Fed.
Reg. 51394 et seq. (December 21, 1988) and 55 Fed. Reg. 8666 et seq. (March 8, 1990)
respectively; EP A guidance documents regarding ARARs entitled "Compliance With Other
Laws Manual: Parts 1 and 2" (OSWER Dir # 9234.1-01 and 92341.02 respectively).
ARARS are cleanup standards, standards of control, and other substantive requirements,
criteria or limitations under federal environmental or State environmental and siting laws
that are applicable or relevant and appropriate to a site cleanup action. They are divided
into contaminant-specific, location-specific, and action-specific categories. ARARs and
TECs are further defined and explained m 40 CFR sS~ 300.400(g). Compliance with
ARARs is a mandatory requirement, unless an appropriate waiver is granted. 40 CFR
sS~sS~ 300.430( e )(9)(iii)(B); 300.430(f)(i)(A); and .300.430(f)(ii)(B).
-------
The new NCP was issued after the Feasibility Study for the Warm Springs Ponds operable
unit was released. EP A has used the regulations found in the final NCP to formulate the
ROD and to identify the final list of ARARs. Definitions of "applicable" and "relevant and
appropriate" found in the new NCP at
40 CFR sS~ 300.05 were used in this identification. The criteria for determining "relevant
and appropriate" requirements found at 40 CPR sS~ 300.400(g)(2), were used where
pertinent.
The primary commentor on ARARs was th~ Atlantic Richfield Company, the potentially.
responsible party (PRP) identified for this action. These responses to ARCO's comm'ents
follow the comments contained in ARCO's document entitled "Review CommentS Silver
Bow Creek Investigation Feasibility Study for the Warm Springs Ponds Operable Unit",
dated January, 1990, Chapter 3.
1.
Comment: ARAR's only pertain to on-site remedial actions.
Response: EP A agrees with this comment.
2.
Comment: Only substantive requirements can be ARARs.
Response: EP A agrees with this comment, and will use the current guidance
in determining which promulgated provisions are substantive ARARs. See
the Preamble to the final NCP, 55 FR 8756 - 8757, for a discussion of the
distinction between substantive and administrative and procedural
requirements. EP A notes that certain State procedural and administrative
requirements have been listed as TBCs, because they will be useful in aiding
agency personnel in determining the adequacy of deliverables and othe r
activities during remedial design and remedial action implementation.
. 3. ,
Comment: TBCs are not ARARs.
-------
5.
Response: TECs are defined in 40 CFR sS~ 300.400(g)(3), and their use is
discussed in the preambles to the proposed and final NCP. 53 FR 51440; 55
FR 8744 - 8745. Basically, TECs are used to examine the risk to human
health and the environment at a site, including the level of risk at a site
without remediation and the appropriate cleanup levels to be achieved at a
site. TECs can also be used as cleanup levels themselves, when there are no
existing promulgated levels. TBCs are also used to develop and examine an
ongoing remedy, such as using RCRA closure guidance to aid agency
personnel in reviewing desi~ plans and specifications for a given remedial.
action.
4.
Comment: ARARs need only be obtained at the completion of a remedial
action, not during remedial action.
Response: EP A disagrees with this statement. As stated in
40 CFR sS~ 300.435(b )(2), appropriate ARARS must be met during the
implementation of the remedial action (the RD/RA stage), as well as on
completion of the remedial action. EP A believes that this interpretation will
ensure that remedial actions will be carried out in a sound and safe manner.
The discussion at 55 FR 8755 - 8756 elaborates further on EP A's rationale
for thi~ position. EP A also notes that removal actions must also meet
ARARs to the extent practicable.
The ARARs list, Attachment 1 to the ROD, attempts to define in a clear
manner which ARARs are to be complied with during the RD /RA phase.
which are to be complied with at the completion of the RD/RA stage and
thereafter, and which are to be complied with in both circumstances.
Comment: Variances, exceptions, exemptions and waivers may be ARARs.
-------
Response: EP A agrees with this statement.
6.
Comment: Where two or more requirements are potential ARARs, the
agency must. select the most appropriate requirement for the site, not
necessarily the most stringent.
Response: EP A disagrees with this statement. ARCO is confusing the
. .
initial determination of relevance and appropriaten.ess with a fmal decision
on compliance. As the. pr~amble to tbe final NCP states, "CERCLA .
requires that. remedial actions comply with ~ requirements that. are
applicable or relevant and appropriate. Therefore, remedial action has to
comply with the most stringent requirement that. is an ARAR to ensure that
all ARARs are attained. ... (T)he degree of stringency of a requirement is
not relevant to the determination of whether it is an ARAR at a site and
must be attained (except for state ARARS)." 55 FR 8741.
7. Comment: Location-specific ARARs primarily contain administrati.ve
requirements which are not ARARs.
Response: Location specific ARARs contain both consultative and other
administrative provisions and substantive provisions. Only tbe substantive
provisions are ARARs, but often, it is appropriate and helpful to consult
with the agency with expertise in the particular area of concern, to
determine the exact nature of the substantive provisions of the ARAR. For
example, compliance with the Endangered Species Act requires, in some
circumstances, mitigative measures to be undertaken during construction on
site. The U.S. Fish and Wildlife Service is experienced in addressing these
concerns, and will often be consulted by EP A in determining what exact
mitigative measures should be undertaken for a particular site cleanup.
-------
9.
10.
8.
Comment: ARARs must be promulgated and effective. as of the date of the
Record of Decision.
Response: EP A generally agrees with this statement. ARARs are "frozen"
at the signing on the ROD. One exception to this general policy occurs
when a component of the remedy was not identified when the ROD is
signed. EP A reserves its rights to identify ARARs at the. time the
component is identified, in such a situation. Additiona1ly~ EP A will consider
ARARs promulgated after th~ date of the ROD during the five-year reviews. .
EP A will look at the new ARARs in determining whether the remedy
remains protective of human health and the environment.
Comment: General goals and policies contained in statutes and regulations
are not ARARs.
Response: General goals which are not specific and directive in intent, such
as general legislative findings or statements of public policy, are not
ARARS. However, if general goals are specific and directive in intent, and
otherwise meet the ARAR criteria, they can be ARARs. The preamble to
the final NCP specifically list State anti-degradation laws as goals which can
be ARARs. H goals are further refined by specific regulations, those
regulations will determine compliance with the goal. This concept is further
addressed in 55 FR 8746 - 8747.
Comment: Only environmental laws and regulations may be ARARS.
Response: Federal environmental promulgated standards, and State
environmental and siting standards are eligible for ARAR identification.
Worker safety and public health and safety laws and regulations fit into this
-------
12.
description, and are contained in the final ARARs'list for this ROD.
ARAR guidance specifically lists such laws and regulations.
11.
Comment: It was improper to compare existing water quality standards
within the Warm Springs Ponds with water quality standards.
Response: Although ,the Ponds themselves are not classified as waters
subject to the State's water classification scheme under the Montana Water
Quality Act and accompanying regulations, comparison of the existing water,
quality within the Ponds to water quality standards provides a useful way to
evaluate the health of the Ponds system. The Ponds are not only industrial
treatment ponds, but are the home of various fish and wildlife. Superfund's
mandate is to protect the environment, including fish and wildlife, regardless
of the location. The ARARs standards based on water quality standards are
identified for water discharged from the Ponds and for surface water outside
of the Ponds only.
Comment: Public water supply standards for ground and surface waters are
not appropriate for the ground and surface water at the Ponds. .
Response: As more fully explained in the response to Appendix B
comments, the ground water within the operable unit is potentially a
drinking water source, pursuant to Montana's ground water classification
scheme and EP A's guidance on ground water remediation. Therefore,
MCLs, which protect public health, are the appropriate standards for ground
water, and should be met at the waste unit boundary. Establishing these
standards at the waste unit boundary will also protect the Qark Fork River
from contamination from the ground water plume. Using public health
standards for surface water is a moot point at this site, given the existence of
-------
14.
applicable water quality standards for the surface water compliance points.
This is further explained in the appendix B response to comments.
13.
. Comment: The point of compliance for surface water ARARs should be at
the confluence of the Mill-Willow Bypass and Warm Springs Creek.
Response: As explained further in the response to comments on Appendix
B, there are two points of compliance for surface water standards. One will
be at the point of discharge ~rom Pond 2, for the .point source discharge. .
This is consist~nt with the Clean Water Act and the State's Water Quality
Act. In addition, compliance for ambient standards must be obtained just
above the confluence of Mill-Willow Bypass and Warm Springs Creek,the
second compliance point. This will enable the agencies to judge the
adequacy of the remedy in meeting these standards without interference
from contamination from Warm Springs Creek. Warm Springs Creek will be
cleaned up through other activities of the Clark Fork Superfund project.
Eventually, all surface water, including the Mill-Willow Bypass, mill and
Willow Creeks, the Warm Springs Creek, and the Clark Fork River will be
required to meet appropriate ARARs. .
Comment: An MPDES water permit should not be required for the Pond 2
discharge.
Response: EPA agrees that the discharge is an "on-site" regulated activity,
and is not subject to administrative permit requirements. Although not
required by CERCLA, EPA notes that the activity was a pre-existing
permitted activity, and continuing to renew and comply with the permi t
would ensure effective post cleanup monitoring, and maintain consistency
within the State's MPDES program. Therefore, EPA encourages the
continued application for and issuance of the permit by ARCa.
-------
EP A notes that the permit requirements will be identiCal to the ARAR
requirements identified by this ROD, and that obtaining. the permit renewal will not
substantilly increase ARCO's costs at the site.
15.
Comment: RCRA Subtitle C standards should never be relevant and
appropriate standards for mining waste which is excluded from. RCRA
subtitle C regulation through the Bevill amendment, and should not be
identified for this particular site.
. .
Response: As explained further in the response to comments on Appendix
B, EP A agrees that RCRA requirements are not applicable to the cleanup
activities at the site, but does find that certain selected RCRA requirements
are relevant and appropriate to certain activities at the site. Because it has
been demonstrated that the waste at the site has and continues to cause
problems to surface water, ground water, and human health at the site, and
because the waste has been or will be gathered in discrete units, certain of
the RCRA subtitle C standards are relevant and appropriate to the site.
This selective use of RCRA subtitle C standards for mining waste is
permitted and described in EPA guidance and the preamble to the new
NCP. CERCLA Compliance with Other Laws Manual: Part TI, pp. 6-2 - 6-4;
55 FR 8763 - 8764.
16.
Comment: The State's Strip and Underground Mine Reclamation Act and
Surface Mining Control and Reclamation Act are not relevant and
appropriate to this action.
Response: EP A agrees with the State's identification of SUMRA and
SMCRA regulations concerning revegetation of excavated or capped areas as
relevant and appropriate .requirements for this action. These requirements
are designed to ensure a stable, long lasting, and permanent revegetated
-------
17.
18.
19.
cover over mining wastes, and that is a goal of the Superfund program. A
more detailed analysis of this issue is presented in the response to comments
on Appendix B.
Comment: Mixing zones should be established for the ground water and
surface water point source discharge ARARs.
Response: Mixing zones are given under the State's regulatory statutes in
the discretion of the State. The State has advised EP A that such zones are
inappropriate for this action and should not be given. In addition, mixing
zones for ground water discharge are not in accordance with NCP
regulations and EP A groundwater remediation guidance. Setting compliance
at the waste unit boundary for ground water and at the point of discharge
for the Pond 2 discharge will ensure adequate protection of human health
and the environment, by reducing potential and actual exposure to
contaminants in areas outside of the Ponds themselves.
Comment: A water hardness of 175 mg/l is appropriate for setting hardness
based concentration limits for the WSP discharge. H use of a different
hardness value is warranted, the surface water ARARs will be adjusted
accor~gly.
Response: The current hardness based ARARs are set at 100 mg/l. EPA
and the State will continue to evaluate information concerning the Ponds to
determine if this is appropriate under the State Water Quality Act and
implementing regulations.
Comment: Federal water quality criteria for arsenic are not based on
current scientific data, and should not be used here.
-------
20.
Response: The arsenic number identified for surface water is based upon
the State's adoption of water quality standards (WQS). WQS are applicable
to the water, and EPA has no discretion to ignore the number. It is only
when FWQC are used as relevant and appropriate ARARs that EPA has
discretion to accept or reject numeric standards, based on designated use or
invalid scientific basis.
In any case, the WQS for arsenic has been waived by EP A as unachieveable,
and a replacement number based on the State's antidegradation statute and
regulations has been established.
Comment: ARCa agrees that it is appropriate to waive mercury and arsenic
water quality standards. ARCa disagrees with the replacement criteria
identified for mercury and arsenic.
Response: EP A is waiving mercury and arsenic WQS for the point source
discharge and the ambient surface water based upon technical
impracticability from an engineering perspective, pursuant to 42 V.S.C. ss~
9621(d)(4)(C), and on the basis that this is an interim action, 42 U.S.c. ss~
9621(d)(4)(A). Mercury is waived because the number is below current
detection limits. Arsenic is waived because the number is not achievable
using current technology. EP A and the State identify replacement numbers
of those waived standards as follows:
Mercury 0.2 ug/l
Arsenic 0.02 mg/l
The mercury number is based upon the current detection limit for mercury,
according to current agency standards. Waiving. standards on this basis, and
using replacement standards based on the detection limit, is endorsed by
-------
21.
22.
current guidance. EP A disagrees with ARoo' that the current Pond
discharge standard of 0.001 mg/l is sufficiently protective of human health
and the environment. The 0.2 ug/l standard is nearer the water quality
standard, and is the agencies' determination of what is protective.
The replacement standard for arsenic is based on the State's anti degradation
statute and regulations. Currently, water quality in the Clark .Fork River is
at or below 0.02mg/1 for arsenic. Waiving the WQS and establishing a
number above this standard, such as ARCO's suggestion of 0.05' mg/l, .
would violate the State's antidegradation statute. To comply with the
antidegradation ARARs and to ensure protection of human health and the
environment, the 0.02 ug/l standard is the more appropriate replacement
standard.
Comment: ARCO disagrees that the surface water ARAR for iron, 0.3
mg/l, is appropriate.
Response: EP A retains the iron standard in its final list of ARARs. The
standard is a secondary Drinking Water Standard, and is also protective of
environmental and aesthetic concerns. Achievement of the standard for
,Othe.r contaminants, which are environmentally based, should result in
compliance with this standard at no extra cost to ARoo or the agency.
Comment: The pH standard for the pond discharge should be kept at the
current permit level of 9.5, to ensure that pH dependent treatment can work
with optimal results within the Ponds.
Response: EP A agrees with this comment, and will retain the current 9.5
maximum pH standard for the point source discharge standards.
-------
23.
Comment: The designation of the berms throughout the Pond system as
high hazard dams through the State's Dam Safety Act is improper. Pond 1
is not a dam, as defined in the Dam Safety Act.
Response: The State's regulations are clear in stating that JillY classification
of a dam a high hazard by the 'Corps of Engineers is grounds for automatic
qualification as high hazard pursuant to State law.. The variation is criteria
between the Corps' designation and the. State's normal, independent
classification is irrelevant to this automatic classification.
Pond 1 is a dam, as defined under the Dam Safety Act, because it does
impound and divert water. The sources of this water are various, but that
does not impact the primary purpose of wet portions of Pond 1, which is to
impound water. Even if the 50-acre feet capacity were changed, these
standards would still be relevant and appropriate for Pond 1, given the site
specific circumstances at the operable unit.
24.
Comment: The selected design floods for determining the appropriate
fraction PMF standard should be further explained.
Response: Other parts of this responsiveness summary address design
floods, and the exact nature of the PMF determination for the berms. EP A
and the State have determined that a 0.05 PMF standard is the applicable
standard for all Ponds within the operable unit.
3.2 PUBUC REALm AND ENVIRONMENTAL ASSESSMENT SUMMARY
Some of the issues addressed in this chapter that relate to human health assessments are
addressed in more detail in Appendix A
-------
ARCO raised a number of comments about the data and methodology used to calculate
baseline risk and cleanup levels at the site. Responses to those specific points are given
here. However, as explained in the ROD, the selection of final cleanup levels in soils
sediments, tailings at the site is deferred at this time. EP A will continue to evaluate risk
issues, and intends to present a final cleanup number at a later date.
3.2.1 Introduction
No comments.
3.2.2 Site Setting
1.
Comment: This comment questions why days with precipitation
greater than 0.01 inch (limited dust generation) during the summer
were not accounted for in the exposure assessment and resulting
estimation of risk through the dust inhalation pathway.
Response: Exclusion of summer precipitation days only affects the
inhalation pathway as days in which precipitation exceeds 0.01 inch
are less likely to generate dust but are assumed to have no effect on
ingestion of soil from hand-to-mouth activity. In the risk assessment
it was acknowledged that precipitation greater than 0.01 inch would
affect dust generation and, therefore, days in which dust would be
inhaled from outdoor sources, but did not adjust exposure days in an
effort to remain conservative.
However, the effects of precipitation days was evaluated
scenario considered by the agencies to be applicable.
calculations and risks were recalculated discounting the
for the
Intake
average
-------
number of days in which precipitation is greater than 0.01 inch in the
months of May, June, and July.
These months account for 46 percent of the annual precipitation.
There are an average of 101.5 days in the year when precipitation
exceeds 0.01 inches or about 47 days in the months of May ~ough
July. Since three months contain 92 total days, about half have
rainfall greater than 0.01 inch (51.1 percent).
For the- recreational scenario, it was assumed in the risk assessment
that 16 days in May, June, and July would be used for fishing at the
ponds. If precipitation greater than - 0.01 inch occurred - on
approximately 50 percent of these days, 8 days may have rain greater
than 0.01 inch reducing the potential for inhalation of dust while at
the ponds. The split fishing season from mid-August through
September yields an additional 12 days available for fishing or other
recreational activities at the ponds when dust could be generated. In
addition, there are approximately 4 days in the hunting season
without snow on the ground. A total of 24 days are available for dust
generation and recreational use of the ponds.
When this reduced exposure frequency accounting for summer
rainfall is used in conjunction with the inhalation intake (as presented
in the risk assessment), the excess lifetime cancer risk that results
from using the revised parameters is not different from that obtained
in the risk assessment, 1 x 10=1.
For the occupational scenario, 120 days of dust exposure are possible:
if you accOunt for precipitation in the months of May, June, and July
(60 work days, half of which could experience rainfall greater than
-------
0.01 inch), as compared with the 150 days of exposure used in the risk
assessment. If this new exposure duration is used in conjunction with
maXimum and average inhalation rates and the estimated annual dust
concentration (from dust modeling done for the risk assessment), the
resulting excess lifetime cancer risk through occupational exposure at
the ponds is 9 x 10:1 using average inhalation rates and 2 x 1~ using
maximum inhalation rates. . The average risk changed slightly, from 1
x 1~ to 9 x 10:1, while the maximum risk did not change from that
in the risk assessment, 2 x 1~.
2.
Comment: This comment states that population stability of the area
has been omitted from discussion.
Response: An extensive discussion on the population and
demographics of the area is presented in Chapter 2 of the risk
assessment. ARCa acknowledged, in the executive summary of the
review comments, the discussion of decreasing population presented
in the FS.
3.2.3 Nature and Extent of Contamination
1. Comment: This comment identified a statement made in the risk
assessment as misleading to the reader, and stated that the comment
indicated concentrations of contaminants at the site posed a threat to
human health and the environment. The comment also states that
concentrations of many constituents were below detection limits.
Response: The statement identified is: "Site investigations to date
indicate inorganic constituents in groundwater (particularly shallow
groundwater immediately downgradient of Pond 1), surface watt=r
-------
(although contaminant concentrations decrease 'in pond outflow as
compared to inflow from Silver Bow Creek), sediments and exposed
tailings, and biological tissue (liver only) at concentrations above
background level."
This statement was meant as a simple summary statement. The
statement does not mention human health or the environment, nor
does it attempt to draw inferences to .health implications of
concentrations greater than background.
Detailed discussions of constituent concentrations are found in the
full risk assessment. Chapter 3 of the risk assessment presents. the
frequency of detection for each constituent analyzed in each
medium. Tables in Chapter 3 indicate the majority of the
constituents analyzed were detected at concentrations greater than
the detection limit.
2.
Comment: This comment states that the term "biological tissue" is
vague.
Response: The agencies agree. The use of the terms fish tissue,
edible fish tissue, waterfowl tissue, waterfowl breast tissue, or edible
waterfowl tissue would have been more descriptive than biological
tissue.
3.2.4 pxposure Assessment
1.
Comment: This comment requests more information on contaminant
migration mechanisms. The comment states that wind direction and
-------
magnitude and drinking water well locations are not included in the
referenced figure.
Response: Contaminant migration mechanisms are discussed in more
detail in Chapter 4 of the full risk assessment. All details could not
be included in the summary of the risk assessment as presented in the
FS.
Both wind direction an~ magnitude and the location of drinking water
wells in the vicinitY of the ponds are shown on Figure 3-1 of the. }
-------
3.
Comment: This comment questions the additivity of exposure
pathways.
Response: EP A guidance stresses the importance of analyzing
multiple exposure pathways (EP A, 1989b). It is conceivable that an
individual could be exposed to each pathway within a developed
scenario, although unlikely that the individual would experience'
exposure to the maximum concentration of contaminants in each
pathway. The risk assessment,provided a range' of possible risk; the
maximum possible risk is used to provide an upper limit to' the
possible risk. The risk assessment is not concerned with risk
obtained from any media other than that which could or is affeCted
by contaminants found at the site.
4.
Comment: This comment questions the use of the potential future
residential scenario in the risk assessment.
Response: EP A policy requires that risk to public health be
estimated using conservative assumptions. Residential. land use is
associated with the greatest exposures and is therefore, the most
conservative choice to make for estimating future land use of a site.
EP A believes that such a scenario is within the reasonable maximum
exposure possible at the site, and therefore is appropriately included
in the baseline risk assessment. The residential scenario provided an
upper bound estimate of the potential risk that could be incurred.
In the ROD, MDHES and EPA have made risk management
decisions based on occupational use of the site which does occur ano
will continue into the future. Institutional controls .to prevent futur~
residential development are necessary. Cleanup of contaminant'
-------
As discussed during the meetings, there are two very important items. during
the PMP. One of these is the timing of the maximum I-hour peak and the
other is the amount of rainfall during that time. ESA used a peak I-hour
rainfall amount of 1.07 inches. This value is less than the calculated 100-
year, I-hour rainfall. Also values of 1.1 and 1.2 inches in less than 1 hour
have been recorded at the Butte Airport climatological recording s~ation.
CHZM HILL used a value of 1.67 inches for the peak I-hour rainfall during
the PMP. This value was .obtained using Hydrometeorological Report.
. .
No. 43 (HMR 43) and procedures suggested by the National Oceanic and
Atmospheric Administration (NOAA) Precipitation Frequency Atlas of the
Western United States. In comparing these values with procedures
suggested in HMR 55A, the 1.67 inches for the peak hour rainfall would be
low. The value of 1.07 inches used by ESA is not correct for use as the
peak I-hour value for the PMP.
ARCO states in the comment that "CHZM HILL's PMP was calculated
through the summation of a convergence component and an. orographic
(evaluation dependent) component. However, in the FS, an orographic
adjustment factor was applied to both components, yielding an erroneously
high PMP in some parts of the basin and that this single error increased the
PMF estimate in the FS to 146,500 cfs, instead of 123,000." This is not
correct in that ARCO used 1.07 inches for the peak I-hour precipitation
instead of 1.67 as used by CH2M HILL, and the reduction in the peak 1-
hour value is alone responsible for the 23,500 cfs reduction in the PMF
calculated by ARCO.
Concerning the orographic factor being applied to both the convergence and
. orographic component of the PMP, this factor yielded results that, even
though they were lower, were consistent with HMR 55. Even though
-------
treated and the design parameters for the Pond 3 berms. The difference
between volumes predicted by the two ~odels is minimal; however, ARCa
agreed to use the storm hydrograph and resultant volume of 13,000 acre-feet
predicted by'the CH2M lllLL model in order to be conservative. It was
also agreed to use ARCa's peak flow of 3,300 cfs as the design peak flow
for the inlet structure to Pond 3 to reduce scour and resuspension concerns
within the pond system.
The increase in the design ~olume of the 100-year event of 500 acre-feet,
(from 12,500 to 13,000 acre-feet) means that the total outlet design 'flow
from Pond 3 will have to be increased from the 700 cfs indicated in ARCa's
plan 3A to approximately 750 cfs. This is required to avoid exceeding the
total allowable storage capacity of Pond 3 of 12,500 acre-feet.
4.
Comment: The comment notes that the PMF was included in the flood
modeling study because it is the standard design flood that is considered in
dam safety rules. Although the approach of using the HEC-1, flood
simulation model is appropriate for calculating the PMF, some assumptions
made by CH2M HILL are not appropriate. Critical errors were made in the
development of the maximum precipitation (PMP) and in the amount of
snow on the ground at the time of the PMF. The comment concludes that,
by modifying the input parameters to what would be ARCO's best estimate,
the PMF would be 80,000 cfs rather than the 146,500 cfs estimated by
CH2M HilL
Response: The items contained in this comment have been discussed in
meetings with ARCa, ESA (ARCO's technical consultant), CH2M HILL,
MDHES, EPA, and others. The following are responses that were given in
detail to ARCa as a result of various meetings concerning the flood
modeling study and particularly the calculation of the PMF.
-------
3.
Comment: The co_mment agrees that the use of the 100-year flood in the FS
as the primary flood for design is acceptable. However, the comment states
that the method used for estimating flood discharge by' CH2M HILL is not
acceptable. The comment further ,states that a regional analysis method
relating flood frequency to basin characteristics (as used by ARCa) is a
more appropriate method than the simulation model used by CH2M HILL.
Response: There are several items contained in this comment that have
been discussed in meetings and correspondence with' ARCa throughout the
. .
hydrologic analyses of the Silver Bow Creek studies. ARCa has developed
its oWn procedures and analyses on the hydrology of the Silver Bow Creek
Site. Some differences still exist between the studies conducted by ARCa
and by CH2M HILL.
Because of' the uniqueness of a hydrologic analysis, several methods exist
that wili provide quantitative results on flooding values for any basin.
Because different approaches are possible, there is potential for different
parties to use different assumptions to derive parameters that are needed in
a flood modelling study. Because of such differences, ARCa's values,
methods of analyses, and procedures were somewhat different than the
procedures used by CH2M HILL. The result is that ARCa's study predicts
a 100-year flood with a peak flow of 3,300 cfs and a total 5-day volume of
12,500 acre-feet, while CH2M HILL's study predicts a peak flow of 4,000 cfs
and a total 5-day volume of 13,000 acre-feet.
A meeting was held on August 2 and 3, 1990, between ESA Consultants
(ARCa's technical advisors) and CH2M Hll.L to discuss design issues,
including the hydrology of the 100-year event. At this meeting, it was agreed
that the most important parameter of the 100-year flood is the design
volume of runoff, because this governs the amount of flow that will he
-------
CHAPTER 4.0
RESPONSES TO ARCO COMMENTS, CHAPTER 4.0
PROBLEM DEFINITION.
4.1 ENVIRONMENTAL AND HUMAN HEAL1H PROBLEMS
4.1.1 MEDIA 1- POND BOTTOM SEDIMENTS
1.
Comment: The dams should ~ot have been listed as high hazard. Inflow.
design floods have not been adequately documented.
Response: Montana law states that dams which are ranked as high hazard
dams by the U.S. Corps of Engineers automatically are treated as high
hazard dams under Montana dam safety law and regulation. That ranking
was done for the Warm Springs Ponds dams.
Additionally, dams which contain the volume and hazardous substance
content of the Warm Springs Ponds dams should be treated as extremely
hazardous, for protection of human health and the environment.
2.
Comment: The comment indicates that two reports (ESA, 1987, and CH2M
H~ 1988) are described in the FS as having reached similar conclusions,
without stating clearly what the similar conclusions were. It further indicates
that the only point of clear concurrence is that some degree of upgrading of
the flood routing capacity of the pondsjbypass is necessary.
Response: As indicated in the comment, the point of concurrence between
the two reports is that some degree of upgrading the berms is required.
-------
through active measures is and will be based' on. risks other than
future residential risks.
5.
Comment: This comment states that the only realistic future use of
the site is continued use of the area for wildlife management.
Response: The agencies agree that continued use of the. area for.
wildlife management is a realistic use of the area. However, it is not
the only possible use.
6.
Comment: This comment states that land use restrictions should
have been considered in the assessment as a means of restricting
human exposure to contaminated media.
Response: The purpose of the baseline risk assessment is to
tjetermine the potential risk to human health and the environment
given that no action is taken at the site to remediate the
contamination. Land use restrictions, deed restrictions, and zoning
ordinance are all considered as remedial actions take.n to prevent
contact with contaminants at the site.
7.
Comment: This comment states that the risk assessment used poor
judgement in the selection of exposure assumptions for the
recreational scenario.
Response: Assumptions used in the assessment were based on
discussions with Montana State Fish and Parks personnel, Montana
residents, and best professional judgement because site-specific
surveys were not available. The 41 days per year (164 hours per
year) spent at the ponds for recreational activities does not seem
-------
umeasonable. It represents approximately 4 percent of the. "free"
time available each year (considering 8 hours per day are spent
sleeping and 8 hours per week day are spent working).
The Exposure Factors Handbook cited by ARCa in its comment was
not available when the assessment was being prepared. It has been
subsequently reviewed. Results of a recreational fishing frequency .
survey included in the Handbook indicated more than 50 percent of
the fishermen intervie~ed fish weekly, while 13 percent fish daily.
This survey was conducted in a relatively urban area. It is not unrea-
sonable to assume that fishermen at this particular operable unit, may
fish the ponds twice a week during the limited fishing season
available.
In addition, activity patterns presented for males 18 to 24 years; 25 to
44 years; 45 to 64 years; and older than 65 years showed that active
leisure accounted for 9 hours/week (standard deviation (sd) of 10.7);
5 hours/week (sd 5.66); 6 hours/week (sd 7.8); and 7 hours/week (sd
11.3), respectively, for the listed age groups (EP A, 1989a). In a rural
setting, fishing or other recreational use of the ponds could account
for a large portion of time spent at active leisure activities. Activity
patterns will also vary with the season, with potentially greater levels
of activity during the nonsnow months.
8.
Comment: This comment states that intake values are not consistent
with the values and methodology of the EPA 1989 Exposure Factors
Handbook (EFH).
Response: The EP A 1989 EFH was not available when the
assessment was conducted. Ingestion rates in the EFH represent
-------
averages; 100 mg/day as a reasonable maximum. may be appropriate
(Porter, 1989). See the response to Section 3.2.4, Comment 9. The
assessment followed the guidance. of the Superfund Public Health
Evaluation Manual (EPA, 1986a), which was the existing guidance at
the time the assessment was prepared. EP A will continue to evaluate
risk based on currently available guidance and literature, as it
prepares to decide a final cleanup action level for the site. .
9.
Comment: This conunent questions the selection of soil ingestion
values and lack of activity pattern data in the risk assessment. .
Response: A range of soil ingestion .values were used in. the
assessment as EP A standardized soil ingestion rates (Porter, 1989)
were not available when this assessment was initially prepared. Best
professional judgement was used in the selection of soil ingestion
rates from data available in the literature. Data for adults are
extremely limited. Professional judgement, site-specific data, and
conversations with persons living near and working on the site were
used for selecting other assumptions necessary to derive exposure
values. Prior to the completion of the assessment, a memorandum
from EPA Assistant Administrator J. Winston Porter, was sent to
EPA Regional offices early in 1989, which recommended a
. .
standardized soil ingestion rate (Porter 1989). A daily ingestion rate
of 200 mg of soil for the ages 0 to 6 years and 100 mg of soil for ages
6 years and older was given in this memorandum. These ingestion
rates were not used in the assessment as it was nearing completion at
the time the rates were provided. If these EP A standardized rates
were used along with current EP A suggested exposure parameters
(EP A 1990) selected to mimic human activity patterns, the following
would result:
-------
For the Recreational Scenario, if the assurnp"tion is made that
recreationists spend 4 hours per day (out of 16 hours available in the
day assuming 8 hours of sleep) at the ponds, the contribution of soil
at the ponds (from a contaminated source) is 100 mg/day x 4/16 or
25 mg/day. It is then assumed that an individual would obtain 25
mg/ day of soil from the ponds each of the 41 days visited, ex~ept for
the 9 days during the winter when snow could be on the" ground.
(However, it is possible to ingest soil, even if it is raining.) These
altered parameters are . used in the revised intake equations along
with an altered life "expectancy of 75 years (EP A 1990) instead of the
70 years used in the PHEA. If this intake is used in conjunction with
. the maximum and average exposure point concentrations of the risk
assessment, the resulting risk is 4 x 1~ using average exposure point
concentrations and 1 x 1~ using maximum exposure point
concentrations.
In the Occupational Scenario, workers spend 8 hours per day at the
ponds, 5 days per week, over a 40 year work life. A 40 year duration
of employment is used in these revised calculations instead of the 30
years used in the PHEA as EP A (1990) recommended 40 years as the
upper bound exposure duration. As adults, using the soil intake
factors of the Porter memo and additional factors of the intake
equation from the new EPA guidance (1989), the total daily soil
intake from the ponds would be 50 mg/day (100 mg/day . 8 exposure
hours/16 hours out of a total of 24 hours available for soils ingestion
each day). Workers are assumed to spend 250 days per year on the
job. Five months of the year the ground is covered with snow,
making ingestion of soil more difficult. During those 5 months,
approximately 100 days are work days. Subtracting the 100 snow
cover days from the 250 work days results in 150 days of potential
-------
exposure to soil (as was previously stated, soil can be ingested, even
during rainy days). If these altered parameters are used in the
revised intake equations along with an altered ,life expectancy of 75
years (EP A 1990) instead of the 70 years used in the PHEA The
slope factor for arsenic is currently reported as 1.65 (mg/kg/dayr1 (as
opposed to the value of 1.5 (mg/kg/dayr1 which was used in the
PHEA). Using the maximum concentration of each constituent
detected in soils to which workers could be exposed, the resulting
excess lifetime cancer, risk due to the presence of carcinogenic
compounds in soils is 2 x 1~. using average exposure point
concentrations the resulting excess lifetime cancer risk due to
the presence of carcinogenic compounds in soils is 9 x 1~. EPA is
continuing to evaluate these assumptions and will explain its choice
when it proposes and selects a final action level.
EPA (1989) currently recommends using the 95th upper confidence
interval on the arithmetic mean for exposure point concentrations.
Because of the limited data available for statistical use and the
potential for the upper 95th confidence interval on that data to be
greater than the maximum value detected, the maximum
concentration is appropriate to use as the exposure point
concentration.
The excess lifetime cancer risk is driven by arsenic. Although
beryllium is now included in the calculation of risk (an oral slope
factor is now available that was not available when the PHEA was
compiled), it does not influence the total excess lifetime cancer risk.
Lead is not included in the estimate of potential excess lifetime
cancer risk or the analysis of potential noncarcinogenic health
-------
impacts~ This is because the EP A has not established a slope' factor
or reference dose with which to evaluate the potential effects of lead
in a manner similar to other compounds. Recent data indicate lead
is a potential carcinogen. . It has been rated as a B2 carcinogen
through ingestion and inhalation by EP A, but a slope factor has not
been assigned. In addition, lead exhibits detrimental health effects
other than cancer. As such, a reference dose for lead would be
expected. However, reference doses are based on the principle of a
threshold effect, which means there is a level of exposure which will
not illicit a detrimental effect. EP A believes there is no threshold for
lead exposure; that all exposures to lead will result in some
detrimental health effect.
As has been demonstrated by the above calculations, altering the
exposure parameters to match activities patterns and adjusting the
intake parameter to meet EP A standardized values does not
significantly influence the resulting risk range.
10.
Comment: This comment states the value used for incidental
ingestion of surface water is arbitrary.
Response: As was previously stated, when standardized intake values
were not available for a potential pathway of exposure, best
professional judgement was used in conjunction with any available
literature that dealt with the medium of exposure to select a
reasonable intake and an upper bound intake.
11.
Comment: This comment questions the use of the Industrial Source
. Complex (ISC) model for determining dust concentrations at receptor
locations.
-------
Response: The Industrial Source Complex model was recommended
by. EP A as the appropriate model to use for estimating dust
concentrations at the Warm Springs Ponds. The methodology used in
the anaIysis is a standard .approach.
3.2.5 Risk Assessment
1.
Comment: This comment questions the representativeness. of risks
presented in the assessment to the actual risks of the site.
Response: Risk are presented in the FS as a range from most probahle to
maximum plausible. The upper bound estimates likely overestimate actual
risk. It is anticipated that the actual risk from the site would be represented
within the range of risk presented in the assessment.
2.
Comment: This comment questions the use of aquatic water quality criteria
in determining potential impacts to the aquatic ecosystem at the site. The
comment states that remedial actions will improve conditions at the site arid
potential impacts will not occur.
Response: The baseline risk assessment evaluates the potential impacts, if
no actions are taken at the site. It is, therefore, inappropriate to evaluate
the potential for impacts, if remedial actions are taken at the site to improve
current conditions.
Chronic effects most often are manifested at the species population level
rather than the individual level. Chronic effects can include decreased
survival, because of physiological stress that makes a species less
competitive; restricted development of eggs or unviable sperm; reduction in
a food source for a species and gradual population decline. All of these can
affect the growth, reproduction and mortality of a species. The risk
-------
assessment did not unequivocally state that these impacts will occur, but that
chronic stress can and likely would manifest itself in th~se forms.
3.2.6 Uncertainities and Limitations
Comment: This comment states that specific issues that. add uncertainty to the
assessment were not evaluated. Further, it questions whether uncertainties and
, limitations were accounted for during the interpretation of the risk assessment.
Response: Specific. sourCeS of 'uncertainty and their qualitative magnitude were
discussed in the full risk assessment presented in Appendix A of the FS. Limita-
tions and uncertainties were considered by the risk managers during the decision
making process in selecting the proposed cleanup levels used in the FS.
3.3 RISK-BASED CLEANUP GOAlS DEVELOPMENT
3.3.1 Cleanup Goal Calculation ~
1.
Comment: This comment questions the proposed soil cleanup concentration
for arsenic and proposes an alternative methodology with a resulting cleanup
goal of 10,000 ppm arsenic in soils.
Response: The cleanup goal of 10,000 ppm calculated by ARCO is
inappropriate for a number of reasons. For purposes of discussion, the
ARCO calculation is compared to the current calculation, using current
guidance, shown below.
Basic Equation:
R = C x HIP x SF
C = Rj(HIF x SF)
where:
-------
R = Risk
C = Chemical concentration (mg/kg)
HIP = Human intake factor (kg/kg/d~y)
SF = Slope factor (mg/kg/day)=!
Current MDHES Calculation (Occupational Scenario Intakes):
R = 1.0E-4
HIF = 50 mg/day x 150/365 days/yr x 40/75 yr x ~~ kg/mg x
= 1.6E-7 kg/kg/day
SF = 1.65 (mg/kg/day)=!"
C~ = 380 mg/kg
ARCO Calculation
R = 1.0E-4
IDF = 5 mg/day x 150/365 days/yr x 30/75 yr x 0.5 x 1~ kg/mg x 1 /70
kg
1/70kg
= 6.3E-9 kg/kg/day
SF = 1.75 (mg/kg/day):!
C~ = 9,100 mg/kg
Comparison of these calculations reveals several areas of difference. First,
ARCO assumes that workers ingest only 5 mg/ day of sediment. This is in
conflict with recent EP A guidance on soil ingestion (Porter 1989). Second,
ARCOs inclusion of a factor of 0.5 for "absorption of arsenic from soil" is
not appropriate, since there are no reliable data to show that in the
absorption of arsenic in pond sediments is less than" that of arsenic not
mixed in soil (see response to Comment #67 for a further discussion of
absorption of arsenic). Third, the SF of 1.75 used by ARCa is based on the
same data and calculations as the value of 1.65 used by the agencies, with
the difference being the result of differences in rounding. If the value of
1.75 was used, the calculated clean-up goal would be lower, not higher.
-------
With respect to the assertions that any cleanup level calCulated as above will
be over protective because (a) arsenic-induced skin cancers are typically
nonlethal, and (b) the dose response curve is nonlinear, these issues are
addressed in subsequent responses. ,
2.
Comment: This comment questions the toxicity information used for arsenic
in that it does not account for the nonlethality of the type of cancer
associated with arsenic. It also questions that the risk presented in the
assessment due to arsenic was not adjusted downward' as suggested by the,
EPA administrator. (Thomas, 1988).
Response: The assessment did adjust the cancer toxicity value available' for
arsenic at the time of the assessment by an order of magnitude, as suggested
by Thomas (1988), prior to use in the assessment. The Thomas (1988)
memo states that it is the toxicity values that should be adjusted, not the
resulting risk as suggested by this comment. At the time of the assessment,
the published cancer potency factor for arsenic was 15 mg/kg-dar. This
cancer potency factor was adjusted to 1.5 mg/kg-dar for use in the risk
assessment and has since been adjusted to 1.65(mg/kg-day)=l.
It is agreed that arsenic-induced skin cancers have a low mortality rate when
properly treated, and this information is indicated in the toxicological
assessment for arsenic. However, just because skin cancer is rarely lethal
does not make it acceptable. Indeed, modem treatment technologies have
reduced the death rates of a number of cancers (Hodgkin's, thyroid, breast,
uterine, testicular), but this is not cited as a reason for accepting an increase
in the incidence of these cancers. Moreover, there is a growing body of
evidence that ingestion of arsenic increases the risk of more deadly internal
cancers, as well as skin cancer (Chen et al 1985; Chen et ale 1986; Chen et
ale 1988; EP A, 1988).
-------
In any event, treatment of arsenic-induced skin cancer may in some cases be
uncomfortable or painful, especially if treatment is not timely, and costs both
time and money. EP A does not consider that the" nonlethal nature of
arsenic induced skin cancers justifies acceptance of a higher than usual risk
from this contaminant.
It is also agreed that "the low mortality of arsenic-induced cancers can be
weighed in the risk interpretation and risk management process. However,
the risk assessment can. only present information to be used in risk
management decisions.
Furthermore, risk assessment, in general, does not consider lethality in
developing potency factors and, thus, does not estimate mortality associated
with cancer. Rather, it is an estimate of the potential excess lifetime cancer
incidence associated with exposure to carcinogenic agents.
3.
Comment: This comment questions the cleanup concentration selected for
lead. ARCa suggests the methodology developed by the Society for
Environmental Geochemistry and Health (SEGH) is best. for deriving
cleanup levels. Using this methodology, ARCa has proposed a cleanup
level of 6,000 ppm lead.
Response: EP A believes that the SEGH method has merit, but that any
calculated soil cleanup level is very dependent upon the input parameters
used. The input valu~s selected by ARca are all highly debatable, and
certainly cannot be characterized as "conservative" (p. 3-5 of ARCO's
comments). A much different outcome results when more appropriately
conservative values are used:
-------
3.
4.
1.
Target blood lead level. ARCa selects 25 p. g/ dlbased on the Center
for. Disease Control 1985 level recommended for medical
intervention. As discussed in subsequent responses (Appendix A,
Section 6.2.1, Comment 1», it is incompatible with current medical
thinking to characterize 25 p. g/ dl as a "health protective blood level"
for children. Although there is considerable debate concerning what
value (if any) is. safe, for illustrative purposes a tatget of 10 p.g/dl will
be used. Such a value (or even lower) is suggested by a number of
recent studies (Davis ~d Svendsgaard 1987; Bellinger et al. 1989; .
Chaney et al. 1989). In addition, the assumption inherent in the
SEGH method that soil may contribute the entire difference between
background and maximum allowable blood lead is questionable. .
2.
Baseline blood lead level. ARCa's calculated number of 2.24 p. g/ dl
could not be replicated based on the description of its derivation
provided by ARCa. An attempt to use the method suggested by
ARCa resulted in a geometric mean baseline of 3.42 p.g/dl with an
upper 95th percentile value of 6.1 p.g/dl.
Blood lead:soillead slope. ARCa used a value of 2.0 (}lg/dl)/(1,000
ppm) based on averaging an undocumented value of 1.8 with a value
of 2.2 derived from a mining study (see response to Appendix A,
Section 6.1.2 comment). Using the mining study value of 2.2, which is
the default value for the disaggregate model, is more appropriate.
Geometric standard deviation. ARCO uses 1.42 for the geometric
standard deviation, which is the value identified by EPA (1989c, p.
ID-6) as the midpoint of the range for children living near a point
source of lead. Since exposure to mining wastes is not a point source
-------
of lead, it is more appropriate to use the high -end of the range, a
vahle of 1.53, as identified by EPA (1989c).
5.
Number of standard deviations. ARCO correctly uses N = 1.64 to
calculate the upper 95th percentile of blood lead levels. It should be
noted, however, that this procedure accounts for human variability in
the blood lead level caused by a uniform source of lead, but does not
provide any margin of error for human variability in the sensitivity to
the toxic effects of lead.
Using the above reasonable values yields a soil lead cleanup level of 708
ppm, rather than 6,000 ppm. If the background blood lead level were set at
the upper 95th percentile, rather than the geometric mean (6.1 instead of
3.42), the calculated soil cleanup level would be less than zero. These
calculations illustrate that the output of the SEGH model can range from
6,000 ppm to less than zero as the input parameters are adjusted from more
liberal to more health conservative. It is precisely this uncertainty and
variability that justifies using the OSWER Directive (No. 9355-02 dated
September 7, 1989) as guidance in establishing soil cleanup levels based on
site specific considerations.
The comment continues with the provision of Attachment 6, which presents
the methodologies and assumptions ARCO used to generate cleanup
concentrations.
Response: ARCO's contention that the future residential use scenario is not
appropriate is addressed in the response to Section 3.2.4, Comment 4. As
stated in response to Appendix A, Section 6.2.1, Comment 1, the agencies
agree that in general, adults would probably receive a lower dose of lead at
-------
4.
5.
a given soil level than would children. However, the quantitative statements
made by ARCa are not supported:
1.
Soil ingestion rates. Soil ingestion rates in both children and adults
depend upon many factors and a reliable quantitative evaluation of
the ratio of adult to child soil ingestion is not possible from available
data (refer to response to Appendix ~ Section 6.2.1, Comment #1).
2.
Absorption of lead. ARCa is. internally inconsistent in its citation of .
absorption (30 percent for children and 10 percent for adults, and
then a reference to 25 percent on page A-21). This inconsistency
reflects variability in the data and the influence of factors such as. age
and nutritional status. Thus, a single ratio such as 3 is not supported.
by the data.
3.
Lead retention. ARCa did not attempt to quantify this difference.
Slope value. ARca's quantification of this difference as a factor of 3
to 6 is not referenced here, but on p. A-23 is attributed to a personal
communication from Bornschein or coworkers. Such preliminary and
undocumented statements do not provide reliable estimates.
Target blood lead level. ARCO's use of OSHA values of 40 or 30
",g/dl are inappropriate and internally inconsistent with the
acknowledged need to protect pregnant women (p A-22). In addition,
EPA (1989c) identified middle-aged men as a sensitive subpopulation,
based on studies. showing a correlation between blood lead and
elevated blood pressure with no apparent threshold down to 7 ",g/dl
(Pirkle et ale 1985).
-------
In summary, present data do not provide a reliable quantitative method to
convert cleanup goals calculated for children using the SEGH method to
cleanup levels protective for adults. In addition, it should be noted that
ignoring exposure to soil lead through inhalation. of dust by attributing it to
"background" is not appropriate. All lead exposures due to site
contamination must be evaluated.
> 3.3.2 Potential Cleanup Goa~he Opera~
1.
Comment: This comment questions the placement of the seCtion
"Risk-Based Cleanup Goals Development" in the FS.
Response: ~is section provides background on the development of
the risk based cleanup goals that were presented in the ARARs
section. The agencies believe this is a necessary and important
discussion.
2.
Comment: This comment states that the risk estimates, as altered by
ARCO, did not provide an adequate basis for requiring or
determining cleanup goals for the site.
Response: The requirements for and determination of cleanup goals
at the site is a decision made by the MDHES and EP A and will be
made according to the NCP, EP A guidance, and sound scientific
judgment.
-------
HMR 43, applies directly to the Silver Bow Creek drainage, HMR 55 should
not be overlooked. Since the Silver Bow Creek drainage is bordered on
three sides by the Continental Divide, the pro.cedures 'and values contained
both in HMR 43 and HMR 55 were reviewed. HMR 55 has now been
revised and republished as HMR 55A During the work on revising
HMR 55A, there were concerns that developed with HMR 43. Th~ current
status is that HMR 55A has been republished and accepted by all' agencies,
and HMR 43 is being revised. The values that are now contained in
HMR 43 will be increased for. the Silver Bow Creek drainage and will more.
closely agree with the values contained in the new HMR 55A
Since the two reports were in discussion during the flood modeling study,
EPA and MDHES requested in September 1989 (after the draft version of
the flood modeling report was published) that values for the PMF be
calculated using both HMR 43 and HMR 55A reports as they now exist.
This is the reason for the "range" in PMF flows, as shown in the final flood
modeling report published in November 1989. After the review and
comments by ARCa and others, EP A and MDHES believe that the value to
be used for the PMF should be 146,500 cfs.
Another concern raised by ARca is the assumption of a snow water content
of 4 inches for the elevation zone between 5,000 and 6,000 feet. As
previously addressed in a response letter to ARCa, along with the timing
and amount of the peak hourly rainfall value, this is one of the most
significant items in determining the PMF. An important point concerning
this value is that during the June 1908 flood event, newspaper records give
account of all power lines being down in Butte due to heavy snows. Also.
the newspaper account mentioned 9 inches of snow on the ground. By
definition, a PMF is estimated utilizing "critical conditions" for hydrologic
factors within a watershed. After researching historical data, it was felt that
-------
the factors leading up to the June 1908 flood conditions were "critica1."
Based on these accounts, and additional snow records in the drainage basin,
4 inches of snow water content in the 5,000- to 6,OOO-foot elevation is a very
reasonable value.
An additional concern raised by ARca in the comment is that if the
snowfall occurs immediately prior to the PMP, so that it has not melted,
then it is probably part of the PMP storm event and should be subtracted
from that event. The Butte area experienced 19 days of precipitation prior.
to the June 6, 1908, flood. The PMP, as used for the .flood modeling study,
only lasts 72 hours. There could be a "critical" set of conditions leading up
to the PMP, as has and did occur in the 1908 storm. These conditions
leading up to the PMP should not be subtracted out of the PMP.
Comment: Page 10 Paragraph 5: Because of the lack of data, both spatial
and temporal, numerous assumptions were made about precipitation
distribution during calibration events. For example, for the winter storm
calibrations, the FMS states that "precipitation data were available for low
elevation stations, but no data existed for the upper elevation stations where
the runoff was originating."
Response: ARCa interpreted the meaning of this paragraph differently
than intended. As shown in the flood modeling report the upper elevation
stations where the runoff was originating refers to the spring storm
calibration and not the winter storm calibration. Runoff from upper
elevations would occur during spring storms whereas runoff from the lower
elevations only would occur during the winter where there is an area of
frozen ground and elevations low enough so that the precipitation would not
be "soaked in" by the existing snowpack.
-------
Comment: Page 16 Paragraph 1: The FMS provides few details of the
method followed to develop temporal distribution other than the comment
that "the distribution of the PMP was arranged in I-hour increments
following the procedure outlined in NOAA Atlas 2."
Response: The amount of the PMP was calculated in I-hour increments
following the procedures as outlined in NOAA Atlas 2. The .temporal
distribution was arranged using HMR43 and HMR55A The Figure 5 that is
in Attachment 5 shows. only CH2M HILL's hyetograph. The PMP .
hyetograph determined for this review at the bottom of the page is in error.
As previously discussed, ARCO did not use the same peak hourly value as
CH2M HILL, (1.67 inches) but only used 1.07 inches. The presentation in
Figure 5 is not correct.
Comment: Page 17 Paragraph 1, Item 4. Summary and Conclusions.
Holding all other criteria of the HEC-l simulation model the same as in the
FMS, while using the PMP illustrated in Figure 5, the peak discharge for a
PMF flood event would be approximately 120,000 cfs, rather than 145,000 cfs
which is presented in the FMS.
Response: The minimal value should be 129,000 cfs and not 120,000 cfs.
This lower value was incorrectly calculated by ARCa.
5.
Comment: The comment questions some of the assumptions used and
conclusions reached in the section that deals with the ability of the pond
berms to withstand earthquake damage. It points out several details
regarding the specific locations of the Warm Springs Ponds relative to
subdivisions of the Intermountain Seismic Belt.
-------
8.
Response: The comment does not disagree with the conclusion of the FS
that the berms are susceptible to damage. by earthquakes, or that
modifications of the berms to withstand earthquakes is required. As clearly
stated in the FS, the initial look into the earthquake potential for the area is
not definitive. A detailed earthquake study will be required as part of the
design phase. The earthquake study should take into account all data and
reports developed recently.
6.
Comment: The comment not.es that. no reference was provided for a study.
. .
by ESA for Anaconda Minerals Company.
Response: No published report was available. .The paragraph in question
described preliminary study results given to EPA and MDHES, by Roger
Hail of ESA, during a meeting with ARCa and ESA
7.
Comment: The comment notes that the studies ARCa has performed to
date do not specifically examine the earthquake stability of the berms as
they currently exist, and that the FS misrepresents ARCa's studies in this
regard.
Response: EP A and MDHES appreciate the clarification regarding the
conclusions of the studies. The understanding by EP A and MDHES of the
work done to date was based on statements made by ARCa's engineers
during a meeting held with ARCa and its consultants during the
development of the FS.
Comment: The statement that the pond berms may fail at accelerations
from 0.05 to O.07g appears unreasonable. The FS may be confusing pseudo-
static earthquake coefficients with ground accelerations.
-------
9.
10.
Response: The comment is correct. The 0.05 and 0.07 are pseudo-static
earthquake coefficients used in the stability analysis. These coefficients are
~omparable to peak ground accelerations of 0.10 to 0.14g, using 0.50 for the
ratio of peak ground surface acceleration to pseudo-static earthquake
coefficient. Ratios of 0.40 to 0.67 are generally used to reduce the peak
ground surface accelerations to pseudo-static earthquake coefficients.
The conclusion of the FS remains unchanged: the berms are susceptible to
earthquake damage and requ~re modifications to improve their earthquake
stability. This conclusion is not in contention. ARCO's own plan calls for
upgrades to the berms to improve their earthquake stability.
Comment: This comment presented further considerations that will need to
be taken into account when the detailed earthquake study is performed.
\,
RespOI.1se: EP A and MDHES, while not agreeing or disagreeing with the
presentations made in this comment, believe they should be deferred until
the detailed earthquake study called for in the FS, and apparently allowed
for in Comment 7 above, is undertaken.
Comment: The comment states that the discussions of the possible conse-
quences of a failure of the pond berms during an earthquake should be
deleted from the FS because there is insufficient data and analyses to
support such discussions.
Response: EP A and MDHES disagree. It is appropriate for the FS to
include indications of the possible consequences of a failure of the berms 50
the public can appreciate the hazard the berms pose. EPA and MDHES
also disagree with the apparent statement that the FS should have included
detailed analyses of earthquake failure scenarios. The purpose of the FS i,
-------
not to determine the exact consequences that would occur if the berms fail
during an earthquake, but only to determine whether a threat of a release
exists, and to describe the potential consequences of such a release.
A more detailed study of the likely failure scenarios would perhaps be
appropriate if the level of protection required for the berms were in
dispute. But, in fact, ARCO's own Alternative Plan 3A includes upgrading
the berms to withstand a maximum credible earthquake (once the MCE is
determined); exactly the same standard of protection included in the
. .
Proposed Plan. Because neither the need to modify the berms to improve
their earthquake stability, nor the level of earthquake they should be able to
withstand is in dispute, detailed modeling of the likely failure scenarios is
not warranted and would only delay actions to upgrade the berms.
In any event, EP A and MDHES continue to believe, based on historical
experience with other tailings ponds, that the potential exists for the tailings
to contaminate an extensive area downstream if released from the ponds by
earthquake failure.
4.1.2 Media 2 - Surface Water
1.
Comment: The comment states that the FS should include all available data
in its entirety that was used in establishing the likely causative mechanism
for the fishkills. The comment implies that some data are missing, but does
not indicate what information is missing.
Response: The argument developed by EP A and MDHES in determining
the most likely cause of the fishkills is presented in the FS in some detail.
The data relied on are either presented in the FS or in the RI reports.
-------
2.
Comment: The comment questions the method used to calculate the
average metals loads in the Mill-Willow Bypass.
Response: The comment is correct in how the loading should have been
calculated. However, EPA and MDHES believe this is now a moot point
since ARCa has proposed to, and would presumably agree to, remove the
sources of the metals "loads in Mill and Willow Creeks as an alternative to .
rc;mting their flows into the pond system for treatment. EPA and MDHES
believe this would be a more appropriate approach to the problem. The FS . .
. .
did not explore the option of source removal, because the sources lie outside
the limits of the operable unit. Data to determine the extent of the sources
and the potential costs of their removal were not available during. the
preparation of the FS. See also the response to Comment 5, Section 8.3.2.
3.
Comment: This comment questions the contribution of Mill and Willow
Creeks to the contamination leaving the operable unit and the need to divert
Mill-Willow flows into the pond system. The comment also questions the
use of total metal analyses rather than dissolved metals for calculations of
loadings.
Response: The text on page 4-29 of the FS would have been clearer if it
had stated that the Mill and Willow Creeks contribute a significant portion
of the total metals reaching the Clark Fork River from Silver Bow Creek.
The comment is correct that the examples presented are for total metal
analysis. However, the use of total metals concentrations is appropriate,
because the harm to aquatic life caused by the metals is due to both
dissolved and undissolved copper; and the standard for copper is based on
acid soluble copper, not on dissolved copper only. Also, see responses to
Section 7.5.1, Comments 3 and 4.
-------
4.1.3 ~ia 3- ThilingLDeposits and Contaminated Soils
Comment: This comment questions the methods used to calculate the volumes of'
contaminated soils in the FS.
Response: XRF and laboratory analyses of surface samples were used to define the
areal extent of tailings and contaminated soils. Hand-auger borings were used to .
. estimate associated thicknesses and, therefore, volumes of .contaminated materials.
The estimates of volumes of tailing and contaminated soils developed for the FS
. .
are suitable for the purposes oJ a FS and provide a reasonable basis on which
remedial technologies and associated costs can be evaluated. Additional data on
contaminated soils and tailings may be necessary to refine area and volume
estimates in support of remedial design, once the preferred alternative for the site
has been selected and contaminant action levels have been established in the ROD
for site soils. The field map mentioned in the response to Section 2.2.7,
Comment 2, sLows the areas of contamination at the bypass area. See the response
to Section 2.2.7, Comment 2, for a more detailed discussion of criteria used to
define contaminated soil and estimate volumes and the acreages of contaminated
soils within the operable unit.
4.1.3 Media 4 - Ground Water
Comment: The comment states that the FS is wrong in implying that the
groundwater discharge to the Clark Fork River (estimated to be approximately
1 cubic foot per second) is a problem.
Response: The comment is correct that the discharge of 1 cubic foot per second of
groundwater into the much larger flow of the Clark Fork River may not cause a
measurable impact on water quality in the river. This section of .the FS was written
only to record the findings of the RI.
-------
4.2 RELATION TO TIm ENVIRONMENTAL CONCERNS FOR TIlE WHOLE SITE
2.
4.2.1 Pond Bottom Sediment
1.
Comment: The comment states that the FS is wrong in asserting that the
problem of upgrading the pond berms to withstand floods and earthquakes is
not affected by approaches that might be taken in the future to remedy
upstream issues, and then goes on to state that this ~rror caused MDHES to
overlook a common approa~h to upgrading the" berms and addressing
upstream issues, namely ARCO's own proposal of raising the Pond 2 "and
Pond 3 berms and using these ponds to settle tailings out of flood flows.
Response: EP A and MDHES disagree with the initial statement of
comment. Upstream issues do not affect reasonable designs for upgrading
the pond berms to withstand earthquakes and floods. EPA and MDHES
agree, however, that certain elements of ARCO Alternative 3A have
sufficient merit to be incorporated into the selected alternative in the ROD.
For a detailed discussion of how the EPA and MDHES Alternative 3 and
the ARCO Alternative 3A have been combined into the ROD's selected
alternative, see the response to Section 7.5.
Comment: The comment states that the FS should either evaluate the
impacts of damage from flood or earthquakes or not assert that the failure
of the ponds during a flood or earthquake ''would have a large, potentially
catastrophic impact on the Clark Fork River, possibly as far as the Milltown
Reservoir, which is also a Superfund site,...approximately 145 river miles
from Butte, [which] already contains tailings from the Butte Area." The
comment also states that the FS should assert that the tailings have
accumulated in the Milltown Reservoir over the past century "through
normal sediment transport."
-------
Response: EP A and MDHES are concerned that failure of the berms would
have large impacts on the Clark Fork River, and that those impacts could
potentially be catastrophic to the life forms in the river and could extend
well down river, possibly as far as Milltown Reservoir. The EPA and
MDHES have considered the po~sibility of conducting an incremental risk
assessment of potential pond failure, but have concluded that such studies
would be too costly and time consuming, and would not be definitive enough
to warrant the effort. EPA and MDHES note that they are in agreement
with ARCO in the need to protect the ponds to withstand the MCE and, the'
0.5 PMF.
4.2.2 Surface Water
Comment: This comment repeats the point of other comments on earlier
sections of the FS in stating that remedial alternatives for the Warm Springs
Ponds Operable Unit must take into consideration upstream source controls,
which will be part of an integrated remedy for the Silver Bow Creek Site.
Response: It is beyond the scope of the FS for the Warm Springs Ponds
Operable Unit to explore, in detail, upstream source controls. If it had
become apparent during the FS that upstream source controls must be part
of any logical solutions to the problems at the operable unit, then it would
have been necessary to redefine the operable unit, reconsider the objectives,
and develop different alternatives.
The remedial action at the Warm Springs Ponds will be the first action along
Silver Bow Creek because of the need to provide immediate protection from
potential earthquake or flood events. The implemented action must be
,capable of providing adequate handling and treatment of Silver Bow Creek
and potentially Mill/Willow Creek sediments and waters for the near term.
-------
That action must realistically assume that upstream improvements will not
take place for some years. When remediation activities upstream are
underway or in place, then it will be appropriate to revisit, and perhaps
revise appropriately, the Warm Springs Pond treatment system.
4.2.3 Tailings Deposits and Contaminated Soils
Comment: This comment refers the reader to the comment in Section 4.2.2
regarding surface water, app~rently intending that it be applied to this
medium.
Response: See the response to Section 4.2.2, which details the EP A and
MDHES position regarding upstream source controls.
4.2.4 Groundwater
Comment: The comment states that ARCa agrees that groundwater
contamination in the operable unit does not pose a problem for areas
outside the operable unit.
Response: The section of the FS to which this comment refers is not meant
to imply that the groundwater contamination in the operable unit does not
pose a problem for areas outside of the operable unit. Rather, the point is
made that selecting and designing a remediation for the groundwater
contamination can be done independently of any problems outside of the
operable unit. This is true because the contamination is largely caused by
the contaminated materials in the operable unit, and because the means
exist to solve the problem by actions taken within the operable unit.
-------
CHAPTER 5.0
RESPONSES TO ARCO COMMENTS, CHAPTER 5.0
IDENTIFYING REMEDIAL OBJECfIVES
AND GENERAL RESPONSE ACflONS
5.1 STATUTORY AND REGULATORY REQUIREMENTS
No comments.
5.2 REMEDIAL ACTION OBJECTIVES
1.
Comment: Inflow design floods and surface water standards were not
correctly identified.
Response: These issues are addressed in the response to Appendix B or
Chapter 3 or 4 responses.
2.
Comment: The comment states that reducing the migration of upstream
tailings is not an objective for this operable unit, but that the existence of
this concern is only an acknowledgement that successful remediation of the
Warm Springs Ponds Operable Unit is dependent on implementation of
source controls upstream.
Response: As discussed in response to Comments 4.2.2 and 4.2.3, source
controls upstream are not considered appropriate remedial objectives for this
operable unit. However, the point of this objective is to reduce migration of
upper Silver Bow Creek and Mill/Willow Creek tailings to the upper Clark
Fork River and the cleaned Mill-Willow Bypass. The remedial objective is
appropriate. Note that both Alternative 3 and ARCO Alternative 3A
propose to intercept most tailing migration from Silver Bow Creek.
-------
5.2.1 Pond Bottom Sediments
Comment: ARCa states that the MDNRC dam safety rules do not address,
and do not apply to the pond bottom sediments, and notes that the FS states
that "the dam safety rules are established primarily for dams containing only
water and not contaminated sediments."
Response: EP A and MDHES disagree with the comment. The dam safety
rules are intended to address .the more limited problem of a dam breach
releasing only uncontaminated water. The FS goes on to note that because
the dam safety rules were not developed to deal with the more hazardous
situation of a dam releasing contaminated' water and contaminated
sediments, they may not be sufficiently protective in this situation, and more
stringent standards may be needed. EPA and MDHES have evaluated the
standards, and determined that a 0.5 PMF standard for all berms within the
operable unit is 'the appropriate ARAR standard, and is necessary to
adequately protect human health and the environment.
5.2.2 Surface Water
Comment: The comment states that the FS appears to imply that
compliance with the water quality standards will improve aquatic life, and
goes on to state that directly below the Warm Springs Ponds outlet, aquatic
life has acclimated to the existing water quality conditions and is thriving
and that the only problem is the periodic fishkills.
Response: EP A and MDHES disagree with the comment. EP A and
MDHES believe that exceedances of the ~,mbient water quality criteria
(A WQC) are assumed to have negative impacts on the aquatic life in the
streams, and that compliance with the water quality standards will improve
-------
aquatic life. The aquatic water quality criteria were established after careful
investigation of the levels of contaminants that various aquatic life forms can
tolerate without chronic or acute impacts.. EP A and MDHES know of no
site-specific study of the stream conditions immediately downstream of the
outlet that would indicate the populations of the species currently there exist
in the proportions and levels that they would if the water quality levels were
not exceeded. CERCLA establishes ARARs as requirements that Superfund
remediations must meet, and the water quality stand~ds are ARARs for this
operable unit.
5.2.3 Tailings Deposits and Contaminated Soils
Comment: The justification for the cleanup levels is not well documented,
and the levels do not bear a clear relationship with actual site hazards.
Responr.e: Detailed responses to ARCQ's comments regarding risk
assessment are contained in other parts of this responsiveness summary.
EPA has determined that a risk is posed at the site by the contaminants and
conditions present, and has selected an interim remedy for thi~ site. EP A
notes that a decision on final cleanup levels is deferred, pending further
evaluation by EP A of the appropriate scenario and methodology and recent
EP A guidance on risk assessments.
5.2.4Ground Water
Comment: Restoration of the ground water at the site is not appropriate.
given site conditions.
Response: EPA and MDHES classify the aquifer as a Class II aquifer.
MCL standards are relevant and appropriate standards for the site as th~
-------
ARCO comments admits, and must be met. Institutional controls are useful
to prevent ingestion or exposure until cleanup occurs, but institutional
controls are not generally considered pemianent or reliable and should not
generally be used as a replacement for active cleanup measures, are
explained in the preamble to the NCP.
5.3
GENERAL RESPONSE ACTIONS
1. Comment: The comment states that the process and criteria used in
selecting the general response actions should be given, along with 'a '
description of each general response action.
Response: The criteria and process used to select the general response
actions are those specified in the RI/FS guidance document. Descriptions of
the general response actions are given in Table C-l, in which the meaning
and scope of each general response action is revealed by the remedial
technologies and process options listed under it. The reasons for eliminating
possible general response actions are also noted in Table C-1.
2.
Comment: The ~omment points out that the FS speaks of both ''Treatment''
and "In situ Treatment," and asks the difference. The comment also requests
that access controls be included in Table 5-2.
Response: Treatment includes those process options listed for it. In situ
treatment includes many of the same process options. The difference is that
in situ treatment is applied to the contaminated medium where it exists,
witbout removing it from its current containment for treatment. The in situ
treatment of the pond bottom sludges developed for Alternative 1 in the FS
is an example of in situ treatment, since it would not require removal of the
sediments for treatment.
-------
Access controls have been considered remedial technologies under the
General Response Action category "Institutional Controls," which is included
in Table 5-2. The two access controls considered are flooding and fencing.
Please see Table C-1 for more detail. EP A and MDHES regret the
confusion caused by including access controls as a separate general response
action in Tables 6-1 and 6-2.
, 3.
Comment: The comment states that the FS should. more' clearly distinguish
between the east-west berms. and the north-south. berms surrounding the.
ponds.
Response: EP A and MDHES agree that there are some distinctions,. but
believes the FS is sufficiently clear on this point. Briefly, the most important
distinction is that the north-south berms along the bypass would have to be
raised to protect the ponds from being damaged by a flood along the Mill-
Willow Bypass, while most of the east-west berms would not have to be
raised for this purpose. Most of the other discussions of the berms,
. particularly regarding the need to upgrade the berms to withstand earth-
quakes, apply to both the north-south and the east-west berms. The raising
of the Pond 3 berms to contain and treat the lOO-year flood event also
applies to both the north-south and the east-west basins.
5.4
INSTITUTIONAL CONTROLS
1. Comment: ARCO generally criticizes statements made concerning the use
of institutional controls at this site.
Response: The Feasibility Study section quoted by ARCO is a fair, general
summary of current EP A policy regarding the use of institutional controls.
EP A agrees that institutional controls may supplement long term remedial
-------
2.
5.5
action, as well as provide short term protection. However, the' NCP
preamble states clearly that institutional controls should not be used as the
sole component of a remedy, except in extreme, circumstances. This was the
general thrust of the FS and the State's institutional controls memorandum.
EPA notes that institutional controls for this ROD have been id~ntified,
both for short term management, and as supplements to engineering
controls. The implementation of institutional controls regarding land use
will focus the selection of final cl~anup action levels on occupational,
recreational, current residential, and environmental scenarios.
Comment: Skepticism over the use of local, county, or state governments to
maintain and enforce institutional control is unwarranted.response: Before
zoning or easement institutional controls can be accepted, EP A believes that
concerns such as permanence and enforceability must be carefully examined.
The willingness of local, county, or state governments to adopt and
administer appropriate laws and programs and the financial ability to
administer these properly are key components. EP A will continue to
examine these issues carefully as it considers institutional controls at the
Clark Fork Basin sites, and other sites.
AREAS AND VOLUMES OF CONTAMINATED MEDIA
Comment: The comment states that the estimates of tailings and
contaminated soils used in the FS are not accurate enough to justify the
necessity or selection of remedial actions in the FS.
-------
Response: EP A and MDHES disagree with the comment. The areas
identified as containing the tailings and contaminated soils are indicated on
Figures ES-1 and 2-2. The estimates of volumes were made using standard
techniques, and are believed to be sufficiently accurate to allow cost
estimating that falls within the accuracy range specified by EP A for
feasibility studies. See the response to Comment 1 in Section 2.3.1 for a
discussion of the methods used to develop the estimates.
-------
CHAPTER 6.0
RESPONSES TO ARCO COMMENTS, CHAPTER 6.0
IDENTIFYING A RANGE OF MEDIA-SPECIFIC ACTIONS
6.1 ALTERNATIVES IDENTIFICATION PROCESS
No comments.
6.1.1 Remedial Technologies and Process Options
1.
Comment: . This comment points out that Access Controls are listed
differently in the two tables that summarize. the screening of remedial
technologies and process options (Tables 6-1 and C-1).
Response: EP A and MDHES regret the confusion caused by minor
inconsistencies in the tables, generally typographical or editing errors. These
inconsistencies did not have an impact on the conclusions reached or the
alternatives developed for the operable unit. The tables were constructed to
summarize what was done and were not a determining factor in the actual
screening.
2.
Comment: "See Appendix e regarding inconsistent and/or inappropriate
selection and screening of remedial technologies and process options."
Response: AReO's comments on Appendix e are addressed in the response
to Appendix e of this Responsiveness Summary. There are six comments,
each of which is individually addressed.
-------
Response: The two tables have been checked, and no instance of a
remedial technology or process option being dropped between the tables
could be identified. This comment may have arisen from the fact that
Table 6-2 is divided into separate sections addressing each. of the four
media. Some remedial technologies and process options retained in the
screening step summarized in Table C-l were retained for use with only one
or some of the four media. In such cases (of which there are many), the
technology or process option was not repeated under all four media, but
instead only under those to which it may potentially be. relevant.
4.
Comment: The comment states that the remedial techno~ogies and process
options screening appeared to be subjective, and in some cases inconsistent.
Two specific examples are given.
.
Comment: "Institutional Controls" was screened out for the pond
bottom sediments medium, but not for the other media.
Response: No institutional or access control was identified that would
be useful in preventing the release of the pond bottom sediments
during earthquakes or floods, which is the remedial objective for the
pond bottom sediments.
.
Comment: The remedial technology "Sediment Control Barriers" and
its two process options ("Capping Barriers" and "Settling
Basins/Ponds") were screened out, in spite of the fact that settling
ponds are the current containment method being used for the pond
bottom sediments.
Response: The comment is correct in a sense. The process option
"Settling Basins/Ponds" is screened out. The reason is that the
-------
sediments are already contained in such ponds, and thus this process
option is not needed in describing new actions that would be taken
but that would leave the sediments where they are. It was not
considered to be worthwhile to propose moving the pond bottom
sediments to new settling basins or ponds at this time. $0 this
process option was not needed.
, 5.
Comment: This comment notes that in Table 6-2, a general response action
is indicated as screened out, although one if its remedial technologies is
retained.
Response: Table 6-2 contains an editorial error. The "treatment" general
response action was actually retained.
6.1.2 Selecting Representative Process Options
1.
Comment: This comment repeats the idea that Institutionalj Access Controls
should have been retained to address flood and earthquake stabilization of
the pond berms.
Response: No institutional or access control was identified that would be
useful in preventing the release of the pond bottom sediments during
earthquakes or floods, which is the remedial objective for the pond bottom
sediments.
2.
Comment: This comment refers to the use of institutional controls in the
development of alternatives.
-------
Response: The use of institutional controls at the site is addressed in other
sections of the Responsiveness Summary, and certain institutional controls
are identified as part of the selected "remedial action. "
3&4. Comment: These comments indicate minor inconsistencies in the screening
tables.
Response: To be consistent with Table 6-3, the remedial technology listed
as "Landfill" in Table 6-2 sho11:ld have been listed as "Land Disposal." The .
media "Surface Water" and "Groundwater" in the titles of Table 6-2 should
have been listed as "Contaminated Surface Water" and "Contaminated
Groundwater."
6.1.3 Assembling Media-Specific Actions
1.
Comment: This comment asks the methodology used in assembling media
specific actions, and the goals that were the guiding principle in assembling
the media-specific actions.
Response: Media-specific actions were assembled to represent feasible
approaches to addressing the problems identified at the operable unit.
Feasibility in this context has the meaning given to it by the RI/FS guidance
document for this phase of the FS process: a combination of
implementability, effectiveness, and cost.
The guidance given in the RI/FS guidance document on the developme n t
of alternatives indicates that "Alternatives [or media-specific actions] should
be developed that will provide decision makers with an appropriate range of
options..."; that. they should be "viable or appropriate alternatives for
addressing site problems;" that they should "represent a range of treatme n t
-------
and containment combinations, as appropriate"; and. that, 'To assemble
alternatives [or media specific actions], general. response actions should be
combined using different technology types and different volumes of media
and/or areas of the site." This guidance does not give detailed procedures
or narrow criteria by which the media-specific actions should be assembled.
In light of the rather general guidance given in the guidance document, the
objective for this FS was to put together media-specific actions that would
represent the range of alternatives that would ultimately prove to be
implementable and effective; .specifically, the goal was to not leave out of .
the media-specific actionS viable alternatives for the individual media that
might prove to be implementable and effective. Cost was to be determined
and evaluated in detail at later stages in the process.
The media-specific actions are, necessarily, rather broad in scope. Being, as
they are, combinations of remedial technologies and process options, which
are thi~mselves broad concepts, the media-specific actions are broadly
drawn. It is clear on inspection that the media-specific actions assembled in
the FS provide no details; each represents a range of possible approaches.
For example, ARCO's proposed alternative is contained, in co~cept, within
the media-specific actions assembled in the FS.
The goals that guided the assembling of media specific actions were the
remedial objectives identified for the operable unit in Chapter 5.
2.
Comment: The comment states that Figure 6-1 in the report is inadequate.
Response: EPA and MDHES disagree with the comment. The figure serves
the purpose for which it is intended: to note that some interactions exis t
between the media at the operable unit.
-------
9.
3,4,
Comment: These comments point out minor inconsistencies in the
screening
5,7.
tables.
Response: Table 6-4 should have noted for both Problems 1 and 2 that
MSA 2 would require draining the pond. MSA 6 should read in all places
Surface Controls/Diversion Structures. MSA 10 should note use of. a non-
RCRA landfill.
6.
Comment: This comment repeats an earlier comment made about the
distinction between the east-west berms and the north-south berms.
Response: See the response to Section 5.3, Comment 3 and Section 6.2.1,
Comment 1. The north-south berms need to be raised and modified for
both flood and earthquake protection. Most of the east-west berms need
only be modified for earthquake protection.
8.
Comment: This comment repeats the comment made in several places
about the use of institutional controls to address the groundwater
contamination below the Pond 1 berm.
Response.:
This comment is addressed in other portions of the
responsiveness summary.
Comment: This comment repeats an earlier comment about the impacts of
the groundwater discharge to the Clark Fork River.
Response: The comment that the discharge of 1 cfs to the Clark Fork River
would not cause a measurable impact on water quality in the river is correct.
-------
6.2
RANGE OF MEDIA-SPECIFIC ACTIONS
6.2.1 Pond Bottom Sediments
1.
Comment: This comment states that the FS should be clarified to indicate
. that containment of the pond bottom sediments during floods only requires
raising and armoring the north-south berms.
Response: EP A and MDHES believe the FS is sufficiently clear regarding
which berms require modifica~ions for which purposes.. The three east-west.
berms that originally created the ponds appear not to require raising or
armoring (except in a limited area of -the Pond 1 berm) to provide
protection from flood erosion damage. However, the newer berm (trending
northwest-southeast) at the southern end of Pond 3 will require significant
modifications to provide protection from a 0.5 PMF.
2.
Comment: The comment states that references should be provided on the
use of the Pozzolanic In Situ Stabilization on a large scale.
Response: See the response to Comment 1 under Section 7.2. That
response briefly describes four demonstration-scale applications of this
technology.
6.2.2 Surface Water
1.
Comment: This comment repeats earlier comments that the need for
upstream source controls must be considered as a part of developing
remedies for the Warm Springs Ponds Operable Unit.
Response: It is beyond the scope of the FS for the Warm Springs Ponds
. Operable Unit to investigate upstream source controls. These will be
-------
investigated during future feasibility studies.
Section 4.2.2 comments.
See also the response to
2.
Comment: This comment repeats the general points of ARCO's proposed
alternative, mainly that an improved Pond 3 can be used to contain and
. .
adequately treat flows up to the tOO-year flood event.
Response: EP A and MDHES are now developing a response' for the
operable unit that is a com};>ination of. elements- of. Alternative 3 and
ARCO's proposed plan. See the response to comments in Section 7.5 "for
additional details.
6.2.3 Tailings Deposits and Contaminated Soils
No comments.
6.2.4 Ground Water
This comment is addressed in other portions of the Responsiveness Sununary.
-------
3.
CHAPTER 7.0
RESPONSES TO ARCO COMMENTS, CHAPTER 7.0
DETAILED DEVELOPMENT OF THE MEDIA-SPECIFIC ACTIONS
1.
Comment: See Comments 1, 2, and 3, under Section 3.1.4, for discussion regarding
the correct level of required flood protection for the ponds.
Response: See responses to Section 3.1.4, Comments 1, 2; and 3 and Appendix B.
The 0.5PMF standard is the appropriate ARAR for- this action, and is necessary to
protect human health and the environment.
2.
Comment: The statement that "a seismic factor of safety of 1.2 will also be
required" (page 7-1) is incorrect. The MDNRC dam safety rules indicate that dams
the size of the Warm Springs Ponds embankments must be designed in accordance
with principles at least equivalent to those in TR-60, Earth Dams and Reservoirs
(Soil Conservation Service, 1981). TR-60 requires that the minimum factor of
safety for slope stability, when analyzed with seismic forces (Le., pseudo-static
analyses), must be either 1.0 or 1.1.
Response: There was a error in the text. The factor of safety used for the seismic
analysis was 1.0.
Comment: The FS should include an additional MSA option 5D in Table 7-1 for
enlargement/upgrade of the existing Warm Springs Ponds system to serve as both a
sedimentation and treatment facility. This option. should then be developed to at
least a level of detail equivalent to the other actions. See also comments in
Section 6.2.2.
-------
Response: MDHES and EP A agree that an additional MSA should have been
included in the FS to address enlargement/upgrading of the existing Warm Springs
Ponds. An evaluation of this concept has been performed, and it has been
concluded that this concept will provide a viable alternative to those presented in
the FS. See the response to Section 7.5 comment for details of this evaluation.
7.1 MEDIA-SPECIFIC ACTION 1: STABILIZE POND BERMS TO WITIISTAND
-. FLOODS
No comments.
7.1.1Media-Specific Action 1A: Stabilize Pond Berms To Withstand A Probable
Maximum Flood
1.
Comment: This comment questions the peak flow of the PMF as used in
the FS (146,500 cis) and states that the actual value of the PMF should be
lower.
Response: The PMF calculations in the Silver Bow Creek Flood Modeling
Study (November 30, 1989) are correct as presented. As discussed in meet- .
ings with ARCO, ESA (ARCO's technical consultant), MDHES, EP A,
CH2M HILL, and HYDMET, the assumptions used by ARCO in.changing
input parameters to the HEC-l model developed by CH2M HILL were
inappropriate. An independent review was conducted by USGS on
development of the hydrologic parameters in the Silver Bow Creek drainage,
and they stated that the HEC-1 model prepared by CH2M HILL was well-
calibrated for use in the upper Clark Fork basin.
The HEC-1 model was used for the calculation of the PMF. As discussed in
the above-mentioned meeting, for PMF calculations there has to be snow in
-------
, ,
the 5,000- to 6,000-foot elevation zone. ESA assumed no snow in this
elevation zone; however, during the June 1908 flood, newspaper reports
discussed power lines being down due to. heavy snows. Also, one report
indicated 9 inches of snow on the ground in Butte and still snowing. Since,
by definition, the PMF is calculated assuming all hydrologic conditions being
at "critical conditions" and since there is historic eVidence that snow cover
has existed in the 5,000- to 6,000-foot elevation zone, it must be used during
calculation of the PMF. Also, the peak I-hour PMP valtie as used by ESA
(1.07 inches) has been exceeded twice as recorded a~ the Butte airport. This
wO"uld indicate there was an inappropriate assumption and calculation made
in obtaining the 1.07 inches.
Based on the thorough review of the calibrated HEC-l model by USGS and
the agreement made to its accuracy, and with proper hydrologic parameters
being calculated and input to the calibrated model, the estimates for the
PMF are 129,000 cfs and 201,000 cfs depending upon which value for the
PMP is utilized (HMR 43 or HMR 55A). It was agreed during negotiations
for the 1990 Mill-Willow Bypass removal action that the design value to be
used for the PMF is 140,000 cfs.
2.
Comment: The assumption that the pond berms along the Mill-Willow
Bypass would be raised using "imported, well-graded gravels" (page 7-6) is
inappropriate. There is no reason not to use suitable natural (Le.,
unprocessed) soils readily available in the immediate site vicinity.
Response: EPA and MDHES agree with this comment. If suitable
materials are available in the immediate site vicinity, they should be used in
the berm construction. The cost estimates included in Appendix D assume
some materials from the local vicinity could be used.
-------
3.
Comment: The FS should include other slope armoring "alternatives in: Foot-
note 4 on page 7-6, such as groutedriprap,gabions, and/or geowebs.
Although these alternatives mayor may not be technically suitable or cost
effective for broad application at the site, they may well prove appropriate
for local use. Also, the FS should use the terminology "armoring," rather
than "stabilization," when discussing erosion or scour protection.
Response: Footnote 4 on page 7-6 was included to clarify the cost-
estimating basis, not to provide a" complete..list of all possible armoring
. "
techniques. EP A and MDHES agree that "armoring" is the more technically
correct term. It should be noted that it has been agreed that soil-cement is
the most cost-effective method for armoring the slopes of the berms along
.
the Mill-Willow Bypass, and soil-cement will be the armoring technique
utilized.
'-
4.
Comment: The FS states that where the existing Mill-Willow Bypass would'
be covered by relocation of the pond embankments, the new channel would
be designed with the same flow capacity as the existing channel (page 7-9).
This suggests that the FS design does not consider balancing the required
embankment raises with excavation of the bypass channel to the west. The
FS design should evaluate the potential for improved hydraulics and
substantial cost savings by balancing cutting and filling. Also, the FS should
clarify whether the additional flow capacity of the bypass due to removal of
tailings and contaminated soils was accounted for in the design.
Response: Detailed balancing of earthwork in the preliminary design of the
Mill-Willow channel/embankment raise was not performed. Feasibility level
designs need only be carried to a point such that reasonable comparative
cost estimates can be performed. The EPA and MDHES agree that
balancing of the earthwork between Mill- Willow excavations and
-------
embankment raises should be done as much as is feasible. This has. been
accomplished during the final design for the summer, 1990 Removal Action.
The increase in flow capacity as a result of contaminated soil and tailings re-
moval in the bypass was not factored into the flood modeling during
preparation of the FS. The primary reason is that. the precise depth and
extent of removal was not determined in sufficient detail to justify adjusting
the flood modeling program. The increase in capacity resulting from
contaminated soil and tailings removal was factored. into revised flood
modeling performed by ARca during final design for the 1990 Removal
Action.
5.
Comment: The FS relies on a previous study by IECa (1981) to evaluate
the effects of PMF runoff from the hills east of the Ponds. The CH2M
HILL (1988) PMF model developed specifically for the Warm Springs Ponds
RI/FS does not include the hills east of the ponds in the modelled area.
Response: The Silver Bow Creek Flood Modeling Study (November 1989)
calculates the PMF for Silver Bow Creek to a point downstream of the
Warm Springs Ponds. This includes the hills to the east of the Warm
Springs Ponds. Specifically, these basins are numbered Node 30 -
Subbasin P - Pond 3 Local, Node 33 - Subbasin Q - Pond 2 Local, and local
drainage for Pond 1 is contained in Node 36 - Subbasin R - Cook Creek
Local, as shown in the flood modeling report Figure 3, page 35. The
contributing flow to the Warm Springs Ponds and subsequently into Silver
Bow Creek is calculated for these subbasins during a PMF on Silver Bow
Creek. The model, as calibrated and run, presents the flows from the hills
to the east of the Warm Springs Ponds.
-------
2.
6.
Comment: Table 7-2 should' be amended to include the elevation
corresponding to the "Storage Available within Existing Berms." A1s~, the
data and methodology used to derive storage volumes should be described.
Response: The elevation corresponding to the available storage is the top of
the berms at the .lowest spot for each pond. The storage volumes are
estimates based upon existing topographic mapping and bathymetric
mapping.
7.1.2 Media-Specific Action 1BL Stabilize Pond .aerms to Witpstand Flows Less
Than Probable Mwaximum Flood
1.
Comment: This comment states that page 7-12 of the FS should be
modified to indicate that the MDNRC dam safety rules do not specify flood
flows for the protection of the ponds.
Response: While the MDNRC Dam safety rules obviously do not specify
actual discharge values for each pond, the values shown are derived from
applying the dam safety rules. For more detail about the development of
the PMF values, refer to the responses to comments under Section 4.1.1.
Comment: The partial PMF values used in developing designs and cost esti-
mates for MSA IB should be revised to match the PMF values presented by
ARCO in Section 7.1.1, Comment 1.
Response: EPA and MDHES maintain that a PMF value of 146,500 cfs is
appropriate for the Silver Bow Creek drainage. Through negotiations on the
Mill-Willow Bypass removal this summer, it was determined that the level of
protection needed for all three ponds was 70,000 cfs, which is approximately
half of the PMF noted in the FS.
-------
5.
3.
Comment: The FS indicates on Figure 7-2 that "compacted granular fill" is
assumed for berm remediation. See Comment 2 under Section 7.1.1.
Response: EP A and MDHES agree with. this comment. If suitable
materials are available in the immediate site vicinity, they should be used in
the berm construction. The cost estimates included in Appendix D assume
some materials from the local vicinity could be used.
4.
Comment: Figure 7-2 indicat~s that no freeboard has been allowed for the
berms along the Mill-Willow Bypass.. Some freeboard should be provided.
Response: The water surface elevation shown in Figure 7-2 is incorrect A
minimum of 2 feet of freeboard should be indicated on Figure 7-2. Two feet
of freeboard was assumed for quantity and cost estimating. The cost
estimates in Appendix D include a minimum of 2 feet of freeboard for the
berms along the Mill-Willow Bypass and for the eastern hills flood control
berm.
Comment:
The FS should provide a reference for the flood volumes
included in Table 7-3.
Response: The "Designated Flood Volumes" in Table 7-3 are percentages of
the PMF values listed in IECO (1981). It should be noted that the PMF
values shown in this reference are different from the PMF values prepared
by ARCO. The differences are:
-------
. IECO (1981) ARCO (1989)
Peak Peak
Discharge Volume Discharge Volume.
(cfs) (ac-ft) (cfs) (ac-ft)
Pond 1 13,800 2,200 17,100 1,980
Pond 2 19,200 . 3,000 25,100 3,210
Pond 3 9,700 1,000 23,500 2,840
Since the PMF values listed by ARCO .are generally more conservative than IECO (1981),
the ARCO PMF values have been accepted by EPA and MDHES.
6.
Comment:
Table 7-3 should be amended to include the elevation
corresponding to the "Storage Available within Existing Berms."
Response: The elevation corresponding to the available storage is the top of .
the berms at the lowest spot for each pond. The storage volumes are
estimates based upon existing topographic mapping and bathymetric
mapping.
7.2
MEDIA-SPECIFIC ACTION 2: STABILIZE/SOLIDIFY POND SEDIMENTS
1.
Comment: The FS should document and describe the existing successful
applications of in situ stabilization/solidification in the United States or
proposed applications of this technology that are alluded to on page 7-17,
especially those involving CERCLA sites.
Response: Perhaps the FS should have used the term "successful
demonstrations." Applications of in situ stabilization/solidification that have
been demonstrated in the United States include:
-------
.
In December 1988, Soliditech, Inc. of Houston, Texas, demonstrated
its solidification and stabilization process, which chemically and
physically immobilizes hazardous constituents cQntained in slurries, at
the Imperial Oil Company Superfund site, an oil recycling facility in
Morganville, New Jersey. During the process, the proprietary reagent
URRICHEMTM was dispersed throughout the waste to micro-
. .
encapsulate hazardous compounds by cross linking organic and
inorganic particles, coating large particles, and sealing small pores
and spaces. This sealing process signific.antly reduces leaching
potential. Wastes .treated during the demonstration were from three
sources at the facility: contaminated soils from a marshy area, used
filter-clay material from an existing waste pile, and an oily sludge
from an unused storage tank. Approximately 2 to 5 cubic yards of .
each waste were processed in Solidi tech's mobile mixing unit.
Contaminants included polychlorinated biphenyls (PCBs), heavy
metals, petroleum hydrocarbons, and low levels of volatile organic
chemicals. Samples of untreated and treated waste were analyzed for
a wide range of organic and inorganic contaminants, and other
physical and chemical characteristics. Samples were also subjected to
several leaching tests. The treated wastes were evaluated for strength
and durability through tests such as the unconfined compressive
strength test, freeze/thaw tests, wet/dry tests, and microstructural
analysis. Preliminary results are not yet available.
.
Chemfix Technologies, Inc. of Metairie, Louisiana, has developed a
proprietary process (CHEMFIX) that stabilizes high-molecular-weight
organic and inorganic constituents in waste slurries. The CHEMFIX
process uses soluble silicates, silicate setting agents, and additives to
crosslink with waste components to produce a stable solid matrix.
The equipment utilized was capable of processing soils at a rate of up
-------
to 100 tons per hour. This technique was demonstrated at the
. Portland Equipment Salvage Company S1Jperfund site in Clackamas,
Oregon. The site operated. as a transformer and metal-salvaging
facility. The major waste contaminants are lead and copper, present
at maximum concentrations between 10 and 15 percent, and PCBs
present at maximum total concentrations of 2,000 to 5,000 Pl?m.
Several leaching and extraction tests were conducted on products of
the solidification/stabilization process to -indicate the long-term
stability of the processed material. The toxicity characteristic
leaching procedure (TCLP) extracts from processed wastes contained
,
lead in concentrations 94 to 99 percent less than in leachate extracts
of untreated wastes. The wet/dry and freeze/thaw durability tests
were good, showing little or no strength loss after 12 cycles. The
unconfined compressive strength at 28 days ranged from 27 to
307 psi. Permeability of the treated material ranged between 1 x 10-
6 em/sec and 6.4 x 10-7 cm/sec.
.
The solidification/stabilization process developed by H~con, Inc. of
Katy, Texas, was demonstrated at the Douglassville Superfund site,
near Reading Pennsylvania, in October of 1987. This process blends
contaminated soil or sludge with cement, pozzolans, and a proprietary
ingredient, called Chloranan, which neutralizes the inhibiting effects
of organics. The result is a concrete-like mass that contains the
contaminants. The contaminated soil wastes stabilized at the
Douglassville site came from six sources: one each from two large
lagoons once filled with waSte oil sludges and, subsequently, drained
and backfilled with soil; an oily filter cake disposal area; an oil dru m
storage area; an oil reprocessing area; and a waste land farm.
-------
During the field demonstration, 5 cubic yards of each waste were
treated. S'amples were taken from the untreated soils, the blended
slurry after 7 days of curing, and core samples, from the 28-day old
blocks: The samples were analyzed for soil characteristics,
leachability, permeability, unconfined compressive strength,
microstructure changes, and contaminant levels. , Samples of the
treated waste will also be taken over a period of 3 to 5 years.
Results indicated that the volume of the, solidified soil was
approximately double that of the undisturbed waste feed, but'the
permeabilities of the treated soil were very low, in the range of 10-8
to 10-9 em/sec. Durability tests results were good and no los5 of
strength was observed after 12 wet/dry and freeze/thaw weathering
test cycles. The TCLP test showed that heavy metals were
immobilized over the range of oil and grease encountered. TCLP
leach tests performed on untreated and treated soils showed
equivalent concentrations of volatile and semi-volatile organics in
their respective leachates. In addition, special leaching tests were run
that simulated leaching of the intact solidified cores.
.
In cooperation with General Electric, Inc., Intern~tional Waste
Technologies (IWf) demonstrated its in situ stabilization/
solidification process at a closed electric service shop in Hialeah,
Florida, in April 1988. At that site, approximately 13,000 cubic yards
of soil are contaminated with PCBs and some volatile organics and
heavy metals. IWf used the Geo-Con, Inc. Deep Soil Mixing system
to drill and blend waste material with IWf's patented bonding agent.
The IWf process bonds organic and inorganic compounds, creating
macromolecules that are highly resistant to acids and other
deteriorating factors.
-------
The major objectives of the demonstration were to evaluate the
ability of the process to immobilize PCBs in the soil; determine the
level of performance and reliability of the mechanical equipment
being used; assess the effectiveness of the process for in situ
stabilization; and to observe the integrity of the solidified soil over a
period of 5 years. During the demonstration, the stabilization/
solidification process was tested on two sectors totaling about
400 square feet. The soil was blended and stabilized in depths up to
18 feet. Samples of the untreated .soil. and the treated material were.
. .
analyzed for soil characteristics, leachability, permeability, unconfined
compressive strength, microstructural changes, and contaminant
levels. These analyses will be repeated on borings of the treated soil
to be taken one to two times a year. Preliminary results show that
the mixing system achieved a homogeneous soil/reagent blend with
minimal difficulties. The mass solidified at low permeability and high
density, but lost its integrity during freeze/thaw <:ycles. Data on the
effect of treatment on the mobility of PCBs is still being evaluated.
2.
Comment: The FS should provide brief descriptions of the "several full-
scale prototype projects" in the U.S. involving in situ stabilization, and
. .
document the success and failure of this technology (page 7-18).
Response: The words "full scale" should not have been included in this sen-
tence. The projects were prototypes (larger than laboratory tests, but not
"full-scale") and are described in the response to Comment 1 above.
-------
7.3
MEDIA-SPECIFIC ACflON 3: STABIUZE POND BERMS TO WITHSTAND
A MAXIMUM CREDmLE EAR1HQUAKE
1.
Comment: The FS. should document and describe the "conceptual level
stability analysis" performed to determine necessary corrective measures.
Response: The stability analysis was. performed at the conceptual level to
determine preliminary designs for cost-estimating purposes. A formal
stability analysis should be performed based on site-specific information
prior to final design.
2.
Comment: The FS states on page 7-20 that the factor of safety used for
seismic loading was 1.0, while on page 7-1 it is listed as 1.2. Which is
correct?
Response: As noted in the response to Section 7.0, Comment 2, the
analyses used a factor of safety of 1.0.
3.
Comment: The FS should reevaluate the conclusions of the IECO (1981)
report in terms of slope stability under earthquake loadings. Those
conclusions were necessarily based on a now outdated assumption that the
Continental Fault was the controlling source of damaging earthquakes that
might impact the Warm Springs Site (see Section 4.1.1 Comment 9). The FS
should be revised as necessary to accommodate the current understanding
that the Continental Fault is not capable of generating significant
earthquakes.
.
Response: The FS was based on previous reports concerning the geology
and seismicity of the area only as a basis to prepare cost estimates of the
-------
corrective measures. As stated in the FS, during the design phase the site
seismicity should be reviewed in detail. .
4.
Comment: See Section 7.1.1, Comment 2, regarding the inefficiency of using
imported granular materials for slope remediation.
Response: EP A and MDHES agree that onsite materials may be .adequate
for slope remediation. See response to Section 7.1.1, Comment 2.
-------
7.4
3.
MEDIA-SPECIFIC ACTION 4: REMOVE AND DISPOSE. OF
CONTAMINATED SURFACE SOILS FROM ALONG mE MILL-WILLOW
BYPASS
1.
Comment: The FS assumes that it will be necessary to remove and dispose
of 130,000 cubic yards of contaminated soil. This estimated soil volume is
not based on action levels. The actual volumes of soils requiring
remediation are not known.
Response: The estimate' of 130,000 cubic yards of contaminated soil.was
prepared prior to the final determination of action levels and was for use in
cost estimating only. During the design phase, detailed field testing of
contaminant levels was performed and detailed plans were developed for the
areal extent and depths of removal. When the Mill-Willow Bypass removal
is complete, the amount of tailings and contaminated soil removed will be'
known.
2.
Comment: See Section 7.7.2.3, Comment 2, regarding the inapplicability of
transporting tailings and contaminated soil to an offsite RCRA facility.
Response: This MSA was developed to assemble alternatives that exceed
the requirements of ARARs. See response to Section 7.7.2.3, Comment 2.
Comment: On page 7-22, the FS indicates that disposal location options for
tailings and contaminated soils include "Pond 1, 2 or 3 prior to solidification
or capping." In other places the FS specifies that disposal would be in
Pond 1. The FS should evaluate locating two or more disposal facilities in
the ponds closest to the source of the materials to be disposed.
-------
Response: During the FS, it was determined that plaCing the materials in
Pond 1 prior to capping was more protective of the environment, and that
option was selected for cost estimating purposes. The' discussion on page 7.
22 should have been changed to reflect this decision for the cost estimating
basis. It has since been determined during negotiations for the 1990 Mill.
Willow removal that two disposal sites (one in Pond 1 and one in. Pond 3)
are equally protective' of the environment.
7.5 MEDIA-SPECIFIC ACTION 5: IMPROVE.1HEPOND TREATMENT SYSTEM
Comment: This comment states that the FS is deficient because it does not include
ARCO's preferred alternative.
Response: The concept of raising the Pond 3 berms to contain the flood flows was
one' of many concepts considered during the early stages of development of MSA's
for addressing upstream tailings and sediment transport. This concept was not,
pursued in depth, because the option of an upstream impoundment was more
promising in terms of protecting human health and the environment. It was
determined during preparation of the FS that a separate upstream impoundment
would offer greater flexibility in terms of operation of the overall pond system. By
separating the functions of flow equalization/primary sedimentation from
biological/chemical treatment, the operation could be more reliable and flexible in
dealing with large flood flows. In addition, providing primary sedimentation
upstream of Pond 3 would extend the life of Pond 3 by capturing a considerable
portion of the sediments during large runoff events.
In light of adverse reaction by ARCO and the public to the concept of an upstream
impoundment, EPA and MDHES re-evaluated the concept of flood storage within
Pond 3. It was concluded that utilizing Pond 3 for flood storage of the 100-year
-------
"
event was capable of providing protection to human health and the environment
equal to an upstream impound~ent.
It was also concluded upon re-evaluation that Pond 2, if upgraded as suggested by
ARCO, could provide polishing and additional treatment during normal operations.
These concepts were incorporated into the proposed plan iricluded in the Record of
Decision (ROD) as Alternative 3 + 3A. The primary flood storage and treatment
features of this concept can be summarized as follows:
.
Allow the ponds to remain in place; Ponds 2 and 3 will continue to function
as treatment ponds.
.
Raise and strengthen all pond berms to protect against dam failure in the
event of major earthquakes or floods and increase volume capacity of
Pond 3 to store flows up to the lOO-year flood volume.
.
Construct new inlet and hydraulic structures to reduce debris plugging of the
Pond 3 inlet and to safely route flows in excess of the design peak flow of
the 100-year flood around the ponds.
.
Comprehensively upgrade the treatment capability of Ponds 2 and 3 to treat
flood volumes up to the 100-year flood and construct spillways for decanting
excess flood water into the bypass channel.
The EP A and MDHES still have concerns. regarding the potential to resuspend
pond bottom materials during high flow or high wind events and the potential
effects upon the biota in the Clark Fork system, if these materials are released to
the environment. Accordingly, th~re is a provision in the ROD requiring additional
study of the potential to resuspend pond bottom materials and the effects upon th~
environment of any resuspension. If resuspension is shown to be a significant
-------
problem, the ROD includes provisions requiring that additional measures be' taken
by ARCO to mitigate the resuspension of these materials.
7.S.1Media-Specific Action SA: Overall Vpgrade oUhe PondJreatment
System
1.
Comment: The FS should document the data, .methodology, and
specific analyses used to arrive at a design maximum flow rate of
600 cfs for Pond 3. The FS should also elaborate on the potential to .
. .
adjust the maximum flow rate during remediation design.
Response: The alternatives presented in the FS were developed in
sufficient detail to provide conceptual-level design numbers according
to standard engineering practice. The design maximum flow rate
provided (600 cfS) was developed to provide adequate detention time
I
for metals removal based upon engineering judgment and information,
in the literature. Footnote 13 on page 7-24 acknowledges that the
results can be adjusted during the remedial design phase of the work.
The optimization of design should be based on field-scale pilot
treatability testing using various flow assumptions and artificially
created flood waters (since actual flood waters would not be
available). As noted in the response to Comment 7.5, Pond 3 will be
used for flood storage. The influent structure will be designed for a
peak flow of 3,300 cfs, and the decant towers (outlet structures) will .
be designed for a combined peak flow of 750 cfs during the lOO-year
event.
2.
Comment: This comment states that the federal water quality
criterion for copper used in the FS (12 J.£g/l) is incorrect, because it is
based on a hardness of 100 J.£l. The comment claims that the
7 - 18
-------
,~
measured average hardness in Warm Springs Ponds is 175 mg/l, and
that the FWQC for copper at a hardness of 175 mg/l is 19 p.g/I. The
comment states that the standards. for cadmium, copper, lead, and
zinc should also reflect a hardness of 175 mg/I.
Response: . ARCa has used the average hardness value for the
Pond 2 discharge (PS-12), rather than that for the inflow to the Mill- .
Willow Bypass. The hardness in the ponds is not. relevant, since, as
ARCa has pointed out in a separate commeJ;lt, the water quality
standards are not currently applied to the ponds. The average
hardness of water entering the Mill-Willow Bypass, as measured
during the Phase I RI at Station SS-18, was 129 mg/I. This value is
much closer to the stated basis for calculating the aquatic standard
for copper than the 175 mg/l hardness suggested by the comment.
The value of 100 mg/l hardness used in Table 7-4 was used to
provide a common standard to which concentrations measured in
inflow water to the Mill-Willow Bypass and Pond 2 discharge water
could be compared. Use of average values or specific values and
associated hardnesses does not change the conclusions that were
developed from the table.
3.
Comment: This comment notes that Table 7-4 shows an average
copper concentration of 31 p.g/l at the inlet to the Mill-Willow
Bypass. The data used to calculate this value were taken from the
1987 RI (MultiTech, 1987). The data from the Phase IT RI (CH2M.
HILL, 1989) show an average copper concentration of 13.5 p. g/l
entering the Mill-Willow Bypass. The more recent data indicates
ambient concentrations of copper in Mill and Willow Creeks do not
exceed the chronic level FWQC for aquatic life under normal flow
conditions.
-------
Response: . Phase I RI surface water data are much' more
comprehensive than Phase II RI surface water data. The Phas~ I RI
data collection effort was designed t9 characterize surface water
chemistry over a variety of flow conditions, as compared to Phase II
data, which were collected to characterize diurnal changes in surface
water quality on a seasonal basis. Phase I data were collecte~ during
15 sampling episodes conducted over a 9-month period, while
. .
Phase II data were collected during' four 24-hour periods during
September, January, April, and-July.. The 'Phase I data are more,
. ,
representative of site conditions than the Phase II RI data, which
were collected for a different purpose. The MDHES is not aware of
any major changes in the site conditions that would indicate that
conditions during the Phase II RI were different from those charac-
terized during the Phase I RI. Sampling by ARCa during 1990 in
Mill and Willow Creeks has confirmed exceedances of the water
Guality standards in Willow Creek.
Section 7.5.1, Comment 4, below.
See also the response to
~,
4.
Comment: This comment, along with Comments 3 and 15 of this
section, makes several points regarding the contributions of Mill and
Willow Creeks to the ambient concentrations leaving the operable
unit. Specifically, the comments state that the FS is in error
regarding the relative contributions of Mill and Willow Creeks to the
levels of contaminants in the surface water as it leaves the operable
unit, and that the FS should have more clearly stated that diverting
Mill and Willow Creeks into the pond system would elinrinate the
fisheries on those streams.
Response: The comment is correct in pointing out that the
contributions of Mill and Willow Creeks to the surface water
-------
'.1
contamination may not be as large as stated in the report. ARCa
has suggested that the design investigation for Warm Springs Ponds
evaluate the contamination in the two creeks further, determine its
sources, and evaluate source-specific actions to address the
contamination, if necessary. EP A and MDHES concur with this
approach.
5.
Comment: The capacity of the Pond 3 inlet structure is stated to be
700 cfs in the FS. Previous - studies by MCa (ESA, 1987) have
indicated that the capacity. is approximately 1,400 cfs. The- FS
estimate of inflow capacity should either be justified or corrected.
Response: Calculation of the inlet structure capacity (Silver Bow
Creek Flood Modeling Study, CH2M ~ 1988) indicated a
maximum flow rate of approximately 900 cfs under ideal conditions.
This estimate was downrated to 700 cfs to account for plugging, age,
and actual field conditions.
6.
Comment: Footnote 16 on page 7-29 should reference the basis for
the estimated flood flows. Diversion structure facilities proposed in
the FS are sized for a 100-year flood of 4,900 cfs. As was stated in
Section 4.1.1 of this review, the lOO-year flood on Silver Bow, Mill
and Willow Creeks that should be used in the MSA for facility sizing
is 4,000 cfs.
Response: The USGS has confirmed that the HEC-l runoff-
simulation model, as prepared by CH2M HILL, is appropriate for the
conditions in the upper Clark Fork basin. The peak flow calculated
in the combined Mill-Willow-Silver Bow Creek system for the 100-
year event was 4,900 cfs without considering any diversion into
-------
Pond 3. See also the responses to Section 4. L 1, Comment 1, and
Section 7.1.1, COInn1ent 1.
7.
Comment: ARCa believes. that the comprehensive treatment pond
upgrades proposed in MSA 5A will not consistently attain FWQC for
surface water quality at normal flows and especially during higher
flows approaching 600 cfs. A review of data from the Phase II RI .
. .
(CH2M HILL, 1989) shows that during normal flows, full utilization
of Pond 3 and Pond 2.volumeis required to. achieve FWQC for both
copper. and zinc. Samples taken at the effluent from both ponds
showed that the additional retention time provided by Pond 2 was
required to meet Gold Book criteria (EP A, 1986) for chronic aqu'atic
life during September and January sampling events for zinc. These
data indicate that the longer retention time provided by Pond 2 in
series with Pond 3 was required to meet discharge criteria at pond
pH levels up to 9.0 and flows ranging from 2 to 105 cfs. Treatability
studies conducted as part of the Phase II RI confirmed that a large
fraction of precipitated metal hydroxide particles are very small and
typically require long retention times (on the order of 20 days and
longer) to settle out completely.
Response: Upon evaluation of ARCa's Plan 3A, EPA and MDHES
agree that Pond 2 can provide polishing and additional metals
removal after the pond operating depth is increased by 2.0 feet and
flows are limited to 200 cfs, as proposed by ARCa. EP A and
MDHES believe, however, that Pond 2 will provide effective
additional treatment only during low flow and low wind conditions.
The RI data indicate that Pond 2 currently can act as a source for
metals during high wind events. The degree to which Pond 2 will
continue to serve as a source for metals following an increase in
-------
operating water level is unknown. Thus, as .a result of further
deliberation, it has been decided that the system will be designed to
allow discharge directly from Pond 3 into the Mill-Willow Bypass, if
required. This will allow Pond 2 to be bypassed during any
conditions in which the discharge from Pond 3 is of higher water
quality than the discharge from Pond 2.
8.
Comment: Overall upgrades to the pond system proposed in the FS
call for lime addition into a lined channel approximately 2,000 feet in
length from the discharge into Pond 3. The advantage of the new
channel is said to be that it will allow better mixing of lime and
prevent lime and metal hydroxides from settling in the influent
channel. However, current lime addition in the winter, which occurs
at the intake structure to Pond 3 shows no visual evidence of lime or
metal hydroxide accumulation in the creek at or below the intake
structure. lime is currently added as dry-powdered quicklime during
cold weather conditions and any problems with undissolved lime
accumulating in the creek would most likely occur during this time.
Response: As stated on page 7-30 of the FS, the primary reason for
channelizing the flow from the inlet structure to the ponds would be
to better control the direction of the flow and to limit the interaction
with tailings deposits above Pond 3. Better lime mixing is a
secondary benefit of the channel. Upon re-evaluation, it has been
decided that the potential benefits of a lined channel do not justify
the additional expenditure, and. this concept has been deleted from
the Recommended Alternative (3+3A) detailed in the ROD.
9.
Comment: The lime addition treatment system in the FS i~
. conceptual. An estimated cost accuracy of + 50 percent to .
-------
30 percent cannot be met based on the level of detail provided in this
process description. Also, a redundancy of 100 percent for all '
components of a treatment system, as assumed in the FS, is not
required to provide a sufficient margin of safety..
Response: FS cost estimates are required to have a +50j-30 percent
overall accuracy. for each remedial alternative, taken as a whole. It is
. .
not necessary for each line item to meet this accuracy requirement.
The overall cost estimates .for .the. alternatives. are designed to be .
order-of-magnitude estimates for comparative purposes. An order-of-
magnitude estimate is defined as an approximate estimate made
without detailed engineering data. The lime addition system pre-
sented in the FS meets these requirements.
The 100 percent redundancy specified for the system is necessary due
to the remoteness of the site, its exposure to extreme natural
elements, and the continuous nature of the natural stream flow.
10.
Comment: Predictions of hydroxide sludge volumes resulting from
treatment by lime to pH 9 (FS, page 7-35) are based on 0.66 ml of
sludge per litet of water treated. However, this value conflicts with
data from lab-scale treatability studies in the Pbase II RI (CH2M
HIL4 1989), which show that less than 0.5 ml of sludge per liter of
water will result from lime treatment at pH 9. Also tbe value of
.0.66 mljl does not distinguish between sludge volume resulting from
hydroxide precipitation and sludge volume resulting from settling of
sediments and previously suspended solids.
Response: The predicted volume of sludge generated due to lime
treatment (0.66 mIjl) used in the FS is taken from the results of the
-------
barrel-scale treatability tests performed as part. of the Phase II RI.
The barrel-scale tests are thought to more accurately predict pond
deposition rates.
The value of 0.66 mIll (FS, page 7-35) represents the hydroxide
. sludge generation rate. Generation due to the settling of suspended
solids and bio-solids is included in the additional 2~000 cu~.ic yards
per year identified in the text.
11.
Comment: The comprehensive treatment pond upgrades presented in
the FS included use of Pond 3 at its present volume with the addition
of a berm partition and new influent and effluent structures. When
constructed, the berm will create a relatively narrow (1,000 foot)
opening to channel the full normal and flood flows through Pond 3.
At the berm opening, flows will accelerate to higher velocities,
resulting in resuspension of settled materials in this part of the pond.
Construction of the new influent structure will narrow the full flow
inlet to a single opening of only .100 feet in width. Erosion and
resuspension of settled particles will occur in this area as well,
especially during flood conditions.
Convergence of flows to a single discharge point will also likely cause
disturbance of settled solids. The combination of new inlet and
effluent structures with the proposed internal berm structure will
result in a shift of the existing flow pattern across the pond. The
pond will establish a new equilibrium during which lime-settled
,
material will be redistributed and substantial quantities of
precipitated metal contaminants may be discharged from the pond.
In addition, the construction of the berm will cause disruption of the
pond bottom sediments underneath and around the. area of
-------
construction, resulting in substantial short-term. di~charges of. metal
contaminants into the Mill-Willow Bypass.
The FS contends that the new berm will reduce the chance of flows
channeling through Pond 3. However, there is no evidence that
channeling is currently a problem. The new berm is also. said to
reduce the effects of wave action caused by high winds. If winds are
demonstrated to create a problem, a log boom or similar structure
could be designed to work as effectively as a berm in reducing wave
action.
Response: The opening at the east end of the berm would be sized
to avoid resuspension of sediments in this area. The average velocity
through the opening would be approximately 0.13 fps at 600 cfs. This
velocity is approximately one-half of the velocity necessary to induce
resuspension of the pond bottom sediments (as noted in the report of '
erosion potential prepared for ARCa by Simons and Associates, 1990).
The EP A and MDHES agree that the new influent structure would
resuspend some sediment in the vicinity of the structure during high
flows. Given the nature of this system, it is virtually impossible to
design an influent structure that will not tend to resuspend some
materials during higher flows. The reason is straightforward. During
normal flow periods, some of the materials entrained in the flow will
tend to settle out (as ,they reach their respective settling velocities) in
. a particular pattern around the influent structure. When the flows
increase during runoff events, the increased velocities will cause some
of these materials to resuspend. However, since these are the
particles. that have comparatively high settling velocities, they \\;11
. merely settle out further downstream in the pond system. In othc: r
-------
words, the sediments around any influent structure will be in a state
of flux as long as influent flows vary significantly.
The convergence at the outlet structure was considered during
preparation of the FS. The preliminary design was based on limiting
the average approach velocity to 0.3 fps around the structure. . Similar
to the phenom~non noted above, however, approach velocities around
the structure vary with the flow rate. Thus, there may be some
unavoidable resuspension around any outlet. structure during higher
flows.
The agencies agree that construction of the berm would cause
. .
disturbance of the sediments in the vicinity of the berm. The
construction should be scheduled during low flow periods in Silver
Bow Creek to maximize the detention time to allow the disturbed
sediments to resettle. The MDHES does not agree that changing
flow patterns in the pond would result in "substantial quantities" of
materials being discharged from the pond.
The FS did not state that "channeling" of flows was a current
problem. The FS stated that the berm would help prevent short
circuiting. It is a well-established design principle that maximizing
the flow path in a settling basin is a desirable goal. This helps
promote better settling efficiency and reduces short circuiting,
especially at higher flows.
The reduction of wave action is only one aspect of the issue of wind
action. The winds during a storm event tend to create currents in the
ponds that are sufficient to resuspend pond bottom sediments.
Reduction of these wind-induced currents is a major function of th~
-------
12.
berm across Pond 3 that could not be addressed by log booms. In
addition, the long-term efficacy of log booms in ponds subject to
freezing conditions is questionable.
It has been decided to delete the specific requirement for a berm
across Pond 3. Instead, the Recommended Alternative (3 + 3A) in the
ROD contains provisions to study and potentially model the effects of
resuspension. The ROD also contains provisions requiring additional
mitigative measures, if. resuspension is demonstrated to be a problem. .
A berm across Pond 3 is one option that would likely be considered
as a mitigative measure.
Comment: Upgrading the lime addition facility, as proposed in the
FS, will improve water quality by maintaining a pond pH of 9.
However, the Phase n RI (CH2M HILL, 1989) showed that the
additional capacity of Pond 2 was still required at a pH of 9.
Furthermore, routing Mill and Willow Creeks into Pond 3 for
treatment will result in higher average flows through the pond requir-
ing increased settling capacity. The comment further states that
elements of ARCO Plan 3A, such as including the use of Pond 2 and
expanding Pond 3, address the problem of insufficient storage
capacity and retention time.
Response: EP A and MDHES agree that use of the flood storage and
treatment upgrades, provided by ARCO's Alternative 3A, would
provide improved water quality equivalent to that which would be
provided by MSA SA. Thus, these upgrades have been incorporated
into the Recommended Alternative (3 + 3A) in the ROD.
-------
14:
13.
Comment: The comprehensive upgrade of the pond treatment. in the
FS' as MSA SA, even when combined with an upstream flood
impoundment (MSA 6A), will not effectively treat the full 100-year
flood run-off.
Response: The flood impoundment identified in MSA 6A would be
sized to store up to the 10o-year event, while allowing a 600-cfs
bypass. Following the storm, the waters would be inetered out of the
upstream impoundment at a maximum rate. of. 600 cfs. The design
capacity of the treatment system identified in MSA SA is 600, cfs;
thus, the comprehensive treatment scheme would be able to treat the
full volume of a lOo-year event.
Comment: The FS contends that the remaining treatment life of the
existing Pond 3, operating under the range of conditions described in
this action, is from less than 10 to approximately 80 years. The FS
then concludes that a new treatment pond would be necessary if flows
from Mill, Willow, and Silver Bow Creeks still required treatment.
Assuming the remaining life estimate is accurate, the more
appropriate approach to resolve this issue is to increase the storage of
the existing Pond 3, as per the ARCa P~an.
Response: The ARCa Plan indicates that Pond 3 would operat~ at
approximately its current water surface elevation under normal
conditions. While it is true that the useful life of Pond 3 could be
extended if the normal operating elevation of Pond 3 were allowed to
increas~ over time, this would necessarily result in a associated
reduction in the flood storage capacity over time.. In fact, routing of
floods into Pond 3 will cause it to fill with sediments much more
quickly than if the majority of those sediments could be settled out in
-------
16.
17.
an upstream impoundment. This will cause a reduction in the useful
life of Pond 3 under ARCQ's Altemative3A.
15.
Comment: See Comment 4 above concerning the need to route Mill
and Willow Creeks into the pond system for the treatment.
Response: See the response to Comment 4 above in which EP A and
MDHES concur with ARCQ's plan to investigate source control for
Mill and Willow Creek!;.
Comment: The FS should address whether the proposed internal
berm in Pond 3 can be constructed without substantive additional
cost, given the poor foundation. conditions inherent in the pond
bottom sediments.
l~esponse: The berm across Pond 3 would likely be constructed using "
techniques similar to cofferdam construction in nvenne
environments. It would involve beginning at the existing western
berm and constructing toward the east by pushing rock and gravel
ahead as the berm is constructed. The dumped rock and gravel
would tend to displace the pond bottom sediments until stable foun-
dation materials are reached. It may be necessary to add additional
materials in the future, if settlement occurs in portions of the berm.
To account for the foundation conditions, the cost estimates for this
berm included in Appendix D of the FS assume that more material
would be required than if the berm were constructed in the dry.
Comment: The FS must justify the need to replace the current PV C
siphons feeding the Wildlife Ponds from Pond 3.. ARCQ is unaware:
-------
~,
18.
19.
of any significant problems with the historic operation of the existing
system.. If significant problems. are documented, the FS should
provide additional detail regarding location. and design of the
proposed ''buried pipe system" (page 7-43). The estimated cost for
this design feature has not been included in Appendix D of the FS.
Response: It is not normally considered good engineering practice in
installations such as this to utilize. exposed inverted PVC siphons,
except as 'temporary measures. They are sJ.lbject to loss of prime,
deterioration by ultraviolet rays, and potential freezing. This is,
however, an insignificant issue and will not affect water quality in the
Clark Fork River. It was included in the FS only because it would be
a good idea to install buried piping during the stabilization of the
berms. The costs were considered insignificant in comparison to the
rest of the project and have been included in the contingencies.
Comment: The FS is deficient by not evaluating the alternative
approach of raising the Pond 2 dike to increase its storage and,
thereby, improve the treatment capacity and extend the remaining life
of this pond.
Response: The EP A and MDHES evaluated this concept and have
included it in the selected alternative (3 + 3A) in the ROD. See the
response to the Section 7.5 comment.
Comment: This comment asks for the data, methodology and .specific
results for "the calculated detention time in Pond 2" (page 7-43).
Response: The detention time is calculated from the volume of
water contained in Pond 2 at normal elevation and the average flow
-------
rate into Pond 2. At a surface elevation of 4;832 ms!, the water
volume of Pond 2 is 211 acre-feet (incorrectly listed in Table 2-1 as
860 acre-feet). At an average flow of 53 cfs into Pond 2, the
detention time is 2 days. (Reported as 3 days in the FS).
20.
Comment: The FS does not provide sufficient detail regarding the
location and design of the ''buried discharge system. . . to discharge
the water from Pond 3 to Pond 2." (page 7-44) Also, a cost estimate
for this design feature J.las not been included in Appendix D of the.
FS.
Response: The details of this system should be addressed in the
design phase. It would likely consist of a concrete headwall with a
slidegate, and a buried 8- to 12-inch pipe into Pond 2. The costs
were considered insignificant in relationship to the overall costs and
are included in the contingencies.
21.
Comment: The existing inlet structure is incompatible with this
. proposed MSA The existing structure and berms in the vicinity will
have to be raised substantially to exclude the PMF-series floods.
However, the structure does not have the physical space to
accommodate such a raise. In addition, the existing wood gates are
in poor condition and, without repair or replacement, could not be
used to control Pond 3 . inflows to a 600 cfs maximum flow.
Consequently, a new structure should be built to control inflows into
Pond 3.
Response: The concept shown in the FS includes. raising the berms
on the iriside (northeast) side in the vicinity of the inlet structure.
This would require extensions of the outlet pipes beneath the raised
-------
22.
"
I
berm. However, the EP A and MDHES agree that operation ,of the
gates under major flood conditions would be impossible using this
scheme. The EP A and MDHES also agree that the condition of the
existing gates may warrant their replacement, or the construction of
an entirely new structure. A new structure is included in the
Recommended Alternative (3 + 3A) in the ROD.
Comment: The construction of the new effluent structure proposed
in the FS will be very difficult and expensive, dlle to the necessity of '
placing the structure on a sound foundation. To cons,truct' the
foundation, a cofferdam will have to be built about 200 feet into the
pond to protect the construction area. The pond sludge should be
removed before the' cofferdam is pushed into the pond to ensure an
adequate foundation for the cofferdam. In addition, the existing em-
bankment will have to be removed to allow installation of the
discharge pipes. The embankment will then have to be rebuilt and
most of the cofferdam removed.
This action also calls for outright abandonment of the two existing
outlet structures. The structures have deteriorated concrete in some
areas and several of the existing 60-inch discharge pipes show signs of
duress evidenced by cracking and damaged joints.
Response: ARCO's concerns regarding difficulties in construction of
the new effluent structure are valid and the same concerns were
taken into account during preparation of the FS. The costs shown in
Appendix D include the measures necessary to address these
concerns.
-------
H MSA 5A were to be implemented, the two existing outlet structures
should be abandoned such that they could withstand the MCE. This
would probably include plugging the structures. and discharge pipes
with concrete but the details would be evaluated during final design.
7.5.2 ,Media-Specific Action 5B:
.Treatment System
Less Comprepensive U:pgrad~
1.
Comment: Many of th~ comments in Section 7.5.1 also apply to this
section,. as the actions are quite siinilar in many regards.
Response: Comment noted.
2.
Comment: This comment questions why the FS evaluates MSA 5B,
which limits the treatable inflow to 210 cfs to accommodate the
existing condition of Pond 2, rather than evaluating the possibility of \.
increasing the storage capacity of Pond 2.
Response: Media-Specific Action 5B was developed to provide a
comparison to the more comprehensive upgrade (MSA 5A) for the
purposes of evaluating costs against treatment efficiency. The
concept was to utilize as much of the current system as possible to
minimize construction costs. To include the expense involved in
raising the Pond 2 berms and modifying the outlet structure would
have defeated the purpose of developing a minimum cost treatment
alternative.
-------
7.5.3 Media-Specific Action 5C: Construct a New Treatment Pond
v
.1.
Comment: Comments under Section 7.5.1 and/or 7.5.2 regarding:
deficiency in data, methodology and specific results to support pond
sizing; the proposed lime addition system; the proposed effluent
system; and the proposed internal berm also apply to this. media-
specific action. .
Response: See the specific responses to t~e appropriate technical
areas in Section 7.5.1 and/or 7.5.2.
2.
Comment: The FS should describe the design rationale an~/or
constraints that result in limiting the inflow capacity of the proposed
new treatment pond to 600 cfs.
Response: The capacity of the new treatment pond is size limited.
As stated on page 7-47 of the FS, the new treatment pond would
have a detention time of just under 2 days with a flow of 600 cfs.
This is considered to be the minimum detention time for adequate
settling.
3.
Comment: This comment repeats earlier comments suggesting that
the capacity of the pond treatment system could be increased per
ARCa Plan 3A .
Response: EP A and MDHES agree, with certain qualifications. See
response to the Section 7.5 comment.
4.
Comment: This comment repeats earlier comments concernjng the
need to treat flows from Mill and Willow Creeks.
-------
Response: EP A and MDHES agree with ARCO's concept to
investigate source control on Mill and. Willow Creeks. See the ~
response to Section 7.5.1, Comment 4.
MEDIA-SPECIFIC ACTION 6: CONSTRUcr AN UPSTREAM FLOOD
IMPOUNDMENT ISETI1..ING BASIN
7.6
> Comment: This comment repeats Section 7.5 comment stating that the FS should
have considered ARCO's Plan 3A in. terms of the upstream impoundment.
Response: EP A and MDHES agree with ARCO's basic concept to store floods in
Pond 3. See the response to the Section 7.5 comment."
7.6.1 Media-Specific Action 6A: Construct an Upstream Flood Impoundment
1.
Comment: The FS should provide estimates of the sediment loading::.
\.
anticipated at Pond 3 in the "less than 600 cfs" (page 7-52) flows to
be released from the upstream flood impoundment during a lOO-year
flood, and the loading assuming the upstream pond is not
constructed. The first estimate must include consideration of the
sediment deficit (i.e., increased sediment carrying capacity) of the
flows released from the flood impoundment. It is not possible to
independently evaluate the contention that "the potential would be
substantially reduced for future erosion of the estimated
200,000 cubic yards of tailings that exist along Silver Bow Creek
between the new pond site and Pond 3" (page 7-52).
Response: The analyses performed during the FS did not include
calculations of the anticipated differences in sediment loading to
Pond 3 with and without the upstream impoundment. The stateme n t
-------
concerning potential reduction of erosion of the. tailings between the
new pond site and Pond 3 was based on the physical parameters of
Silver Bow Creek. Even though the 600 cfs flows would be in a
sediment deficit condition following release from the upstream
impoundment, the potential to resuspend significant contaminated
sediment is limited.. This is because Silver Bow Creek at flows below
600 cfs is generally confined to its channel and would not significantly
erode the tailings (which are generally outside of the channel). It
should be noted that regardless of the quantity of contaminated
sediments that were eroded at 600 cfs, the flows would still be
diverted into Pond 3 for treatment.
2.
Comment: The FS does not include the data; design assumptions,
methodology and specific results of the "preliminary mass balance"
performed to size the upstream flood impoundment. Without this
,nformation, an independent evaluation of the feasibility of the
proposed impoundment and its operation is difficult.
Response: The FS is a summary of numerous evaluations and
calculations. Including all backup information in the FS is not
~ecessary. The referenced calculations and backup data are
available, but the need to review these calculations should be moot at
this time because EPA and MDHES have decided to eliminate the
upstream impoundment from the Recommended Alternative.
3.
Comment: The FS is inconsistent regarding the design release rate of
the proposed upstream. flood impoundment. It is listed as 600 cfs in
one place and "about 500 cfs" in another place.
-------
6.
Response: The precise release rate would be adjustable through a
gated structure. The maximum release rate for design of the outlet
should be at least the 600 cfs that Pond 3 can treat; but, good design
practices would dictate a higher maximum release rate (say 700 or
800 cfs) for flexibility of operation.
4.
Comment: The FS should define/clarify the meaning of the modifier
"normal" in reference to the lOO-year flood (see pages 7-52 and 7-53
of the FS).
Response: The term "normal" should not have been used in the text.
The reference should have been to the shape of the calculated
hydrograph for a lOO-year run-off event. It is likely that tbe actual
flows would vary somewhat from the calculated hydrograph. The
extent of variation cannot be determined without experiencing an
actual flood.
5.
Comment: This comment agrees that selection of the lOO-year flood
is appropriate as a maximum design criterion for interception of
sediment carried by Silver Bow Creek, which includes tailings and/or
contaminated soils. .
Response: Comment noted.
Comment: Subsurface investigation data from. the site of the
upstream flood impoundment are not available. The design of such a
structure cannot be technically supported without sufficient subsurface
data. The cost-effectiveness and technical feasibility of this MSA
cannot be evaluated with any degree of certainty without such
information.
-------
7.
8.
Response: The site of the upstream flood impoundment was visited
during preparation of the FS. The basic embankment section and
foundation preparation were based on reasonable assumptions for the
sole purpose of developing the cost estimates and would have to be
verified during the design phase. Subsurface explorations and a
geotechnical investigation of the site would be completed during the
design of the upstream flood impoundment.
Comment: The FS should provide the rat~onale for designing the
. .
upstream flood impoundment against a full PMF (page 7-68). Based
upon the applicable MDNRC dam safety criteria, a 0.5 PMF flood is
the maximum appropriate design criteria.
Response: EP A and MDHES agree. The appropriate design flood
for protection of the upstream impoundment embankment is
0.5 PMF. The 0.5 PMF was used in the cost estimates included in.
Appendix D.
Comment: Because Mill-Willow flows will be routed into Pond 3, a
release of 600 cfs from the upstream impoundment cannot occur at
all times. In fact, the flow from Mill-Willow alone may exceed the
600-cfs inflow limitation of Pond 3 during larger flood events.
Response: H Mill and/or Willow Creek flows ~e routed into Pond 3,
the detailed design should provide a structure capable of routing Mill
and Willow flows during major floods into the bypass. This would
allow the treatment capacity of Pond 3 to be used primarily for
treating Silver Bow Creek flows.
7 - 39
-------
11.
9.
Comment: The. proposed 20-foot high diversion dam 'would
accumulate sediment and debris. The decrease of velocities behind
the diversion dam would cause the coarser sediment to drop 'out.
The FS does not address this problem.
RespoIlSe: . As with any diversion dam, periodic maintenan,ce after
large runoff events would be required. The O&M costs for this
. .
maintenance are included in Chapter 8 of the FS.
10.
. .
Comment: The upstream impoundment inlet structure proposed -in
the FS will actually pass in excess of 8,000 cfs to the reservoir at
maximum water level, not the 4,000 cfs design flow cited. In addition,
there is no means to shut off inflow in the event of an emergency. A
structure with sluice gate control and less construction than the
proposed orifices would alleviate most of the problems and should be
substituted for the design in the FS.
Response: EP A and MDHES acknowledge that the inlet structure
will allow more than 4,000 cfs into the upstream impoundment during
flood events greater than the 100-year event. The preliminary design
presented in the FS was developed primarily for cost-estimating
purposes. Refinements of the FS design would be investigated during
the design phase.
Comment: Thelow-Ievel impoundment outlet incorporated in the FS
design will remobilize and discharge sediment to Silver Bow Creek.
A tower intake may alleviate this problem.
-------
13.
Response: The design of the outlet structure was developed in
sufficient detail to generate the .cost estimates. Alternative designs
would be investigated during the design phase. .
12.
Comment: The spillway width in Figure 7-10 is 150 feet, while the
width in Figure 7-15 is 100 feet. A 150-foot-wide spillway would be
required to pass the maximum inflow, which is in excess of 8,000 cfs.
Which is correct?
Response: The 15~foot width for the spillway is correct.
Comment: It appears that a part of the PMF flow could escape to
the back (south) side of the proposed upstream flood impoundment.
If this were to happen, the south dike could be washed out, allowing
the PMF flow to enter the impoundment and cause failure. Only the
east dike has rip rap protection. The design should be reevaluated
and the FS revised as necessary, including revising the cost for this
item in Appendix D.
Response: The note shown in Figure 7-11 is incorrect. Riprap
protection should be provided for the south berm also. Costs for this
riprap were included in Appendix D.
7.6.2 Media-Specific Act~: Construct an Upstream Sett~
1.
Comment: The FS should provide a reference or appropriate
analyses to support the conclusion that a typical reduction in
detention time of 40 percent is applicable to the proposed upstream
settling basin.
-------
Response: It is difficult to determine the actual efficiency of a
settling basin such as this without detailed modeling studies. The
reduction in detention time was used to. account for the inefficiencies
of the real world as opposed to ideal conditions used in theoretical
Stokes' Law calculations. The 40 percent reduction in efficiency was
an estimate based upon experience with similar basins. If it were
necessary to determine actual removal efficiencies, detailed -modeling
studies would be performed during the design phase.
2.
Comment: The FS concludes on page 7-70 that the - spillway
discharge from the upstream settling basin will carry substantially less
sediment than the flow entering the basin during the 100-year flood.
Given this fact, the FS must examine the potential for increased
erosion and transport of sediment along Silver Bow Creek between
the settling basin and Pond 3. It appears that a significant percentage
of the sediment load settled out in the upstream basin would be re-
gained as the relatively clean basin discharge flows down the
intervening stretch of Silver Bow Creek. The effectiveness of this
media-specific action must be reconsidered based on the results of
appropriate sediment transport modelling. See also Section 7.6.1,
Comment 1.
Response: EPA and MDHES agree that there is a potential to
mcrease erosion and transport of sediment between the settling basin
and Pond 3, since these flows would be in a sediment deficit
condition during major flood events. The extent of the regain in
sediment load and the potential contamination associated with these
sediments is not known. If the decision is made to construct an
upstream impoundment in the future, an evaluation should be made
of the erosion potential of flows between the impoundment and
-------
Pond 3, including the potential for dissolved metals contamination
from the tailings eroded. If the evaluation indicates unacceptable
erosion potential, the contaminated soils and. tailings between the
upstream basin and Pond 3 should be removed prior to completion of
the upstream basin.
3.
Comment: The FS should reference the particle size data used as a
basis for concluding that the settling basin will remove approximately
80 percent of particulates from the 100-year flood flow~
Response: Because particle size data are not available for sediments
carried by an actual 100-year flood, it was necessary to estimate. the
sediment characteristics. The estimates were made by evaluating and
averaging particle size distributions for streamside sediments collected
at several points along Silver Bow Creek. The original data was
developed from the "Streamside Tailings and Revegetating Studies,
STARS Phase I, Appendix B, October 1989."
4.
Comment: The FS should provide an estimate of the possible
variation in size of the upstream settling basin which might result
from optimization during design.
Response: There are several design studies that would be required
for the optimization process to determine the size of the basin.
These should include as a minimum:
.
Erosion and Sediment Transport Study-This study would
examine the potential of the waters released from the basin to
erpde and transport tailings between the basin and Pond 3.
As discussed in the response to Section 7.6.2, Comment 1.
-------
these waters would be in a sediment deficit condition: The I
modeling should be done for a variety of flow and s'ediment
deficit conditions to adequately define the design parameters.
If this study indicated a significant potential to erode and
transport these tailings, it might be necessary to remove the
tailings in conjunction with the construction of the basin.
.
Sediment Contamination Study-- This s.tudy would exaInine the
relationship between - transported sediments and associated,
metals contamination. It would entail modeling sediments
transported from upstream of the ponds in terms of particle
size distribution against associated metals, both dissolved and
total metals. The study should model several different sized
storms to adequately defme the design parameters.
.
Basin Modeling Study-- This study would model the settling
efficiency of the upstream basin to determine sediment
removal rates. It should be based on a model of the physical
dimensions of the basin, taking into account wind-induced
currents, density currents and particle size distributions. The
study should include modeling of different sizes and shapes of
basins to adequately define the design parameters.
The results from these studies would be evaluated together to
determine the optimum size for the upstream impoundment. Since
the results of these studies are not available at the present time, any
estimate of the possible variation in size of the upstream
impoundment would be strictly speculation.
-------
5.
Comment: Comments under Section 7.6.1 are. also applicable to
similar design features for the upstream settling basin.
Response:
Section 7.6.1.
See the responses to the specific comments in
7.7 MEDIA-SPECIFIC ACTIONS 7, 8, 9, AND 10: ISOlATE TAILINGS DEPOSITS
>ANDCONTAMITNATEDSO~
No comments.
7.7.1 Locations of ThilingLDeposits anq Contaminated....smls
1.
Comment: This comment states that a discussion of the locations of
tailings deposits in Chapter 7 appears to be misplaced.
/'
Response: This discussion was placed in this section next to the
discussions of possible remedial actions for the tailings and
contaminated soils. These discussions could fit either in Chapter 7 or
in Chapter 2.
2.
Comment:
This comment refers the reader to Comment 1 ,in
Section 7.4 and Comment 1 in Section 4.1.3 regarding quantities of
t&lings and contaminated soils.
Re~ponse: The responses to these specific comments are included in
the respective sections.
-------
7.7.2 Media-Specific Actions.
No comments.
7.7.2.1
Media-Specific Action 7: Cap and Revegetate Tailings
Deposits and Contaminated Soils
Comment: This comment asks the rationale for the proposed
design of the soil. cap--6 inches of tiUed-in agricultural lime
and 18 inches of imported soil. The comment also asks for a
description of the nature and source of the imported soil,. and
states that local soil, if amended, should be suitable for use in
capping contaminated areas.
Response: The use of agricultural lime was proposed to
. .
reduce the mobility of the metal contaminants in the materials
that are capped. The 18 inches of cover soil is derived from
the mining reclamation ARARs discussed in Appendix B of
the FS. The use of the word "imported" was not meant to
imply that the soil would come from a great distance, only that
the cover soil would not be merely an amendment of the
contaminated materials. The imported soil would be
uncontaminated soil brought either from other areas within
the operable unit or from nearby areas outside the operable
unit.
7.7.2.2
Media-Specific Action 8:
Contaminated Soils
Flood Tailings Deposits and
-------
~;-.
1.
Comment: The FS should document the reasons for
concluding that flooding areas of tailings deposits will cause
increased groundwater contamination (p~ge 7-78).
Response: The FS did not conclude that flooding will result in
"increased" groundwater contamination. The FS stated that
the contjnued presence of water in Pond 1 provides a source.
of recharge ~nd contaminants to the shallow aquifer below
Pond 1. The analyses supporting this conclusion are
summarized in the FS.
2.
Comment: It is our opinion that dry closure of Pond 1. will
not eliminate the saturated tailings within the pond as a source
of long-term groundwater contamination as is contended in the
FS. The FS provides no analyses to document this contention,
as discussed in Section 7.7.2.6, Comment 2.
Response: The decision to dry close Pond 1 is primarily based
on current groundwater quality conditions beneath Pond 1.
Monitoring well WSP-GW-10S is located in the western-most
cell of Pond 1, a considerable distance from the flooded
portion of Pond 1. Tailings were penetrated at this location to
a depth of approximately 10 feet, with groundwater levels
ranging from 9 to 11 feet below ground surface. Dissolved
arsenic concentrations measured in samples collected from
monitoring well WSP-GW-10S were considerably less than
MCLs of 16.8 pg/l.
Monitoring well WSP-GW-llS was completed near the
eastern-most cell of Pond 1, proximal to the flooded portion of
-------
the pond. Water levels in monitoring well WSP-GW-llS
indicate that up to 3 feet of the tailings are saturated.
Dissolved arsenic concentrations measured in monitoring well
WSP-GW-llS were as high as 105 p,g/l.
. Dissolved metals concentrations measured in monitoring wells
completed downgradient of Pond 1 offer additional supporting
. .
evidence that groundwater quality. improves to~ard the
western portion..of Pond.1. Dissolved arsenic concentrations .
. .
in monitoring well WSP-GW-03S, located north of th~ western
portion of Pond 1, were below the laboratory detection limit of
2 p,g/l. Dissolved arsenic concentrations measured in
monitoring well WSP-GW-13S, located downgradient of the
flooded portion of Pond 1, were as high as 61.2 p,g/l. These
data offer empirical evidence that the saturated tailings and ,""'-
. I
ponded water in the eastern portion of Pond 1 are a source of
metals contamination in the shallow groundwater system.
3.
CommeI!t: The FS should provide the design height of the
proposed new berms across Pond 2 and Pond 3 under this
action. This information is necessary to allow independent
evaluation of technical feasibility and cost estimates for this
feature. The FS. should also describe the anticipated
foundation conditions and constructability of such a berm, and
. accommodate these issues in the design and cost estimate.
Response: The berms will vary from 0 to approximately
18 feet in height, depending upon topography. The exact
dimensions of the berms would be determined during detailed
design. It should be noted that the FS is not intended to be a
-------
-,\
7.7.2.3
design. document and that preliminary ,designs were, only
carried to a level of detail sufficient to perform cost
estimating. It was anticipated that difficult foundation
'conditions would be encountered during construction of
portions of these berms. These anticipated difficulties were
factored into the cost estimates in Appendix D.
4.
Comment: The FS should clarify whether or not the flooded
Pond 2 is intended to provide physical. and biologic treatment
of the water discharged into the Mill-Willow Bypass. It is
assumed on the basis of the information provided that either
this is the case or the FS does not anticipate such discharges
will violate applicable standards.
Response: Minimal flows would be maintained through
Pond 2. Because of wind action, the concentrations of metals
in the discharge may at times exceed the standards. Howevert
given the low outflow from the pond, the metals in the
discharge should not cause violations at the compliance point,
once the small outflow from Pond 2 is mixed with the Pond 3
outflow and the Mill-Willow Bypass flows.
Media-Specific Action 9: Excavate From All Applicable Areas
and'Disposal In An Off site RCRA Facility
1.
Comment: This comment refers the reader to several previous
comments regarding soil removal action levels.
Response: These comments are addressed in other portions of
the responsiveness summary.
-------
3.
2.
Comment: This comment states that conSideration of offsite
disposal of tailings and contaminated soil~ in a RCRA
hazardous waste facility is not appropriate, because the RCRA
requirements are not. appropriate for mining waste manage-
ment. Mining wastes are disposed of throughout the country
1n a manner protective of public health and the environment.
Response: This media-specific action was developed in order
to assemblealtematives. that exceed the requirements of .
ARARs. The. MDHES and EP A have concluded that, while
the RCRA requirements are not applicable to the wastes,
certain RCRA requirements are relevant to the situation at
Warm Springs Ponds, and it will be appropriate to follow
them. It is not only appropriate to consider actions that
exceed the requirements of ARARs, but it was required by the
National Contingency Plan in effect at the time that the draft
FS was completed.
,..
I
Comment:
This comment repeats earlier comments about
action levels for soils. It also notes a discrepancy in the FS
regarding which materials along the bypass would be removed.
Response: See the response to the comments on Chapter 3
and in Appendix A for a discussion of the risk assessment and
action levels. The discrepancy concerning exposed tailings and
contaminated soils is noted. The intent of the tabulated values
at the bottom of page 7-83 was to include both exposed
tailings and contaminated soils.
-------
7.7.2.4
/
4.
Comment: This comment states that removing nearly
1 million cubic yards of. soils and tailings below Pond 1,
exposed in Pond 2, along the Mill-Willow Bypass, and above
. Pond 3 for offsite disposal is not feasible or necessary, to pro-
tect human health and the environment.
Response: The option of disposal of contaminated materials
at an off site RCRA facility was investigated to provide a cost
comparison with onsite disposal. Although such disposal. is
feasible, EP A and MDHES agree that this action is.. not
necessary. Thus, it is not part of the recommended
alternative.
Media-Specific Action 10: Excavate from All Applicable Areas and
Disposal Onsite in a Non-RCRA Facility
1.
Comment: This comment notes an inconsistency in the
description of Media-Specific Action 10. In Chapter 6,
MSA 10 is described as including disposal in a non-RCRA
facility. However, in Chapter 7, MSA 10 is described as
involving disposal in a "RCRA-equivalent facility."
Response: The comment is correct. As originally developed,
MSA 10 was to involve an onsite (i.e., somewhere within the
Silver Bow Creek Site) disp.osal facility to be developed to
serve purposes other than just the Warm Springs Ponds
Operable Unit. It was realized during the study that a more
specific disposal option would have to be assumed in order to
develop cost estimates that would allow all of the alternatives
to be compared on an equal basis. Pond 1 was identified as a
-------
suitable site for disposal of material~ excavated from various
locations within the operable unit. .
ConcernS about the oxidized state of some of the sediments in
Pond 1, together with the knowledge that much of the
. groundwater contamination below Pond 1 was a result of
leaching. of contaminants from Pond 1, led to the conclusion
. .
that a relatively impermeable cap would be needed to reduce
the. potential. for infiltration,... and. that monitoring the
.. .
. groundwater below Pond 1 would be necessary to ensure. that
the current releases were effectively reduced. The RCRA
standards were referenced for relevant and appropriate design
and monitoring standards for such a situation, and the RCRA
standards were adopted for the conceptual design.
Late in the preparation of the FS report, it was decided that
the intent of MSA 10 would be more clear if the resulting
disposal facility to be developed at Pond 1 were described as a
RCRA-equivalent facility. Most of the substantive
requirements that a RCRA landfill would have to meet
regarding the cap and groundwater monitoring would be met
by MSA 10. The Chapter 6 table mentioned in the comment
. .
should have been changed to describe the onsite disposal as
being in a RCRA-equivalent facility.
2.
Comment: The FS should explain and justify the rationale for
excavating the tailings from all applicable areas and hauling
them all to Pond 1 for capping.
-------
Response: This media-specific action. was develop~d to
provide the decisionmakers with an alternative that would
consolidate the scattered wastes into a. single disposal area
. where they could be protected from floods, isolated by a better
cap, and monitored more effectively. Since publication of the
. FS, it has been decided that it would be equally protective of
human health and the environment to have two disposaI areas:
within Pond 1 and within the dry area of Pond 3. The disposal
site for the 1990 Removal Action is .within Pond 3. Future
disposal of contaminated materials will be either within
Pond 3 or Pond 1, depending upon the. economics of haul
distances.
3.
Comment: This comment notes that the cap for Pond 1
developed as a part of MSA 10 would include a limited
permeability clay layer, and that the cap appears to be
designed to meet a subset of the RCRA closure requirements.
The comment states that the RCRA requirements are not
ARARs for this action.
Response: The MDHES and EP A have determined that the
RCRA regulations do contain relevant and appropriate
requirements for the actions to be taken at Warm Springs
Ponds. However, upon further consideration of RCRA
ARARs, EP A has determined that an impermeable cap is not
appropriate for the disposal facilities, which are part of the
selected remedial action.
-------
7.7.2.5
3.
Media-Specific Action 11: Collect Groundwater and Treat in
a Wetland Below Pond No.1
1.
Comment: The FS should provide references and/ or
descriptions of the methodology, data, and results from which
the wetlands design parameters were developed.
Response: The design parameters used for the development
of the wetland.MSA represent. a conservative estimate of .
current wetland operating parameters based upon engineering
judgment and information available in the literature on the
technology. The primary limiting .design factor of 1 gpm .per
500 square feet of available surface area was presented on
page 7-87 of the FS.
. I
2.
Comment: An arrow in Figure 7-18 suggests that Cell 2 will
receive flow from the realigned Mill-Willow Bypass. If this is
the case, the reason for this operation should be explained; if
. not, the figure should be corrected.
Response: What appears to be an arrowhead in Figure 7-18 is
a smudge on the. drawing.
Comment: The FS incorrectly implies that the entire area
below Pond 1 is covered by tailings and/or contaminated soils
to a depth averaging 3-1/2 feet. A large majority of these
materials are located along the historic Silver Bow Creek
channel, as shown in FS Figure 2-2.
-------
7.7.2.6
Response: The FS was not intended to imply that the ,entire
area is covered in tailings. As noted in the comment, the
figures make it clear that the' tailings, are concentrated in
several areas. The intention of the statement on page 7-87 is
to convey to the reader that the tailings deposits are
substantial and not, merely surface contamination. The
474,400 cubic yards of tailings and contaminated soil deposits
in this 76-acre area calculates out to an average of 3.87 feet
deep.
Media-Specific Action 12: Groundwater Collection In
and Below Pond No.1 With Treatment in Pond No.3
1.
Comment: The FS should describe the data, methodology,
and specific results which result in the estimate of 2 cfs as the
pumping rate from the combined groundwater trenches in this
action (page 7-91).
Response: The FS includes summaries of many calculations
and analyses performed during its preparation. These
calculations were carried to a sufficient level of detail to
perform cost estimating. The .calculations should be refined
during remedial' design based upon site-specific data to
determine the design parameters for pump station and
pipeline sizing.
2.
Comment: The FS contends that "the groundwater collection
trench at the toe of the Pond 2 berm...would be expected
to...eventually drain saturated material in Pond 1 to below the
base of the tailings contained in the pond" (page 7-90). The
-------
FS should be expanded to reference or to provide
appropriately detailed analyses to document this contention,
mc1uding an estimate of the tim~ required to drain below the
base of the tailings. On the basis of the meager information in
the FS, it is ARCO's opinion that at least the base of the
. tailings would remain saturated over the long-term, as is the
case at nearby. Opportunity Ponds.
Response: A computer modeling analysis was performed .to .
. .
evaluate the effects of installing a groundwater interception
trench in the eastern portion of Pond 1. Results of this
analysis were presented in a Technical Memorandum (Chen
Northern, 1989).
The model showed that steady state conditions would be
reached in the alluvial groundwater system underlying the
tailings within 500 days after installation of the interception
trench. During the initial 500 days, and for an unknown
period of time following, the saturated tailings would gravity
drain to the underlying alluvial system. The tailings would
release water until the field capacity of the fine-grained
tailings is reached. Vertical and horizontal permeability values
of the tailings were assumed based upon experience and data
from similar areas within th~ operable unit. Actual
measurements of these permeabilities would be required in
order to estimate the actual time required for the saturated
tailings to completely drain.
EPA and MDHES realize that a more detailed analysis should
be completed for design considerations prior to constructin~
-------
the interception trench. However, based on the ~rrent
conditions in the western portion of Pond 1, groundwater
levels beneath the tailings are expected to drop below the base
of the tailings. The volume of water draining vertically into
the underlying alluvIal aquifer is expected to be minimal
relative to the volume of water migrating to the north in the
sand and gravel aquifer.
-------
CHAPTER 8.0
RESPONSES TO ARCO COMMENTS, CHAPTER 8.0 ASSEMBLY AND
ANALYSIS OF ALTERNATIVES
8.1
DEVELOPMENT OF ALTERNATIVES.
1.
Comment: This comment repeats earlier comments regarding the action
levels for soils assumed in the FS.
Response: These comments are addressed -in the Chapter 3.0 and App~ndix .
A responses.
2.
Comment: This comment notes an inconsistency in how onsite disposal of
tailings and contaminated soils is described under some of the alternatives. .
It also repeats previous comments about whether or not the RCRA
regulationS can be considered to be ARARs for this site.
Response: The inconsistency noted in the comment is acknowledged. The
intention is that, if any of these alternatives are chosen, the tailings and
. contaminated soils would be disposed of in an onsite facility, and that the
most likely place for such a facility is assumed, for purposes of the FS, to be
Pond 1.. The facility is described as a "RCRA-equivalent facility" to indicate
that certain of the RCRA landfill requirements would be adopted as
relevant and appropriate for the design of the cover and for the monitoring
requirements for Pond 1. The Mill-Willow Bypass removal includes use of a
disposal site within Pond 3.
3.
Comment: This comment repeats earlier comments regarding the
incorporation of institutional controls in the alternatives.
-------
Response: See the responses to these issues in Chapter 6.0.
8.2
DETAILED ANALYSIS OF ALTERNATIVES
1.
Comment: This comment notes that the discussions of the alternatives in
Chapter 8 of the FS do not repeat discussions that are relevant to more than
one alternative, but instead refer the reader to the initial occurren~e of each
point. The comment notes that this makes it more di"rficult to keep the
alternatives straight.
Response: ARCO's description of the structure of the FS is accurate. But,
EP A and MDHES disagree that this makes it more difficult to distinguish
between the alternatives. In fact, this format was chosen in order to make
the differences between the alternatives more clear by focusing on the
differences. The alternative approach of repeating discussions of each
feature each time it occurs in an alternative is more difficult to follow; the (
\
differences between the alternatives get lost in a mass of largely repetitive
discussions that tend to mask small distinctions.
8.3
INDIVIDUAL ANALYSIS OF ALTERNATIVES
1.
Comment: Thiscomrnent states that the upstream impoundment would
result in significant changes in terrestrial and aquatic wildlife habitat, and
that such changes must be addressed in the FS.
Response: EPA and MDHES acknowledge that this action, like any of t~;:
actions discussed in the study, would have impacts on some wildlife. Due:,
adverse reaction by the public and ARCO, EPA and MDHES have decide:
not to pursue the upstream impoundment or settling basin options.
-------
~~~~.~I
2.
Comment: This comment notes that Alternatives 1 to 6 developed in the FS
would have impacts on wetlands, endangered species, and historical
resources.
,The comment further notes that the FS should identify adverse impacts and
probable costs to mitigate these impacts. Finally, the comment notes that
ARCa's proposed plan would result in "no net loss of existing wetlands."
Response: It is true that the alternatives in the F~ would, as ARCa's plan
would, result in impacts on wetlands, endangered species, and historical
, ,
resources. The required action in such cases is exactly as given in the FS:
when the proposed action becomes concrete enough to determine, the
probable impacts (usually during the design stage), the responsible party
must consult with the appropriate agencies in order to incorporate mitigative
measures in the project plan or design. Until the alternative to be
implemented is chosen and partially designed, the regulatory agencies will
not be able to give specific guidance on the mitigative measures likely to be
required.
While ARCa's plan may not, as currently conceived, result in a net loss in
.1QaJ wetlands, certainly some of the existing wetlands would be affected.
Any of the alternatives in the FS could be designed and implemented so as
to result in no net loss of total wetlands, and such a mitigative measure may
be required by the responsible regulatory agencies once the remedial
alternative for the operable unit is chosen and the impacts on wetlands can
be determined in detail. The impacts of ARCa's plan on endangered
species or historical resources are likely to be similar to those for the
alternatives in the FS.
-------
3.
8.3.1 Alternative 1:
1.
Comment: This comment repeats an earlier comment that the flood flows
. .
on Mill and. Willow Creeks were not properly considered in sizing the
upstream impoundment.
Response: H Mill and/or Willow Creek flows are routed into Pond 3, the
detailed design should provide a structure capable of routing Mill and
Willow flows during major floods in the bypass. . This would allow the.
treatment capacity of Pond. 3 to be used primarily for treating Silver Bow
Creek flows.
2.
Comment: This comment points out ambiguous language in the FS that
could be interpreted to imply that a 100-year life span is intended for the
upstream impoundment.
Response: EP A and MDHES acknowledge that the language was
ambiguous. The intent was to state that the upstream impoundment could
treat floods up to the peak flow of a 100-year flood.
\
I...'
Comment: This comment states that the FS should establish the point of
compliance for the contaminated groundwater as the downgradient property
boundary or other boundary based on the use of institutional controls and
that the FS incorrectly implies that all MCI.s are exceeded.
Response: The point of compliance is established in the ROD. Certain
MCI.s are exceeded at the site.
-------
6.
4.
Comment: This comment repeats an earlier comment about the need for
the FS to include costs of wetland, endangered species, and historical
resources impact mitigation.
Response: Proposed mitigation for these resources will be developed as part
of the remedial design phase. Adequacy and cost of mitigation will be
evaluated at that time. For more detail, see response to Section 8.3,
Comment 2.
5.
Comment: ARCa concurs with the' FS characterization of any flood flows
. .
on Silver Bow, Mill, and Willow Creeks in excess of the IOO-year event as
extreme cases for which treatment for suspended solids and dissolved metals
is unjustified (page 8-29).
Response: Comment noted.
Comment: This comment notes the following problems that might be
associated with an upstream impoundment, such ~ the one included in
Alternative 1.
A.
Blowing dust from the surface of any accumulated tailings, even if
periodically hauled offsite.
B.
Institutional and design considerations related to periodic transport of
settled tailings for disposal offsite.
C.
Potential for groundwater contamination if settled tailings are left in
place.
D.
Final closure requirements.
-------
E.
Utility relocation (i.e., high-voltage power lines).
F.
Creation of a contaminated. area in what was previously an
uncontaminated area.
G.
Land acquisition (both in terms of legal issues and costs) and the
resulting sacrifice of agricultural lands and the tax base tpey repre-
sent.
Response: These issues are no longer relevant because EP A and
MDHES have decided not to pursue the concept of an upstream
impoundment.
7.
Comment: This comment states that costs to remove and dispose of tailings
settled in the upstream impoundment should be included in the operation
and maintenance costs for the alternatives.
Response: The operation and maintenance costs were estimated from the
level of work that would be required to carry out all of the O~M activities
related to each alternative. Allowances were made for full- and/or part-
time staff at the ponds, periodically hiring a contractor to perform certain
duties, and materials and other costs. Periodically removing small amounts
of collected tailings and sediments from the impoundment, if necessary.
could be performed within the costs estimated. Costs for removal of larger
amounts of tailings f~om major floods were not addressed because the
probability of such an event is difficult to account for in a cost estimate.
These issues are no longer relevant because EPA and MDHES have decided
not to pursue the concept of an upstream impoundment.
-------
2.
8.
Comment: This comment states several reasons why ARCO believes
Alternative 1 would not be implementable or reliable.
Response: While not necessarily agreeing with each point made in the com-
ment, EP A and MDHES agree that Alternative 1 should not be selected.
That is why it was not identified as the preferred alternative in the Proposed
Plan.
8.3.2 Alternative 2: Exceeds the Requirements otARARs
1.
Comment: This comment repeats an earlier comment regarding sizing of
the upstream impoundment.
Response: EP A and MDHES acknowledge that this action, like any of the .
actions discussed in the study, would have impacts on some wildlife.
Howev~r, the impacts would not likely be as severe as apparently envisioned
by ARCO. The primary reason is that the upstream impoundment, or
settling basin, would be empty except during serious floods. The impacts of
the impoundment or basin when empty would largely be only the impacts of
the berms themselves. Since the area is currently privately owned grazing
land, the effects on terrestrial wildlife should be minimal. Impacts on
aquatic wildlife should be positive, not negative, given that the purpose of
the impoundment or basin would be to improve surface water quality. EP A
and MDHES have decided not to pursue the upstream impoundment or
settling basin options.
Comment: This comment states that the inclusion of protection of the pond
system from a full PMF in this alternative is not justifiable. The comme n t
goes on to state that the appropriate level of protection to be provided for
-------
5.
the ponds should be "derived from an analysis of the MDNRC dam" safety
regulations. "
Response: The purpose of a FS is to explore a range of options. It is
entirely proper to examine the costs, benefits, and impacts of providing
protection from the full PMF as part of one of the alternatives in the FS.
EP A and MDHES agree that implementing full PMF protection would be
excessive; the preferred plan included partial PMF protection. The level of
protection proposed was deriv~d from MDNRC dam safety regulations. "
3.
Comment: Thiscomment repeats Section 8.3.1, Comments 6 and 7, which
note potential problems with an upstream impoundment.
Response: These issues are no longer relevant because EP A and MDHES
have decided not to pursue the concept of an upstream impoundment.
4.
Comment: This comment agrees with an assertion in" the FS regarding the
potential for recontamination of the Mill-Willow Bypass by floods over the
100-year event.
Response: Comment noted.
Comment: This comment repeats previous comments regarding the effects
on fisheries" of diverting Mill and Willow Creek flows into Pond 3.
Response: In response to opposition from the public and ARCO to the
concept of routing the flows of these two creeks into the pond system, the
selected remedy for the operable unit will not include this action. Instead, if
the impacts will" cause future exceedances of applicable or relevant and
appropriate requirements, then action to address those exceedances, such ;~,
-------
source control actions, will be investigated. .These investigations would be
conducted in conjunction with other investigations at the Anaconda
Superfund site. . EP A and MDHES reserve the authority to require the
. .
. .
diversion of Mill and/or Willow Creek into the pond system for treatment.
See also the responses to Section 7.5.1, Comments 3 and 4.
6.
Comment: The FS estimate on page 8-49 of the total volume of tailings and
contaminated soils to be excavated and hauled offsite (160,000 cUbic yards)
is inconsistent with the estimate given under Media-Specific Action 9 on
. .
. .
page 7-83 (940,000 cubic yards). . Also, the FS is confusing in that MSA 9.
addresses removal of the tailings and sediments in Pond 2, while
Alternative 2 incorporates that action, but uses wet closure to remediate
Pond 2. Even when this factor is accommodated, a significant discrepancy in
removal volumes still exists (i.e., 160,000 cubic yards versus 290,000 cubic
yards).
Response: The volume noted on page 8-49 is in error. The total volume of
tailings and contaminated soils should have been listed as "Approximately
290,000 cubic yards. . ." on page 8-49. This is the quantity used in the cost
estimates for Alternative 2 in Appendix D.
As noted in the introduction to Chapter 8, the media-specific actions (and
options) were combined to form the various alternatives. There was no
intention and no requirement to limit the alternatives to discreet
combinations of intact media-specific actions. Alternative 2 uses a
combination of Media-Specific Actions 8 and 9 for addressing tailings and
contaminated soils. Other alternatives use similar combinations of media-
specific actions.
-------
2.
3.
8.3.3 Alternative 3:
1.
Comment: As previously discussed in Comment 1 . of Section 8.3.1, the
operation of. the upstream settling basin ignores the Mill. Willow Creek
flows. The FS should fully evaluate as an alternative source controls on Mill
and Willow Creeks and enlargement of the existing Ponds 2 and 3 as
proposed in the ARCQ Plan; These actions would eliminate the need for an
upstream impoundment and treatment of Mill. Willow. Creek flows. As
noted previously, the need f~r remediation of Mill and/or Willow Cre,ek ,
flows must' first be documented with new data more definitive than.' that
. .
referenced or presented in the FS.
Response: EPA and MDHES concur with the concept of investigating
source controls. See the response to Section 8.3.1., Comment 1.
Comment: See Comments 1, 2, and 3 under Section 3.1.4 and Comment 4
under Section 4.1.1 regarding the inflow design flood levels proposed in the
FS.
Response: The inflow design floods have been agreed to by ARCa and
EPA/MDHES as noted in the response to Section 4.1.1, Comment 4.
Comment: Alternative 3 would bypass, without treatment in Pond 3, the
majority of the volume of a 100-year flood. As noted in Comment 5 under
Section 7.6.1, we concur that the resulting surface water quality impacts from
such an occurrence would be minimal, due to the significant dilution under
such high flows. In any case, the ARCa Plan incorporates raises of both the
existing Ponds 2 and 3 to permit treating flows from Silver Bow Creek up to
the 100-year flood, as well as providing sediment trapping up to that event.
-------
6.
7.
The FS should be revised to evaluate this much more efficient and cost-
effective approach.
Response: EP A and MDHES concur. The Recommended Alternative in
the ROD incorporates the concept of treating the lOO-year flood in Pond 3.
4.
Comment: The issues raised in Comments 6 and 7 under Section 8.3.1
regarding an upstream flood impoundment also apply. to the upstream
settling basin inco.rporated in Alternative 3. The orily difference is the scale
of the problems.
Response: These comments are no longer valid, because EP A and MDHES
have decided not to pursue an upstream impoundment.
5.
Comment: See Comments 11 and 14 under Section 3.1.2 regarding the point
of compliance issue and the applicability of a mixing-zone for groundwater.
Response:
These comments are addressed as other portions of the
responsiveness summary.
Comment: See C0II1II.1ent 2 under Section 7.7.2.4 regarding the absence of
justification for the proposed excavation and hauling of tailings and
contaminated soils for on-site disposal in Pond 1.
Response: Contaminated soils and tailings will be disposed of within both
Ponds 1 and 3, as noted in the response to Section 7.7.2.4, Comment 2.
Comment: This comment refers to previous comments regarding RCRA as
an ARAR for Warm Springs Ponds.
-------
10.
Response:
These comments are addressed in other portions .of the..
responsiveness summary.
8.
Comment: This comment states that the FS is misleading in that it states
that Alternative 3 would comply with the MDNRC dam safety regulations up
to lower flow rates than Alternatives 1 and 2.
. .
Response: Alternative 3 is clearly identified as meeting ARARs. The intent
of the sentence on page 8-55 is to: 1) state that this alternative would meet.
the dam safety regulations, and 2) distinguish the protection provided. as
being at lower levels than would be provided under Alternatives 1 and 2.
9.
Comment: This comment focuses mainly on whether or not mining
reclamation requirements can be considered ARARs for this operable unit.
It also notes that the FS mentions the possibility of designing in-place
capping actions to meet mining reclamation ARARs.
Response: These cominents are addressed in the response to comment~,
Chapter 3.0 and Appendix B.
Comment: This comment notes that failure of the berms during a flood or
earthquake would be unlikely to release all of the 19 million cubic yards of
toxic tailings and sludges in the ponds. The comment' also repeats an earlier
comment regarding the potential release scenarios for the tailings ant!
sludges.
Response: EP A and MDHES agree that neither a flood nor an earthqua ~~.
would be likely to release 100 percent of the tailings and sludges, a r, :
believe that the FS is clear on this point in several sections. The sente r., .
pointed out in this comment was misworded. The correct wording, wr.
-------
appears on page 8-42 for Alternative 2, deletes the words "release of," and
reads: ''The primary risk at the site, the 19 million cubic yards of sediments
in the treatment ponds,..." The potential for release of a substantial fraction
of the tailings and sludges should be the focus of concern, and that potential
threat is not in question.
11.
Comment: This comment notes an error on page 8-59, where it" is stated
that the tailings that collect in the upstream impoundment '~ be re-
moved" periodically, while page 8-58 states thai the tailings "would be
removed."
Response: Page 8-59 should have used the phrasing "would be removed,"
since this is the assumption used in the cost estimates for this alternative.
8.3.4 Alternative 4: Compiles Wit~
r' ,
\
1.
2.
Comment: Most of the comments on Alternative 3 also apply to this
alternative, since the only significant difference involves capping in-place
versus removal of tailings and contaminated soils below Pond 1 and above
Pond 3.
Response: Comment noted.
Comment: .Comment 5 under Section 8.3.3 regarding a groundwater
compliance point and the applicability of a mixing-zone also applies to this
alternative.
Response: These comments are addressed elsewhere.
-------
3.
Comment: ARCa concurs that "Capping contaminated soils and. tailings
deposits in-place wherever possible would effectively isolate them from
direct contact and limit their mobility." (page 8-66), and that this action
would "isolate the material to inhibit human and environmental exposure"
(page 8-69). As discussed in Comments I and 2 under Section 7.7.2.4 and
Comment 5 below, there is no justification for any other action, except in the
case of the Mill-Willow Bypass or where flooding-closure is technically more,
feasible and cost effective.
Response: Cpmment noted.
4.
Comment: This comment states that the FS does not support the conclusion
that removing tailings from below Pond I into Pond I to cap and better
contain them would be cheaper and more effective than leaving them in the
floodplain and capping them there.
')
Response: During the FS, the possibility of protecting the tailings below
Pond I from flood erosion was considered. A soil cap could not be relied
upon to provide protection from erosion due to a 0.5 PMF. The agencies do
not believe that it would be appropriate to protect a capped area to only a
100-year flood. The possibility of protecting the cap with soil-cement,
concrete, or rip rap was briefly explored. Because of the flow rates that would
exist, and the uncertainty that any of these methods would be reliable, they
were not pursued further. It was obvious on brief examination that it would
be cheaper and more effective to remove the tailings from the potential
threat of erosion than to protect against it.
Moving the tailings to Pond 1 as envisioned in Alternative 3 would be much
less expensive. The cost for moving the tailings is estimated to be $400,000
to $500,000. This compares to a very preliminary estimate of $3 to
-------
r--
I'
I
$4 million to provide a cap, which includes soil cement erosion protection.
This estimate is based upon an assumption of 6 inches of agricultural lime,
and a 24-inch cap comprised of 12 to 18 inches of clay and 6 to 12 inches of
soil cement erosion protection.
5.
Comment: This comment makes further points regarding the issue of
capping the tailings below Pond.1 in place or moving them to Pond 1. The
main point of the comment is that the tailings would pose no more risk if
capped in place than if moved to Pond 1. The con:unent further states that
the FS does not support the conclusion that operation and maintenance costs
. .
would be lower if the tailings were moved first to Pond 1, and that the
savings in operation and maintenance costs cannot outweigh the costs of
excavating the tailings and moving them.
Response: First, regarding what appears to be the main comment, EP A
and MDHES believe that the advantages of moving the tailings to an area
that would be protected from floods would be clear. EPA and MDHES
believe that the added protection that would be provided by moving the
tailings to a flood-protected area warrants the statement in the FS that "it
would be ... more effective to move these materials to Pond 1 and cap them
there, where they would be protected from floods by the pond berms."
Regarding the statement that operation and maintenance costs would be
lower if the tailings were moved to Pond 1, EPA and MDHES believe that
the costs of maintaining caps at the operable unit would be lower if the re
were fewer caps to maintain. It was not intended that this savings would
independently offset the costs of moving the tailings to Pond 1. The
justification for moving the tailings to Pond 1 would more likely be based 0 n
the added protection afforded by this option and on the avoided costs of the
additional cap and annoring the cap against floods.
-------
2.
3.
6.
Comment: (This comment is numbered 8 in ARCO's comments, tmt no
Comment 6 or 7 appears.) This comment states that the FS should provide
the basis for the estimate of 325 acres of tailings below Pond 1 and above
Pond 3 that would potentially require capping..
Response: The data and methods used to develop this estimate are
described in Section 2. Briefly, a combination of analytical data and
screening data from XRF studies was used to delineate contaminated areas.
See the response in Section 2.2.7, Comment 2, . and the response to
Section 2.3.1, Comment L
8.3.5 Alternative 5:
1.
Comment: This comment notes that many of the comments made for
Alternatives 3 and 4 also apply to this alternative.
--',
Response: Comment noted.
Comment: This comment requests an explanation of the terminology
"unvegetated material" as used in describing what materials would be capped
under this alternative.
Response: Unvegetated materials would be those soils, sediments, or
tailings which are not revegetated adequately, according to ARAR standards
identified.
Comment: See Comment 18 under Section 7.5.1 and Comment 2 under
Section 7.52 regarding the inefficiency of imposing a 210 cfs treatment
constraint as assumed for this alternative.
-------
"
'\
Response:, The capacity of the new treatment pond is size limited. As
stated on page 7-47 of the FS, the new treatment pond would have a
detention time of just under 2 days with a flow of 600 cfs. This is considered
to be the minimum detention time for adequate settling.
4.
Comment: This comment makes several points regarding the use of
wetlands treatment for the contaminated groundwater. The' primary
comment is that the FS notes several potential problems with the use of
wetlands, but does not document those potential pro~lems with references to
experiences at other sites., Th~ comment also questions whether it would be
, ,
necessary to remove the biomass from the wetlands, and states that the
sediments in the wetland would be no different, in terms of the hazards they
would present, than are the other volumes of sediments (presumably in the
ponds) that would be left in place. The comment suggests three additional
considerations that should be taken into account in considering the use of
wetlands for groundwater treatment, which are addressed below.
Response: EP A and MDHES believe that a wetland treatment system could
be used below Pond 1 for treatment of contaminated groundwater.
However, there are significant problems with this option that have to be
considered, and the FS properly raised these issues.
The general discussion of use of wetlands for treatment of contaminated
water is based on experience with such usage at other sites. The comment is
correct that references should have been given for this discussion. All of the
potential problems noted in the discussion have been observed in previous
uses of wetlands, as can be seen by referring to the following:
-------
Girts, M.A and R.L.P. Kleinrn~ Constructing Wetlands for Treatment of
Mine Waier, presented at the 1986 Society of Mining Engineers Fall
Meeting, St. Louis, MO, September 1986. .
Girts, M.A and Robert Knight, Operations Optimization Draft, CH2M
HILL, 1987. .
Final Technical Report, Tasks 1 and 2, Utilization of Geothermal Effluents
to Create Waterfowl Wetland~, CH2M HILL, 1980.
Chan, E., T.A Bursztynsky, N. Hantzsche, and Y J. Litwin, The Use of
Wetlands for Water Pollution Control, EPA-600/2-82-086, U.S. EPA, 1982.
Girts, M.A and R.LP. Kleinmann, Constructed Wetlands for Treatment of
Acid Mine Drainage: A Preliminary Review, 1988.
Howard, E.A. and T.R. Wildeman, Conceptual Design and Preliminary Cost
Estimates for the Passive Treatment of Drainage from the National and
Quartz Hills Tunnels, Blackhawk and Central City, Colorado, Camp Dresser
and McKee, 1987.
Erickson, P.M., M.A. Girts, and J. Holbrook, Use of Constructed Wetlands
for Coal Mine Drainage, presented at the National Western Mining
Conference and Exhibition, Denver, Colorado, The Colorado Mining
Association, February 1987.
EPA and MDHES disagree with the unqualified contention that th~
sediments in the wetland would pose no different hazards than the othe r
sediments that would be left in place. The wetland would be in J ;.
unprotected area in a floodplain, unlike the sediments in the ponds, J r \ ~
-------
would be subject to release and dispersal during floods.. The wetland could
be provided with some flood protectio~ at a considerable expense. The
wetland concept developed in the FS does not include flood protection for
the wetland, but such protection could be incorporated.
The comment specifically states that the wetlands option should be
reconsidered with four points in mind:
A
The existing wildlife ponds have operated for. more than 15 years and
required little startup effort.
Response: The wildlife ponds are not wetlands designed for treating
contaminated groundwater and are not monitored for treatment
effectiveness. It is not known how long it took before the Wildlife
Ponds began to yield the incidental treatment they now provide.
Therefore, it is difficult to be certain of the intended point of the
comment. Based on experience that has been seen at other sites
where wetlands have been established for treatment of contaminated
water, there is a potential for difficulties and delays while getting the
wetland operating so that it can receive extracted groundwater. The
FS properly pointed out this source of potential difficulty.
B.
The biologic treatment [that would be provided by] wetlands is
already proven by the operation of the existing ponds (Ponds 2 and
3).
Response: The functional (settling and treatment) portions of the
ponds are not wetlands. Some of the biologic processes that would
operate in a wetland are the same as those that likely operate in the
ponds. Other processes would occur as well. A straight compariso i1
-------
D.
of the wetland concept considered in the FS and the existing pond
system is not meaningful.
c.
Wetlands expansion is a significant national priority.
Response: Wetlands retention (or "no net loss") is a significant
national priority. However, the wetlands expansion that is desired is "
somewhat different from the new wetland that would be constructed
below Pond 1. The m~jor difference is that the treatment wetland"
would be fed with contaminated water, and in time would be
contaminated with toxic metals in the sediments and substrate, where
much of the desirable biologic activity in "a wetland takes place. "
The wetlands option would avoid the risk of, or resources and
expenses to mitigate, releases of contaminants associated with the
excavation and hauling required in the removal option.
Response: The wetlands option would not entirely avoid the need to
move the tailings. Whether or not the tailings and contaminated soils
were left in the area below Pond 1, considerable excavation, hauling,
and grading of the tailings and soils in that area would be required to
prepare it for construction of a wetland. Thus, the potential releases
ARCO is concerned about would exist either way. Additionally, it is
not clear why the risks in moving these tailings should be any greater
than the risks involved in moving the tailings in the bypass, which
ARCO is undertaking during the summer of 1990. The question of
whether or not the tailings should be removed is driven instead by the
desire to remove these materials from the floodplain where it would
be expensive to adequately protect them from floods, or where they
might hamper efforts to establish an effective wetland.
-------
-
5.
Comment: This comment refers the reader to a previous comment
(Section 7.2.2.2, Comment 1) regarding the potential for increased
groundwater contamination below Pond 1 due to wetlands remediation.
Response: The FS did not conclude that flooding will result in "increased"
groundwater contamination. The FS stated that the continued presence of
water in Pond 1 provides a source of recharge and contaminants to the
shallow aquifer below Pond 1. The analyses supporting .this conclusion are
summarized in the FS.
6.
Comment: This comment asks why Alternative 5 includes removal of the
tailings below Pond 1 prior to construction of the wetland. ARCa states
that the tailings would be in a reducing environment and would effectively
be no different from the sediments left in place in other alternatives. If
initial dissolution of metal salts from the oxidized metals is an expected
problem, the comment suggests that a method to pump the wetlands water
to the ponds prior to release could be used in the startup period.
Response: Although it might be possible to construct and operate a wetland
on top of the tailings deposits, there are several reasons that it might be
more advantageous to remove the tailings first. Because it seemed more
likely, for the reasons given below, that the tailings would be removed, the
FS assumed that they would be and incorporated this step in the conceptual
design and the cost estimate.
Four reasons that it would be advantageous to remove the tailings are:
A
The tailings are in a floodplain. While it would be possible, at least
in theory, to protect the wetland from flood damage, it seemed more
. reasonable, from an engineering value perspective, to assume that am
-------
flood damage would be repaired and not. attempt instead an
engineered solution to avoid such damage. That they are in a
floodplain is, as pointed out above, a significant difference betWeen
the sediments that would be in the wetland and the sediments in the
ponds. The sediments in the ponds are to be protected against the
. 0.5 PMF. The sediments in the wetland under Alternative 5 would
not have such protection.
.B.
The tailings and other. soils in the area below Pond 1 would require.
considerable regr.ading to allow construction of a wetland.'- The
handling of the material ~o complete this regrading would present the
same types of potentials for releases, which ARCO expresses concern
about, as would excavation and hauling to Pond 1. Thus, leaving the
tailings. in place does not eliminate the potential for releases
associated with excavation and moving of the tailings, as stated by the
comment. If some of the tailings have to be disturbed, it seems that
moving them out of the floodplain at the same time would be
worthwhile.
C.
It is not. clear that establishing the wetland over large deposits of
tailings would be acceptable environmentally. ARCO mentions the
problem of initial dissolution of metal salts from the tailings as a
potential problem. In addition to this problem, the existence of these
large deposits of tailings in the wetland would have impacts on the
types and quantities of species that could be established in t r. e
wetland. Also, given the large quantities of metal ions that would be
in the wetland environment from the existence of the tailings, it,
reasonable to assume that the potential effectiveness of the wetb:' :
system in removing low concentrations of dissolved metals from:'
influent groundwater would be reduced. Soluble metals from'.
-------
/
~
8.
tailings would load up the substrate. In short, the tailings represent a
large source of metal contaminants that would no doubt have some
negative impacts on the possibility of establishIng a successful
wetland. Wetland treatment technology is, as mentioned in the FS, a
relatively new technology. . It seemed prudent during the FS to
assume that the wetland would be given the best chance of success by
removing the tailings prior to construction.
D.
EP A and MDHES believe that the poten~ial. exists for increased
groundwater contamination if a wetland were constructed on top of
. .
the tailings deposits.. Studies during the design phase may establish
that the impacts would be small enough riot to represent a significant
threat to the environment; but, that conclusion would have to await
further studies during the design phase.
7.
Coniment: This comment points out that a negative impact of diverting the
flows of Mill and Willow Creeks into the pond system is mentioned in the
FS only under Alternative 5, but should also have been mentioned for
Alternatives 1 through 4.
Response: The comment is correct in part. The negative impact would exist
for all alternatives that would divert Mill and Willow Creeks into the pond
system. Given the format of this chapter of the FS, it should have been
mentioned just once, under Alternative 1 (instead of under Alternative 5),
since the impact would not differ among the alternatives.
Comment: This comment repeats an earlier request for references to the
information used in developing the discussion of potential problems in
establishing and maintaining wetlands. It also states that the magnitude of
-------
3.
such disadvantages should be documented so that the advantages can be
weighed against the disadvantages. ~. :
Response: References are given above (Section 8.3.5, Comment 4) that de-
scribe the current state of understanding of wetlands treatment systems. The
references provided describe the problems encountered in other uses of wet-
lands for treatment of contaminated water.
8.3.6 Alternative 6:
Environment.
S{gnifjcant protection o~h. Welfare. and-1M
1.
Comment: This comment refers to previous comments on the use of institu-
tional controls in constructing remedial alternatives.
Response:. See the responses to the comments on the use of institutional,
controls in Chapter 3 and 6.
2.
Comment: This comment repeats earlier comments regarding the adoption
of ARCO's plan for modifying the existing pond system for use as a settling
system during major floods.
Response: EP A and MDHES have decided to include storage and
treatment of the tOO-year event into the Recommended Alternative in the
ROD. See the response to the Section 7.5 comment.
Comment: This comment refers the reader to a previous comment
(Section 7.2.2.2, Comment 1) regarding the potential for increased
groundwater contamination below Pond 1 due to wetlands remediation.
-------
~
Response: The FS did not conclude that flooding will result in "in~reased
groundwater contamination. The FS stated that the continued presence of
water in Pond 1 provides a source of recharge and. contaminants to the
shallow aquifer below Pond 1. The analyses supporting this conclusion are
summarized in the FS. See also the response to Section 7.2.2.2, Comment 2.
4.
Comment: This comment states that the FS is misl~ading in noting that
"Poor control of water level [in areas wet closed by shallow flooding] could
lead to reexposure of the material." The comment .states that control of the
. .
water level would be no. more diffiCult than other actions proposed in the
alternatives.
Response: EP A and MDHES believe that the statement in the FS is
accurate. It is a proper function of the FS to point out operation and
maintenance requirements for the alternatives. Maintaining a shallow level
of water over wide expanses does require some special care to see that the
overflows remain open and operate properly. .
5.
Comment: This comment states that conversion of an area that is currently
mixed wetland and terrestrial habitat to all wetland should be characterized
as an environmental benefit rather than as an impact.
Response: EPA and MDHES disagree with the comment. Both wetlands
and terrestrial habitat have advantages to wildlife. Loss of terrestrial habitat
is an impact that should be noted in the FS.
8.3.7 Alternative 7: No-Action Alternative
1.
Comment: See Comment 3 in Section 8.3.1 regarding the incorrect
implication in the ~S that all groundwater MCLs are currently exceeded.
-------
Response: The- point of compliance is established in- the ROD. Certain
MCLs are exceeded.
-'~
2.
Comment: This comment repeats earlier comments regarding whether or
not the RCRA regulations can be considered as ARARs for this operable
uni t.
Response: These comments are addressed in the response to comments,
Chapter 3.0 and Appendix B. .
8.4
COMPARATIVE ANALYSIS OF ALTERNATIVES
1.
Comment: The comment - asks again why the FS concluded that excavating
materials from the floodplain below Pond 1 and taking them to Pond 1 to be
covered by the cap that would be placed over Pond 1 would be more
protecti ve than capping them in place.
Response: The primary advantage would be that the capped materials
would no longer be in a floodplain. Because susceptibility to flood erosion
would be eliminated (assuming that the Pond 1 berms are raised and
armored as necessary to provide protection from major floods), this
approach would be more protective of the environment than leaving the
materials in the floodplain. Additional advantages of moving the materials
to Pond 1 prior to capping them include avoiding the cost of an additional
cap, avoiding the cost of protecting a cap below Pond 1 from flood damage.
and lower maintenance - costs, since there would be less capped area at the
site to maintain. See also response to Section 8.3.4, Comments 4 and 5.
2.
Comment: This comment states that Alternatives 3, 4, 5, and 6 are "more or
.
less equivalent" in effectiveness in treating the flows that would go throu ~:
-------
'.
\.;
the pond system, and states that the differences would be in the volumes
that could be routed through the ponds, which affects the frequency of
events in which flows would have to be routed around the ponds.
Response: The comment is correct in part. The concept common to the
alternatives developed in the FS is that influent flows to the ponds would be
limited to what the ponds could treat adequately. But, the concept of
"effectiveness" includes not only the degree of treatment but also the range
of flows that can be treated. In this matter, the alternatives differ. The FS
properly points out these differences ineffectiveness.
3.
Comment: This comment repeats earlier requests for references on previous
experiences with establishing and using wetlands to treat contaminated
water. This comment particularly requests references to support the
notation in the FS that up to 5 years may be required to establish good
operatirm of the wetland.
Response: The requested references are provided in the response to Sec-
tion 8.3.5, Comment 4.
4.
Comment: This comment requests further basis for the annual operation
and maintenance cost estimates in the FS, and asks why the operation and
maintenance costs for the alternatives are so similar.
Response: The assumptions used in developing the costs estimates are
provided in Appendix D of the FS. The primary reason that the costs are so
similar is that the alternatives themselves are actually quite similar in those
aspects that most determine operation and maintenance costs. All of th~
alternatives would involve the continuing existence of miles of berms that
would bave to be maintained. All of tbe alternatives would include .:
-------
8.4.1
2.
surface water treatment system for the flows in Silver Bow Creek, which I
would require approximately equal levels of operation and maintenance
costs for all of the alternatives. And nearly all of the alternatives would
involve areas.of contamination that are closed. either by flooding, capping, or
revegetation; any of these actions would require certain low levels of
operation and maintenance. . Thus, the primary determinants of operation
and maintenance costs are very similar between the alternatives, and the
estimates of operation and maintenance costs are similar.
1.
Comment: This comment repeats earlier comments on the use of the
RCRA regulations as ARARs for this operable unit. Specifically, the
comment questions the assumption that the cap used in certain areas of the
site would follow certain of the RCRA requirements for caps.
Response: These comments are addressed in the response to comments at
Section 3.0, and Appendix B..
Comment: This comment repeats an earlier comment in Section 7.7.2.2
(Comment 1) that flooding the area below Pond 1 should not increase
groundwater contamination.
Response: The FS did not conclude that flooding will result in "increased"
groundwater contamination. The FS stated that the continued presence 0 f
water in Pond 1 provides a source of recharge and contaminants to the
.
shallow aquifer below Pond 1. The analyses supporting this conclusion are
summarized in the FS. See also the response to Section 7.2.2.2, Comment :!.
-------
A.
3.
Comment: This comment repeats earlier questions regarding the value of
moving the. tailings below Pond 1 to Pond 1 prior to capping them.
Response: The agencies believe there will be several advantages to moving
the contaminated materials below the pond. See the responses to
Section 8.3.4, Comments 4 and 5.
Comment: This comment questions the conclusion in the FS that water
quality standards would be exceeded with incre3:Sing frequency as the
maximum input rate for. the pond ireatment system decreases between
alternatives, because smaller flood flows would be bypassed without
. treatment.
Response: The alternatives would allow treatment of 600 or 210 cfs in the
pond system. 'The comment points out a difficult issue, one that was
considered during the study. As flow rates increase in the creeks, the
QJ,lantities of both dissolved and nondissolved metals would increase due to
the runoff being contaminated with dissolved metals (see Chapter 4 of the
FS) and due to greater erosion of tailings deposits along the stream banks;
the deposits contain both dissolved and nondissolved metal contaminants.
However, the concentrations of the dissolved and nondissolved metals
contaminants would vary in ways that are difficult to predict. As noted in
the FS, water. quality information collected during high flow events was not
available during the FS.
Some information on the effect of precipitation events on water quality are
available. Three studies have been done: two are complete and were
considered during the study; and the other is in draft and being reviewed by
the regulatory agencies (MultiTech, 1987; CH2M HIL4 1987; CH2M HILL,
1990). All three studies revealed that during runoff precipitation events.
-------
very high concentrations of metals are released to Silver Bow Creek. The I
runoff study done at Ramsey Flats (as part of the treatability study for the
FS) (CH2M HILL, 1987) is described on page 4-28 of the FS. . It indicates
that very high concentrations of metals are released to the river during
runoff events. The Phase I remedial investigation for the Silver Bow Creek
Site (MultiTech, 1987) showed high metals levels in the river during a
rainstorm. Results from the more recent study (CH2M lffiL, 1990) also
show very high metals levels in the river during a high flow event..
These results indicate that the river receives additional metals loads during
precipitation events. What the results do not show is the levels of metals
that would be released by erosion of tailings deposits along the creek during
flow events between 200 cfs and 600 cfs, or the levels of total metals that
. would be in the river at these flows. The reason that such data are
unavailable is because, as noted in the FS, no high flow events that could be
sampled and analyzed have occurred since the beginning of the remedial
investigation. Given the frequency of high flow events of this size range (a
probability of several years between events), it is expectable not to have such
information for the FS.
Given the lack of actual high flow concentrations in the river, the FS took
the. approach of developing alternatives that would allow for treatment of alI'
flows up to certain levels. This approach was taken to allow the
decisionmakers to choose among a range of flows to receive treatment. As
noted in the response to the Section 7.5 Comment, the EPA and MDHES
have ~ecided to utilize Pond 3 for detention and treatment of all flows up to
3,300 cfs.
-------
, '
8.4.2
No specific comments were made by ARCa in this section.
8.4.3 Lon~- Term Effectiveness
1.
Comment: This comment states that the evaluations of residual risks in this
section of the report are qualitative in nature, and that the FS should
quantify such risks where feasible.
Response: The discussions of residual risks are descriptive and comparative,
as required by the RI/FS guidance document. . According to the guidance
document, the purpose of this section of the FS is "to identify the advantages
and disadvantages of each alternative relative to one another so that the key
tradeoffs the decisionmaker must balance can be identified." 'The
presentation of differences among the alternatives can be measured either
qualitatively or quantitatively,..." (emphasis added). A qualitative approach
was adopted because quantification of residual risks would have been
difficult and the results would have been of uncertain accuracy.
2.
Comment: "A reference or other documentation should be provided to
support the contention that the' probability of occurrence' of a probable
maximum flood is only once in several thousand years.' (page 8-108) As
noted in the same sentence of the FS, 'no specific return intervals are
associated with probable maximum floods.' The fact is that estimation of a
PMF-series flood is a wholly detenninistic procedure, and thus a
probabilistic estimate of its occurrence is invalid. It is of some interest to
note that the Institute of Civil Engineers (ICE London, recommends using
either a 0.5 PMF or a 10,OOO-year recurrence flood (whichever is greater) for
'significant'hazard dams (National Research Council, 1985). Without sayin~
-------
that a 0.5 PMF flood has a recurrence interval of 10,000 years, it is apparent
that ICE views these events of at least comparable order of magnitude."
(Emphasis added in the original comment).
Response: The comment has been quoted in its entirety. EP A and
MDHES have examined this comment and can fmd no point of contention.
It appears that ARCa is trying to make the same point made in the FS:
. .
while PMFs have no specific return interval associated with them, they are
not likely, to occur more often than once every several thousand years on a .
probabilistic basis.
3.
Comment: See Comment 6 under Section 8.3.5 regarding our disagreement
that "eventual recontamination of soils, sediments, and groundwater" (page
8-110) would be a significant issue associated with wetlands treatment.
Responst;: The agencies believe there are four reasons why removal of the
tailings would be advantageous. See the response to Section 8.3.5,
Comment 6, for a complete explanation of those reasons.
8.4.4 Re
1.
Comment: The comment states that the FS should provide an analysis of
the advantages of Alternative 3 in reducing the mobility of contaminants in
the tailings and contaminated soils compared to the advantages of AI.
ternatives 4 and 5. The comment states that any advantage in prote~tivene~'
of Alternative 3 would be "minimal in comparison to the total costs."
Response: The FS does compare the relative advantages of t ~~.
Alternative 3 versus Alternatives 4 and 5. The difference in actions I'. .
. these alternatives is that under Alternative 3 the tailings and contarnina t.:
-------
soils would be moved to Pond 1, where feasible, and capped along with the
materials already in Pond 1. Under Alternative 4 or 5, the tailings would be
capped and revegetated in place, where possible.
The primary advantage to ARCa in capping in place is to save the costs of
excavating the tailings and soils and transporting them to Pond 1. The
disadvantages are the costs of additional areas requiring caps and the
additional maintenance costs for the additional caps.
As pointed out in responses to previous comments, the primary (but not the
only) advantage of moving the tailings and contaminated soils below Pond 1
to Pond 1 prior to capping is that they would then be protected from release
due to floods. The primary advantage in moving the soils above Pond 3 to
either a consolidated location within Pond 3 or to Pond 1 is the ability to
construct a well designed cap that would meet the requirements of the
RCRA regulations for hazardous waste landfill caps. Capping in place (for
the tailings and contaminated soils below Pond 1 or the tailings and
contaminated soils above Pond 3) would not allow for carefully graded and
designed caps to be constructed over all of the contaminated areas. Greater
infiltration of rainfall, and/or greater potential for erosion of such caps will
be a necessary consequence of capping in place, unless substantial regrading
of the tailings and contaminated soils is envisioned.
MDHES and EP A have determined that the certain RCRA cap
requirements are relevant and appropriate requirements for this action. A
cap will provide better long-term assurance that the materials will remain
isolated from the environment.
-------
8.4.5 Short-Term Effectiveness
\..
1.
Comment: This comment - repeats earlier -comments regarding
diverting Mill and Willow Creeks into the pond system.
Response: In response to considerable public opposition to routing
the flows of these two creeks into the pond system, the selected
remedy for the operable unit will not include this action. Instead,
ARCa will continue i~ sampling and analysis activities to determine -
. . .
the actual impact - of these two creeks. If the impacts will cause
future exceedances of applicable or relevant and appropriate
requirements, then action to address those exceedances, such as
source control actions, will be required as part of another operable
unit. EP A and MDHES reserve this authority to require diversion of
Mill and/or Willow Creeks into the pond system for treatment in the
interim or long-term if required to protect the environment. See also
the responses to Section 7.5.1, Comments 3 and 4.
1.
2.
Comment: The comment states that technical feasibility (emphasis in
the original) is not dependent on the magnitude of the project. The
commenter notes that, constructing a 50-foot-high embankment is just
as feasible as constructing a 30-foot-high embankment.
Response: EP A and MDHES agree with the comment.
Comment: This comment agrees with a statement in the FS that
alternatives requiring less riprap are more favorable because of the
apparent scarcity of riprap in the area. -
-------
8.4.7 ~
Response: EP A and MDHES agree. However, the difficulty in
obtaining riprap does not affect the technical feasibility of alternatives
that require greater amounts of riprap.
1.
Comment: This comment refers the reader to other comments on
Appendix D of the FS.
Response: See the response to the comments in Appendix D. ..
2.
Comment:
The comment states that the unknown found.ation
conditions for the upstream flood impoundment or settling basin
should be mentioned in the sensitivity analysis.
Response: It is true that site-specific studies of the area considered
for the impoundment or settling basin have not been done. Such
studies are not appropriate at the feasibility study stage of the
Superfund process. However, the fact that site-specific studies have
not been done does not mean that nothing is known about the
foundation conditions in that area.
The regional geology is fairly well known, and the probable
subsurface conditions can be inferred from the known regional
geology. It was assumed in. the FS, based on the regional geology and
field observations, that suitable foundation conditions would be
found.
Further, the type of structure considered--earthen berms with 3: 1
slopes--do not ordinarily require special foundation conditions. n e
-------
gravels that likely underlie the surface would probably provide
excellent foundation conditions for that type of structure. The '.
existing pond berms have been in existence for many years on
foundations that are likely to be similar to those found at the
. .
proposed location. The existing pond berms are not as well designed
as the new berms for an upstream flood impoundment or settling
basin would have been.
-------
CHAPTER 9.0
REFERENCES
ARCOt 1989. "ARCO Plan Alternative ,3A, FeaSibility Level Design for Remediation of
the Warm Springs Ponds, Operable Unit," 2 Volumes, prepared by ESA Consultantst
Ft. Collins, Colorado, November 1989..
CH2M HIL~ 1980. Final Technical Report, Tasks I, and 2, U,tilization of Geothermal
Effluents to Create Waterfowl Wetlands.
CH2M HIL~ 1987. "Data Summary Report Supplemental Remedial Investigation; Silver
Bow Creek Sitet Butte, Montana." EPA Work Assignment No. 26-8122.0.
CH2M HIL~ 1989a. "Phase II Remedial Investigation Data Summary Report." Prepared
, .
for Montana DepartMent of Health and Environmental Sciences by CH2M HI~ Helenat
Montanat May 1989.
CH2M HILL, 1989b. "Streamside Tailings and Revegetation Studies, ST~S Phase I,"
Appendix B, CH2M HILL, Boise, Idaho, October 1989.
CH2M HILL, 1989c. "Silver Bow Creek Flood Modeling Study." Prepared for Montana
Department of Health and Environmental Sciences, by CH2M HILL, Helena, Montana.
November 1989.
CH2M HILL, 1990. "Draft Final--Silver Bow Creek CERCLA Phase II Remedial
Investigation Data Summary; Area I Operable Unit. Volumes I-D." Prepared for
Montana Department of Health and Environmental Sciences, Helena. Docume n ~
-------
Chan, E., T.A Bursztynsky, N. Hantzsche, and Y.J. Litwin, 1982. The 'Use of Wetlands for
Water Pollution Control~ EPA-600j2-82-086, U.S. EPA
Chen-Northern, 1989. "Technical Memorandum--Hydrologic Analyses of a Groundwater
. .
Interception Trench for Dry-Closure of Pond 1, Warm Springs Ponds Operable Unit, Silver
Bow Creek CERCLA site," prepared by Chen-Northern, Inc., Helena, Montana, April
1989.
EPA, 1988. "Guidance for Conducting R~medial Investigations and Feasibility Studies.
. .
Under CERCLA, Interim Final," U.S. E.P.A., Washington, D.C., October 1988.
Erickson, P.M., M.A Girts, and J. Holbrook 1987. Use of Constructed Wetlands for Coal
Mine Drainage, presented at the National Western Mining Conference and Exhibition,
Denver, Colorado, The Colorado Mining Association, February 1987.
S
. Flood
Studies and Conceptual Desi~ Modifications.
Company, Denver, Colorado.
Prepared for the Anaconda Minerals
Girts, M.A. and R.L.P. Kleinmann, 1986. Constructing Wetlands for Treatment of Mine
Water, presenteQ at the 1986 Society of Mining Engineers Fall Meeting, St. Louis, MO,
September 1986.
Girts, M.A and R.L.P. Kleinmann, 1988. Constructed Wetlands for Treatment of Acid
Mine Drainage: A Preliminary Review.'
Girts, M.A and Robert Knight, Operations Optimization Draft, 1987. CH2M HILL.
Howard, E.A. and T.R. Wildeman, 1987. Conceptual Design and Preliminary Co~:
Estimates for the Passive Treatment of Drainage from the National and Quartz Hill,
-------
IECO, 1981. "Geotechnical and Hydrologic ~tudies, Warm Springs Tailings Ponds,
Anaconda, Montana." Report prepared for AnaConda Copper Company by International
Engineering Company, Denver, Colorado, Mar~ 1981.
MultiTech, 1987. "Silver Bow Creek Remedial :Investigation Final Report. Appendi-
ces A-F." Submitted to Montana Department nr Health and Environmental Sciences,
-------
APPENDIX B
RESPONSES TO ARCO COMMENTS APPENDIX B,
DETERMINATIONOFARARs
1. Comment: ARCO has outlined a variety of sources and methods for analyzing ARARS,
and has given definitions of key terms used in ARAR analysis. '
Response: Attachment 1 to the Record of Decision contains the final list of applicable or
relevant and appropriate cleanup standards, standards of control, and other substantive
requirements, criteria, or limitations (ARARs) for the Record of Pecision, as well as a list
of documents or other sources of information which are To Be Considered during the'
remedy selection or during implementation of the remedy. EP A has identified the list based
upon the statutory provisions addressing ARARs found in CERClA, particularly section
121(d) of CERClA, 42 V.S.C. sS~ 9621(d); the new National Contingency Plan, 40 CFR Part
300 (1990); the preambles to the proposed NCP and the final NCP, 53 Fed. Reg. 51394.e.1
~ (December 21, 1988) and 55 Fed. Reg. 8666 et seQ. (March 8, 1990) respectively; EPA
guidance documents regarding ARARs entitled "Compliance With Other Laws Manual:
Parts 1 and 2" (OSWER Dir # 9234.1-01 and 92341.02 respectively).
Definitions for "applicable" and "relevant and appropriate" are now promulgated in the new
NCP, 40 CFR sS~ 300.5. Those definitions are used by EPA in its identification of final
ARARs. Discussions of "substantive", "administrative", and "promulgated" occurs in the
preambles to the proposed arid new NCP, and in the ARAR guidance. All of those
discussion were used by EP A in its identification of the final ARARs.
2. Comment: Comments found on pages B-2 - B-9 of ARCO's comments, addressing
general ARAR issues are responded to in the responsiveness Summary, Chapter 3.
Comments on Action Specific ARARs
A Surface Water
3. Comment: ARCO disagrees with the identification of the State's nondegradation statute
and regulations, MCA s'l 75-5-303 and ARM sll 16.20.702, because it is a general policy.
and because it applies only to new sources and is not relevant, and appropriate.
Response: Statutes and regulations that contain general goals can be ARARs, if they are
directive in intent and promulgated. Non-degradation statutes such as this one are
specifically cited by EP A as probable ARARs, even though they may be characterized a.s
-------
EPA agrees that the nondegradation statute is not applicable to the site, because the. statute
addresses new sources of water pollution. However, the statute is intended to regulate,
pollution to the same medium as this cleanup (water), addresses the same substances and
activities, places, and structures (pollution from point sources), and is generally designed to
protect human health and the environment from unnecess'ary pollution. It is therefore
relevant and appropriate for the site and this cleanup action.
EP A notes that compliance with tbe specific numeric standards established for tbe point
source discharge and ambient water at tbe site ~l achieve compliance with this .ARAR.
" .
. .
4. Comment: Water quality standards and federal water quality criteria, or numeric
standards from the Safe Drinking Water Act and the Montana Public Water Supply Act for.
human health should not be identified. for "the surface water ARARs, because the Clark
Fork River is not designated as a drinking water source. Instead, only standards and criteria.
for protection of the aquatic environment, or protection for humans who eat fish should be
~d .
Response: Tbe State of Montana has adopted.all Gold Book Criteria, including criteria
protecting public health, as applicable water quality standards for the Clark Fork River, Mill
Creek, and Willow Creek, among other surface water bodies within the State. Tbe adoption
of those standards is not dependent on the designated use of the particular stream.
Therefore, the Gold Book Criteria values for protection of human health through ingestion)
are the applicable standards for tbe site surface water outside of the Ponds, and for the"
discharge from the Ponds into those surface waters.
ARCO's citations to EP A guidance and preamble language discussion of this issue are
addressed to determining when to use federal water quality criteria as relevant and
appropriate requirements. As the same preamble discussions and the guidance make clear,
when state water quality standards exist for a given site, and they are equivalent to or more
stringent than the federal water quality criteria, those applicable standards should be
identified. 55 FR 8754 - 8755.
Additionally, both Mill and Willow Creeks, which will run through the Bypass, are
designated as potential drinking water sources. Therefore, identification of the drinkin~
water standards from the Gold Book Criteria for this action is appropriate.
MCLs for drinking water are not as stringent as the State's standards. Therefore, t~.:
identification of those standards here is a moot issue. EPA and MDHES reserve the rigr.:
to identify these standards for surface water bodies in the Clark Fork Basin, as appropriate:
As a practical matter, use of these contested standards does little to change the fi n..
selection of numeric standards" for the discharge or for the surface water. compliance po i i. :
Only arsenic and mercury standards for drinking water are below the standards for aqu~: .
. .or fish consumption. Both of those standards are waived as for this action, and hig~~"
-------
~
Pirkle, J.L, J. Schwartz, J.R Landix, and W.R. Harlan. 1985. The Relationship
Between Blood Lead Levels and Blood Pressure and Its Cardiovascular Risk Implica-
tions. Am J. Epidemiol. 121:246-258. .
Porter, J.W. 1989. Memorandum: Interim Final Guidance for Soil Ingestion Rates.
OSWER Directive 9850.4, Office of Solid Waste and Emergency Response, Washing-
ton, D.C.
PuIs, R.W. and MJ. Barcelona. 1989. Filtration of Groundwater Samples for Metals
Analysis. Hazardous Waste and Hazardous Materials 6(4):385.
Stark, A.D., R.F. Quah, J.W. Meigs, and E.R. DeLouise. 1982. The Relationship of
Environmental Lead to Blood-lead Levels in Children. Environ. Res. 27: 372-383. .
. .
Thomas, LM. (Administrator, US EP A). 1988. Memorandum to Assistant Adminis-
trators US EP A, re: Recommended Agency Policy on the Risk of Carcinogenicity
Associated with the Ingestion of Inorganic Arsenic. Office of the Administrator,
Washington, D.C. .
US DHRS. 1985. 'Preventing Lead Poisoning in Young Children. CDC 99-2230, U.S.
Department of Health and Human Services, Centers for Disease Control, Center for
Environmental Health, Atlanta, GA.
US EP A. 1984a. Rapid Assessment of Exposure to Particulate Emissions from Sur-
face Contamination Sites. Office of Health and Environmental Assessment.
Washington, D.C.
US EP A 1984b. Health Assessment Document for Inorganic Arsenic. EP A
600/8-83-021F, Office of Health and Environmental Assessment, Washington, D.C.
US EPA 1986a. Superfund Public Health Evaluation Manual. EPA 540/1-86/060
(OSWER Directive 9285.4-1), Office of Emergency and Remedial Response, Wash-
ington, D.C.
US EPA 1986b. Air Quality Criteria for Lead, and addendum. EPA 600/8-83-018F.
Office of Research and Development, Research Triangle Park, N.C.
US EPA 1987. Revisions to the National Ambient Air Quality Standards for Partic-
ulate Matter, Final Rule. Federal Register 52(126):24634.
US EPA 1988. Special Report on Ingested Inorganic Arsenic, Skin Cancer: Nutri-
tional Essentiality. EPA 625/3-87/013, Risk Assessment Forum, Washington, D.C.
US EPA 1989a. Exposure Factors Handbook. EPA 600/8-89/043, Office of Health
-------
US EPA 1989b. Risk Assessment Guidance, Vol.!, Hu'man Health.
540/1-89/002, Office of Emergency and Remedial Response, Washingto~ D.C.
US EPA 1989c. Review of the National Ambient Air Quality Standard for Lead:
. Exposure Analysis Methodology and Validation.' QAQPA Staff Report,
EP A-450/2-89-011, Office of Air Quality Planning and Standards, Research Triangle
. Park, N.C.
EPA/.
US Fish and Wildlife Service. 1988. Arsenic Hazards to Fish, Wildlife, and Inverte-
brates: A synoptic review. Biological Report 85(1.12), Contaminant Hazard Reviews,
Report No. 12, Patuxent Wildlife Research Center, Laurel, MD. .
Vahter, M. and E. Marafante. 1989.. In vivo Methylation and Detoxification of.
Arsenic. PJ. Craig and F. Glocking (eds) in: The Biological Alkylation of Heavy
Elements. . .
Wong, 0., M.D. Whorto~R. Lowenhart, D. Ragland, and S. Ardalan. 1988.. Skin
Cancer Incidence in Four Montana Counties 1980-1986. Prepared by Environmental
Health Associates.
-------
APPENDIX A
RESPONSES TO ARCO COMMENTS, APPENDIX A PUBLIC HEALTH
AND ENVIRONMENTAL ASSESSMENT
1.0 INTRODUCTION
No comments.
2.0 SITE Sl!aTlNG
2~1. PROJECT BOUNDARIES AND LOCATION
No comments.
2.2 . CLIMATE AND METEOROLOGICAL CONDmONS
Comment: This comment questions the exclusion of summer rainfall days in the
exposure assessment.
Response: See the response to Chapter 3, Section 3.2.2, Comment 1.
Summer rainfall days have been included and the risks recalculated.
2.3
SITE FEATURES
No comments.
2.4
LAND USE AND. THE HUMAN POPUlATION
1--3. Comment: These comments discuss the potential for residential development at
the ponds in light of demographic information.
Response: This issue has been discussed. See the response to Chapter 3,
Section 3.2.4, Comment 4.. MDHES agrees with ARCO that future residential
development on the ponds is not likely to occur if appropriate institutional
controls are enacted. MDHES and EPA have made further risk management
decisions based on occupational use of the site which does occur and will
-------
2.5
1.
2.
3
ENVIRONMENTAL SETTING
Comment: This comment states that algae at locations other than the ponds
bioconcentrate metals and other elements.
Response: The statements made in the assessment about the ability of algae to
bioconcentrate metals were based on data presented in the literature from other
locations. MDHES recognizes that this is a general phenomenon and not site
speCific. The assessment does not indicate this phenomenon is uniqu~ to the
ponds.
3.0 DATA EVALUATION AND LIMITATIONS
Comment: This comment questions the use of the term "upgradient'" in
describing the location of wells upgradient of the ponds and states the risk
assessment should define the groundwater flow "upgradient" more completely.
Response: The assessment does state that wells installed and sampled are
"upgradient, adjacent to, and downgradient of the pond system." Again on page
3-2 of the assessment when upgradient wells are discussed, they are referred to
as "upgradient of the pond system." Moreover, data tables presented in Chapter
3 of the risk assessment indicate groundwater in wells upgradient of the pond
system are of better quality than that below Pond 1. The term "upgradient" is
always used in conjunction with "the pond system" and was never used to
implicate "background."
Comment: This comment states that arrows indicating groundwater flow
direction in the risk assessment are incorrect.
Response: The arrows presented on Figure A3-2 in the risk assessment indicate
a general flow direction in the shallow aquifer. One arrow inadvertently
extended beyond the bypass and MDHES acknowledges this error. Flow in the
shallow aquifer at the ponds is toward the bypass, which then flows north.
Comment: This comment requests more information be included in the risk
assessment on hydraulic conductivities and vertical gradients.
Response: As previously stated, the risk assessment uses the information and
data presented in the RI and portions of the FS. The purpose of the risk
assessment is not to repeat technical analysis nor to reiterate technical details
that have been presented in other sections of the FS or in previous RI reports.
Refer to appropriate sections of the RI or FS for detailed information on
-------
7.
4.
Comment: This comment states that the risk assessment should present
QA/QCinformation of data obtained during the Phase I and Phase II RI.
,Response: The risk assessment was not tasked with QA/QC of data obtained
during the RI. Data were validated based on the approved QA/QC plan. For
a discussion of the limitations of the data, the reader is referred to the RI. See
also the response to Appendix ~ Section 3.0, Comment 1.
5.
Comment: This comment seeks an explanation for exceedance of dissolved
fraction over total fractions.
Response: The data passed discussed in the comment were subjected to and
passed the data validation required by the approved QA/QC plan. In the data
used in the assessment, the. parameter' which exhibited this phenomenon was .
zinc, at two sampling locations. On Table 3-11 of the risk assessment (Appen-
dix A of the FS) in the all-flow maximum concentration column, the dissolved
concentration slightly exceeded the total concentration (570 ppm versus 550
ppm), although these two numbers are probably analytically the same.' On
Table 3-14 of the risk assessment, dissolved zinc concentrations exceed the total
zinc concentrations for both maximum and average concentrations in the low
flow samples. The limited number of occurrences of this phenomena did not
warrant an extensive search for the cause. A potential explanation for an
overestimate of the dissolved fraction may be the passage of colloidal material
less than 0.45 Jj m in size, as metals may be transported on mobile colloids (PuIs
and Barcelona, 1989).
6.
Comment: This comment requests a reference for sediment data interpreta-
tions.
Response: Sediment contamination is discussed in the Phase I Remedial
Investigation Report by Multitech (1987), the Silver Bow Creek CERCLA, Data
Summary by CH2M HILL (1988), and the Phase II Remedial Investigation Data
Summary by CH2M HILL (1989).
Comment: This comment requests a reference for the particle sizes given in the
assessment for entrainable particles and respirable particles.
Response: Entrainable particles as referred to in the assessment, are those 75
Jjm or less in size, as stated in EPA (1984a). These are particles that move by
suspension and tend to follow air motions. This size, 75 Jjffi, is the upper size
limit of silt particles that can become suspended and the smallest particle size
for which size analysis by dry sieving is practical.
Respirable fraction is defined in the assessment as 9 Jj m or less in s i z c= .
-------
4.1
1.
2.
(EPA 1987, EPA 1984a), 9 J.'m was used in the assessment as it was the.closest
size fraction to 10 J.'rn reported by the laboratory.
8.
Comment: This comment requests a table comparing the concentrations of
contaminants in "metallic salts" with surface sediment samples of the Mill--
Willow Bypass.
Response: Table 4-4 of the FS presents a summary of the. Mill-Willow Bypass
sediment data by material type. This table indicates concentrations of . zinc and
copper contaminants in metallic salts are an order of magnitude greater than.
typical sediment materials.
9.
Comment: This comment reques~ a discussion of the projected total extent and.
volume of contaminated soils along' the Mill-Willow Bypass.
Response: The risk assessment is not concerned with estimates of the volume
of contaminated soils along the Mill-Willow Bypass. The assessment diet not
estimate the volume nor does it mention it. The pages the comment indicates
refer to what is and is not surficial contamination, not what volume of tailings
are present. Volume estimates are appropriate RI/FS activities.
10.
Comment: This comment states that bioconcentration of metals by fish is not
unique to the Warm Springs Ponds. Comment continues with a statement that,
liver tissue is often higher in metals concentrations than muscle tissue.
Response: Refer to the response to Appendix A, Section 2.5. MDHES
recognizes this is a general phenomenon and not a site specific occurrence. The
assessment makes a statement that higher concentrations of metals are found in
liver tissue than in muscle tissue. It is not identified as a problem.
4.0 -EXPOSURE ASSESSMENT
CONTAMINANT RELEASE MECHANISMS
Comment: This comment requests that metal deposition rates be given for the
ponds, bypass, the Clark Fork River, and groundwater.
Response: The source-receptor relationships described in Chapter 4 of the risk
assessment were derived from data and discussions found in Section 4 of the FS.
Repeating numerical deposition rates in the risk assessment would not provide
a better understanding of the contaminant release mechanisms. Refer to
Section 4 of the FS for numerical values.
Comment: This comment requests probabilities be determined for contaminant
-------
4.2
1:
2.
3.
distribution of contaminants from those mechanisms be determined. CQmment
also questions which flood event is referred to in the risk assessment.
Response: Defming the probability of a release and the aerial deposition of the
resulting release was beyond the scope of the FS~ Refer to Section 4 of the FS
for a discussion of release scenarios.
The risk assessment refers to the 10o-year flood event.
MIGRATION AND FATE
Comment: This comment requests data on the range of metal concentrations
measured in pond outflow.
. .
Response: Tables with surface water data are provided in Chapter 3 of the risk
assessment and in Section 2 of the FS. .
Comment: This comment requests data on metals speciation.
Response: Twelve groundwater samples during the Phase n RI were analyzed
for arsenic In, arsenic V, and total dissolved arsenic (CH2M HILL 1989).
Speciation data were not definitive and did not provide a better understanding
of fate and transport of constituents at the ponds.'
Comment: This comment states that hypothetical fate and transfer mechanisms
should be so stated.
Response: The fate and transfer mechanisms discussed in the ri~k assessment
are termed "potential" mechanisms. The potential for transfer of contamin~nts
between media does not comment directly on the risks, but contributes to the
overall assessment. Risks are discussed in Appendix A, Section 6 of the FS.
4.3 HUMAN EXPOSURE ASSESSMENT
4.3.1 Potential E~osure ~
1.
Comment: This comment questions the inclusion in the assessment of exposure
to constituents in fish and 'waterfowl tissue.
Response: Refer to the response to Chapter 3, Section 3.2.4, Comment 2. The
. assessment is analyzing risk from constituents at the site, in what ever medi3
they are detected and at whatever the concentration. The determination of risk
through exposure to multiple media would not be complete without inclusion of
all media potentially affected by contaminants where exposure could occu r.
-------
2.
2.
Comment: This comment questions the inclusion of fish and waterfowl in the
exposure assessment It reiterates the statement previously made on multiple
pathway exposures to a single receptor and requests a reorganization of the
section dealing with exposure to fish and. waterfowl tissue.' Comment states that
the conclusions should be. presented before the supporting information.
Response: As stated above, the risk assessment is not looking at incremental
risk over background. Whether or not contaminant concentrations in media at
the site are equal to concentrations elsewhere is irrelevant to whether or not
there is ~ health risk from all exposure pathways of a developed scenario that is
known to occur at the site. EP A guidance stresses the importance of. analyzing
multiple exposure pathways (EPA 1989B). It is conceivable that an individual
could be exposed to each pathway within a developed scenario. ' ,
Since the analyses for other media were presented prior to the conclusions
about the medi~ altering the format for the section dealing with fish and
waterfowl would be inappropriate. The risk assessment does state (Section 6)
that risk through fish ingestion could be between zero and 6 x 1~ since it is a
, catch-and-release area, and consumption of the fish may not occur.
4.3.3 Exposure Scenarios
1.
Comment:. This comment requests that exposure scenarios reflect "actual data
regarding activity patterns and characteristics of potentially exposed individuals",
al1d include a quantitative estimation of probability of exposure scenarios being
realized.
Response: Actual data on time and activity patterns for users of the or resi-
dents of the surrounding communities does not exist. Obtaining such data was
out of the scope of the risk assessment. The risk assessment states "no data
exist to determine the true frequency and duration of exposure. Therefore,
assumptions have been made for these variables of exposure for each scenario
based on conversations with persons who live in the Warm Springs area and are
familiar with activities occurring at the ponds." Exposure scenarios were based
on discussions with residents of the State of Montana and best professional
judgement.
Sufficient data to quantitatively estimate the probability of events occurring was
not available. for the risk assessment.
Comment: This. comment questions the assumptions of the recreational
scenario.
Response: See response to Chapter 3, Section 3.2.4, Comment 7. Assumption.,
used in the assessment were based on discussions with Montana State Fish and
Parks personnel, Montana residents and best professional judgement becau,"'
-------
3.
Comment: This comment questions the exposure duratio.n times used. in the
assessment.. .
Response: At the time the risk. assessment was conducted, the guidance from
EP A was to use a 70-year residential duration. Although more data are
becoming available with which to refine exposure assumptions based on actual
observations, current guidance (EP A, 1989b) indicates that 9 years and 30 years
are average and upper-bound residence times. Longer residence times may
more accurately reflect typical conditions in a rural environment.
4.,
Comment: This comment states that a 75-year life expectancy should be used in
the risk assessment.
. .
Response: The risk assessment followed EP A guidance in use at the time the
assessment was prepared which recommended a 70-year life span (EPA, 1986a).
Current guidance suggests that either a 70-year or a 75-year life span can be
used. A 70-year life span can be used by convention (EP A, 1989b) or a 75-year
life span can be used as it is more representative of the present day conditions
(EP A, 1989a). Revised risk estimates were based on a 75 year life span.
4.3.4 Estimation of Human Intake
2.
3.
1.
Comment: This comment questions the soil intake values used in the risk
assessment for the recreational and occupational scenarios.
Response: See response to Chapter 3, Section 3.2.4, Comment 9. A range of
soil ingestion values were used in the assessment as EP A standardized soil
ingestion rates (Porter 1989) were not available when this assessment was
initially prepared. Risk estimates have been revised based on the standardized
intake estimates suggested in the Porter (1989) memo.
Comment: This comment requests that the exposure duration to soils in the
occupational scenario be reduced in the winter months due to the heavy cloth-
ing worn that would reduce the amount of soil contact.
Response: The risk assessment already assumes no contact with soils in the
winter months because of snow cover.
Comment: This comment questions the estimates for surface water ingestion
during recreational or occupational use of the ponds.
Response: Incidental ingestion of surface water by individuals while using the
ponds for recreation activities is a possibility through splashing, contact with
fish, rinsing hands, etc. Best professional judgement was used in determining
the intake parameter for this pathway as standardized values were not available.
-------
water could be consumed each year through incidental and unnoticed ingestion.
The upper bound. estimate of approximately 30 ounces per year may seem high
for this pathway, however personal hygiene habits differ between individuals,
and this value represents a maximum estimate. .These values represent about
0.01 percent to 0.05 percent of the water ingested over a lifetime.
For occupational exposures, State employees are on the ponds every working
day performing activities that include significant contact with surface water. It is
not unreasonable to assume that incidental ingestion of surface water occurs.
The reasonable intake value represents approximately 0.06 percent of the water
ingestion over a lifetime while the upper bound intake estimate is approximately
0.4 percent of ingestion over a lifetime. These percentages assume an adult
ingestion rate of 2 liters per day, which may be an underestimate for persons .
continuously engaged in outdoor activities.
4.
Comment: This comment states that indoor air concentrations of particulate
contaminants may be considerably different from outdoor air concentrations.
. Response: MDHES agrees that indoor air concentrations could be considerably
different from outdoor air concentrations, that the concentration of con-
taminated particulates in indoor air is a function of the concentration of con-
taminated particulates in outdoor air (~) and indoor house dust (Sj), as well as
the concentration from indoor sources (~), and that inhalation of particulates in
indoor air is an important pathway of h-uman exposure. However, EPA does\
not agree that available data are adequate to show that the numerical
coefficients in the algorithm proposed by ARCa are applicable either in the
general case or to the Clark Fork region. Indeed, the "detailed justification"
provided. in Attachment 2 makes clear that all of the quantitative values
proposed in this algorithm are based on little or no data and are largely
intuitive. For example, ARCa notes that if a value of 0.4 is used for the pene-
tration factor (the coefficient of ~) the results appear "plausible" but there is
"no independent information." fn addition, ARCa notes that "in a cold
climate--where windows are closed much of the year-a smaller value might be
appropriate," and so arbitrarily chooses a coefficient of 0.3.
With respect to the second term in the equation (0.5 ~mSj), ARCO notes that
the "rate of which house dust is resuspended has -not -been systematically
studied," but nonetheless proposes a coefficient of 0.5 based on one study of the
effect of an outside air filter on indoor dust levels in one home, and one
observation from one home in Denver during a 4-day period when outdoor
concentrations of lead were low. Clearly, the results in these studies depend on
a number of parameters that may not be representative of conditions at the .
There is similar ~certainty in the values for Sj and ~m.
proposes the algonthm
5j = 0.1559
-------
.4.
for the contribution of outdoor soil (So) to contaminant concentration in indoor
soil (Sj). This coefficient is based on one data set from a mining location in
England, and there is no evidence that this coefficient is applicable to the
conditions at. With respect to the term A.DK (the concentration of dust in air),
ARCO acknowledges that simple activities such as vacuuming or making a bed
can increase values 4 to 50 fold, revealing that this term is highly variable and.
dependent on human activity. The value proposed by ARCO (55 ~g/m3) is not
measured, but is in fact calculated from measurements of ~, ~, and Sj for
arsenic at Mill Creek, using the very algorithm that contains tliis term. .
4.3.5 Exposure Point Concentrations.
1.
Comment: This comment states that the updated version of AP-42 should be
used for wind erosion of open areas... . .
Response: At the time of the emission rate calculation, the most recent AP-42
equation was not available. Therefore, emission estimates used the previous
AP-42 equation. The current equation uses data that may not be readily avail-
able for this site. Its use is probably not feasible in this case.
2.
Comment: This comment questions the use of 0.35 to 0.5 for reducing emis-
sions due to ~urface crusting.
ResPQnse: The 0.35 to 0.5 factor was an assumed factor. Referencing the
depth of soil eroded makes several assumptions that are not clear: that no new
dust was blown into the area for replenishment, and that erodible particles
present are evenly distributed on the surface. Furthermore, 57 JJm is a very
small quantity, only three-quarters of one particulate diameter in depth. The
entire section was calculated with available data, but ,when appropriate data
were not available, conservative assumptions were made and documented in the
report.
3.
Comment: This comment questions the use of a factor of 0.1 for relating I-hour
estimated concentrations to annual average dust concentrations.
Response: The methods used to determine I-year average concentrations from
the 1-hour maximum values are appropriate and standard EPA methodology
(EPA, 1988).
The available wind rose (from the Opportunity Ponds) did not discuss the length
of sampling time, data quality achieved, etc. More information is requiretl
before the wind rose data could be used in modeling scenarios. Also, havin~
16 wind directions has no bearing on developing a "dilution factor" as suggested
. Comment: This comment questions why the particle settling algorithm was nl'l
-------
2.
3.
Response: Information is not available to determine .the particle .settling
velocity for use in ISC. Particles less than 75 JJ m were assumed in the analysis,
and the fine particles were treated as gaseous emissions, a standard "worst-case"
assumption. ..
4.4
.ENVIRONMENTAL ASSESSMENT.
Comment: This comment states that impacts to aquatic or terrestrial life have
not been identified and no basis established for future impacts.
Response: The risk assessment does. state that terrestrial and aquatic life
appear to be productive and improving from past years with the re-establish-
ment of ground dwelling anim8:ls (based on communications with Montana.
Department of Fish, Wildlife and. Parks personnel). Chronic effects on
individuals of a species is a potential. However, it is unlikely they would" be
measurable even if costly studies had been performed. Impacts to terrestrial
organisms are difficult to determine unless they are acute or cumulative.
However, chronic aquatic life criteria have been and continue to be exceeded
for selected contaminants. This has been identified as a problem at the ponds,
which could impact downstream aquatic life. Also identified as a potential .
future impact to downstream aquatic and terrestrial life is a natural disaster that
may result from a breach of the dams and subsequent release of tailings and
contaminated waters.
5.0 COMPARISONS TO ARARS
1.
Comment: This comment states that compounds that affect the odor, color and
taste of groundwater under the site are unrelated to the location of the ponds.
Response: The presence of particulates, iron and othe~ compound$ responsible
for discoloration and odor in groundwater below the site are a result of the
presence of contaminants in the ponds.
Comment: This comment states that MCLGs should not be used in comparison
to groundwater under the site as they are nonenforceable health goals.
Response: MCLGs are concentrations of contaminants in water, whi~h would
result in no known or anticipated adverse health effects, with an adequate
margin of safety to protect sensitive subpopulations. These goals are strictly
health based and their comparison to groundwater quality under the site is
appropriate. Their regulatory status has no relevance to their use as a com-
parison to water quality at the site.
Comment: This comment states that ARCa agrees with the statement made in
the risk assessment that surface water at the site cannot be construed as ::\
-------
/
7.
Response: Comment noted.
4.
Comment: This comment states that use of an assumed value for hardness (100
mg/l) for use in comparing hardness dependent criteria wi,th measured values is
inappropriate.
, ,
Response: An assumed value of 100 mg/l hardness was used to calculate hard-
ness dependent criteria for use in comparing a criteria to a sample value for
high flow data. High flow data were the only samples in which hardness was
not measured. Because spring high flow is primarily due to snow melt runoff, a
value of 100 mg/l hardness is not unreasonable as hardness values' for snow
melt runoff are typically 50 to 60 mg/l (Heinle, 1990). The average measured
hardness of all-flow data was below 200 mg/l, except f9r waters that are limed.
Thus, the use of 100 mg/l hardness is appropriate. ' ,
5.
Comment: This comment requests that ambient water quality criteria for pro-
tection of aquatic life for arsenic be removed from the risk assessment, as
ARCa believes it is not valid based on current information. '
Response: The freshwater aquatic criteria for protection of aquatic life for
arsenic was revised in 1985. The criteria remain in effect until revised criteria
are published in the Federal Register. A 1988 document (U.S. Fish and Wild-
life Service, 1988), recommends a revision downward as adverse effects on
aquatic life have been demonstrated at concentrations well below the current
chronic criteria of 190 p. g/1.
6.
Comment: This comment states that comparisons between water quality criteria
and water quality as measured within the ponds are not valid as the ponds are
not classified as water bodies under the state.
Response: Comparisons between ambient water quality criteria for freshwater
aquatic life and water quality as measured within the ponds are appropriate as
the ponds support aquatic life and release water to the river. The classification
as a water body under State law has no bearing on the surface water quality and
its ability to support aquatic life in the ponds and the ponds influence on down-
stream aquatic life. Comparisons are valid and reasonable in the context of the
risk assessment.
Comment: This comment questions the comparison of concentrations of arsenic
detected in waterfowl breast tissue with FDA allowable levels of arsenic in
swine muscle.
Response: Specific. FDA allowable levels of arsenic or other contaminants in
edible waterfowl tissue were not available for comparison to the concentrations
detected in waterfowl at the ponds. The purpose of the comparison was to ha\'~
-------
2.
concentrations of contaminants detected in edible tissue frOm waterfowl at the
site. The MDHES agrees the comparison may not be appropriate.
\
6.0 HUMAN RISK CHARACTERIZATION
6.1
TOXICITY ASSESSMENT
6.1.1 Arsenic Toxicity .
1.
Comment: This comment states that the risk assessment overestimated the
carcinogenic risk through exposure to arsenic by not adjusting the estimates
downward by an order of magnitude. .
Response: See response to Chapter 3, S.ection 3.3.1, Comment 2. At the time
of the assessment,.the published cancer potency factor (CPF) for arsenic was 15
(mg/kg-day)=!. This CPF was adjusted to 1.5 (mg/kg-day)=! for use in the risk
assessment. It has been further adjusted to 1.65 (mg/kg-day)=! to reflect current
EP A estimates.
Comment: This comment discusses the nonlinearities in the dose-response
curve for arsenic.
Response: 7'he Risk Assessment Forum of EP A has carefully considered the
available evidence on the epidemiology, metabolism, and genotoxicity of arsenic
and its possible significance in the derivation of a cancer slope factor (EP A.
1988). All of the data reviewed by the Science Advisory Board (SAB) were
available to the Forum, including the studies of Vahter and Valentine. Based
on its review, the Forum concluded:
"While consideration of these data on the genotoxicity,
metabolism, and pathology of arsenic has provided infor-
mation on the possible mechanism by which arsenic may
produce carcinogenic effects, a more complete under-
standing of these biological data in relation to
carcinogenesis is needed before they can be factored with
confidence into the risk assessment process."
As more extensive data are developed on the toxicokinetics of arsenic metabo.
lism, the mechanism of arsenic genotoxicity and carcinogenesis, and the potenc:
of organic and inorganic forms of arsenic, these factors may be incorporated
into the risk assessment for arsenic as suggested by the SAB. Until that time. ::
is believed that the cancer slope factor derived by the Risk Assessment Forur:~
is the most appropriate value from the data currently available.
Absence of observable increases in skin cancer rate in the U.S. populati(l'
-------
,/
4.
Comment: This comment discusses the bioavailability of- arsenic and indicates
that risks in the FSare overestimated by using 100 percent absorption.
Response: The MDHES agrees that bioavailability of arsenic (and other
metals) is a relevant issue, and that metals in soil may be less bioavailable than
metals dissolved in water. However" bioavailability is a site-specific term that
depends on the physical and chemical form of the arsenic and on the nature of
the soil at the site. Thus, detailed site-specific physical-chemical data and or
site-specific tests of bioavailability are needed in order to estimate this para-
meter. Arsenic speciation was not evaluated in soils or sediments at the site. In
the absence of such data, it is necessary and prudent to assume that arsenic in
soil is as bioavaiIable as the form of arsenic involved in' those studies used to
derive the oral and inhalation slope factors. If "continuing investigations" pro-,
vide convincing data to justify a' reduction in bioavailability, EP A will consider
this and act accordingly. In the meantime, it is not appropriate to use an,
absorption fraction of 50 percent to account for bioavailability for several
reasons. First, the data cited do not indicate that humans absorb only 50 per-
cent of ingested inorganic arsenic, only that 50 percent is excreted in the urine.
Indeed, most studies suggest that humans absorb nearly all ingested inorganic
arsenic (EPA, 1984b). Second, the results of the pilot study (Johnson et al. .
1989) are based on only one animal per test group, and clearly are not adequate
to establish 10 to 20 percent as a reliable estimate of arsenic uptake from soil.
Moreover. it is not clear that the soil sample selected for study is representative
of the sedIment of. Also, the relevant data item is not how much is absorbed
per se, but the ratio of the gastrointestinal absorption of arsenic in sediment
from compared to the gastrointestinal absorption of arsenic in the water
consumed by the Taiwanese population studied by Tseng.
6.1.2~
1.
Comment:, This comment questions the methodology used in the derivation of
clean-up levels for lead in soils. Specifically, it questions the 500 to 1,000 ppm
recommended target cleanup level for Superfund sites.
Response: See response to Chapter 3, Section 33.1, Comment 2. EP A believes
that the method suggested by ARCa has merit. However, the ARCa suggested
method is dependent upon the input parameters used (as all models) and the
values selected by ARca for use in the model are debatable' and not
conservative.
Comment: This comment continues with a presentation of a figure (A6-1) that
relates a soil concentration of 10,000 ppm to a 25l£g/dl blood lead.
Response: Figure A6-1 is incorrectly characterized both here and in its legend
as presenting the relationship between soil lead concentration and blood lead in
-------
of 6.8 is a reasonable upper bound (p. V -10). ARCa st~tes that the slope at
mining sites is about 2, and that lead in soil at mining sites is less bioavailable:"
and has less influence on blood lead levels than at nonmining sites (smelters,
other urban areas). If so, this implies that a slope higher than 2.2 could be
appropriate at smelter and other nonmining sites" This would be consistent with
the fact that the slope calculated from toxicokinetic parameters is about 5, and
that empirical estimates are expected to yield lower-than-actual slopes due to
the effects of confounding factors (EPA, 1989c, p. V-14).
ARca states that Gradient Corporation analysis identified slope factors from
two studies of mining sites. However, concerning the Park City Utah study,
Gradient states: "Because no soil lead concentrations are available for
individual children, a slope valu~ cannot be.calculated." Thus, this slope of. 1.8.
is completely undocumented. The other. cited mining study, from Telluride
Colorado (Bomschein et al. 1988), presents a slope of 3.7 for soil lead values.
between 100 and 1,000 ppm and a slope of 2.2 for soil lead values between 500
to 1,000 ppm. Thus, in agreement with other studies, the calculated .slope
depends upon the soil lead concentration, among other factors. The cited
Bornschein et ale (1988) slope of 2.2 for 500 to 1,000 ppm is in fact the average
value for the disaggregate slope suggested by EPA (1989c), based on Stark et ale
(1982). However, this agreement is coincidental and does not indicate that
there is a single value for this slope that can be used without uncertainty. .
3.
Comment: This comment questions a statement in the FS that blood lead level
of 10-15 p,g/dl are of concern in yo1:}ng children.
Response: The statement made in the FS is accurate; ARCa is incorrect in
stating that blood lead levels of 10-15 p,g/dl are not of concern in children.
Such concern can be found even in the selected references cited by ARCa:
1.
David and Svendsgaard (1987). This review summarized studies linking
prenatal lead exposure with adverse outcomes, but did not, as implied by
ARCa, dismiss concern about postnatal exposure. The conclusions of
this paper were: ''There can now be little doubt that exposure to lead,
even at blood levels as low as 10-15 P,g d}=l and possibly lower, is linked
with undesirable developmental outcomes in human fetuses and children"
(p. 29~).
2.
Bellinger et ale (1989). ARCa selectively summarized the results of this
study for middle-class children. These investigators found that prenatal
exposure above 6 p,g/dl and postnatal blood levels higher than 10 p,g/dl
were associated with neurological deficits in lower-class children.
3.
Chaney et ale (1989). [Cited by ARCa on p. A-22 and p. 16 of Attach-
ment 4.] This article states "...the mean Pb-B of children 0.5 to 5 yea
-------
to be at or above the threshold for adverse effects of ingested .Pb on
neurologic development of children" (p.123). This article also concludes
"We interpret these data that soil must be less than or equal to 150 mg
Pb/kg in order to prevent excessive Pb absorption in children."
Thus, although it may be correct that fetuses are more sensitive than infants and
children to a given blood lead level, it is certainly not correct to characterize 25
~g/d1 as a "health protective blood lead level" (ARCa Attachment 6, p. 3 and
p. ES-8) for infants and children. Furthermore, ARCOs reliance on the Centers
for Disease Control (CDC) guideline of 25 ~g/d1 is inappropriate. for two
reasons. First, the CDC did not identify this as the no-effect level, but as the
level of sensitivity of the erythrocyte protoporphyrin (EP) screening test which
utilizes capillary blood. The CDC stated: "Although t.he biologic threshold for
lead toxicity, as manifested by increasing EP.levels, is less than 20 ~.g/dl, the.
criteria for a screening program have to take into account additional factors: (1)
acceptability, sensitivity, and specificity of the screening procedure; (2)
cost-effectiveness; and (3) the feasibility of effective. intervention and follow-up"
(DHHS, 1985). Second, this recommendation is five years old and a number of
more recent publications have reported effects at PbB levels lower than 25
~g/dl (Bellinger et al. 1989; Fulton et al. 1987; McMichael et al. 1988;
Needleman 1989). The Clean Air Scientific Advisory Committee to EPA
concluded that blood lead levels of 10 to 15 ~g/dl in children are associated
with the onset of subtle biomedicallyadverse effects (ATSDR, 1988). Indeed,
the CDC is currently considering a downward revision of their guideline.
Although there is no consensus among researchers, many investigators believe
that if a target blood lead level for children were to be selected, it would be 10
~ g/ dl or lower.
The MDHES agrees that in principal, adults (includ41g pregnant women) would
in many instances receive a lower dose of lead at a given soil level than would
children. However, quantitative evaluation of the magnitude of the difference is
not possible with current data, and the 6.5 fold factor cited by ARCO cannot be
evaluated because it is based on a personal communication from Bornschein or
coworkers in 1989. ARCO's quarititative conclusion that a soil lead level that
protects young children will also protect fetuses goes beyond currently available
d~~ .
4.
Comment: This comment disputes the FS presentation of background blood
levels in United States children of 16 ~g/d1.
Response: The baseline blood lead level is indeed dropping in the United
States and EPA has estimated that the geometric mean blood lead level for
2-year-old children in 1990 will be 4.2 to 5.2 ~g/dl (EPA, 1989c). The following
attempt was made to replicate the value of 2.24 ~g/dl due to nonsoil source'
based on the description of the method provided by ARCa. The following i" .:
-------
4.
5.
1.
Table 5-1 of EPA (1989c, p. V-9) was used as the basic source of
numbers.
2.
Values in that table for ingested dust and inhaled air were dependent
upon the air lead content and the lowest air content in the table was 0.25
Jjg/mg/m3. Since ARCO stated that a site-specific value of 0.1 p.g/m3
ambient lead was used (Attachment 6, p. 3), corresponding levels for
ingested dust (0.6) and inhaled air (0.2) were extrapolated from the
table.. .
3.
Corresponding totals were 1.9 for nonsoil/dust and. 0.7 for soil/dust
sources, or a 73 percent contribution from nonsoil/dust sources.
The geometric mean of 4.2 and 5.2 Jjg/dl was calculated (4.67 Jjg/dl)..and
multiplied by 73 percent to yield 3.42 Jjg/dl as a background lead level.
This value is more than 50 percent higher than the value provided by
ARCO. .
The upper 95th percentile background using a geometric standard
deviation of 1.42 was calculated to be 6.1 Jj g/ dl.
Thus, although baSeline blood lead levels are declining in the United States,
these calcul~tions illustrate that sufficient information is not presently available
to make a quantitative evaluation of the appropriate value.
6.2
MECHANISMS OF QUANTIFYING RISK
No comments.
6.3
RISK ESTIMATES
Comment: This comment disputes the conservative nature of the risk estimates
provided in the FS.
Response: Estimates of risk are, in general, conservative. EP A has determined.
this is an appropriate approach for protection of the public health. As several
responses have shown, when intake assumptions or exposure factors are 'altered,
little change in the risk results.
Risk estimates between 1~ to 10:1 are not considered "acceptable" by EPA, but"
-------
7.0 ENVIRONMENTAL RISK CHARACTERIZATION
7.1
QUAUTATIVE ENVIRONMENTAL RISKS
Comment: This comment implies the environmental assessment fosters mis-
conceptions about actual environmental risks. .
Response: The risk assessment does not present misleading information. The
description of a potentially stressed ecosystem does not state nor iIIlply that
these conditions are occurring at the ponds. No data are available on the status
of the ecosystem at the ponds with which to determine if any. adverse conditions
currently exist. From all outward appearances, the ecosystem of the ponds is
stable and improving. However, subtle effects cannot .beascertained by casual
observation, and lack of data prevents accurate prediction of future effects. .
As has been previously stated, the risk assessment did not imply that bio-
concentration of metals in fish and algae is unique to the ponds.
Lack of diversity in an aquatic ecosystem is commonly associated with stressed
conditions. MDHES agrees that many explanations are possible. No data exist
with which to evaluate this condition. .
8.0 UNCERTAINTIES AND LIMITATIONS
Comment: This comment states that assumptions and parameters used in the
risk assessment exhibit a bias toward overestimating the risk.
Response: It is EP A policy to provide conservative estimates of risk in order to
protect the public health. The risk estimates are provided as a range 'of values,
with the maximum plausible intending to provide an upper bound that actual
risk will not exceed.
9.0 REFERENCES
ATSDR. 1988. The Nature and Extent of Lead Poisoning in Children in the United
States: A Report to Congress. U.S. Department of Health and Human Services,
Atlanta, GA.
Angle, C.R., and M.S. McIntire. 1982. Children, the Barometer of Environmental
Lead. Adv. Pediatr. 27:3-31.
Bellinger, D., A Leviton, C. Watemaux, H. Needleman, and M. Rabinowitz. 1989.
Low-level. Lead Exposure, Social Class, and Infant Development Neurotoxicol.
-------
Bornschein, R.L, C.S. Clark, J. Grote, B. Peace, S. .Roda, and P~ Succop. 1989. Soil
Lead Blood Lead RelationShip in a Former Lead Mining Town. Paper presented at :
"Lead in Soil: Issues and Guidelines" Chapel Hill N.C. .
CH2M HILL. 1988. Silver Bow Creek.C.ERClA Pata Summary.
CH2M HILL 1989. P
Chaney, R.L, H.W. Mielke, and S.B Sterrett. 1989. Speciation, Mobility 'and Bio-
availability of Soil Lead. Environ. Geochem. Health 11(suppl.):105-129.
Chen, CJ., Y.C. Chuang, S.L You, T.M. Lin, and H.Y. Wu: 1986. A Retrospective
Study on Malignant Neoplasms of Bladder, . Lung.. and. Liver in Blackfoot Disease.
Endemic Area in Taiwan. Br. J. Cancer 53:399-405. .
Chen, CJ., Y.C. Chuang, T.M. Lin, and H.Y. Wu. 1985. Malignant Neoplasms Among
Residents of a Blackfoot Disease-endemic Area in Taiwan: High Arsenic Artesian
Well Waters and Cancer. Cancer Res. 45:5895-5899.
Chen, CJ., T.L Kuo, and M.W. Wu.
1(8575/6):414-415.
1988.
Arsenic and Cancers.
Lancet
Davis, J.M. and DJ. Svendsgaard. 1987. Lead and Child Development. Nature
329:297-300. .
Fulton, M., G. Raab, G. Thomson, D. Laxen, R. Hunter, and W. Hepburn. 1987.
Influence of Blood Lead on the Ability and Attainment of Children in Edinburgh.
Lancet 1 (8544):1221-1226. .
Heinle, D. 1990. Personal communication. CH2M HILI.., Water Quality Specialist,
Seattle, W A
Johnson, J., G.B. Freeman, and J.M. Killinger. 1989. Pilot Bioavailability Study of
Lead and Arsenic in Soil Following Oral Administration to Rabbits, unpublished.
Prepared by Battelle Laboratory for ARCO Coal Company. Provided by ARCa.
McMichael et al. 1988. Port Pirie Cohort Study: Environmental Exposure to Lead
and Children's Abilities at the Age of Four Years. New Engl. J. Med. 319:468-478.
Multitech. 1987. P
Needleman, H.L. 1989. The Persistent Threat of Lead: A Singular Opportunity. Am
-------
5. Comment: Point source discharge requirements for the Pond 2 discharge should' not be
set at water quality staridards. Instead, site specific criteria and other factors outlined in
ARM sS~ss~ 16.20.622 and 16.20.631, and the mixing zone discUssed at 16.20.634, should be
sued to set discharge limits which will meet water quality standards in the ambient water.
Response: The factors cited by ARCa may be used to set a point source discharge, but
their use is not mandatory. The MDHES Solid and Hazardous Bureau consulted with the
MDHES Water Quality Bureau, who is the agency assigned with the administration of the
Water Quality Act, and the factors outlined by ARCa were determined to be inappropriate.
Using the appropriate water quality standards as the end-of-pipe point source discharge was
determined to be the appropriate way to ensure that ambient water quality standards will
be met at the beginning of the Clark Fork River, since the point soUrce discharge from Pond
2 will be a primary source of water.to.the.Clark.;Fork-River.' . This will also protect the'
waters within the Bypass, where sensitive fish populations and hatcheries are likely to exist.
6. Comment: The point source discharge standards established at 40 CPR Section 440;104,
and discussed at ARM sS~ 16.20.631 are not relevant and appropriate or applicable to the
conditions found at the Warm Springs Ponds operable unit.
Response: These standards are not applicable to the discharge, but EP A continues to
believe that the standards are relevant and appropriate. The standards are developed for
mining and dressing effluent from copper, lead, zinc, gold, silver, or molybdenum ores.
These contaminants are the same contaminants found at the Warm Springs Ponds and other
operable units with the Clark Fork Basin project. In addition, the technology on which
these standards were developed is essentially settling and treatment, which is similar to the
Warm Springs Pond system. Therefore, the standards are relevant and appropriate for the
Warm Springs Ponds system. .
These standards are less stringent that the water quality standards for the Pond discharges
discussed earlier. Therefore, the water quality standards are the only point source discharge
standards listed in the final list of ARARs.
7. Comment: Monitoring and reporting requirements set forth in 40 CPR sS~ 122.41 and
best management practices of 40 CPR sS~ 125.100 are not ARARs, because they are not
substantive requirements.
Response: Monitoring requirements are not administrative, but substantive. 55 FR 8757.
Best Management practices too are substantive requirements, which are similar to action
specific ARARs and will ensure compliance with the numeric limitations of the Pond 2 point
source discharge. Therefore, these remain as identified ARARs.
As noted elsewhere, the point source discharge from Pond 2 should continue to be subject
-------
ARARs, including the best management practices and the monitoring requireme~ts, will
ensure approriate monitoring and reporting for the Pond discharge.
B. Ground Water
8. Comment: A mixing zone should be established to determine the ground water point
of compliance, and the appropriate point is at the property boundary rather than the waste
unit boundary.
Response: Granting a mixing zone for ground water compliance is entirely within the
discretion of the administering agency, under State law, ARM sS~ 16.20.634.' EP A, in
consultation with the MDHES, has determined that such a zone would not be appropriate
here, because the ground water below the Pond 1 waste unit should be available for public
use and other uses, and because establishing the point of compliance at the waste unit.
boundary rather is consistent with current EP A guidance. .
9. Comment: MCA sS~ 85-2-504 should not be an ARAR because it is a general statute
from a non-environmental statute.
Response: EP A agrees that this statute is too general to be considered an ARAR.
10. Comment: MCA sS~ 85-2-505 should not be an ARAR because it is a general statute
from a non-environmental statute.
Response: This statute establishes certain directive conditions concerning the drilling of
ground water wells, to protect the spread of contamination during construction of a well.
This is obviously addressed to prevent pollution and contamination, and is therefore an
environmental standard. Ground water wells will be required at the site, to measure
compliance with the ground water standards. Therefore, these requirements are applicable
requirements to any drilling which must occur at the site during or after remedial action.
11. Comment: Ground Water Pollution Control Regulations and standards, which
incorporate federal MCLs, should not be ARARs for ground water, because there is no
public water supply system in the area, and because use of the aquifer as a ground water
supply is unlikely.
Response: The State has classified the aquifer at the operable unit as a Qass n aquifer,
potentially suitable for public use. This classification corresponds with EPa guidance, which
directs that potentially usable aquifers should be restored to potential use as part of
Superfund cleanups. EP A's guidance is based upon Congress' clear concern for the
protection of ground water, as evidenced in CERCLA and other environmental laws.
Therefore, MCL standards are identified as the relevant and appropriate ARARs for th
-------
12. Comment: The State's nondegradation regulations and statute are not ARARs because
. they are general goals, and because the aquifer is not likely to be used, and because the
aquifer is already degraded.
Response: Statutes and regulations that contain general goals can be ARARs, if they are
directive in intent and promulgated. Non-degradation statutes such as this one are
specifically cited by EP A as probable ARARs, even though they may be characterized as
general goals. 55 FR 8746 - 8747, 53 FR 51438. .
The State interprets its statute to be applicable here, because it prevel1ts the spreading of
existing plumes of groundwater plumes, as well as the creation of new plumes. EP A defers
to this interpretation by the State. EP A notes that compliance ~th the MCL standards
identified for this action will result in-~ompliance with, these ARARs. .
As preViously explained, this aquifer is potentially usable as a drinking water source, and
should be cleaned up to appropriate -standards.
C. Air
13. Comment: MCA s5fi 75-2-102 is a general policy statement that should not be an
ARAR.
Response: This statute is direct in nature, in part, and is appropriately identified as an
ARAR.
14. Comment: Lead and PM-10 ambient standards should be identified as federal
standards, using appropriate 40 CFR Part 50 citations, rather than state citations, because
the State standards are not more stringent.
Response: When State programs are delegated or authorized, the appropriate citation
should be to the State standard, although the ARAR is considered to be a federal ARAR.
55 FR 8742. Attainment with these ARARs will be determined by the methodologies
described in 40 CFR Part 50, since these are the most definitive and up-to-date methods
available.
15. Comment: The standards for sulfur dioxide and particulate matter found at 16.8.925
are not ARARs, because they were formulated for major stationary sources, and no such
source is expected or present at the operable unit.
Response: BPA agrees with this comment, and has dropped these standards from the final
-------
16. Comment: MCA sS~ss~ 50-70-102 and 113 are general policies which should.not be
ARARs. . , , '
Response: EP A agrees that section 102 is general and should not. be an ARAR. EP A
believes section 113 is directive and specific in nature, and should be retained. Section 113
is implemented by specific numeric limitations found at cited regulations.
17. Comment: State regulations pursuant to the State's health Noise and Air Con,taminant
Regulations are .not part of an authorized program, and should not be state ~s.
> '
, Response: EP A agrees that the State's OSHA equivalent program has not been approved
and authorized, and is therefore not more stringent, where the. State standards are
duplicative of the federal OSHA standards: For those 'standards, EPA has identified the'
federal regulations and standards. However, the State has promulgated occupational health
standards for lead and arsenic that are not duplicated by the federal program. These
standards are applicable standards, are more stringent than, federal standards, and are '
identified as State standards in the final list. '
EP A notes that nose regulations do not address contaminants, pollutants or hazardous
substances, and are therefore not ARARs, in a strict sense. Those State standards are listed
in the other laws section of Attachment 1, and should be complied with during the remedial
action implementation and thereafter.
II. Location ARARs
18. Comment: MCA sS~ss~ 76-5-102 is a general goals statute, and should not be considered
an~.
Response: This section is directive in nature, and is identified as an~. This approach
is explained above.
19. Comment: Only substantive portions of MCA sS~ 76-5-402 are ARARs.
Response: EP A agrees with this comment.
20. Comment: MCA sS~s.f 76-5-1101 and 1102 are general goals which should not be
ARARs, and is administrative in nature.
Response: These statutes are of a directive nature, and are therefore legitimate ARAR ~
-------
21. Comment: Only substantive ponions of ARM sS~ 36.15.216 are ARARs - administrative
and permit requirements should not be identified as ARARs.. .
Response: EP A agrees with this statement.
regulation are identified. .
Only substantive requirements of this
22. Comment: Only certain portions of ARM sS~ 36.15.606 should be identified as relevant
and appropriate. .
Response: EP A believes that all portions of the regulation are relevant and appropriate,
because they address situations similar to the situation at the Warm Springs Ponds, namely
strengthening Pond 3 to hold a 100 year flood flow, and doing appropriate reconstruction
of the Mill-Willow Bypass. It should be. noted that.the.ROD is exp.ected to comply with this.
provision, as currently understood and designed.
23. Comment: Only substantive portions of ARM sS~ 36.15.801 are substantive and
therefore ARARs.
Response: EP A agrees with this statement, and has identified only substantive portions of
this regulation as ARARs.
24. Comment: MCA sS~ 75-7-102 is a general goal and should not be an ARAR.
Response: This section is directive in nature, and is a valid ARAR, as explained above.
25, Comment: ARM sS~ 36.2.404 should not require a permit for this action, pursuant to
the permit exemption of section 121(e)(4) of CERCLA
Response: EP A agrees that the permit exemption applies to berm strengthening, and permit
requirements should not be identified as ARARs. These criteria have been identified as
TBCs, to aid the agencies in evaluating this project as it proceeds through the RD IRA
process.
26. Comment: MCA Sl~ 87-5-501 is a general policy that should not be an ARAR.
Response: EPA agrees with this comment and has not identified this as an ARAR in the
final list.
27. Comment: Only the substantive portions of federal locations specific requirements
-------
Response: EP A agrees that only substantive requirementS emanating from the location
specific requirements described in the FS are ARARs. However, EP A described the
consultation processes for these statutes in the FS, to demonstrate how it would identify
substantive requirements for location-specific ARARs. See section 7 of the response to
comments, Chapter 3. The final ARARs list identifies substantive requirements only for
these ARARs.
III. Action ARARs
A. Dam Construction
28. Comment: MCA sS~ss~ 85-15-207 and 208 are general policy s.tatements and should not
be ARARs.
Response: These statutory provisions are directive in nature, and are included as ARARs
in the final list.
29. Comment: ARCa agrees that ARM ss~ 36.14.501 is an ARAR.
Response: This key ARAR is listed in the final ARAR list. All berms within the operable
unit are classified as high hazard, and this ARAR is applicable to all of them. The most
important portion of this provision is the need to strengthen berms to the MCE standard
30. Comment: ARCa agrees that sS~ 36.14.502 is an ARAR.
Response: This key ARAR is listed in the final list. Upon further review of this ARAR,
and further consultation with the Department of Natural Resources and Conservation, who
administers the Dam Safety Act and its regulations, it has been determined that the
provision will require all berms within the operable unit to meet the 0.5 PMF standard.
This is based on the inflow design flood expected for the Ponds, and the fact that the Ponds
are in a series, and therefore the highest standard must be applied to all three.
31. Comment: Revegetation and reclamation requirements identified by the State, pursuan t
to the Strip and Underground Mine Act and implementing regulations, are not relevant and
appropriate, because the use of those standards, designed to address coal and uranium mine
reclamation, is not well suited to the characteristics at the Warm Springs Ponds operable
uni t.
Response: The current ROD will require excavation of some contaminated soils, sediments.
and tailings (Within the Mill-Willow Bypass and elsewhere), consolidation of that material
into two disposal facilities, and covering of certain contaminated soils and sediments in plac
-------
identified by the State.
The contaminated soils, sediments, and tailings at issue within this operable unit are the
result of several years of downstream transport of I11etals ore mining in Butte and Anaconda,
including processing, bonification, and extraction waste. The Ponds and surrounding areas
are similar to settling ponds used in the mining industry to capture and separate
contaminants from waste water, before that water enters a surface water body.
EP A has compared the site activities and conditions with the regulations cited by the State,
using the criteria found at . .
40 CFR sS~ 300.400(g)(2). The svMRA regulations purpose is to ensur.e that revegatation
and reclamation activities are done in such a manner that they are permanent and long
. lasting, such that cover of mining wastes remains despite adver~e weather conditions or
other factors. That is also the purpose. of-this' Superfund .remedial action - to pr~vide .
permanent protection against the furiher release of contaminants found at the. site,
especially those located in the disposal area. The substances addressed by the requirements
is different than that found at the site (coal and uranium waste as opposed to hard rock
mine waste), but the substances are not unrelated. Both concern metals contamination as
opposed to organic chemical contamination, and both seek to protect against phytotoxic
effect and human health exposure. Both the regulations and the Superfund cleanup regulate
reclamation and revegetation after mining disturbance, and both concern soils replacement.
Both look to restor the affected area to the fullest possible post-mining use, such as human
exposure and wildlife habitat.
- Therefore, EP A has concluded that these requirements are relevant and appropriate, and
has listed many of the requirements identified by the State in the FS in the final ARAR list.
EP A has examined this issue carefully, and has dropped certain of the State's identified
ARARs in this area. These. include general policy statements, administrative requirements,
and standards which did not fit sensibly in applying the standards to the planned cleanup.
32. Comment: MCA sS~ss~ 75-5-101 and -605 are general policy statements that should not
be ARARs.
Response: EPA agrees with this statement, and has dropped these requirements from the
final ARAR liSt.
33. Comment: The point source discharge from Pond 2 of the Warm Springs Pond operable
unit should not be subject to a continuing permit pursuant to the State's Water Quality Act.
because it is an on-site response action exempt from permit requirements, pursuant to
section 121(e)(4) of CERClA, 42 V.S.C. slA 9621(e)(4). The discharge should he
considered "on-site" even though the discharge will.enter waters off of the site eventuall~
-------
MPDES permit which has existed for the site for several years. should be suddenly
discontinued. ARCa correctly points out that EP A's discussions of "on-site" ave specifically
described situations such as this, as being on-site despite the downstream transport of
contaminants. EP A accepts this rationale, and finds that this point source discharge, and
other point source discharges which may be effected or created by Superfund actions along
Clark Fork Basin Rivers, are "on-site" and therefore exempt from local, State, and federal
permits, as long a substantive standards within ARAR determinations are met.
Neertheless, EPA believes that the existing permit should be continued at the site, in the
interest of ensuring adequate monitoring and reporting and in the interest of maintaining.
consi~tency within the State's MPDES program.
Therefore, the point source discharge from ~ond 2 which will result after completion of this.
remedial action should continue to be permitted under the State Water Quality. Act.
Numeric standards for the discharge and other substantive standards are identified in the
final ARAR list, and the State's reissued permit will reflect these same requirements.
ARCa will be required to apply for the permit in the same manner as any permitted u:nder
the Water Quality Act. .
34. Comment: ARM sS~ 16.20.633 should not be an ARAR because it is a general goal.
Response: Section 16.20.633 is directive in nature, and is therefore an appropriate ARAR.
Compliance with the numeric standards for the point source discharge will achieve.
compliance with this ARAR.
35. Comment: ARM sS~ 16.20.904 is not ~ ARAR because it is administrative in nature.
Response: EP A agrees that this is not an ARAR, but notes, as explained above, that a
permit renewal must be obtained by ARCa for the Pond 2 discharge, according to these
and other applicable procedures.
36. Comment: MCA sS~ss~ 75-6-112 is a general policy goal and is not relevant and
appropriate, because the Clark Fork River is not a public water supply.
Response: The statute is directive in nature and is therefore appropriate for use as an
ARAR. EPA agrees that the Qark Fork River is not designated for public water use, so
subsection(1) has been eliminated from the list. However, subsection 2 discusses the
discharge of po~lutants to any State waters, and therefore its proscriptions are relevant and
appropriate to this action. Compliance with the numeric standards for the point source
-------
37. Comment: ARM sS~ 16.20.1016 should not be an ARAR because it is a general policy.
Response: Certain provisions of this regulation are directive and substantive, and are
included in the final ARAR list. .
38. Comment: ARM sS~ 16.44.702 is not relevant and appropriate, because RCRA
regulations should not be considered at this site.
Response: As more fully explained below, EP A believes that certain RCRA standards are
relevant and appropriate for the site deanup. Accordingly, EP A has identified the standards
at 40 CFR sS~ 264.97 only from the 40 CFR Part 264, subpart F regulations as relevant and
appropriate here (these standards are incorporated by reference in ARM sS~ 16.44.702).
Other portions of the subpart F regulations. are not identified in the . final list. In addition,.
these standards may be applied at the .site in a manner which treats all of the units at. the
site as .one cluster of units. This limited use of RCRA is sensible in view of the ground
water contamination problems already present here, and the n~ed for adequate monitoring
to detennine further ground water contamination from the Ponds, and to monitor
compliance with the contamination-specific ground water ARARs identified for this site.
These specific requirements are relevant and appropriate for the site.
Montana and EP A Hazardous Waste Management Requirements
39. Comment: ARCO objects generally to the use of any statutes or regulations to from
the Montana hazardous Waste Management Act or the federal Resource Conversation and
Recovery Act (RCRA) as ARARs as applicable or relevant and appropriate. ARCO
contends that much of the waste at the operable unit is bonification or extraction waste,
which is specifically excluded from regulation under RCRA by the Bevill Amendment,
42 V.S.C. sS~ 9601(b)(ill)(A)(ii). Further, ARCO does not think that use of RCRA can be
justified at this site as relevant and appropriate requirements.
Response: EP A agrees that RCRA is not applicable to the wastes found at the operable
unit. However, EP A believes that certain RCRA requirements as relevant and appropriate
requirements applied to certain waste within this and other Clark Fork Basin Superfund
sites is warranted. - -
Use of RCRA standards for mining waste sites is endorsed in the Guidance and in the new
NCP. Although EPA has concluded that application of.all RCRA requirements to .all
mining waste across the country is not warranted, and has therefore exempted mining
extraction and beneficiation waste from the RCRA Subtitle C regulatory scheme, the
guidance and preamble to the new NCP note that this does not the prevent the case-by-case
use of some RCRA requirements as relevant and appropriate, if the site conditions warrant
it. 55 FR 8763 - 8764; Guid~ce Part n at 6-4.
At this site, two disposal facilities will be created, and two treatment ponds will be
-------
of waste, which is contaminated with high levels of arsenic, lead, cadmium, zinc, and copper.
The remedial investigation for the site has demonstrated that ground water contamination
is already a source of release for these contaminants. The risk assessment for demonstrates
that unrestricted exposure to these contaminants would present a risk to human health and
the environment. The sudden release of contaminants of this waste into the surrounding
rivers has probably caused fish kills, and would cause catastrophic damage to the
environment and to human health. EP A believes that these site conditions sufficiently
distinguishes the operable unit from the general wastes studied in EPA's RCRA
determination document, and that comparison of these conditions to the RCRA statute and
regulations discussed below meets the relevant and appropriate criteria given i.Q.
40 CfR sS~ 300.400(g)(2).
Given these site conditions, EP A believes it is reasonable and within its discretion to use.
certain, limited RCRA standards to control operation of the active Ponds, and to ensure
adequate disposal of wastes within the two disposal facilities. For the active Ponds, these
standards require design to prevent overtopping and to prevent massive failure. 40 CFR ss~
264.221(f)and (g). For the disposal areas, waste must be drained of free liquids and
stabilized, and the facilities must be designed to minimize maintenance, and minimize or
eliminate further releases to the ground water or surface waters or the atmosphere to the
extent necessary. The dirt cover must be designed to function with minimum maintenance,
promote drainage and minimize erosion or abrasion, and accommodate settling and
subsidence. This does not require the facilities' to be lined with impermeable liners or
capped with an impermeable cap, which the normal RCRA closure would require.
40 CFR sS~ 264.228(a)(2)(iii)(B)(C)and (D). Limited ground water monitoring after closur "
is required as described above. Finally, survey plats to local land use planning authorities; .
and deed notices must be submitted for the disposal units. 40 CPR sS~ss~ 264.116 and .119.
These requirements are but a very small portion of the total number of RCRA operational
and post closure requirements which could possibly have been identified, and .make sense
under the conditions at the site.
Operation and Maintenance plans for this site will need to include appropriate post closure
care for the disposal areas. Eventually, the active Ponds will be discontinued as treatment
facilities and Silver Bow Creek will be routed directly into the Bypass and the Cark Fork
River. Appropriate care and maintenance of the ponds as wetlands and wildlife ponds
should be required in the operation and maintenance plans.
EP A notes that specific regulations governing extraction and beneficiation waste dispos~:
and post disposal care are being developed pursuant to Subtitle D of RCRA Until thos~
standards are promulgated, EP A believes that the use of limited RCRA Subtitle (
requirements for certain wastes only is warranted.
Solid Waste Requirements.
. .
. 40. Comment: MCA SI~SI~ 75-10-202,-212, and -214 are general goals which should r
-------
Response: EP A agrees that sections 202 and 212 are general statements of legislative intent
'or enforcement provisions, an~ should not be an ARARs. Section 214 are directive in
nature, and are included in the final ARAR list. Only certain provisions of ARM sS~ss~
16.44.505 and .523 are listed in the final ARARs list, as relevant and appropriate
requirements.
Air Quality Requirements
41. Comment: MCA Sl~ 75-2-102, and ARM ss~ 16.8.1427 should not be ARARs~ because
they are general in nature or do not address expected conditions at the site. '
. ,
Response: EP A agrees with this comment and has not included these requirements in the
final list. In addition, EPA has not included.ARM'ss~16.8;1l03 iIi the final list, because no '
new or altered sources are expected for this action.'
Safety and Health Conditions
42. Comment: MCA sS~ 50-71-201 is not an environmental law and is a general statement,
and should not be an ARAR.
Response: EP A believes that this provision is addressed to the protection of human health
and the environment, at least in part, and is directive in nature. Therefore, it is included
in the final list of ARARs. '
43. ,Comment: MCA sS~ss~ 50-778-202, -203, -204, and -307 are administrative requirements
and should not be ARARs. .
Response: Substantive provisions similar to action ARARs are included in these statutory
provisions, and these requirements have been identified as ARARs in the final list, to the
extent they are applicable requirements.
44. Comment: Provision of the Montana Water Development Program Act, the Natural
Streambed and Land Preservation Act of 1975, and the Montana Streambed Protection Act
which address water rights are not environmental laws, and should not be ARARs.
Response: EP A agrees that these provision do not meet the definitions of ARARs, and has
listed them in the Other Laws section of the ARARs list.
45. Comment: Federal Surface Mining Control and Reclamation Act 'provisions are not
-------
. Response: Use of these standards at CERCLA mining sites is endorsed in EPA's guidance
documents. CERCLA Compliance with Other Laws Manual: Part fi, pp. 6-1 - 6-3. The
requirements are intended to ensure that reclamation and revegetation actions are long
lasting and permanent. This is the same purpose as the CERClA cleanup. For these
reasons, the requirements are well suited to the cleanup a1 this operable unit, and are
relevant and appropriate.
46. Comment: The State of Montana's TBC List contains non-promulgated standards,
which should not be considered ARARs.
Response: Non-promulgated standards are appropriately listed in the TBC list for
consideration by the agencies considering the remedy selection. If promulgated standards
do not exist for the listed standards, the TBC standards can be selected as remediation goals
for the cleanup. 55 FR 8744. Here, promulgated standards existed for the contaminants
of concern, and TBC criteria were not selected. EP A and the State will continue to list such
criteria, and continue to consider their selection if appropriate.
47. Comment: MCLGs should not be listed, as MCLs are sufficiently protective.
Response: Although EP A's proposed NCP, cited by ARCO, agreed with this position,
EPA's final NCP changed this position. Promulgated MCLGs which are not zero will be .
identified as ARARs at Superfund sites. 55 FR 8750 - 8753. 40 CFR ss~
300.430(e)(2)(i)(B). Here, no promulgated MCLGs exist for the contaminants of concern.
Should MCLGs be appropriate (of other sites, EP A and the State may list them in final
ARARs lists.
48. Comment: The State's non-comprehensive list of other laws is improper, because
CERCLA preempts all other laws, and only ARARs should be required for Superfund
cleanups. .
Response: EP A agrees that CERCLA preempts State and federal environmental or siting
laws, and that ARARs are the only method for assertion of these laws at a Superfund
cleanup. 55 FR8741- 8742. (The State of Montana disagrees with this position). However,
it is unclear of the relation of other, non-environmental laws to Superfund cleanups. In the
face of this lack of clarity. EP A has listed certain of the State's non-environmental laws
which should be complied with during the cleanup. The laws listed should be able to be
complied with little difficulty. EP A's listing of these laws in no way prevents EP A from
arguing about the exact applicability of non-environmental laws in the future.
EPA agrees that the permit exemption of section 121(e)(4) exempts CERClA remedial and
-------
APPENDIX C
RESPONSES TO ARCO COMMENTS, APPENDIX C
INITIAL SCREENING OF REMEDIAL TECHNOLOGIES
AND PROCESS OPI'IONS .
1.
Comment: This comment notes that nine general response actions were identified
in the FS, and asks what the basis was for their identification.
Response: The nine general response actions were all the general response actions
that could be identified as being relevant to the problems enumerated for the
> operable unit. They were taken from a list of general response actions developed for
Superfund feasibility studies. The entire list is included m the left-hand column of
Table C-l. .
2.
Comment: This comment states that increasing the volumes of the ponds was not
included as a general response action. .
Response: The comment reflects a misunderstanding of the nature of general
response actions. As can be seen by looking at any of the general response actions
in Table C-I, they are very general in nature and not so specific as to identify
increasing the pond volumes for the purposes indicated. Instead, general response
actions are very general approaches that could be taken, without any specific
approach being preferred over others.
In fact, the more specific approach mentioned by ARCa is represented both in Table
C-I, where its general response actions would be 'Treatment" and "Containment" (of
the stream flows), and in Table 5-2, where it would be included in the general
response actions described as 'Treat the stream flow (in this case, improve or replace
the current treatment system) (Treatment)" and "Provide a settling basin to reduce
the flow rates and allow for settling (Containment)." Increasing the volume of Pond
3 to improve its treatment capacity and to use it as the settling basin would be
included within these concepts.
3.
Comment: This comment repeats an earlier comment regarding the difference
between the general response actions 'Treatment" and "In Situ Treatment".
Response: This comment is addressed in the Chapter 5 responses.
4.
Comment: This comment notes that coagulation and flocculation were screened au t
in Appendix C because of the volumes of water that would have to be treated. The
comment suggests that these process options should not have been screened out, and
that Pond 2 could be used as a polishing step with the use of such settling aids.
-------
Response: The agencies believe that it was reasonable to screen' out these two
process options. However, if ARCO wants to incorporate the use of such compounds
in the selected remedy, pay their costs over the long-term, and accept the shortening
of the life of Pond 3 that would be involved, MDHES may be able to agree to their
use. However, the use of Pond 2 as a treatment unit, even with the use of settling
aids, appears unacceptable without major modifications to the pond.
5.
Comment: This comment notes that "Chemical Sealants or Stabilizers" was screened
out, and that it should have been retained because such substances can be used
during remediation and for temporary stabilizations of surface contaminants.
Response: The agencies disagree with the comment. Process options useful only as
aids during the remediation are not the interest of this screening step, which is to.
identify process options that can.be used to attain the remedial action objectives over
the long-term. Process options that would only be used during the construction phase
of the remediation are generally screened out.
Use of chemical sealants or stabilizers to reduce mobility of surface contaminants
would not be useful in meeting any of the identified remedial action objectives.
6.
Comment: This comment notes that the process option "Organic
Agents/Polymers/Foams" was retained even though it is similar to "Chemical
Sealants/Stabilizers," which was screened out.
Response: The comment is correct. "Organic Agents/Polymers/Foams" should have
been screened out.
-------
APPENDIX D
RESPONSES TO ARCO COMMENTS, APPENDIX D
COST ANALYSIS FOR REMEDIAL ALTERNATIVES
ARCO prepared several pages of comments on the cost-effectiveness of the alternatives and
the CERClA requirements for cost-effectiveness, as well as several pages of comments on
the cost estimates included in the FS. The discussions of the CERClA process and cost-
effectiveness requirements do not require a response. EP A does not agree. with the
co~ents, and believes that the CERClA statute and the NCP provide adequate criteria
on the use of cost in remedy selection. .
The comments on the cost estimates general1y deal with minor inconsistencies in the backup
information for the estimates and requests for additional information on hauling distances
and other matters that determine the unit costs. None of the comments suggests that the
estimates for the alternatives fail to meet the objective standard of the RIfFS guidance
document, which is that the cost estimates should be within + 50 percent and -30 percent
of the actual cost that would be incurred to complete the remediation as scoped in the
conceptual design developed for each alternative. As noted in the responses to the Chapter
8 comments, this requirement does not apply to individual line items, or even to the
individual MSAs. It applies only to the overall cost estimates for the complete alternatives.
Many of the comments made by ARCO are more in the mode of comments that might be
made on an "engineer's estimate," which is a very different kind of cost estimate prepared
for a final design. The designs in the FS are conceptual only, as is appropriate when dealing
with several conceptual approaches. Once a general approach is agreed upon and a ROD
is signed for the operable unit, a specific detailed design can be developed, and accurate
cost estimates prepared. It would be a waste of resources to prepare detailed designs and
accurate cost estimates for all six alternatives in the FS.
~
Many of the ARCO's comments are either correct, or potentially correct. But, there is no
indication that modifying the cost estimates as suggested in the comments would affect the
selection of the remedy for the operable unit. Some of the comments, if implemented, would
increase the cost estimates, and some would decrease the estimates. The result would be
to make the estimates more accurate and reliable than the Superfund process requires, but
not to greatly change the overall cost estimates.
The MDHES included a sensitivity analysis in the FS to determine which cost items had the
potential to significantly affect the overall cost estimates. The conclusion that can be drawn
from the sensitivity analysis is that few of the line items have this potential. All of the
al ternatives include upgrading the berms, cleaning up the Mill-Willow Bypass, and upgrading
the pond treatment system. These are substantial costs that tend to dominate in the total
. .
-------
estimates. Minor discrepancies in estimating the costs for these items do not affect the
comparative analysis that the cost estimates are used for, because such discrepancies are
common to all of the action alternatives. Also, because these items are such a large portion
of the total costs for each of the alternatives, the 'potential for other cost items to put the
estimates outside the required + 50 percent to -30 percent range is limited; their small size
does not allow them to have a large overall effect., :, " ,
, ,
. The individual comments made by ARCa have been considered, but separate responses are
not included in this responsiveness summary.,',': ",
. -,::. ~... .
~ . ".. '.
':",' ,,-. . ....
.. : .r .. ~
'.' '\ ',:.' '" ':v
. . .j.
D - 2
";, . '"':.-'
. .. ~ .. .
. ..: ",' . ':1." ':
------- |