Second External Review Draft | EPA910-R-12-004Ba | April 2013
United States
Environmental Protection
Agency
An Assessment of Potential Mining Impacts
on Salmon Ecosystems of Bristol Bay, Alaska
Volume 1 - Main Report
^B
U.S. Environmental Protection Agency, Seattle, WA
www.epa.gov/bristolbay
Second External Review Draft - Do Not Cite or Quote
-------
DRAFT EPA910-R-12-004Ba
DO NOT CITE OR QUOTE April 2013
Second External Review Draft
www.epa.gov/bristolbay
AN ASSESSMENT OF POTENTIAL MINING IMPACTS ON
SALMON ECOSYSTEMS OF BRISTOL BAY, ALASKA
VOLUME 1—MAIN REPORT
NOTICE
THIS DOCUMENT IS AN EXTERNAL REVIEW DRAFT. It has not been formally released
by the U.S. Environmental Protection Agency and should not be construed to represent
Agency policy. It is being circulated for comment on its technical accuracy and policy
implications.
U.S. Environmental Protection Agency
Seattle, WA
-------
DISCLAIMER
This document is distributed solely for the purpose of pre-dissemination peer review under applicable
information quality guidelines. It has not been formally disseminated by the U.S. Environmental
Protection Agency (USEPA). It does not represent and should not be construed to represent any Agency
determination or policy. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
-------
This report evaluates the potential impacts of large-scale mining development on salmon and other fish
populations, wildlife, and Alaska Native cultures in the Nushagak River and Kvichak River watersheds of
Bristol Bay, Alaska. It is not an assessment of a specific mine proposal for development, nor does it
outline decisions made or to be made by the U.S. Environmental Protection Agency (USEPA). The
assessment was conducted as an ecological risk assessment and starts with a review and
characterization of the fisheries, wildlife, and Alaska Native cultures of the Bristol Bay watershed,
particularly the Nushagak and Kvichak River watersheds. We developed realistic mine scenarios that
include an open pit mine producing 0.25, 2.0, and 6.5 billion tons of ore and a 138-km transportation
corridor. Based on these mine scenarios, we conclude that mining would, at minimum, cause the loss of
spawning and rearing habitat for multiple salmonids (Pacific salmon, rainbow trout, and Dolly Varden).
The mine footprint in each of the three scenarios would likely result in the direct loss of 38, 90, and 145
km of streams and 5.0,12.4 and 19.4 km2 of wetlands, respectively. Water withdrawals for mine
operations would significantly diminish habitat quality in an additional 15, 26 and 54 km of streams.
Leakage of tailings and waste rock leachates during routine operations would result in instream copper
levels sufficient to cause direct effects on salmonids in 29 and 57 km of streams beyond the mine
footprint in the 2.0- and 6.5-billion-ton scenarios unless additional mitigation measures were taken.
These leakages would not be likely to cause direct effects in streams under the 0.25-billion-ton scenario.
Under a reasonable upper bound failure scenario for the wastewater treatment plant, copper
concentrations would be sufficient to cause direct effects on salmonid fish in 45,100, and 100 km of
streams, respectively, under each mine scenario. The transportation corridor would cross 53 streams
and rivers known or likely to support migrating and/or resident salmonids. At those road crossings,
culvert failures could inhibit fish migration and degrade habitat, truck accidents could spill industrial
chemicals, and runoff could reduce water quality. Failure of a tailings dam has a very low probability of
occurrence, buta spill of 20% of the tailings from a single tailings storage facility would destroy more
than 30 stream km, and more streams and rivers would have greatly degraded habitat for decades. A
spill of product concentrate slurry along the transportation corridor would result in toxicity to fish in
streams between the road and Iliamna Lake. Reductions in the populations of salmon would be expected
from these habitat losses and toxic effects, but cannot be quantified. These losses would adversely affect
the Alaska Native cultures and the wildlife of the region. The Nushagak River and Kvichak River
watersheds contain multiple sites under consideration for large-scale mining. Potential risks of mining
development on salmon and other fish populations are likely to increase as a result of the cumulative
impacts of multiple mines.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . April 2013
-------
Abstract i
Contents ii
Authors, Contributors, and Reviewers xxiv
Authors (listed alphabetically) xxiv
Contributors (listed alphabetically) xxiv
Reviewers of Internal Review Draft (listed alphabetically) xxv
Reviewers of External Review Draft (listed alphabetically) xxv
Photo Credits xxvi
Acknowledgements xxviii
Executive Summary ES-1
Scope of the Assessment ES-2
Ecological Resources ES-5
Alaska Native Cultures ES-8
Economics of Ecological Resources ES-9
Geological Resources ES-9
Mine Scenarios ES-10
Risks to Salmon and Other Fishes ES-14
Mine Footprint ES-14
Water Quality ES-15
Transportation Corridor ES-16
Tailings Dam Failure ES-17
Pipeline Failures ES-23
Common Mode Failures ES-24
Fish-Mediated Risks to Wildlife ES-24
Fish-Mediated Risks to Alaska Native Cultures ES-25
Cumulative Risks ES-25
Mitigation and Remediation ES-26
Summary of Uncertainties in Mine Design and Operation ES-27
Summary of Uncertainties and Limitations in the Assessment ES-28
Uses of the Assessment ES-29
Chapter 1. Introduction 1-1
1.1 Assessment Approach 1-2
1.2 Use of this Assessment 1-4
Chapter 2. Overview of Assessment 2-1
2.1 Structure 2-1
2.1.1 Data Used in the Assessment 2-2
2.1.2 Types of Evidence and Inference 2-3
2.2 Scope 2-5
2.2.1 Topical Scope 2-5
2.2.2 Spatial Scales 2-7
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment •• April 2013
-------
Chapters. Region 3-1
3.1 Physiographic Divisions 3-1
3.2 Hydrologic Landscapes 3-13
3.3 Groundwater Exchange and Flow Stability 3-13
3.4 Quantity and Diversity of Aquatic Habitats 3-18
3.4.1 Stream Reach Characterization: Attributes 3-18
3.4.2 Stream Reach Characterization: Results 3-26
3.5 Water Quality 3-28
3.5.1 Water Chemistry 3-28
3.5.2 Water Temperature 3-31
3.6 Seismicity 3-32
3.7 Existing Development 3-35
3.8 Climate Change 3-36
3.8.1 Climate Change Projections for the Bristol Bay Region 3-39
3.8.2 Potential Climate Change Effects 3-44
Chapter 4. Type of Development 4-1
4.1 Mineral Deposits and Mining in the Bristol Bay Watershed 4-1
4.2 Porphyry Copper Deposits and Mining Processes 4-3
4.2.1 Genesis of Porphyry Copper Deposits 4-3
4.2.2 Chemistry and Associated Risks of Porphyry Copper Deposits 4-5
4.2.3 Overview of the Mining Process 4-6
4.2.4 Timeframes 4-19
Chapter 5. Endpoints 5-1
5.1 Overview of Assessment Endpoints 5-1
5.2 Endpoint 1: Salmon and Other Fishes 5-2
5.2.1 Species and Life Histories 5-8
5.2.2 Distribution and Abundance 5-10
5.2.3 Economic Implications 5-23
5.2.4 Biological Complexity and the Portfolio Effect 5-24
5.2.5 Salmon and Marine-Derived Nutrients 5-26
5.2.6 Bristol Bay Fisheries in the Global Context 5-27
5.3 Endpoint 2: Wildlife 5-27
5.3.1 Life Histories, Distributions, and Abundances of Species 5-28
5.3.2 Recreational and Subsistence Activities 5-31
5.4 Endpoint 3: Alaska Natives 5-32
5.4.1 Alaska Native Populations 5-32
5.4.2 Subsistence and Alaska Native Cultures 5-32
Chapter 6. Mine Scenarios 6-1
6.1 Basic Elements of the Mine Scenarios 6-1
6.1.1 Location 6-4
6.1.2 Mining Processes 6-4
6.1.3 Transportation Corridor 6-16
6.2 Specific Mine Scenarios 6-19
6.2.1 Mine Scenario Footprints 6-20
6.2.2 Water Balance 6-23
6.3 Closure and Post-Closure Site Management 6-32
6.3.1 Mine Pit 6-32
6.3.2 Tailings Storage Facilities 6-33
6.3.3 Waste Rock 6-33
6.3.4 Water Management 6-34
6.3.5 Premature Closure 6-35
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment jjj April 2013
-------
6.4 Conceptual Models 6-36
6.4.1 Sources Evaluated 6-36
6.4.2 Stressors Evaluated 6-37
6.4.3 Endpoints Evaluated 6-42
6.4.4 Conceptual Model Diagrams 6-4
Chapter 7. Mine Footprint 7-1
7.1 Abundance and Distribution of Fishes in the Mine Scenario Watersheds 7-1
7.1.1 Fish Distribution 7-1
7.1.2 Spawning Salmon Abundance 7-12
7.1.3 Juvenile Salmon and Other Salmonid Abundance 7-14
7.2 Habitat Modification 7-15
7.2.1 Stream Segment Characteristics in the Mine Scenario Watersheds 7-15
7.2.2 Exposure: Habitat Lost to the Mine Scenario Footprints 7-16
7.2.3 Exposure-Response: Implications of Stream and Wetland Loss for Fish 7-26
7.2.4 Risk Characterization 7-31
7.2.5 Uncertainties 7-33
7.3 Streamflow Modification 7-33
7.3.1 Exposure: Streamflow 7-33
7.3.2 Exposure-Response: Streamflow 7-50
7.3.3 Risk Characterization 7-57
7.3.4 Uncertainties and Assumptions 7-58
7.4 Summary of Footprint Effects 7-60
Chapter 8. Water Collection, Treatment, and Discharge 8-1
8.1 Water Discharge Sources 8-1
8.1.1 Routine Operations 8-4
8.1.2 Wastewater Treatment Plant Failure 8-19
8.1.3 Post-Closure Wastewater Sources 8-21
8.1.4 Probability of Contaminant Releases 8-22
8.2 Chemical Contaminants 8-23
8.2.1 Exposure 8-23
8.2.2 Exposure-Response 8-25
8.2.3 Risk Characterization 8-35
8.2.4 Additional Mitigation of Leachates 8-56
8.2.5 Uncertainties 8-57
8.3 Temperature 8-61
8.3.1 Exposure 8-61
8.3.2 Exposure-Response 8-63
8.3.3 Risk Characterization 8-63
8.3.4 Uncertainties 8-64
Chapter 9. Tailings Dam Failure 9-1
9.1 Overview 9-1
9.1.1 Causes of Tailings Dam Failures 9-2
9.1.2 Probability of Tailings Dam Failures 9-7
9.1.3 Uncertainties 9-11
9.2 Material Properties 9-12
9.2.1 Tailings Dam Rockfill 9-12
9.2.2 Tailings Solids and Liquids 9-12
9.3 Tailings Dam Failure via Flooding and Overtopping 9-13
9.3.1 Hydrologic Characteristics 9-15
9.3.2 Sediment Transport and Deposition 9-19
9.3.3 Remediation 9-23
9.3.4 Uncertainties 9-23
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment T April 2013
-------
9.4 Scour, Sediment Deposition, and Turbidity 9-24
9.4.1 Exposure through Sediment Transport and Deposition 9-26
9.4.2 Exposure-Response 9-27
9.4.3 Risk Characterization 9-28
9.4.4 Uncertainties 9-31
9.5 Post-Spill Water Quality 9-32
9.5.1 Suspended Tailings Particles 9-32
9.5.2 Tailings Constituents 9-34
9.5.3 Weighing Lines of Evidence 9-49
9.6 Summary of Risks 9-51
9.6.1 Tailings Spill 9-51
9.6.2 Remediation of a Tailings Spill 9-51
Chapter 10. Transportation Corridor 10-1
10.1 Introduction 10-1
10.2 Fish Habitats and Populations along the Transportation Corridor 10-7
10.3 Potential Risks to Fish Habitats and Populations 10-14
10.3.1 Wetland Filling and Alteration 10-19
10.3.2 Stream Crossings 10-20
10.3.3 Chemical Contaminants in Stormwater Runoff 10-29
10.3.4 Fine Sediment 10-32
10.3.5 Dust 10-35
10.3.6 Invasive Species 10-37
10.4 Overall Risk Characterization for the Transportation Corridor 10-40
10.5 Uncertainties 10-40
Chapter 11. Pipeline Failures 11-1
11.1 Causes and Probabilities of Pipeline Failures 11-5
11.2 Potential Receiving Waters 11-7
11.3 Concentrate Pipeline Failure Scenarios 11-7
11.3.1 Sources 11-7
11.3.2 Exposure 11-9
11.3.3 Exposure-Response 11-12
11.3.4 Risk Characterization 11-12
11.3.5 Uncertainties 11-18
11.4 Return Water Pipeline Failure Scenarios 11-19
11.5 Diesel Pipeline Failure Scenarios 11-20
11.5.1 Sources 11-20
11.5.2 Exposure 11-21
11.5.3 Exposure-Response 11-23
11.5.4 Risk Characterization 11-27
11.5.5 Uncertainties 11-31
Chapter 12. Fish-Mediated Effects 12-1
12.1 Effects on Wildlife 12-1
12.2 Effects on Alaska Natives 12-6
12.2.1 Subsistence Use 12-8
12.2.2 Perception of Food Security 12-10
12.2.3 Economic Impacts 12-11
12.2.4 Social, Cultural, and Spiritual Impacts 12-12
12.2.5 Mitigation and Adaptation 12-15
12.3 Uncertainties 12-16
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment April 2013
-------
Chapter 13. Cumulative Effects of Large-Scale Mining 13-1
13.1 Overview 13-1
13.1.1 Definition of Cumulative and Induced Impacts 13-1
13.1.2 Vulnerability of Salmonids to Cumulative Impacts 13-6
13.1.3 Nature and Extent of Past, Present, and Future Impacts 13-6
13.2 Cumulative Impacts from Multiple Mines 13-7
13.2.1 Pebble South/PEB 13-9
13.2.2 Big Chunk South 13-22
13.2.3 Big Chunk North 13-23
13.2.4 Groundhog 13-24
13.2.5 AUDN/lliamna 13-25
13.2.6 Humble 13-26
13.2.7 Impacts of Multiple Mines 13-28
13.3 Cumulative Impacts from Induced Development 13-31
13.4 Potential Effects on Assessment Endpoints 13-32
13.4.1 Fishes 13-32
13.4.2 Wildlife and Alaska Native Cultures 13-33
13.5 Summary 13-35
Chapter 14. Integrated Risk Characterization 14-1
14.1 Overall Risk to Salmon and Other Fish 14-1
14.1.1 Routine Operation 14-1
14.1.2 Accidents and Failures 14-4
14.2 Overall Loss of Wetlands 14-11
14.3 Overall Fish-Mediated Risk to Wildlife 14-11
14.4 Overall Fish-Mediated Risk to Alaska Native Cultures 14-12
14.5 Summary of Uncertainties and Limitations in the Assessment 14-13
14.6 Summary of Uncertainties in Mine Design and Operation 14-16
14.7 Summary of Risks under the Mine Scenarios 14-16
14.8 Summary of Cumulative and Watershed-Scale Effects of Multiple Mines 14-17
Chapter 15. References 15-1
15.1 References by Chapters 15-1
15.1.1 Chapter 1—Introduction 15-1
15.1.2 Chapter 2—Overview of Assessment 15-2
15.1.3 Chapter 3—Region 15-2
15.1.4 Chapter 4—Type of Development 15-9
15.1.5 Chapter5—Endpoints 15-11
15.1.6 Chapter 6—Mine Scenarios 15-19
15.1.7 Chapter 7-Mine Footprint 15-22
15.1.8 Chapter 8—Water Collection, Treatment, and Discharge 15-29
15.1.9 Chapter 9—Tailings Dam Failure 15-36
15.1.10 Chapter 10—Transportation Corridor 15-44
15.1.11 Chapter 11-Pipeline Failures 15-51
15.1.12 Chapter 12-Fish-Mediated Affects 15-56
15.1.13 Chapter 13—Cumulative Effects of Large-Scale Mining 15-59
15.1.14 Chapter 14—Integrated Risk Characterization 15-64
15.2 CIS Base Map Citations 15-66
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . April 2013
-------
List of Appendices
Volume 2: Appendices A-D
Appendix A. Fishery Resources of the Bristol Bay Region
Appendix B. Non-Salmon Freshwater Fishes of the Nushagakand Kvichak River Drainages
Appendix C. Wildlife Resources of the Nushagak and Kvichak River Watersheds, Alaska
Appendix D. Ecological Knowledge and Cultures of the Nushagak and Kvichak Watersheds, Alaska
Volume 3: Appendices E-J
Appendix E. Bristol Bay Wild Salmon Ecosystem: Baseline Levels of Economic Activity and Values
Appendix F. Biological Characterization: Bristol Bay Marine Estuarine Processes, Fish, and Marine Mammal
Assemblages
Appendix G. Foreseeable Environmental Impact of Potential Road and Pipeline Development on Water
Quality and Freshwater Fishery Resources of Bristol Bay, Alaska
Appendix H. Geologic and Environmental Characteristics of Porphyry Copper Deposits with Emphasis on
Potential Future Development in the Bristol Bay Watershed, Alaska
Appendix I. Conventional Water Quality Mitigation Practices for Mine Design, Construction, Operation, and
Closure
Appendix J. Compensatory Mitigation and Large-Scale Hardrock Mining in the Bristol Bay Watershed
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment .. April 2013
-------
Tables
Table ES-1. Mine scenario parameters ES-11
Table ES-2. Summary of estimated stream lengths potentially affected under the three mine
scenarios, assuming routine operations ES-17
Table ES-3. Summary of estimated wetland areas potentially affected under the three mine
scenarios, assuming routine operations ES-18
Table ES-4. Probability and consequences of potential failures under the mine scenarios ES-18
Table 2-1. Spatial scales considered in the assessment 2-8
Table 3-1. Physiographic divisions (Wahrhaftig 1965) of the Nushagak and Kvichak River
watersheds 3-3
Table 3-2. Distribution of hydrologic landscapes in the Nushagak and Kvichak River watersheds 3-15
Table 3-3. Proportion of stream channel length within the Nushagak and Kvichak River
watersheds (Scale 2) classified according to stream size (based on mean annual
discharge in m3/s), channel gradient (%), and potential floodplain influence 3-31
Table 3-4. Mean background surface water characteristics of the mine scenario watersheds 3-32
Table 3-5. Examples of earthquakes in Alaska 3-33
Table 3-6. Average annual and seasonal air temperature for historical and projected periods
(SNAP 2012), and the difference between these periods across two spatial scales 3-39
Table 3-7. Average annual and seasonal precipitation for historical and projected periods, and the
difference between these periods across two spatial scales (SNAP 2012) 3-40
Table 3-8. Average annual water surplus for historical and projected periods, and the difference
between these periods across two spatial scales (SNAP 2012) 3-40
Table 4-1. Characteristics of past, existing, or potential large mines in Alaska 4-2
Table 4-2. Global grade and tonnage summary statistics for porphyry copper deposits 4-3
Table 5-1. Fish species reported in the Nushagak and Kvichak River watersheds 5-3
Table 5-2. Life history, habitat characteristics, and total surveyed occupied stream length for
Bristol Bay's five Pacific salmon species in the Nushagak and Kvichak River
watersheds 5-9
Table 5-3. Mean annual commercial harvest (number of fish) by Pacific salmon species and
Bristol Bay fishing district, 1990 to 2009 5-11
Table 5-4. Summary of regional economic expenditures based on salmon ecosystem services 5-23
Table 5-5. Life-history variation within the Bristol Bay sockeye salmon populations 5-25
Table 6-1. Summary of scenarios considered in the assessment 6-2
Table 6-2. Mine scenario parameters 6-10
Table 6-3. Summary of water balance flows (million m3/year) during operations for the three mine
scenarios 6-15
Table 6-4. Characteristics of pipelines in the mine scenarios 6-19
Table 6-5. Estimated areas for individual mine components under the Pebble 0.25 scenario 6-21
Table 6-6. Estimated areas for individual mine components under the Pebble 2.0 scenario 6-22
Table 6-7. Estimated areas for individual mine components under the Pebble 6.5 scenario 6-23
Table 6-8. Summary of water balance flows (million m3/year) during post-closure period for all
mine scenarios 6-34
Table 6-9. Stressors considered in the assessment and their relevance to the assessment's
primary endpoint (salmonids) and USEPA's regulatory authority 6-38
Table 6-10. Screening benchmarks for metals with no national ambient water quality criteria 6-39
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
VIM
April 2013
-------
Table 7-1. Highest reported index spawner counts in the mine scenario watersheds for each year,
2004 to 2008 7-13
Table 7-2. Average 2008 index spawner counts by stream reach 7-14
Table 7-3. Highest index counts of selected stream-rearing fish species from mainstem habitats 7-15
Table 7-4. Distribution of stream channel length classified by channel size (based on mean annual
flow in m3/s), channel gradient (%), and potential floodplain influence for streams and
rivers in the mine scenario watersheds 7-16
Table 7-5. Stream length (km) eliminated, blocked, or dewatered by the mine footprints under the
Pebble 0.25, 2.0, and 6.5 scenarios 7-24
Table 7-6. Distribution of stream channel length classified by channel size (based on mean annual
discharge in m3/s), channel gradient (%), and potential floodplain influence for streams
under the Pebble 6.5 mine footprint 7-25
Table 7-7. Wetland areas (km2) eliminated, blocked, or dewatered by the mine footprints under
the Pebble 0.25, 2.0, and 6.5scenarios 7-25
Table 7-8. Total documented anadromous fish stream length and stream length documented to
contain different fish species in the mine scenario watersheds 7-26
Table 7-9. Stream gages and related characteristics for the South and North Fork Koktuli Rivers
and Upper Talarik Creek 7-37
Table 7-10. Measured mean monthly pre-miningflow rates (m3/s) and estimated mean monthly
flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for gages along
the South Fork Koktuli River 7-39
Table 7-11. Measured mean monthly pre-miningflow rates (m3/s) and estimated mean monthly
flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for gages along
the North Fork Koktuli River 7-39
Table 7-12. Measured mean monthly pre-miningflow rates (m3/s) and estimated mean monthly
flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for gages along
Upper Talarik Creek 7-40
Table 7-13. Measured minimum monthly pre-miningflow rates (m3/s) and estimated minimum
monthly flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for
gages along the South Fork Koktuli River 7-40
Table 7-14. Measured minimum monthly pre-miningflow rates (m3/s) and estimated minimum
monthly flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for
gages along the North Fork Koktuli River 7-41
Table 7-15. Measured minimum monthly pre-miningflow rates (m3/s) and estimated minimum
monthly flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for
gages along Upper Talarik Creek 7-41
Table 7-16. Pre-mining watershed areas, mine footprint areas, and flows in the mine scenario
watersheds, for the Pebble 0.25 mine scenario 7-42
Table 7-17. Pre-mining watershed areas, mine footprint areas, and flows in the mine scenario
watersheds, for the Pebble 2.0 mine scenario 7-43
Table 7-18. Pre-mining watershed areas, mine footprint areas, and flows in the mine scenario
watersheds, for the Pebble 6.5 mine scenario 7-44
Table 7-19. Estimated change in streamflow (%) and subsequent stream lengths affected (km) in
the mine scenario watersheds under the Pebble 0.25, Pebble 2.0, and Pebble 6.5
scenarios 7-49
Table 7-20. Estimated change in streamflow (%) at selected stream gages 7-60
Table 8-1. Effluent and receiving water flows at each gage under the Pebble 0.25 scenario. All
values are presented in m3/yr 8-5
Table 8-2. Effluent and receiving water flows at each gage under the Pebble 2.0 scenario. All
values are presented in m3/yr 8-7
Table 8-3. Effluent and receiving water flows at each gage under the Pebble 6.5 scenario. All
values are presented in m3/yr 8-9
Table 8-4. Aquatic toxicological screening of tailings supernatant against acute (criterion
maximum concentration) and chronic (criterion continuous concentration) water quality
criteria or benchmark values 8-13
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
April 2013
-------
Table 8-5. Aquatic toxicological screening of tailings humidity cell leachates against acute
(criterion maximum concentration) and chronic (criterion continuous concentration)
water quality criteria or benchmark values 8-14
Table 8-6. Composition of test leachate from Tertiary waste rock in the Pebble deposit and
quotients relative to the acute (criterion maximum concentration) and chronic (criterion
continuous concentration) water quality criteria or benchmark values 8-15
Table 8-7. Composition of test leachate from Pebble East pre-Tertiary waste rock and quotients
relative to acute (criterion maximum concentration) and chronic (criterion continuous
concentration) water quality criteria 8-16
Table 8-8. Composition of test leachate from Pebble West pre-Tertiary waste rock and quotients
relative to acute (criterion maximum concentration) and chronic (criterion continuous
concentration) water quality criteria 8-17
Table 8-9. Estimated concentration of contaminants of concern in effluents from the wastewater
treatment plant, tailings, non-acid-generating waste rock, and potentially acid
generating waste rock 8-18
Table 8-10. Mean and coefficient of variation of background surface water characteristics of the
mine scenario watersheds, 2004-2008 8-25
Table 8-11. Results of applying the biotic ligand model to mean water chemistries in the mine
scenario watersheds to derive copper criteria specific to receiving waters 8-27
Table 8-12. Results of applying the biotic ligand model to mean water chemistries in waste rock
leachates to derive effluent-specific copper criteria 8-28
Table 8-13. Rainbow trout site-specific acute and chronic copper toxicity derived by applying the
biotic ligand model to mean water chemistries in the mine scenario watersheds 8-28
Table 8-14. Rainbow trout site-specific benchmarks for sensory effects 8-29
Table 8-15. Hardness-dependent acute water quality criteria (criterion maximum concentration)
and chronic water quality criteria (criterion continuous concentration) for the three
potential receiving streams under the mine scenarios 8-32
Table 8-16. Estimated concentrations of contaminants of concern and associated risk quotients for
the Pebble 6.5 mine scenario, at locations in the mine scenario watersheds 8-39
Table 8-17. Estimated concentrations of contaminants of concern and associated risk quotients for
the Pebble 6.5 mine scenario assuming wastewater treatment plant failure, at
locations in the mine scenario watersheds 8-40
Table 8-18. Estimated total toxicity of metals of concern for each mine scenario, under routine
operations and with wastewater treatment plant failure, at locations in the mine
scenario watersheds 8-41
Table 8-19. Background copper concentrations and, for each mine scenario, copper concentrations
in contributing loads and ambient waters and associated risk quotients for routine
operations 8-42
Table 8-20. Background copper concentrations and, for each mine scenario, copper concentrations
in contributing loads and ambient waters and associated risk quotients under
wastewater treatment plant failure 8-44
Table 8-21. Description of stream reaches affected in the mine scenarios and sources of the
concentration estimates applied to the stream reaches 8-46
Table 8-22. Copper concentration benchmarks exceeded in ambient waters in each reach and for
each mine scenario during routine operations 8-48
Table 8-23. Copper concentration benchmarks exceeded in ambient waters in each reach and for
each mine scenario duringa wastewater treatment plant failure 8-49
Table 8-24. Length of stream (km) in which copper concentrations would exceed levels sufficient to
+cause toxic effects, under routine operations and wastewater treatment plant failure,
for each of the three mine scenarios 8-56
Table 8-25. Copper concentrations (mg/L) in waste rock leachates for two water quality models 8-60
Table 9-1. Number and causes of tailings dam failures at active and inactive tailings dams 9-4
Table 9-2. Summary of Alaska's classification of potential hazards of dam failure 9-8
Table 9-3. HEC-RAS model results for the Pebble 0.25TSFdam failure analysis 9-17
Table 9-4. HEC-RAS model results for the Pebble 2.0 TSF dam failure analysis 9-18
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
April 2013
-------
Table 9-5. Tailings potentially mobilized and deposited during Pebble 0.25 and Pebble 2.0 dam
failures at TSF1 9-22
Table 9-6. Sediment size distributions surveyed at the South and North Fork Koktuli Rivers, Upper
Talarik Creek, and 77 wadeable stream sites in the Nushagak and Kvichak River
watersheds 9-25
Table 9-7. Aquatic toxicological screening of tailings supernatant against acute water quality
criteria (CMC) and chronic water quality criteria (CCC) 9-37
Table 9-8. Aquatic toxicological screening of tailings humidity cell leachates against acute water
quality criteria (CMC) and chronic water quality criteria (CCC) 9-38
Table 9-9. Comparison of mean metal concentrations of tailings (Appendix H) to threshold effect
concentration and probable effect concentration values for fresh water sediments and
sums of the quotients (XTU) 9-40
Table 9-10. Results of applying the biotic ligand model to mean water chemistries in tailings
leachates and supernatants to derive effluent-specific copper criteria 9-44
Table 9-11. Summary of evidence concerning risks to fish from a tailings dam failure 9-50
Table 10-1. Proportion of stream channel length in stream watersheds intersected by the
transportation corridor (Scale 5) classified according to stream size (based on mean
annual discharge in m3/s), channel gradient (%), and potential floodplain influence 10-8
Table 10-2. Average number of spawning adult sockeye salmon at locations near the transportation
corridor 10-10
Table 10-3. Proximity of transportation corridor to National Hydrography Dataset streams 10-16
Table 10-4. Proximity of transportation corridor to National Wetlands Inventory wetlands 10-17
Table 10-5. Proximity of transportation corridor to water (within 200 m of National Hydrology
Dataset streams or National Wetland Inventory wetlands) 10-18
Table 10-6. Road-stream crossings along the transportation corridor, upstream lengths of streams
of different sizes likely to support salmonids (based on stream gradients of less than
12%), and downstream length to Iliamna Lake 10-21
Table 10-7. Lengths downstream of road-stream crossings, by stream size 10-24
Table 10-8. Lengths of different stream sizes that occur upstream of road-stream crossings and are
likely to support salmonids (based on stream gradients of less than 12%) 10-25
Table 11-1. Studies that examined pipeline failure rates 11-6
Table 11-2. Conditions of concentrate pipeline spill to Chinkelyes Creek and Knutson Creek 11-9
Table 11-3. Comparison of mean metal concentrations in copper concentrate from the Aitik
(Sweden) porphyry copper mine (Appendix H) to threshold effect concentration and
probable effect concentration values for fresh water 11-10
Table 11-4. Aquatic toxicological screening of leachates from Aitik (Sweden) mine copper
concentrate (Appendix H) based on acute and chronic benchmarks (water quality
criteria or equivalent values) and quotients of concentrations divided by benchmark
values 11-14
Table 11-5. Summary of evidence concerning risks to fish from a product concentrate spill 11-18
Table 11-6. Conditions of return water pipeline spill to Chinkelyes and Knutson Creeks 11-20
Table 11-7. Conditions of diesel pipeline spill to Chinkelyes and Knutson Creeks 11-21
Table 11-8. Toxicity of diesel fuel to freshwater organisms in laboratory tests 11-25
Table 11-9. Cases of diesel spills into streams and the diesel pipeline failure scenarios 11-26
Table 11-10. Summary of evidence concerning risks to fish from a diesel spill 11-30
Table 13-1. Mining prospects, in addition to the Pebble deposit, with more than minimal recent
exploration in the Nushagak and Kvichak River watersheds 13-3
Table 13-2. Waters, fish, and subsistence uses potentially affected by a mine at the Pebble
South/PEB prospect 13-11
Table 13-3. Waters, fishes, and subsistence uses potentially affected by a mine at the Big Chunk
South prospect 13-12
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
XI
April 2013
-------
Table 13-4 Waters, fishes, and subsistence uses potentially affected by a mine at the Big Chunk
North prospect 13-13
Table 13-5. Waters, fishes, and subsistence uses potentially affected by a mine at the Groundhog
prospect 13-14
Table 13-6. Waters, fishes, and subsistence uses potentially affected by a mine at the
AUDN/lliamna prospect 13-16
Table 13-7. Waters, fishes, and subsistence uses potentially affected by a mine at the Humble
prospect 13-18
Table 13-8. Streams, water bodies, and wetlands potentially eliminated by additional large-scale
mines in the Nushagak and Kvichak River watersheds 13-21
Table 14-1. Summary of probability and consequences of potential failures under the mine
scenarios 14-5
Table 14-2. Summary of estimated effects on streams under the three mine scenarios, assuming
routine operations 14-17
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment .. April 2013
-------
Figures
Figure ES-1. The Nushagak and Kvichak River watersheds of Bristol Bay 3
Figure ES-2. Reported salmon (sockeye, Chinook, coho, pink, and chum combined) distribution in
the South and North Fork Koktuli Rivers and Upper Talarik Creek watersheds 6
Figure ES-3. Total sockeye salmon run sizes by (A) region and (B) watershed in the Bristol Bay region 7
Figure ES-4. The mine scenario footprints for the three scenarios evaluated in the assessment:
Pebble 0.25 (0.25 billion tons of ore), Pebble 2.0 (2.0 billion tons of ore), and Pebble
6.5 (6.5 billion tons of ore) 12
Figure ES-5. The transportation corridor area, comprising 27 subwatersheds in the Kvichak River
watershed thatdrain to Iliamna Lake 13
Figure ES-6. Streams and wetlands lost (eliminated, blocked, or dewatered) under the Pebble 6.5
scenario 19
Figure ES-7. Salmon, Dolly Varden, and rainbow trout distribution along the transportation corridor 20
Figure ES-8. Height of the dam at TSF 1 in the Pebble 2.0 and Pebble 6.5 scenarios relative to U.S.
landmarks 21
Figure 2-1. Conceptual model illustrating sources, stressors, and responses potentially associated
with large-scale mine development in the Bristol Bay watershed 2-6
Figure 2-2. Five spatial scales considered in this assessment 2-9
Figure 2-3. The Bristol Bay watershed (Scale 1), comprisingthe Togiak, Nushagak, Kvichak,
Naknek, Egegik, and Ugashik River watersheds and the North Alaska Peninsula 2-10
Figure 2-4. The Nushagak and Kvichak River watersheds (Scale 2) 2-11
Figure 2-5. The mine scenario watersheds—South Fork Koktuli River, North Fork Koktuli River, and
Upper Talarik Creek—within the Nushagak and Kvichak River watersheds (Scale 3) 2-12
Figure 2-6. The mine footprints for the three scenarios evaluated in the assessment (Scale 4) 2-13
Figure 2-7. The transportation corridor area (Scale 5), comprising 27 subwatersheds in the Kvichak
River watershed thatdrain to Iliamna Lake 2-14
Figure 3-1. Hydrologic landscapes within the Nushagak and Kvichak River watersheds, as defined
by physiographic division and climate class 3-4
Figure 3-2. Distribution of mean annual precipitation (mm) across the Nushagak and Kvichak River
watersheds, 1971 to 2000 (SNAP 2012) 3-5
Figure 3-3. Generalized geology of the Bristol Bay watershed (adapted from Selkregg 1974) 3-6
Figure 3-4. Occurrence of permafrost in the Bristol Bay watershed (adapted from Selkregg 1974) 3-7
Figure 3-5. Dominant soils in the Bristol Bay watershed (adapted from Selkregg 1974) 3-8
Figure 3-6. Erosion potential in the Bristol Bay watershed (adapted from Selkregg 1974) 3-9
Figure 3-7. Dominant vegetation in the Bristol Bay watershed (adapted from Selkregg 1974) 3-10
Figure 3-8. Physiographic divisions of the Nushagak and Kvichak River watersheds of Bristol Bay 3-11
Figure 3-9. Groundwater resources in the Bristol Bay watershed (adapted from Selkregg 1974) 3-16
Figure 3-10. Mean monthly runoff for selected streams and rivers in the Nushagak and Kvichak
River watersheds 3-17
Figure 3-11. Examples of different stream size and gradient classes in the Nushagak and Kvichak
River watersheds 3-22
Figure 3-12. Valley gradient classes in the Nushagak and Kvichak River watersheds 3-23
Figure 3-13. Likelihood of floodplain connectivity, as measured by the percent flatland in lowland
areas, for the Nushagak and Kvichak River watersheds 3-29
Figure 3-14. Stream size classes in the Nushagak and Kvichak River watersheds as determined by
mean annual flow 3-30
Figure 3-15. Seismic activity in southwestern Alaska 3-34
Figure 3-16. Mean annual temperature across the Bristol Bay watershed under (A) historical
conditions (1971 to 2000) and (B) the A2 emissions scenario (2071 to 2099), and (C)
the temperature change between these two climate scenarios (SNAP 2012) 3-41
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~jjj April 2013
-------
Figure 3-17. Mean annual precipitation across the Bristol Bay watershed under (A) historical
conditions (1971 to 2000) and (B) the A2 emissions scenario (2071 to 2099), and (C)
the precipitation change between these two climate scenarios (SNAP 2012) 3-42
Figure 3-18. Mean annual water surplus (precipitation minus evapotranspiration) across the Bristol
Bay watershed under (A) historical conditions (1971 to 2000) and (B) the A2 emissions
scenario (2071 to 2099), and (C) the water surplus change between these two climate
scenarios (SNAP 2012) 3-43
Figure 4-1. Porphyry copper deposits around the world 4-4
Figure 4-2. Neutralizing potential at the Bingham Canyon porphyry copper deposit, Utah 4-7
Figure 4-3. Simplified schematic of mined material processing 4-15
Figure 4-4. Cross-sections illustrating (A) upstream, (B) downstream, and (C) centerline tailings
dam construction 4-17
Figure 5-1. Approximate extents of popular Chinook and sockeye salmon recreational fisheries in
the vicinity of the Nushagakand Kvichak River watersheds 5-5
Figure 5-2. Subsistence harvest and harvest effort areas for salmon and other fishes within the
Nushagakand Kvichak River watersheds 5-6
Figure 5-3. Diversity of Pacific salmon species production in the Nushagak and Kvichak River
watersheds 5-14
Figure 5-4. Reported sockeye salmon stream distribution in the Nushagak and Kvichak River
watersheds 5-15
Figure 5-5. Reported Chinook salmon distribution in the Nushagak and Kvichak River watersheds 5-16
Figure 5-6. Reported coho salmon distribution in the Nushagakand Kvichak River watersheds 5-17
Figure 5-7. Reported chum salmon distribution in the Nushagakand Kvichak River watersheds 5-18
Figure 5-8. Reported pink salmon distribution in the Nushagakand Kvichak River watersheds 5-19
Figure 5-9. Total sockeye salmon run sizes by (A) region and (B) watershed within the Bristol Bay
region 5-20
Figure 5-10. Reported rainbow trout occurrence and distribution in the Nushagak and Kvichak River
watersheds 5-21
Figure 5-11. Reported Dolly Varden occurrence and distribution in the Nushagak and Kvichak River
watersheds 5-22
Figure 5-12. Subsistence use intensity for salmon, other fishes, wildlife, and waterfowl within the
Nushagakand Kvichak River watersheds 5-37
Figure 6-1. Footprint of the Pebble 0.25 scenario 6-5
Figure 6-2. Footprint of the Pebble 2.0 scenario 6-6
Figure 6-3. Footprint of the Pebble 6.5 scenario 6-7
Figure 6-4. Height of the dam atTSF 1 relative to U.S. landmarks 6-11
Figure 6-5. Water management and water balance components for the three mine scenarios 6-14
Figure 6-6. Transportation corridor connecting the Pebble deposit area to Cook Inlet 6-17
Figure 6-7. Hydraulic conductivity in the Pebble deposit area 6-26
Figure 6-8. Water flow schematic for the Pebble 0.25 scenario 6-28
Figure 6-9. Water flow schematic for the Pebble 2.0 scenario 6-29
Figure 6-10. Water flow schematic for the Pebble 6.5 scenario 6-30
Figure 6-11. Approximate locations of stream gages and wastewater treatment plant discharges
represented in Figures 6-8 through 6-10 6-31
Figure 6-12. Conceptual model illustrating potential effects of routine mine construction and
operation on physical habitat 6-43
Figure 6-13. Conceptual model illustrating potential effects of routine mine construction and
operation on water chemistry 6-44
Figure 6-14. Conceptual model illustrating potential effects of unplanned events on physical habitat
and water chemistry 6-45
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~T April 2013
-------
Figure 7-1. Conceptual model illustrating potential linkages between sources associated with the
mine scenario footprints, changes in physical habitat, and fish endpoints 7-3
Figure 7-2. Reported sockeye salmon distribution in the mine scenario watersheds 7-5
Figure 7-3. Reported coho salmon distribution in the mine scenario watersheds 7-6
Figure 7-4. Reported Chinook salmon distribution in the mine scenario watersheds 7-7
Figure 7-5. Reported chum salmon distribution in the mine scenario watersheds 7-8
Figure 7-6. Reported pink salmon distribution in the mine scenario watersheds 7-9
Figure 7-7. Reported Dolly Varden occurrence in the mine scenario watersheds 7-10
Figure 7-8. Reported rainbow trout occurrence in the mine scenario watersheds 7-11
Figure 7-9. Cumulative frequency of stream channel length classified by mean annual flow (m3/s),
reach gradient (%), and floodplain potential (measured as %flatland in lowland) for the
mine scenario watersheds (Scale 3) versus the Nushagak and Kvichak River
watersheds (Scale 2) 7-17
Figure 7-10. Streams and wetlands lost (eliminated, blocked, or dewatered) under the Pebble 0.25
scenario 7-18
Figure 7-11. Streams and wetlands lost (eliminated, blocked, or dewatered) under the Pebble 2.0
scenario 7-19
Figure 7-12. Streams and wetlands lost (eliminated, blocked, or dewatered) under the Pebble 6.5
scenario 7-20
Figure 7-13. Cumulative frequency of stream channel length classified by mean annual flow (m3/s),
reach gradient (%), and floodplain potential (measured as %flatland in lowland) for the
mine footprints (Scale 4) versus the Nushagak and Kvichak River watersheds (Scale 2) 7-23
Figure 7-14. Comparison of MIKE-SHE modeled groundwater upwellingareas and inferred upwelling
areas based on PLP (2011) aerial surveys 7-30
Figure 7-15. Stream segments in the mine scenario watersheds showing flow changes (%)
associated with the Pebble 0.25 footprint 7-34
Figure 7-16. Stream segments in the mine scenario watersheds showing flow changes (%)
associated with the Pebble 2.0 footprint 7-35
Figure 7-17. Stream segments in the mine scenario watersheds showing flow changes (%)
associated with the Pebble 6.5 footprint 7-36
Figure 7-18. Monthly mean streamflows for stream gages in the (A) South Fork Koktuli River,
(B) North Fork Koktuli River, and (C) Upper Talarik Creek watersheds, based on water
years 2004 through 2010 7-45
Figure 7-19. Monthly mean pre-miningstreamflow for South Fork Koktuli River gage SK100F (bold
solid line) illustrating 10 and 20%sustainability boundaries (gray lines) and projected
monthly mean streamflows under the Pebble 0.25 scenario (dashed line) 7-52
Figure 8-1. Conceptual model illustrating the pathways linking water treatment, discharge, fate,
and effects 8-3
Figure 8-2. Processes involved in copper uptake as defined in the biotic ligand model (USEPA
2007) 8-27
Figure 8-3. Comparison of copper concentrations in leachates and background water to state
hardness-based acute (CMC) and chronic (CCC) water quality criteria for copper 8-37
Figure 9-1. Conceptual model illustrating potential pathways linking tailings storage facility failure
and effects on fish endpoints 9-5
Figure 9-2. Annual probability of dam failure due to slope failure vs. factor of safety (modified from
Silvaetal.2008) 9-11
Figure 9-3. Representative particle size distributions for tailings solids (bulk and pyritic tailings)
and tailings dam rockfill 9-13
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment April 2013
-------
Figure 10-1. Streams, wetlands, ponds, and lakes along the transportation corridor 10-3
Figure 10-2. High-impact areas along the transportation corridor 10-4
Figure 10-3. Conceptual model showing potential pathways linking the transportation corridor and
related sources tostressors and assessment endpoints 10-5
Figure 10-4. Cumulative frequency of stream channel length classified by mean annual flow (m3/s),
reach gradient (%), and floodplain potential (measured as %flatland in lowland) for
watersheds intersected by the transportation corridor (Scale 5) versus the Nushagak
and Kvichak River watersheds (Scale 2) 10-9
Figure 10-5. Location of sockeye salmon surveys and number of spawners observed along the
transportation corridor 10-11
Figure 10-6. Reported salmon, Dolly Varden, and rainbow trout distribution along the transportation
corridor 10-13
Figure 11-1. Conceptual model illustrating potential stressors and effects resulting from a
concentrate pipeline failure 11-2
Figure 11-2. Conceptual model illustrating potential stressors and effects resulting from a return
water pipeline failure 11-3
Figure 11-3. Conceptual model illustrating potential stressors and effects resulting from a diesel
pipeline failure 11-4
Figure 12-1. Conceptual model illustrating potential effects on wildlife resulting from effects on fish 12-3
Figure 12-2. Conceptual model illustrating potential effects on Alaska Native cultures resulting from
effects on fish 12-4
Figure 13-1. Claim blocks with more than minimal recent exploration in the Nushagak and Kvichak
River watersheds 13-4
Figure 13-2. Conceptual model illustrating potential cumulative effects of multiple large-scale
mines 13-5
Figure 13-3. Location of claim blocks in relation to subsistence use intensity for salmon, other
fishes, wildlife, and waterfowl in the Nushagak and Kvichak River watersheds 13-34
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . April 2013
-------
Boxes
Box 1-1. Stakeholder Involvement in the Assessment 1-6
Box 2-1. Conceptual Models 2-2
Box 2-2. Salmonid Fishes in the Bristol Bay Watershed 2-8
Box 3-1. Methods for Characterizing Valley Gradient 3-20
Box 3-2. Methods for Characterizing Mean Annual Flow 3-25
Box 3-3. Methods for Characterizing Percent Flatland in Lowland 3-27
Box 3-4. Methods for Climate Change Projections 3-38
Box 4-1. Reducing Mining's Impacts 4-8
Box 4-2. Permitting Large Mine Projects in Alaska 4-9
Box 4-3. Financial Assurance 4-10
Box 4-4. Block Caving and Subsidence 4-12
Box 4-5. Chemicals Used in the Flotation Process 4-14
Box 5-1. Subsistence Use Methodology 5-7
Box 5-2. Commercial Fisheries Management in the Bristol Bay Watershed 5-12
Box 5-3. Testimony on the Importance of Subsistence Use 5-39
Box 6-1. Cumulative Footprint of a Large-Scale Porphyry Copper Mine 6-3
Box 6-2. Mine Pit Drawdown Calculations 6-25
Box 7-1. Calculation of Streams and Wetlands Affected by Mine Scenario Footprints 7-21
Box 7-2. Compensatory Mitigation 7-32
Box 8-1. An Accidental Tailings Water Release: Nixon Fork Mine, Alaska, Winter 2012 8-20
Box 8-2. Potential Failures of Reverse-Osmosis Wastewater Treatment Plants 8-20
Box 8-3. Use of Risk Quotients to Assess Toxicological Effects 8-36
Box 8-4. The Fraser River 8-54
Box 9-1. Examples of Historical Tailings Dam Failures 9-3
Box 9-2. Selecting Earthquake Characteristics for Design Criteria 9-9
Box 9-3. Modeling the Probable Maximum Flood Hydrograph atTSF 1 9-14
Box 9-4. Modeling Hydrologic Characteristics of Tailings Dam Failures 9-15
Box 9-5. Using Hydrologic Models to Estimate Tailings Deposition After a Tailings DAM Failure 9-21
Box 9-6. Background on Relevant Analogous Tailings Spill Sites 9-35
Box 10-1. Calculation of Stream Lengths and Wetland Areas Affected by Transportation Corridor
Development 10-15
Box 10-2. Culvert Mitigation 10-29
Box 10-3. Stormwater Runoff and Fine Sediment Mitigation 10-33
Box 10-4. Mitigation for Invasive Species 10-39
Box 10-5. Likely Effectiveness of Mitigation Measures 10-42
Box 12-1. Testimony on Potential Effects of Mining on Alaska Native Cultures 12-11
Box 13-1. Methods for Estimating Impacts of Other Mines 13-9
Box 13-2. Examples of Mine Characterization Errors 13-31
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~. April 2013
-------
Box 14-1. Failure Probabilities 14-6
Box 14-2. Climate Change and Potential Risks of Large-Scale Mining 14-15
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ... April 2013
-------
Acronyms and Abbreviations
AAC
AD EC
ADF&G
ADNR
ADOT
AFFI
AP
APDES
ASME
AVS
AWC
BBAP
BLM
BMP
CCC
CFR
CH
CIBB
CMC
CPU
CWA
DBB
DEM
EBD
ECso
E-R
ERA
FA
FERC
FK
FR
FS
GCM
GIS
GMU
HEC-HMS
HEC-RAS
HUC
IA
1C
IC2o
ICso
IFIM
IGTT
kg CaCOs/metric ton
LCso
LFP
MCE
micrograms per liter
Alaska Administrative Code
Alaska Department of Environmental Conservation
Alaska Department of Fish and Game
Alaska Department of Natural Resources
Alaska Department of Transportation and Public Facilities
Alaska Freshwater Fish Inventory
acid-generation potential
Alaska Pollutant Discharge Elimination System
American Society of Mechanical Engineers
acid volatile sulfides
Anadromous Waters Catalog
Bristol Bay Area Plan for State Lands
biotic ligand model
best management practice
criterion continuous concentration
Code of Federal Regulations
channel
Cook Inlet-to-Bristol Bay
criterion maximum concentration
Climate Research Unit
Clean Water Act
Dillingham/Bristol Bay
digital elevation model
Environmental Baseline Document 2004 through 2008
effective concentration
exposure-response relationship
ecological risk assessment
fish avoidance
Federal Energy Regulatory Commission
fish kill
fish reproduction
fish sensory
global climate model
geographic information system
Game Management Unit
Hydrologic Engineering Center's Hydrologic Modeling System
Hydrologic Engineering Center's River Analysis System
hydrologic unit code
invertebrate acute
invertebrate chronic
20% inhibitory concentration
median inhibitory concentration
Instream Flow Incremental Methodology
Intergovernmental Technical Team
kilograms of calcium carbonate per metric ton of waste material
median lethal concentration
leftfloodplain
maximum credible earthquake
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
XIX
April 2013
-------
MDE
MDN
MOA
NA
NAG
NANA
NDM
NED
NHD
NNP
NP
NPR
NWI
QBE
OHW
PAG
PEC
PEL
PET
PHABSIM
PLP
PMF
PMP
PRISM
Reclamation
RFP
SCADA
SEM
SNAP
SWATP
SWPPPs
IDS
TEC
TEL
TSF
USAGE
USEPA
USFWS
USGS
WWTP
maximum design earthquake
marine-derived nutrients
memorandum of agreement
not applicable
non-acid-generating
NANA Regional Corporation, Inc.
Northern Dynasty Minerals
National Elevation Dataset
National Hydrography Dataset
net neutralization potential
neutralization potential
neutralizing potential ratio
National Wetlands Inventory
operating basis earthquake
ordinary high water
potentially acid-generating
probable effect concentration
probable effect level
potential evapotranspiration
Physical Habitat Simulation
Pebble Limited Partnership
probable maximum flood
probable maximum precipitation
Parameter-elevation Regressions on Independent Slopes Model
Bureau of Reclamation
right floodplain
supervisory control and data acquisition
simultaneously extracted metals
Scenarios Network for Alaska and Arctic Planning
Southwest Alaska Transportation Plan
stormwater pollution prevention plans
total dissolved solids
threshold effect concentration
threshold effect level
tailings storage facility
U.S. Army Corps of Engineers
U.S. Environmental Protection Agency
U.S. Fish and Wildlife Service
U.S. Geological Survey
wastewater treatment plant
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
xx
April 2013
-------
Units of Measure
|jg microgram
°C degrees Celsius
cm centimeter
g gram
ha hectare
kg kilogram
km kilometer
km2 square kilometers
km-yr kilometer-year
L liter
m meter
m2 square meter
m3 cubic meter
mg milligram
mm millimeter
s second
S.U. standard units
t ton
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . April 2013
-------
Unit of Measure Conversion Chart
Metric
1 iJg(microgram)
1 mg (milligram)
Ig(gram)
1 kg (kilogram)
1 metric ton
1 mm (millimeter)
1 cm (centimeter)
1 m (meter)
1 m2 (square meter)
1 m3 (cubic meter)
1 km (kilometer)
1 km2 (square kilometer) or 100 ha (hectares)
1 ha (hectare)
1 L (liter)
1°C (degrees Celsius)
Standard
3.527396 x lQ-°8 ounces
3.527396 x 10-°5ounces
0.035 ounces
2.202 pounds
1.103 tons
0.039 inches
0.39 inch
3.28 feet
10.764 square feet
35.314 cubic feet
0.621 miles
0.286 square miles
2.47 acres
0.264 gallons
1.8C + 32 Fahrenheit
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
XXII
April 2013
-------
Elements and Chemical Symbols
Ag Silver
Al Aluminum
As Arsenic
B Boron
Ba Barium
Be Beryllium
Bi Bismuth
Ca Calcium
CaCOs Calcium carbonate
Cd Cadmium
Cl Chlorine
CN Cyanide
Co Cobalt
Cr Chromium
Cu Copper
F Fluorine
Fe Iron
Ga Gallium
Hg Mercury
In Indium
K Potassium
Mg Magnesium
Mn Manganese
Mo Molybdenum
Na Sodium
Ni Nickel
Pb Lead
Sb Antimony
Se Selenium
Sn Tin
S04 Sulfate
Sr Strontium
Te Tellurium
Th Thorium
Tl Thallium
U Uranium
V Vanadium
Zn Zinc
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ... April 2013
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Authors (listed alphabetically)
Rebecca Aicher, AAAS Fellow, USEPA-ORD, Washington, DC.
Greg Blair, ICF, Seattle, WA
Barbara Butler, USEPA-ORD, Cincinnati, OH
Heather Dean, USEPA-Region 10, Anchorage, AK
Joseph Ebersole, USEPA-ORD, Corvallis, OR
Sheila Eckman, USEPA-Region 10, Seattle, WA
Tami Fordham, USEPA-Region 10, Seattle, WA
Jeffrey Frithsen, USEPA-ORD, Washington, DC
Ralph Grismala, ICF, Lexington, MA
Michael Kravitz, USEPA-ORD, Cincinnati, OH
Phil North, USEPA-Region 10, Soldotna, AK
Richard Parkin, USEPA-Region 10, Seattle, WA
Jim Rice, ICF, Lexington, MA
Dan Rinella, University of Alaska, Anchorage, AK
KateSchofield, USEPA-ORD, Washington, DC
Steve Seville, ICF, Portland, OR
Glenn Suter, USEPA-ORD, Cincinnati, OH
Jason Todd, USEPA-ORD, Washington, DC
Parker J. Wigington, Jr., USEPA-ORD, Corvallis, OR
Contributors (listed alphabetically)
Dave Athens, Kenai River Center, Soldotna, AK
Alan Barnard, ICF, Sacramento, CA
Deborah Bartley, ICF, Seattle, WA
David Bauer, ICF, Fairfax, VA
Alan Boraas, Kenai Peninsula College, Soldotna, AK
Philip Brna, USFWS, Anchorage, AK
Laura Cooper, ICF, Portland, OR
Eric Doyle, ICF, Seattle, WA
John Duffield, Bioeconomics, Inc., Missoula, MT
Lorraine Edmond, USEPA-Region 10, Seattle, WA
Ginny Fay, University of Alaska, Anchorage, AK
Rachel Fertik, USEPA-OW, Washington, DC
Christopher Frissell, Pacific Rivers Council, Poison, MT
Cindi Godsey, USEPA-Region 10, Anchorage, AK
Oliver Scott Goldsmith, University of Alaska, Anchorage, AK
Michael Griffith, USEPA-ORD, Cincinnati, OH
Stephen Hoffman, USEPA-OSWER, Washington, DC
Palmer Hough, USEPA-OW, Washington, DC
Maureen Johnson, USEPA-ORD, Washington, DC
Gunnar Knapp, University of Alaska, Anchorage, AK
Catherine Knott, Kenai Peninsula College, Homer, AK
Douglas Limpinsel, NOAA, Anchorage, AK
James Lopez-Baird, USEPA-Region 10, Seattle, WA
Michael McManus, USEPA-ORD, Cincinnati, OH
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
XXIV
April 2013
-------
Chris Neher, Bioeconomics, Inc., Missoula, MT
Grant Novak, ICF, Seattle, WA
Corrine Ortega, ICF, Sacramento, CA
David Patterson, Bioeconomics, Inc., Missoula, MT
Ryan Patterson, ICF, Los Angeles, CA
Rori Perkins, ICF, Portland, OR
Ken Rock, ICF, Fairfax, VA
Tobias Schworer, University of Alaska, Anchorage, AK
Robert Seal, USGS, Reston, VA
Sacha Selim, ICF, San Francisco, CA
Rebecca Shaftel, University of Alaska, Anchorage, AK
Michael Slimak, USEPA-ORD, Washington, DC
Judy Smith, USEPA-Region 10, Portland, OR
Danny Stratten, ICF, Bellingham, WA
Greg Summers, ICF, Portland, OR
Jenny Thomas, USEPA-OW, Washington, DC
Lori Verbrugge, USFWS, Anchorage, AK
Michael Wiedmer, University of Washington, Anchorage, AK
Reviewers of Internal Review Drafts (listed alphabetically)
Dwight Atkinson, USEPA-OW, Washington, DC
Ned Black, USEPA-Region 9, San Francisco, CA (check)
Adrianne Fleek, USEPA-Region 10, Anchorage, AK
Cami Grandinetti, USEPA-Region 10, Washington, DC
Scot Hagerthey, USEPA-ORD, Washington, DC
James Hanley, USEPA-Region 8, Denver, CO
Stephen Hoffman, USEPA-OSWER, Washington, DC
Chris Hunter, USEPA-OW, Washington, DC
Thomas Johnson, USEPA-ORD, Washington, DC
Phil Kaufman, USEPA-ORD, Corvallis, OR
Stephen LeDuc, USEPA-ORD, Washington, DC
Julia McCarthy, USEPA-Region 8, Denver, CO
Caroline Ridley, USEPA-ORD, Washington, DC
Carol Russell, USEPA-Region 8, Denver, CO
DaveTomten, USEPA-Region 10, Seattle, WA
Felicia Wright, USEPA-OW, Washington, DC
Reviewers of External Review Draft (listed alphabetically)
David Atkins, Watershed Environmental, LLC
Steve Buckley, WHPacific
Courtney Carothers, University of Alaska Fairbanks
Dennis Dauble, Washington State University
Gordon Reeves, USDA Pacific Northwest Research Station
Charles Slaughter, University of Idaho
John Stednick, Colorado State University
Roy Stein, The Ohio State University
William Stubblefield, Oregon State University
Dirk van Zyl, University of British Columbia
Phyllis Weber Scannell, Scannell Scientific Services
Paul Whitney, Independent Consultant
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~~~April 2013
-------
PHOTO CREDITS
Front cover Main photo: Upper Talarik Creek (Joe Ebersole, USEPA)
Thumbnail 1: Brown bear (Steve Hillebrand, USFWS)
Thumbnail 2: Fishing boats at Naknek, Alaska (USEPA)
Thumbnail 3: Iliamna Lake (Lorraine Edmond, USEPA)
Thumbnail 4: Sockeye salmon in Wood River (Thomas Quinn, University of Washington)
Title Pages
Executive Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Summary Tributary of Napotoli Creek, near the Humble claim (Michael Wiedmer)
Area of mine scenario's tailings storage facility 1 (Michael Wiedmer, ADFG)
Chapter 1 Sockeye salmon near Gibraltar Lake (Thomas Quinn, University of Washington)
Kvichak River below Iliamna Lake and IgiugigfJoe Ebersole, USEPA)
Salmon art on a building in Dillingham (Alan Boraas, Kenai Peninsula College)
Chapter 2 Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Lodge on the Kvichak River (Joe Ebersole, USEPA)
Floodplain beaver ponds on Upper Talarik Creek (Joe Ebersole, USEPA)
Chapter 3 Groundwater upwelling near Kaskanak Creek, Lower Talarik basin (Joe Ebersole, USEPA)
Homes in Nondalton (Alan Boraas, Kenai Peninsula College)
Knutson Creek draining into the Knutson Bay area of Iliamna Lake (Keith Denton)
Chapter 4 Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Tributary of Napotoli Creek, near the Humble claim (Michael Wiedmer)
Area of mine scenario's tailings storage facility 1 (Michael Wiedmer, ADFG)
Chapter 5 Sockeye salmon near Gibraltar Lake (Thomas Quinn, University of Washington)
Kvichak River below Iliamna Lake and IgiugigfJoe Ebersole, USEPA)
Salmon art on a building in Dillingham (Alan Boraas, Kenai Peninsula College)
Chapter 6 Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Lodge on the Kvichak River (Joe Ebersole, USEPA)
Floodplain beaver ponds on Upper Talarik Creek (Joe Ebersole, USEPA)
Chapter 7 Groundwater upwelling near Kaskanak Creek, Lower Talarik basin (Joe Ebersole, USEPA)
Homes in Nondalton (Alan Boraas, Kenai Peninsula College)
Knutson Creek draining into the Knutson Bay area of Iliamna Lake (Keith Denton)
Chapter 8 Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Tributary of Napotoli Creek, near the Humble claim (Michael Wiedmer)
Area of mine scenario's tailings storage facility 1 (Michael Wiedmer, ADFG)
Chapter 9 Sockeye salmon near Gibraltar Lake (Thomas Quinn, University of Washington)
Kvichak River below Iliamna Lake and IgiugigfJoe Ebersole, USEPA)
Salmon art on a building in Dillingham (Alan Boraas, Kenai Peninsula College)
Chapter 10 Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Lodge on the Kvichak River (Joe Ebersole, USEPA)
Floodplain beaver ponds on Upper Talarik Creek (Joe Ebersole, USEPA)
Chapter 11 Groundwater upwelling near Kaskanak Creek, Lower Talarik basin (Joe Ebersole, USEPA)
Homes in Nondalton (Alan Boraas, Kenai Peninsula College)
Knutson Creek draining into the Knutson Bay area of Iliamna Lake (Keith Denton)
Chapter 12 Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Tributary of Napotoli Creek, near the Humble claim (Michael Wiedmer)
Area of mine scenario's tailings storage facility 1 (Michael Wiedmer, ADFG)
Chapter 13 Sockeye salmon near Gibraltar Lake (Thomas Quinn, University of Washington)
Kvichak River below Iliamna Lake and IgiugigfJoe Ebersole, USEPA)
Salmon art on a building in Dillingham (Alan Boraas, Kenai Peninsula College)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
XXVI
April 2013
-------
Chapter 14 Sockeye salmon in the Wood River (Thomas Quinn, University of Washington)
Lodge on the Kvichak River (Joe Ebersole, USEPA)
Floodplain beaver ponds on Upper Talarik Creek (Joe Ebersole, USEPA)
Chapter 15 Groundwater upwelling near Kaskanak Creek, Lower Talarik basin (Joe Ebersole, USEPA)
Homes in Nondalton (Alan Boraas, Kenai Peninsula College)
Knutson Creek draining into the Knutson Bay area of Iliamna Lake (Keith Denton)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment .. April 2013
-------
ACKNOWLEDGEMENTS
Assistance for this assessment was provided by ICF International under USEPA contract number
EP-C-09-009 and by NatureServe under USEPA contract number EP-W-07-080. Additional assistance
was provided by Colleen Matt (USEPA Contract EP-12-H-000099), Northern Ecologic LLC
(USEPA Contract EPA-12-H-001001), Charles Schwartz (USEPA Contract EP-12-H-000105) and Halcyon
Research (USEPA Contract EP-12-H-000107). The initial external peer review of the assessment was
coordinated by Versar, Inc., under USEPA contract number EP-C-07-025. The external peer review of
specific supplemental materials provided during the public comment period was coordinated by Versar,
Inc., under USEPA contract number EP-C-12-045. Assistance with the management of public comments
was provided by Horsley-Witten under USEPA contract EP-C-08-018. Contractors contributing to this
report we re required to certify that they had no organizational conflicts of interest. As defined by
Federal Acquisition Regulations subpart2.101, an organizational conflictof interest may exist when,
"because of other activities or relationships with other persons, a person is unable or potentially unable
to render impartial assistance or advice to the Government, or the person's objectivity in performing the
contract work is or might otherwise be impaired or a person has an unfair competitive advantage."
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
xxvill
April 2013
-------
The Bristol Bay watershed in southwestern Alaska supports the largest sockeye salmon fishery in the
world, is home to 25 federally recognized tribal governments, and contains large mineral resources. The
potential for large-scale mining activities in the watershed has raised concerns about the impact of
mining on the sustainability of Bristol Bay's world-class commercial, recreational, and subsistence
fisheries and the future of Alaska Natives tribes in the watershed, who have maintained a salmon-based
culture and subsistence-based way of life for at least 4,000 years.
The U.S. Environmental Protection Agency (USEPA) launched this assessment to determine the
significance of Bristol Bay's ecological resources and evaluate the impacts of large-scale mining on these
resources. It uses the well-established methodology of an ecological risk assessment, which is a type of
scientific investigation that provides technical information and analyses to foster public understanding
and to inform future decision making. As a scientific assessment, it does not discuss or recommend
policy, legal, or regulatory decisions, nor does it outline or analyze options for future decisions.
The purpose of the assessment is to characterize the biological and mineral resources of the Bristol Bay
watershed, increase understanding of the impacts of large-scale mining on the region's fish resources,
and inform future government decisions related to protecting and maintaining the physical, chemical,
and biological integrity of the watershed.
The assessment is intended to be a technical resource for the public and for federal, state, and tribal
government entities as they consider how best to address the challenges of mining and ecological
protection in the Bristol Bay watershed. It will inform the ongoing discussions of the risks of mine
development to the sustainability of the Bristol Bay salmon fisheries and thus will be of value to the
many stakeholders in this debate.
The assessment also could inform the consideration of options for future action by government bodies.
This includes USEPA, which has been petitioned by multiple groups to address mining activity in the
Bristol Bay watershed using its authority under the Clean Water Act (CWA). Should specific mine
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-1
April 2013
-------
projects reach the permitting stage, the assessment will enable state or federal permitting authorities to
make informed decisions to grant, deny, or condition permits and/or conduct additional research or
assessment as a basis for such decisions. USEPA is conducting this assessment consistent with its
authority under the CWA Section 104(a) and (b).
Scope of the Assessment
This assessment reviews, analyzes, and synthesizes information relevant to impacts of large-scale mine
development on Bristol Bay fisheries and subsequent effects on the wildlife and Alaska Native cultures
of the region. Given the economic, ecological, and cultural importance of the region's salmonids
(sockeye, Chinook, coho, chum, and pink salmon, as well as rainbow trout and Dolly Varden) and the
concern of stakeholders and the public that a mine could affect those species, the primary focus of the
assessment is the abundance, productivity, and diversity of these fishes. Because wildlife and Alaska
Native cultures in Bristol Bay are intimately connected to and dependent upon these and other fishes,
changes in these fisheries are likely to affect the abundance and health of wildlife populations and the
viability and welfare of Alaska Native populations. Therefore, wildlife and Alaska native cultures are also
considered as assessment endpoints, but only as affected by changes in salmonid fisheries.
The assessment considers multiple spatial scales. The largest scale is the Bristol Bay watershed, which is
a largely undisturbed region with outstanding natural, cultural, and mineral resources. Within the larger
Bristol Bay watershed, the assessment focuses on the Nushagak and Kvichak River watersheds
(Figure ES-1). These are the largest of the Bristol Bay watershed's six major river basins, containing
about 50% of the total watershed area and are identified as mineral development areas by the State of
Alaska. The Pebble deposit, the most likely site for near-term, large-scale mine development in the
region, is located in the headwaters of tributaries to both the Nushagak and Kvichak Rivers. Therefore,
both of these watersheds are subject to potential risks from mining. The third spatial scale is the
watersheds of the three tributaries that originate within the potential footprint of a mine on the Pebble
deposit: the South Fork Koktuli River, which drains the Pebble deposit area and converges with the
North Fork west of the Pebble deposit; the North Fork Koktuli River, located to the northwest of the
Pebble deposit, which flows into the Nushagak River via the Mulchatna River; and Upper Talarik Creek,
which drains the eastern portion of the Pebble deposit and flows into the Kvichak River via Iliamna
Lake, the largest undeveloped lake in the United States (Figure ES-1). The mine footprints under the
three realistic mine scenarios evaluated in the assessment make up the fourth spatial scale. These
scenarios—Pebble 0.25, Pebble 2.0, and Pebble 6.5—define three potential mine sizes, representing
different stages in the potential process of mining the Pebble deposit. The final spatial scale is the
combined area of the subwatersheds between the mine footprints and the Kvichak River watershed
boundary that would be crossed by a transportation corridor linking the mine site to Cook Inlet.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ 2 APril 2013
-------
Figure ES 1. The Nushagak and Kvichak River watersheds of Bristol Bay.
Cook Inlet
Bristol Bay
N
A
25
25
50
] Kilometers
50
] Miles
w Approximate Pebble Deposit Location
• Towns and Villages
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-3
April 2013
-------
The assessment also addresses two periods for mine activities. The first is the development and
operation phase, during which mine infrastructure would be built and the mine would be operated. This
phase may last from 20 to 100 years or more. The second is the post-mining or post-closure phase,
during which the site would be monitored. As necessary, water treatment and other waste management
activities would continue and any failures would be remediated. Because mine wastes would be
persistent, this period could continue for centuries and potentially in perpetuity.
We began the assessment with a thorough review of what is known about the Bristol Bay watershed, its
fisheries and wildlife populations, and its Alaska Native cultures. We also reviewed information about
copper mining and publicly available information outlining proposed mining operations for the Pebble
deposit, which has been the focus of much exploratory study and has received much attention from
groups in and outside of Alaska. With the help of regional stakeholders, we developed a set of conceptual
models to show potential associations between salmon populations and the environmental stressors
that might reasonably be expected as a result of large-scale mining. Then, folio wing the USEPA's
ecological risk assessment framework, we analyzed the sources and exposures that could occur and the
potential responses to those exposures. Finally, we characterized the risks to fish habitats, salmon, and
other fish populations; and the implications of those risks to the wildlife and Alaska Native cultures that
use them.
This is not an in-depth assessment of a specific mine, but rather an examination of impacts of reasonably
foreseeable mining activities in the Bristol Bay region, given the nature of the watershed's mineral
deposits and the requirements for successful mine development. The assessment analyzes mine
scenarios that reflect the expected characteristics of mine operations at the Pebble deposit. It is
intended to provide a baseline for understanding the impacts of mine development not just at the Pebble
deposit, but throughout the Nushagak and Kvichak River watersheds. The mining of other existing
porphyry copper deposits in the region would likely include the same types of mining activities and
facilities evaluated in this assessment for the Pebble deposit (open pit mining, waste rock piles, tailings
storage facilities [TSFs]), and therefore would present potential risks similar to those outlined in this
assessment. However, those mines would likely be most similar to the smallest of the mine scenarios
analyzed in this assessment (Pebble 0.25), because the other ore bodies are believed to be much smaller
than the Pebble deposit.
This assessment does not consider all impacts associated with future large-scale mining in the Bristol
Bay watershed. Although the mine scenarios assume development of a deep-water port on Cook Inlet to
ship product concentrate elsewhere for smelting and refining, impacts of port development and
operation are not assessed. The assessment does not evaluate impacts of the one or more large-capacity,
electricity-generating power plants that would be required to power the mine and the port. It also does
not assess the effects of induced development that could result from large-scale mining in the region.
However, it is recognized that a large-scale mine development could induce the development of
additional support services for mine employees and their families, recreational facilities due to
increased access, vacation homes, and transportation infrastructure beyond the main corridor (i.e.,
airports, docks, and roads).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ 4 April 2013
-------
Ecological Resources
The Bristol Bay watershed provides habitat for numerous animal species, including 29 fish species, more
than 40 terrestrial mammal species, and more than 190 bird species. Many of these species are essential
to the structure and function of the region's ecosystems and economies. Chief among these resources are
world-class commercial and sport fisheries for Pacific salmon and other salmonids. The watershed
supports production of all five species of Pacific salmon found in North America: sockeye (Oncorhynchus
nerka], coho (0, kisutch), Chinook [0, tshawytschd), chum [0, ketd), and pink [0, gorbuschd)
(Figure ES-2). Because no hatchery fish are raised or released in the watershed, Bristol Bay's salmon
populations are entirely wild. These fishes are anadromous, meaning that they hatch and rear in
freshwater systems, migrate to sea to grow to adult size, and return to freshwater systems to spawn and
die.
The most abundant salmon species in the Bristol Bay watershed is sockeye salmon. The watershed
supports the largest sockeye salmon fishery in the world, with approximately 46% of the average global
abundance of wild sockeye salmon (Figure ES-3). Between 1990 and 2009, the annual average inshore
run of sockeye salmon in Bristol Bay was approximately 37.5 million fish. Annual commercial harvest of
sockeye over this same period averaged 25.7 million fish. Approximately half of the Bristol Bay sockeye
salmon production is from the Nushagak and Kvichak River watersheds, the main area of focus for this
assessment (Figure ES-3).
Chinook salmon are also abundant in the region. Chinook returns to the Nushagak River are consistently
greater than 100,000 fish per year and have exceeded 200,000 fish in 11 years between 1966 and 2010,
frequently placing Nushagak River Chinook runs at or near the world's largest. This is noteworthy given
the Nushagak River's small watershed compared to other Chinook-producing rivers such as the Yukon
River, which spans Alaska, and the Kuskokwim River in southwestern Alaska, just north of Bristol Bay.
The Bristol Bay watershed also supports populations of non-salmon fishes that typically (but not
always) remain in the watershed's freshwater habitats throughout their life cycles. The region contains
highly productive waters for sport and subsistence fish species, including rainbow trout (0. mykiss),
Dolly Varden (Salvelinus malmd), Arctic char (S. alpinus), lake trout (S. namaycush), Arctic grayling
(Thymallus arcticus), northern pike [Esox lucius), and humpback whitefish [Coregonus pidschian). These
fishes occupy a variety of habitats in the watershed, from headwater streams to wetlands to large rivers
and lakes. The Bristol Bay region is especially renowned for the abundance and size of its rainbow trout:
between 2003 and 2007, an estimated 183,000 rainbow trout were caughtin the Bristol Bay
Management Area.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ 5 April 2013
-------
Figure ES 2. Reported salmon (sockeye, Chinook, coho, pink, and chum combined) distribution in
the South and North Fork Koktuli Rivers and Upper Talarik Creek watersheds. Designation of
species spawning, rearing, and presence is based on the Anadromous Waters Catalog. Life stage
specific reach designations are likely underestimates, given the challenges inherent in surveying all
streams that may support life stage use throughout the year.
SOUTH FORK KOKTULI
Present
Spawning
Rearing
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-6
April 2013
-------
Figure ES 3. Total sockeye salmon run sizes by (A) region and (B) watershed in the Bristol Bay
region. Values are averages from 1956 to 2005 and 1956 to 2010 for A and B, respectively.
I Bristol Bay
I Russia Mainland & Islands
I West Kamchatka
I East Kamchatka
I Western Alaska (excluding Bristol Bay)
I South Alaska Peninsula
iKodiak
Cook Inlet
: Prince William Sound
Southeast Alaska
North British Columbia
South British Columbia, Washington & Oregon
B
• Togiak
• Nushagak
Kvichak
Naknek
Egegik
Ushagik
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-7
April 2013
-------
The exceptional quality of the Bristol Bay watershed's fish populations can be attributed to several
factors, the most important of which is the watershed's high-quality, diverse aquatic habitats, which are
untouched by human-engineered structures and flow management controls. Surface and subsurface
waters are highly connected, enabling hydrologic and biochemical connectivity between wetlands,
ponds, streams, and rivers, thus increasing the diversity and stability of habitats able to support fish.
These factors all contribute to making the Bristol Bay watershed a highly productive system. High
aquatic habitat diversity also has supported the high genetic diversity offish populations. This diversity
in genetics, life history, and habitat acts to reduce year-to-year variability in total production and
increase the stability of the fishery.
The return of spawning salmon from the Pacific Ocean brings marine-derived nutrients into the
watershed and fuels terrestrial and aquatic food webs. Thus, the condition of Bristol Bay's terrestrial
ecosystems is intimately linked to the condition of salmon populations as well as to almost totally
undisturbed habitats. The watershed continues to support large carnivores such as brown bears (Ursus
arctos), bald eagles (Haliaeetus leucocephalus], and gray wolves (Canis lupus); ungulates such as moose
[Alces alcesgigas) and caribou [Rangifer tarandusgranti); and numerous waterfowl species. Brown
bears are abundant in the Nushagak and Kvichak River watersheds. Moose also are abundant, with
populations especially high in the Nushagak River watershed where felt-leaf willow, a preferred forage
species, is plentiful. The Nushagak and Kvichak River watersheds are used by caribou, primarily the
Mulchatna caribou herd. This herd ranges widely through these watersheds, but also spends
considerable time in other watersheds.
Alaska Native Cultures
The predominant Alaska Native cultures present in the Nushagak and Kvichak River watersheds—the
Yup'ik and Dena'ina—are two of the last intact, sustainable, salmon-based cultures in the world. In
contrast, other Pacific Northwest salmon-based cultures are severely threatened by development,
degraded natural resources, and declining salmon resources. Salmon are integral to the entire way of life
in these cultures as subsistence food and subsistence-based livelihoods, and are an important
foundation for language, spirituality, and social structure. The cultures have a strong connection to the
landscape and its resources. In the Bristol Bay watershed, this connection has been maintained for at
least the past 4,000 years and is in part both due to and responsible for the continued undisturbed
condition of the region's landscape and biological resources. The respect and importance given salmon
and other wildlife, along with the traditional knowledge of the environment, have produced a
sustainable subsistence-based economy. This subsistence-based way of life is a key element of Alaska
Native identity and serves a wide range of economic, social, and cultural functions in Yup'ik and
Dena'ina societies.
Fourteen of Bristol Bay's 25 villages and communities are within the Nushagak and Kvichak River
watersheds, with a total population of 4,337 in 2010. Thirteen of the 14 communities have federally
recognized tribal governments and a majority Alaska Native population. Many of the non-Alaska Native
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ a April 2013
-------
residents in the watersheds also have strong cultural ties to the region and practice a subsistence way of
life. In the Bristol Bay region, salmon constitute approximately 52% of the subsistence harvest, and for
some communities this proportion is substantially higher.
The subsistence way of life in many Alaska Native villages is augmented with activities supporting cash
economy transactions. Alaska Native villages, in partnership with Alaska Native corporations and other
business interests, are considering a variety of economic development opportunities. Some Alaska
Native villages have decided that large-scale hard rock mining is not the direction they would like to go
in, while a few others are seriously considering this opportunity. All are concerned with the long-term
sustainability of their communities.
Economics of Ecological Resources
The Bristol Bay watershed supports several economic sectors that are wilderness-compatible and
sustainable: commercial, sport, and subsistence fishing; sport and subsistence hunting; and non-
consumptive recreation. Considering all these sectors, the ecological resources of the Bristol Bay
watershed generated nearly $480 million in direct economic expenditures and sales in 2009, and
provided employment for over 14,000 full- and part-time workers. The Bristol Bay commercial salmon
fishery generates the largest component of economic activity: it was valued at approximately
$300 million in 2009 (sales from fishers to processors), and provided employment for over 11,000 full-
and part-time workers at the season's peak. These estimates do not include retail expenditures from
national and international sales. The Bristol Bay sport-fishing industry supports approximately 29,000
sport-fishing trips, generates approximately $60 million per year, and directly employs over 800 full-
and part-time workers (based on 2009 data). Sport hunting—mostly of caribou, moose, and brown
bear—generates more than $8 million per year and employs over 100 full- and part-time workers. The
scenic value of the watershed, measured in terms of wildlife viewing and tourism, is estimated to
generate an additional $100 million per year and supports nearly 1,700 full- and part-time workers. The
subsistence harvest offish also contributes to the region's cash economy when Alaskan households
spend money on subsistence-related supplies. These contributions are estimated to be over $6 million
per year. This does not include the replacement value of subsistence resources. These economic data
provide background only. The economic effects of mining are not assessed.
Geological Resources
In addition to significant and valuable ecological resources, the Nushagak and Kvichak River watersheds
contain considerable mineral resources. The potential for large-scale mine development in the region is
greatest for copper deposits and, to a lesser extent, for intrusion-related gold deposits. Because these
deposits are low-grade—meaning that they contain relatively small amounts of metals relative to the
amount of ore—mining will be economic only if conducted over large areas, and mining will produce
large amounts of waste material.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ 9 April 2013
-------
The largest known and most explored deposit is the Pebble deposit. If fully mined, the Pebble deposit
could produce more than 11 billion tons of ore, which would make it the largest mine of its type in North
America. Although the Pebble deposit represents the most imminent and likely site of mine
development, other mineral deposits with potentially significant resources exist in the Nushagak and
Kvichak River watersheds. Ten specific claims with more than minimal recent exploration have been
filed for copper deposits, most near the Pebble deposit. Findings of this assessment concerning the
impacts of large-scale mining are generally applicable to these other sites.
Mine Scenarios
Like all risk assessments, this assessment is based on scenarios that define a set of possible future
activities. To assess mining-related stressors that could affect ecological resources in the watershed, we
developed realistic mine scenarios that include a range of mine sizes and operating conditions. These
mine scenarios are based on the Pebble deposit because it is the best-characterized mineral resource
and the most likely to be developed in the near term. The mine scenarios draw on plans developed for
Northern Dynasty Minerals, consultation with experts, and baseline data collected by the Pebble Limited
Partnership to characterize the likely mine site, mining activities, and surrounding environment. Details
of any future mine plan for the Pebble deposit or for other deposits in the watershed will differ from our
mine scenarios. However, our scenarios reflect the general characteristics of mineral deposits in the
watershed, modern conventional mining technologies and practices, the scale of mining activity
required for economic development of the resource, and the necessary development of infrastructure to
support large-scale mining. Therefore, the mine scenarios evaluated in the assessment realistically
represent the type of development plan that can be anticipated for a porphyry copper deposit in the
Bristol Bay watershed. Uncertainties associated with the mine scenarios are discussed later in this
executive summary.
The three mine scenarios evaluated in the assessment are based on the amount of ore processed:
Pebble 0.25 (approximately 0.25 billion tons [0.23 billion metric tons] of ore and duration of 20 years),
Pebble 2.0 (approximately 2.0 billion tons [1.8 billion metric tons] of ore and duration of 25 years), and
Pebble 6.5 (approximately 6.5 billion tons [5.9 billion metric tons] of ore and duration of 78 years). The
major parameters of the three mine scenarios are presented in Table ES-1, and their layouts are
presented in Figure ES-4. The largest features of a mine would be an open pit, waste rock piles, and
TSFs. Other significant features include an ore-processing facility and a water collection and treatment
system. An underground extension of the mine could increase the size of the mine to 11 billion tons of
ore, is not included in this assessment.
The mine scenarios include a 138-km (86-mile) transportation corridor of which 113 km (70 miles)
would be within the assessment watersheds (Figure ES-5). This corridor would include a gravel-
surfaced road and four pipelines (one each for product concentrate, return water, diesel fuel, and
natural gas).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment EsTo April 2013
-------
The assessment considers risks from routine operation of a mine designed using modern conventional
mitigation practices and technologies and with no significant human or engineering failures. The
assessment also considers various failures that have occurred during the operation of other mines and
could occur in this case, including failures of a tailings dam, pipelines, a wastewater treatment plant, and
culverts.
Table ES 1. Mine scenario parameters.
Parameter
Amount of ore mined (billion metric tons)
Approximate duration of mining
Ore processing rate (metric tons/day)
Mine Scenario
Pebble 0.25
0.23
20 years
31,000
Pebble 2.0
1.8
25 years
198,000
Pebble 6.5
5.9
78 years
208,000
Mine Pit
Surface area (km2)
Depth (km)
1.5
0.30
5.5
0.76
17.8
1.24
Waste Rock Pile
Surface area (km2)
PAG waste rock (million metric tons)
NAG waste rock (million metric tons)
2.3
95
350
13.0
580
2,200
22.6
4,700
10,900
TSFl"
Capacity, weight (billion metric tons)
Surface area, exterior (km2)
Maximum dam height (m)
0.25
5.88
90
1.96
15.8
209
1.96
15.8
209
TSF 2"
Capacity, weight (billion metric tons)
Surface area, exterior (km2)
NA
NA
NA
NA
3.7
21.5
TSF 3"
Capacity, weight (billion metric tons)
Surface area, exterior (km2)
Total TSF surface area, exterior (km2)
NA
NA
5.88
NA
NA
15.8
0.96
8.3
45.6
Notes:
a Final value, when TSF is full.
NA = not applicable; TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-11
April 2013
-------
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-12
April 2013
-------
Subwatersheds defined at the HUC 12 level according to the National Hydrography Dataset.
W" Approximate Pebble Deposit Location
= = — — Transportation Corridor (Outside Assessment Area)
^=^= Transportation Corridor
Existing Roads
Transportation Corridor Area
Subwatersheds within Area
N
A
0 5 10
0 5 10
] Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-13
April 2013
-------
Risks to Salmon and Other Fishes
Based on the mine scenarios, the assessment defines mining-related stressors that could affect the
Bristol Bay watershed's fishes and consequently have impacts on wildlife and human welfare. The
scenarios include both routine operations (Tables ES-2 and ES-3) and several potential failure scenarios
(Table ES-4).
Mine Footprint
Effects on fishes resulting from habitat loss and modification would occur directly in the area of mining
activity and indirectly downstream because of habitat destruction.
• Loss of 38, 90, and 145 km (24, 56 and 90 miles) of streams in the footprint of the mine pit, TSFs
and waste rock piles, due to elimination, blockage, or dewatering of streams under the Pebble 0.25,
2.0, and 6.5 scenarios, respectively. These losses would translate to losses of 8,24, and 35 km (5,15,
and 22 miles) of streams known to provide spawning or rearing habitats for coho salmon, sockeye
salmon, Chinook salmon, and Dolly Varden. (Figure ES-6.)
• Altered streamflow due to retention and discharge of water used in mine operations, ore processing,
transport, and other processes would reduce the amount and quality offish habitat. Reductions in
streamflow exceeding 20% would adversely affect habitat in an additional 15, 26 and 54km (9.3,16
and 34 miles) of streams under the Pebble 0.25, 2.0, and 6.5 scenarios, respectively, reducing
production of sockeye salmon, coho salmon, Chinook salmon, rainbow trout, and Dolly Varden.
Reduced flows would also result in an unquantifiable area of riparian floodplain wetland habitat
being lost or altered in terms of hydrologic connectivity with streams.
• Loss of 5.0,12.4 and 19.4 km2 (1,200, 3,000 and 4,800 acres) of wetlands in the footprints of the
Pebble 0.25, 2.0, and 6.5 scenarios, respectively, would reduce off-channel habitat for salmon and
other fishes (Figure ES-6). Wetland loss would reduce availability of and access to hydraulically and
thermally diverse habitats that can provide enhanced foraging opportunities and important rearing
habitats for juvenile salmon.
• Indirect effects of stream and wetland losses would include reductions in the quality of
downstream habitat for coho salmon, sockeye salmon, Chinook salmon, rainbow trout, and Dolly
Varden. These indirect effects cannot be quantified, but likely would diminish fish production
downstream of the mine site. Indirect effects would be caused by the following alterations.
o Reduced food resources would result from the loss of organic material and drifting
invertebrates from the streams and streamside wetlands lost to the mine footprint.
o The balance of surface water and groundwater inputs to downstream reaches would shift,
potentially reducing winter fish habitat and making streams less suitable for spawning and
rearing.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES~14April 2013
-------
o Seasonal temperatures could be altered by water treatment and reduced groundwater
flowpaths, making streams less suitable for salmonids.
Water Quality
Leakage during Routine Operations
Water from the mine site could enter streams through the waste water treatment plant discharges and in
uncollected runoff and leakage of leachates from the waste rock piles and tailings storage facilities.
Wastewater treatment is assumed to meet all state and national standards and criteria, or equivalent
benchmarks for chemicals that have no criteria. However, water quality would be diminished by
uncollected leakage of tailings and waste rock leachates from the containment system. Test leachates
from the tailings and non-ore-bearing waste rocks are mildly toxic. They would require an
approximately two-fold dilution to achieve water quality criteria for copper, but are not estimated to be
toxic to salmonids. Waste rocks associated with the ore body are acid-forming with high copper
concentrations in test leachates, and would require 2,900- to 52,000-fold dilution to achieve water
quality criteria. Several metals could be sufficiently elevated to contribute to toxicity, but copper is the
dominant toxicant.
Because leachates could leak during routine operations, instream copper levels would be sufficient to
cause direct effects on salmonids in 29 and 57 km (18 and 35 miles) of streams beyond the mine
scenario footprints in the Pebble 2.0 and Pebble 6.5 scenarios, but not in the Pebble 0.25 scenario (Table
ES-2). These effects would range from aversion and avoidance of the affected habitat to rapidly induced
death of many or all fish in 12 km of streams under the Pebble 6.5 scenario. Copper would cause death
or reduced reproduction of aquatic invertebrates in 15, 62, and 83 km (9.3, 38, and 51 miles) of streams
in the Pebble 0.25, 2.0, and 6.5 scenarios, respectively. These invertebrates are the primary food source
for juvenile salmon and all life stages of other salmonids, so reduced invertebrate productivity would
reduce fish productivity. These results are sensitive to the assumed efficiency of the leachate capture
system, and a more efficient system could be devised. However, greater than 99% capture efficiency
would be required to prevent exceedance of the copper criteria for the South Fork Koktuli River under
the Pebble 6.5 scenario, which would require technologies beyond those specified in our scenarios or
identified in the most recent preliminary mine plan.
Wastewater Treatment Plant Failure
Based on a review of historical and currently operating mines, some failure of water collection and
treatment systems would be likely during operation or post-closure periods. A variety of water
collection and treatment failures are possible, ranging from operational failures resulting in short-term
releases of untreated or partially treated leachates to long-term failures to operate water collection and
treatment systems in perpetuity. A reasonable upper bound failure scenario would involve a complete
loss of water treatment and release of untreated wastewater. Under that scenario, copper
concentrations would be sufficient to cause direct effects on salmonids in 45,100, and 100 km (28, 63,
and 64 miles) of streams and on aquatic invertebrates in 100,110, and 130 km (62, 68, and 80 miles) of
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES~15April 2013
-------
streams in the Pebble 0.25, 2.0, and 6.5 scenarios, respectively (Table ES-2). Under the Pebble 6.5
scenario, death offish would occur rapidly in 31 km (19 miles) of stream following treatment failure, but
effects on fish would be less severe in the other scenarios.
Transportation Corridor
Construction and Routine Operation
In the Kvichak River watershed, the transportation corridor would cross 53 streams and rivers known
or likely to support migrating and resident salmonids, including 20 streams designated as anadromous
waters at the location of the crossing (Figure ES-7). The corridor would run near Iliamna Lake and cross
multiple tributary streams near their confluence with the lake. These habitats are important spawning
areas for sockeye salmon, putting sockeye particularly at risk from the road. Diminished habitat quality
in streams and wetlands below road crossings would result primarily from altered flow, runoff of road
salts, and siltation of habitat for salmon spawning and rearing and production of invertebrate prey
(Tables ES-2 and ES-3).
Culvert Failures
The most likely serious failure associated with the road would be blockages or other failures of culverts
that would inhibit fish passage. Culverts commonly become blocked by debris or ice that may not stop
water flow but create a barrier to fish movement. Fish passage may also be blocked or inhibited by
erosion below a culvert that "perches" the culvert, resulting in a waterfall, or by shallow water caused by
a wide culvert and periodic low stream flows. If blockages occurred during adult salmon immigration or
juvenile salmon emigration and were not cleared for several days, production of a year-class (i.e., fish
spawned in the same year) could be lost or diminished from that stream above the culvert.
Culverts can also fail to convey water as a result of landslides or, more commonly, floods that wash out
culverts that are too small or improperly installed. In such failures, the stream could be temporarily
impassible to fish until the culvert is repaired or until erosion re-establishes the channel. If the failure
occurs during a critical period in salmon migration, the effects would be the same as with a debris
blockage (i.e., a lost or diminished year-class).
Culvert failures also could result in the downstream transport and deposition of silt, which could cause
returning salmon to avoid a stream if they arrived during or immediately following the failure. More
likely, deposition of silt would smother salmon eggs and alevins, if they were present, and would
degrade downstream habitat for salmonids and the invertebrates that they eat.
Extended blockage offish passage at road crossings is unlikely during operation in our scenarios, which
specify daily inspection and maintenance. However, after mine operations cease, the road may be
maintained less carefully by the operator or may be transferred to a government entity that likely would
not be able to support daily inspection and maintenance. In either case, the proportion of culverts that
are impassable would be expected to revert to levels found in published surveys of public roads (range
of 30 to 58%, mean of 47%). Of the approximately 46 culverts that would be required, 35 would be on
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES~16April 2013
-------
streams that are believed to support salmonids. Hence, over the long term, 10 to 20 streams would be
expected to lose passage of salmon, rainbow trout, or Dolly Varden for an indefinite period of time, and
some proportion of those streams would have degraded downstream habitat resulting from
sedimentation from washout of the road.
Truck Accidents
Trucks would carry ore processing chemicals to the mine site. Truck accidents are likely over the long
period of mine operation and could release process chemicals to streams, resulting in toxic effects on
invertebrates or fish. The risk of spills might be mitigated by using impact-resistant containers.
Tailings Dam Failure
Tailings are the waste materials produced during ore processing, which, in our scenarios, would be
stored in TSFs consisting of tailings dams and impoundments. The probability of a tailings dam failure
increases with the number of dams. The Pebble 0.25 scenario would include one TSF with a single dam,
the Pebble 2.0 scenario would include one TSF with three dams, and the Pebble 6.5 scenario would
include three TSFs with a total of nine dams. Because there is no plan for their removal when mining
activities cease, the TSFs and their component dams are likely to be in place for hundreds to thousands
of years, long beyond the life of the mine. Available reports from the PLP suggest tailings dams as high as
209 m (685 feet) at TSF 1. At this height, the tailings dam would be higher than the St. Louis Gateway
Arch and the Washington Monument (Figure ES-8). We evaluated two potential dam failures in this
assessment: one when TSF 1 was partially full (under the Pebble 0.25 scenario) and one when it was
completely full (under the Pebble 2.0 scenario). In both cases we assumed 20% of the tailings would be
released, a conservative estimate that is well within the range of historical tailings dam failures. Failures
in the Pebble 6.5 scenario, which includes three TSFs, were not analyzed but would be similar.
Table ES 2. Summary of estimated stream lengths potentially affected under the three mine
scenarios, assuming routine operations.
Effect
Eliminated, blocked, or dewatered
Eliminated, blocked, or dewatered— anadromous
>20% flow reduction3
Direct toxicity to fisha
Direct toxicity to invertebrates3
Downstream of transportation corridor
Stream Length Affected (km)
Pebble 0.25
38
8
15
0
15
Pebble 2.0
90
24
26
29
62
Pebble 6.5
145
35
54
57
83
290
Notes:
"" Stream reaches with flow reductions partially overlap those with toxicity.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-17
April 2013
-------
Table ES 3. Summary of estimated wetland areas potentially affected under the three mine
scenarios, assuming routine operations.
Effect
Lost to the mine footprint
Lost to reduced flow from footprint
Filled by road bed
Influenced by the road (within 100 m)
Wetland Area Affected (km2)
Pebble 0.25 Pebble 2.0 Pebble 6.5
5.0 12.4 19.4
unquantified
0.11
2.4
Table ES 4. Probability and consequences of potential failures under the mine scenarios.
Failure Type
Tailings dam
Product concentrate pipeline
Concentrate spill into a stream
Concentrate spill into a wetland
Return water pipeline
Diesel pipeline spill
Culvert, during operation
Culvert, post-operation
Truck accidents
Water collection and treatment,
operation
Water collection and treatment,
managed post-closure
Water collection and treatment,
after site abandonment
Probability3
4 x ID-4 to 4 x ID-6 per dam-year =
recurrence frequency of 2,500 to
250,000 yearsb
10-3 per km-year = 95% chance
per pipeline in 25 years
1.5xlO-2 per year = Ito 2
stream-contaminating spills in 78
years
3 x 10-2 per year = 2 wetland-
contaminating spills in 78 years
Same as product concentrate
pipeline
Same as product concentrate
pipeline
Low
3 x 10-1 to 6 x 10-1 per culvert;
instantaneous = 11 to 21 culverts
1.9 x 10-7spills per mile of travel =
4 accidents in 25 years (up to
2 near-stream spills in 78 years)
0.60 to 0.93 = proportion of
recent U.S. mines with reportable
water collection and treatment
failures. Better practices might
reasonably be expected to reduce
this to 0.1.
Somewhat higher than operation
Certain
Consequences
More than 30 km of salmonid stream would be
destroyed and more streams and rivers would have
greatly degraded habitat for decades.
Most failures would occur between stream or
wetland crossings and might have little effect on fish.
Fish and invertebrates would experience acute
exposure to toxic water and chronic exposure to toxic
sediment in a stream and potentially extending to
Iliamna Lake.
Invertebrates and potentially fish would experience
acute exposure to toxic water and chronic exposure
to toxic sediment in a pond or other wetland.
Fish and invertebrates would experience acute
exposure to toxic water.
Acute toxicity would reduce the abundance and
diversity of invertebrates and possibly cause a fish
kill if spilled to a stream or wetland.
Frequent inspections and regular maintenance would
result in few impassable culverts.
If culvert failures revert to those seen in surveys of
roads, 11 to 21 salmonid streams would have
impeded fish passage.
Accidents that spill processing chemicals into a
stream or wetland could cause a fish kill.
Water collection and treatment failures are very likely
to result in exceedance of standards potentially
including death offish and invertebrates, but not
necessarily as severe or extensive as in the failure
scenario.
Collection and treatment failures are highly likely to
result in release of untreated or incompletely treated
leachates for days to months, but the water would be
less toxic due to elimination of PAG waste rock.
When water is no longer managed, untreated
leachates would flow to the streams. However, the
water would be less toxic due to elimination of PAG
waste rock.
a Because of differences in derivation, the probabilities are not directly comparable.
b Based on expected state safety requirements. Observed failure rates for earthen dams are higher (about 5 x 1CH per year or a recurrence
frequency of 2,000 years).
PAG = potentially acid-generating
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-18
April 2013
-------
Pebble 6.5 Footprint
Groundwater Drawdown Zone
Eliminated, Blocked, or Dewatered Streams
Eliminated, Blocked, or Dewatered Wetlands
1 2
Kilometers
1 2
Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-19
April 2013
-------
Figure ES 7. Salmon, Dolly Varden, and rainbow trout distribution along the transportation corridor. Designation of salmon presence is
based on the Anadromous Waters Catalog; designation of Dolly Varden and rainbow trout presence is based on the Alaska Freshwater Fish
Inventory.
N
A
10
] Kilometers
10
] Miles
Lake Clark
Note: Sampling intensity is greatly reduced away from the Pebble deposit area.
Streams without data points may not have been surveyed: thus, it is unknown
whether or not they provide suitable habitat for this species.
Approximate Location of Pebble Deposit
Transportation Corridor (Outside Assessment Area)
Dolly Varden (AFFI)
Rainbow Trout (AFFI)
Transportation Corridor
Transportation Corridor Area
Sub watersheds within Area
Salmon (AWC)
Cook Inlet
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-20
April 2013
-------
Figure ES 8. Height of the dam at TSF 1 in the Pebble 2.0 and Pebble 6.5 scenarios relative to U.S.
landmarks. The Pebble 0.25 and Pebble 2.0 TSFs are the two TSF failure scenarios evaluated in the
assessment.
260
240-
220-
200
180
160
£ 140
S 120
100
80
60
40
20
0
Transamerica Building - 260 Meters
Tailings Dam TSF 1 - 209 Meters
Gateway Arch -192 Meters
Washington Monument -169 Meters
Maximum Pebble 2.0 Tailings
Tailings Reservoir
Maximum Pebble 0.25 Tailings
The range of estimated probabilities of dam failure is wide, reflecting the great uncertainty concerning
such failures. The most straightforward method of estimating the annual probability of a tailings dam
failure is to use the historical failure rate of similar dams. Three reviews of tailings dam failures
produced an average rate of approximately 1 failure per 2,000 dam years, or 5 x 10~4 failures per dam
year. The argument against this method is that the record of failure does not fully reflect current
engineering practice. Some studies suggest that improved design, construction, and monitoring
practices can reduce the failure rate by an order of magnitude or more, resulting in an estimated failure
probability within the range assumed here. The State of Alaska's guidelines suggest that an applicant
follow accepted industry design practices such as those provided by the U.S. Army Corps of Engineers
(USAGE), the Federal Energy Regulatory Commission (FERC), and other agencies. Based on safety factors
in USAGE and FERC guidance, we estimate that the probability of failure for all causes requires a
minimum factor of safety of 1.5 against slope instability, for the loading condition corresponding to
steady seepage from the maximum storage facility. An assessment of the correlation of dam failure
probabilities with slope instability safety factors suggests an annual probability of failure of 1 in 250,000
per year for facilities designed, built, and operated with state-of-the-practice engineering (Category I
facilities) and 1 in 2,500 per year for facilities designed, built, and operated using standard engineering
practice (Category II facilities). The advantage of this approach is that it addresses current regulatory
guidelines and engineering practices. The disadvantage is that we do not know whether standard
practice or state-of-the practice dams will perform as expected, particularly given the potentially large
size of tailings dams and subarctic conditions in these scenarios.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
ES-21
April 2013
-------
Failure of the dam atTSF 1 (the only TSF in all three mine scenarios) would result in the release of a
flood of tailings slurry into the North Fork Koktuli River, scouring the valley and depositing many
meters of tailings fines in a sediment wedge across the entire valley near the TSF dam, with lesser
quantities of fines deposited at least as far as the North Fork's confluence with the South Fork Koktuli
River. The North Fork Koktuli River currently supports spawning and rearing populations of sockeye,
coho, and Chinook salmon; spawning populations of chum salmon; and rearing populations of Dolly
Varden and rainbow trout. The tailings slurry flood would continue down the mainstem Koktuli River
with similar effects, the extent of which cannot be estimated at this time due to model and data
limitations.
The tailings dam failures evaluated in the assessment would be expected to have the following severe
direct and indirect effects on aquatic resources, particularly salmonids.
• It is very likely that the North Fork Koktuli River below the TSF 1 dam and much of the mainstem
Koktuli River would not support salmonids in the short term (less than 10 years).
o There would be a complete loss of suitable salmon habitat in the North Fork Koktuli River along
at least 30 km (18.6 miles) of stream habitat, which was the spatial limit of the modeling
conducted for this assessment.
o Deposited tailings would degrade habitat quality for both fish and the invertebrates they eat.
Based largely on their copper content, deposited tailings would be toxic to benthic
macroinvertebrates, but existing data concerning toxicity to fish are less clear.
o Deposited tailings would continue to erode from the North Fork Koktuli River and mainstem
Koktuli River valleys.
o Suspension and redeposition of tailings would likely cause serious habitat degradation in the
mainstem Koktuli River and downstream into the Mulchatna River; however, the extent of these
effects cannot be estimated at this time due to model and data limitations.
• The affected streams would provide low-quality spawning and rearing habitat for a period of
decades.
o Recovery of suitable substrates via mobilization and transport of tailings would take years to
decades, and would affect much of the watershed downstream of the failed dam.
o Ultimately, spring floods and stormflows would carry some proportion of the tailings into the
Nushagak River.
o For some years, periods of high flow would be expected to suspend sufficient concentrations of
tailings to cause avoidance, reduced growth and fecundity, and even death offish.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES 22 April 2013
-------
• Near-complete loss of North Fork Koktuli River fish populations would likely result from these
habitat losses.
o The Koktuli River watershed is an important producer of Chinook salmon. The Nushagak River
watershed, of which the Koktuli River watershed is a part, is the largest producer of Chinook
salmon in the Bristol Bay region, with annual runs averaging over 190,000 fish.
o A tailings spill would be expected to eliminate 25% or more of the Chinook salmon run in the
Nushagak River due to loss of the Koktuli River watershed population; an additional 10 to 20%
could be lost due to tailings deposited in the Mulchatna River and its tributaries.
o Sockeye are the most abundant salmon returning to the Nushagak River watershed, with annual
runs averaging more than 1.9 million fish. The proportion of sockeye and other salmon species
of Koktuli-Mulchatna origin is unknown.
o Similarly, the North Fork Koktuli River populations of rainbow trout and Dolly Varden would be
lost for years to decades if they could not be successfully maintained entirely in headwater
networks upstream of the affected zone. Quantitative estimates of these losses are not possible
given available information.
Effects would be qualitatively similar for both the Pebble 0.25 and Pebble 2.0 dam failures, although
effects from the Pebble 2.0 dam failure would extend farther and last longer. Failure of dams at the two
additional TSFs under the Pebble 6.5 scenario (TSF 2 and TSF 3) were not modeled, but would have
similar types of effects in the South Fork Koktuli River and downstream rivers.
Pipeline Failures
Under the mine scenarios, the primary mine product would be a copper concentrate with traces of other
metals that would be pumped in a pipeline to a port on Cook Inlet. Water that carried the sand-like
concentrate would be returned to the mine site in a second pipeline. Based on the general record of
pipelines and further supported by the record of metal concentrate pipelines at existing mines, one to
two near-stream failures of each of these pipelines would be expected to occur over the life of the Pebble
6.5 scenario (approximately 78 years). Failure of either the product or the return water pipeline would
release water that is expected to be highly toxic due to dissolved copper with potential contributions to
toxicity by processing chemicals. Invertebrates, and potentially early life stages offish, would be killed in
the affected stream over a relatively brief period. If concentrate spilled into a stream, it would settle and
form highly toxic bed sediment based on its high copper content and acid generation. The mean
velocities of many streams crossed by the pipelines are sufficient to carry the concentrate downstream
to Iliamna Lake, but some would collect in low-velocity areas of the receiving stream. If the spill
occurred during low flows, dredging could recover some concentrate but would cause physical damage
to the stream. Concentrations in Iliamna Lake could not be predicted, but near the pipeline route,
Iliamna Lake contains important beach spawning areas for sockeye salmon that could be exposed to a
toxic spill. Sockeye also spawn in the lower reaches of streams that could be directly contaminated by a
spill.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES~23 April 2013
-------
The diesel fuel pipeline also would be expected to spill near a stream over the life of the mine. Based on
evidence from modeling the dissolved and dispersed oil concentrations in streams, laboratory tests of
diesel toxicity, and studies of actual spills in streams, a diesel spill at a stream crossing would be
expected to kill invertebrates and probably fish as well. Remediation would be difficult and recovery
would likely take 1 to 3 years. Failure of the natural gas pipeline would also be likely, but significant
effects on fish are unlikely.
Common Mode Failures
Multiple, simultaneous failures could occur as a result of a common event, such as a severe storm with
heavy precipitation (particularly one that fell on spring snow cover) or a major earthquake. Over the
long period that tailings impoundments, a mine pit, and waste rock piles would be in place, the
likelihood of multiple extreme precipitation events, earthquakes, or combinations of these events
becomes much greater. Multiple events further increase the chances of weakening and eventual failure
of facilities that remain in place.
Such an event could cause multiple dam failures that would spill tailings slurry into both the South and
North Fork Koktuli Rivers; road culvert washouts that would send sediments downstream and
potentially block fish passage; and pipeline failures that would release product slurry, return water, or
diesel fuel. The effects of each of these accidents individually would be the same as discussed previously,
but their co-occurrence would cause cumulative effects on salmonid populations and make any remedial
responses more difficult.
Fish-Mediated Risks to Wildlife
Although the effects of salmonid reductions on wildlife—that is, fish-mediated risks to wildlife—cannot
be quantified given available data, some reduction in wildlife would be expected under the mine
scenarios. Changes in the occurrence and abundance of salmon have the potential to change animal
behavior and reduce wildlife population abundances. The mine footprints would be expected to have
local effects on brown bears, wolves, bald eagles, and other wildlife that consume salmon, as a result of
reduced salmon abundance from the loss and degradation of habitat in or immediately downstream of
the mine footprint. Any of the accidents or failures evaluated would increase effects on salmon, which
would further reduce the abundance of their predators.
The abundance and production of wildlife also is enhanced by the marine-derived nutrients that salmon
carry upstream on their spawning migration. These nutrients are released into streams when the
salmon die, enhancing the production of other aquatic species that feed wildlife. Salmon predators
deposit these nutrients on the landscape, thereby fertilizing terrestrial vegetation, which, in turn,
provides food for moose, caribou, and other wildlife. The loss of these nutrients from a reduction in
salmon would likely reduce the production of riparian or upland species.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES~24 April 2013
-------
Fish-Mediated Risks to Alaska Native Cultures
Under routine operations with no major accidents or failures, the predicted loss and degradation of
salmonid habitat in South and North Fork Koktuli Rivers and Upper Talarik Creek would be expected to
have some impact on Alaska Native cultures of the Bristol Bay watershed. Fishing and hunting practices
are expected to change in direct response to the stream, wetland, and terrestrial habitats lost to the
mine footprints and the transportation corridor. It is also possible that subsistence use of salmon
resources could decline based on the perception of reduced fish or water quality resulting from mining.
The potential for significant effects on Alaska native cultures is much greater from mine failures that
would reduce or eliminate fish populations in affected areas, including areas significant distances
downstream from the mine. In the case of the TSF failures described in the assessment, the significant
loss of Chinook salmon populations would have severe consequences, especially for villages in the
Nushagak River watershed.
Any loss offish production from these failures would reduce the availability of these subsistence
resources to local Alaska Native villages, and the reduction of this highly nutritious food supply could
have negative consequences on human health. Because salmon-based subsistence is integral to Alaska
Native cultures, the effects of salmon losses go beyond loss of a food resource. If salmon quality or
quantity was adversely affected (or perceived to be affected), the nutritional, social, and spiritual health
of Alaska Natives and their culture would decline.
Cumulative Risks
This assessment has focused on the effects of a large mine at a single deposit on salmon and other
resources in the Nushagak and Kvichak River watersheds, including the cumulative effects of multiple
stressors associated with that mine. However, multiple mines and associated infrastructure may be
developed in these watersheds. Each mine would pose risks similar to those identified in the mine
scenarios. Estimates of the stream and wetland habitats lost would differ across different deposits,
based on the size and location of mine operations within the watersheds. Individually, each mine
footprint would eliminate some amount offish-supporting habitat and, should operator or engineering
failures occur, affect fish habitats well beyond the mine footprint.
We considered development of mines at several sites in the Nushagak River watershed, including the
Pebble South/PEB, Big Chunk South, Big Chunk North, Groundhog, AUDN/Iliamna, and Humble claims.
These sites were chosen because all contain copper deposits that have generated exploratory interest. If
all six mine sites were developed, the cumulative area covered by the six mine footprints could be 35 to
53 km2 (8,600 to 13,000 acres). Stream habitats lost to eliminated or blocked streams could be 41 to 64
km (25 to 40 miles). Cumulative wetland losses could be 7.4 to 25 km2 (1,800 to 6,100 acres).
These are conservative estimates of lost habitats, because we did not estimate the hydrologic drawdown
zones around each mine pit as was done for the Pebble scenarios. Inclusion of the drawdown area in the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES~25 April 2013
-------
Pebble 0.25 scenario increased the area of stream and wetlands losses by 84%. A similar increase might
be expected at the other mine sites, depending on local geology.
In addition, mine operations are assumed to be of average size, and would modify flows and diminish
water quality to approximately the same extent as the Pebble 0.25 scenario. Waters on these claim
blocks include the Chulitna River and Rock, Jensen, Yellow, Napotoli, Klutuk, and Kenakuchuk Creeks, as
well as over 250 unnamed tributaries and over 50 unnamed lakes and ponds. Although not all support
salmon, many do. Loss of substantial habitat across the watersheds could contribute to diminishing the
genetic diversity of salmon stocks and thereby increasing annual variability in salmon returns.
Mitigation and Remediation
The mine scenarios assessed here include modern conventional mitigation practices as reflected in
Northern Dynasty Mineral's published plan for the Pebble deposit, plus practices suggested in the
mining literature and consultations with experts. These practices include, but are not limited to,
processing all potentially acid-generating waste rock before closure, managing effluent water
temperatures, inspecting and maintaining roads daily, and providing automatic monitoring and remote
shut-off for the pipelines. However, we recognize that risks could be further reduced by unconventional
or even novel mitigation measures, such as dry stack tailings disposal or the use of armored tanks on the
trucks carrying process chemicals to the site. These practices may be unconventional because they are
expensive, unproven, or impractical. However, these obstacles to implementation might be overcome, as
justified by the large mineral resource and the highly valued natural and cultural resources of the Bristol
Bay watershed.
Although remediation would be considered if spills contaminated streams, features of the Pebble
deposit area would make remediation difficult. Pipeline crossings of streams would be near Iliamna
Lake, so the time available to block or collect spilled material would be short. Spilled return water and
the aqueous phase of the product concentrate slurry would be unrecoverable. The product concentrate
itself would resemble fine sand, and mean velocities in many receiving streams would be sufficient to
suspend and transport it. Hence, concentrate spilled or washed into streams could be recovered only
where it collected in low-velocity locations. Diesel spills would dissolve, vaporize, and flow as a slick to
Iliamna Lake. Booms and absorbents are not very effective in moderate- to high-velocity streams. Spilled
tailings from a dam failure would flow into streams, rivers, and floodplains that are in roadless areas and
not large enough to float a barge-mounted dredge. Recovery, transport, and disposal of hundreds of
millions of metric tons of tailings under those conditions would be extremely difficult and would result
in additional environmental damage. Compensatory mitigation measures could offset some of the
stream and wetland losses, although there are substantial challenges regarding the efficacy of these
measures to offset adverse impacts.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES 26 April 2013
-------
Summary of Uncertainties in Mine Design and Operation
This assessment considers realistic mine scenarios that are based on specific characteristics of the
Pebble deposit and plans proposed by Northern Dynasty Minerals and are generally applicable to
copper deposits in the Bristol Bay watershed. If the Pebble deposit is mined, actual events will
undoubtedly deviate from these scenarios. This is not a source of uncertainty, but rather an inherent
aspect of a predictive assessment. Even an environmental assessment of a specific plan proposed for
permitting by a mining company would be an assessment of a scenario that undoubtedly would differ
from the actual development.
Multiple uncertainties are inherent in planning, designing, constructing, operating, and closing a mine.
These uncertainties, summarized below, are inherent in any complex enterprise, particularly when that
enterprise involves an incompletely characterized natural system. However, the large scales and long
durations required of mining the Pebble deposit make these inherent uncertainties more prominent.
• Mines are complex systems requiring skilled engineering, design, and operation. The uncertainties
facing mining and geotechnical engineers include unknown geologic features, uncertain values in
geological properties, limited knowledge of mechanisms and processes, and human error in design
and construction. Models used to predict the behavior of engineered systems represent idealized
processes, and by necessity contain simplifications and approximations that potentially introduce
errors.
• Accidents are unplanned and inherently unpredictable. Although systems can be put into place to
protect against system failures, seemingly logical decisions about how to respond to a given
situation can have unexpected consequences due to human error (e.g., the January 2012 overflow of
the tailings dam at the Nixon Fork Mine near McGrath, Alaska). Further, unforeseen events or events
that are more extreme than anticipated can negate apparently reasonable operation and mitigation
plans.
• The ore deposit would be mined for decades, and wastes would require management for centuries
or even in perpetuity. Engineered waste storage systems of mines have been in existence for only
about 50 years and their long-term behavior is not known. The response of the current technology
in the construction of tailings dams is untested and unknown in the face of centuries of extreme
events such as earthquakes and weather.
• Mine management or ownership may change over time. Over the long time span (centuries) of
mining and post-mining care, generations of mine operators must exercise due diligence. Priorities
are likely to change in the face of financial circumstances, changing markets for metals, new
information about the resource, political priorities, or any number of currently unforeseeable
changes in circumstance.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES 27 April 2013
-------
Summary of Uncertainties and Limitations in the Assessment
The most important uncertainties concerning estimated effects of the mine scenarios, as judged by the
assessment authors, are identified below.
• Consequences of the loss and degradation of habitat on fish populations could not be quantified
because of the lack of quantitative information concerning salmonid populations in freshwater
habitats. The occurrence of salmonid species in rivers and major streams is known, but information
on abundances, productivities, and limiting factors in each of the watersheds is not available.
Estimating changes in populations would require population modeling, which requires knowledge
of life-stage-specific survival and production and of the limiting factors and processes. Further, it
requires knowledge of how temperature, habitat structure, prey availability, density dependence,
and sublethal toxicity influence life-stage-specific survival and production. Obtaining this
information would require more detailed monitoring and experimentation. Furthermore, salmon
populations naturally vary in size due to many factors that vary among locations and years. At
present, data are insufficient to establish reliable salmon population estimates and obtaining such
data takes many years. Estimated effects of mining on fish habitat thus become the surrogate for
estimated effects on fish populations.
• Standard leaching test data are available for test tailings and waste rocks from the Pebble deposit,
but these results are uncertain predictors of the actual composition of leachates from tailings
impoundments, tailings deposited in streams and on their floodplains, and waste rock piles.
• Capture efficiencies for leachates are uncertain. For waste rock outside of the mine pit drawdown
zone, we assume 50% capture. To avoid exceeding water quality criteria for copper, more than 99%
capture would be required.
• The effects of tailings and product concentrate deposited in spawning and rearing habitat are
uncertain. It is clear that they would have harmful physical and toxicological effects on salmonid
larvae or sheltering juveniles, but the concentration in spawning gravels required to reduce
salmonid reproductive success is unknown.
• The estimated annual probability of tailings dam failure is uncertain because it is based on design
goals. Historical experience is presumed to provide an upper bound. Features that should reduce
failure frequencies have not been tested for the thousands of years that they must function properly.
Hence, actual failure rates could be higher or lower than the estimated probability.
• The proportion of tailings that would spill in the event of a dam failure could be larger than the
largest value modeled (20%).
• The long-term fate of spilled tailings in the event of a dam failure could not be quantified. As in other
cases, it is likely that tailings would erode from areas of initial deposition and move downstream
over more than a decade. However, the data needed to model that process and the resources needed
to develop that model are not available.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES~28 April 2013
-------
• The actual response of Alaska Native cultures to any impacts of the mine scenarios is uncertain.
Interviews with village Elders and culture bearers and other evidence suggest that responses would
involve more than the need to compensate for lost food, and would likely include some degree of
cultural disruption. It is not possible to predict specific changes in demographics, cultural practices,
or physical and mental health.
Uses of the Assessment
This assessment is a scientific investigation. It does not reflect any conclusions or judgments about the
need for or scope of government action, nor does it offer or analyze options for future decisions. Rather,
it is a scientific product intended to provide a characterization of the biological and mineral resources of
the Bristol Bay watershed, increase understanding of the risks from large-scale mining to the fish
resources, and inform future government decisions. The assessment will also better inform dialogue
among interested stakeholders concerning the resources in the Bristol Bay watershed and the impacts
of large-scale mining on those resources.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ES 29 April 2013
-------
The Bristol Bay watershed in southwestern Alaska supports the largest sockeye salmon fishery in the
world, is home to 25 federally recognized tribes, and contains abundant natural resources, including
large mineral reserves. Worldwide attention to this watershed has increased because of widespread
mineral exploration activities and the discovery of a large ore deposit in the watershed's northeast-
central region. The potential for large-scale mining activities has raised concerns about the quality and
sustainability of Bristol Bay's world-class fisheries and the future of Alaska Natives who have
maintained a salmon-based culture and a subsistence-based way of life for at least 4,000 years.
Public interest in the Bristol Bay watershed has centered on the ecological goods and services provided
by the watershed and on potential mining activity. The watershed is most noted for its abundant fish
resources. The Bristol Bay watershed supports production of all five species of Pacific salmon found in
North America (sockeye, Chinook, chum, coho, and pink), including almost half of the world's
commercial sockeye salmon harvest. In 2009, Bristol Bay's ecosystems, which support the watershed's
commercial, recreational, and subsistence fisheries, generated $480 million in direct annual economic
expenditures in the region, and provided employment for over 14,000 full- and part-time workers. This
consistently large fish production results from the watershed's high hydrologic diversity and pristine
quality, both of which contribute to highly diverse fish populations.
In addition to these biological resources, 16 mine claim blocks have recently been explored in the Bristol
Bay's Nushagak and Kvichak River watersheds. Eleven of these claims are associated with porphyry
copper deposits, the largest belonging to the Pebble Limited Partnership (PLP). This partnership was
created in 2007 by co-owners Northern Dynasty Minerals, Ltd., and Anglo American to design, permit,
construct, and operate a long-life mine at the Pebble deposit (PLP 2013). Although PLP has notyet
submitted a permit application for a mine, preliminary mine plans have been developed and publicly
available information strongly suggests that a mine at the Pebble deposit has the potential to become
one of the largest mining developments in the world. The Pebble deposit is a large, low-grade deposit
containing copper, gold, and molybdenum-bearing minerals. Extraction is expected to require the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
1-1
April 2013
-------
Chapter 1 Introduction
creation of a large open pit, the production of large amounts of waste rock and mine tailings, the
creation of a transportation corridor connecting the deposit area to Cook Inlet, and the development of a
deep-water port. Revenues from such a mine have been estimated at between $300 billion and
$500 billion over the mine's life.
In light of these factors, nine Bristol Bay federally recognized tribes, the Bristol Bay Native Association,
the Bristol Bay Native Corporation, other tribal organizations, and many groups and individuals
petitioned the U.S. Environmental Protection Agency (USEPA) to restrict or prohibit the disposal of
dredged or fill material associated with large-scale mining activities in the Bristol Bay watershed, using
its authorities under Clean Water Act (CWA) Section 404(c). These groups are concerned that large-
scale mining could adversely affect the region's valuable natural resources, particularly its fisheries.
Four Bristol Bay federally recognized tribes, other tribal organizations, the governor of Alaska, and
others groups and individuals, including PLP, have asked USEPA to wait until formal mine permit
applications have been submitted and an environmental impact statement developed.
USEPA initiated this assessment in response to these competing requests. The assessment's purpose is
to characterize the biological and mineral resources of the Bristol Bay watershed, increase
understanding of the potential impacts of large-scale mining on the region's fish resources, and inform
future governmental decisions related to protecting and maintaining the physical, chemical, and
biological integrity of the watershed. The assessment represents a review and synthesis of available
information to identify and evaluate potential risks of future large-scale mining development on the
Bristol Bay watershed's fish habitats and populations and subsequent indirect effects on the region's
wildlife and Alaska Native cultures.
1.1 Assessment Approach
This assessment of potential large-scale mining in the Bristol Bay watershed was conducted as an
ecological risk assessment (ERA). ERA is a scientific process used to determine whether exposure to one
or more stressors may result in adverse ecological effects, the findings of which are used to inform
environmental decision making. USEPA routinely uses ERA methods to evaluate the potential impacts of
current and future actions when considering management decisions (USEPA 2002a, 2002b, 2002c, 2008,
2011). USEPA is conducting this assessment consistent with its authority under CWA Section 104. CWA
Sections 104(a) and (b) provide USEPA with the authority to study the resources of the Bristol Bay
watershed, evaluate the effect of pollution from large-scale mining development on those resources, and
make such an assessment available to the public. This assessment is not an environmental impact
assessment, an economic or social cost-benefit analysis, or an assessment of any one specific mine
proposal.
Risk assessors, decision makers, and community stakeholders determine the topical, spatial, and
temporal scope needed to effectively address the decision the ERA is informing. Within this scope, risk
assessments consider the potential effects of an activity and use one or more scenarios, or sets of
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T April 2013
-------
Chapter 1 Introduction
assumptions, to identify how resources of interest (in this case, fish habitat and populations) could be
exposed to stressors generated by some activity (in this case, porphyry copper mining).
Assessment endpoints are explicit expressions of the environmental resources of interest to the risk
assessors, decision makers, and stakeholders. We selected fish as the primary endpoint because of its
universal importance to stakeholders and future decision making in the watershed. The sustainability of
the Bristol Bay fishery is a concern shared by all Bristol Bay stakeholders—including those who support
mining—and the ecological, economic and cultural importance of the region's commercial, sport and
subsistence fisheries has been emphasized consistently by all stakeholders throughout the process. Our
preliminary technical consultations with federal, state, and tribal representatives indicated that
evaluating the potential risks of large-scale mining on the region's fishery resources was a top priority.
During our public engagement efforts, stakeholders consistently emphasized that fish are the crucial
resource of concern. Thus, because of its critical importance to stakeholders and future decision making
in the watershed, we selected fish, and specifically salmon and other salmonid fishes, as our primary
assessment endpoint.
We also considered two key secondary endpoints: wildlife and Alaska Native cultures. Fish-mediated
effects on wildlife were considered because fish, particularly salmon, are an important food resource for
wildlife, via both direct consumption and as a source of marine-derived nutrients that contribute to the
overall productivity of the watershed. Fish-mediated effects on Alaska Natives were considered because
sustainability of the region's fish populations is critical to the future of Alaska Natives in Bristol Bay, and
because concern about the region's fishery resources prompted the original requests from Alaska
Natives that USEPA examine issues in the Bristol Bay watershed.
Multiple geographic scales are considered in the assessment. Background and characterization
information is presented for the entire Bristol Bay watershed. The evaluation of potential large-scale
mining impacts focuses on the Nushagak and Kvichak River watersheds. These two watersheds are the
most likely to be affected by large-scale mining development, given the location of current mine claims
and current federal and state restrictions on development in other portions of the Bristol Bay
watershed. These two watersheds are responsible for approximately half of the Bristol Bay sockeye
salmon production, and are also home to approximately half of the region's Alaska Native communities
and federally recognized tribes. There are 31 federally recognized tribes in the Bristol Bay area, and 25
of these tribal communities are within the watershed boundary defined in this assessment. Fourteen of
these communities (13 of which have federally recognized tribes) are within the Nushagak and Kvichak
River watersheds.
The assessment also considers smaller geographic scales for risk analysis and characterization. Because
the Pebble deposit is the largest known and most explored deposit in the region, we use it as a case
study for potential risks. Because none of the parties holding mine claims in the Bristol Bay watershed
has submitted a formal application and mine plan, we developed a set of realistic mine scenarios for the
assessment. The foundations for these scenarios are industry documents outlining approaches for
mining porphyry copper deposits, as well as specific documents from the PLP outlining a basic,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T April 2013
-------
Chapter 1 Introduction
preliminary mine plan for the Pebble deposit. The mine scenarios were used to complete the risk
analyses and characterizations in the assessment. Although these mine scenarios were developed for the
Pebble deposit, the potential risks evaluated are expected to be qualitatively similar to potential risks
associated with any mine of the same resource type (porphyry copper) anywhere in the Nushagak and
Kvichak River watersheds.
Throughout the assessment, we have reached out to interested parties to ensure transparency of the
assessment process (Box 1-1). Through public comment opportunities and by engaging an
Intergovernmental Technical Team (IGTT) of federal, state, and tribal representatives, we were able to
identify additional information helpful for characterizing the biological and mineral resources of the
watershed. These interactions with members of the community were also helpful in narrowing the
scope of the assessment to what was most important to stakeholders.
Detailed background characterizations for the resources of the watershed are included in the
assessment's appendices. We used these background characterization studies and input from the IGTT
to develop a series of conceptual models illustrating potential linkages between sources and stressors
associated with large-scale mining and the assessment endpoints. These models were then used to
develop a plan for analyzing and characterizing risks. During the analysis, available data were used to
assess potential exposure to stressors and potential effects on assessment endpoints stemming from
those exposures. In the final phase, results of these analyses were integrated to provide a
comprehensive picture of the risks to assessment endpoints (within the defined scope of the
assessment). The uncertainties and limitations associated with these analyses are also identified.
1.2 Use of this Assessment
This assessment is a scientific investigation. It does not reflect any conclusions or judgments about the
need for or scope of possible government action, nor does it offer or analyze options for future decisions.
Rather, it is a scientific product intended to provide a characterization of the biological and mineral
resources of the Bristol Bay watershed, increase understanding of the potential risks to fish resources
from large-scale mining, and inform future government decisions.
USEPA and other stakeholders may use this assessment in several ways. The assessment will inform the
public and interested government entities about the biological and mineral resources of the Bristol Bay
watershed. Much of the information about these resources was previously found in a variety of sources.
With this assessment, we have synthesized and integrated available literature and provided a useful
summary characterizing the Bristol Bay watershed's resources.
The assessment will inform the public and interested government entities about the potential impacts of
large-scale mining. USEPA recognizes the high level of interest concerning the impacts of potential mine
development on the watershed's ecological resources. That interest originates from Alaska Native
communities within the watershed, other Alaska residents, and interested parties throughout the United
States. It is expressed by those interested in protecting the Bristol Bay fishery and by those interested in
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~T April 2013
-------
Chapter 1 Introduction
developing the watershed's large mineral resources. This assessment is a scientific and technical
resource that is useful to members of the public as they weigh the challenges of both mining and
protecting the ecological resources in the Bristol Bay watershed.
Our findings concerning the potential impacts of large-scale mining help to inform future government
decisions regarding mine development in the Bristol Bay watershed and potential actions to protect and
maintain the integrity of the watershed's aquatic resources. One of the initiators for the assessment was
the multiple petitions to USEPA to use its authority under Section 404(c). It is expected that the
assessment will provide an important base of information for any agency decision about whether or not
to use CWA Section 404(c), either now or in the future, and will facilitate a thoughtful decision regarding
whether application of this authority is or is not warranted.
The assessment may also assist federal and state scientists and resource managers involved in the
evaluation of future applications for mine permits submitted for the deposits in the Bristol Bay
watershed. It is likely that future mines in the watershed would require the filling of streams and
wetlands and would require a Section 404 permit be obtained from the U.S. Army Corps of Engineers
(USAGE). USEPA review and comments on proposed Section 404 permit applications and this
assessment will be a valuable resource in the development and review of such permit applications.
This assessment will also inform any future environmental assessment conducted under the National
Environmental Policy Act (NEPA) related to mining in the Bristol Bay watershed. If a Section 404 permit
or other major federal action is required for a future mine in the watershed, it would trigger review of
the proposed mine under NEPA. This assessment, particularly its identification and analysis of the
direct, indirect, and cumulative effects of large-scale mining, will be a valuable resource in the
development and review of any environmental impact statement for mines proposed in the watershed.
This assessment is also likely to lead to a more efficient and timely NEPA review.
Perhaps the most important use of this assessment is to better inform dialogue among interested
stakeholders concerning the resources in the Bristol Bay watershed and the potential impacts of large-
scale mining on those resources.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , K April 2013
_L~O
-------
Chapter 1 Introduction
BOX 11. STAKEHOLDER INVOLVEMENT IN THE ASSESSMENT
Meaningful engagement with stakeholders was essential during development of the assessment to ensure that the
U.S. Environmental Protection Agency (USEPA) heard and understood the full range of perspectives on the draft
assessment itself and the potential effects of mining in the region. USEPA has used a variety of tools to involve and
inform stakeholders prior to and during release of the draft assessment, including a community involvement plan to
ensure that a robust outreach effort is in place and a project webpage and listserv to ensure that assessment-
related information is shared with the public. Additional ways in which stakeholders and tribal governments were
involved in the assessment process are summarized below.
• Meetings with stakeholders. Prior to the release of the draft assessment, USEPA visited many Bristol Bay
communities and met with representatives from Bristol Bay tribes, tribal corporations, the mine industry,
commercial fisherman, seafood processors, hunters and anglers, chefs and restaurant owners, jewelry
companies, conservation organizations, members of the faith community, and elected officials from Alaska and
other states. USEPA heard from hundreds of people at these meetings and from thousands more via phone and
email.
• Intergovernmental Technical Team. In March 2011, USEPA invited representatives from tribes and state and
federal agencies to participate on an Intergovernmental Technical Team (IGTT), which was established to
provide USEPA with input on the structure of the assessment and to identify potential data sources. IGTT
participants included tribal representatives from Ekwok, Newhalen, Iliamna, South Naknek, New Koliganek,
Curyung, Nondalton, and Levelock and agency representatives from the Alaska Department of Public Health, the
Alaska Department of Fish and Game, the National Park Service, the U.S. Fish and Wildlife Service, the National
Oceanic and Atmospheric Administration, and the Bureau of Land Management. In August 2011, USEPA held a
2-day workshop to share information and solicit feedback from the IGTT on the assessment effort; this feedback
was used to inform the early stages of problem formulation. USEPA also updated the IGTT on assessment
progress in January 2012, via webinar.
• Tribal consultation. USEPA's policy is to consult on a government-to-government basis with federally
recognized tribal governments when USEPA actions and decisions may affect tribal interests. Consultation is a
process of meaningful communication and coordination between USEPA and tribal officials. In February 2011,
USEPA invited all 31 federally recognized tribal governments of the Bristol Bay watershed to enter formal
consultation on the assessment, to ensure their involvement and to include their concerns and relevant
information in the assessment. Not all tribes elected to participate in consultation. Currently, USEPA has met
with representatives from 18 of the 31 village councils, either in person or on the phone.
• Public meetings. Between February 2011 and May 2012, USEPA met with many Bristol Bay communities,
tribal governments and representatives, and organizations, including the communities of Ekwok, Dillingham,
Kokhanok, NewStuyahok, Koliganek, Iliamna, Newhalen, Nondalton, Naknek, King Salmon, Igiugig, and
Levelock; the Iliamna Development Corporation; Nuna Resources; the Bristol Bay Native Corporation; Nunamta
Aulukestai; Iliamna Natives Ltd.; the Alaska Peninsula Corporation; the Pedro Bay Native Corporation; the Lake
and Peninsula Borough; Paug-vik; Dillingham City Schools; Trout Unlimited; and the Bristol Bay Seafood
Processors Association. USEPA was also invited to participate in numerous conferences and meetings to discuss
the assessment.
• Public comments: Following release of the draft assessment on May 18, 2012, USEPA held a 60-day public
comment period. Approximately 233,000 comments were submitted to the USEPA docket during this period,
and all of these comments can be accessed online. In addition, USEPA held public meetings during the first
week of June 2012 in Dillingham, Naknek, NewStuyahok, Nondalton, Levelock, Igiugig, Anchorage, and Seattle,
to hear spoken public comments, collect written comments, and share information. In total, approximately
2,000 people attended these meetings. An overview of these meetings was shared via two webinars in July
2012.
• Public involvement in peer review: USEPA provided multiple opportunities for stakeholder involvement in the
peer review process. In February 2012, the public was invited to nominate qualified scientists as potential peer
reviewers; these nominations were submitted to the peer review contractor for consideration. In March 2012,
USEPA requested public comments on the charge questions to be given to peer reviewers, and these questions
were revised in response to comments received. In August 2012, the public was invited to participate in the first
2 days of the peer review meeting in Anchorage. On the first day of this meeting, over 100 members of the
public provided oral comments to the peer reviewers; on the second day, members of the public observed
discussions among the peer reviewers.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 77April 2013
-------
2.1 Structure
We based this assessment on U.S. Environmental Protection Agency (USEPA) guidelines for ecological
risk assessment (ERA) (USEPA 1998). We began by reviewing existing literature to synthesize
background information on Bristol Bay, particularly the Nushagak and Kvichak River watersheds. This
information focused on several topics, including the ecology of Pacific salmon and other fishes; the
ecology of relevant wildlife species; mining and mitigation, particularly in terms of porphyry copper
mining; potential risks to aquatic systems due to road and pipeline crossings; fishery economics; and
Alaska Native culture. These detailed background characterizations are provided in the appendices to
this assessment.
In accordance with the different phases of an ERA, the assessment document itself is organized into two
main sections: Problem Formulation (Chapters 2 through 6) and Risk Analysis and Characterization
(Chapters 7 through 14). Problem formulation is the first phase of an ERA, during which the purpose
and scope of the assessment are defined (USEPA 1998). Risk assessors, decision makers, and
stakeholders determine the topical, spatial, and temporal scope needed to effectively address whatever
decision process the assessment is meant to inform. Assessment endpoints, or explicit expressions of the
environmental entities of interest (USEPA 1998), are identified. Conceptual models illustrating potential
linkages among sources, stressors, and endpoints considered in the assessment, as well as a plan for
analyzing and characterizing risks, are developed (Box 2-1).
The risk analysis and characterization phases follow problem formulation (USEPA 1998). During the
risk analysis phase, available data are used to assess potential exposure to stressors and potential
effects on assessment endpoints stemming from those exposures. In the risk characterization phase,
information on exposures and effects is integrated, and the uncertainties and limitations associated with
the assessment's analyses are identified.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-1
April 2013
-------
Chapter 2
Overview of Assessment
BOX 2 1. CONCEPTUAL MODELS
Throughout this assessment we use conceptual model diagrams to illustrate potential ways in which large-
scale mine development may adversely affect the Bristol Bay watershed's biota and Alaska Native cultures.
These conceptual model diagrams show hypothesized pathways linking common sources associated with
mining to potential stressors, and those stressors to potential responses of interest. Inclusion of a pathway
indicates that the pathway can occur, not that it will definitely occur. Thus, these diagrams are not meant to
illustrate worst-case scenarios in which all pathways occur simultaneously; rather, they are meant to provide
overviews of potential linkages among sources, stressors, and responses, one or more of which may
plausibly result from mine development.
The conceptual model diagrams contain the following elements
(note that not all elements are found in each diagram).
Sources: entities associated with mining that may directly or
indirectly result in one or more stressors.
Steps in causal pathways: processes or states that may link
sources to stressors or stressors to responses.
Stressors: physical or chemical entities that may directly induce a
response of concern.
Modifying factors: processes, states, or other factors that may
influence the delivery, expression, or effect of stressors.
Biotic responses: potential effects on salmon, other fish, and
wildlife.
Human responses: potential effects on Alaska Native people and
culture.
When viewing these diagrams, it helps to keep the following
principles in mind:
Arrows leading from one shape to another indicate a hypothesized
cause-effect relationship, whereby the first (or originating) shape
could plausibly cause or result in the second shape.
Arrows leading from a shape to another arrow indicate that the
originatingshape (always categorized as a modifying factor) could
plausibly influence the cause-effect relationship illustrated by the
second arrow (e.g., by increasing or decreasing its probability or
intensity of occurrence).
Shapes bracketed under another shape are specific examples of
the more general shape under which they appear.
Bold lines, arrows, and outlines indicate high-priority pathways that
were evaluated; dashed lines, arrows, and outlines indicate lower-
priority pathways that were not evaluated.
2.1.1 Data Used in the Assessment
An ERA requires data of sufficient quantity and quality, from a variety of sources. Throughout the
problem formulation, risk analysis, and risk characterization phases, relevant data are identified and
acquired. These data may result from different kinds of studies, including field studies at the site of
interest, field studies at other sites somehow relevant to the site or issue of interest, laboratory tests,
and modeling applications.
In this assessment, we prioritized peer-reviewed, publicly accessible sources of information to ensure
that the data we incorporated were of sufficient quality. In many cases, however, peer-reviewed data—
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-2
April 2013
-------
Chapter 2 Overview of Assessment
particularly those directly relevant to potential mining in the Bristol Bay region—were not available.
Thus, we also incorporated non-peer-reviewed data from government sources, most notably the State of
Alaska (e.g., Alaska Department of Fish and Game [ADF&G], Alaska Department of Natural Resources
[ADNR]). When data collected under the auspices of the Pebble Limited Partnership (PLP) (e.g., as
presented in Ghaffari et al. 2011, PLP 2011) were the only data available, those data were included in
the assessment; PLP is currently conducting its own peer review of the data presented in the
Environmental Baseline Document2004 through 2008 (EBD) (PLP 2011). Other non-governmental
organizations have collected data specific to the Pebble deposit site. USEPA subjected some of these
documents to external peer review before incorporating this information into the assessment. Some
minor sources of information were used without peer review. In all cases, data included in the
assessment are appropriately cited (Chapter 15) to identify sources of information.
2.1.2 Types of Evidence and Inference
As in other ERAs, the risk analysis and characterization phase of this assessment is based on weighing
multiple types of evidence. Available and relevant pieces of evidence from a variety of sources are used
to follow different lines of inference and reach the best-supported conclusions.
In this risk analysis, we use general scientific knowledge, mathematical and statistical models, and data
from the Bristol Bay region, other sites (e.g., mines in other regions), and laboratory studies to evaluate
potential consequences of mine scenarios—that is, realistic potential mines of different sizes, the
characteristics of which are based largely on a mining company report (Ghaffari et al. 2011)—in terms
of sources, exposure to different stressors, and exposure-response relationships. First, we estimate the
magnitude of exposure potentially resulting from the mine scenarios, such as elevated aqueous copper
concentrations, kilometers of streams eliminated, and kilometers of streams upstream of road crossings.
Then, we consider the effects of these exposures—that is, the exposure-response relationships—on our
endpoints of interest (e.g., the relationship between water withdrawal and loss of salmon habitat,
concentration-response relationships for copper and fish). We describe and quantify, where possible,
exposure-response relationships for the endpoints and estimated exposures. For some issues, multiple
lines of evidence are available (e.g., state standards, federal criteria, effects models, field studies, and
toxicity tests as lines of evidence for copper toxicity); for other issues, lines of evidence are more limited.
Evidence from existing mines and other analogous facilities is used where relevant. Prior mining
activities in comparable watersheds provide examples of what can happen to the environment when
metals are mined. Some components of our mine scenarios have analogues in other industries (e.g., oil
and gas pipelines). These inferences by analogy reduce the uncertainties that come with modeling and
prediction, but introduce other uncertainties related to industry-specific or site-specific differences in
environmental conditions and potential changes in practices. Because no analogue is similar in all
aspects to potential mines and their components in the Bristol Bay region, we choose analogues to fit the
specific issues being assessed, and take care to use analogues that are defensible despite their
differences from our mine scenarios. For example, the Fraser River watershed can be considered an
analogous system to the Bristol Bay watershed because it has similar mines and a similar salmon
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TT April 2013
-------
Chapter 2 Overview of Assessment
resource, but we also recognize there are important differences between these systems (e.g., extensive
urban development, forestry, and agriculture in the Fraser River watershed). Metal mines in the Rocky
Mountain metal belt (e.g., sites at Coeur d'Alene River, Idaho, and Clark Fork, Montana) were developed
using mining practices that would not be allowed under current mining laws. However, the fate and
effects of tailings in streams and floodplains at these sites, which also supported trout and salmon
populations, offer some parallels to the fate and effects of tailings following potential tailings dam
failures in the Bristol Bay region, should they occur—even if the underlying causes of failure differ.
The use of data from the historical, operational records of mines, pipelines, and roads is necessary but
controversial. It is essential and conventional for risk assessments to use the history of a technology to
estimate failure rates. However, developers argue, with some justification, that the record of older
technology is not relevant because of technological advances. Despite advances, no technology is perfect,
and rates of past failures may be a better guide to future outcomes than the expectation that developers
can design a system that will not fail. A classic example is the NASA space shuttle program, which denied
the relevance of the failure rate of solid rocket boosters and declared that the shuttle's rate of failure on
launch would be one in a million. The Challenger failure showed that the prior failure rate was still
relevant, despite updated technology.
For most potential failures, historical failure rates are the only available evidence. New technologies
typically have not been in use long enough or widely enough to provide failure rates, and measures to
correct past failure modes may unwittingly introduce new failure modes. Thus, in this assessment we
choose failure rates that are most relevant and interpret them cautiously, using them to provide an
upper bound estimate of future failure rates.
After these analyses and lines of evidence are presented, we characterize risk for each line of evidence
by combining exposures and exposure-response relationships to estimate effects, and by considering
uncertainties. We weigh different lines and types of evidence based on evidence strength and quality.
The resulting qualitative or quantitative estimates of risk and uncertainty are based on either the best
line of evidence or a combined estimate from multiple lines of evidence and inferences. Bounding
analyses, which set upper and lower limits for key parameters, are used to express uncertainties
concerning future mine activities and their effects. In particular, multiple mine sizes and durations are
included in the mine scenarios (Chapter 6). Bounding is also used to express stochasticity. For example,
the occurrence and magnitude of tailings dam failures are random variables that cannot be reasonably
defined. Hence, a range of tailings dam failure probabilities and a range of tailings release magnitudes
are evaluated (Chapter 9).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment » 4 APril 2013
-------
Chapter 2 Overview of Assessment
2.2 Scope
2.2.1 Topical Scope
Construction and operation of a large-scale mining operation require the development of extensive
infrastructure and involve numerous processes and components, each of which may have repercussions
for receiving environments. In this assessment, we do not consider all potential sources of risk
associated with the development of large-scale mining in the Bristol Bay watershed, all the stressors
that may result from these sources, and all the endpoints that may be affected. Rather, we focus on a
more limited set of sources, stressors, and endpoints based on decision-maker needs (Chapter 1). These
focal components are described in broad terms below. In Chapters 3 through 6, we consider these
components in greater detail, and more specifically define the focus of the assessment—in terms of
geographic region, type of mining development, and ecological endpoints—for risk analysis and
characterization purposes.
In terms of sources, we consider the mine infrastructure and transportation corridor components of a
large-scale surface mining operation (Figure 2-1). Certain sources associated with mining but not
directly related to mine operations are not evaluated here, including power generation and transmission
facilities and activities, ancillary facilities such as housing for mine workers and wastewater treatment
plants to serve an increased human population, and construction and operation of a deep-water port at
Cook Inlet (Figure 2-1). A thorough evaluation of induced development—development that is not part of
the mine project, but for which the mine project provides the impetus or opportunity, such as residential
and commercial growth resulting from increased accessibility—is also outside the scope of this
assessment, although its importance is considered qualitatively in Chapter 13.
In terms of stressors, we focus on potential environmental effects on freshwater habitats (Figure 2-1).
We focus on freshwater habitats, because the most exceptional ecological feature of the Bristol Bay
watershed is its fish populations, and these populations are intimately linked to the watershed's
freshwater habitats. Although we recognize that large-scale mining could also have significant direct
impacts on terrestrial and marine systems, as well as direct economic and cultural repercussions, we do
not evaluate these impacts here.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment » 5 APril 2013
-------
Chapter 2
Overview of Assessment
Figure 2 1. Conceptual model illustrating sources, stressors, and responses potentially associated with large scale mine development in
the Bristol Bay watershed. Pathways explicitly evaluated in this assessment are in bold; dashed pathways may be considered qualitatively in
parts of the assessment, but are generally considered outside its scope. See Box 2 1 for a general discussion of how conceptual models are
used and structured in the assessment.
power generations '; [other ancillary "\ [transportation
transmission facilities,' ', facilities • L corridor
" l~
port
V
induced l|
development /
LEGEND
Polcl arrows & outlines indicate
topics within assessment's scope
dashed arrows^ outlines indicate
topics outside scope of assessment
environmental impacts
on freshwater habitats
environmental impacts
on terrestrial habitats
environmental impacts
on marine habitats
1 other ;
1 environmental impacts ]
i
economic
impacts
i
cultural
impacts
/
,.. y.
,-' effects on
'- other biota
effects on
Alaska Native culture
effects on
hum an health
f"" other effects on "%
"5*^ Alaska residents .•'
effects on
recreational sectors
effects on
commercial fisheries ,-'
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-6
April 2013
-------
Chapter 2 Overview of Assessment
Given the ecological and cultural significance of fishery resources in the Bristol Bay watershed, and the
fact that the health and sustainability of the watershed's fish populations are primary concerns shared
by all stakeholders interested in the Bristol Bay area, including those who support mining, we focus on
effects on salmonid fishes (Box 2-2) and resulting secondary effects on wildlife and Alaska Native
cultures as assessment endpoints (Chapter 5). Direct effects of mining on wildlife and Alaska Native
cultures, while potentially significant, are not evaluated in this assessment. For example, construction
and operation of a transportation corridor would likely directly affect wildlife populations (Forman and
Alexander 1998); however, because the assessment focuses on freshwater habitats, these direct wildlife
effects are not considered here. The only effects on wildlife and Alaska Native cultures evaluated in the
assessment are those resulting from impacts on fish populations (Chapter 12). We also recognize that
many other endpoints may be directly affected by large-scale mining operations, including other biota
(e.g., vegetation, small mammals), other recreational and commercial fisheries, and human health
(Figure 2-1), butthese topics are also outside the scope of the assessment.
It is important to keep in mind that exclusion of a source, stressor, or endpoint from this assessment
does not imply that it would be insignificant or unaffected. We recognize that many of the pathways we
identify as outside of the assessment's scope could have significant repercussions for the region's biota
and people.
2.2.2 Spatial Scales
Throughout this assessment, we consider data across five spatial scales (Table 2-1, Figure 2-2).
• The Bristol Bay watershed (Scale 1, Figure 2-3) includes all the basins and waterways that flow into
Bristol Bay.
• The Nushagak and Kvichak River watersheds (Scale 2, Figure 2-4) include those drainage areas that
contain stream segments that flow either directly or via downstream segments into the mainstem
Nushagak River or Kvichak River.
• The mine scenario watersheds (Scale 3, Figure 2-5) include the cumulative drainage areas of the
South and North Fork Koktuli Rivers to their junction and Upper Talarik Creek to its junction with
Iliamna Lake.
• The mine scenario footprints (Scale 4, Figure 2-6) include the areas that would be covered by mine
infrastructure components (e.g., mine pit, waste rock piles, and tailings storage facilities) for each
mine scenario (Chapter 6).
• The transportation corridor area (Scale 5, Figure 2-7) includes the 27 subwatersheds in the Kvichak
River watershed that drain to Iliamna Lake (Chapter 6); the transportation corridor does not cross
into the Nushagak River watershed.
In problem formulation, we use broader spatial scales to describe the physical, chemical, and biological
environment in the Bristol Bay region (Table 2-1), and consider the effects of multiple mines across the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TZ April 2013
-------
Chapter 2
Overview of Assessment
landscape. In risk analysis and characterization, we use smaller spatial scales to evaluate the potential
effects of mining operations.
BOX 2 2. SALMONID FISHES IN THE BRISTOL BAY WATERSHED
The Bristol Bay watershed's freshwater habitats support a diverse and robust assemblage of fishes,
dominated by the family Salmonidae. This family comprises three subfamilies—Salmoninae (salmon, trout,
and char), Thymallinae (grayling), and Coregoninae (whitefish)—all of which are represented in the region. In
this assessment, we focus on fishes in the subfamily Salmoninae, particularly the five North American
Pacific salmon species (sockeye, Chinook, coho, chum, and pink), rainbow trout, and Dolly Varden (a species
of char). Collectively, we refer to these seven species as salmonids throughout this report.
All Salmonidae fishes spawn in freshwater, but they can differ in their life histories. Some populations (e.g.,
Bristol Bay's Pacific salmon) are anadromous, meaning that individual fish migrate to marine waters to feed
and grow before returning to fresh water to reproduce. Other Bristol Bay populations (e.g., lake trout, Arctic
grayling) are non-anadromous (resident), mean ing that essentially all individuals remain in fresh waters to
feed. Other populations (e.g., rainbow trout, Dolly Varden) can exhibit either anadromous or non-
anadromous life histories.
Table 2 1. Spatial scales considered in the assessment.
Scale
1
2
3
4
5
Description
Bristol Bay watershed
Nushagakand
Kvichak River
watersheds
Mine scenario
watersheds
Hydrologic Unit Codes (HUCs)3
19030202-19030206
19030301-19030306
1903010101-1903010113
1903010201-1903010203
1903020101-1903020110
19030301-19030304
19030205, 19030206"
190303021103, 190303021104
190303021101- 190303021102
1903020607
Area or Length
(% of scale above)
115,500 km2 (NA)
59,890 km2 (52%)
925 km2 (2%)
Representative
Chapters
2,3,4,5,13
2,3,4,5,13
6, 7,8, 9, 12
Mine scenario footprints
Pebble 0.25
Pebble 2.0
Pebble 6. 5
Transportation
corridor
NA
NA
NA
190302051403-190302051406
190302060101- 190302060104
190302060201- 190302060206
190302060301- 190302060302
190302060701- 190302060702
190302060704
190302060901- 190302060905
190302060907, 190302060914
9.74 km2 (1%)
34.6 km2 (4%)
75.0 km2 (8%)
113 km (NA)
6, 7,8, 9, 12
6,7,8,9,12
6, 7,8, 9, 12
6, 10, 11
Notes:
a From the National Hydrography Dataset (USGS 2012L Scale 1 is defined by 8-digitand 10-digit HUCs; Scale 2 by 8-digitand 12-digit HUCs;
Scale 3 by 10-digit and 12-digit HUCs; Scale 5 by 12-digit HUCs.
b Except for 190302062301-190302062311.
NA = not applicable
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-8
April 2013
-------
Chapter 2
Overview of Assessment
Figure 2 2. Five spatial scales considered in this assessment. Only selected towns and villages are shown on this map. See Figures 2 3
through 2 7 for detailed views of each scale.
^f Approximate Pebble Deposit Location
• Towns and Villages
| | Scale 1: Bristol Bay Watershed
| | Scale 2: Nushagak & Kvichak River Watersheds
Scale 3: Mine Scenario Watersheds
I I Scale 4: Mine Scenario Footprints
I I Scale 5: Transportation Corridor Area
N
A
0 50 100
0 50 100
] Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-9
April 2013
-------
Chapter 2
Overview of Assessment
Figure 2 3. The Bristol Bay watershed (Scale 1), comprising the Togiak, Nushagak, Kvichak, Naknek, Egegik, and Ugashik River
watersheds and the North Alaska Peninsula. Only selected towns and villages are shown on this map.
Salmon "*/• **">
National Preserve
AKNEK
100
Kilometers
100
Miles
K Approximate Pebble Deposit Location
• Towns and Villages
Watershed Boundary
Parks, Refuges, or Preserves
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-10
April 2013
-------
Chapter 2
Overview of Assessment
Figure 2 4. The Nushagak and Kvichak River watersheds (Scale 2).
Cook Inlet
Bristol Bay
N
A
25
Approximate Pebble Deposit Location
Towns and Villages
Parks, Refuges, or Preserves
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-11
April 2013
-------
Chapter 2
Overview of Assessment
Figure 2 5. The mine scenario watersheds South Fork Koktuli River, North Fork Koktuli River, and
Upper Talarik Creek within the Nushagak and Kvichak River watersheds (Scale 3). See Figure 2 6
for descriptions of the three mine scenario footprints.
S(xrm TORK KOKTULI
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watershed
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-12
April 2013
-------
Chapter 2
Overview of Assessment
Figure 2 6. The mine footprints for the three scenarios evaluated in the assessment (Scale 4).
Pebble 0.25 represents 0.25 billion tons of ore; Pebble 2.0 represents 2.0 billion tons of ore; Pebble
6.5 represents 6.5 billion tons of ore. Each footprint includes mine pit, waste rock, and tailings
storage facility areas. See Figures 6 1, 6 2, and 6 3 for more detailed maps of individual mine
scenario footprints. Blue areas indicate streams and lakes from the National Hydrography Dataset
(USGS 2012) and wetlands from the National Wetlands Inventory (USFWS 2012).
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-13
April 2013
-------
Chapter 2
Overview of Assessment
Figure 2 7. The transportation corridor area (Scale 5), comprising 27 subwatersheds in the Kvichak River watershed that drain to Iliamna
Lake. Subwatersheds are defined at the HUC 12 level according to the National Hydrography Dataset (USGS 2012).
K Approximate Pebble Deposit Location
- - - Transportation Corridor (Outside Assessment Area)
Transportation Corridor
Existing Roads
Transportation Corridor Area
Subwatersheds within Area
5 10
Kilometers
5 10
I Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
2-14
April 2013
-------
Bristol Bay is a large gulf of the Bering Sea located in southwestern Alaska. The land area draining to
Bristol Bay consists of six major watersheds—from west to east, the Togiak, Nushagak, Kvichak, Naknek,
Egegik, and Ugashik River watersheds—and a series of smaller watersheds draining the north Alaska
Peninsula (Figure 2-3). The Bristol Bay region encompasses complex combinations of physiography,
climate, geology, and hydrology, which interact to control the amount, distribution, and movement of
water through a landscape shaped by processes such as tectonic uplift, glaciation, and fluvial erosion
and deposition. The region's freshwater habitats are varied and abundant, and support a diverse and
robust assemblage of fishes (Chapter 5).
The Nushagak and Kvichak River watersheds account for more than half the land area in the Bristol Bay
watershed. The Pebble deposit, the largest known porphyry copper deposit in the region, is located in
the headwaters of both watersheds (Figure 2-4), and represents the most likely site for near-term, large-
scale mining development in the Bristol Bay watershed. In this chapter, we consider key aspects of the
Bristol Bay watershed's physical environment, with particular emphasis on the Nushagak and Kvichak
River watersheds (Figure 2-4).
3.1 Physiographic Divisions
The Nushagak and Kvichak River watersheds comprise five distinct physiographic divisions (Wahrhaftig
1965): the Ahklun Mountains, the Southern Alaska Range, the Aleutian Range, the Nushagak-Big River
Hills, and the Nushagak-Bristol Bay Lowland (Table 3-1, Figure 3-1). Precipitation is greatest in the
Southern Alaska Range, the Aleutian Range, and the Ahklun Mountains (Figures 3-1 and 3-2), and these
divisions serve as major water source areas for lower portions of the watersheds. Annual water balance,
especially in the mountains and hills, is dominated by snowpack accumulation and subsequent melt,
although late summer and fall rains are also important contributors to the hydrologic cycle, particularly
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-1
April 2013
-------
Chapter 3 Region
in the Nushagak-Bristol Bay Lowland division (Selkregg 1974). Additional key attributes of each
physiographic division are discussed below.
The Ahklun Mountain physiographic division, in the western portion of the Nushagak River watershed,
is dominated by rolling hills to sharp, steep, glaciated mountains that receive high snowfall (Table 3-1,
Figure 3-1) (Wahrhaftig 1965, Selkregg 1974, Gallant etal. 1995). Parent bedrock is deformed
sedimentary rocks, intruded in several locations by igneous batholiths and stocks (Figure 3-3). A few
small glaciers occur in high mountains cirques, and isolated masses of permafrost occur sporadically
(Figure 3-4). Glacially carved lowland valleys are now filled with large, deep lakes, and adjacent streams
are often incised in bedrock gorges. The surrounding area is mantled with colluvium, alluvium, and
glacial drift and moraines (Figure 3-3). Soils are generally well drained and have medium erosion
potential (Figures 3-5 and 3-6). Dwarf scrub is the dominant vegetation in the mountains and tall scrub
and herbaceous plants are common in the valleys and lower mountain slopes (Figure 3-7).
The Southern Alaska Range physiographic division comprises a series of high, steep, glaciated
mountains with land surfaces covered by rocky slopes, glacial drift and moraines, and glaciers
(Table 3-1, Figure 3-1) (Wahrhaftig 1965, Selkregg 1974). Bedrock is a complex of granitic batholiths
intruded into metamorphosed sedimentary and volcanic rock (Figure 3-3). Soils are shallow or not
present (Figure 3-5) and permafrost occurs as isolated masses (Figure 3-4). Alpine tundra is the
predominant vegetation (Figure 3-7). Streams are frequently swift and braided with several headwaters
originating in glaciers (Figure 3-8). Several large, deep lakes occur in the glaciated valleys within the
division. Braided, turbid streams flow into lakes, allowing sediment to settle, before flowing into the
Nushagak and Kvichak River systems.
Within the Bristol Bay watershed, the Aleutian Range physiographic division consists of rolling hills to
steep, glaciated mountains built of volcanic and intrusive bedrock (Table 3-1, Figure 3-1) (Wahrhaftig
1965, Selkregg 1974). Cirque glaciers remain atop mountains in the extreme southeast corner of the
Kvichak River watershed (Figure 3-3). This division is generally free of permafrost (Figure 3-4). Soils
have formed in volcanic ash over glacial deposits at lower elevations, whereas rocky lands dominate at
higher elevations (Figure 3-5). Erosion potential is high for some soils in the Aleutian Range division
(Figure 3-6). Large, deep, moraine- and sill-impounded lakes are found in the ice-carved valleys. The
Alagnak River, which drains most of the Aleutian Range physiographic division within the Bristol Bay
watershed, is highly braided as it flows across the Nushagak-Bristol Bay Lowland to the Kvichak River.
Dwarf scrub vegetation is common (Figure 3-7) (Selkregg 1974, Gallant etal. 1995).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o 2 APril 2013
-------
Chapter 3
Region
Table 3 1. Physiographic divisions (Wahrhaftig 1965) of the Nushagak and Kvichak River watersheds.
Physiographic Division
Ahklun Mountains
Southern Alaska Range
Aleutian Range
Nushagak-Big River Hills
Nushagak-Bristol Bay
Lowland
Description
Rolling hills to sharp, steep, glaciated mountains
separated by broad lowlands, with a few small
glaciers in high mountain cirques
Rolling hills to steep, glaciated mountains covered
by glacial drifts and moraines, rocky slopes, and
glaciers
Rolling hills to sharp, steep glaciated mountains,
separated by broad lowlands, with a few small
glaciers in high mountain cirques
Rounded ridges with broad, gentle slopes and
broad, flat or gently sloping valleys
Flat to rolling landscape with low local relief and
deep morainal, drift, and outwash deposits, but no
glaciers
Elevation
(meters)
10-1,600
14-2,800
14-1,600
14-1,300
0-800
Permafrost
Extent
Sporadic
Unknown
Unknown
Sporadic
Sporadic or
absent
Freshwater Habitats
Mix of unconstrained and constrained streams;
Wood and Tikchik Lakes in U-shaped valleys
Swift, braided streams and rivers, some with
glacial headwaters; Lake Clark and other large
lakes in glaciated valleys
Large lakes associated with ice-carved valleys and
terminal moraines; glacially fed lake tributaries
Glacial moraines and ponds in eastern part of
region; upper reaches of the Nushagak and
Mulchatna Rivers
Moraine and thaw lakes; western half of Iliamna
Lake; Kvichak, Alagnak, Nushagak, Nuyakuk, and
Mulchatna River mainstems
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-3
April 2013
-------
Chapter 3
Region
Figure 3 1. Hydrologic landscapes within the Nushagak and Kvichak River watersheds, as defined
by physiographic division and climate class. Physiographic divisions (Wahrhaftig 1965) are classified
as Ahklun Mountains, Nushagak Bristol Bay Lowland, Aleutian Range, Nushagak Big River Hills, and
Southern Alaska Range. Climate classes (Feddema 2005) were defined as very wet, wet, moist, dry,
and semiarid, and calculated using Scenarios Network for Alaska and Arctic Planning data (SNAP
2012). Points labeled A through H indicate approximate locations where photos in Figure 3 8 were
taken.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Towns and Villages
Watershed Boundary
N
A
25
25
50
] Kilometers
50
D Miles
Ahklun Mountains Nushagak-Big River Hills Southern Alaska Range Nushagak-Bristol Bay Lowlands Aleutian Range
Moist Dry Semiarid Dry Moist
H Wet Moist DrV Molst H Wet
H Very Wet _ [ Wet H Moist Wet H Very Wet
H Wet H Very Wet
B Very Wet
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-4
April 2013
-------
Chapter 3
Region
Figure 3 2. Distribution of mean annual precipitation (mm) across the Nushagak and Kvichak River
watersheds, 1971 to 2000 (SNAP 2012).
I
Cook Inlet
Bristol Bay
N
A
25 50
25
] Kilometers
50
] Miles
Precipitation
3,725 mm/yr
2,025 mm/yr
325 mm/yr
Approximate Pebble Deposit Location
Towns and Villages
I Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-5
April 2013
-------
Chapter 3
Region
Figure 3 3. Generalized geology of the Bristol Bay watershed (adapted from Selkregg 1974).
^f Approximate Pebble Deposit Location
• Towns and Villages
| | Watershed Boundary
Generalized Geology
Moraine and Drift
Glaciolacustrine
Glaciofluvial
Alluvial
Coastal
| Eolian
| | Undifferentiated
Quaternary Volcanics
Intrusives
Tertiary
I Jurassic to Cretaceous
Late Paleozoic to Middle Mesozoic
| Triassic to Early Jurassic
| Paleozoic and Older
Glacier
Lake
Cook Inlet
NORTbTALASKAPENINSULA
N
A
0 50 100
0 50 100
JMiles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-6
April 2013
-------
Chapter 3
Region
Figure 3 4. Occurrence of permafrost in the Bristol Bay watershed (adapted from Selkregg 1974).
Mountainous
Isolated masses of permafrost
Lowland
| Thick to thin permafrost in fine-grained sediments
Isolated deep relicts or shallow lenses in fine-grained sediments
Permafrost free; coarse-grained sediments
Permafrost free except for few isolated masses
:•:::,#;•.•: Glacier
Lake
^f Approximate Pebble Deposit Location
• Towns and Villages
| | Watershed Boundary
Cook Inlet
NORTbTALASKAPENINSULA
N
A
50
50
100
] Kilometers
100
] Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-7
April 2013
-------
Chapter 3
Region
Figure 3 5. Dominant soils in the Bristol Bay watershed (adapted from Selkregg 1974).
Entisols
Well-drained loamy or gravelly gray soils
| Well-drained soils in stratified materials on floodplainsand low terraces
Inceptisols
Po oily drained soils with peaty surface layer; shallow permafrost ta bl e
Poorly drained soils; shallow to deep permafrost table
Well-drained dark soils formed in fine volcanic ash
Well-drained dark soils formed infine volcanic ash; shallow bedrock
Well-drained soilsformed in dominantly coarse volcanic ash or in shallow ash over other material
I Well-drained soils with dark, acid surface layer; shallow bedrock
| Well-drained thin soils with dark, acid surface layer; deep permafrost table
Spodosols
Well-drained thin strongly acid soils; deep permafrost table
Well-drained strongly acidsoils; very darksubso II
Histosols
Cither
Poorly drained fibrous peat; freezes in winter
Poorly drained fibrous peat; shallow permafrost table
Po oily drain ed fi brous peat; I enses of vo lean ic ash or a lluvial material ; seldom freezes deeply
Very steep rocky or ice-covered land
Fresh volcanic ash or cinders; little or no vegetation
Cook Inlet
Approximate Pebble Deposit Location
• TownsandVillages
Watershed Boundary
N
A
50 100
Kilometers
50 100
Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-8
April 2013
-------
Chapter 3
Region
Figure 3 6. Erosion potential in the Bristol Bay watershed (adapted from Selkregg 1974)
NORThTALASKAPENINSULA
•fa Approximate Pebble Deposit Location
• Towns and Villages
| | Watershed Boundary
Erosion Potential
| Low
Low-Medium
Medium
Medium-High
High
Undetermined
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-9
April 2013
-------
Chapter 3
Region
Figure 3 7. Dominant vegetation in the Bristol Bay watershed (adapted from Selkregg 1974)
NORTbTALASKAPENINSULA
X Approximate Pebble Deposit Location
• Towns and Villages
| | Watershed Boundary
Dominant Vegetation
| Coastal Western Hemlock-Sitka Spruce Forest
| Bottomland Spruce-Poplar Forest
Upland Spruce-Hardwood Forest
Lowland Spruce-Hardwood Forest
Moist Tundra
Wet Tundra
Alpine Tundra and Barren Ground
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-10
April 2013
-------
Chapter 3
Region
Figure 3 8. Physiographic divisions of the Nushagak and Kvichak River watersheds of Bristol Bay.
The Nushagak and Kvichak River watersheds contain a wide range of aquatic habitats within five
distinct physiographic divisions; see Figure 3 1 for a map of these divisions and the general location
where each photo was taken. All photos taken between August 2003 and August 2010, courtesy of
Michael Wiedmer.
ain south of the lower Nushagak River, Nushagak-Bristol
Bay Lowland division
Tributary to Nishlik Lake in the upper Nushagak River watershed,
Ahklun Mountains division
©
Klutuk Creek in the lower Nushagak River watershed, western
Nushagak-Bristol Bay Lowland division
Confluence of the Upper Nushagak Riverand the Nuyakuk River,
Nushagak-Bristol Bay Lowland division
Source of the Mulchatna River, Southern Alaska Range division
Village of Igiugig and the Kvichak River immediately downstream of
Iliamna Lake outlet, Nushagak-Bristol Bay Lowland division
Lake Clark Southern Alaska Range division of the upper Kvichak River Nofth fork Swan Rjver_ Nushagak_Bjg River Hi||s djvision
watershed
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-11
April 2013
-------
Chapter 3 Region
The Nushagak-Big River Hills physiographic division is largely rounded ridges that have moderate
elevations and broad, gentle slopes and broad, flat or gently sloping valleys (Table 3-1, Figure 3-1)
(Wahrhaftig 1965, Selkregg 1974). Major geologic formations include graywacke, argillite,
conglomerate, and greenstone flows (Figure 3-3). No modern glaciers are present, but glacial drift and
moraines are common throughout lower elevations and colluvium and alluvium mantle higher
elevations. The Nushagak River headwaters are the only part of the Nushagak and Kvichak River
watersheds that have not been glaciated. In most of this division falling within the Nushagak and
Kvichak River watershed, permafrost is found only in isolated masses or lenses (Figure 3-4). Soils
throughout the province are typically shallow, occur in well-drained to poorly drained conditions, and
have medium erosion potential (Figures 3-5 and 3-6). Rivers in the Mulchatna and Newhalen River
systems originate from glaciers in the Southern Alaska Range. Sediment from these glaciers is trapped in
large lakes, providing clearer water for downstream reaches.
The Pebble deposit is located in the eastern portion of the Nushagak-Big River Hills and is heavily
influenced by past glaciation (PLP 2011: Chapter 3). At various times, Pleistocene glaciers blocked the
South Fork Koktuli River, the North Fork Koktuli River, and Upper Talarik Creek, the three tributaries
draining the Pebble deposit area (Figure 2-5). Unconsolidated glacial deposits, ranging from a few to
several tens of meters in thickness, cover most of the lower elevations of the area (Detterman and Reed
1973). All three of the stream valleys in the Pebble deposit area have extensive glacial sand and gravel
deposits (PLP 2011: Chapter 8). Based on extensive studies in the Pebble area, the Environmental
Baseline Document 2004 through 2008 (EBD) (PLP 2011) concluded that the presence of permeable
shallow aquifers, upward hydraulic gradients, and strong local relief indicate that local and intermediate
groundwater flow systems dominate regional groundwater flow systems. Further, the EBD noted the
presence of many local, cross-cutting faults with high hydraulic conductivities in the Pebble deposit
area.
The Nushagak-Bristol Bay Lowland physiographic division (Table 3-1, Figure 3-1) is mantled with
glacial drift and moraine deposits up to hundreds of meters deep, forming a rolling landscape with low
local relief (15 to 75 m) and maximum elevations of 90 to 150 m near the transitions from the lowland
to adjacent mountains or hills (Wahrhaftig 1965, Detterman 1986, Lea et al. 1991, Stilwell and Kaufman
1996). Arc-shaped bands of morainal deposits ranging from 1.6 to 8 km wide enclose Iliamna Lake and
are frequent in the lowlands between the Nushagak River and the Ahklun Mountains division
(Figure 3-3). Steep outliers of the Wood River Mountains in the Ahklun Mountains physiographic
division arise from the western part of the lowland. A small area with sand dunes occurs east of the
Nushagak River (Lea and Waythomas 1990). Glacial drift is coarser near the mountains because of high
amounts of outwash and grades to fine sand along the coast (Wahrhaftig 1965). The remainder of the
lowland is dominated by low-relief (less than 20 m), rolling expanses of tundra underlain by Holocene
peat and wind-born deposits (Lea et al. 1991). Glaciers do not occur today in the Nushagak-Bristol Bay
Lowland division, and permafrost is sporadic or absent (Figure 3-4) (Wahrhaftig 1965). Morainal and
thaw lakes are common, and mainstem rivers draining this area exhibit high channel complexity
(Figure 3-8). Poorly drained soils predominate in the southern portions, whereas well-drained soils
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTI April 2013
-------
Chapter 3 Region
predominate across the remainder of the physiographic division (Figure 3-5). Soil erosion potential is
moderate throughout the province (Figure 3-6). Extensive dwarf scrub communities occur on relatively
well-drained soils and wet communities vegetate large areas as well (Figure 3-7) (Selkregg 1974, Gallant
etal. 1995).
3.2 Hydrologic Landscapes
To better evaluate the influence of inherent river basin attributes on streamflows, and thus fish
populations, we used the physiographic divisions discussed above to define different hydrologic
landscapes across the Nushagak and Kvichak River watersheds. These landscapes can be considered
hydrologic building blocks, and provide a broad-scale approach to spatially characterizing climate and
watershed factors controlling the amount, timing, and flowpaths of water within watersheds (Winter
2001).
We defined hydrologic landscapes by calculating water surplus (precipitation minus potential
evapotranspiration) across the basins in each of the five physiographic divisions, using Scenarios
Network for Alaska and Arctic Planning (SNAP) data (SNAP 2012) and procedures outlined in Feddema
(2005). Feddema (2005) defined six annual climate classes ranging from very wet to arid conditions.
The very wet, wet, and moist classes have an annual water surplus, whereas the dry, semi-arid, and arid
classes have an annual water deficit. Combining these climate classes with the physiographic divisions
(Section 3.1), we identified 18 different hydrologic landscapes across the Nushagak and Kvichak River
watersheds (Table 3-2, Figure 3-1), which represent the range of hydrologic characteristics across the
region.
3.3 Groundwater Exchange and Flow Stability
A key aspect of the Bristol Bay watershed's aquatic habitats is the importance of groundwater exchange.
Because salmon rely on clean, cold water flowing over and through porous gravels (from upwelling and
downwelling) for spawning, egg incubation, and rearing (Bjornn and Reiser 1991), areas of groundwater
exchange create high-quality salmon habitat (Appendix A). For example, densities of beach spawning
sockeye salmon in the Wood River watershed were highest at sites with strong groundwater upwelling,
and zero at sites with no upwelling (Burgner 1991). Portions of the Nushagak-Bristol Bay Lowland and
Nushagak-Big River Hills physiographic divisions, including the Pebble deposit area, contain coarse-
textured glacial drift with abundant, high-permeability gravels and extensive connectivity between
surface waters and groundwater (Figures 3-3, 3-4, and 3-9). Abundant wetlands and small ponds also
contribute disproportionately to groundwater recharge (Rains 2011). This strong connection between
groundwater and surface waters helps to moderate water temperatures and streamflows. For example,
groundwater contributions that maintain water temperatures above 0°C are critical for maintaining
winter refugia in streams that might otherwise freeze (Power et al. 1999).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment T~77 April 2013
-------
Chapter 3 Region
These groundwater contributions to streamflow, along with the influence of large and small lakes,
support flows in the region's streams and rivers that are more stable than those typically observed in
many other salmon streams (e.g., in the Pacific Northwest or southeastern Alaska). The lower mainstem
Nushagak and Kvichak Rivers illustrate this tendency toward moderated, consistent streamflows
(Figure 3-10). Coarse-textured glacial drift in the Kaskanak and Upper Talarik Creek drainages promotes
high groundwater contributions to these streams, resulting in stable flows through much of the year
(Figure 3-10). High baseflow in the Nushagak River also is consistent with increased interactions
between surface water and groundwater, as water flows from the Southern Alaska Range, Ahklun
Mountains, and Nushagak-Big River Hills into the coarse-textured glacial drift of the Nushagak-Bristol
Bay Lowland (Figure 3-10).
Water storage in upstream lakes plays a role in flow stabilization, as well. In the Kvichak River
watershed, Iliamna Lake dampens high flows from the Iliamna and Newhalen Rivers before they reach
the mainstem. The attenuating effect of upstream lakes on streamflow is also evident in the Newhalen
River, located downstream of Lake Clark (Figure 3-10). In the Nushagak River watershed, large lakes
occur in the Wood River Mountains headwaters, and their moderating influence can be seen in the
Nuyakuk River (Figure 3-10).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o .14 April 2013
-------
Chapter 3
Region
Table 3 2. Distribution of hydrologic landscapes in the Nushagak and Kvichak River watersheds. Values represent percentage of total area
in the two watersheds.
Physiographic Division
Climate class
Ahklun Mountains
V
W
M
Southern Alaska Range
V
W
M
D
Aleutian Range
V
W
M
Nushagak-B g River Hills
V
W
M
D
Nushagak-Bristol
Bay Lowland
V
W
M
Nushagak River watershed
Nushagak River (whole
watershed)
Nushagak River at Ekwoka
Nuyakuk River
Mulchatna River
Nushagak River at Mulchatna
River
Koktuli River
South Fork Koktuli Riverb
North Fork Koktuli River0
7
4
19
8
-
-
16
9
43
18
-
-
1
2
1
-
-
1
2
4
-
-
-
2
3
7
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
25
40
3
53
30
99
100
100
9
14
22
9
-
-
-
-
-
-
-
1
-
-
-
24
27
32
14
35
1
-
15
1
-
-
-
-
Kvichak River watershed
Kvichak River (whole watershed)
Kvichak River at lgiugigd
Kaskanak Creek near Igiugig8
Iliamna River near Pedro Bay'
Upper Talarik Creeks
-
-
-
-
-
-
-
-
-
16
25
94
-
13
20
6
-
8
12
-
-
1
2
-
-
2
-
-
-
11
-
-
-
2
6
-
-
-
-
-
7
10
21
-
100
7
11
-
-
1
-
-
-
-
-
3
-
28
-
-
28
11
50
-
-
Notes:
Blank values (-) indicate hydrologic landscapes that are not found in that portion of the Nushagak or Kvichak River watersheds. Climate classes are defined as very wet (V), wet (W), moist (M), and dry
(D) according to Feddema (2005); no semi-arid or arid climates are found in the region.
a USGS gage 15302500.
b USGS gage 15302200.
c USGS gage 15302250.
d USGS gage 15300500.
e USGS gage 15302520.
' USGS gage 15300300.
e USGS gage 15300250.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-15
April 2013
-------
Chapter 3
Region
Figure 3 9. Groundwater resources in the Bristol Bay watershed (adapted from Selkregg 1974). Yields presented in gallons per minute.
Groundwater Resources
Unconsolidated Deposits; mostly sand and gravel (100-1,000 gal/min yield)
Unconsolidated Deposits; mostly sand and gravel, silt and clay (10-100 gal/min yield)
Unconsolidated Deposits; mostly sand and gravel, silt and clay (0-10 gal/min yield)
Bedrock (0-10 gal/min yield)
Approximate Pebble Deposit Location
Towns and Villages
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-16
April 2013
-------
Chapter 3
Region
Figure 3 10. Mean monthly runoff for selected streams and rivers in the Nushagak and Kvichak
River watersheds. USGS gages and dates used to generate each line: A. Nushagak River watershed:
Nushagak River (15302500, Oct 1977 Sep 1993); Nuyakuk River (15302000, Jun 1953 Sep 2010);
North Fork (NF) Koktuli River (15302250, Sep 2004 Sep 2010); South Fork (SF) Koktuli River
(15302200, Sep 2004 Sep 2010). B. Kvichak River watershed: Kvichak River (15300500, Aug
1967 Sep 1987); Kaskanak Creek (15300520, Jun 2008 Sep 2011); Iliamna River (15300300, Jun
1996 Sep 2010); Upper Talarik Creek (15300250, Sep 2004 Sep 2010); Newhalen River
(15300000, Jul 1951 Sep 1986).
600
500
400
E
300
200
TOO
-»-Nushagak River
Nuyakuk River
NF Koktuli River
SF Koktuli River
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep
B
600
-»-Kvichak River
-•-Kaskanak Creek
-*-lliamna River
-—Upper Talarik Creek
Newhalen River
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-17
April 2013
-------
Chapter 3 Region
3.4 Quantity and Diversity of Aquatic Habitats
Differences in hydrology, geology, and climate across the Bristol Bay watershed interact to create the
region's diverse hydrologic landscapes (Table 3-2, Figure 3-1), ultimately shaping the quantity, quality,
diversity, and distribution of aquatic habitats throughout the watershed. These diverse habitats support
a high level of biological complexity, in part supported by enhanced ecosystem productivity associated
with anadromous salmon runs, contributing to the environmental integrity and resilience of the
watershed's ecosystems (Lisi et al. 2012, Ruff et al. 2011, Schindler et al. 2010).
In general, conditions in the Bristol Bay watersheds are highly favorable for Pacific salmon. The
Nushagak and Kvichak River watersheds encompass an abundant and diverse array of aquatic habitats,
supporting a diverse salmonid assemblage (Section 5.2). Habitats range from headwater streams to
braided rivers, small ponds to large lakes, side channels to off-channel alcoves. These watersheds
contain over 53,000 km of streams, 14% of which have been documented as anadromous fish streams
(Johnson and Blanche 2012). This percentage is likely a significant underestimate of the actual extent of
anadromous waters across the watersheds (Box 7-1, Appendix A).
Lakes and associated tributary and outlet streams are key spawning and rearing areas for sockeye
salmon. Lakes cover relatively high percentages of watershed area in the Bristol Bay region: 7.9% for the
entire Bristol Bay watershed area and 13.7% for the Kvichak River watershed (Luck et al. 2010). In
other North Pacific river systems supporting sockeye salmon populations, from northern Russia to
western North America, these values tend to be much lower (e.g., 0.2 to 2.9%) (Luck et al. 2010).
Relatively low watershed elevations (especially in the extensive Nushagak-Bristol Bay Lowland
physiographic division) and the absence of artificial barriers to migration (e.g., dams and roads) mean
that not only are streams, lakes, and other aquatic habitats abundant in the Bristol Bay region, but they
also tend to be accessible to anadromous salmonids. With very few exceptions, all major lakes in the
watershed are accessible to anadromous salmon (Appendix A). Lakes and ponds also play a key role in
groundwater dynamics and flow stability (Section 3.3).
Overall physical habitat complexity in the Bristol Bay watershed is higher than in many other systems
supporting sockeye salmon populations. Of 1,509 North Pacific Rim watersheds, the Kvichak, Wood, and
Nushagak (exclusive of Wood) Rivers ranked third, fourth, and forty-fourth, respectively, in physical
habitat complexity, based on an index that included variables such as lake coverage, stream junction
density, floodplain elevation and density, and human footprint (Luck et al. 2010).
3.4.1 Stream Reach Characterization: Attributes
To characterize the stream and river habitats in the Nushagak and Kvichak River watersheds, we
described stream and river valley attributes for each of the 65,701 stream and river reaches in the
Nushagak and Kvichak River watersheds documented in the National Hydrography Dataset (NHD)
(USGS 2012). For each reach, we estimated the mean annual flow (m3/s), mean valley gradient (%), and
percent of flatland in the contributing watershed lowland (% flat); each attribute is described in detail in
the following sections. These attributes were selected because they represent fundamental aspects of
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TT7 April 2013
-------
Chapter 3 Region
the physical and geomorphic settings in streams, providing context for stream and river habitat
development and subsequent fish habitat suitability (Burnett et al. 2007). It also was feasible to obtain
these attributes for the entire area given available data. These attributes have been used to model
habitat suitability for salmon at large scales, for example via intrinsic potential modeling (Burnett et al.
2007, Shallin Busch et al. 2011). We did not develop intrinsic potential models for salmon species in this
assessment, as that effort would require multiple years of field data collection for model validation and
testing and those data are not currently available. However, our characterization results do provide
insights into the distribution of broad-scale habitat conditions within the watersheds, and could provide
the basis for future intrinsic potential model development.
3.4.1.1 Valley Gradient
Valley gradient broadly characterizes channel steepness and geomorphic form. Valley gradient and
associated aspects of channel morphology influence channel capacity to transport sediment, affecting
the channel response to disturbance (Montgomery and Buffington 1997). Channel morphology can
strongly influence suitability for salmon rearing and spawning. Specific substrate and hydraulic
requirements vary slightly by species (Appendix A), but stream-spawning salmon generally require
relatively clean gravel-sized substrates with interstitial flow, and sufficient bed stability to allow eggs to
incubate in place for weeks to months prior to fry emergence (Quinn 2005).
Montgomery and Buffington (1997) proposed a process-based classification of mountain streams. Field
data from their study indicated that gradients estimated by digital elevation models (DEMs) provide a
useful predictor of channel morphology. We estimated the valley gradient of each stream reach in the
Nushagak and Kvichak River watersheds by assessing the gradient of correlated flowpaths across a
30-m-cell National Elevation DatasetDEM (Gesch etal. 2002, Gesch 2007) (Box 3-1). We adapted the
classification scheme put forth by Montgomery and Buffington (1997) to define four gradient classes
and predicted channel morphologies for stream reaches at different watershed scales.
• Less than 1 %, dune-ripple or pool-riffle morphology.
• At least 1 % and less thanS %, plane-bed morphology.
• At least 3 % and less than 8 %, step-pool morphology.
• At least 8 %, cascade morphology.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o H g April 2013
-------
Chapter 3 Region
BOX 3 1. METHODS FOR CHARACTERIZING VALLEY GRADIENT
The valley gradient of each stream reach in the Nushagak and Kvichak River watersheds was estimated by
assessing the gradient of correlated flowpaths along across a 30-m cell National Elevation Dataset (NED)
digital elevation model (DEM) (Gesch etal. 2002, Gesch 2007). We found the measured gradient of the NHD
flowlines (based on the elevation of the underlying DEM) was not an accurate representation of channel
gradient because of inconsistencies between the mapped streams and rivers in the NHD and the topography
described by the DEM. Channel traces in the NHD did not reliably follow the valley floor, and upslope traces
and misalignment with the DEM resulted in inaccurate measures of stream gradients and sampled
elevations.
We determined that the gradient of streams in a drainage network described by a flow analysis across the
DEM would more accurately represent channel morphology given the data available. The drainage network of
the DEM paralleled the network of the NHD flowlines, but included or excluded some small tributaries and
lacked the sinuosity mapped in the NHD.
Gradients of flowlines across the DEM were determined using the hydrology tools of the Spatial Analyst
extension of ArcGIS. First, the hydraulic network was generated based on the topography of the NHD DEM.
Generation of the hydraulic network involved the following tools:
• Fill. Sinks in the DEM were filled so that continuous flowpaths could be described.
• Flow Direction. The steepest path or flow direction was determined from each cell in the DEM.
• Flow Accumulation. Based on the direction of flow, the total number of cells, or receiving area for each
cell in the DEM, was determined.
• Reclassify. A threshold value of 0.25 km2 was applied to the total receiving area output from the previous
step to distinguish streams from non-streams.
• Stream Link. The resulting network was processed to assign unique identifiers to each link in the drainage
network.
To determine the gradient of each of the stream links in the drainage network, and to generate geometry that
could assign these values to the reaches of the NHD flowlines, the following tools were used:
• Extract by Mask. Elevation values underlying the drainage network were isolated from the DEM so that
cross-valley slopes would not be measured when determining gradient.
• Slope. Gradient along the drainage network was measured between each cell of the isolated drainage
network DEM. The drainage DEM confined the slope measures to the flowpath of the drainage network,
providing an estimate of stream gradient at each 30-m cell.
• Watershed. The output of the Stream Link tool (see above) and the results of the flow direction analysis
were used to delineate the drainage basin for each stream link. This geometry was then used to transfer
gradient values to the NHD stream reaches.
• Zonal Statistics. In the drainage basin for each stream segment, the average gradient was determined for
all cells with values (i.e., a mean gradient of the stream segment.) The mean gradient values were then
assigned to the drainage basin geometry.
The mean gradient for each drainage basin was then transferred to the NHD flowlines usingthe Zonal
Statistics as Table tool. This tool measured the length-weighted mean of the gradients for each reach (as
defined by the NHD Reach Code attribute) assigned to the drainage basins. Typically, the NHD flowlines
occupied no more than two drainage basins. The resulting gradient estimates were appended to the table of
the NHD flowlines.
The substrate and hydraulic conditions required by stream-spawning salmon are most frequently met in
stream channels with gradients less than 3% (Montgomery et al. 1999). At the lowest gradients, the
channel's capacity to transport fine sediments will be low, and substrates may be dominated by sands
and other fines, providing sub-optimal salmon spawning habitat A notable exception to this generality
occurs in low-gradient, off-channel habitats that may be dominated by fine sediments but that contain
areas of upwelling and are used by riverine-spawning sockeye salmon (Eiler et al. 1992). At gradients
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TT7 April 2013
-------
Chapter 3 Region
above 3%, channels develop step pool or cascade morphologies and the size, stability, and frequency of
pockets of suitable spawning substrates decrease substantially (Montgomery and Buffington 1997). In
the Bristol Bay region, gradients of productive stream reaches for salmon are typically less than 3%,
with gradients less than 1% characterizing the most productive reaches; these habitats include lake
outlets and lower tributary reaches, and most of the major spawning reaches and tributaries of the
Nushagak and Kvichak River watersheds (Figures 3-11 and 3-12) (Demory et al. 1964). We note,
however, that low-gradient watersheds in the coastal plain region of the Nushagak-Bristol Bay Lowland
that lack upland headwaters are generally not productive salmon habitats. These streams tend to be
characterized by fine-textured substrates with high proportions of organic material, and may lack
substrates coarser than sand, presumably due to lack of higher-gradient source areas for gravel
recruitment (Wiedmer pers. comm.).
Environmental conditions determining suitability for juvenile salmon and adult resident salmonids
(e.g., resident Dolly Varden; Box 2-2) are also influenced by gradient. Fish movement can be restricted
by the high water velocities and frequent drops found in streams with gradients exceeding 12%,
although Dolly Varden have been found at gradients exceeding 15% in southeast Alaska streams
(Wissmar et al. 2010). Gradient and channel roughness also influence the distribution of water velocities
and hydraulic conditions in streams, influencing food delivery rates and availability and subsequent
energetic demands of drift feeding fishes (Hughes and Dill 1990).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 32± April 2013
-------
Chapter 3
Region
Figure 3 11. Examples of different stream size and gradient classes in the Nushagak and Kvichak
River watersheds.
<1% gradient headwater stream, Kaskanak Creek drainage, Kvichak
River watershed. Photo: Michael Wiedmer, ADF&G, 8/27/03
5% gradient headwater stream, North Fork Koktuli River drainage,
Nushagak River watershed. Photo: Michael Wiedmer, ADF&G, 8/19/06
3% gradient medium stream, Mulchatna River drainage, Nushagak
River watershed. Photo: Michael Wiedmer, USGS, 8/19/10
1.2% gradient medium stream, Koktuli River drainage, Nushagak
River watershed. Photo: Michael Wiedmer, AOF&G, 8/24/03
<1% gradient small river, South Fork Koktuli River, Nushagak River <1% gradient large river, Nushagak River between Portage Creek and
watershed. Photo: Joe Buckwalter, ADF&G, 8/17/06 Ekwok. Photo: Michael Wiedmer, ADF&G, 8/12/06
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-22
April 2013
-------
Chapter 3
Region
Figure 3 12. Valley gradient classes in the Nushagak and Kvichak River watersheds. Valley gradient
was assessed by measuring drainage channel slope across the watersheds' landscapes (Box 3 1).
Cook Inlet
Bristol Bay
^f Approximate Pebble Deposit Location
• Towns and Villages
I Watershed Boundary
< 1% Gradient
1 - 3% Gradient
3-8% Gradient
> 8% Gradient
I N
A
25
25
50
] Kilometers
50
3 Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-23
April 2013
-------
Chapter 3 Region
3.4.1.2 Mean Annual Flow
Mean annual flow is a metric of stream size, an important determinant of available habitat space
(capacity) for stream fishes. The relationship between mean annual flow and habitat capacity for rearing
juvenile salmon can vary with streamflow regime and other limiting factors, but is generally positive
when other factors are not constraining.
Mean annual flow for each stream reach within the Nushagak and Kvichak River watersheds was
estimated using regression equations for the prediction of mean annual streamflow, based on drainage
area and historical mean annual precipitation in southwestern Alaska (Parks and Madison 1985) (Box 3-
2). We defined four classes of stream size based on these mean annual flow calculations.
• Small headwater streams (less than 0.15 m3/s), including many of the tributaries of the South and
North Fork Koktuli Rivers and Upper Talarik Creek.
• Medium streams (0.15 to 2.8 m3/s), including the upper reaches and larger tributaries of the South
and North Fork Koktuli Rivers and Upper Talarik Creek.
• Small rivers (2.8 to 28 m3/s), including the middle to lower portions of South and North Fork
Koktuli Rivers, and Upper Talarik Creek, and the mainstem Koktuli River.
• Large rivers (greater than 28 m3/s), including the Mulchatna River below the confluence with the
Koktuli River, the Newhalen River, and other larger rivers.
All five species of salmon present in the Bristol Bay region use portions of the large and small rivers and
medium streams for migration, spawning, and/or rearing habitat. Salmon also use small streams in the
Bristol Bay region for spawning and rearing, but use of these habitats may be constrained by shallow
depths, insufficient flow to allow passage, the unavailability of open water in winter, or other limitations
related to stream size.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o 24 APril 2013
-------
Chapter 3 Region
BOX 3 2. METHODS FOR CHARACTERIZING MEAN ANNUAL FLOW
Mean annual flow for each stream reach in the Nushagakand Kvichak River watersheds was estimated
using regression equations, based on drainage area and historical mean annual precipitation data in
southwestern Alaska (Parks and Madison 1985). Total drainage area was determined for reaches along the
NHD flowlines by developing a drainage-corrected DEM based on the National Elevation Dataset (NED).
While the underlying topography and catchments described by the NED remained the same, the elevations
underlyingthe NHD flowlines and in their immediate vicinity were lowered and smoothed such that runoff
conformed to the geometry of the NHD flowlines.
Using the drainage-corrected DEM, we estimated total catchment area above any location in the drainage
network. The NED DEM was corrected to better conform to the NHD flowlines and drainage areas were
calculated using the following tools of the ArcHydro and Spatial Analyst tools of the ArcGIS suite:
• DEM Reconditioning. The elevations of the DEM were altered along the NHD flowlines and in their
immediate vicinity. Parameters used for this tool were a 10-m reduction of elevations along the flowline, a
5-cell (150-m)-wide transition zone on either side of the flowline, and a post-process 1000-m reduction in
elevations along the flowlines. The initial elevation reduction and transition width were found to
adequately capture flows and maintain those flows within the channel geometry. The post-processing
adjustment is a more arbitrary value intended to confine flows to the channels once captured.
• Fill. Sinks in the reconditioned DEM were filled so that continuous flowpaths could be described.
• Flow Direction. The steepest path or flow direction was determined from each cell in the DEM.
• Flow Accumulation. Based on the direction of flow, the total number of cells, or receiving area for each
cell in the DEM, was determined. These values were multiplied by 0.0009 to convert the area of each cell
(900 m2) to square kilometers.
Precipitation data from 1971 to 2000 were used to calculate average annual precipitation for any zone in
the study area, based on Scenarios Network for Alaska and Arctic Planning data (SNAP 2012). The output
drainage area raster from the steps above and raster coverage of average annual precipitation were used as
inputs for the mean annual flow regression equation developed by Parks and Madison (1985) for
southwestern Alaska:
• Q = (10-1.38)*(DA0.98)*(P1.13)
Where Q is mean annual flow in cubic feet per second, DA is drainage basin area in square miles, and P is
mean annual precipitation in inches per year. We used the mean annual flow value from the approximate
midpoint of each stream reach as the estimate of mean annual flow for the reach.
Salmonid species differ in their propensities for small streams. Dolly Varden have been documented
using all stream sizes, including some of the smallest channels. Of the Pacific salmon species, coho
salmon are most likely to use small streams for spawning and rearing, and have been observed in many
of the smaller streams near the Pebble and other deposits. Larger-bodied Chinook salmon adults are less
likely to access smaller streams for spawning (Quinn 2005). However, juvenile Chinook salmon are
observed in small tributaries where spawning has not been documented.
3.4.1.3 Proportion of Flatland in Lowland
Stream channels in mountainous and foothill terrain are, to varying degrees, laterally constrained by
their valley walls. Degree of channel constraint influences channel form, including the development of
off-channel habitats, variability in local channel gradient, and hydraulic conditions during over-bank
flows. Generally, unconstrained channels have higher complexity of channel habitat types and hydraulic
conditions, and higher frequencies of off-channel habitats such as side channels, sloughs, and beaver
ponds. Such habitat complexity can be beneficial to salmon by providing a diversity of spawning and
rearing habitats throughout the year (Stanford etal. 2005).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ 25 April 2013
-------
Chapter 3 Region
To provide an index of the degree of channel constraint expected within each stream reach, we
estimated the percent of flatland (less than 1% slope) within lowland (area below median elevation) for
each stream reach's drainage basin (Box 3-3). Visual inspection of portions of the study area where high-
resolution aerial photographs were available showed that channels were typically unconstrained when
the proportional flat lowland exceeded 5%. This threshold was used to identify two classes:
• Less than 5% flatland in lowland, indicating reaches are constrained and not floodplain prone.
• At least 5% flatland in lowland, indicating reaches are unconstrained and floodplain prone.
In the Bristol Bay region, streams that are unconstrained and able to develop complex off-channel
habitats are more likely to provide a diversity of channel habitat types and hydraulic conditions,
creating favorable conditions, particularly for salmonid rearing. For Chinook and coho salmon, as well as
river-rearing sockeye salmon that may overwinter in streams, such habitats may be particularly
valuable. This metric is not a perfect index of channel constraint, however. Channels in flat lowlands
such as the coastal Nushagak-Bristol Bay Lowland physiographic division (Figure 3-1) may actually be
incised into fine-grained sediments with very little off-channel habitat complexity. In the glacially
worked landscapes of the Bristol Bay region, streams may be constrained by valley terraces and
moraine deposits that are not distinguishable on the coarse-scale DEM available for the region. Terraces
are a common feature in portions of the region, but the degree to which terrace constraint influences
these results could not be determined from the existing DEM. In steep, mountainous terrain, narrow
valleys may occasionally allow for unconstrained stream channel development across low-gradient
floodplains, but these features are likely not always detected with the resolution of OEMs currently
employed for this effort.
3.4.2 Stream Reach Characterization: Results
We estimated the three stream-reach attributes discussed above in four geographically defined areas
that vary in scale and location.
• The Nushagak and Kvichak River watersheds (Scale 2).
• The mine scenario watersheds—the South Fork Koktuli River, North Fork Koktuli River, and Upper
Talarik Creek (Scale 3).
• The streams lost to the Pebble 6.5 mine scenario footprint (Scale 4).
• The subwatersheds of the transportation corridor area (Scale 5).
In this section, we summarize results for the Nushagak and Kvichak River watersheds; results for the
other three scales are reported in Sections 7.2.1 and 10.2.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 095 April 2013
-------
Chapter 3 Region
BOX 3 3. METHODS FOR CHARACTERIZING PERCENT FLATLAND IN LOWLAND
The relative degree of channel constraint in the Nushagak and Kvichak River watersheds was estimated by
calculating the percent of flatland (<1% slope) within lowland (area below median elevation) in each stream
reach's drainage basin. These calculations included the delineation of drainage basins of the drainage-
corrected drainage network (developed for the mean annual flow analysis; see Box 3-2) as well as elevation
and slope analyses of the unaltered DEM.
To establish the drainage basin geometry of the drainage-corrected flow analysis, the following Spatial
Analyst tools were applied within an ArcGIS workspace.
• Reclassify. Athreshold value of 0.25 km2 was applied to the total receivingarea output from the
drainage-corrected flow analysis to distinguish streams from non-streams.
• Stream Link. The resulting network was processed to assign unique identifiers to each link in the
drainage network.
• Watershed. The output of the Stream Link tool (see above) and the results of the flow direction analysis
were used to delineate the drainage basin for each stream link. This geometry was used as the
geographic extent of analysis for each stream segment.
Areas of flatland (<1% slope) and lowland were then identified for each drainage basin. The unaltered NED
DEM was processed with the following Spatial Analyst tools from ArcGIS.
• Slope. The original (not drainage-corrected) DEM was analyzed to determine slope (%) across the extent
of the Nushagak and Kvichak River watersheds.
• Reclassify. Athreshold value of l%was applied to the slope analysis, and attributes were assigned
across the study area as meeting or not meeting the flatland criteria.
• Zonal Statistics. In the drainage basin for each stream segment, the minimum and maximum elevations
were determined using the Zonal Statistics tool. These values were used to identify the median elevation
for each watershed.
• Reclassify. The DEM was classified as meeting or not meeting the lowland criteria based on results of the
previous step.
Finally, the percent flatland in lowland for each stream reach's drainage basin was calculated using the
following steps.
• Times. Areas of flatland outside of lowland areas were eliminated by multiplyingthe flatland and lowland
rasters. The flatland and lowland rasters used 1 and 0 values for true and false, respectively, so both
conditions were required to return a positive result for flatland in lowland.
• Zonal Statistics. The total areas of lowland and flatland within lowland were calculated for each drainage
basin.
• Divide. The percent flatland in lowland was determined for each drainage basin by dividing the area of
flatland in lowland by the area of lowland in each drainage basin.
• Zonal Statistics as Table. The average value of percent flatland in lowland for each stream reach was
calculated and added to a table, which was then appended to the NHD flowline data table. Although the
mean statistic was used to ascertain these values for the NHD flowlines, the flowlines typically had a one-
to-one correlation with drainage basins, as the basins were based on the drainage-corrected flow
analysis.
We characterized over 52,900 km of stream channel and 79,535 stream and river reaches in the
Nushagak and Kvichak River watersheds. Reach attributes reflected the hydrologic landscapes in which
the reaches occurred, and upstream within the reach's drainage (Section 3.2). Relatively low-gradient
stream channels extend far up into the headwaters of the upper Mulchatna and Nushagak River
watersheds (Figure 3-12), allowing salmon to access streams in the headwaters. High-gradient
conditions are primarily found in the headwaters of Lake Clark and Iliamna Lake tributaries, and the
headwaters of the Alagnak, Wood, Kokwok, and Nuyakuk Rivers (Figure 3-12). Valley flatland is heavily
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTZ April 2013
-------
Chapter 3 Region
concentrated in the Nushagak-Bristol Bay Lowland physiographic division and along the larger rivers,
but includes significant wider-valley reaches in the Nushagak-Big River Hills, Southern Alaska Range,
and Aleutian Range regions (Figure 3-13).
The majority of stream channel length (78%) in the Nushagak and Kvichak River watersheds is
composed of medium and small (less than 2.8 m3/s mean annual flow), low-gradient (less than 3%)
streams (Table 3-3, Figures 3-12 and 3-14). The extent of flatland in valley lowlands is strongly
associated with gradient. For streams with less than 1% gradient, 89% are associated with floodplains
(at least 5% flatland in lowland), versus less than 12% for streams with greater than 1% gradient.
Stream reaches with greater than 3% gradient were only found in landscapes where floodplains were
minimal (less than 5% flatland in lowland). Overall, these results reveal the high proportion of stream
channels in these watersheds possessing the broad geomorphic and hydrologic characteristics that
provide the context for the development of stream and river habitats highly suitable for fish species such
as Pacific salmon, Dolly Varden, and rainbow trout.
3.5 Water Quality
3.5.1 Water Chemistry
Water quality of streams near the Pebble deposit has been characterized extensively (PLP 2011,
Zamzow 2011). The watersheds in the Pebble deposit area (Figure 2-5) are neutral to slightly acidic,
with low conductivity, hardness, dissolved solids, suspended solids, and dissolved organic carbon
(Table 3-4). In those respects, they are characteristic of undisturbed streams. However, as would be
expected for a metalliferous site, levels of sulfate and some metals (copper, molybdenum, nickel, and
zinc) are elevated, particularly in the South Fork Koktuli River. PLP found that copper levels in some
samples from the South Fork Koktuli River exceeded Alaska's chronic water quality standard. However,
most of the exceedances were in or close to the deposit and the number and magnitude of exceedances
decreased with distance downstream (PLP 2011: Figure 9.1-35, 60, 61, 65, and 66).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ JR April 2013
-------
Chapter 3
Region
Figure 3 13. Likelihood of floodplain connectivity, as measured by the percent flat land in lowland
areas, for the Nushagak and Kvichak River watersheds. Percent flatland refers to land with less than
1% slope; lowland areas are defined as areas below the midpoint elevation within the drainage basin
of each stream reach (Box 3 3).
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Towns and Villages
Watershed Boundary
< 5% Flatland in Lowlands
5 -10% Flatland in Lowlands
10 - 25% Flatland in Lowlands
25 - 50% Flatland in Lowlands
> 50% Flatland in Lowlands
N
A
25
25
50
] Kilometers
50
H Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-29
April 2013
-------
Chapter 3
Region
Figure 3 14. Stream size classes in the Nushagak and Kvichak River watersheds as determined by
mean annual flow. Mean annual flow of streams and rivers was estimated using drainage area and
mean annual precipitation (Box 3 2).
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Towns and Villages
Watershed Boundary
Small Headwater Stream (< 0.15 m3/s)
Medium Stream (0.15 - 2.8 m3/s)
Small River (2.8-28 m3/s)
Large River (> 28 m3/s)
IN
A
25
25
50
] Kilometers
50
H Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-30
April 2013
-------
Chapter 3
Region
Table 3 3. Proportion of stream channel length within the Nushagak and Kvichak River watersheds
(Scale 2) classified according to stream size (based on mean annual discharge in m3/s), channel
gradient (%), and potential floodplain influence. Gray shading indicates proportions greater than 5%;
bold indicates proportions greater than 10%.
Stream Size
Small headwater streams3
Medium streams'5
Small rivers0
Large riversd
Gradient
<1%
FP
31%
20%
6%
2%
NFP
4%
3%
0%
0%
>1 % and <3 %
FP
3%
1%
0%
0%
NFP
12%
3%
0%
0%
>3 % and <8 %
FP
0%
0%
0%
0%
NFP
8%
2%
0%
0%
>8%
FP
0%
0%
0%
0%
NFP
2%
0%
0%
0%
Notes:
a 0-0.15 m3/s; most tributaries in the mine footprints.
b 0.15-2.8 m3/s; upper reaches and larger tributaries of the South Fork Koktuli, North Fork Koktuli, and Upper Talarik Creek.
c 2.8-28 m3/s; mid to lower portions of the South Fork Koktuli, North Fork Koktuli, and Upper Talarik Creek, including the mainstem Koktuli
River.
d >28 m3/s; the Mulchatna River below the Koktuli confluence, the Newhalen River, and other large rivers.
FP = floodplain influence; NFP = no floodplain influence.
3.5.2 Water Temperature
Water temperature data (PLP 2011: Appendix 15.IE, Attachment 1) indicate significant spatial
variability in thermal regimes. Average monthly stream water temperatures in the Pebble deposit area
in July or August can range from 6°C to 16°C. Longitudinal profiles of temperature provided in the EBD
indicate that stream temperatures in the Pebble deposit area do not uniformly increase with decreasing
elevation. This is often due to substantial inputs of cooler water from tributaries or groundwater (PLP
2011). Extensive glacially reworked deposits with high hydraulic conductivity allow for extensive
connectivity between groundwater and surface waters in the region (Power et al. 1999). This
groundwater-surface water connectivity has a strong influence on the hydrologic and thermal regimes
of streams in the Nushagak and Kvichak River watersheds, providing a moderating influence against
both summer heat and winter cold extremes in stream reaches where this influence is sufficiently
strong. The range of spatial variability in temperatures provided in the EBD (PLP 2011) is consistent
with streams influenced by a variety of thermal modifiers, including upstream lakes, groundwater, or
tributary contributions (Mellina et al. 2002, Armstrong et al. 2010).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-31
April 2013
-------
Chapter 3
Region
Table 3 4. Mean background surface water characteristics of the mine scenario watersheds.
Analyte
IDS (mg/L)
pH (field)
DO (mg/L)
Temperature (°C)
Specific Conductivity (pS/cm)
TSS (mg/L)
Ca (mg/L)
Mg(mg/L)
Na (mg/L)
K (mg/L)
Alkalinity (mg/L)
S04 (mg/L)
Cl (mg/L)
F (mg/L)
Hardness (mg/L)
Al (Mg/L)
As (Mg/L)
Ba (Mg/L)
Cd (Mg/L)
Cu (Mg/L)
Fe (Mg/L)
Mn(Mg/L)
Mo (Mg/L)
Ni(Mg/L)
Pb (Mg/L)
Zn (Mg/L)
CN (Mg/L)
DOC (mg/L)
South Fork
Koktuli River
44
7.0
10.2
4.77
55.5
2.21
6.34
1.41
2.35
0.38
17.4
8.78
0.69
0.05
21.6
11
0.31
4.1
0.013
1.3
120
20
0.66
0.41
0.087
2.7
2.8
1.36
North Fork
Koktuli River
37
6.74
10.2
4.39
46.0
1.39
5.09
1.32
2.38
0.41
20.5
2.26
0.66
0.03
14.4
13
0.2
3.1
0.012
0.39
110
10
0.19
3.0
0.39
1.8
1.9
1.5
Upper
Talarik Creek
51.2
6.99
10.5
4.04
73.4
2.52
8.77
2.12
2.82
0.44
31.8
5.48
0.29
0.39
26.5
13
0.67
5.5
0.012
0.42
110
21
0.2
0.63
0.067
2.0
1.5
1.57
Notes:
Filtered concentrations are used for hardness and trace elements.
Source: PLP2011.
3.6 Seismicity
The Alaska Earthquake Information Center and U.S. Geological Survey (USGS) collect data on
earthquakes occurring in Alaska at seismological monitoring stations throughout the state. Earthquakes
in Alaska range from minor events detected only by sensitive instruments, to the largest earthquake
ever recorded in North America (the 1964 Good Friday earthquake near Anchorage, magnitude 9.2)
(Table 3-5, Figure 3-15).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-32
April 2013
-------
Chapter 3
Region
Table 3 5. Examples of earthquakes in Alaska.
Date
March 28, 1964
Novembers, 2002
September 25, 1985
July 13, 2007
March 25, 2012
Depth (km)
25
4.2
184
6.2
12
Magnitude3
9.2
7.2
4.9
4.3
3.0
Distance and Direction from the Pebble
Deposit
469 km east-northeast
593 km northeast
61 km southeast
30 km west-southwest
122 km east
Notes:
a Local magnitude as reported by the Alaska Earthquake Information Center. Note that earthquakes in the range of magnitudes 1.5 to 3.6
occur regularly in the Lake Clark area (data not shown). These earthquakes are centered at a depth of 100 km or greater.
Southwestern Alaska experiences a large number of earthquakes related to the presence of four active
moving blocks of crust associated with large fault systems. These faults are, from north to south, the
Tintina-Kaltag Fault, the Iditarod-Nixon Fork Fault, the Denali-Farewell Fault, the Lake Clark-Castle
Mountain Fault system, the Bruin Bay Fault, and the Border Ranges Fault (Figure 3-15). Some sections
along these faults are seismically active and have generated earthquakes in the past. The size of an
earthquake is directly related to the area of the fault that ruptures; therefore, longer faults are capable of
producing larger earthquakes. The damage caused by an earthquake is related to the size of and distance
from the earthquake. The effects of an earthquake diminish with distance, so more damage occurs at the
epicenter than at a point several kilometers away.
The Lake Clark-Castle Mountain Fault system, with a mapped length of 225 km, is the fault located
nearest to the Pebble deposit. The northeast-southwest trending Lake Clark Fault is the western
extension of the Castle Mountain Fault (Koehler and Reger 2011). The western terminus of the Lake
Clark Fault has not been identified, but was originally interpreted to be near the western edge of Lake
Clark. Recent studies by USGS reinterpreted the position of the Lake Clark Fault further to the
northwest, potentially bringing it as close as 16 km to the Pebble deposit (Haeussler and Saltus 2004).
Haeussler and Saltus (2004) acknowledge that the fault could extend closer than 16 km, but data are not
available to support this interpretation.
There are few residents and no long-term seismic monitoring station records in the area of the Pebble
deposit, which make it difficult to assess accurately the recent seismic history of the area. As a result, the
paleoseismic history of the western part of the Lake Clark Fault is unknown (Koehler and Reger 2011).
USGS has concluded that there is no evidence for fault activity or seismic hazard associated with the
Lake Clark Fault in the past 1.8 million years, and no evidence of movement along the fault northeast of
the Pebble deposit since the last glaciations 11,000 to 12,000 years ago (Haeussler and Waythomas
2011).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-33
April 2013
-------
Chapter 3
Region
Figure 3 15. Seismic activity in southwestern Alaska. Location and magnitude of significant,
historic earthquakes (USGS 2010) that caused deaths, property damage, and geological effects, or
were otherwise experienced, are shown. Fault lines are based on Haeussler and Saltus (2004),
including the preferred drawing of the Lake Clark Fault (dashed purple line).
O 5.1-6.0
Approximate Pebble Deposit Location
Transportation Corridor
Fault
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-34
April 2013
-------
Chapter 3 Region
The 1980 USGS map of the structural geology of the Iliamna Lake quadrangle shows several mapped
faults in the Tertiary-age volcanic rocks that host the area's mineral deposits. Geologic mapping
conducted by consulting firms for the Pebble Limited Partnership (PLP) identified numerous faults in
the Pebble deposit area. The mapped faults shown in both these sources are all considerably shorter
than the Lake Clark Fault, and therefore by themselves have a very limited capability to produce
damaging earthquakes. The largest mapped fault in the Pebble deposit area is an unnamed northwest-
trending fault approximately 13 km southwest of the deposit, approximately 16 km in length. There are
several short (less than 4 km) faults mapped within and near the mine site watersheds (the Z-series
faults), about half of which have northeast-south west orientations. The faults show vertical
displacement ranging from tens of meters to over 900 m, and are interpreted to have formed coincident
with mineralization (Ghaffari et al. 2011). Although there is no evidence that the Lake Clark Fault
extends closer than 16 km to the Pebble deposit, and there is no evidence of a continuous link between
the Lake Clark Fault and the northeast-trending faults at the mine site, mapping the extent of subsurface
faults over long, remote distances is difficult and has a high level of uncertainty.
Not all earthquakes occur along the mapped sections of faults. In some instances, stresses build up and
cause earthquakes in rock outside of known pre-existing faults. Earthquakes can occur on previously
unidentified, minor, or otherwise inactive faults, or along deeper faults that are not exposed at the
surface. While these floating earthquakes are generally smaller and less frequent than those associated
with faults, they may occur at locations closer to critical structures than the nearest mapped capable
fault. Small earthquakes can be induced when reservoirs or impoundments are constructed (Kisslinger
1976), altering the soil and rock stresses and increasing pore pressure along pre-existing zones of
weakness. Induced earthquakes are generally small, but can occur frequently and cause landslides and
structural damage to earthen structures.
Interpreting seismicity in the Bristol Bay area is difficult because of the remoteness of the area, its
complex bedrock geology overlain by multiple episodes of glacial activity, and the lack of historical
records on seismicity. Thus, there is a high degree of uncertainty in determining the location and extent
of faults, their capability to produce earthquakes, whether these or other geologic features have been
the source of past earthquakes, and whether they have a realistic potential for producing future
earthquakes. Large earthquakes have return periods of hundreds to thousands of years, so there may be
no recorded or anecdotal evidence of the largest earthquakes on which to base future predictions.
3.7 Existing Development
Unlike most other areas supporting Pacific salmon populations, the Bristol Bay watershed is
undisturbed by significant human development. It is located in one of the last remaining virtually
roadless areas in the United States (Section 6.1.3). Large-scale, human-caused modification of the
landscape—a factor contributing to extinction risk for many native salmonid populations (Nehlsen et al.
1991)—is absent, and development in the watershed consists of only a small number of towns, villages,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ 35 April 2013
-------
Chapter 3 Region
and roads. The Bristol Bay watershed also is home to Iliamna Lake, the largest undeveloped lake in the
United States.
The primary human manipulation of the Bristol Bay ecosystem is the marine harvest of approximately
70% of salmon returning to spawn. However, commercial salmon harvests are the Alaska Department of
Fish and Game's (ADF&G's) second priority for fish management; its first priority is to ensure that
sufficient fish migrate into rivers to maintain a sustainable fishery, and thus sustainable salmon-based
ecosystems. No hatchery fish are reared or released in the Bristol Bay watershed, whereas
approximately 5 billion hatchery-reared juvenile salmon are released annually across the North Pacific
(Irvine etal. 2009).
3.8 Climate Change
Thus far, this chapter has focused on the current physical environment in the Bristol Bay watershed. In
the future, over time scales at which large-scale mining will potentially affect these watersheds, this
physical environment is likely to change substantially—particularly in terms of climate and, by
extension, hydrology. Over the past 60 years, much of Alaska has been warming at twice the rate of the
United States and many parts of the world (ACIA 2004). Throughout Alaska, changes such as warmer
temperatures, melting glaciers, declining sea ice, and declining permafrost have already occurred
(Serreze et al. 2000, Stafford et al. 2000, ACIA 2004, Hinzman et al. 2005, Listen and Hiemstra 2011,
Markon et al. 2012). However, there is limited evidence over the last decade that suggests air
temperature in much of Alaska has cooled, due to changes in the Pacific Decadal Oscillation and
weakening of the Aleutian low (Wendler et al. 2012). Climate models suggest that warming throughout
Alaska is projected to continue, and it is likely to lead to changes in the type and timing of precipitation,
decreased snowpack and earlier spring snowmelt, and subsequent changes in hydrology similar to
projections in Arctic regions (Hinzman et al. 2005).
Using methods detailed in Box 3-4, we used the multi-model average A2 emissions scenario developed
by SNAP (2012) to generate 30-year means for future temperature and precipitation patterns in the
Bristol Bay region. We focus on characterizing possible climate change impacts using the A2 emissions
scenario 30-year mean for the end of this century (2071-2100) as an upper bound estimate of climate
change effects expected for this region with current modeling. Effects earlier in the century or with more
benign emission scenarios show similar direction in temperature and precipitation, but smaller
magnitude.
By the end of the century, based on SNAP (2012) data for the A2 emissions scenario, the multi-model
average annual air temperature in the Bristol Bay region is projected to increase by approximately 4°C,
with an approximately 6°C increase occurring in the winter months. Increases in air temperature are
likely to affect the accumulation and melt of snowpack, the extent of lake ice, and the timing of spring ice
break up, and result in increased water temperatures. Although we are unable to predict a change in
extreme events, changes in precipitation patterns are likely to occur (Christensen et al. 2007, Peacock
2012, Markon etal. 2012, Salathe 2006), with rain-on-snow events becoming more common. The effect
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ og April 2013
-------
Chapter 3 Region
of increases in rain-on-snow events on the frequency or volume of floods is unclear. Storm patterns also
may change, although the increased likelihood of extreme events occurring and potential impacts on
flooding are unknown. Changes in the seasonality of precipitation, snowpack, and the timing of
snowmelt will likely affect the flow regimes and may result in water availability changes, particularly in
terms of decreased water availability in summer. Based on temperature, precipitation, and
evapotranspiration projections, the landscape will likely be warmer and wetter annually; however, due
to method limitations we are not able to determine how evapotranspiration will affect water availability
on the landscape seasonally (Box 3-4).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o 37 April 2013
-------
Chapter 3 Region
BOX 3 4. METHODS FOR CLIMATE CHANGE PROJECTIONS
To project temperature and precipitation changes over the next century, we used data from Scenarios
Network for Alaska and Arctic Planning (SNAP). A full description of the SNAP data and methodology used is
available on the SNAP website (SNAP 2012).
From the SNAP dataset, we used downscaled values of monthly mean temperature and precipitation. The
historical dataset is derived from the Climate Research Unit (CPU) at the University of East Anglia for 1901
to 2009 (CPU 2012). The CPU data are downscaled using the Parameter-elevation Regressions on
Independent Slopes Model (PRISM) 1971 to 2000 monthly climatologies for Alaska (Prism Climate Group
2012), which take into account elevation, slope, and aspect. SNAP then developed downscaled monthly
projections of temperature and climate for Alaska under three emissions scenarios developed by the
Intergovernmental Panel on Climate Change for the Coupled Model Intercom pa rison Project. SNAP uses five
global climate models (GCMs) [cccma_cgcm31, mpi-echam5, gfdl_cm21, ukmo_hadcm3, and
miroc3_2_medres] that best characterize the Arctic region up to the year 2100 (Walsh et al. 2008). These
emissions scenarios are:
• the Bl scenario, which represents a best-case emissions scenario;
• the A1B scenario, which represents a middle-of-the-road emissions scenario; and
• the A2 scenario, which represents a worst-case emissions scenario.
For this assessment, we use the SNAP5-model average for the A2 scenario of the best-performing GCMs to
consider a worst-case climate change scenario for the Bristol Bay region. Although uncertainty is inherent in
climate modeling due to many factors, the SNAP 5-model average tends to perform better than any single
model under the A2 scenario. Using the SNAP model, we calculated 30-year normal values, or average
values over a 30-year period, for temperature and precipitation for 1971 to 2000 (historical) and for 2011
to 2040, 2041 to 2070, and 2071 to 2100 under the three emissions scenarios. We focus on the A2
scenario for the years 2071 to 2099 (the year 2100 is not included because one of the GCMs used in the
average did not include that year). Using the SNAP data, we calculated changes in temperature and
precipitation at three scales: the Bristol Bay watershed (Figure 2-3), the Nushagak and Kvichak River
watersheds (Figure 2-4), and the mine scenario watersheds (Figure 2-5). We also calculated annual
potential evapotranspiration (PET) (Hamon 1961) and annual water surplus (annual precipitation minus
PET) for the Bristol Bay watershed and the Nushagak and Kvichak River watersheds.
Data for the appropriate watersheds were extracted from the SNAP dataset, which covers the entire state of
Alaska. The resolution of the SNAP dataset is a 771-m grid. Any grid pixel intersecting a watershed boundary
was included, even if the intersection was minimal, to account for the full range of possible
temperature/precipitation values across the watersheds. In all cases, the values reported in the assessment
represent the geographic spatial average across the entire watershed over an average of 30 years.
Precipitation and temperature differences between the two periods are calculated as the geographic spatial
average across the entire watershed of the raster representing the A2 scenario (2071 to 2099), minus the
present period. Precipitation percent differences are calculated as the geographic spatial average across
the entire watershed of the raster representing the difference between the A2 scenario (2071 to 2099) and
the present period, divided by the present period and multiplied by 100.
Water surpluses under historical and future periods were calculated for each calendar month and summed
to arrive at annual values. Differences between periods were calculated by subtracting the present value
from the A2 scenario (2071 to 2099) value. It is important to remember that surplus measurements are
calculated at the annual level and do not represent monthly or seasonal differences across a single scenario
or between multiple scenarios.
Uncertainty is an inherent issue when dealing with projected temperature, precipitation, and water surplus
values because of local variability and uncertainty in global climate models. Using average values for the five
best-performing GCMs for the Arctic, and calculating mean values over 30-year periods, help to reduce
uncertainty; however, this averaging also leads to the decrease in precision in predicting extreme events.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o og April 2013
-------
Chapter 3
Region
3.8.1 Climate Change Projections for the Bristol Bay Region
Across the entire Bristol Bay watershed, average temperature is projected to increase by approximately
4°C by the end of the century (Table 3-6, Figure 3-16), and winter temperature is projected to increase
the most (Table 3-6). Similar patterns are projected in the Nushagak and Kvichak River watersheds
(Table 3-6).
By the end of the century, precipitation is projected to increase on the order of 30% across the Bristol
Bay watershed, for a total increase of approximately 250 mm annually (Table 3-7, Figure 3-17). In the
Nushagak and Kvichak River watersheds, precipitation is projected to increase on the order of 30%, for
a total increase of approximately 270 mm of precipitation annually (Table 3-7). At both spatial scales,
increases in precipitation are expected to occur in all four seasons (Table 3-7). Based on
evapotranspiration calculations, annual water surpluses of 144 mm and 165 mm are projected for the
Bristol Bay watershed and the Nushagak and Kvichak River watersheds, respectively (Table 3-8,
Figure 3-18). Our simulated changes of temperature and precipitation using the SNAP (2012) data for
the Bristol Bay region are within the range of changes projected by other studies concentrating on
Alaska and the Arctic (Christensen et al. 2007, Peacock 2012, Markon et al. 2012).
Table 3 6. Average annual and seasonal air temperature for historical and projected periods (SNAP
2012), and the difference between these periods across two spatial scales. Temperature was
calculated as average values over each 30 year period. Number in parentheses one standard
deviation.
Scale
Bristol Bay Watershed
(Scale 1)
Nushagak and Kvichak
River Watersheds
(Scale 2)
Season
Annual
Winter
Spring
Summer
Fall
Annual
Winter
Spring
Summer
Fall
Historical Temperature
(1971-2000)
(°C)
1(1)
-8(2)
0(1)
11(2)
1(2)
1(1)
-9(1)
0(1)
11(2)
0(2)
Projected Temperature
(2017-2099)
(°C)
5(1)
-2(2)
4(1)
14(2)
5(2)
5(1)
-3(1)
4(1)
14(2)
5(2)
Difference
(°C)
4 (0.2)
6(1)
4 (0.2)
3 (0.07)
4 (0.3)
4 (0.2)
6 (0.4)
3 (0.2)
3 (0.05)
4 (0.07)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-39
April 2013
-------
Chapter 3
Region
Table 3 7. Average annual and seasonal precipitation for historical and projected periods, and the
difference between these periods across two spatial scales (SNAP 2012). Precipitation was
calculated as average values over each 30 year time period. Number in parentheses equals one
standard deviation.
Scale
Bristol Bay Watershed
(Scale 1)
Nushagak and Kvichak
River Watersheds
(Scale 2)
Season
Annual
Winter
Spring
Summer
Fall
Annual
Winter
Spring
Summer
Fall
Historical Precipitation
(1971-2000)
(mm)
847 (421)
177 (121)
150 (91)
234 (97)
286 (141)
795 (336)
160 (79)
138 (67)
226 (84)
271 (123)
Projected Precipitation
(2017-2099)
(mm)
1,095 (512)
229 (143)
196 (112)
303 (117)
367 (170)
1,062 (430)
215 (97)
189 (90)
300 (107)
357 (152)
Difference
(mm)
248 (104)
52 (27)
45 (25)
69 (25)
81 (34)
267 (95)
55 (21)
51 (23)
75 (24)
86 (32)
Table 3 8. Average annual water surplus for historical and projected periods, and the difference
between these periods across two spatial scales (SNAP 2012). Number in parentheses equals one
standard deviation.
Scale
Bristol Bay Watershed
(Scale 1)
Nushagak and Kvichak River
Watershed
(Scale 2)
Historical Surplus
(1971-2000)
(mm)
400 (441)
341 (359)
Projected Surplus
(2017-2099)
(mm)
544 (534)
506 (456)
Difference
(mm)
144 (106)
165 (99)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-40
April 2013
-------
Chapter 3
Region
Figure 3 16. h
emissions scenario (2071 to 2099), and (C) the temperature change between these two climate scenarios (SNAP 2012). See Box 3 4 for
additional details.
A. Historical Conditions
Temperature (°C)
C. Temperature Change
4.5 - 5.0
4.0-4.5
3.5-4.0
3.0 - 3.5
2.5-3.0
2.0-2.5
1.5 - 2.0
1.0 -1.5
0.5 -1.0
0.0 - 0.5
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-41
April 2013
-------
Chapter 3
Region
A. Historical Conditions
Precipitation (mm)
>2000
1,500 - 2,000
1,000 -1,500
750 -1,000
600 - 750
550 - 600
500 - 550
450 - 500
400 - 450
350 - 400
325 - 350
C. Precipitation Change
Precipitation Difference (mm)
>600
500-600
400 - 500
300 - 400
250-300
200 - 250
150 - 200
100 -150
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-42
April 2013
-------
Chapter 3
Region
Figure 3 18. h
conditions (1971 to 2000) and (B) the A2 emissions scenario (2071 to 2099), and (C) the water surplus change between these two climate
scenarios (SNAP 2012). See Box 3 4 for description of surplus calculations.
A. Historical Conditions
Surplus (mm)
>2,000
1,500 - 2,000
1,000 -1,500
750 -1,000
500 - 750
400 - 500
300 - 400
200 - 300
100 - 200
0-100
-200 - 0
C. Water Surplus Change
Difference in Surplus (mm)
>500
250 - 500
200-250
150-200
100 -150
50 -100
0-50
-10-0
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
3-43
April 2013
-------
Chapter 3 Region
3.8.2 Potential Climate Change Effects
There are likely to be hydrological impacts associated with projected changes in temperature,
precipitation, and evapotranspiration in the Bristol Bay watershed, including changes in the magnitude
and timing of flow that are likely to affect salmon habitat and populations. When temperature increases
in freshwater environments, community structure, habitat, and salmon populations can be affected
(Eaton and Scheller 1996, Hauer et al. 1997). With warmer temperatures and changes in the type,
timing, and amount of precipitation, there will likely be changes in snowpack, a shift in the timing of
spring snowmelt, and changes in the type of precipitation falling (Barnett et al. 2005). With these
changes, there will be alterations to the natural flow regime in both magnitude and timing, and a likely
decline in seasonal water availability, mirroring already observed changes in other systems such as the
Pacific Northwest (Mote etal. 2003).
These hydrologic flow regime changes may affect salmon populations during spawning and smolt
migrations, and can scour streambeds leading to the loss of salmon eggs (Lisle 1989, Montgomery et al.
1996, Steen and Quinn 1999, Mote et al. 2003, Lawson et al. 2004, Stewart et al. 2004). Changes in
hydrology are likely to affect existing habitat via changes in water volume and velocity along with
channel forms, which may lead to declines in habitat availability for spawning and rearing salmon
populations. Changes to baseflow, depending on the groundwater and surface water interactions, are
likely to affect the amount of wetlands in the Bristol Bay watershed, such that if there are drier baseflow
conditions, wetlands are likely to decrease. Although we are unable to predict whether baseflow will
increase or decrease, any changes in baseflow will likely affect water temperature (in addition to the
direct effects of increased air temperature on water temperature).
Both the hydrology and water temperature of the freshwater system affect critical life stages of
salmonid species. Furthermore, these hydrological changes are likely to have differential effects on
populations of salmon species depending on the amount of time they spend rearing in freshwater
habitats, their life stage, and their ability to adapt to changes in environmental conditions. Pink and
chum salmon are likely to be affected by temperature increases early in egg incubation, which can affect
timing of emergence, migration to the ocean, and potential mismatch in the timing of peak food
abundance in the marine environment (Bryant 2009). For example, the average migration time for one
population of pink salmon in southeast Alaska occurs nearly 2 weeks earlier than it did 40 years ago
(Kovach et al. 2012). For sockeye salmon that typically rear in the freshwater environment for 1 to
2 years, temperature increases may affect the timing of life stages, including spawning and emergence of
fry, as well as the growth and survival of the fry that rear in lakes (Healey 2011, Martins et al. 2012).
Changes in precipitation and hydrology also may affect access to lakes and spawning locations, and high-
intensity rainfall may increase sedimentation in spawning streams and rearing lakes for sockeye salmon
(Bryant 2009). Rich etal. (2009) hypothesized that warmer temperature was a factor in poor
recruitment of sockeye salmon in the Kvichak River system. For Chinook salmon, increases in
temperature are likely to affect incubation and fry emergence (Beer and Anderson 2001), which may
affect growth, survival, and timing of migration to the ocean (Heming et al. 1982, Taylor 1990, Berggren
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment I~TT April 2013
-------
Chapter 3 Region
and Filardo 1993). Coho salmon incubation and timing of emergence are also affected by increases in
temperature (Tangetal. 1987).
Populations of Pacific salmon species are likely to respond and adapt to changes in temperature,
precipitation, and hydrology in different ways and the geographic location of populations is likely to
affect their ability to adapt to these changes. Studies have predicted that the reproductive success of
salmon populations in Washington is likely to decline over the next century (Battin et al. 2007, Mantua
et al. 2010), and freshwater temperature increases in the Fraser River will negatively affect growth and
survival of sockeye salmon at all life stages (Healey 2011). The genetic and life history diversity within
and among the Bristol Bay Pacific salmon populations (Section 5.2.4) will likely be crucial for
maintaining the resiliency of the salmon stocks under a future environment characterized by climate
change and increased anthropogenic stressors (Hilborn et al. 2003, Schindler et al. 2010, Rogers and
Schindler2011).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o 45 April 2013
-------
4.1 Mineral Deposits and Mining in the Bristol Bay Watershed
Significant mineral resources are located in Alaska, and the state has a long mining history. Russian
explorers began searching for placer gold in the early 1800s, and substantial placer deposits have been
found in many areas of the state; more recently, hard rock exploration has increased throughout the
region. Alaska mines range in size from small, recreational suction dredging operations to large-scale
commercial operations, for a variety of deposit types (Table 4-1).
Several known mineral deposits with potentially economically significant resources are located in the
Nushagak and Kvichak River watersheds (Table 13-1), and active exploration of deposits is occurring in
a number of claim blocks (Figure 13-1). Of deposit types occurring or likely to occur in the region,
porphyry copper, intrusion-related gold, and copper and iron skarn may indicate economically viable
mining, thereby prompting large-scale development. Thus, the development of a number of mines, of
varying sizes, is plausible in this region—and once the infrastructure for one mine is available, it would
likely facilitate the development of additional mines (Chapter 13).
The potential for large-scale mining development within the watershed is greatest for porphyry copper
deposits, most notably at the Pebble deposit. Significant exploration activity has been ongoing at this
deposit for many years, and the information available provides the most complete description of
potential mining in the region. Because the Pebble deposit is the most likely deposit to be developed in
the near term, this assessment focuses exclusively on porphyry copper deposits. However, much of the
discussion of mining methods (Section 4.2.3) applies to all types of disseminated ore deposits (i.e., ores
with low concentrations of metal spread throughout the body of rock).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-1
April 2013
-------
Chapter 4
Type of Development
Table 4 1. Characteristics of past, existing, or potential large mines in Alaska.
Mine
Location
Target metals
Ore type
Ore grade
quality
Operational life
(years)
Extraction type
Total resource
(million metric
tons)
Ore processing
rate (metric
tons/day)
Total waste rock
(million metric
tons)
Tailings disposal
Tailings amount
(million metric
tons)
Tailings footprint
(km=)
Dam height (m)
Acid mine
drainage
potential
Kennecott
Copper River
basin, in
Wrangell-St.
Elias National
Park
Copper, silver
Massive sulfide
Very high
27 (1911-
1938)
Underground
slope mining
-4.5
-91
<0.9
On Kennicott
Glacier
<0.9
NA
NA
No
Donlin
13 miles N of
village of Crooked
Creek and
Kuskokwim River
Gold
Gold-bearing
quartz
Moderate
22
Open pits (2)
491b
48,524
1900
Dams/ponds (2)
426
5.4
143 (largest of
multiple dams)
Yes
Fort Knox
26 miles NEof
Fairbanks
Gold
Oxide ore body
Low
20
Open pit
401
33,000-45,000
338
Dam/pond
181
4.5
111
No
Greens Creek
18 miles SW of
Juneau, in
Admiralty Island
National
Monument
Zinc, lead, silver,
gold
Massive sulfide
High
35-50
Underground
slope mining
29
1,524
-1.8
Dry tailings
-13.6
0.25
NA
Yes
Kensington
45 miles NW of
Juneau, belween
Berners Bay and
Lynn Canal
Gold
Gold-bearing
quartz
Moderate
10
Underground slope
mining
24
1,134
1.5
Lake disposal
4.1
0.24
2/b
No
Pogo
85 miles ESEof
Fairbanks
Gold
Gold-bearing
quartz
Moderate
11
Underground
slope mining
9.1
2,267
1.7
Drylailings
4.9
0.12
NA
No
Red Dog
Western Brooks
Range, 82 miles N
of Kolzebue and 46
miles from Ihe
Chukchi Sea
Zinc, lead
Massive sulfide
High
42 (1989-2031)
Open pils (2)
171
7,500-8,300
142
Dam/pond
91
3
63
Yes
Pebble (78-yr)a
Headwaters of
Ihree slreams
running into Ihe
Nushagak and
Kvichak Rivers
Copper, gold,
molybdenum
Porphyry copper
Low
78
Open pil
5,920
208,000
14,600
Dams/ponds
(multiple)
5,910
46
209 (largesl of
multiple dams)
Yes
Notes:
a Ghaffari etal. 2011.
b Novagold 2012.
NA = not applicable.
Source: Levit and Chambers 2012, except as noted.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-2
April 2013
-------
Chapter 4
Type of Development
4.2 Porphyry Copper Deposits and Mining Processes
4.2.1 Genesis of Porphyry Copper Deposits
Porphyry copper deposits are found around the world, often occurring in clusters (Lipman and Sawyer
1985, Singer et al. 2001, Anderson et al. 2009) in areas with active or ancient volcanism (Figure 4-1).
They are formed when hydrothermal systems are induced by the intrusion of magma into shallow rock
in the Earth's crust. Water carries dissolved sulfur-metallic minerals (sulfides) into crustal rock where
they precipitate (John et al. 2010). Minerals containing sulfur and metals are disseminated and
precipitate throughout the affected rock zone in concentrations typically less than 1% (Table 4-2)
(Singer et al. 2008). Porphyry copper deposits range in size from millions to billions of tons (Table 4-2).
The well-delineated Pebble deposit is at the upper end of the total size range; thus, any additional
deposits found in the Nushagak and Kvichak River watersheds are likely to be much smaller than the
Pebble deposit.
Table 4 2. Global grade and tonnage summary statistics for porphyry copper deposits.
Parameter
Tonnage (Mt)
Copper grade (%)
Molybdenum grade (%)
Silver grade (g/t)
Gold grade (g/t)
10th Percentile
30
0.26
0.0
0.0
0.0
50th Percentile
250
0.44
0.004
0.0
0.0
90th Percentile
1,400
0.73
0.023
3.0
0.20
Pebble Deposit3
10,777
0.34
0.023
unknown
0.31
Notes:
a Pebble deposit information is based on 0.3% copper cut-off grade, and includes measured, indicated, and inferred resources from PLP.
Mt = million tons; g/t = grams per ton.
Sources: Singer etal. 2008; Appendix H.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-3
April 2013
-------
Chapter 4
Type of Development
Figure 4 1. Porphyry copper deposits around the world. Values are from the database compiled by and described in Singer et al. 2008. Other
mines and mining regions mentioned in the text also are shown on the map.
Clark Fork River
Soda Butte Creek
Kingston Fossil Plant
Slavs
-AurulSA
Los Frailes -
Chuquicamata
Bajo de la Alumbrera -
•« \ ^
I *"
f &
e .,
c * =; •••..
• Other Mines/Mine Areas
Porphyry Copper Deposits
Ore (millions of metric tons)
• 0 - 500
s 501-2,000
I 2,001-8,000
: 8.001-16,000
C 16,001-24,000
N
A
0 1,500 3.000
0 1.500 3,000
] Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-4
April 2013
-------
Chapter 4 Type of Development
4.2.2 Chemistry and Associated Risks of Porphyry Copper Deposits
Exposure to hazards associated with mining porphyry copper deposits can pose risks to aquatic and
terrestrial ecosystems and to human health. These risks can range from insignificant to extremely
harmful depending on a variety of factors that control the hazards, including site geology (both local and
regional), hydrologic setting, climate, and mining and ore processing methods. There are a variety of
geochemical models and approaches to understand and predict the water quality of releases to the
environment; however, our ability to make predictions is limited because of data insufficiency and the
inherent complexity of natural materials and their environment.
Sources of hazards from porphyry copper mines can be grouped into four broad, interrelated categories:
acid-generating potential, trace elements and their mobilities, mining and ore processing methods, and
waste disposal practices. The relative importance of these categories will vary from deposit to deposit,
but some generalization can be made for porphyry copper deposits as a whole. In this section we
consider those categories related to environmental chemistry, acid-generating potential, and trace
elements (categories related to mining processes are described in Section 4.2.3).
Mining processes expose rocks and their associated minerals to atmospheric conditions that cause
weathering, which releases minerals (e.g., copper minerals) from the rock matrix. Grinding methods
used in these processes create materials that have a high specific surface area, which accelerates the rate
of weathering. Porphyry copper deposits are characterized by the presence of sulfide minerals, and
oxidation of sulfide minerals creates acidity that can further accelerate weathering rates. Because most
metals and other elements become more soluble as pH decreases, the acid-generating or acid-
neutralizing potentials of waste rock, tailings, and mine walls are of prime importance in determining
potential environmental risks associated with exposure to metals and certain elements in the aquatic
environment.
One way to predict if acid generation has the potential to occur is to perform acid-base accounting tests.
Acid-base accounting tests are rapid methods to determine the acid-generation potential (AP) and
neutralization potential (NP) of a rock or mining waste material, independent of reaction rates. These
potentials are then compared to one another by either their differences or their ratios, with the net
neutralization potential (NNP) being NP-AP and the neutralizing potential ratio (NPR) being NP/AP. AP,
NP, and NNP typically are expressed in units of kilograms of calcium carbonate per metric ton of waste
material (kg CaCOs/metric ton). Positive NNP values are net alkaline and negative values are net acidic.
Although methods used for acid-base accounting have known limitations, it is common industry practice
to consider materials that have an NPR of 1 or less as potentially acid-generating (PAG) and materials
that have an NPR greater than 4 (Brodie etal. 1991, Price and Errington 1998) as being non-acid-
generating (NAG). Materials that have a ratio between 1 and 4 require further testing via kinetic tests
(e.g., ASTM D5744-07el) and geochemical assessment for classification (Brodie etal. 1991, Price 2009,
Price and Errington 1998). This further testing and assessment are necessary because if neutralizing
minerals react before acid-generating minerals, the neutralizing effect may not be realized and acid
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 77April 2013
-------
Chapter 4 Type of Development
might be generated at a later time—that is, pH of the system may decrease over time as neutralizing
materials are used up. Additionally, some toxic elements (e.g., selenium and arsenic) may be released
from mining materials under neutral or higher pH conditions, which would be observed during kinetic
leaching tests conducted at variable pH values.
In general, the rocks associated with porphyry copper deposits tend to straddle the boundary between
being net acidic and net alkaline, as illustrated by Borden (2003) for the Bingham Canyon, Utah
porphyry copper deposit (Figure 4-2 [A]). AP values for porphyry copper deposits typically correlate
with the distribution of pyrite. The pyrite-poor, low-grade core corresponds to the central part of the
Bingham Canyon deposit, where NNP values are greater than zero. Moving outward from the core to the
ore shell and pyrite shell, pyrite abundance increases and NNP values become progressively more
negative (Figure 4-2 [B]).
4.2.3 Overview of the Mining Process
Developing a mine requires establishing surface or underground mine workings that allow access to the
ore body. The scope and complexity of development-related activities vary depending on the
characteristics of each project, but typically include the following components.
• Site preparation (clearing, stripping, and grading). Topsoil and overburden are removed and
typically stockpiled for later use in mine reclamation.
• Construction of mine site infrastructure. Specific requirements depend on the size and type of mine
operation, its location, and proposed mining, milling, and processing methods. Typical infrastructure
includes facilities for ore crushing, grinding, and other mineral separation processes; ore stockpiling
and waste rock disposal facilities; tailings storage facilities; water supply, treatment, and
distribution facilities; roads; pipelines; conveyers; and other infrastructure (e.g., offices, shops,
housing).
• Establishment of mine workings. Once the site is prepared and infrastructure is constructed, mine
workings are established: ore is extracted and processed, water at the site is managed and treated,
and tailings and waste rock are stored and managed.
At each stage of mine development, potential impacts on the environment and human health can be
reduced by ensuring effective implementation of proper design, construction, operation, and
management techniques and protocols (Box 4-1).
Any mining company must comply with a number of federal, state, and local laws when developing and
operating a mine. Compliance is facilitated through the regulatory permitting process and involves
multiple state and federal agencies (see Box 4-2 for additional detail on these regulatory requirements).
Regulations also serve to hold an operator accountable for potential future impacts, through
establishment of financial assurance requirements and imposition of fines or compliance orders upon
non-compliance with permit requirements (Box 4-3).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 77April 2013
-------
Chapter 4
Type of Development
Figure 4 2. Neutralizing potential at the Bingham Canyon porphyry copper deposit, Utah. (A) Plot of
neutralizing potential (NP) vs. acid generating potential (AP) for mineralized rock types. PAG denotes
potentially acid generating. Note that the range of uncertainty is indicated as 1 to 2 in this figure; in
the assessment, we use the more conservative range of 1 to 4. (B) Plan view of the distribution of net
neutralizing potential (NNP) values. Plots modified from Borden (2003).
1,000
100
B
1 £
Uncertain
\ ,'
X'
non-PAG
+ • ,4*
• 7 A.1
PAG
_ +
i-
+
-HH
10 100
AP(kgCaCO3/t)
1,000
NNP (kg CaCOs/t)
>0
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-7
April 2013
-------
Chapter 4 Type of Development
BOX 4 1. REDUCING MINING'S IMPACTS
Reducing mining's impacts on the environment and on human health requires proper planning, design,
construction, and operation; appropriate management and closure of waste and water containment and
treatment facilities; and monitoring and maintenance over all mine-life phases, including post-closure. Some
general methods for reducing adverse impacts of miningare provided here, along with information about
how these concepts are incorporated into the assessment.
Best Management Practices refer to specific measures for managing non-point source runoff (40 CFR
130.2(m)). Measures for minimizing and controlling sources of pollution in other situations are often
referred to as best practices, state of the practice, or simply mitigation measures. We assume that these
types of measures would be applied throughout a mine as it is constructed, operated, closed, and post-
closure. While we describe some measures as they are relevant to a discussion, it is not necessary, for the
purpose of this assessment, to describe them all.
Mitigation refers to all steps taken to avoid, minimize, treat, or compensate for potential adverse impacts on
the environment from a given activity. One example of a mitigation measure for avoidance is to avoid mining
a particularly reactive type of rock that might make future leachate management too difficult. Minimization
of an impact is conducted when avoidance is not feasible, and includes measures taken to lessen the
amount of contaminant released. An example of a mitigation measure to minimize an impact is to blend
known acid-producing material with sufficient neutralizing material. Treatment is required when
contaminants are released. An example is the diversion and collection of seepage from a waste rock pile for
passage through a wastewater treatment plant (e.g., precipitation, reverse osmosis, others) to meet
appropriate water quality criteria prior to release to the environment. Many elements of our mine scenarios
include mitigation measures and all are assumed to meet minimum regulatory requirements; Appendix I
contains further discussion of these and other mitigation measures.
Compensatory mitigation refers to the restoration, establishment, enhancement, and/or preservation of
wetlands, streams, or other aquatic resources to offset environmental losses resulting from unavoidable
impacts on waters of the United States authorized by Clean Water Act Section 404 permits issued by the
U.S. Army Corps of Engineers (40 CFR 230.93(a)(l)). This becomes an option only after all opportunities for
aquatic resource impact avoidance and minimization have been exhausted. See Box 7-2 and Appendix J for
a more complete discussion of compensatory mitigation.
Reclamation refers to restoration of a disturbed area to an acceptable form and planned use, following
closure of a mining operation. Our scenarios present some options that are feasible and common, and
assume that the site would be reclaimed according to statutory requirements, but it is outside the scope of
the assessment to evaluate a specific post-closure plan.
Remediation refers to fixing a problem that has become evident, such as an accidental release or spill of
product or waste material. For example, a tailings slurry spill would require remediation. The dam may have
been designed and constructed to properly mitigate (i.e., avoid or minimize) the potential for a spill, but an
accident or failure could cause contaminant release, thereby creating the need for remediation.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . „ April 2013
-------
Chapter 4
Type of Development
BOX 4 2. PERMITTING LARGE MINE PROJECTS IN ALASKA
Large mine projects in Alaska must comply with federal and state environmental laws, and many federal,
state, and local government permits and approvals are required before construction and operation of a
large hard rock mine can begin. The specific permits and approvals vary from project to project, depending
on the unique challenges posed by each mine.
Federal Laws and Agencies. The involvement of federal agencies varies for each mine, but most projects at
least require authorizations from the U.S. Army Corp of Engineers. Other agencies that may be involved
include (but are not limited to) the U.S. Environmental Protection Agency, the U.S. Fish and Wildlife Service,
the National Marine Fisheries Service, the U.S. Coast Guard, and the U.S. Department of Transportation.
Federal agency authorizations ensure that projects comply with the following applicable federal laws.
Clean Water Act
Clean Air Act
National Environmental Policy Act
National Historic Preservation Act
Resource Conservation and Recovery Act
Rivers and Harbors Act
Endangered Species Act
Bald Eagle Protection Act
Migratory Bird Act
Magnuson-Stevens Act
Mine Safety and Health Act
Alaska Department of Natural Resources Permits and Approvals. The Alaska Department of Natural
Resources (ADNR) Office of Project Management and Permitting coordinates the permitting of large mine
projects via the establishment of a large mine project team for each project. This project team is an
interagency group, coordinated by ADNR, that works cooperatively with large mine permit applicants and
operators, federal resource agencies, and the Alaskan public to ensure that projects are designed, operated,
and reclaimed in a manner consistent with the public interest.
ADNR may require the following permits and approvals.
• Plan of operations approval
Reclamation plan and bond approval
Right-of-way for access and utilities (roads, power
lines, pipelines)
Millsite lease
Permit to appropriate water
Dam safety certification (certificates of approval
to construct and operate a dam)
Upland ortideland leases
Material sale
Winter travel permits
Cultural resource authorization
Mining license
Alaska Department of Environmental Conservation Permits and Approvals. The Alaska Department of
Environmental Conservation may require the following permits related to wastewater management and
water and air quality.
• Waste management permit
• Alaska pollutant discharge elimination permit
• Domestic and non-domestic wastewater disposal
permits
Air quality permits
Approval to construct and operate in a public
water supply system
Plan review for non-domestic wastewater
treatment system
Plan review and construction approval for
domestic sewage system
• Oil discharge prevention and contingency plan
Other State Permits and Approvals. The state may require the following permits and approvals.
Certificate of reasonable assurance for 404
permits
Storm water discharge pollution prevention plan
Fish passage permit
Fish habitat permit
Utility permit on right of way
Driveway permit
Approval to transport hazardous materials
Life and fire safety plan check
State fire marshal plan review certificate
Certificate of inspection for fired and unfired
pressure vessel
Employer identification number
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-9
April 2013
-------
Chapter 4
Type of Development
BOX 4 3. FINANCIAL ASSURANCE
Many of the regulatory checks listed in Box 4-2 help to reduce potential impacts of mining on the
environment—although they do not ensure that a permitted mine will have negligible effects on the
environment. Even with the most stringent requirements, accidents and human error may cause mine
systems to fail. Thus, regulations also serve to hold an operator accountable for potential future impacts,
through establishment of financial assurance requirements and imposition of fines for non-compliance with
permit requirements.
Operators of hard rock miningfacilities in Alaska, including those facilities miningfor copper and gold, are
required by the State of Alaska to demonstrate financial assurance for reclamation, waste management,
and dam safety costs.
• Prior to commencement of hard rock mining operations on state-owned, federal, municipal, or private
land, a reclamation plan must be approved by the Alaska Department of Natural Resources (ADNR) and
financial assurance must be demonstrated in an amount necessary to ensure performance of the plan
(Alaska Statute 27.19).
• Hard rock mining operations disposing of solid or liquid waste material or heated process or cooling water
into the waters or onto the land of the state under a waste management and disposal permit may be
required by the Alaska Department of Environmental Conservation to demonstrate financial assurance in
an amount based on the estimated costs of required closure activities and post-closure monitoring for
the waste management area (Alaska Statute 46.03.100(f)).
• Operators of hard rock mines on state-owned or privately owned land seekingADNR approval to construct
mine tailings dams must demonstrate financial assurance to cover the cost of reclamation and post-
closure monitoring and maintenance of the dam (Alaska Statute 46.17).
Operators of hard rock miningfacilities in Alaska on land managed by the Bureau of Land Management or
U.S. Forest Service, including those facilities used in miningfor copper and gold, can be required by these
agencies to demonstrate additional financial assurance for reclamation (43 CFR 3809 and 36 CFR 228
SubpartA, respectively).
In addition to the financial assurance requirements of the state and the BLM, facilities operating under
leases, permits, or other forms of agreements for the development of hard rock minerals on tribal lands can
be required by the Bureau of Indian Affairs to demonstrate financial assurance to ensure compliance with
the terms and conditions of the mineral agreement and applicable statutes and regulations (25 CFR 211.24
and 225.30).
The State of Alaska allows several types of assurance, including cash, gold bullion, surety bonds,
reclamation trust funds, or irrevocable letters of credit. Surety bonds have become extremely difficult for
operators to secure and most companies currently provide an irrevocable letter of credit (ADNR 2012).
Financial assurance does not address chemical or tailings spills because of the greater degree of
uncertainty related to these accidents. Reclamation and mine closure can be estimated, but the cost of
cleaning up a spill is unpredictable.
Example Financial Assurance Amounts for Alaska Mines
Mine
Fort Knox
Kensington
Pogo
Red Dog
Amount EIA
$68,852,293
$7,345,015
$44,430,000
$305,150,000
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document Is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-10
April 2013
-------
Chapter 4 Type of Development
4.2.3.1 Extraction Methods
The low concentrations of disseminated metals in porphyry copper deposits require large amounts of
ore to enable a return on investment. Bulk or large-scale mining methods have been developed for this
purpose, and specific mining methods depend on ore quality and depth. A long-range mining plan is
usually developed first to match the final mine design with the available ore reserves, weighing
economics against engineering restrictions. This plan will be re-evaluated throughout the life of the
mine to reflect changes in the economy, increased knowledge of the ore body, and potential changes in
mining technology.
Porphyry copper deposits are most commonly mined using open pit and, less commonly, underground
methods (John etal. 2010). Open pit mining is typically used to extract ore where the top of a deposit is
within 100 m of the surface (Blight 2010). Excavation of a pit begins at the surface, with drilling and
blasting to strip overburden from the ore body surface. The equipment and materials used will fit the
economies of scale for the project (e.g., mine life, daily production). The ore is drilled and blasted
according to a blasting pattern. The size and spacing of the drill holes and the amount of explosives used
determine the size of the material that is loaded and hauled to the crushing plant. The pit is successively
enlarged until the pit limits are established by the extent of ore that can be profitably mined.
Pit design depends on the material characteristics of the ore and waste rock. The moisture content,
strength, and load-bearing capacity of the ore and waste help determine the angle of the pit slopes,
which generally are designed to be as steep as possible while still maintaining stability. A properly
designed pit reduces the stripping ratio, or the volume of waste rock to ore, thereby increasing
efficiency, potentially decreasing costs, and optimizing the amount of ore that can be mined
economically.
Block caving is an underground mining method used for large deposits with rock mass properties
amenable to sustainable caving action (Singer et al. 2008, Lusty and Hannis 2009, Blight 2010). Such
deposits typically have mineralization throughout the rock (e.g., porphyry copper deposits) and are too
deep to be mined economically by open pit methods. Block caving uses gravity to reduce the amount of
drilling and blasting required to extract ore. It involves tunneling to the bottom of the ore and
undercutting it, so that the deposit caves under its own unsupported weight. As ore is removed from
below, fractures spread throughout the block, which breaks into fragments and is removed from the
bottom of the enlarging void (Box 4-4).
Underground mining via block caving has a different set of costs than open pit mining, because of the
extensive drilling of tunnels and shafts through non-ore-bearing rocks needed to gain access to the ore.
Once begun, block caving generally requires less drilling and blasting, allows for less ore selectivity in
the mining process, and may require less labor relative to open pit mining. As with other types of
mining, the economics of block caving are determined by the prices of metals being extracted,
operational costs, and a number of other factors. If block caving allows the mining of additional ore that
could not be mined using open pit mining methods, it creates the need for additional tailings storage
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 777April 2013
-------
Chapter 4 Type of Development
capacity, increased capacity at the mill, increased consumption of utilities such as water and power,
increased production of metal concentrates, and possible extension of the mine life.
BOX 4 4. BLOCK CAVING AND SUBSIDENCE
Subsidence at the ground surface is an inevitable result of block caving, as the void formed by ore removed
from the ground is filled with overlying materials. The extent and rate at which subsidence occurs depends
on a number of factors, but it may occur within the first few years after mining begins, as the mined area
void grows and cracks in the overlying materials propagate to the surface. The subsided area is commonly
much larger than the footprint of the block cave operation, and the land form can remain unstable for a
period of years.
In addition to altering surface topography, subsidence can affect both the quantity and quality of surface
water and groundwater systems. As the ground subsides, cracks, fissures or pits may form. If these features
are connected either directly and indirectly to surface waters, it could lead to complete or partial dewatering
of streams and lowering of the groundwater table, as water flows to lower strata or the mine workings. Water
entering the underground mine from above would come in contact with broken mineralized rock that
remains underground in the mined areas. Unlike the situation prior to mining, when the sulfide ore body was
essentially isolated from oxygen, the mined area would be in contact with water and oxygen. This could lead
to oxidation of the sulfide minerals exposed during mining operations and, depending on the hydrogeology,
the potential generation of groundwater with elevated metals content from the mined area. Sealing of mined
surfaces may minimize interaction with oxygen and water, but complete sealing of all cracks and fissures is
unlikely. In addition to the lowering of groundwater levels, there also may be changes in groundwater flow
rates and impacts on water quality from changes in the chemical reactions and reaction rates with the
minerals orsurroundingstrata.
4.2.3.2 Water Treatment and Management
Because mine workings must be kept dry for the duration of mining activities, dewatering is required for
both open pit mines and block caving operations. Dewatering is accomplished by pumping water either
from the pit or underground workings, or from wells surrounding these areas. This pumping of water
may create a cone of depression, or a cone-shaped reduction in water level extending outward from the
point of water withdrawal, where water levels are lowest. Water extracted during dewatering typically
is pumped to lined process water ponds for use in the milling process. Excess water typically is tested
and, if necessary, treated before discharge.
In hard rock metal mining, most water use occurs during milling and separation operations. This water
is obtained from the mine site area, and then held in storage facilities until its use. However, much of the
water used in the mining process is recycled and reused. For example, the water used to pump tailings
slurry from the mill to the tailings storage facility (TSF) becomes available when the tailings solids settle
and excess overlying water is recycled back to the mill. Other sources of water use include power plant
cooling and transport of metal concentrate slurry (where transport occurs via pipeline).
In general, storm water runoff is diverted around mining components to keep it from becoming
contaminated, and then collected in sedimentation ponds to settle out suspended solids prior to use or
discharge to a stream. Stormwater runoff that contacts mining component areas (e.g., waste rock piles or
an open pit) may be contaminated with pollutants. Such water is directed to collection ponds and
treated before being used in mine processes or released. Seepage and leachate are directed to storage
ponds for containment, treated, and released to the environment. Tailings may be dewatered, and
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 771April 2013
-------
Chapter 4 Type of Development
reclaimed water directed to process water holding ponds for reuse. Surface water and groundwater are
monitored for contamination throughout mine operations, and are routed to a storage facility for
treatment if significant contamination is detected.
Water treatment options include physical or chemical methods—for example, reverse osmosis
(physical) and formation of precipitated solids (chemical)—used together or independently. The choice
of treatment methods and the chemicals used for treatment depend on the site's specific water
chemistry and the water's end use.
Once mining ceases, an open pit is typically allowed to fill with water. Acid-generating waste rock and
other potentially acid-generating (PAG) materials (e.g., pyrite-rich tailings) may be placed at the bottom
of the pit, to submerge these materials and reduce the potential for acid mine drainage once the pit fills.
In block caving, ore is removed from the ground and the resulting void is filled by overlying materials
(Box 4-4). After mining operations cease, groundwater fills in the remaining pore spaces in the void.
4.2.3.3 Ore Processing
Generally, two streams of materials come from a mine: ore and waste rock (Figure 4-3). Ore is rock with
sufficient amounts of metals to be economically processed. Waste rock is material that has little or no
economic value at the time of disturbance, although it may have recoverable value at a future time
(i.e., under different technology or economic conditions).
Ore blasted from a porphyry copper mine typically is hauled to a crushing plant near or in the mine pit
(Figure 4-3). The crushing plant reduces ore to particle sizes manageable in the processing mill (e.g., less
than 15 cm; Ghaffari et al. 2011). Crushed ore is carried by truck or conveyer to a ball mill, where
particle size is further reduced (e.g., less than 200 um; Ghaffari et al. 2011) to maximize the recovery of
metals. The milled ore is subjected to a flotation process with an aqueous mixture of chemical reagents
(Box 4-5) to collect valuable copper, molybdenum, and gold minerals in a copper-molybdenum
concentrate, which also contains gold. Bulk tailings are the non-acid-generating materials left after the
first flotation circuit, and are directed to a TSF (Figure 4-3). The copper-molybdenum (+gold)
concentrate may be fed through a second ball mill to regrind the particles (e.g., to less than 25 um;
Ghaffari et al. 2011). Once sufficiently sized, the regrind concentrate is directed into a second flotation
process and then to a copper-molybdenum separation process. Final products are a copper concentrate
that includes gold, a molybdenum concentrate, and pyritic tailings (Figure 4-3).
The most profound influence that ore processing can have on long-term management of a mine site
centers on the fate of pyrite (Fuerstenau et al. 2007). Traditionally, PAG and NAG tailings were
discharged together, thereby contributing to the acid-generating potential of the TSF. It is possible to use
a technique called selective flotation to separate most of the pyrite into the cleaner circuit tailings (PAG)
with the rougher tailings (bulk tailings in Figure 4-3) comprising predominantly NAG minerals. The PAG
tailings would need to be stored separately and kept isolated from oxygen.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . . o April 2013
-------
Chapter 4 Type of Development
BOX 4 5. CHEMICALS USED IN THE FLOTATION PROCESS
The froth flotation process is used to separate minerals from gangue (rock barren of target minerals), and to
separate one mineral from another. Reagents are added to a water-ore slurry to modify the surface of
particles either chemically or physically and facilitate separation. The amounts and types of reagents used in
flotation are site-specific, and depend on many factors such as particle size variation, particle density, ore
grade, and host rock character. Although some of these reagents can be transported to a mine site as
powder or pellets, most material arrives in liquid form.
The reagents used in flotation generally fall into three categories:
• Collectors (e.g., xanthates, dithiophosphates) increase the ability of air bubbles to stick to a particle.
• Frothers (e.g., methylisobutyl carbinol) increase the stability of air bubbles so they do not burst before
bringing a particle to the surface.
• Modifiers (e.g., carboxymethylcellulose) make collectors more effective by either activating or depressing
certain reactions.
The gold in porphyry copper deposits is partitioned among the copper-sulfide minerals (chalcopyrite,
bornite, chalcocite, digenite, and covellite), pyrite, and free gold (Kesler et al. 2002). Gold associated
with the copper minerals would stay with the copper (+gold) concentrate and be recovered at an off-site
smelter. Gold associated with pyrite would end up in the TSF unless a separate pyrite concentrate were
produced, and gold could be recovered from this concentrate by a vat leaching cyanidation process
(Logsdon et al. 1999, Marsden and House 2006). The solution that remains after this cyanidation
process could be passed through a cyanide destruction unit and either treated in a wastewater
treatment plant or stored in the TSF, where cyanide concentrations may decrease further through
natural attenuation (e.g., volatilization, photodegradation, biological oxidation, and precipitation)
(Logsdon et al. 1999). Tailings from this process, which have high concentrations of acid-generating
sulfides, typically are directed to the TSF, where they would be encapsulated in non-acid-generating
tailings and kept saturated to minimize oxidation.
Porphyry copper deposits (and other metal deposits) often have marketable quantities of metals other
than the primary target metals. These metals are carried through the flotation process and might be
removed at some later point. As an example, the Pebble deposit is reported to have marketable
quantities of silver, tellurium, rhenium, and palladium (Ghaffari etal. 2011), which are not sufficiently
concentrated in the ore to warrant separation and production of an additional metal concentrate.
The process for removing metals from ore is not 100% efficient. At some point the cost of recovering
more metals exceeds their value, so the amount of metals left in the tailings represents a tradeoff
between revenues from more complete ore processing and extraction costs. The process proposed by
Ghaffari etal. (2011) would recover 86.1% of the copper, 83.6 % of the molybdenum and 71.2% of the
gold from the Pebble deposit ore. The residual metals remaining with the tailings are discharged to a
TSF with the residue of blasting agents, flotation reagents, and inert portions of the ore.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . ,. April 2013
-------
Chapter 4
Type of Development
Figure 4 3. Simplified schematic of mined material processing.
Open Pit Mining
Waste Rock
Waste Rock Pile
Processing Facility
Product
Destination
Crushing Plant
Ball Mill
Milled Ore
Flotation Mill
Rougher Concentrate
Regrind Ball Mill
Milled Concentrate
Flotation Mill
BulkTailings
(non-acid-generating)
Tailings Storage
Facility
±
Copper (+gold)
Concentrate
Port Site
(via Pipeline)
Molybdenum
Concentrate
Port Site
(via Truck)
Pyritic Tailings
(potentially acid-generating)
Tailings Storage
Facility *
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-15
April 2013
-------
Chapter 4 Type of Development
4.2.3.4 Tailings Storage
Tailings are a mixture of fine-grained particles, water, and residue of reagents remaining from the
milling process. The most common method of tailings storage is disposal in an impoundment (i.e., a TSF)
(Porter and Bleiwas 2003). Tailings are transported from the mill to a TSF as a slurry, of which solids—
silt to fine sand particles (0.001 to 0.6 mm) with concentrations of metals too lowto interact with
flotation reagents—typically make up 30 to 50% by weight. Tailings may be thickened, or dewatered,
prior to disposal. Thickening reduces evaporation and seepage losses and allows recycling of more
process water back to the processing plant, thereby reducing operational water demand. It also
minimizes the amount of water stored in the TSF.
Tailings impoundments are water-holding structures typically built by creating a dam in a valley.
Tailings dams are generally earthen or rockfill dams constructed from waste rock or the coarse fraction
of the tailings themselves. The majority of existing tailings dams are less than 30 m in height, but the
largest exceed 150 m.
The engineering principles governing the design and stability of tailings dams are similar to the
geotechnical principles for earthen and rockfill dams used for water retention. They are typically built in
sections, called lifts, over the lifetime of the mine, such that dam height increases ahead of reservoir
level, using upstream, downstream, or centerline methods (Figure 4-4). Tailings dams built by the
upstream method are less stable against seismic events than dams built by either the downstream or the
centerline method (ICOLD 2001) because part of the dam rests on the tailings, which have a lower
density and a higher water saturation than the dam materials (USEPA 1994), and it is not possible to
compact them. Although upstream construction is considered unsuitable for impoundments intended to
be very high or to contain large volumes of water or solids (State of Idaho 1992), this method is still
employed (Davies 2002). An upstream dam lift was recently designed and constructed on the Fort Knox
Mine tailings impoundment dam near Fairbanks, Alaska (USAGE 2011). The downstream method is
considered more stable from a seismic standpoint, but it is more expensive to implement than the
upstream method. Centerline construction has characteristics of both upstream and downstream types
(USEPA 1994, Martin et al. 2002).
As they fill with tailings, TSFs must store immense quantities of water (Davies 2011). Water level is
controlled by removing excess water either for use in the mining process or for treatment and
subsequent discharge to local surface waters. Tailings are deposited against the embankment through
spigots or cyclones. Coarser-grained sands are directed at the embankment to create a beach, causing
water and fines to drain away from the dam to form a tailings pond. Care must be taken to prevent the
formation of low-permeability lenses or layers on tailings beaches, as these layers may perch water in
the TSF such that saturation of or flow through the dam may occur, leading to erosion or failure.
Although most of the tailings dam mass consists of fairly coarse and permeable material, the dams often
have a low permeability core to limit seepage, as well as internal drainage structures to collect seepage
water and to control pore pressures. Mitigation measures for seepage through or beneath a tailings dam
may include any combination of liners, seepage cutoff walls, under-drains, or decant systems.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 777April 2013
-------
Chapter 4
Type of Development
Figure 4 4. Cross sections illustrating (A) upstream, (B) downstream, and (C) centerline tailings
dam construction. In each case, the initial dike is illustrated in light gray, with subsequent dike raises
shown in darker shades (modified from Vick 1983). Tailings dams in our mine scenarios are assumed
to use initially the downstream construction method and at some point change to centerline
construction.
Tailings
B
Tailings
Tailings
Liners may cover the entire impoundment area (e.g., as proposed for the Donlin Creek Mine TSF in
Alaska), or only the pervious bedrock or porous soils. Full liners beneath TSFs are not always used and
may not be practicable for large impoundments; however, the use of liners to minimize risks of
groundwater contamination is increasingly required (Commonwealth of Australia 2007). Liners can
include a high-density polyethylene, bituminous, or other type of geosynthetic material (geomembrane)
and/or a clay cover over an area of higher hydraulic conductivity. A clay liner may have a saturated
hydraulic conductivity of 10-8 m/s, whereas a geomembrane may have a hydraulic conductivity of
approximately 10'10 m/s (Commonwealth of Australia 2007). However, geomembrane technology has
not been available long enough to know the service life of these liners. Laboratory tests and data from
landfills estimate that high-density polyethylene liner lifespans range from 69 to 600 years, depending
on whether it is the primary (upper) or secondary (lower or backup) liner (Rowe 2005, Koerner et al.
2011). In general, longer lifespans are expected at lower temperatures and exposures to light (Rowe
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
4-17
April 2013
-------
Chapter 4 Type of Development
2005, Koerner et al. 2011). Breakdown of the liner material and punctures by equipment or rocks may
limit the effective life of liners (Rowe 2005). Overly steep slopes also may put stresses on
geomembranes and cause them to fail. Service life data for other types of geomembranes are anecdotal
and based on field performance, since no laboratory studies have been conducted (Koerner et al. 2011).
If seepage is expected or observed, mitigation or remedial measures such as interception trenches or
seepage recovery wells can be installed around the perimeter and downstream of the TSF to capture
water and redirect it to a treatment facility. Precipitation runoff from catchment areas up-gradient of the
TSF is typically diverted away from the impoundment to reduce the volume of stored liquid.
Dry stack tailings management, in which tailings are filtered and "stacked" for long-term storage, is a
newer, less commonly used tailings disposal method. Dry stacked tailings require a smaller footprint,
are easier to reclaim, and have lower potential for structural failure and environmental impacts (Martin
etal. 2002). However, the high energy cost of dry stack technology remains a barrier for mining low-
grade ores such as porphyry copper. In addition, this type of storage is inappropriate for acid-generating
tailings and is less feasible in larger operations, where tailings impoundments serve to store water as
well as tailings. It is most applicable in arid regions, although dry stacks are also used in wet climates, or
in cold regions where water handling is difficult (Martin et al. 2002). Currently, the only mines in Alaska
that use dry stack disposal of tailings are underground mines with high-grade ore and relatively low
quantities of tailings (e.g., Greens Creek, a lead, silver, zinc mine in southeast Alaska; and Pogo, a gold
mine in eastern interior Alaska).
4.2.3.5 Waste Rock
Waste rock is rock overlying or removed with the ore body that contains uneconomic quantities of
metals. A waste-to-ore ratio of 2:1—that is, the removal of 2 tons of waste rock for each ton of ore—is
not uncommon for porphyry copper deposits (Porter and Bleiwas 2003). Waste rock is stored
separately from tailings (Blight 2010). Some waste rock that contains marketable minerals may be
stored such that it can be milled if commodity prices increase sufficiently or if higher than usual metal
concentrations in ore require dilution to optimize mill operation. However, the potential for
environmental impacts must be managed if the waste rock is PAG, via selective handling, drains,
diversion systems, or other means. PAG waste rock also may be blended with ore in the mill to maintain
a steady and predictable composition of feed material for the flotation process over time. NAG waste
rock may be placed in piles near the open pit, with ditches to divert stormwater around the piles and
drains (or other systems) to capture leachate or direct it toward the open pit. At closure, a dry cover
(e.g., encapsulation) can be placed over the waste rock pile to isolate it from water and oxygen, or the
pile could be placed into the completed open pit and kept below the water line if PAG material, with
choices dependent on site specifics (O'Kane and Wels 2003). With small pits and in some settings, it is
beneficial to fill the pit with waste rock and other waste material and then construct a dry cover over the
filled pit area.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . , „ April 2013
-------
Chapter 4 Type of Development
4.2.4 Timeframes
The mining process described above can be thought of in terms of three distinct periods:
• Operation refers to the period during which the mine is active—that is, the period when mine
infrastructure is being built and ore is being extracted and processed.
• Closure refers to the period following completion of mining operations (either as planned or
prematurely) when mining has ceased and activities related to reclamation and preparation of the
site for future stability continue. During this period, waste areas are reclaimed and facilities needed
to support ongoing monitoring and maintenance activities—such as stormwater management
ditches, monitoring wells, engineered covers on waste materials (if required), waste water treatment
plants, and roads—are created, retained from the operational period, or replaced or remediated if
they had become compromised.
• Post-closure refers to the extended period following closure activities when monitoring and
maintenance activities continue. During this time, water leaving the site is monitored and treated for
as long as contaminants were present at levels exceeding regulatory standards. The post-closure
phase may last decades, centuries, or longer, until only minimal oversight is required. Such minimal
oversight is necessary, perhaps into perpetuity, to ensure the structural integrity and minimal
environmental impact from the anthropogenic changes to the prior landscape. Given the limited
lifetime of human institutions, continued monitoring and maintenance of the site might become
increasingly unlikely as the time from mine closure increases.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . . p April 2013
-------
5.1 Overview of Assessment Endpoints
Selection of assessment endpoints is a key component of the problem formulation stage of an ecological
risk assessment. Each endpoint is an explicit expression of the environmental values of concern in the
assessment, in terms of both the entity valued (e.g., a species, community, or ecological process) and a
potentially at-risk characteristic or attribute of that entity (USEPA 1998). Endpoints can be defined at
any level of ecological organization, from within an organism to across ecosystems, depending on the
needs of the assessment. In all cases, however, selected endpoints should be relevant, to both ecology
and decision-maker needs, as well as susceptible to potential stressors (USEPA 1998).
We consider three endpoints in this assessment: (1) the abundance, productivity, or diversity of the
region's Pacific salmon and other fish populations; (2) the abundance, productivity, or diversity of the
region's wildlife populations; and (3) the viability of Alaska Native cultures. Each of these endpoints
meets the criteria of ecological relevance, management relevance, and potential susceptibility to
stressors associated with large-scale mining.
Given the economic, ecological, and cultural importance of Bristol Bay's fish resources, the assessment
focuses most heavily on Endpoint 1. Only Endpoint 1 is considered in terms of direct effects—that is, it is
the only endpoint for which we evaluate potential direct effects of large-scale mining (Section 2.2.1).
Most of the analyses center on Pacific salmon, rainbow trout, and Dolly Varden. This focus reflects the
ecological, economic, and cultural significance of these fish species, as well as data availability issues.
Other components of the region's aquatic ecosystems, including algae, aquatic invertebrates, and smaller
resident fishes such as sculpins, also may be affected by large-scale mining. However, these taxa are not
as relevant to decision makers and data on their distribution and abundance are more limited.
We evaluate Endpoints 2 and 3 indirectly, via the effects of large-scale mining on Endpoint 1. This focus
on indirect effects is not meant to suggest that mining would directly affect only fish populations, or that
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-1
April 2013
-------
Chapter 5 Endpoints
direct effects of mining on wildlife and Alaska Native populations would be inconsequential; rather, it
reflects the ecological and regulatory importance of the region's fisheries. Under Endpoint 2, we focus
on wildlife species that depend on salmon for food (e.g., brown bear, bald eagles, gray wolves,
waterfowl) or that are important subsistence foods for Alaska Natives (e.g., moose, caribou). Although
Alaska Natives are not the only people that would potentially be affected by mining in the region,
Endpoint 3 focuses on Alaska Native populations because of the centrality of salmon and other salmon-
dependent resources to their way of life and well-being, and because this assessment was initiated in
response to requests from federally recognized tribal governments to restrict large-scale mining in the
Swatersheds. We focus on the primary Alaska Native cultures of the Nushagak and Kvichak River
watersheds, the Yup'ik and Dena'ina. Within the greater Bristol Bay watershed region, there are
Aleut/Alutiiq people who traditionally lived along the Alaska Peninsula and who still live in the region;
however, because the Alaska Peninsula falls outside the Nushagak and Kvichak River watersheds, these
cultures were not included in the assessment. We also recognize that diverse non-Native people have
lived in the Bristol Bay region for hundreds of years, and also consider salmon an integral part of their
way of life. Further discussion of the scope of the assessment and how this scope was defined can be
found in Chapters 1 and 2.
In the following sections, we discuss each of the three assessment endpoints in greater detail. We
present information on the fish and wildlife species considered, including what is known about their life
histories, distributions, and abundances both across the Bristol Bay watershed (Scale 1) and within the
Nushagak and Kvichak River watersheds (Scale 2). We discuss the Alaska Native populations in the
region and examine why the region's salmon fisheries are an ecologically, economically, and culturally
important resource.
5.2 Endpoint 1: Salmon and Other Fishes
The Bristol Bay watershed is home to at least 29 species offish, representing at least nine different
families (Table 5-1). The region is renowned for its fish populations, and it supports world-class
fisheries for multiple species of Pacific salmon and other game fishes (Dye and Schwanke 2009). These
resources generate significant benefit for commercial fishers, support valued recreational fisheries
(Figure 5-1), and provide sustenance for Alaska Native populations (Figure 5-2, Box 5-1).
In this section we summarize key fish species found in the Bristol Bay watershed, their distributions and
abundances in the region, and some of the factors contributing to the significance of these resources.
This background information is provided to underscore the uniqueness of the region's fisheries and
support the assessment's focus on potential impacts of large-scale mining on these fisheries. More
detailed discussion of the region's fishes can be found in Appendices A and B.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g » APril 2013
-------
Chapter 5
Endpoints
Table 5 1. Fish species reported in the Nushagak and Kvichak River watersheds. (H) indicates
species considered harvested fish that is, they are well distributed across these drainages and are or
have been targeted by sport, subsistence, or commercial fisheries. List does not include primarily
marine species that periodically venture into the lower reaches of coastal streams. See Appendix B,
Table 1, for additional information on the abundance and life history of each species.
Family
Salmonids
(Salmonidae)
Lampreys
(Petromyzontidae)
Suckers
(Catostomidae)
Pikes
(Esocidae)
Mudminnows
(Umbridae)
Species
Bering Cisco
(Coregonaus laurettae)
Humpback whitefish (H)
(C. pidschian)
Least Cisco
(C. sardinella)
Pygmy whitefish
(Prosopium coulterii)
Round whitefish
(P. cylindraceum)
Coho salmon (H)
(Oncorhynchus kisutch)
Chinook salmon (H)
(0. tshawytscha)
Sockeye salmon (H)
(0. nerka)
Chum salmon (H)
(0. keta)
Pink salmon (H)
(0. gorbuscha)
Rainbow trout (H)
(0. mykiss)
Arctic char (H)
(Salvelinus alpinus)
Dolly Varden(H)
(S. ma/ma)
Lake trout (H)
(S. namaycush)
Arctic grayling (H)
(Thymallus arcticus)
Arctic lamprey
(Lethenteron camtschaticum)
Alaskan brook lamprey
(L. a/askense)
Pacific lamprey
(Entosphenus tridentatus)
Longnose sucker
(Catostomus catostomus)
Northern pike (H)
(Esox /L/C/US)
Alaska blackfish
(Da ///a pectoralis)
Relative Abundance
Very few specific reports
Common in large upland lakes; locally and seasonally common in large
rivers
Locally common in some lakes (e.g., Lake Clark, morainal lakes near
Iliamna Lake)
Locally common in a few upland lakes or adjacent streams
Abundant/widespread throughout larger streams in upland drainages;
not found in headwaters or coastal plain areas
Juveniles abundant and widespread in upland flowing waters of
Nushagak River watershed and in some Kvichak River tributaries
downstream of Iliamna Lake; uncommon upstream of Iliamna Lake
Juveniles abundant and widespread in upland flowing waters of
Nushagak River watershed and in some Kvichak River tributaries
downstream of Iliamna Lake; rare upstream of Iliamna Lake
Abundant
Abundant in upland flowing waters of Nushagak River watershed and in
some Kvichak River tributaries downstream of Iliamna Lake; rare
upstream of Iliamna Lake
Abundant (in even years), with restricted distribution, in the Nushagak
River watershed and in some Kvichak River tributaries downstream of
Iliamna Lake; rare upstream of Iliamna Lake
Frequent/common; in summer, closely associated with spawning
salmon
Locally common in upland lakes
Abundant in upland headwaters and selected lakes
Common in larger upland lakes; absent from the Wood River lakes
Abundant/widespread
Juveniles common/widespread in sluggish flows3
Rare
Common in slower areas of larger streams
Common/widespread in still or sluggish waters
Locally common/abundant in still or sluggish waters in flat terrain
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-3
April 2013
-------
Chapter 5
Endpoints
Table 5 1. Fish species reported in the Nushagak and Kvichak River watersheds. (H) indicates
species considered harvested fish that is, they are well distributed across these drainages and are or
have been targeted by sport, subsistence, or commercial fisheries. List does not include primarily
marine species that periodically venture into the lower reaches of coastal streams. See Appendix B,
Table 1, for additional information on the abundance and life history of each species.
Family
Smelts
(Osmeridae)
Cods
(Gadidae)
Sticklebacks
(Gasterosteidae)
Sculpins
(Cottidae)
Species
Rainbow smelt
(Osmerus mordax)
Pond smelt
(Hypomesus olidus)
Eulachon
(Thaleichthys pacificus)
Burbot
(Lota lota)
Threespine stickleback
(Gasterosteus aculeatus)
Ninespine stickleback
(Pungitius pungitius)
Coastrange sculpin
(Cottus aleuticus)
Slimy sculpin
(C. cognatus)
Relative Abundance
Seasonally abundant in streams near coast
Locally common in coastal lakes and rivers, Iliamna Lake, and inlet
spawning streams; abundance varies widely interannually
No or few specific reports; if present, distribution appears limited and
abundance low
Frequent/common in deep, sluggish, or still waters
Locally abundant in still or sluggish waters; abundant in Iliamna
Lake
Abundant/widespread in still or sluggish waters
Abundant/widespread15
Notes:
a These species are combined here, because juveniles, the most commonly encountered life stage for each, are indistinguishable.
b These species are combined here, because they are not reliably distinguished in field conditions, although slimy sculpin is thought to be more
abundant and widely distributed.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-4
April 2013
-------
Chapter 5
Endpoints
Figure 5 1. Approximate extents of popular Chinook and sockeye salmon recreational fisheries in
the vicinity of the Nushagak and Kvichak River watersheds. Areas were digitized from previously
published maps (Dye et al. 2006). Recreational rainbow trout fisheries are also distributed throughout
the watersheds.
Cook Inlet
Bristol Bay
N
A
25 50
25
] Kilometers
50
] Miles
f Approximate Pebble Deposit Location
• Towns and Villages
Watershed Boundary
Chinook Salmon
Sockeye Salmon
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-5
April 2013
-------
Chapter 5
Endpoints
Figure 5 2. Subsistence harvest and harvest effort areas for salmon and other fishes within the
Nushagak and Kvichak River watersheds. Other fishes are defined as those non salmon and
whitefish species discussed in the text. Each fish category is designated by a representative individual
color and includes all harvest points, lines, or polygons meeting that classification. See Box 5 1 for
more detailed discussion of methodology.
Cook Inlet
Clark's Point
South Naknek
Bristol Bay
N
A
25 50
25
] Kilometers
50
] Miles
w Approximate Pebble Deposit Location
° Nonsurveyed Towns and Villages
• Surveyed Towns and Villages
V^1 Other Fish Harvest Areas
\-;/ Salmon Harvest Areas
j Watershed Boundary
\_»/ Existing Roads
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-6
April 2013
-------
Chapter 5 Endpoints
BOX 5 1. SUBSISTENCE USE METHODOLOGY
Subsistence use and harvest data are extracted from data collected by the Alaska Department of Fish and
Game in collaboration with Stephen R. Braund and Associates (Fall etal. 2006, Kriegetal. 2009, Holen and
Lemons 2010, Holen et al. 2011, Holen et al. 2012). These data are a compilation of a multi-year study to
document and examine baseline subsistence use and harvest, along with demographic and economic data
within the communities near the Pebble deposit. Eighteen communities were interviewed: Aleknagik, Clark's
Point, Dillingham, Igiugig, Iliamna, King Salmon, Kokhanok, Koliganek, Levelock, Lime Village, Manokotak,
Naknek, NewStuyahok, Newhalen, Nondalton, Pedro Bay, Port Alsworth, and South Naknek.
Members of participating households within each community were asked to document where they hunted,
fished, and gathered subsistence resources during the previous year by adding points (used for harvest
locations), polygons (used for harvest effort areas), and lines (used to depict trap lines or courses travelled
during fish trolling) to various maps. Interviews were conducted from 2004 to 2011, although not every
community was interviewed in the same year so the reported years differed between communities. Following
completion of interviews, hand-drawn maps were digitized and data compiled for use within a geographic
information system. In this assessment, only towns and villages documenting subsistence use and harvest
within the Nushagak and Kvichak River watersheds were considered; data points or sections of polygons
and lines falling outside the boundary of these watersheds were omitted.
Subsistence use and harvest data were extracted for four representative use categories: salmon, other fish,
wildlife, and waterfowl, based on tables found within each report (e.g., Table 1-16 in Holen et al. 2012).
Species within each category are as follows:
• Salmon', chum salmon, Chinook (king) salmon, pink salmon, salmon, coho (silver) salmon, sockeye
salmon, and spawning sockeye (red) salmon
• Other Fish (i.e., non-salmon fish species and whitefishes)'. Arctic char, broad whitefish, Dolly Varden,
humpback whitefish, lake trout, least Cisco, rainbow trout, round whitefish, steelhead trout, trout, and
whitefish
• Wildlife', black bear, brown bear, caribou, and moose
• Waterfowi. black scoter, brant, Canada goose, eggs, geese, gull eggs, lesser snow goose, mallard, pintail,
sandhill crane, teal, tern eggs, tundra swan, waterfowl, and white-fronted goose
Data were extracted for all points, lines, and polygons in each category, for each interviewed community.
Data were then summed across all communities to produce a cumulative layer for the entire Nushagak and
Kvichak River watersheds. Subsistence intensity across the landscape was derived by first generating a 1-
km square grid across the extent of the Nushagak and Kvichak River watersheds. Each documented point,
line, and polygon shapefile was spatially joined and summed across the 1-km grid to account for multiple or
overlapping points, lines, and polygons within the same 1-km pixel. Therefore, each pixel represents the total
number of points and sections of lines and polygons within its boundaries. Subsistence use was then
summed across the four species categories to derive a total cumulative subsistence use across the
Nushagak and Kvichak River watersheds.
This subsistence use metric provides a coarse measure of areas on the landscape that are used for
subsistence uses more than others within the area. However, it is important to note some of the limitations
of the subsistence intensity metric. Points represent harvest locations, but the way these data are tabulated
does not confer abundance of species harvested within the pixel. Therefore, a point may represent either a
single capture or multiple captures of a given species. Although abundance information was collected by the
researchers, it was not consistently reported in the geospatial data. Further, the line and polygon files
represent general catch areas and not point of actual capture, allowing broad areas to have the same value
as an actual point of capture. Finally, since this assessment is focused on fish as the main assessment
endpoint, the intensity of subsistence uses is biased towards aquatic species and habitats. Many other plant
and animal species included in the subsistence use databases were not used to arrive at our metric of
subsistence intensity in this report.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment c -, April 2013
-------
Chapter 5 Endpoints
5.2.1 Species and Life Histories
5.2.1.1 Salmon
Five species of Pacific salmon spawn and rear in the Bristol Bay watershed's freshwater habitats:
sockeye or red (Oncorhynchus nerka], coho (0, kisutch), Chinook or king (0. tshawytscha), chum (0, ketd),
and pink [0, gorbuschd). Because no hatchery fish are raised or released in the watershed, Bristol Bay's
salmon populations are entirely wild.
All five salmon species share a trio of life-history traits, which contribute to the success and significance
of these species in the Bristol Bay region. First, they are anadromous: they hatch in freshwater habitats,
migrate to sea for a period of relatively rapid growth, and then return to freshwater habitats to spawn.
Second, the vast majority of adults return to their natal freshwater habitats to spawn. This homing
behavior fosters reproductive isolation, thereby enabling populations to adapt to the particular
environmental conditions of their natal habitats (Blair etal. 1993, Dittman and Quinn 1996, Eliason etal.
2011). Homing is notabsolute, however, and a small amount of straying increases the probability that
suitable habitats will be colonized by salmon (e.g., Milner and Bailey 1989). Finally, each species is
semelparous: adults die after spawning a single time. After completing their upstream migration,
females excavate nests (redds) in the gravel and release eggs into them. These eggs are fertilized by one
or more competing males as they are released, and the females bury them in the nests. The females and
males then die, depositing the nutrients incorporated into their body tissues in their spawning habitats
(Section 5.2.5).
The seasonality of spawning and incubation is roughly the same for all five species, although the timing
can vary somewhat by species, population, and region. In general, salmon spawn from summer through
fall, and the fry emerge from spawning gravel the following spring to summer. Freshwater habitats used
for spawning and rearing vary across and within species, and include headwater streams, larger
mainstem rivers, wetlands, and lakes (Table 5-2). With some exceptions, preferred spawning habitat
consists of gravel-bedded stream reaches of moderate water depth (30 to 60 cm) and current (30 to
100 cm/s) (Quinn 2005). Sockeye are unique among the species, in that most populations rely on lakes
as the primary freshwater rearing habitat (Table 5-2).
Both chum and pink salmon migrate to the ocean soon after fry emergence (Heard 1991, Salo 1991).
Because sockeye, coho, and Chinook salmon spend a year or more rearing in the Bristol Bay watershed's
streams, rivers, and lakes before their ocean migration (Table 5-2), these species are more dependent on
upstream freshwater resources than chum and pink salmon. As a result, potential large-scale mining in
this region likely poses greater risk to sockeye, coho, and Chinook salmon, and the assessment thus
focuses primarily on these three species of Pacific salmon.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g „ April 2013
-------
Chapter 5
Endpoints
Table 5 2. Life history, habitat characteristics, and total surveyed occupied stream length for
Bristol Bay's five Pacific salmon species in the Nushagak and Kvichak River watersheds.
Species
Sockeye
Coho
Chinook
Chum
Pink
Freshwater
Rearing Period
(years)
0-3
1-3
1+
0
0
Freshwater Rearing
Habitat
Lakes, rivers
Headwater streams to
moderate-sized rivers,
headwater springs, beaver
ponds, side channels,
sloughs
Headwater streams to
large-sized mainstem
rivers
Limited
Limited
Ocean
Feeding
Period (years)
2-3
1+
2-4
2-4
1+
Spawning Habitat
Beaches of lakes, streams
connected to lakes, larger
braided rivers
Headwater streams to
moderate sized rivers
Moderate-sized streams
to large-sized mainstem
rivers
Moderate-sized streams
and rivers
Moderate-sized streams
and rivers
Surveyed Stream
Length Occupied
(kilometers)
4,624
5,860
4,788
3,435
2,155
Notes:
Data compiled from Appendix A, pages 4-13.
5.2.1.2 Other Fishes
In addition to the five Pacific salmon species discussed above, the Bristol Bay region is home to more
than 20 species of resident fishes, or fishes that typically (but not always) remain within the watershed's
freshwater habitats throughout their life cycles. The region contains highly productive waters for such
sport and subsistence fish species as rainbow trout (0, mykiss), Dolly Varden [Salvelinus malmd), Arctic
char (S. alpinus), Arctic grayling (Thymallus arcticus), and lake trout (S. namaycush) (Dye and Schwanke
2009), as well as numerous other species that are nottypically harvested (Table 5-1). These fish species
occupy a variety of habitats throughout the watershed, from headwater streams to rivers and lakes.
In this assessment, we focus primarily on two resident fishes: rainbow trout and Dolly Varden; together
with the five Pacific salmon species, we refer to these fishes as salmonids (Box 2-2). This focus is not
meant to imply that other fish species found in the Bristol Bay watershed are not economically,
culturally, or ecologically important, or that they are unlikely to be affected by potential mining-related
activities. Rather, it reflects the value of rainbow trout and Dolly Varden as both sport and subsistence
fisheries throughout the region, the potential sensitivity of these species to mine development and
operation, and the relatively greater amount of information available for these species, particularly in
terms of distribution and abundance.
The species 0. mykiss includes both a non-anadromous or resident form (commonly referred to as
rainbow trout) and an anadromous form (commonly referred to as steelhead). In the Bristol Bay
watershed, steelhead generally are restricted to a few spawning streams near Port Moller, on the Alaska
Peninsula; thus, most populations throughout the region of the assessment are the non-anadromous
form.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-9
April 2013
-------
Chapter 5 Endpoints
The spawning habitat and behavior of rainbow trout are generally similar to that of the Pacific salmon
species, with a few key exceptions. First, rainbow trout are iteroparous, meaning that they can spawn
repeatedly. Second, spawning occurs in spring, versus summer and early fall for salmon. Juveniles
emerge from spawning gravels in summer (Johnson et al. 1994, ADF&G 2012), and immature fish may
remain in their natal streams for several years before migrating to other habitats (Russell 1977).
Rainbow trout in the Bristol Bay watershed exhibit complex migratory patterns, moving between
spawning, rearing, feeding, and overwintering habitats. For example, many adults in the region spawn in
inlet or outlet streams of large lakes, then migrate shortly after spawning to feeding areas within those
lakes; some mature fish may seasonally move distances of 200 km or more (Russell 1977, Burger and
Gwartney 1986, Minard et al. 1992, Meka et al. 2003). Often, these migratory patterns ensure that
rainbow trout are in close proximity to the eggs and carcasses of spawning salmon, which provide an
abundant, high-quality food resource (Meka etal. 2003). The variety of habitat types utilized by rainbow
trout is reflected by different life-history types identified in the region, including lake, lake-river, and
river residents (Meka et al. 2003). See Appendix B (pages 6-11) for additional information on rainbow
trout life history.
Dolly Varden is a highly plastic fish species, with multiple genetically, morphologically, and ecologically
distinct forms that can co-exist in the same water bodies (Ostberg et al. 2009). Both anadromous and
non-anadromous Dolly Varden are found in the Bristol Bay watershed, and both life-history forms can
exhibit complex and extensive migratory behavior (Armstrong and Morrow 1980, Reynolds 2000,
Scanlon 2000, Denton et al. 2009). Anadromous individuals usually undertake three to five ocean
migrations before reaching sexual maturity (DeCicco 1992, Lisac and Nelle 2000, Crane et al. 2003).
During these migrations, Dolly Varden frequently leave one drainage, travel through marine waters, and
enter a different, distant drainage (DeCicco 1992, DeCicco 1997, Lisac 2009). Non-anadromous
individuals also may move extensively between different habitats (Scanlon 2000).
Dolly Varden spawning occurs in fall, upstream of overwintering habitats (DeCicco 1992). Northern-
form anadromous Dolly Varden (the geographic form of Dolly Varden found north of the Alaska
Peninsula) overwinter primarily in lakes and in lower mainstem rivers where sufficient groundwater
provides suitable volume of free-flowing water (DeCicco 1997, Lisac 2009). Within the Nushagak and
Kvichak River watersheds, juveniles typically rear in low-order, high-gradient stream channels (ADF&G
2012). Because Dolly Varden occur in upland lakes and high-gradient headwater streams (ADF&G
2012)—farther upstream than many other fish species and above migratory barriers to anadromous
salmon populations—they may be especially vulnerable to mine development and operation in these
headwater areas. See Appendix B (pages 19-25) for additional information on Dolly Varden life history.
5.2.2 Distribution and Abundance
Fish populations throughout the Bristol Bay watershed have not been sampled comprehensively; thus
estimates of total distribution and abundance across the region are not available. However, available
data (e.g., the Anadromous Waters Catalog, escapement and harvest data) provide at least minimum
estimates of where key species are found and how many individuals of those species have been caught.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TT7 April 2013
-------
Chapter 5
Endpoints
More information on the distribution and abundance of key fish species can be found in Appendices A
and B; see Sections 7.1.1 and 7.2.5 for additional information on the interpretation of available fish
distribution data.
5.2.2.1 Salmon
Most (63%) of the subwatersheds in the Nushagak and Kvichak River watersheds are documented to
contain at least one species of spawning or rearing salmon within their boundaries, and 12 % are
documented to contain all five species (Figure 5-3). Reported distributions for each species in the
Nushagak and Kvichak River watersheds are shown in Figures 5-4 through 5-8.
Sockeye is by far the most abundant salmon species in the Bristol Bay watershed (Table 5-3) (Salomone
et al. 2011). Bristol Bay is home to the largest sockeye salmon fishery in the world, with 46% of the
average global abundance of wild sockeye salmon between 1956 and 2005 (Figure 5-9A) (Ruggerone et
al. 2010). Between 1990 and 2009, the average annual inshore run of sockeye salmon in Bristol Bay was
approximately 37.5 million fish (ranging from a low of 16.8 million in 2002 to a high of 60.7 million in
1995) (Salomone etal. 2011). Annual commercial harvest of sockeye over this period averaged
25.7 million fish (Table 5-3), and 78% of the average annual subsistence salmon harvest
(140,767 salmon) over this period were sockeye (Dye and Schwanke 2009, Salomone et al. 2011).
Escapement goals—that is, the number of individuals allowed to escape the fishery and spawn, to
ensure long-term sustainability of the stock—vary by species and stock. For example, the current
sockeye escapement goal for the Kvichak River ranged from 2 to 10 million fish (Box 5-2). Annual sport
harvest of sockeye in recentyears has ranged from approximately 8,000 to 23,000 fish (Dye and
Schwanke 2009).
Table 5 3. Mean annual commercial harvest (number of fish) by Pacific salmon species and Bristol
Bay fishing district, 1990 to 2009a.
Salmon
Species
Sockeye
Chinook
Coho
Chum
Pinkb
Bristol Bay Fishing District
Naknek-
Kvichak3
8,238,895
2,816
4,436
184,399
73,661
Egegik
8,835,094
849
27,433
78,183
1,489
Ugashik
2,664,738
1,402
10,425
70,240
138
Nushagak3
5,478,820
52,624
27,754
493,574
50,448
Togiak
514,970
8,803
14,234
158,879
43,446
Total
25,732,517
66,494
84,282
985,275
169,182
Notes:
a Naknek-Kvichak district includes the Alagnak River; Nushagak district includes the Wood and Igushik Rivers.
b Pink salmon data are from even-numbered years; harvest is negligible during odd-year runs.
Source: Appendix A, Table 1.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-11
April 2013
-------
Chapter 5
Endpoints
BOX 5 2. COMMERCIAL FISHERIES MANAGEMENT IN THE BRISTOL BAY WATERSHED
Commercial fisheries management in Alaska is largely focused on achieving escapement goals—management
goals based on the optimum range of fish numbers allowed to escape the fishery and spawn—rather than harvest
rates (Fair et al. 2012). Thus, management involves allowing an adequate number of spawners to reach each river
system while maximizing harvest in the commercial fishery (Salomone et al. 2011). Bristol Bay's commercial
salmon fisheries are considered a management success (Hilborn et al. 2003, Hilborn 2006). Several factors have
contributed to this success, including a clear management objective of maximum sustainable yield, the
escapement goal system, management responsibility fallingto a single agency, a permit system that limits the
number of fishers, and favorable freshwater habitats and ocean conditions (Hilborn et al. 2003, Hilborn 2006).
Escapement goals for sockeye salmon in the nine major rivers draining the Bristol Bay watershed are listed in the
table below. The Alaska Department of Fish and Game (ADF&G) regularly reviews escapement goals for the major
salmon stocks in Bristol Bay. These reviews include updates to escapement estimates, revisions to how catch is
partitioned to stocks, and revisions to stock-recruit models used to recommend escapement goals. For example,
data on sockeye genetic stock composition, age composition, and run timing were used to reconstruct brood
tables for the major stocks in 2012 (Cunningham et al. 2012, Fair et al. 2012).
The Kvichak River frequently did not meet its sockeye escapement goal from 1991 through 1999, and in 2001 it
was placed into special management status due to chronic low yields (Fair 2003). The cause of this low
productivity in Kvichak River sockeye is not entirely known, but marine conditions likely led to this decline (see
Appendix A, pages 30-32, for a more detailed discussion of this decline). However, the Kvichak River stock is
considered to be rebuilding: escapement goals have been met for the last 5 years, and in 2012 ADF&G
recommended that it be removed from special management status (Morstad and Brazil 2012).
Sockeye Salmon Escapement Goals in the Bristol Bay Watershed
River
Kvichak
Alagnak
Naknek
Egegik
Ugashik
Wood
Igushik
Nushagak-Mulchatna
Togiak
Escapement Range
(thousands offish)
2,000-10,000
320 minimum
800-1,400
800-1,400
500-1,200
700-1,500
150-300
370-840
120-270
Once escapement goals are set, the timing and duration of commercial fishery openings are adjusted throughout
the fishing season to ensure that escapement goals are met and any additional fish are harvested. Fishery
openings are based on information from a number of sources, including pre-season forecasts (expected returns of
the dominant age classes in a given river system, based on the number of spawning adults that produced each
age class); the test fishery at Port Moller on the Alaska Peninsula; early performance of the commercial fishery;
and in-river escapement monitoring. At the beginning of the fishing season, the frequency and duration of
openings are primarily based on pre-season forecasts and are managed conservatively. As the season progresses
and additional information becomes available, fishing times and areas are continuously adjusted via emergency
orders. If the escapement goal is exceeded at a given monitoring station, the fishery is opened longer and more
frequently. If the escapement goal is not reached, the fishery is closed.
This type of in-season management is also used to meet a Chinook salmon escapement goal for the Nushagak
River (55,000-120,000 fish). There is a chum salmon escapement goal for the Nushagak River (200,000 fish
minimum) and there are Chinook salmon escapement goals for the Alagnak and Naknek Rivers; however, in-
season management is not used to help attain these goals (Baker et al. 2009).
See Appendix A for a more detailed discussion of historical and current fisheries management in the Bristol Bay
region.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-12
April 2013
-------
Chapter 5 Endpoints
More than half of the Bristol Bay watershed's sockeye salmon harvest comes from the Nushagak and
Kvichak River watersheds (Figure 5-9B). Sockeye returns to the Kvichak River averaged 10.5 million fish
between 1963 and 2011, and this number climbs to 12.1 million fish when returns to the Alagnak River
are included (Cunningham et al. 2012). Kvichak River sockeye runs have exceeded 30 million fish three
times since 1956, with 48.6, 34.9, and 37.9 million fish in 1965,1970, and 1980, respectively
(Cunningham etal. 2012).
Tributaries to Iliamna Lake, Lake Clark, and the Wood-Tikchik Lakes are major sockeye spawning areas,
and juveniles rear in each of these lakes (Figure 5-4). Iliamna Lake provides the majority of sockeye
rearing habitat in the Kvichak River watershed, and historically has produced more sockeye than any
other lake in the Bristol Bay region (Fair et al. 2012). Riverine sockeye populations spawn and rear
throughout the Nushagak River watershed (Figure 5-4).
Chinook salmon spawn and rear throughout the Nushagak River watershed and in many tributaries of
the Kvichak River (Figure 5-5). Although Chinook is the least common salmon species across the Bristol
Bay region, the Nushagak River watershed supports a large Chinook salmon fishery and its commercial
and sport-fishing harvests are greater than those of all other Bristol Bay river systems combined
(Table 5-3). In addition, Chinook salmon are an important subsistence food for residents of the
Nushagak River watershed. Chinook returns to the Nushagak River are consistently greater than
100,000 fish per year, and have exceeded 200,000 fish per year in 11 years between 1966 and 2010.
This frequently places the Nushagak at or near the size of the world's largest Chinook runs, which is
notable given the Nushagak River's small watershed area compared to other Chinook-producing rivers
such as the Yukon, Kuskokwim, Fraser, and Columbia.
Coho salmon spawn and rear in many stream reaches throughout the Nushagak and lower Kvichak River
watersheds (Figure 5-6), and juveniles distribute widely into headwater streams, where they are often
the only salmon species present (Woody and O'Neal 2010, King et al. 2012). Production of juvenile coho
is often limited by the extent and quality of available wintering habitats (Nickelson et al. 1992, Solazzi et
al. 2000).
Chum salmon is the second most abundant salmon species in the Nushagak and Kvichak River
watersheds (Table 5-3). Both chum and pink salmon spawn throughout the Nushagak and lower
Kvichak River watersheds (Figures 5-7 and 5-8), but do not have an extensive freshwater rearing stage.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment c . o April 2013
-------
Chapter 5
Endpoints
Figure 5 3. Diversity of Pacific salmon species production in the Nushagak and Kvichak River
watersheds. Counts of species (sockeye, Chinook, coho, pink, and chum) spawning and rearing, based
on the Anadromous Waters Catalog (Johnson and Blanche 2012), are summed by 12 digit hydrologic
unit codes. See Section 7.2.5 for details on interpretation of distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location Number of Species Documented
Towns and Villages
Mine Scenario Watersheds
Watershed Boundary
None
1
2
3
4
5
IN
A
25 50
25
J Kilometers
50
]Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-14
April 2013
-------
Chapter 5
Endpoints
Figure 5 4. Reported sockeye salmon stream distribution in the Nushagak and Kvichak River
watersheds. "Present" indicates species was present but life stage use was not determined;
"spawning" indicates spawning adults were observed; "rearing" indicates juveniles were observed.
Present, spawning, and rearing designations are based on the Anadromous Waters Catalog (Johnson
and Blanche 2012). Life stage specific reach designations are likely underestimates, given the
challenges inherent in surveying all streams that may support life stage use throughout the year. See
Section 7.2.5 for details on interpretation of fish distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Present
Spawning
Rearing
Mine Scenario Watersheds
Watershed Boundary
N
A
25
25
50
] Kilometers
50
]Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-15
April 2013
-------
Chapter 5
Endpoints
Figure 5 5. Reported Chinook salmon distribution in the Nushagakand Kvichak River watersheds.
"Present" indicates species was present but life stage use was not determined; "spawning" indicates
spawning adults were observed; "rearing" indicates juveniles were observed. Present, spawning, and
rearing designations are based on the Anadromous Waters Catalog (Johnson and Blanche 2012). Life
stage specific reach designations are likely underestimates, given the challenges inherent in
surveying all streams that may support life stage use throughout the year. See Section 7.2.5 for
details on interpretation offish distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Present
Spawning
Rearing
Mine Scenario Watersheds
Watershed Boundary
IN
A
25
25
50
] Kilometers
50
]Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-16
April 2013
-------
Chapter 5
Endpoints
Figure 5 6. Reported coho salmon distribution in the Nushagak and Kvichak River watersheds.
"Present" indicates species was present but life stage use was not determined; "spawning" indicates
spawning adults were observed; "rearing" indicates juveniles were observed. Present, spawning, and
rearing designations are based on the Anadromous Waters Catalog (Johnson and Blanche 2012). Life
stage specific reach designations are likely underestimates, given the challenges inherent in
surveying all streams that may support life stage use throughout the year. See Section 7.2.5 for
details on interpretation offish distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Present
Spawning
Rearing
Mine Scenario Watersheds
Watershed Boundary
N
A
25
25
50
] Kilometers
50
]Mi!es
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-17
April 2013
-------
Chapter 5
Endpoints
Figure 5 7. Reported chum salmon distribution in the Nushagak and Kvichak River watersheds.
"Present" indicates species was present but life stage use was not determined; "spawning" indicates
spawning adults were observed; "rearing" indicates juveniles were observed. Present, spawning, and
rearing designations are based on the Anadromous Waters Catalog (Johnson and Blanche 2012). Life
stage specific reach designations are likely underestimates, given the challenges inherent in
surveying all streams that may support life stage use throughout the year. See Section 7.2.5 for
details on interpretation offish distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Present
Spawning
Rearing
Mine Scenario Watersheds
Watershed Boundary
IN
A
25
25
50
] Kilometers
50
]Mlles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-18
April 2013
-------
Chapter 5
Endpoints
Figure 5 8. Reported pink salmon distribution in the Nushagak and Kvichak River watersheds.
"Present" indicates species was present but life stage use was not determined; "spawning" indicates
spawning adults were observed. Present and spawning designations are based on the Anadromous
Waters Catalog (Johnson and Blanche 2012). Life stage specific reach designations are likely
underestimates, given the challenges inherent in surveying all streams that may support life stage
use throughout the year. See Section 7.2.5 for details on interpretation of distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Present
Spawning
Mine Scenario Watersheds
Watershed Boundary
N
A
25 50
25
] Kilometers
50
DMiles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-19
April 2013
-------
Chapter 5
Endpoints
Figure 5 9. Total sockeye salmon run sizes by (A) region and (B) watershed within the Bristol Bay
region. Values are averages from 1956 2005 and 1956 2010 for A and B, respectively (Tables A2
and A3 in Appendix A).
• Bristol Bay
• Russia Mainland & Islands
• West Kamchatka
• East Kamchatka
• Western Alaska (excluding Bristol Bay)
• South Alaska Peninsula
• Kodiak
• Cook In let
Prince WilliamSound
Southeast Alaska
North British Columbia
South British Columbia, Washington & Oregon
• Togiak
• Nushagak
Kvichak
Naknek
Egegik
Ushagik
5.2.2.2 Other Fishes
Extensive sampling for rainbow trout and Dolly Varden has not been conducted throughout the Bristol
Bay region, so total distribution and abundance are unknown. Figures 5-10 and 5-11 show the reported
occurrence and distribution of rainbow trout and Dolly Varden throughout the Nushagak and Kvichak
River watersheds, and provide a minimum estimate of their extents.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-20
April 2013
-------
Chapter 5
Endpoints
Figure 5 10. Reported rainbow trout occurrence and distribution in the Nushagak and Kvichak River
watersheds. Designation of species presence is based on the Alaska Freshwater Fish Inventory (AFFI
point data, ADF&G 2012). Note that points shown on land actually occur in smaller streams not
shown on this map. Absence cannot be inferred from this map. See Section 7.2.5 for details on
interpretation offish distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Present (AFFI)
Mine Scenario Watersheds
Watershed Boundary
IN
A
25 50
25
] Kilometers
50
]Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-21
April 2013
-------
Chapter 5
Endpoints
Figure 5 11. Reported Dolly Varden occurrence and distribution in the Nushagak and Kvichak River
watersheds. Designation of species presence is based on the Alaska Freshwater Fish Inventory (AFFI
point data, ADF&G 2012) and the Anadromous Waters Catalog (AWC line data, Johnson and Blanche
2012). Note that points shown on land actually occur in smaller streams not shown on this map.
Absence cannot be inferred from this map. See Section 7.2.5 for details on interpretation of fish
distribution data.
Cook Inlet
Bristol Bay
Approximate Pebble Deposit Location
Present (AFFI)
Present (AWC)
Mine Scenario Watersheds
Watershed Boundary
N
A
25 50
25
] Kilometers
50
]Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-22
April 2013
-------
Chapter 5
Endpoints
Between 2003 and 2007, an estimated 183,000 rainbow trout were caught in the Bristol Bay
Management Area (Dye and Schwanke 2009). Radio telemetry, tagging, and genetic studies indicate that
multiple rainbow trout populations are found within Bristol Bay watersheds (Gwartney 1985, Burger
and Gwartney 1986, Minard etal. 1992, Krueger etal. 1999, Meka etal. 2003).The most popular
rainbow trout fisheries are found in the Kvichak River watershed, the Naknek River drainage, portions
of the Nushagak and Mulchatna River watersheds, and streams of the Wood River lakes system (Dye and
Schwanke 2009).
Dolly Varden populations are a significant subsistence resource. In the mid-2000s, subsistence harvests
of Dolly Varden and Arctic char combined (Alaska's fisheries statistics do not distinguish between the
two species) were estimated at 3,450 fish for 10 communities in the Nushagak and Kvichak River
watersheds (Fall et al. 2006, Krieg et al. 2009). From the mid-1970s to the mid-2000s, these two species
were estimated to represent between 16.2 and 26.9% of the total weight of the Kvichak River
watershed's non-salmon freshwater fish subsistence harvest (Krieg et al. 2005). Dolly Varden also
supports a popular sport fishery.
5.2.3 Economic Implications
The Bristol Bay watershed supports several sustainable, wilderness-compatible economic sectors,
including commercial fishing, sport fishing, subsistence hunting and fishing, recreational hunting, and
wildlife viewing and other non-consumptive recreation. Each of these sectors generates expenditures or
sales that drive the region's economy, generating more than $479 million (in 2009 dollars) in total direct
annual economic benefit (Table 5-4).
Table 5 4. Summary of regional economic expenditures based on salmon ecosystem services.
Values are regional expenditures in different economic sectors, expressed in 2009 dollars. Note that
estimates of certain year specific total harvest and sales values vary slightly throughout this report,
due to differences in how data were aggregated and reported. See Appendix E for additional
information on these values.
Economic Sector
Commercial fisheries, wholesale value
Sport fisheries
Sport hunting
Wildlife viewing/ tourism
Subsistence harvest
TOTAL
Estimated Direct Expenditure
(sales per year, in $ millions)
300.2
60.5
8.2
104.4
6.3
479.6
Roughly 75% of this annual economic benefit results directly from the commercial, sport, and
subsistence fishing supported by the Bristol Bay watershed. The commercial salmon fishery currently
provides the region's greatest source of economic activity. From 2000 through 2010, the annual
commercial salmon catch averaged 23 million fish (170 million pounds). The average annual
commercial value of all Bristol Bay salmon fisheries from 1990 to 2010 totaled $116.7 million,
$114.7 million of which resulted from the sockeye harvest (Salomone et al. 2011). Thus, sockeye salmon
represent the principal species of economic value throughout the Bristol Bay region.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-23
April 2013
-------
Chapter 5 Endpoints
In 2009, fishers received $144 million for their catch, and fish processors received approximately
$300 million, which is referred to as the first wholesale value of the fish (Table 5-4, Appendix E). The
commercial salmon fishery, which is largely centered in the region's salt waters rather than its
freshwater streams and rivers, is closely managed for sustainability using a permit system (Box 5-2).
Approximately 26% of permit holders are Bristol Bay residents. The commercial fishery also provides
significant employment opportunities, directly employing over 11,000 full- and part-time workers at the
season's peak.
The uncrowded, pristine wilderness setting of the Bristol Bay watershed attracts recreational fishers,
and aesthetic qualities are rated as most important in selecting fishing locations by Bristol Bay anglers.
Sport fishing in Bristol Bay accounts for approximately $60.5 million in annual spending (Table 5-4),
$58 million of which is spent in the Bristol Bay region. In 2009, approximately 29,000 sport-fishing trips
were taken to the Bristol Bay region (12,000 trips by people living outside of Alaska, 4,000 trips by
Alaskans living outside the Bristol Bay area, and 13,000 trips by Bristol Bay residents). These sport
fishing activities directly employ over 800 full- and part-time workers. In 2010, 72 businesses and
319 guides were operating in the Nushagak and Kvichak River watersheds alone, down from a peak of
92 businesses and 426 guides in 2008 (Appendix A, Table 4).
Many households participate in the subsistence harvest offish, which generates regional economic
benefits when Alaskan households spend money on subsistence-related supplies. In total, individuals in
Bristol Bay communities harvest about 2.6 million pounds of subsistence harvest per year. In 2010, the
U.S. Census Bureau reported an estimated 1,873 Alaska Native and 666 non-native households in the
Bristol Bay region. Goldsmith et al. (1998) estimated that Alaska Native households spend an average of
$3,054 on subsistence harvest supplies, whereas non-native households spend an estimated $796 on
supplies (values updated to 2009 price levels). Based on these estimates, subsistence harvest activities
resulted in expenditures of approximately $6.3 million (Table 5-4). It is important to note that these
estimates of expenditures reflect only the annual economic activity generated by these activities. It may
be useful to consider calculations such as net economic value, or the value of the resource or activity
over and above regular expenditures associated with it. These types of calculations, as well as the
regional economic significance of Bristol Bay's salmon fishery, are discussed in Appendix E.
5.2.4 Biological Complexity and the Portfolio Effect
As the previous sections illustrate, the Bristol Bay watershed supports world-class salmon fisheries.
These fisheries result from numerous, interrelated factors (Chapter 3). Closely tied to the Bristol Bay
region's physical habitat complexity (Chapter 3) is its biological complexity, which greatly increases the
region's ecological productivity and stability. This biological complexity operates at multiple scales and
across multiple species, but it is especially evident in the watershed's Pacific salmon populations. As
discussed in Section 5.2.1.1, the five Pacific salmon species found in Bristol Bay vary in many life-history
characteristics (Table 5-5), allowing them to fully exploit the range of habitats available throughout the
watershed. Even within a single species, life histories can vary significantly. For example, sockeye
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment T7T April 2013
-------
Chapter 5
Endpoints
salmon may spend anywhere from 0 to 3 years rearing in freshwater habitats, then 1 to 4 years feeding
at sea, before returning to the Bristol Bay watershed anytime within a 4-month window (Table 5-5).
Table 5 5. Life history variation within the Bristol Bay sockeye salmon populations.
Element of Biological Complexity
Location within the Bristol Bay watershed
Time of adult return to freshwater habitats
Time of spawning
Spawning habitat
Body size of adults
Body shape of adults
Egg size
Time between entry into spawning habitat and death
Time spent rearing in freshwater
Time spent at sea
Range of Traits or Options
7 major subwatersheds, ranging from maritime-influenced systems on
the Alaska Peninsula to more continental systems
June-September
July-November
Major rivers, small streams, spring-fed ponds, mainland beaches,
island beaches
130 to 190-mm body depth at 450-mm male length
Sleek, fusiform to very deep-bodied, with exaggerated humps and jaws
88-116 mg at 450 mm female length
Days-weeks
0-3 years
1-4 years
Source: Hilbornetal. 2003.
This life-history variability, together with the Pacific salmon's homing behavior, results in distinct
populations adapted to their own specific spawning and rearing habitats (Hilborn et al. 2003).
Variations in temperature and streamflow associated with seasonality and groundwater-surface water
interactions create a habitat mosaic supporting a range of spawning times across the watersheds.
Spawning adults return at different times, to different locations, creating and maintaining a degree of
reproductive isolation and allowing development of genetically distinct stocks (Hilborn etal. 2003,
McGlauflin et al. 2011). Thus, Bristol Bay watershed's sockeye salmon "population" is actually a sockeye
salmon stock complex, or a combination of hundreds of genetically distinct populations, each adapted to
specific,localized environmental conditions (Hilbornetal. 2003, Schindler etal. 2010). This stock
complex structure can be likened to a financial portfolio in which assets are divided among diverse
investments to increase financial stability. Essentially, it creates a biological portfolio effect (Schindler et
al. 2010), stabilizing salmon productivity across the watershed as a whole as the relative contribution of
sockeye with different life-history characteristics, from different regions of the Bristol Bay watershed,
changes over time in response to changes in environmental conditions (Hilborn et al. 2003). For
example, salmon stocks that spawn in small streams may be negatively affected by low-flow conditions,
whereas stocks that spawn in lakes may not be affected (Hilborn etal. 2003). Thus, any population
containing stocks that vary in spawning habitat is better able to persist as environmental conditions
change.
Without this high level of system-wide biological complexity, annual variability in the size of Bristol
Bay's sockeye salmon runs would be expected to more than double and fishery closures would be
expected to become more frequent (Schindler et al. 2010). In other watersheds with previously robust
salmon fisheries, such as the Sacramento River's Chinook fishery, losses of biological complexity have
contributed to salmon population declines (Lindley et al. 2009). These findings suggest that even the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-25
April 2013
-------
Chapter 5 Endpoints
loss of a small stock within an entire watershed's salmon population may have more significant effects
than expected, due to associated decreases in biological complexity of the population's stock complex.
5.2.5 Salmon and Marine-Derived Nutrients
Most of the nitrogen, phosphorus, and other elements in adult salmon bodies are derived from the
marine environment (Larkin and Slaney 1997, Schindler et al. 2005). Adult salmon returning to their
natal freshwater habitats import nutrients that they obtained during their ocean feeding period—that is,
marine-derived nutrients (MDN)—back into those habitats (Cederholm et al. 1999, Gende et al. 2002).
Approximately 95 to 99% of the carbon, nitrogen, and phosphorus in an adult salmon's body is derived
from the marine environment (Larkin and Slaney 1997, Schindler et al. 2005), and MDN from salmon
accounts for a significant portion of nutrient budgets in the Bristol Bay watershed (Kline et al. 1993). For
example, sockeye salmon are estimated to import approximately 12,700 kg of phosphorus and
101,000 kg of nitrogen into the Wood River system annually, and 50,200 kg of phosphorus and
397,000 kg of nitrogen into the Kvichak River system annually (Moore and Schindler 2004).
Productivity of the Bristol Bay region's fish and wildlife species is highly dependent on this influx of
MDN into the region's freshwater habitats. When available, salmon-derived resources in the form of live
adult salmon, eggs, carcasses, and invertebrates that feed upon carcasses are important dietary
components for numerous animal species, including fishes (e.g., rainbow trout, Dolly Varden, Pacific
salmon, Arctic grayling), mammals (brown bears, wolves, foxes, minks), and birds (bald eagles,
waterfowl) (Brna and Verbrugge 2013; this document was originally published as Appendix C of this
assessment, but has since been released as a U.S. Fish and Wildlife Service [USFWS] report). Availability
and consumption of salmon-derived resources can have significant benefits for these species, including
increased growth rates, energy storage, litter size, nesting success, and population density (Appendix A,
Brna and Verbrugge 2013). The abundance of trophy-sized rainbow trout in the Bristol Bay system
likely results from MDN imported by spawning salmon. Terrestrial systems of the Bristol Bay watershed
also benefit from these MDN (Cederholm etal. 1999, Gende etal. 2002). Bears, wolves, and other wildlife
transport carcasses and excrete wastes throughout their ranges (Darimont et al. 2003, Helfield and
Naiman 2006), which provide food and nutrients for other terrestrial species.
Finally, by dying in the streams where they spawn, adult salmon subsidize the next generation by adding
their nutrients to the ecosystem that will feed their young. This positive feedback is missing from
freshwater systems with depleted salmon runs, which probably inhibits attempts to renew those runs
(Gresh etal. 2000).
Eggs from spawning salmon are a major food source for Bristol Bay rainbow trout and are likely
responsible for much of the growth attained by these fish. Upon arrival of spawning salmon in the Wood
River basin, rainbow trout shifted from consuming aquatic insects to primarily salmon eggs for a five-
fold increase in ration and energy intake (Scheuerell et al. 2007). With this rate of intake, a bioenergetics
model predicts a 100-g trout to gain 83 gin 76 days; without the salmon-derived subsidy, the same fish
was predicted to lose 5 g (Scheuerell et al. 2007). Rainbow trout in Lower Talarik Creek were
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment T77 April 2013
-------
Chapter 5 Endpoints
significantly fatter (i.e., had a higher condition factor) in years with high spawner abundance of salmon
than in years with low abundance (Russell 1977).
5.2.6 Bristol Bay Fisheries in the Global Context
The Bristol Bay region is a unique environment supporting world-class fisheries, particularly in terms of
Pacific salmon populations. The region takes on even greater significance when one considers the status
and condition of Pacific salmon populations throughout their native geographic distributions. These
declines are discussed briefly below; for additional information on threatened and endangered salmon
stocks, see Appendix A (pages 36-38).
Although it is difficult to quantify the true number of extinct Pacific salmon populations around the
North Pacific, estimates for the western United States (California, Oregon, Washington, and Idaho) range
from 106 to 406 populations (Nehlsen etal. 1991, Augerot 2005, Gustafson etal. 2007). Pacific salmon
are no longer found in 40% of their historical breeding ranges in the western United States, and where
populations remain, they tend to be significantly reduced or dominated by hatchery fish (NRC 1996).
For example, 214 salmon and steelhead stocks were identified as facing risk of extinction in the western
United States; 76 of those stocks were from the Columbia River basin alone (Nehlsen et al. 1991). In
general, these losses have resulted from cumulative effects of habitat loss, water quality degradation,
climate change, overfishing, dams, and other factors (NRC 1996, Schindler et al. 2010). Species with
extended freshwater rearing periods—that is, species like sockeye, which dominate salmon production
in the Bristol Bay watershed—are more likely to be extinct, endangered, or threatened than species
which spend less time in freshwater habitats (NRC 1996). No Pacific salmon populations from Alaska
are known to have gone extinct, although many show signs of population declines.
The status of Pacific salmon throughout the United States highlights the value of the Bristol Bay
watershed as a salmon sanctuary or refuge (Rahr et al. 1998, Pinsky et al. 2009). The Bristol Bay
watershed contains intact, connected habitats that extend from headwaters to ocean with minimal
influence of human development. These characteristics, combined with the region's high Pacific salmon
abundance and life-history diversity, make the Bristol Bay watershed a significant resource of global
conservation value (Pinsky et al. 2009).
5.3 Endpoint 2: Wildlife
Unlike most terrestrial ecosystems, the Bristol Bay watershed has undergone little development and
remains largely intact. Thus, it still supports its historical complement of species, including large
carnivores such as brown bears (Ursus arctos), bald eagles (Haliaeetus leucocephalus], and gray wolves
[Canis lupus); ungulates such as moose [Alces alces gigas) and caribou [Rangifer tarandusgranti); and
numerous waterfowl species. Wildlife populations tend to be relatively large in the region, due to the
increased productivity associated with Pacific salmon runs (Section 5.2.5). MDN provide a foundational
element for the food web in these watersheds and are important for many species of wildlife. Wildlife, in
turn, distribute these nutrients from the aquatic to the terrestrial environment, cycling them through
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment T7Z April 2013
-------
Chapter 5 Endpoints
the entire ecosystem. Thus, interactions between salmon and other wildlife species are complex and
reciprocal.
In this section we summarize key wildlife species in the Nushagak and Kvichak River watersheds, with
particular focus on how these species are related to salmon resources. The species selected for
characterization—brown bear, moose, barren ground caribou, wolf, bald eagle, waterfowl (as a guild),
shorebirds (as a guild), and land birds (as a guild)—are important to ecosystem function, have a direct
link to salmon, and/or are important to Alaska Native and non-Native residents. There are no known
breeding or otherwise significant occurrence within the Nushagak and Kvichak River watersheds of any
species listed by USFWS as being threatened or endangered under the Endangered Species Act, and
there is no designated critical habitat within these watersheds. For additional information on wildlife
species, readers should consult Brna and Verbrugge (2013). In many cases, little abundance data specific
to the Bristol Bay watershed are available, but it is reasonable to assume that species distribution and
abundance patterns in this region mirror those observed in similar habitats across southwestern Alaska.
Although this assessment focuses on inland aquatic and nearshore habitats of the Bristol Bay watershed,
it should be noted that once the region's Pacific salmon populations migrate to the ocean, they also
provide food for marine predators (Appendix F). Marine mammals such as northern fur seals, harbor
seals, stellar sea lions, orcas and beluga whales are known to feed on Pacific salmon. These interactions
also can be important in freshwater habitats, as one of two freshwater harbor seal populations in North
America is found in Iliamna Lake (Smith et al. 1996).
5.3.1 Life Histories, Distributions, and Abundances of Species
5.3.1.1 Brown Bears
Brown bears are wide-ranging and use many different plant and animal communities for food. They
typically spend July through mid-September near streams supporting salmon runs, then move to higher
elevations in the fall to feed on berries and other food items before denning in October/November. They
emerge in spring and feed on vegetation and carrion, as well as moose and caribou calves. Because of
their wide-ranging behavior, they distribute MDNs both through deposition of salmon carcasses and
through excretion of wastes throughout their ranges.
Brown bear density estimates range from roughly 40 bears per 1,000 km2 in the northern Bristol Bay
region (Togiak National Wildlife Refuge and the Bureau of Land Management's Goodnews Block) (Walsh
etal. 2010) to 150 bears per 1,000 km2 along the shore of Lake Clark. From July 2006 to July 2007,
621 brown bears were reported harvested from the Alaska Department of Fish and Game's (ADF&G's)
Game Management Unit (GMU) 9, which includes the Kvichak River watershed and the Alaska Peninsula.
Brown bears are not as abundant in the Nushagak River watershed as the Kvichak River watershed, and
densities in both watersheds are lower than on the Alaska Peninsula's Pacific coast, which is home to the
highest documented brown bear density in North America (551 bears per 1,000 km2) (Miller et al.
1997). Brown bears are common in the area surrounding the Pebble deposit, with a 2009 estimated
density of 18.4 to 22.5 per 1,000 km2 (PLP 2011).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~April 2013
-------
Chapter 5 Endpoints
5.3.1.2 Moose
Moose habitat is determined by forage opportunities and includes both aquatic and upland areas.
Alluvial habitats along the Nushagak and Mulchatna Rivers, where willows and other plants regenerate
after scouring and subsequent deposit of river silt, support an abundant moose population. High-quality
summer forage, especially near wetlands, is important for nursing cows and calves. It is likely that MDN
contribute to increased plant productivity in these alluvial areas (Cederholm et al. 1999, Gende et al.
2002).
Moose abundance in the Nushagak and Kvichak River watersheds was estimated at 8,100 to 9,500 in
2004 (Butler 2004, Woolington 2004). Populations are especially high in the Nushagak River watershed
(ADF&G 2011), where felt-leaf willow, a preferred plant species, is abundant (Bartz and Naiman 2005).
Moose were considered "low density" (0.04 moose/km2) in the immediate area of the Pebble deposit
and the transportation corridor, but there is a large variance around this estimate (PLP 2011).
5.3.1.3 Caribou
Caribou feed in open tundra, mountain, and sparsely forested areas and can travel for long distances.
The Nushagak and Kvichak River watersheds are primarily used by caribou from the Mulchatna herd,
one of 31 caribou herds found in Alaska. The Mulchatna herd ranges widely through the Nushagak and
Kvichak River watersheds, but also spends considerable time in other watersheds. It numbered roughly
200,000 in 1997 but had decreased to roughly 30,000 by 2008 (Valkenburg et al. 2003, Woolington
2009). Recent surveys reported only a few caribou near the Pebble deposit area and potential
transportation corridor (PLP 2011). However, caribou populations and ranges in the Bristol Bay region
fluctuate significantly over time, and in previous years the herd was much larger and there was higher-
density use of the Pebble deposit area (PLP 2011), The Barren-ground caribou on the North Slope of
Alaska have demonstrated avoidance of exploration activities (Fancy 1983), and some tribal Elders in
the Nushagak and Kvichak River watersheds believe that mining exploration has contributed to
avoidance of the Pebble deposit area (Brna and Verbrugge 2013).
5.3.1.4 Gray Wolf
Gray wolf abundance is influenced by prey abundance and availability, but populations are primarily
limited by mortality caused by humans. Wolves have flexible diets and can shift to non-ungulate prey
species when ungulate prey are scarce, or take advantage of seasonally abundant species such as
salmon. Wolves often transport salmon away from streams for consumption or to feed pups through
regurgitation.
Gray wolf populations have not been well-studied in the Bristol Bay region, and it is difficult to assess
population numbers. Wolves are currently thought to be abundant in the Nushagak River watershed;
between 2003 and 2008, reported annual wolf harvest ranged from 60 to 141 in GMU 17, which includes
the Nushagak and Togiak River watersheds. In the Kvichak River watershed, numbers are believed to be
lower, although populations have increased since the 1990s (Butler 2009).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~ April 2013
-------
Chapter 5 Endpoints
5.3.1.5 Bald Eagle
Bald eagles generally nest near riparian and beach areas and eat primarily fish, although they have a
variable diet. Nesting bald eagles rely on salmon resources (Hansen 1987), and inland bald eagles
nesting near spawning streams have higher nesting success than those with more distant nests (Gerrard
et al. 1975). Birds and other fish are also important prey for bald eagles. Salmon abundance in the
Nushagak and Kvichak River watersheds affects bald eagle abundance, distribution, breeding, and
behavior. Bald eagles, in turn, distribute MDNs in their excretions.
Although no comprehensive survey of bald eagles or bald eagle nests has been conducted in the Bristol
Bay watershed, limited count data are available for parts of the region. For example, 50 bald eagle nests
were recorded along portions of the Nushagak, Mulchatna, and Kvichak Rivers in 2006 (Brna and
Verbrugge 2013); approximately half of those nests were categorized as active. The USFWS Bald Eagle
Nest Database contains approximately 230 nest records for the Nushagak and Kvichak River
watersheds, with 169 of those records collected between 2003 and 2006. Raptor studies in the Pebble
deposit area indicate that bald eagles were the most abundant nesting raptor (30% of all raptor nests in
2005) (PLP 2011).
5.3.1.6 Waterfowl
More than 30 species of waterfowl, including ducks (e.g., northern pintail, scaup, mallard, and green-
winged teal), geese (e.g., white-fronted, Canada), swans, and sandhill cranes, regularly use the Bristol
Bay region (PLP 2011). Diversity of habitat and extent of wetlands and waters provide habitat for
migrants and wintering waterfowl, and the region is an important staging area for many species,
including emperor geese, Pacific brant, and ducks, during spring and fall migrations.
The Alaska Yukon Waterfowl Breeding Population Survey found average late-May abundance indices of
497,000 ducks, 7,700 geese, 15,400 swans, and 5,300 sandhill cranes in the Bristol Bay Lowlands
between 2002 and 2011 (Brna and Verbrugge 2013). Salmon are used by some waterfowl as direct
sources of prey and carrion, and used indirectly through invertebrates and vegetation. Of the 24 duck
species in the Bristol Bay region, at least 11 prey on salmon eggs, parr, or smolts, or scavenge on salmon
carcasses (Brna and Verbrugge 2013).
5.3.1.7 Shorebirds
Thirty of 41 shorebird species or subspecies that regularly occur in Alaska can be found in the Bristol
Bay watershed (see Brna and Verbrugge [2013] for a summary of different shorebird surveys).
Shorebirds use the Bristol Bay watershed primarily during migration and breeding. Significant areas of
intertidal habitat exist at Kvichak Bay (530 km2) and Nushagak Bay (400 km2). Important foods include
abundant intertidal invertebrates, and fruits and tubers in upland areas. Shorebirds likely play an
important role in the distribution of MDNs to terrestrial ecosystems. Adults, young, and eggs also
provide a source of food for predatory birds and terrestrial mammals. Although there is not a strong
direct link between salmon and shorebirds, it is reasonable to assume that MDNs contribute to the
abundance of invertebrates in the intertidal zone.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTI April 2013
-------
Chapter 5 Endpoints
The Bristol Bay/Alaska Peninsula lagoon system, which includes the Nushagak and Kvichak River deltas,
is one of the most important migratory shorebird stop-over areas in Alaska. Surveys of the Pebble
deposit area in 2004-2005 identified 14 shorebird species in the Pebble deposit area (PLP 2011).
5.3.1.8 Land Birds
Approximately 80 species of land birds, both migratory and year-round residents, breed in and adjacent
to the Nushagak and Kvichak River watersheds. Land birds eat vegetation (seeds, berries),
invertebrates, and vertebrates. Studies indicate that the abundance of many songbird species is related
to the presence of salmon carcasses (Christie and Reimchen 2008; Gende and Willson 2001; Willson et
al. 1998). Salmon carcasses provide food for aquatic invertebrate larvae, and MDNs provide increased
plant productivity (Cederholm etal. 1999, Gende etal. 2002), both important food sources for land
birds. Few abundance studies have focused on the Nushagak and Kvichak River watersheds, but 2004-
2005 surveys in the Pebble deposit area identified 28 land bird species in the Pebble deposit area (PLP
2011).
5.3.2 Recreational and Subsistence Activities
Many of the species discussed in the preceding sections are important subsistence resources. For
example, a 2002 survey of Bristol Bay residents found that 86% and 88% of respondents have
consumed moose and caribou meat, respectively (Ballew et al. 2004). Between 1983 and 2006, moose
harvest in GMU 17 increased from 127 to 380 per year; the upper Nushagak River watershed alone
(GMU 17B) has a mean annual harvest of 149 moose (Brna and Verbrugge 2013). Caribou harvest
ranged from 1,573 to 4,770 per year between 1991 and 1999, but this estimate is for the entire
Mulchatna herd, including those taken outside of the Nushagak and Kvichak River watersheds
(Valkenburg et al. 2003).
Waterfowl support recreational and subsistence harvests, as well as wildlife viewing opportunities.
There are no reliable estimates of recreational harvests specific to the Nushagak and Kvichak River
watersheds. Subsistence harvest of waterfowl is very important in the watershed. The spring harvest
provides fresh meat early in the season after winter food supplies are depleted. Harvest data from 1995
through 2005 for Dillingham, Nushagak River, and Iliamna subregions (Wentworth 2007, Wong and
Wentworth 1999) estimates annual harvest of roughly 10,000 ducks, 2,500 to 2,900 geese, up to
300 tundra swans, as well as fewer than 500 waterfowl eggs (Brna and Verbrugge 2013).
Sport hunting for caribou, moose, brown bear, and other species also plays a role in the local economy of
the Bristol Bay region. In recent years approximately 1,323 non-residents and 1,319 non-local residents
of Alaska traveled to the region to hunt. Miller and McCollum (1994) estimate that non-residents and
non-local residents spend approximately $5,170 and $1,319 per trip (values updated to 2009 dollars),
respectively. These hunting activities result in an estimated $8.2 million per year in direct hunting-
related expenditures (Table 5-4) and directly employ over 100 full- and part-time workers.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment c OH April 2013
-------
Chapter 5 Endpoints
5.4 Endpoint 3: Alaska Natives
5.4.1 Alaska Native Populations
Fourteen of the Bristol Bay watershed's 25 Alaska Native villages and communities are within the
Nushagak and Kvichak River watersheds, with a total population of 4,337 in 2010 (U.S. Census Bureau
2010). Dillingham (population 2,329) is the largest community; other communities range in size from
two (year-round) residents (Portage Creek) to 510 residents (New Stuyahok). Because population in
some communities is seasonal, these numbers increase during the subsistence fishing season. Thirteen
of these 14 villages—all but Port Alsworth—have federally recognized tribal governments and had an
Alaska Native population majority in 2010.
Overall population in the region grew 55% from 1980 to 2000, and remained relatively stable from
2000 to 2010 (U.S. Census Bureau 2010). Population has fluctuated in individual villages since 1980
(Appendix D, Table 2). From 2000 to 2010, nine villages lost and five villages gained population. The
extent to which these changes reflect natural population fluctuations and whether any gains or losses
indicate a long-term trend are unknown. Four of the villages that lost population (Dillingham, Igiugig,
Aleknagik, and Kokhanok) and one of villages that gained population (Iliamna) changed less than 10%.
Port Alsworth has experienced steady population growth since 1980. Its economy is more closely tied to
Lake Clark National Park, and its population contains the smallest proportion of Alaska Natives among
the 14 villages. Portage Creek is the smallest village in the region, and its population has fluctuated
significantly over the past 40 years (e.g., 48 in 1980, 5 in 1990, 36 in 2000, 2 in 2010), making it difficult
to draw conclusions about trends.
5.4.2 Subsistence and Alaska Native Cultures
5.4.2.1 Importance of Salmon to Alaska Native Cultures
The primary Alaska Native cultures present in the Nushagak and Kvichak River watersheds—the Yup'ik
and Dena'ina—are part of the last intact, sustainable salmon-based cultures in the United States. This is
especially significant as other Pacific Northwest salmon-based cultures struggle with degraded
resources (Colombi and Brooks 2012). Cultures associated with salmon fishing appeared in these
watersheds as early as 2000 BC and intensified around AD 1000 (Appendix D). Currently, the percentage
of Alaska Native population in the region's villages ranges from 21.4% (Port Alsworth) to 95.7%
(Koliganek) (Appendix D, Table 2). The Yup'ik and Dena'ina cultures still provide the framework and
values for everyday life in the region. Among the Yup'ik, over 40% of the population continues to
maintain their native language, one of the highest percentages among native cultures in the United
States (Appendix D).
In the Bristol Bay region, the subsistence way of life is irreplaceable. Subsistence resources provide high
quality foods, foster a healthy lifestyle, and form the basis for social relations for both Alaska Natives
and non-Alaska Natives in the villages. These resources—particularly salmon—are integral to the entire
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7TIApril 2013
-------
Chapter 5 Endpoints
way of life in Yup'ik and Dena'ina cultures. The Alaska Federation of Natives (2010) describes
subsistence as follows.
The hunting, fishing, and gathering activities which traditionally constituted the economic base of
life for Alaska's Native peoples and which continue to flourish in many areas of the state
today...Subsistence is a way of life in rural Alaska that is vital to the preservation of communities,
tribal cultures, and economies. Subsistence resources have great nutritional, economical, cultural,
and spiritual importance in the lives of rural Alaskans...Subsistence, being integral to our
worldview and among the strongest remaining ties to our ancient cultures, are as much spiritual
and cultural as it is physical.
For Alaska Natives today, subsistence is more than the harvesting, processing, sharing, and trading of
land and sea mammals, fish, and plants. Subsistence holistically subsumes the cultural, social, and
spiritual values that are the essence of Alaska Native cultures. There is a strong tradition and practice of
sharing and trading subsistence resources. Food is shared with tribal Elders, family living outside of the
watershed, and others who may not be able to fully participate in subsistence (Appendix D). This
practice was confirmed by tribal Elders interviewed for Appendix D and those who testified at public
meetings on the draft assessment (Box 5-3).
Tribal Elders and culture bearers continue to instruct young people, particularly at fish camps where
cultural values as well as fishing and fish processing techniques are shared. The social system that forms
the backbone of the culture, nurturing the young, supporting the producers, and caring for the tribal
Elders, is based on the virtue of sharing the wild foods harvested from the land and waters. Sharing
networks extend to family members living far from home. The first salmon catch of the year is
recognized with a prayer of thanks and shared in a continuation of the ancient First Salmon Ceremony
(Appendix D), when those who have caught the first king salmon in the spring share them with tribal
Elders and all those in need, as well as with friends and family. Cultural and personal identity largely
revolves around traditional cultural practices such as hunting, fishing, and gathering of wild food
resources—that is, subsistence.
Traditional and more modern spiritual practices place salmon in a position of respect and importance,
as exemplified by the First Salmon Ceremony and the Great Blessing of the Waters (Appendix D). The
salmon harvest provides a basis for many important cultural and social practices and values, including
the sharing of resources among the people, fish camp, gender and age roles, and the perception of
wealth. While a small minority of tribal Elders and culture bearers interviewed expressed a desire to
bring in more market economy opportunities (including large-scale mining), most equated wealth with
stored and shared subsistence foods (Appendix D). Consistently, in the interviews conducted for
Appendix D, the Yup'ik and Dena'ina communities of the Nushagak and Kvichak River watersheds define
a "wealthy person" as one with food in the freezer, a large extended family, and the freedom to pursue a
subsistence way of life in the manner of their ancestors. Their ability to continue their reliance on
subsistence and their concept of wealth have contributed to the maintenance of vital and viable cultures
for the last 4,000 years.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g 03 April 2013
-------
Chapter 5 Endpoints
5.4.2.2 Use of Subsistence Resources in the Bristol Bay Watershed
Alaska Native populations, as well as non-Alaska Native residents, have continual access to a range of
subsistence foods. As described by Fall et al. (2009), these subsistence resources are the most consistent
and reliable component of the local economies in the Bristol Bay watershed—even given the world-
renowned commercial fisheries and other recreational opportunities the region supports (Table 5-4).
Subsistence uses on state lands are given priority by state law and regulations (i.e., the 1978 State of
Alaska Subsistence Act). All citizens of Alaska benefit from a subsistence priority in areas specifically
designated as subsistence areas by the State of Alaska. State hunting and fishing regulations apply to
lands of the Alaska Native Corporations. These lands were often selected because of their significant
value for subsistence activities, and Alaska Native peoples have the exclusive right to occupy and use
these lands for subsistence. These rights are not recognized in the State of Alaska Constitution; however
the Alaska Federation of Natives has passed resolutions for several years asking for the constitution to
be revised. In addition, the Alaska Federation of Natives recommended improvements to management of
state and federal subsistence programs. Indigenous hunting and fishing rights are recognized by statute
only and therefore can be diminished over time. Their lack of special status makes these rights
vulnerable to constitutional challenges, especially challenges based on the right to equality (Duhaime
and Bernard 2008).
No watershed data are available for the proportion of Bristol Bay watershed residents' diets made up of
subsistence foods, as most studies focus on harvest data and not dietary surveys. A study that included
the nearby Yukon-Kuskokwim region found that 22.8 % of calories came from Native (subsistence)
foods (Johnson et al. 2009). In 2004 and 2005, annual subsistence consumption rates in the Nushagak
and Kvichak River watersheds were over 300 pounds per person in many villages, and reached as high
as 900 pounds per person (Appendix D, Table 12; for comparison, an average American consumes
1,996 pounds of food per year). Villages with the highest per-capita subsistence usage were Koliganek,
Ekwok, Newhalen, Kokhanok, Igiugig, and Levelock. Subsistence use varies throughout the Bristol Bay
watershed, as villages differ in the per-capita amount of subsistence harvest and the variety of
subsistence resources used. Salmon and other fishes provide the largest portion of substantial
subsistence harvests of Bristol Bay communities. On average, about 50% of the subsistence harvest by
local community residents (measured in pounds usable weight) is Pacific salmon, and about 10% is
other fishes (Fall et al. 2009). The percentage of salmon harvest in relation to all subsistence resources
ranges from 29% to 82% in the villages (Appendix D, Table 11). Salmon accounts for an especially high
percentage compared to all subsistence resources for Iliamna, Kokhanok, and Pedro Bay. Igiugig,
Levelock, and New Stuyahok show the lowest percentage of salmon usage relative to other subsistence
resources. Villages in the Nushagak River watershed, especially New Stuyahok, Ekwok, and Dillingham,
rely on Chinook salmon to a great extent, whereas villages in the Kvichak River watershed and Iliamna
Lake area (e.g., Iliamna, Kokhanok, Iguigig, Newhalen, Nondalton, Pedro Bay, and Port Alsworth) rely
primarily on sockeye salmon. All communities also rely on non-salmon fishes, but to a lesser extent than
salmon (Table 5-1). These fishes are taken throughout the year by a variety of harvest methods and fill
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment T7 April 2013
-------
Chapter 5 Endpoints
an important seasonal component of subsistence cycles (Fall etal. 2009). For example, whitefish and
other freshwater species provide fresh fish during winter ice-fishing season (Appendix D).
The ADF&G overview of subsistence fisheries in the Bristol Bay watershed (Fall etal. 2009) provides the
following information.
• The number of Bristol Bay subsistence salmon permits issued has been stable since 1990, and the
recent 10-year average is 1,146 permits. Most permit holders (84%) are residents of Bristol Bay
communities, and most permits are issued for the Nushagak and Naknek/Kvichak districts. Sockeye
salmon make up the largest portion of the Bristol Bay subsistence salmon harvest (79% of the
1998-2007 average, based on subsistence salmon permits), followed by Chinook (19%), coho (5%),
chum (5%), and pink (2%).
• Annual subsistence harvests for the Bristol Bay management area vary from year to year. Salmon
harvest declined from the early 1990s to the early 2000s but has slightly recovered since 2002.
Since 1975, the average annual harvest was about 152,371 salmon; the recent 5-year average
(2003-2007) was 126,717 salmon.
• The largest decline over the last 15 years has occurred in the Kvichak River watershed subsistence
sockeye salmon fishery, historically the largest component of the Bristol Bay subsistence salmon
harvest. Declines are due to lower harvests per permit, rather than reduced fishing effort. Since
1996, harvest per day is down 26% in years of escapements under 2 million fish, compared to the
previous 13-year average. The long-term average (45 years, for which permit data are available) for
this fishery is 66,614 sockeye salmon.
• There has been an overall harvest decline in the Nushagak district from a high of 86,400 fish in 1986
to a low of 40,373 salmon in 2006. The 24-year average harvest (the time for which data are
available) is 50,740 fish. However, the number of subsistence salmon permits issued in the
Nushagak district has remained relatively stable since 1983.
• Subsistence salmon harvests in the Nushagak district are similar to those in the Kvichak district in
terms of harvest levels. In 2007, for example, based on permit returns, the communities in the
Nushagak district harvested 44,944 salmon, compared to 47,538 salmon in the Kvichak
River/Iliamna Lake subdistrict. However, there are differences in the two fisheries. Whereas salmon
harvest in the Kvichak River watershed is almost all sockeye salmon (47,473 out of 47,538 in 2007),
salmon harvest in the Nushagak district is more varied, with larger harvests of Chinook, coho, and
chum salmon. There are also larger communities in the Nushagak district, including Dillingham,
Manokotak, Aleknagik, New Stuyahok, and Koliganek.
• Chinook salmon returns are higher in the Nushagak River watershed than in the Kvichak River
watershed. In the upper portion of the Nushagak River residents attempt to harvest large numbers
of Chinook salmon, their traditionally preferred salmon resource. Chinook salmon spawn early in
the season, and it is important to put up these fish for subsistence before commercial fishing starts
in earnest (Holen et al. 2012). Substitution of Chinook for sockeye salmon accounts for some, but not
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g 35 April 2013
-------
Chapter 5 Endpoints
all, of the decline in the Nushagak district. Subsistence sockeye salmon harvests in the Kvichak River
watershed, including Iliamna Lake and Lake Clark, historically the largest component of the Bristol
Bay subsistence salmon fishery, declined by more than 50% during the 1990s and early 2000s. Local
subsistence fishers attributed these lowered harvests to poor returns and scarcities of salmon in
once reliable and abundant traditional harvest locations. Effort has increased in harvesting salmon
in these areas since the low harvest levels seen in early 2000.
Figures 5-2 and 5-12 show areas of subsistence use identified by ADF&G in the Nushagak and Kvichak
River watersheds. Clark's Point subsistence use areas overlap with Nushagak and Kvichak River
watersheds for caribou, coho salmon, and moose. Clark's Point high per-capita harvest rate (1,210 Ibs
per capita) resulted from a high harvest rate of salmon in 2008. This was three times higher than the
harvest levels reported in 1973 and 1989 (Holen etal. 2012). Manokotak subsistence use areas overlap
with the Nushagak communities for caribou and moose. Aleknagik moose search areas include part of
Nushagak River area (Holen et al. 2012). South Naknek, Naknek, and King Salmon subsistence use areas
for waterfowl, rainbow, unspecified trout, moose, and berry picking, as well as caribou search areas,
overlap the Nushagak and particularly the Kvichak River watersheds (Holen etal. 2011). It is likely that
subsistence use occurs outside of the areas identified on the figures. Data used to generate the figures
were collected in different years, and at least one village (Ekwok), which has high recorded subsistence
harvests, declined to be surveyed. Also note that these figures do not indicate abundance or harvest,
only use.
Subsistence is a non-market economic activity and is not officially measured (Appendix E). There is a
strong and complex relationship between subsistence and the market economy (largely commercial
fishing and recreation) in the area. Goods and services (e.g., boats, rifles, nets, snow mobiles, and fuel)
are purchased by households and used for subsistence activities. The commercial fishing and recreation
market economy provides seasonal employment for residents, giving them income to purchase goods
and services needed for subsistence, as well as time to participate year-round in subsistence activities.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g og April 2013
-------
Chapter 5
Endpoints
Figure 5 12. Subsistence use intensity for salmon, other fishes, wildlife, and waterfowl within the
Nushagak and Kvichak River watersheds. See Box 5 1 for more detailed discussion of methodology.
I
Cook Inlet
Naknek
South Naknek
Bristo/ Bay
N
A
25 50
25
] Kilometers
50
] Miles
« Approximate Pebble Deposit Location
• Surveyed Towns and Villages
° Nonsurveyed Towns and Villages
Existing Roads
| | Watershed Boundary
Subsistence Use Intensity
High
Low
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
5-37
April 2013
-------
Chapter 5 Endpoints
The salmon-dependent diet of the Yup'ik and Dena'ina benefits their physical and mental well-being in
multiple ways, besides encouraging high levels of fitness based on subsistence activities. The interviews
conducted for Appendix D confirm ADF&G harvest data that people of the Nushagak and Kvichak
primarily eat two species of Pacific wild salmon, sockeye and Chinook. These are consumed in different
ways, including fresh, salted, pickled, canned, dried, and smoked. Salmon and other traditional wild
foods comprise a large part of the people's daily diet throughout their lives, beginning as soon as they
are old enough to eat solid food. (Appendix D). Subsistence foods consumed in rural Alaska have
demonstrated multiple nutritional benefits, including lower cumulative risk of nutritionally mediated
health problems such as diabetes, obesity, high blood pressure, and heart disease (Murphy et al. 1995,
Dewailly et al. 2001, Dewailly et al. 2002, Din et al. 2004, Alaska Department of Health and Social
Services 2005, Chan etal. 2006, Ebbesson and Tejero 2007) and provision of essential micronutrients
and omega-3 fatty acids (Bersamin etal. 2007, Nobmann etal. 2005) (Murphy etal. 1995, Ebbesson and
Tejero 2007).
A disproportionately high amount of total diet protein and some nutrients comes from subsistence
foods. A 2009 study of two rural regions (Johnson et al. 2009) found that 46% of protein, 83% of
vitamin D, 37% of iron, 35% of zinc, 34% of polyunsaturated fat, 90% of eicosapentaenoic acid
(EPA PUFA 20:5n-3), and 93% of docosahexaenoic acid (DHA PUFA 22:6n-3) came from subsistence
foods consumed by Alaska Natives.
In summary, the roles of salmon as a subsistence food source and as the basis for Alaska Native cultures
are inseparable, and the characteristics of these subsistence-based salmon cultures have been widely
documented (Appendix D). The cultures have a strong connection to the landscape and its resources; in
the Bristol Bay watershed, this connection has been maintained for centuries by the uniquely pristine
condition of the region's landscape and resources. In turn, the respect and importance given salmon and
other wildlife, along with the traditional knowledge of the environment, has produced a sustainable
subsistence-based economy (Appendix D). This subsistence-based way of life is a key element of Alaska
Native identity and it serves a wide range of economic, social, and cultural functions in Yup'ik and
Dena'ina societies (Appendix D). Appendix D states the following.
... Salmon and clean water are foundational to the Yup'ik and Dena'ina cultures in the Nushagak
and Kvichak watersheds. The people in this region not only rely on salmon for a large proportion of
their highly nutritional food resources; salmon is also integral to the language, spirituality, and
social relationships of the culture. Because of this interconnection, the cultural viability, as well as
the health and welfare of the local population, are extremely vulnerable to a loss of either quality
or quantity of salmon resources.
It should be noted that, even though the scope of the assessment is focused on villages in the Nushagak
and Kvichak River watersheds, subsistence harvest areas do not necessarily correspond with watershed
boundaries. As noted previously, villages outside of these watersheds use areas in the watersheds for
subsistence activities and obtain subsistence resources from the watersheds through sharing or
bartering.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o April 2013
-------
Chapter 5 Endpoints
BOX 5 3. TESTIMONY ON THE IMPORTANCE OF
SUBSISTENCE USE
The USEPA held a series of public meetings to collect input on the draft Bristol Bay Assessment. Many
Alaska Natives, including tribal Elders and other tribal leaders, provided testimony on the importance of
salmon and the subsistence way of life to Alaska Native cultures in the watersheds. The following are
selected quotes representative of this testimony; complete public meeting transcripts are available at
www.epa.gov/BristolBay.
• "Our subsistence way of life plays a substantial role in our health both spiritually and physically."
• "From traditional knowledge we keep our culture going. My subsistence life is with my family, which
consists of four boys and my wife. I also help my grandmother, grandfather, mother, father, and our other
family members. I hold a Bristol Bay drift permit, my family fishes with me both commercially and
subsistence. My family processes approximately 4,000 pounds of salmon, kings, reds, silvers, etc. We
start when the fish first come into the river, all the way to the very end. My family and I smoke, dry, and
freeze the salmon. I brought you some canned salmon to share that we keep year round."
• "The king salmon is a very important part of our fishery. If you cover that portion of the king [Chinook]
salmon spawning beds, it is going to make it very hard for us to maintain our culture of people who eat
king every year. Is the first fish of the year, it's a very important fish for us and we can't have that huge
loss."
• "Fishing is our life and our livelihood. It's what we do for healthy communities, healthy lifestyles. Going
out and catching the subsistence fish, smoking these. Passing the traditional knowledge on to younger
generations. You hear about how they will make you free, the fish. We have been doing this for 6,000
years and we will want to do it for 6,000 more."
• "The generations that are coming who can be fed from this resource and this land and it's a beautiful
interaction and it's one that we are losing around the world. When we realize that we have lost it we strive
to get it back, but it is taking a long time for this beautiful balance between human, animal and
subsistence lifestyle to come about and evolve."
• "The survival of our culture directly depends on the health of our land, the fish and the wildlife. No
amount of money or jobs can replace our way of life and our culture."
• "I am a Dena'ina, and Athabascan Indian. This village is my home. We are very rich people in our culture,
our resources, plants, animals and salmon. They all need clean water. That includes us, the Dena'ina
people of the land. But only because we are so blessed to have clean water. Salmon have been a great
part of our diet for generations and will be in the future."
• "Right now we are getting excited for the kings to come up our river. For everyone works together cutting
fish. To dry, salt or vacuum pack for the winter. We do not waste anything, because we fish. Around here
it is gold, gold to us which we treasure. When we fill our dry rack, we go walking and help one another."
• "I've lived here for 30 years and I moved here by choice. My experience of living in this area is that people
choose to be here whether born or coming here. It's a choice. It is not a scientific fact, but three reasons
people choose to be in Bristol Bay is because clean water, the fishery and the lifestyle."
• "This environment has sustained our culture for thousands of years. It sustained jobs and commercial
fishing for hundreds of years, and recreation and sport fishing and everything."
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment c on April 2013
-------
6.1 Basic Elements of the Mine Scenarios
For this assessment, we used information on porphyry copper deposits and mining practices
summarized in Chapter 4 to develop three mine size scenarios: Pebble 0.25 with 0.25 billion ton
(0.23 billion metric tons) of ore, Pebble 2.0 with 2.0 billion tons (1.8 billion metric tons) of ore, and
Pebble 6.5 with 6.5 billion tons (5.9 billion metric tons) of ore. The word Pebble in the names of the
scenarios represents the fact that we place our scenarios at the Pebble deposit. These three mine
scenarios, as well as other scenario types considered in later chapters of the assessment, are
summarized in Table 6-1.
These three mine scenarios represent realistic, plausible descriptions of potential mine development
alternatives, consistent with current engineering practice and precedent. The scenarios are not mine
plans: they are not based on a specific mine permit application, and are not intended to be the detailed
plans by which the components of a mine would be designed. However, the scenarios are based on
preliminary mine details put forth in Northern Dynasty Minerals' Preliminary Assessment of the Pebble
Mine (Ghaffari et al. 2011), as well as information from scientific and industry literature for mines
around the world (see Chapter 4 and Appendix H for background information on mining and the geology
of porphyry copper deposits). Thus, the mine scenarios reflect the general activities and processes
typically associated with the kind of large-scale porphyry copper mining development likely to be
proposed once a specific mine application is developed. We use these scenarios to benchmark potential
risks resulting from this type of development, to provide decision makers with a better understanding of
potential risks associated with any specific action proposed in the future.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-1
April 2013
-------
Chapter 6
Mine Scenarios
Table 6 1. Summary of scenarios considered in the assessment.
Scenario Type
Mine size
Water collection,
treatment, and
discharge
Tailings dam failure
Scenario
Pebble 0.25
Pebble 2.0
Pebble 6. 5
Routine operations3
Wastewater treatment
plant (WWTP) failure3
Pebble 0. 25 TSF failure
Pebble 2. OTSF failure
Transportation corridor
Pipeline failure
Product concentrate
pipeline failure15
Return water pipeline
failure15
Diesel pipeline failure15
Description
Mine size of 0.25 billion ton (0.23 billion metric ton) of
ore
Mine size of 2.0 billion tons (1.8 billion metric tons) of
ore
Mine size of 6.5 billion tons (5.9 billion metric tons) of
ore
All water collection and treatment at site works
properly, and wastewater is treated to meet state and
national standards before release; however, some
leachate from waste rock and TSFs is not captured.
WWTP fails and releases untreated wastewater through
its two outfalls.
Failure of 90-m dam at TSF 1
Failure of 209-m dam at TSF 1
113-km gravel road with three pipelines, within the
Kvichak River watershed
Complete break or equivalent failure of the product
concentrate pipeline
Complete break or equivalent failure of the return
water pipeline
Complete break or equivalent failure of the diesel
pipeline
Assessment
Chapters)
7,8
8
9
10
11
Notes:
3 Each water collection, treatment, and discharge scenario was considered for each mine size scenario.
b Each pipeline failure scenario was considered at two locations: Chinkelyes Creek and Knutson Creek.
In the scenarios, we make decisions concerning mine placement; the size of the mine and the time over
which mining would occur; the size, placement, and chemistry of waste rock; the size, placement, and
chemistry of tailings storage facilities (TSFs); on-site processing of the ore; and the removal of processed
ore concentrate from the site. For comparison purposes, Table 4-1 provides similar information on
other past, existing, and potential large mines in Alaska. The mine components described in the
scenarios are placed on the landscape based on information either from Ghaffari et al. (2011), or where,
in our experience, modern mining practice suggests a component would be placed. For example, the pit
is located on the deposit; TSFs are placed in the locations described in Ghaffari et al. (2011) and where
topography provides an efficient location to store a large volume of tailings; waste rock is placed around
the pit to minimize the cost of hauling millions to billions of tons of material; and the transportation
system is located within the corridor described in Ghaffari et al. (2011). We describe only the
components of a mine that have the potential to adversely affect aquatic resources regulated under the
federal Clean Water Act (33 USC 1251-1387) (Box 6-1). For example, mine buildings such as the mill,
offices, and housing would be expected to be placed in uplands to avoid wetlands, ponds, and streams;
thus, they are only addressed as they relate to stormwater runoff.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-2
April 2013
-------
Chapters Mine Scenarios
BOX 6 1. CUMULATIVE FOOTPRINT OF A LARGE SCALE PORPHYRY COPPER MINE
In this assessment, we focus on the footprints of the primary mine components: mine pit, tailings storage
facilities, waste rock piles, and the transportation corridor. The actual infrastructure needed to operate any
large-scale mine would be significantly more extensive than these four components and would result in a
larger cumulative mine footprint. These additional infrastructure needs (based on Ghaffari et al. 2011)
would include, but are not limited to, the following.
• Mining and processing facilities, including grinding mills, ore stockpiles, conveyers, a wastewater
treatment plant, and process water ponds and distribution lines.
• Drainage management structures, such as seepage cutoff walls, stream diversion channels, drainage
ditches, and sediment control ponds.
• Other storage and disposal facilities, such as overburden and topsoil stockpiles, explosives storage, a
non-hazardous waste landfill, process water storage tanks, waste incinerators, a fuel storage compound,
and hazardous waste storage.
• Other operational infrastructure, such as administrative buildings, dormitories, a sewage treatment
plant, a power generation plant, power distribution lines, potable water treatment plant and distribution
lines, and a truck shop.
These additional infrastructure needs were considered outside the scope of this assessment, but it should
be noted that the cumulative footprint of a large-scale mine at the Pebble deposit likely would be much
larger than the footprints evaluated in the mine scenarios.
• The footprint and operational infrastructure for a 25-year mine at the Pebble deposit (similar to the
Pebble 2.0 scenario) would cover approximately 125 km2 (Ghaffari et al. 2011); in comparison, the mine
components considered in this assessment (pit, waste rock piles, and TSF) would cover approximately
34 km2 under the Pebble 2.0 scenario (Table 6-2).
• Net power generation for such a mine would be approximately 378 megawatts (Ghaffari et al. 2011). This
is more than 100 times the maximum electrical load of the largest population center in the Bristol Bay
watershed, the Dillingham/Aleknagik area (Marsik 2009), and slightly less than half of the combined
capacity of the two electric utilities that serve more than 40% of Alaska's total population (CEA 2011,
ML&P2012).
• Dormitories for such a mine would house 2,150 people during construction and more than 1,000 people
during mine operation (Ghaffari etal. 2011), meaningthatthe mine site would rival Dillingham as the
largest population center in the Bristol Bay watershed during construction and would remain the second
largest population center during operation.
• The mine site could contain more than 19 km of main roads, as well as numerous pit and access roads,
and would depend on a fleet of 50 to 100 vehicles, in addition to approximately 150 or more large, ore-
hauling trucks (Ghaffari et al. 2011). Potential risks associated with these roads would be similar in type
to those described in Chapter 10.
We specify that all mine components would be developed using modern conventional design and
practice and operated under standard industry practices. Our purpose in this assessment is to evaluate
the potential effects of mining porphyry copper deposits in the Nushagak and Kvichak River watersheds
given design and operation to these standards. We have included basic descriptions of design features
intended to mitigate potential adverse effects of mine operation.
In spite of these design and operation standards, however, any large-scale mine in the Bristol Bay region
would have a footprint that would affect aquatic resources (Figures 6-1 through 6-3). These footprint-
related impacts are addressed in Chapter 7. Additional impacts that may result from human error,
mechanical failure, accidents, and other unplanned events are considered in Chapters 8 through 11.
Compensatory mitigation for effects on aquatic resources that cannot be avoided or minimized by mine
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ITI April 2013
-------
Chapters Mine Scenarios
design and operation would be addressed through a regulatory process that is beyond the scope of this
assessment (see Box 7-2 and Appendix J).
Section 6.1 describes the mine components common to the three mine scenarios (and most other mines
of this type, as described in Chapter 4). Section 6.2 describes specific characteristics of each mine
scenario relevant to our assessment, including water treatment and discharge. Section 6.3 describes
closure of the mines, and Section 6.4 provides conceptual models of the relationships between mine
components, potential stressors, and biotic responses.
6.1.1 Location
The mine scenarios considered in this assessment are sited at the Pebble deposit, in headwaters of the
Nushagak and Kvichak River watersheds where the South and North Fork Koktuli Rivers and Upper
Talarik Creek originate (Figure 2-5). The Pebble deposit represents the most likely site for near-term,
large-scale mining development in the Bristol Bay watershed. Many other mineral exploration sites in
the Nushagak and Kvichak River watersheds report findings consistent with a porphyry copper deposit
similar to the Pebble deposit (see Table 13-1 and Figure 13-1 for other mineral prospects in the area).
Non-porphyry copper deposits being explored in the area are likely to require similar mining facilities
such as an open pit, a tailings impoundment, and waste rock dumps, and may produce acid-generating
materials. Salmon and other fish occur in streams throughout the Nushagak and Kvichak River
watersheds (Chapter 5; Appendices A and B). Thus, much of our analysis is transferable to other
portions of the two watersheds; a mining operation at any one of these sites could have qualitatively
similar impacts to a mine operation at the site of the Pebble deposit. However, we recognize that specific
component placement could differ based on site-specific factors at each mine. Because our scenarios are
located at the Pebble deposit, we refer to them throughout the text as Pebble 0.25, Pebble 2.0, and
Pebble 6.5. This distinguishes the site of the analysis from other potential mine sites in the Nushagak
and Kvichak River watersheds that are included in the evaluation of cumulative impacts (Chapter 13).
6.1.2 Mining Processes
6.1.2.1 Extraction
Ore associated with the western portion of the Pebble deposit is near-surface and, in our scenarios,
would be mined via conventional open-pit mining methods of drilling and blasting. Pit depth and width
would be increased progressively to recover the ore. Pit walls and benches would be constructed to
stabilize slopes for safety and to optimize runoff. Dusts would be controlled by wetting surfaces with site
water and covering truck beds during transport of excavated rock. Groundwater flow into the pit would
be managed by pumping to storage ponds or TSFs for later treatment or use in mine processes. Although
our scenarios describe open pit mining, underground methods could be used, particularly for the deeper
eastern portion of the ore body. Many of the impacts would be similar in type and magnitude to those of
surface mining (Section 4.2.3.1 and Box 4-4).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment Z~7April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 1. Footprint of the Pebble 0.25 scenario. Individual mine components are the open mine
pit, waste rock area, and the tailings storage facility (TSF). Blue areas indicate streams and lakes
from the National Hydrography Dataset (USGS 2012a) and wetlands from the National Wetlands
Inventory (USFWS 2012).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-5
April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 2. Footprint of the Pebble 2.0 scenario. Individual mine components are the open mine pit,
waste rock area, and the tailings storage facility (TSF). Blue areas indicate streams and lakes from
the National Hydrography Dataset (USGS 2012a) and wetlands from the National Wetlands Inventory
(USFWS2012).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-6
April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 3. Footprint of the Pebble 6.5 scenario. Individual mine components are the open mine pit,
the waste rock areas, and three tailings storage facilities (TSFs). Blue areas indicate streams and
lakes from the National Hydrography Dataset (USGS 2012a) and wetlands from the National
Wetlands Inventory (USFWS 2012).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-7
April 2013
-------
Chapters Mine Scenarios
6.1.2.2 Ore Processing
In the mine scenarios, an in-pit crusher would reduce the ore to a constant maximum size and a
conveyor would bring the crushed ore to processing facilities. Ore would be processed in a flotation
circuit similar to that described in Section 4.2.3.3. The milling process would generate two tailings
streams, one from the rougher flotation circuit (bulk tailings having undergone a single grind sequence)
and another from the secondary cleaner circuit (cleaner scavenger tailings) (Figure 4-3). Selective
flotation would be used to minimize the amount of potentially acid-generating (PAG) tailings. Copper
(+gold) and molybdenum concentrates would be produced as described in Section 4.2.3.3, with the
copper (+ gold) slurry concentrate pumped via pipeline to Cook Inlet and the final molybdenum
concentrate dried, bagged, and trucked off site for processing. Gold associated with the copper minerals
in the slurry concentrate would be recovered at an off-site smelter. Pyrite tailings would be directed
either to the TSF for subaqueous disposal or to a vat leach cyanidation operation for removal of gold,
after which sulfide-rich tailings would be directed to the TSF for subaqueous disposal. A cyanide
destruction unit would be used at the end of the leaching process.
All chemical reagents used in ore processing (Box 4-5) would be transported to the mine site
(Section 10.3.3), then prepared and stored in areas with secondary containment and instrumentation to
detect any spills or leaks. All pipelines would be designed to standards of the American Society of
Mechanical Engineers (ASME), which include the use of liners to minimize abrasion and corrosion,
freeze protection, secondary containment over water bodies, and leak monitoring and detection. Dusts
would be controlled in the processing area through use of cartridges, wet scrubbers, and/or enclosures.
6.1.2.3 Waste Rock
Waste rock consisting of both PAG and non-acid-generating (NAG) materials would be stored around the
mine pit, at least partially within the cone of depression from mine pit dewatering. PAG waste rock
would be stored separately from NAG waste rock and over the life of the mine would be blended with
processed ore to allow consistency in chemical usage and to remove material from surface storage prior
to its expected time of acid generation (e.g., within 20 years of its excavation). Any PAG material
remaining unprocessed at the end of mining would be processed separately prior to closure.
During operation, waste rock piles would be constructed with a 2:1 slope for structural stability and
minimization of the amount of runoff requiring treatment. Waste rock piles would occupy
approximately 2.3 km2,13.0 km2, and 22.6 km2 under the Pebble 0.25, 2.0, and 6.5 scenarios,
respectively (Table 6-2). Water quality of the leachate from waste rock is described in Tables 8-6 and 8-
7. Monitoring and recovery wells and seepage cutoff walls would be placed downstream of the piles to
manage seepage, with seepage and contaminated groundwater directed either into collection ponds for
use in mine processes or for treatment and release to the environment, or into the mine pit. Stormwater
falling upslope of waste rock piles would be diverted around the piles and directed toward
sedimentation ponds for settling of suspended solids prior to discharge to a nearby stream, or for
treatment if determined to be contaminated. Embankments would be constructed above the seepage
cutoff walls to contain any excess storm water runoff that could not be contained in collection ponds.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment JTApril 2013
-------
Chapters Mine Scenarios
Water captured in these embankments would be released or directed to treatment as appropriate.
Because the Tertiary volcanic rocks were classified as NAG (Ghaffari et al. 2011, PLP 2011), they may be
useful for building purposes such as TSF construction. However, because of the potential for metals
leaching, use would be appropriate only where leachate would be collected for treatment as necessary.
6.1.2.4 Tailings Storage Facilities
In the mine scenarios, TSF dam design would proceed as described in Ghaffari et al. (2011). The number
and size of TSFs in each scenario would be commensurate with tailings storage requirements
(Figures 6-1 through 6-3). The water rights application submitted by Northern Dynasty Minerals to the
State of Alaska in 2006 described several potential locations for TSFs (NDM 2006). We assume that the
higher mountain valleys similar to the site of TSF 1, on the flanks of Kaskanak Mountain, are the most
plausible sites for TSFs for a mine at the Pebble deposit given geotechnical, hydrologic, and
environmental considerations. This placement does not imply that these sites would not pose
unacceptable environmental harm, or that they would be the least environmentally damaging
practicable alternatives for purposes of Clean Water Act permitting. Permit-specific study, which is
beyond the scope of this assessment, would determine if these or other sites met these criteria.
At each TSF, a rockfill starter dam would be constructed, with a liner (high-density polyethylene
geomembrane on top of a geosynthetic clay liner) extending up the upstream dam face. Seepage capture
and toe drain systems would be installed at the upstream toe, with perpendicular drains installed to
direct seepage toward collection ponds. The TSF would be unlined other than on the upstream dam face,
and there would be no impermeable barrier constructed between tailings and underlying groundwater.
As tailings accrued near the top of the starter dam, dam height would be raised using the downstream
construction method (Figure 4-4) (Ghaffari et al. 2011). At some point, dam construction would shift to
the centerline method (Figure 4-4), and a new stage would be constructed as the capacity of each
previous stage is approached. TSF 1 would require maximum dam heights of approximately 90 m and
209 m for the Pebble 0.25 and Pebble 2.0 scenarios, respectively (Figure 6-4).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g p April 2013
-------
Chapter 6
Mine Scenarios
Table 6 2. Mine scenario parameters.
Parameter
Amount of ore mined (billion metric tons)
Approximate duration of mining
Ore processing rate (metric tons/day)
Mine Scenario
Pebble 0.25
0.23
20 years
31,000
Pebble 2.0
1.8
25 years
198,000
Pebble 6.5
5.9
78 years
208,000
Mine Pit
Surface area (km2)
Depth (km)
1.5
0.30
5.5
0.76
17.8
1.24
Waste Rock Pile
Surface area (km2)
PAG waste rock (million metric tons)
PAG waste rock bulk density (metric tons/m3)
PAG waste rock area (km2)
NAG waste rock (million metric tons)
NAG waste rock bulk density (metric tons/m3)
NAG waste rock area (km2)
2.3
95
2.08
0.49
350
2.08
1.84
13.0
580
2.08
1.79
2,200
2.08
11.2
22.6
4,700
2.08
6.77
10,900
2.08
15.8
TSFl"
Capacity, weight (billion metric tons)
Surface area, interior (km2)
Surface area, exterior (km2)
Maximum dam height (m)
Maximum number of dams
Capacity, volume (million m3)
Tailings dry density (metric tons/m3)b
NAG density, embankment (metric tons/m3)b
0.25
5.71
5.88
90
1
177
1.42
2.31
1.96
14.2
15.8
209
3
1,380
1.42
2.31
1.96
14.2
15.8
209
3
1,380
1.42
2.31
TSF 2"
Capacity, weight (billion metric tons)
Surface area, interior (km2)
Surface area, exterior (km2)
Maximum dam height (m)
Maximum number of dams
Capacity, volume (million m3)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.7
20.0
21.5
not determined
3
2,600
TSF3"
Capacity, weight (billion metric tons)
Surface area, interior (km2)
Surface area, exterior (km2)
Maximum dam height (m)
Maximum number of dams
Capacity, volume (million m3)
Total TSF surface area, exterior (km2)
NA
NA
NA
NA
NA
NA
5.9
NA
NA
NA
NA
NA
NA
15.8
0.96
7.7
8.3
not determined
3
679
45.6
Transportation Corridor
Total length (km)
Length in assessment watersheds (km)
138
113
138
113
138
113
Notes:
a Final value when TSF is full.
b Values are the same for TSF 2 and TSF 3, so not repeated under those TSFs.
NA = not applicable; TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-10
April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 4. Height of the dam at TSF 1 relative to U.S. landmarks. Dam heights evaluated under the
Pebble 0.25 and Pebble 2.0 TSF failure scenarios (Chapter 9) are also shown.
Transamerica Building - 260 Meters
Tailings Dam TSF 1 - 209 Meters
Gateway Arch -192 Meters
Washington Monument -169 Meters
Maximum Pebble 2.0 Tailings
Tailings Reservoir
Maximum Pebble 0.25 Tailings
lamngsuam
260
240-
220-
200
180
160
I 14°"
EU
5 120-
100-
80-
60-
40-
20-
0-
Given the low grade of ore expected in the region, our mine scenarios would produce large amounts of
tailings: approximately 99% of the mass of ore processed would be tailings, with 85% as NAG bulk
tailings and 14% as PAG (pyritic) tailings (Ghaffari etal. 2011). Both types of tailings would be directed
to TSFs (Figures 6-1 through 6-3). The discharge of bulk tailings would be managed such that the
coarsest mate rials (fine sand) would be discharged at intervals along the inside perimeter of the TSF to
form beaches, while finer materials (silt) would be carried with discharged water toward the center of
the impoundment. Pyritic tailings would be discharged below the water surface of the tailings pond and
encapsulated in NAG tailings to retard the rate of pyrite oxidation.
The capacity and dimensions of each TSF are listed in Table 6-2. Pebble 6.5, the largest size scenario
considered, would require the construction of TSFs 1, 2, and 3, with a combined tailings capacity
exceeding 6 billion metric tons. We estimate that these three TSFs would have a combined surface area
of about 46 km2 (Table 6-2).
During operation, water quality in TSF ponds would be similar to process water. At the end of mining,
process water would no longer enter the tailings facility, so it is expected that, over time, dilution from
precipitation would cause the composition of tailings pond water to approach that of local surface water.
Seepage from the base of the tailings impoundment, either during operation or after closure would be
expected to be similar to water quality estimates produced by pre-mining humidity-cell test results
(Appendix H). The low solubility of oxygen in water (less than 15 mg/L) limits the access of oxygen to
submerged unreacted sulfide minerals in the tailings, reducing dissolution reaction rates and thus the
concentration of solutes. Furthermore, under anoxic conditions commonly encountered in sulfidic
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-11
April 2013
-------
Chapters Mine Scenarios
tailings, trace amounts of carbonate or silicate minerals may partially neutralize acid, further limiting
the solubility of metals and other trace elements (Blowes etal. 2003). However, a good deal of
uncertainty exists because the humidity cell tests used to predict pore water chemistry represent a
small sample of the ore body. Actual water quality in the tailings impoundment may differ significantly
from what is estimated (Appendix H). For example, lower concentrations of metals than those reported
in humidity cells tests would likely be seen in TSF water if pH was buffered by reactions with carbonate
and silicate minerals (see Section 8.1.1.1 for discussion of tailings leachate quality).
Well fields spanning the valley floor would be installed at the downstream base of all tailings dams to
monitor groundwater flow down the valley, including potential uncaptured seepage from the TSF. If
contaminated groundwater was detected, monitoring wells would be converted to collection wells or
new recovery wells would be installed, and water from the well field would be pumped back into the
TSF or treated and released to stream channels.
6.1.2.5 Water Management and Treatment
Water uses in the mine scenarios would include ore processing, tailings slurry transport, and transport
of copper concentrate slurry in the product pipeline. In this section, we provide an overview of water
management and treatment under the three mine scenarios. Figure 6-5 presents a schematic illustration
of these components (but note that, for clarity, diversions of stormwater around mine components are
not shown on the schematic).
• Stormwater runoff that did not contact potential contaminants would be diverted around mine
components (e.g., waste rock piles, processing facilities) in ditches directed toward sediment
settling ponds. After settling, water would be tested for compliance with discharge limits and, if
required, treated prior to being discharged to the environment.
• Stormwater runoff from waste rock piles and water from pit dewatering would be pumped to lined
process water ponds; water reclaimed from tailings impoundments or tailings thickening also would
be stored in the process water storage ponds for reuse in ore processing.
• Stormwater falling onto TSFs would be stored in the tailings impoundments and used in the process
water cycle.
• Seepage collected from waste rock piles and TSFs would be directed to lined seepage collection
ponds or TSFs for later treatment.
• Seepage escaping the waste rock and TSF leachate collection systems would be monitored with
monitoring wells. If groundwater contamination was detected, wells would be converted to recovery
wells, or new recovery wells would be installed, and the groundwater pumped either to a TSF or a
storage pond for later treatment.
• Water reclaimed from the copper concentrate after transport to the port would be returned to the
process water storage ponds via pipeline from the port.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment JTI April 2013
-------
Chapters Mine Scenarios
• Streams blocked by the mine pit or waste rock piles would, where practicable, be diverted around
and downstream of the mine. However, the zone of groundwater depression around the mine pit
and the slow filling of the post operation pit would likely dewater these streams for hundreds of
years.
• Prior to being discharged, water would be treated to meet effluent limits using chemical
precipitation methods and/or reverse osmosis. Water would be discharged to the South and North
Fork Koktuli Rivers (Section 6.2.2.4) according to permit conditions for composition, flow, and
temperature. Sludges and brines from the treatment process would be disposed in the TSF.
Water balances for both the operation and post-closure phases of our mine scenarios are discussed in
detail in Section 6.2.2. Development of these water balances is important, because they estimate the
amount of water available to contribute to downstream flows. Calculating these water balance
components is challenging, however, and requires a number of assumptions (e.g., estimates of the
amount of water needed to support mining operations, the amount of water delivered to the site via
precipitation, the amount of water lost due to evaporation, and the net balance of water to and from
groundwater sources). Information exists to estimate precipitation and evaporation, and estimates of
water needed for mining operations are available based on typical mining practices (Ghaffari et al.
2011). More challenging—and potentially the largest source of uncertainty in these calculations—is the
net balance of water from groundwater sources.
Mining operations would affect the quantity, quality, timing, and distribution of surface flows. Mining
operations always consume some water, so there would be less water available in the landscape during
active mining than before the mine was present. Major stream flow reductions during mine operation
would result from the capture of precipitation falling on the mine pit, waste rock piles, and TSFs
(Table 6-3, Figure 6-5). The mine pit would capture precipitation directly, but pit dewatering would also
draw down the water table beyond the rim of the pit, creating a cone of depression that would extend
underneath the waste rock piles (Figure 6-5). Leachate recovery wells for any detected groundwater
contamination downstream of the waste rock piles would extend the cone of depression. Because the
mine pit would be located on a water divide, we estimate that there would be little net contribution from
groundwater flow into the area defined by the cone of depression, and that the cone of depression
would expand until water flow into the mine pit was balanced by recharge from precipitation over the
cone of depression. The cone of depression would lower the groundwater table, drying up streams,
ponds, and wetlands that depend on groundwater discharge and turning areas of groundwater
discharge into areas of groundwater recharge. Precipitation and other water collected in the mine pit or
from recovery wells would be pumped to a process water pond or to one of the TSFs. Water falling
within the perimeter of a TSF would be captured directly in the TSF, but runoff from catchment areas
up-gradient of the TSF would be diverted downstream. Runoff at the port site would be pumped to the
mine site in the return water pipeline, contributing to the mines water supply and avoiding the need for
treatment at the port.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment J~77April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 5. Water management and water balance components for the three mine scenarios.
Return Water
and Site Runoff
from Port
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-14
April 2013
-------
Chapter 6
Mine Scenarios
Prior to active mining, but after the starter dam was built for TSF 1, site water would be diverted to TSF
1 to allow sufficient water for process plant startup. During mine operation, groundwater and
precipitation would be pumped from the open pit to prevent flooding of the mine workings (Figure 6-5).
Water would be needed for the flotation mill, to operate the TSF, and to maintain concentrated slurry in
the product pipeline.
Table 6 3. Summary of water balance flows (million m3/year) during operations for the three mine
scenarios.
Flow Component
Captured at mine pit area
Captured at TSF 1
Captured at TSF 2
Captured at TSF 3
Captured at mill & other facilities
Potable water supply well(s)
Water in ore (3%)
Total Captured
Cooling tower losses
Water in concentrate to port
Water in concentrate return
Runoff collected from port
Stored in TSF as pore water
Stored in mine pit
Crusher use
Total Consumptive Losses
Returned to streams via wastewater treatment plant
Returned as NAG waste rock leachate
Returned as PAG waste rock leachate
Returned as TSF leakage
Total Reintroduced
Percent of captured water reintroduced
Pebble 0.25
10.3
4.91
0
0
0.559
0.031
0.340
16.1
0.211
0.166
-0.149
-0.125
4.08
0
0.113
4.30
10.2
0.556
0
1.10
11.9
73.9%
Pebble 2.0
24.9
12.2
0
0
2.24
0.124
2.17
41.6
1.32
1.04
-0.934
-0.251
23.2
0
0.722
25.1
12.2
1.78
0.216
2.35
16.5
39.7%
Pebble 6.5
45.4
12.2
17.2
6.62
2.24
0.124
2.27
86.1
1.32
1.04
-0.934
-0.251
24.2
0
0.758
26.1
49.4
2.37
1.03
7.19
60.0
69.7%
Notes:
TSF = tailings storage facility; NAG = non-acid-generating; PAG = potentially acid-generating.
In hard rock metal mining, most water use occurs during milling and separation operations; however,
much of this water is recycled and reused. For example, much of the water used to pump the tailings
slurry from the mill to a TSF becomes available when the tailings solids settle, and excess overlying
water is pumped back to the mill. Water losses occur when there is a consumptive use and that water is
no longer available for reuse. In our mine scenarios, consumptive uses would include cooling tower
losses through drift and evaporation, water in the concentrate sent to the port that would exceed the
amount recovered and returned in the return water line, and water that would fill the pore spaces in the
TSF (Table 6-3). TSF pore water accounts for greater than 90% of the mine operations water demand
(Table 6-3). Consumptive losses would be made up by withdrawing water stored in a TSF or by pumping
directly from the mine pit. Some of this captured water (40 to 73%, Section 6.2.2.3) would not be needed
at the mine site (Figure 6-5). This excess captured water would be treated to meet existing water quality
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-15
April 2013
-------
Chapters Mine Scenarios
standards and discharged to nearby streams, partially mitigating flow lost from eliminated or blocked
upstream reaches.
6.1.3 Transportation Corridor
6.1.3.1 Roads
Development of any mine in the Bristol Bay watershed would require substantial expansion and
improvement of the region's transportation infrastructure. The Bristol Bay watershed is located in one
of the last remaining, virtually roadless regions in the United States. There are no improved federal or
state highways, and no railroads, pipelines, or other major industrial transportation infrastructure.
Roadways presently link Iliamna Lake (Pile Bay) to Cook Inlet (tidewater at Williamsport) and the
Iliamna area (including Iliamna airport) north to the site of a proposed bridge over the Newhalen River
near the village of Nondalton. Two other short road segments link Dillingham to Aleknagik and Naknek
to King Salmon (Figure 6-6). Local roads also exist in villages throughout the Nushagak and Kvichak
River watersheds. Most people travel by air or boat during the ice-free season, and by air or snow
machine in winter.
In our mine scenarios, a 138-km, two-lane (9-meter-wide), gravel surface, all-weather permanent access
road would connect the mine site to a new deep-water port on Cook Inlet (Figure 6-6), from which
concentrate would be shipped elsewhere for processing (Ghaffari et al. 2011). An estimated 113 km of
this corridor would fall within the Kvichak River watershed (this value does not include the portion of
this road occurring within the potential mine site). This route would traverse highly variable terrain and
variable subsurface soil conditions, including extensive areas of rock excavation in steep mountainous
terrain.
The primary purpose of this road would be to transport freight by conventional highway tractors and
trailers, although critical design elements would be dictated by specific oversize and overweight loads
associated with project construction. Material sources for road embankment fill, road topping, and
riprap (e.g., borrow and gravel pits and rock quarries) would be available at regular intervals along the
road route. We assume state-of-the-art practices for design, construction, and operation of the road
infrastructure, including design of bridges and culverts for fish passage. Permanent structures would be
designed for a service life of 50 years. Because the access road would be kept open for ongoing care,
maintenance, and environmental monitoring at the site post-closure, maintenance and resurfacing of the
access road would necessarily be required for the same time, which may extend in perpetuity.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment R ,. R April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 6. Transportation corridor connecting the Pebble deposit area to Cook Inlet
Approximate Pebble Deposit Location
Transportation Corridor
Transportation Corridor (Outside Assessment Area)
Watershed Boundary
Existing Roads
-^ /
KVICHAKCa. r
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-17
April 2013
-------
Chapters Mine Scenarios
The transportation corridor would cross many streams, rivers, wetlands, and extensive areas with
shallow groundwater, including numerous mapped (and unmapped) tributary streams to Iliamna Lake
(Figure 6-6, Section 10.3.2). Approximately 21 bridges would be constructed over larger, anadromous
streams and rivers along the length of the corridor, with spans ranging from approximately 12 to 183 m.
At the remaining water crossings—that is, the majority of these crossings—culverts would be installed.
In addition, there would be a 573-m (1,880-foot) causeway across the upper end of Iliamna Bay, and
approximately 8 km of embankment construction along coastal sections in Iliamna Bay and Iniskin Bay
(Ghaffarietal. 2011).
Avalanche hazards exist in isolated locations along the alignment, but routing would attempt to avoid
any avalanche chutes and runout areas. Because of steep mountain slopes and lack of significant
vegetation at high elevations, storm runoff can rapidly accumulate and result in intense local runoff
conditions. Road areas near the south slope of Knutson Mountain and the southeast slope of the
mountain above Lonesome Bay and Pile Bay may be especially susceptible to these runoff events, as
demonstrated in 2004, when storm runoff washed out several culverts on the state-maintained Pile Bay
Road.
6.1.3.2 Pipelines
The transportation corridor would include four pipelines, which would carry copper (+gold)
concentrate, return water, natural gas (to fuel a natural gas-fired generating plant), and diesel fuel
between the mine site and the Cook Inlet port (Table 6-4). All pipelines would be designed following the
standards of the ASME. Except at stream and river crossings, pipelines would be buried together in a
trench adjacent to the road alignment, in the right-of-way. At short stream and river crossings, pipelines
would be bored under channels to minimize waterway impacts. At longer crossings, pipelines would be
supported aboveground on road bridges. Any aboveground pipeline sections would be constructed of
double-walled pipe. Freeze protection would be provided by insulation (aboveground pipes) or burial
(1.5 meters below ground surface). External corrosion would be prevented by a cathodic protection
system. A leak detection system would be built into the pipelines, which would also assist in the
detection and prevention of slack flows. A supervisory control and data acquisition (SCADA) system
would monitor and control pumping facilities via a fiber optic line buried alongside the pipelines.
Instruments such as pressure and temperature transducers located along the pipeline route would be
tied into the fiber optic link.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g , „ April 2013
-------
Chapter 6
Mine Scenarios
Table 6 4. Characteristics of pipelines in the mine scenarios.
Pipeline
(#of pipes)
Route
Pipe Material
Nominal Diameter (cm)
Along Transportation Corridor
Copper (+gold) concentrate (1)
Reclaimed water (1)
Natural gas (1)
Diesel fuel (1)
Mine to port
Port to mine
Port to mine
Port to mine
HDPE-lined steel
HDPE-lined steel
Steel
Steel
20
18
5
13
At Mine Site
Bulk tailings (2)
Pyritic tailings (2)
Reclaimed water (1)
Reclaimed water (1)
Mine pit dewatering(l)
Process plant to TSF
Process plant to TSF
TSF barge to TSF head tank
TSF head tank to process
pond
Pit to process pond or TSF
Steel with liner
Steel with liner
HOPE
Steel
Steel
86
46
107
107
TBD
Notes:
HOPE = high-density polyethylene; TSF = tailings storage facility; TBD = to be determined.
Source: Ghaffari etal. 2011.
On the mine site, pipelines would carry tailings slurry from the process plant to the TSFs and reclaimed
water from the TSFs to the process water storage ponds (Table 6-4). There also would be smaller
pipelines for water supply, firefighting, and process flows within the plant. In this assessment, we
assume that any leakage from pipelines in the process plant area would be captured and controlled by
the plant's drainage system and either be treated prior to discharge or pumped to the process water
storage pond or the TSFs. Failures of these on-site pipelines could result in uncontrolled releases in the
mine site, but these failures are not evaluated in this assessment. At mine closure, concentrate and
return water pipelines would be removed. Diesel and natural gas pipelines would be retained as long as
fuel was needed at the site for monitoring, treatment, and site maintenance. It is also possible that local
communities would select to retain the pipelines for continued use.
6.2 Specific Mine Scenarios
In this assessment we evaluate three specific mine scenarios, representing mines of different sizes. The
smallest mine scenario, Pebble 0.25, represents the median-sized porphyry copper deposit of
250 million tons (227 metric tons) (Singer etal. 2008). The second mine scenario, Pebble 2.0, is based
largely on the 25-year, 2 billion tons (1.82 billion metric tons) case described in Ghaffari et al. (2011) for
initial development at the Pebble deposit. The third mine scenario, Pebble 6.5, is based largely on the
78-year, 6.5 billion tons (5.92 billion metric tons) case described in Ghaffari et al. (2011) for further
resource development at the Pebble deposit.
Pebble 2.0 and Pebble 6.5 reflect projects based on extensive exploration, assessment, and preliminary
engineering, which are described in Ghaffari et al. (2011) as "economically viable, technically feasible
and permittable". They are among the most likely to be developed in the watershed and are site-specific
to the Pebble deposit. For the purposes of this assessment, we have also placed the Pebble 0.25 mine
scenario at the Pebble deposit because of the availability of site-specific information. If mines are
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-19
April 2013
-------
Chapters Mine Scenarios
developed at other exploration sites in the watershed (Figure 13-1), they are likely to have
characteristics and impacts much closer to those of the Pebble 0.25 mine scenario. Table 6-2 provides
detailed parameters for each of our three mine scenarios, and Figures 6-1 through 6-3 show the general
layout of mine components.
6.2.1 Mine Scenario Footprints
The major mine components contributing to the mine footprint include the mine pit, waste rock piles,
and TSFs. Placement of these components for each of the scenarios is shown in Figures 6-1 through 6-3.
In each case, these layouts represent one possible configuration for the mine; other configurations are
possible, but would be expected to have impacts of similar types and magnitudes. Each footprint would
also include ancillary facilities such as ore-crushing and screening areas, processing mill, laydown areas
(e.g., temporary storage areas for construction materials), camp, fuel storage areas, stockpile areas,
workshops, roads within the mine site, pipeline corridors, and other disturbed areas. To estimate the
unavoidable direct impacts from each mine scenario footprint, we calculated mine component areas for
each scenario (Tables 6-5 through 6-7).
6.2.1.1 Pebble 0.25 Footprint
Figure 6-1 shows the general layout of the mine pit, waste rock piles, and TSF for the Pebble 0.25
scenario. The TSF, identified as TSF 1, is located in a natural valley in the headwaters of the North Fork
Koktuli River located to the west of the Pebble deposit. The valley would be closed off by the
construction of a rockfill dam 90 m in height (Table 6-2). The waste rock pile footprint was determined
by calculating the areas that would be covered by the expected volume of waste rock, taking advantage
of natural landforms near the mine pit. In this scenario, the PAG and NAG waste rock would be separated
in the waste rock pile during mine operation, and processed throughout the mine life, as mill conditions
permit, with the intent to process all of the PAG waste rock before mine closure. The area of the plant
and ancillary facilities is estimated to account for 4% of the total disturbed area (Table 6-5).
The footprint of the drawdown zone created by pit dewatering includes the mine pit and the area
beyond the mine pit perimeter up to the limit of the cone of depression (see Box 6-2 for discussion of
calculations). Much, but not all, of the waste rock pile area lies within this drawdown zone. The
cumulative pit area includes the drawdown zone and the waste rock areas beyond the drawdown zone
(Table 6-5).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g »Q April 2013
-------
Chapter 6
Mine Scenarios
Table 6 5. Estimated areas for individual mine components under the Pebble 0.25 scenario.
Component
Mine pit
Waste rock piles3
NAG waste rock in drawdown zone
NAG waste rock not in drawdown zone
PAG waste rock in drawdown zone
PAG waste rock not in drawdown zone
Drawdown zone
Cumulative mine pit areab
Cumulative plant and ancillary areas0
Total TSF (TSF 1 interior area)
TOTAL
Area (km2)
1.5
2.3
0.5
1.3
0.5
0.0
10.7
12.0
0.7
5.7
18.4
Notes:
a 100-m-high piles.
b Cumulative mine pit area = drawdown zone (incorporating mine pit and part of waste rock piles within the drawdown zone) + NAG and PAG
waste rock piles outside the drawdown zone.
0 Estimated as 25% of Pebble 2.0.
d Total = cumulative mine pit area + cumulative plant and ancillary areas + TSF 1.
NAG = non-acid-generating; PAG = potentially acid-generating; TSF = tailings storage facility.
6.2.1.2 Pebble 2.0 Footprint
Figure 6-2 depicts the general layout of the major mine components for the Pebble 2.0 scenario,
including the mine pit, the waste rock piles, and the TSF. The TSF is located in the same valley as TSF 1
in the Pebble 0.25 scenario (Figure 6-2), but it is increased in size to accommodate the additional
tailings expected under this larger mine size. The plant and ancillary facilities are estimated to account
for 6% of the total disturbed area (Table 6-6).
Footprints of the waste rock piles are located around the perimeter of the mine pit, with separate areas
designated for NAG and PAG waste rock. In our scenario, the PAG and NAG waste rock would be
separated in the waste rock pile during mine operation and processed throughout the mine life as mill
conditions permit with the intent to process all of the PAG waste rock before mine closure. Dewatering
of the mine pit would generate a cone of depression around the pit. Most of the footprint of the waste
rock piles would be within the dewatering cone of depression (Table 6-6).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-21
April 2013
-------
Chapter 6
Mine Scenarios
Table 6 6. Estimated areas for individual mine components under the Pebble 2.0 scenario.
Component
Mine pit
Waste rock piles3
NAG waste rock in drawdown zone
NAG waste rock not in drawdown zone
PAG waste rock in drawdown zone
PAG waste rock not in drawdown zone
Drawdown zone
Cumulative mine pit areab
Cumulative plant and ancillary areas
Total TSF (TSF 1 interior area)
TOTAL'
Area (km2)
5.5
14.7
7.1
5.8
1.3
0.5
22.6
28.9
2.6
14.2
45.7
Notes:
3 100-m-high piles.
b Cumulative mine pit area = drawdown zone (incorporating mine pit and part of waste rock piles within the drawdown zone) + NAG and PAG
waste rock piles outside the drawdown zone.
c Total = cumulative mine pit area + cumulative plant and ancillary areas + TSF 1.
NAG = non-acid-generating; PAG = potentially acid-generating; TSF = tailings storage facility.
6.2.1.3 Pebble 6.5 Footprint
The general layout of the Pebble 6.5 scenario is similar to that of the Pebble 2.0 scenario, with major
differences being a larger open pit, different and expanded areas for the waste rock piles, and the
inclusion of two additional TSFs (TSF 2 and TSF 3) to store the increased tailings volume (Figure 6-3,
Table 6-7). Placement of TSF 2 and TSF 3 in this scenario draws upon some of the TSF options presented
in Northern Dynasty Minerals' water rights application (NDM 2006) and takes advantage of natural
landforms in the Pebble deposit area.
The mine pit is located as shown by Ghaffari et al. (2011), based on the evaluation of the Pebble deposit.
The waste rock piles are located around the perimeter of the expanded mine pit, with some portion of
the PAG waste rock stored within part of the exploited Pebble West pit to utilize storage within the
drawdown zone prior to the PAG waste rock being taken to the surface for processing. This practice
would reduce the amount of PAG waste rock that must be stored outside the drawdown zone and,
therefore, the amount of PAG leachate that could leak into the South Fork Koktuli River.
Areas of the plant and ancillary facilities are the same as those described for the Pebble 2.0 scenario;
because production rates of the Pebble 2.0 and Pebble 6.5 scenarios are similar, no increase in these
areas is needed for the larger mine scenario.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-22
April 2013
-------
Chapter 6
Mine Scenarios
Table 6 7. Estimated areas for individual mine components under the Pebble 6.5 scenario.
Component
Mine Pit
Waste Rocka
NAG waste rock in drawdown zone
NAG waste rock not in drawdown zone
PAG waste rock in drawdown zone
PAG waste rock in mine pit
PAG waste rock not in mine pit
PAG waste rock not in drawdown zone
Drawdown zone
Cumulative mine pit areab
Cumulative plant and ancillary areas
TSF 1 interior area
TSF 2 interior area
TSF 3 interior area
Total TSF
TOTAL'
Area (km2)
17.8
24.2
10.3
7.2
4.4
2.0
2.4
2.4
44.9
52.8
2.6
14.2
20.0
7.7
41.9
97.3
Notes:
3 100-m-high piles.
b Cumulative mine pit area = drawdown zone (incorporating mine pit and part of waste rock piles within the drawdown zone) + NAG and PAG
waste rock piles outside the drawdown zone.
c Cumulative mine pit area + cumulative plant and ancillary areas + total TSF.
NAG = non-acid-generating; PAG = potentially acid-generating; TSF = tailings storage facility.
6.2.2 Water Balance
Many of the most significant impacts of large-scale mining relate to a mine's use of water and its impact
on water resources. To understand potential impacts of water use in our mine scenarios, we developed
an annual water balance that accounts for major flows into (via streamflow and precipitation and out of
(via surface or groundwater flow) the mine area for each scenario. Three major categories of flows make
up each water balance estimate: water inputs, consumptive losses, and water outputs; these categories
are discussed in detail in the following sections. These water balances focus on changes in flows entering
or leaving the mine site, relative to pre-mining conditions. Changes are divided into flows that would be
withdrawn or captured from the natural system and flows that would be released to the natural system.
Each water balance subtracts consumptive water losses within mine operations from water inputs to
determine the water available for release. The water balance analysis does not attempt to describe or
quantify internal flows among mine components, although some are mentioned where necessary to
explain the analysis. The water balance analysis also does not attempt to quantify any flows that are
recycled within the mine site because these do not capture water from the environment or release water
to it.
6.2.2.1 Water Inputs
Water inputs for each of the three scenarios are summarized in Table 6-3. These inputs are derived
primarily from net precipitation (total precipitation minus any losses due to evapotranspiration) that
falls on the mine footprints and is captured by water collection and management systems within the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-23
April 2013
-------
Chapters Mine Scenarios
mine site. We assume that all captured flows would be available for use by the mine operator. Three
gages surrounding the mine site were used to calculate net precipitation at the mine site: gage SK100B
(USGS gage 15302200) on the South Fork Koktuli River, gage NK100A (USGS gage 15302250) on the
North Fork Koktuli River, and gage UT100B (USGS gage 15300250) on Upper Talarik Creek. Monthly
mean flows for each gage were summed across the year, producing an area-weighted average of net
runoff of 860 mm per year.
Water inputs resulting from the mine footprints are calculated as the product of footprint areas
multiplied by annual net precipitation. For the TSFs, the volume of water captured is based on the
interior area of the TSF, defined as the area within the dam crests and excluding the downstream faces
of the rockfill dams.
Dewatering the open pit would create a cone of depression around the mine extending beyond the limits
of the mine pit. Because the mine pit would be located very close to the water divide between the South
Fork Koktuli River, North Fork Koktuli River, and Upper Talarik Creek watersheds, we assume that there
would be negligible net influx of groundwater from beyond the cone of depression. Most of the
groundwater outside the cone of depression would flow away from the site. Therefore, the area of the
cone of depression would be determined by matching net precipitation falling on the drawdown area
with the calculated groundwater inflow into the mine pit (Box 6-2).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g -4 April 2013
-------
Chapters Mine Scenarios
BOX 6 2. MINE PIT DRAWDOWN CALCULATIONS
Groundwaterflow into the mine pit was calculated using a simplified model based on the Dupuit-Forcheimer
discharge formula for steady-state radial flow into a fully penetrating well in a phreatic aquifer with a
diameter equal to the average mine pit diameter. The hydraulic conductivity data gathered in the Pebble
deposit area during geologic investigations show significant scatter (Figure 6-7). We based our analysis on
the hydraulic conductivity (k) varying with depth, with logk varying linearly from the surface to a depth of
200 m (k = 1 xlO'4 m/s at the surface and k = 1 xlO'8 m/s at depths greater than or equal to 200 m). Given
these values, negligible flow occurs below a depth of 200 m, so our model included a no-flow boundary at
that depth. To apply the Dupuit-Forcheimer formula, we needed to transform the cross-section into an
equivalent isotropic section by transforming the vertical dimension so that the thickness at any depth was
proportional to the hydraulic conductivity at that depth. The initial water table in our simplified model was at
the ground surface and assumed to be horizontal.
Our analysis assumed that the drawdown at the mine pit was 100 m, but we also verified that the results
were not very sensitive to this assumption. The radius of influence was determined by balancingthe net
precipitation falling within the cone of depression with the calculated flow into the mine pit. Inflows were
calculated to be 0.27 m3/s (4,330 gpm), 0.52 m3/s (8,200 gpm) and 1.06 m3/s (16,830 gpm) for the
Pebble 0.25, 2.0, and 6.5 mine scenarios, respectively. The Pebble 2.0 mine inflow agrees closely with the
estimate provided in Ghaffari etal. (2011).
The cone of depression was determined to extend 1148 m, 1222 m, and 1260 m from the edge of the
idealized circular mine pit under the Pebble 0.25, 2.0, and 6.5 scenarios, respectively. In a geographic
information system (GIS), we established the boundary of the cone of depression at those distances from
the actual perimeter of the mine pits to derive the drawdown areas presented in Tables 6-5 through 6-7.
The waste rock piles do not lie completely within the drawdown areas. This is important in assessing water
quality because precipitation falling on the waste rock piles within the drawdown area is presumed to be
collected within the mine pit whereas precipitation falling outside of the drawdown area is presumed to
migrate away from the mine pit. To more accurately assess the waste rock pile positions relative to the
drawdown area, we distorted the shape of the cone of depression by superimposing the drawdown area on a
uniform flow field with a southern gradient of 0.0354, approximately equal to the slope of the ground
surface across the mine pit from north to south. The effect of this distortion is a shift in the boundaries of
the cone of depression to the north, resulting in larger areas of waste rock outside of the drawdown area.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment Q25 April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 7. Hydraulic conductivity in the Pebble deposit area. Data are from three test types:
bedrock packer (Lugeon) tests (blue diamonds, with error bars indicating upper and lower limits of
zone tested) (PLP 2011: Chapter 8 and Appendix 8.IN); overburden rising or falling head tests (red
squares) (PLP 2011: Chapter 8 and Appendix 8.1C); and bedrock rising or falling head tests (green
triangles) (PLP 2011: Chapter 8 and Appendix 8.1C). Red line indicates values used in the
assessment's mine pit drawdown and tailings storage facility leakage calculations.
1.6
U
Cfl
1 on
E
0.
LU
Q
200
250 -
300
-09
f
<
HYDRAULIC CONDUCTIVITY (m/s)
l.E-08 l.E 07 l.b 06 l.b Ob
'
I
1
1
i
i
»
;
\
'
1
[
1
I
A
\
•}I
fit
'
T
I
< f
'1
L
X
x
1
H '
1
ii
\
*
i
T1
i
!
-
i
T
4
$
/
•
*
X
,
Lf!
i
5/TI
i
*
1
*
A
X
l.E-04 l.E 03
• taf
>^A
X
1
1
JL 1
"
if
1
.
4 Packer Tests (showing packer limits)
• Overburden Rising/Falling Head Tests
Bedrock Rising/Falling Head Tests
Approximation used in Bristol Bay Assessment
Precipitation falling on areas outside of these disturbed footprints would infiltrate as groundwater or
flow into streams without treatment. Flow in upstream tributaries blocked by the mine footprint would
be piped or otherwise diverted around the footprint and discharged back into streams without
treatment, where practicable. Because this diverted flow is not captured by the mine operations, it is not
explicitly included in the water balance tabulations and is assumed to remain part of the background
flow.
6.2.2.2 Consumptive Losses
Consumptive losses for each mine scenario are summarized in Table 6-3. To estimate the amount of
water available for release, we subtracted consumptive losses associated with mining activities from the
captured flows (Table 6-3). Consumptive losses would include water pumped to the port in the copper
(+gold) concentrate pipeline minus return water, cooling tower evaporation and drift losses, interstitial
water trapped in the pores of stored tailings, water used for dust suppression, and water stored in the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-26
April 2013
-------
Chapters Mine Scenarios
mine pit after closure. The tailings pore water accounts for over 90% of consumptive loss during mine
operations (Table 6-3). When the tailings settle, about 46% of the volume would consist of voids
between solid particles; the water trapped in these pore spaces would no longer be available for use at
the mine or release to streams.
Information on flows in the concentrate and return water pipelines and on cooling tower losses is
reported in Ghaffari et al. (2011). The return water pipeline reduces consumptive losses by returning
water from the port (e.g., from dewatering the copper [+gold] concentrate and from stormwater runoff
collected at the port site). We estimated the area of the port facilities over which runoff was likely to be
collected (137,160 m2) and multiplied that area by the precipitation rate at the port (1,830 mm/year) to
determine contributions from port site runoff (Table 6-2). We also included a consumptive loss at the
crusher and screening site for dust control equal to 1% of the mass of the material being crushed.
6.2.2.3 Water Outputs
When the amount of captured water exceeds consumptive losses, water would be available, after testing
and treatment, for release into area streams. This released water may differ from natural stream water
in chemistry and temperature, but would comply with permitted discharge requirements. Water may be
reintroduced at locations, flow rates, or times of year that differ from baseline conditions.
The water deficit for each scenario—that is, the amount of water extracted from the environment and
not returned to streams—is presented in Table 6-3. These water deficits equal the total consumptive
losses of 4.3 million m3/year, 25 million m3/year, and 26 million m3/year for the Pebble 0.25, 2.0, and
6.5 scenarios, respectively. The percentage of water reintroduced to streams, including uncontrolled
leachate escapes, would equal 74, 40, and 70% of the total water captured in the three scenarios,
respectively.
6.2.2.4 Additional Water Balance Issues
During the early life of each mine, there is one other significant source of water that the mine operator
would need to manage that is not considered in Table 6-3: the water obtained from dewatering the
sandy and gravelly overburden overlying the waste rock and ore. Based on an average overburden
thickness of 30.5 m and a porosity of 0.40, dewatering the overburden would produce one-time
quantities of 19 million m3, 67 million m3, and 220 million m3 of water over the mine pit areas in the
Pebble 0.25, 2.0, and 6.5 scenarios, respectively. This water would be expected to be relatively clean and,
if properly managed to control turbidity, could most likely be released without chemical treatment to
maintain or augment stream flow.
Water treated at the wastewater treatment plant (WWTP) and discharged into streams might not be
reintroduced into the same dewatered streams. The WWTP is assumed to discharge to the South and
North Fork Koktuli Rivers, but not to Upper Talarik Creek (Figures 6-8 through 6-11).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment « April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 8. Water flow schematic for the Pebble 0.25 scenario. Flows include water from the non acid generating (NAG) waste rock pile and
tailings storage facility (TSF) 1 (dashed black arrows), discharge from the wastewater treatment plant (WWTP) (solid black arrows), flow along
the stream channels (solid blue arrows), and known groundwater transfers (dashed blue arrow); for clarity, only flows greater than 5% of total
outflows from the TSF and waste rock pile are shown. Gaging stations are maintained by U.S. Geological Survey and Pebble Limited
Partnership. SK South Fork Koktuli; NK North Fork Koktuli; UT Upper Talarik Creek.
NK10QB
UT10QE
NK100A •
• UT100C2
UT100C1
SK100B
UT100C
UT100B
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-28
April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 9. Water flow schematic for the Pebble 2.0 scenario. Flows include water from the potentially acid generating (PAG) and non acid
generating (NAG) waste rock piles and tailings storage facility (TSF) 1 (dashed black arrows), discharge from the wastewater treatment plant
(WWTP) (solid black arrows), flow along the stream channels (solid blue arrows), and known groundwater transfers (dashed blue arrow); for
clarity, only flows greater than 5% of total outflows from the TSF and waste rock pile are shown. Gaging stations are maintained by U.S.
Geological Survey and Pebble Limited Partnership. SK = South Fork Koktuli; NK = North Fork Koktuli; UT Upper Talarik Creek.
NK100A •
NK100B
>•
UT100E
UT100C2
• UT100C1
SK119A • SK124A
i
SK100B1
SK100B
UT100C
•
UT100B
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-29
April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 10. Water flow schematic for the Pebble 6.5 scenario. Flows include water from the potentially acid generating (PAG) and non acid
generating (I""~ ' " " -—..^ ~ . ~< ,
treatment plant (WWTP) (solid black arrows), flow along the stream channels (solid blue arrows), and known groundwater transfers (dashed
blue arrow); for clarity, only flows greater than 5% of total outflows from the TSFs and waste rock piles are shown. Gaging stations are
maintained by U.S. Geological Survey and Pebble Limited Partnership. SK = South Fork Koktuli; NK = North Fork Koktuli; UT = Upper Talarik
Creek.
NK100B
UT100E
NK100A
UT100C2
• UT100C1
SK100B
UT100C
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-30
April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 11. Approximate locations of stream gages and wastewater treatment plant discharges represented in Figures 6 8 through 6 10.
Gages denoted with CP indicate confluence points, where virtual gages were created for analysis purposes. The Pebble 6.5 scenario footprint
is shown for reference. Data for the USGS gages based on Water Resources of Alaska (USGS 2012b); data for the PLP gages based on the
Environmental Baseline Document 2004 through 2008 (PLP 2011).
NK119CP1 __
' -NK100C
NK119B
Rivers & Streams (National Hydrography Datasel)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-31
April 2013
-------
Chapters Mine Scenarios
6.3 Closure and Post-Closure Site Management
As discussed in Section 4.2.4, the assessment examines potential impacts both during mine operations
and after mining activities have ceased, either as planned or prematurely. In this section, we consider
how the mine scenarios would be handled during and after closure of the mine.
We assume that the mine would be closed after all economically profitable ore is removed from the site,
leaving behind the mine pit, NAG waste rock piles and TSFs. Water at the site would require capture and
treatment for as long as it did not meet water quality standards. Weathering of the waste rock and pit
walls would release ions of potential concern, such as sulfates and metals. Weathering to the point
where these contaminants are present at levels approaching their pre-mining background
concentrations would likely take hundreds to thousands of years, resulting in a need for monitoring and
management of exposed mate rials andleachate over that time (Blight 2 010). We assume that existing
water management structures and the WWTP would be monitored and maintained as part of post-
closure operations.
Seepage and leachate monitoring and collection systems, as well as the WWTP, might need to be
maintained for hundreds to thousands of years. It is impossible to evaluate the success of such long-term
collection and treatment systems for mines—no examples exist, because these timeframes exceed both
existing systems and most human institutions. Throughout this section, we refer to the potential need
for treatment over extended periods. The uncertainty that human institutions have the stability to apply
treatment for these timeframes applies to all treatment options.
6.3.1 Mine Pit
Upon mine closure, pit dewatering pumps would be turned off. The cone of depression would persist
around the pit for a time, and groundwater would flow toward the pit in response to the local gradient.
Eventually, the water level in the pit would recover toward equilibrium with the surrounding water
table. Any water exiting the pit through surface channels or pumped from the pit would be tested (and
treated if necessary) prior to discharge to surrounding surface waters. Based on our calculations for
groundwater and precipitation inflows to the pit after operations have ceased, we estimate that the time
required for the pit to fill ranges from approximately 20 years for the Pebble 0.25 scenario to over
300 years for the Pebble 6.5 scenario.
Upper benches of the pit would be partially backfilled, regraded, covered with growth medium and
vegetated. Some areas may be flattened to enable construction of wetlands for passive water treatment.
At least portions of the pit walls, as well as rocks on the pit bottom or on side benches, would consist of
mineralized rock that was not economical to mine. Any exposed rock containing sulfide minerals would
likely be acid-generating for as long as it remained above the water surface in the pit, resulting in water
with low pH and dissolved metals running down the sides of the pit into the water body at the bottom.
As the water level in the pit rose, pit walls would become submerged and exposure to oxygen would be
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~ April 2013
-------
Chapters Mine Scenarios
reduced. Eventually, acid generation would be expected to cease from rocks below the water's oxic zone.
Exposed rock above the water surface or within the oxic zone would continue to produce acidic metal-
sulfate salts that would run into the pit lake with precipitation and snowmelt. Surfaces anticipated to
produce acidic drainage could be sealed against exposure to oxygen. However, this might not be
effective for a pit of this size: it might be difficult to seal all cracks and fissures present within walls.
There could be degradation of sealants from exposure to sun and air, and freeze-thaw fracturing of rock
could reduce acid-preventing efficacy over time. Predicting pit water quality has a high degree of
uncertainty (Section 8.1.3; Appendix I) (Gammons etal. 2009), but water would need to be monitored
and treated to meet effluent requirements prior to being discharged to streams, for as long as the water
remained contaminated.
6.3.2 Tailings Storage Facilities
At closure, tailings beaches in the TSFs would be covered with NAG waste rock and a growth medium,
then vegetated with native species (Ghaffari et al. 2011). Embankments and crests also would be
covered with a growth medium and vegetated. The tailings pond would be drawn down to prevent
flooding and to maintain stability, but a pond of sufficient depth would be retained to keep the core of
PAG tailings hydrated and isolated from oxidation. Retaining water in the tailings maintains a higher
potential for tailings dam failure than if the tailings were drained; however, draining the tailings to
stabilize them could allow oxygen-rich water to percolate through the tailings and oxidize the sulfides.
As long as a cover of water is maintained, oxygen movement into the tailings would be retarded,
minimizing acid generation. Drawing down the water level in the TSF would also provide capacity for
unusual precipitation events, reducing the likelihood that a storm would provide enough precipitation
to overwhelm capacity and cause tailings dam failure or overtopping. Additionally, wetlands might be
included in reclamation to provide additional stormwater retention, passive water treatment, and
significantly increased evapotranspiration (Reeve and Gracz 2008).
TSFs would require active management for hundreds to thousands of years (Blight 2010). A tailings dam
is an engineered structure that requires monitoring to ensure structural and operational integrity. An
assumption in the mining industry is that tailings continue to compact, expelling interstitial water and
becoming more stable over time. However, there appears to be little data available that document the
magnitude of this stability gain. A recent analysis suggests that densification of oil sands tailings may
stop after a period of time (Wells 2011). Although oil sands tailings are different from porphyry copper
tailings, the principle is the same. Lack of data specific to porphyry copper tailings suggests a cautious
approach, so we do not assume that tailings consolidate to a fully stable land form. Thus, the system may
require continued monitoring to ensure hydraulic and physical integrity in perpetuity.
6.3.3 Waste Rock
Some NAG waste rock would be used to cover tailings beaches, and some would be used to backfill
upper portions of the open pit. The remaining NAG waste rock would be sloped to a stable angle (e.g.,
less than 15 degrees [Blight and Fourie 2003]), covered with soil and plant-growth media, and vegetated
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g -o April 2013
-------
Chapter 6
Mine Scenarios
with native species. No PAG waste rock would remain on the surface, as it would have been processed
either as blending material during operations or at the end of operations.
6.3.4 Water Management
Table 6-8 summarizes the flow components of the water balance after closure, both during the period in
which the pit is filling and the steady state condition after the pit reaches its maximum level. During the
post-closure period, the mine would still capture water from precipitation over the mine pit, waste rock
piles, and the TSFs. Groundwater would continue to flow into the mine pit, so precipitation over the cone
of depression would continue to contribute to the captured water. Consumptive losses from operation
would cease, but water stored in the mine pit would constitute a new consumptive loss until the pit
water level reaches equilibrium with the surrounding groundwater level.
The footprint of the mine would be reduced as land occupied by production facilities is reclaimed. For
purposes of estimating water inputs, we assume that 80% of the areas disturbed by the plant and
ancillary facilities would be reclaimed, but that some facilities (e.g., the fuel depot, the WWTP, some
pipelines, and part of the camp) would remain.
Table 6 8. Summary of water balance flows (million m3/year) during post closure period for all mine
scenarios.
Flow Component
Captured at mine pit area
Captured atTSFl
Captured atTSF2
Captured atTSF3
Captured at mill & other facilities
Potable water supply well(s)
Water in ore (3%)
Total Captured
Cooling tower losses
Water in concentrate to port
Water in concentrate return
Runoff collected from port
Stored in TSF as pore water
Stored in mine pit
Crusher use
Total Consumptive Losses
Returned to streams via waste water treatment plant
Returned as NAG waste rock leachate
Returned as PAG waste rock leachate
Returned as TSF leakage
Total Reintroduced
Percent of captured water reintroduced
During Mine Pit Filling
41.0
12.2
17.2
6.62
0.447
0
0
77.5
0
0
0
0
0
41.0
0
41.0
28.3
0.947
0
719
36.5
47.1%
Post-Closure
24.2
12.2
17.2
6.62
0.447
0
0
60.7
0
0
0
0
0
0
0
0
52.5
0.947
0
719
60.6
100.0%
Notes:
TSF = tailings storage facility; NAG = non-acid-generating; PAG = potentially acid-generating.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-34
April 2013
-------
Chapters Mine Scenarios
As the mine pit fills, the cone of depression would shrink to the point that most or all of the waste rock
would be outside of the drawdown area. Runoff from the reclaimed NAG waste rock piles would either
seep into the ground, travel as overland flow, or be diverted to streams. Some precipitation would be
expected to infiltrate through the NAG waste rock cover, drain through the waste rock pile, and become
groundwater. Runoff from the reclaimed NAG waste rock piles is not anticipated to require treatment,
but would be periodically monitored to confirm this assumption.
The elevation of the north rim of the Pebble 6.5 pit would be over 100 m higher than the elevation of the
south rim, so that even when the mine pit reaches its maximum water level there would still be seepage
into the pit from the higher ground. For water balance purposes, we estimate that the post-closure cone
of depression would extend an average of 100 m beyond the pit rim as a result of surface outflow or
pumping.
Post-closure water management includes construction of wetlands for treatment of surface runoff,
retaining diversions, seepage/leachate capture systems operated during mining and at the WWTP, and
directing all surface water and water collected in seepage collection systems to the WWTP or to the pit.
Precipitation falling on the post-closure tailings would be monitored and diverted to the WWTP as
necessary to meet water quality standards. Otherwise, the water would be discharged to downstream
waters. Stormwater diversions and collection systems from the operations phase would be maintained
and water directed away from the TSF, or, if risk of contamination existed, toward the WWTP for
discharge to streams. Interstitial water within the tailings would continue to seep into naturally
fractured bedrock below the TSF. The well field placed downstream from the TSF during operations
would be retained and monitored post-closure, with water pumped and treated if determined to be
contaminated by leachate from the TSF. The pit water would be monitored and treated prior to being
released to streams, for as long as concentrations of contaminants exceeded effluent limits.
6.3.5 Premature Closure
Many mines close before their ore reserves are exhausted. In one study of international mine closures
between 1981 and 2009, 75% of the mines considered were closed before the mine plan was fully
implemented (Laurence 2011). The Illinois Creek and Nixon Fork mines are examples of mines that
closed prematurely in Alaska (although Nixon Fork has since re-opened).
Closure before originally planned—that is, premature closure—may occur for many reasons, including
technical issues, project fun ding, deteriorating markets, operational issues, or strategic financial issues
of the owner. Premature closures can range from cessation of mining with continued monitoring of the
site to complete abandonment of the site. As a result, environmental conditions at a prematurely closed
mine may be fully reclaimed or equivalent to those under a planned closure, may be severely
contaminated and require extensive remediation, or may fall anywhere between these extremes.
Environmental impacts associated with premature closure may be more significant than impacts
associated with planned closure, as mine facilities may not be at the end condition anticipated in the
closure plan and there may be uncertainty about future re-opening of the mine. For example, PAG waste
rock in our mine scenarios would likely still be on the surface in the event of a premature closure. If the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -c April 2013
-------
Chapters Mine Scenarios
mine closed because of a drop in commodity price, there would be little incentive to incur the cost of
moving or processing millions of metric tons of PAG waste rock. Because premature closure is an
unanticipated event, water treatment systems might be insufficient to treat the excessive and persistent
volume of lowpH water containing high metal concentrations. Some method of financial assurance
generally is required by state and federal agencies to ensure cleanup in the event a mine company
defaults on their responsibility (Box 4-3). In the past, however, financial assurance often has not been
adequate, and taxpayers have been left with substantial cleanup costs (USEPA 1997). This may be
changing, as agencies update bonding requirements to reflect cleanup costs more accurately, but
projecting these costs far into the future is a difficult task.
When a mine re-opens after premature closure, the owners might change the mining plan, implement
different mitigation practices, or negotiate new effluent permits. An example is the Gibraltar copper
mine in British Columbia. The Gibraltar mine began operations permitted as a zero-discharge operation.
However, when it was re-opened under new ownership after having closed prematurely, the new permit
allowed treated water to be discharged to the Fraser River with a 92-m dilution zone for copper and
other metals. On October 1, 2012, an Alaska Pollution Discharge Elimination System permit authorized
the Fort Knox Mine near Fairbanks, Alaska, to discharge wastewater to nearby Fish Creek. Although this
mine has never been closed, it was originally designed and permitted in 1994 as a no-discharge facility.
6.4 Conceptual Models
The development of conceptual models is a key component of the problem formulation stage of an
ecological risk assessment (USEPA 1998), and in Chapter 2 we introduced the use of conceptual models
as tools to help structure ecological risk assessments. At the outset, we broadly define the scope of this
assessment to be potential effects of a large-scale mine and a transportation corridor on freshwater
habitats, resulting effects on fish, and subsequent fish-mediated effects on wildlife and Alaska Native
populations (Section 2.2.1, Figure 2-1). To conduct a risk analysis, this scope needs to be refined and the
specific sources, stressors, and endpoints to be evaluated need to be explicitly identified.
In this section, we summarize the specific sources, stressors, and endpoints considered in the
assessment, as informed by the background information on the region, type of development, and
endpoints of interest presented in the preceding chapters, given the scenarios described in this chapter.
We then integrate these components into conceptual model diagrams that illustrate hypothesized cause-
effect linkages among them.
6.4.1 Sources Evaluated
The two main sources considered in the assessment are the mine and the transportation corridor, each
of which can be subdivided into several components. These components are summarized below, and
discussed in greater detail in Section 6.1.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g -g April 2013
-------
Chapters Mine Scenarios
• The mine infrastructure includes an open pit, potential underground mine, waste rock piles, TSFs,
water collection and storage facilities, a WWTP, ore-processing facilities, and chemical storage
facilities.
• The transportation corridor comprises a road and four pipelines (one each for product slurry, diesel
fuel, natural gas, and return water) connecting the mine site area to Cook Inlet.
6.4.2 Stressors Evaluated
As discussed above and in Chapter 4, large-scale mining is a complex process that typically involves both
physical alteration of the environment and the release of pollutants. The specific stressors considered
for inclusion in the assessment were identified based on their potential to significantly affect our
primary endpoint of interest—the region's salmon resources—and their relevance to the USEPA's
regulatory authority and decision-making context. Stakeholders also identified potential stressors of
concern, which were considered by the assessment team. These stressors are summarized in Table 6-9,
and discussed in detail below. Those stressors that are analyzed in the assessment or are of particular
concern to stakeholders are discussed in the following subsections.
6.4.2.1 Physical Habitat Alteration
Large-scale mining in the Bristol Bay region would necessarily involve the destruction of streams and
wetlands through excavation and filling associated with the mine pit, waste rock piles, tailings
impoundments, borrow pits, and the transportation corridor. This excavation and filling would directly
affect anadromous and resident salmonid habitats and directly involve USEPA under Section 404 of the
Clean Water Act.
Mining-related excavation and filling would also result in water diversion and withdrawal. Stream and
overland flow must be diverted around the mine site to keep it dry and minimize erosion; the mine pit
must be dewatered to continue excavation; and water must be obtained for use in ore processing,
tailings and product transport, and other purposes. These diversions and withdrawals would redirect
and reduce flow and plausibly affect fish via reduced habitat quality or quantity.
6.4.2.2 Water Temperature
Stream and wetland water temperatures could be affected via the capture, storage, use, treatment, and
discharge of water throughout the mining process. Elevated temperatures could result from warm water
discharges or, in summer, from reduced groundwater inputs. In winter, reduced groundwater inputs
could result in reduced temperatures. Because water temperature affects fish development and habitat,
any temperature changes could plausibly influence fish populations.
6.4.2.3 Chemical Contaminants
A range of chemical contaminants associated with mining may enter surface waters and pose risks to
fish. These contaminants include rock-derived inorganic contaminants (metals and pH), ore-processing
chemicals, fuels, and nitrogen compounds.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g 37 April 2013
-------
Chapter 6
Mine Scenarios
Table 6 9. Stressors considered in the assessment and their relevance to the assessment's primary
endpoint (salmonids) and USEPA's regulatory authority.
Stressor
Excavation
Filling
Water diversion
and withdrawal
Water
temperature
Product metal
(copper)
Other metals
PH
Process
chemicals
Nitrogen
Tailings and
other fine
sediment
Diesel fuel
Natural gas
Dust
Noise
Rock slide
Blocked or
perched culvert
Washed out
culvert
Invasive plants
Climate change
Description
Removal of streams and wetlands due to creation
of the mine pit and other excavations.
Filling in of streams and wetlands due to waste
rock piles, tailings impoundments, and roads.
Reduced flow in streams and wetlands due to
removal of water.
Changes in water temperature associated with
discharges of treated water or reduced
groundwater flows.
Copper occurring in the product concentrate,
waste rock, or tailings could enter streams and
wetlands.
Metals other than copper occurring in the product
concentrate, waste rock, or tailings could enter
streams and wetlands.
Oxidation of sulfides could result in acidification
of waste and receiving waters.
Chemicals used in ore processing would occur in
tailings and product concentrate and could spill.
Nitrogen compounds are released during blasting
and would deposit on the landscape.
Tailings, product concentrate, and other fine
particles could fill streams or wetland or, at lower
concentrations, could change substrate texture
and abrade fish gills.
Spilled diesel fuel could enter streams and
wetlands.
Leaking natural gas could combust.
Dust from blasting and vehicle traffic could
deposit on the landscape and be washed into
streams.
Noise from blasting or other activities.
Slides from waste rock piles or roads.
Inhibition offish passage due to malfunctioning
culverts.
Inhibition offish passage or downstream siltation
due to washed out culverts.
Changes in habitat quality due to invasion by
plants carried by road traffic.
Altered risk of mine failures, and changes in
marine and freshwater habitat quality and life
history timing, associated with increased
precipitation and temperature.
Relevance to
Salmonids
Relevant
Relevant
Relevant
Relevant
Relevant
Relevant
Relevant
Relevant
Weakly relevant
Relevant
Relevant
Not relevant
Weakly relevant
Not relevant
Relevant
Relevant
Relevant
Relevant
Indirectly relevant
Relevance to Decision-Making
Directly relevant to Section 404
of the Clean Water Act
Directly relevant to Section 404
of the Clean Water Act
Consequence of excavation and
filling
Consequence of excavation and
filling
Consequence of excavation and
filling
Consequence of excavation and
filling
Consequence of excavation and
filling
Consequence of excavation and
filling
Consequence of excavation
Directly relevant to Section 404
of the Clean Water Act (if
particles act as fill) and
consequence of excavation and
filling
Peripheral to excavation and
filling
Peripheral to excavation and
filling
Consequence of excavation and
filling
Peripheral to excavation and
filling
Consequence of excavation and
filling
Consequence of excavation and
filling for a road
Consequence of excavation and
filling for a road
Peripheral to excavation and
filling
Not related to excavation and
filling, but modifies other
consequences of excavation
and filling
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-38
April 2013
-------
Chapter 6
Mine Scenarios
Rock-Derived Inorganic Contaminants
Mines are developed because rocks at the site have high metal concentrations, which are further
concentrated as ore is isolated from waste rock and as product concentrate is created from the ore.
These metals may enter surface waters from uncollected leachate and runoff, from WWTPs, or from
spills of product concentrate and its associated water. Metals are known to cause toxic effects on aquatic
biota, including fish; however, when combined with low pH (acidity), metals become especially
problematic. Acid rock drainage occurs when PAG rocks are present at the mine site. Acidity can be
directly deleterious to aquatic biota, but it also increases metal concentrations in solution.
Because copper is the major resource metal in the Pebble deposit and is particularly toxic to aquatic
organisms, it is the metal most likely to cause toxic effects at this site. Copper toxicity also has been a
primary concern of stakeholders, including the National Oceanic and Atmospheric Administration, the
federal agency responsible for salmon management. Thus, copper criteria, standards, and toxicity are
considered in detail in the assessment.
Other metals are considered if their concentrations in test leachates from the Pebble deposit indicate
that they are potentially toxic, based on benchmark values. When possible, national ambient water
quality criteria are used as screening benchmarks. Both criterion maximum concentrations (CMCs) and
criterion continuous concentrations (CCCs) are used to account for acute and chronic exposures,
respectively. When U.S. criteria are not available, the most similar available value is used (e.g., Canadian
benchmarks, the lowest acute and chronic values from the USEPA's ECOTOX database, or the European
Chemical Agency and Organization for Economic Cooperation and Development's eChemPortal)
(Table 6-10). Some metals, such calcium, magnesium, and sodium, are not screened because of their low
toxicity.
Table 6 10. Screening benchmarks for metals with no national ambient water quality criteria.
Metal
B
Ba
Co
Fea
Mn
Mo
Sb
Acute/Chronic Benchmarks
(Mg/L)
29,000/1,500
46,000/8,900
89/2.5
350/-
760/693
32,000/72
14,400/1,600
Source and Notes
Canadian acute and chronic guidelines based on SSDs (CCME 2009)
Austroptamobius pallipes 96 hr LCso (Boutet and Chaisemartin 1973) and Daphnia
magna 21 day reproductive ECso (Biesinger and Christensen 1972)
Acute value is the lowest acute test datum and the chronic value is the 5th centile
of a chronic species sensitivity distribution (Environment Canada and Health
Canada 2011)
Chronic data were inadequate to set a value, but the Canadian authors believed
that it would not be much lower than this acute value (BC 2008)
Hardness adjusted (for 20 mg/L) acute and chronic guidelines (BC 2001)
Daphnia magna 48 hr LCso (Kimball 1978) and Canadian chronic guideline (CCME
1999)
Lowest acute and chronic values from a fathead minnow early life-stage test
(USEPA 1980, Swedish Chemicals Inspectorate 2008, Environment Canada and
Health Canada 2010)
Notes:
a The listed U.S. iron criterion, from the 1976 Red Book (USEPA 1976), is less reliable than this more recent benchmark.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-39
April 2013
-------
Chapters Mine Scenarios
Screening against tailings and waste rock leachates are presented in Tables 8-4 through 8-8. The metals
of concern are aluminum, cadmium, cobalt, copper, manganese, nickel, lead, selenium, and zinc based on
the average concentration exceeding either the acute or chronic benchmark for at least one leachate.
However, most of the estimated total toxicity is due to Cu.
Major Ions (Total Dissolved Solids)
Some metals such as calcium, magnesium, potassium, and sodium are not screened because of their low
toxicity, but they contribute to ionic stress. Mining inevitably involves crushing rocks and the leaching of
crushed rock results in relatively rapid dissolution resulting in elevated concentrations of dissolved
major ions (calcium, magnesium, sodium, potassium, chlorine, sulfate, and bicarbonate). This ionic
mixture is measured as total dissolved solids (TDS) or specific conductance (conductivity). Even if this
mixture is not acidic, it can be toxic to aquatic biota, particularly in the extremely dilute waters that
occur in this region. Examples of toxicity due to leaching of major ions from mine-derived waste rock are
discussed in USEPA2011 and Chapman etal. 2000. Also, the history of TDS compliance problems at the
Red Dog Mine near Kotzebue, Alaska, suggests that dissolved major ions should be a stressor of concern.
Ore-Processing Chemicals
Chemicals used to process the ore and separate the product from the tailings have a potential to enter
the environment as a result of truck wrecks, spills on site, tailings slurry spills, product concentrate
slurry spills, or failures of water collection and treatment. Tests of the Pebble deposit ore used alkaline
flotation to separate the product concentrate from the tailings (Ghaffari et al. 2011). The collector was
sodium ethyl xanthate, the frother was methyl isobutyl carbinol, and lime was used to adjust pH.
Molybdenum separation also requires fuel oil as a collector. Of these, the xanthate is clearly a
contaminant of concern because it is highly toxic to aquatic life (Hidalgo and Gutz 2001). Methyl isobutyl
carbinol has been poorly tested but appears to have relatively low toxicity (acute lethality to African
clawed frogs and goldfish at 360 to 656 mg/L from ECOTOX). Lime would contribute to the risk from
major ions (TDS). Fuel oil use for this purpose would be small relative to its use as fuel.
In addition, cyanide might be used to recover gold from pyritic tailings. Although it is likely that the
tailings slurry from this process would exceed the acute and chronic water quality criteria of 22 and
5.2 ug free cyanide per liter, it is likely to be rapidly diluted and degraded in the TSF. Accidental releases
and on-site spills, as recently occurred at the Fort Knox mine (ADEC 2012), are possible but are not
judged to be as directly significant to our endpoints as other accidents considered.
Fuels
Both diesel oil and natural gas would be piped to the mine site and could enter the environment via
pipeline leaks or failures. Diesel spills could enter surface waters, and have been known to adversely
affect aquatic biota, so diesel is considered in the assessment. Natural gas could combust, but a natural
gas fire is unlikely to significantly affect salmon populations.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g . „ April 2013
-------
Chapters Mine Scenarios
Nitrogen Compounds
Nitrogen compounds would be released during the blasting associated with excavation and some of
these compounds would deposit on the landscape and could enter surface water and groundwater. It is
likely these streams are phosphorus-limited, not nitrogen-limited (Moore and Schindler 2004, Goldman
1960). We know of no studies of nitrogen deposited at mine sites, and the consequences of a change in
nitrogen/phosphorus ratio for salmonids are unknown and but judged to be minimal. Thus, nitrogen
deposition is not considered in the assessment.
6.4.2.4 Fine Sediment
If tailings, product concentrate, unpaved road materials, or other fine particles are spilled or eroded,
they could fill streams and wetlands, alter streambed substrates, or abrade the gills offish.
6.4.2.5 Dust
Blasting and vehicle traffic, both at the mine site and along the transportation corridor, would generate
dust. This dust could contribute to the sedimentation of streams and, depending on the composition of
the rock, could contribute toxic metals to surface waters. Dust from unpaved roads is known to affect
streams, so it is included in this assessment. In contrast, the occurrence of dust from blasting is poorly
documented and its effects are unknown. We anticipate that much of the dust generated by blasting will
settle on the site and be collected with runoff water. Wind may carry dust off site, but would also
disperse it across the landscape. Thus, we do not judge dust from blasting to be an important
contributor to risks to salmonids (although this judgment is highly uncertain), and do not consider it in
the assessment.
6.4.2.6 Noise
Noise would be generated by blasting at the mine site and vehicle traffic along the transportation
corridor. Although noise may directly affect wildlife, it is unlikely to affect salmonids and is not
considered in the assessment.
6.4.2.7 Culverts
Blocked or perched culverts could significantly reduce fish passage, thereby reducing salmon migrations
or movement among habitats by resident salmonids. Culverts also may wash out during floods,
temporarily inhibiting fish movement and reducing habitat due to siltation by the deposited roadbed
material. Culverts are a component of roads that fill wetlands and the floodplains of streams. They may
significantly affect salmon in the surface waters they intersect and thus are considered in the
assessment.
6.4.2.8 Invasive Species
Several dozen species of plants, animals, and micro-organisms are considered to be or have the potential
to be invasive in Alaska (Eddmaps 2013, ADF&G 2013). Of those currently present, reed canarygrass
[Phalaris arundinacea) is wide spread on the Kenai Peninsula (HSWCD 2007) and elodea [Elodea
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment JT7 April 2013
-------
Chapters Mine Scenarios
canadensis) exist in Stormy Lake on the northern Kenai Peninsula (Etcheverry 2012). These plants have
the potential to degrade salmon habitat (Merz et al. 2008). The improved and expanded road from Cook
Inlet may facilitate the spread of reed canary grass and elodea from the Kenai Peninsula to the Bristol
Bay watershed where they may adversely affect salmon habitat.
6.4.3 Endpoints Evaluated
In this assessment, the primary endpoint of interest is the region's salmonid populations (Pacific
salmon, rainbow trout, and Dolly Varden) in terms of abundance, productivity, or diversity. Given the
importance of salmonids to the region's ecosystems and culture, we also consider the effects of potential
changes in fish populations on wildlife abundance, productivity, or diversity and on Alaska Native
culture. These endpoints are discussed in detail in Chapter 5.
6.4.4 Conceptual Model Diagrams
To frame the assessment, we developed conceptual model diagrams illustrating potential pathways
linking the sources, stressors, and endpoints detailed above (see Box 2-2 for an overview of how the
assessment's conceptual models are structured). These diagrams went through several iterations, from
initial brainstorming of all potential pathways associated with large-scale mining development in the
Bristol Bay region (both with the assessment team and other stakeholders) to focusing on those
pathways considered both within the assessment's scope (Chapter 2) and likely to affect endpoints of
interest
Through this iterative process, we developed a series of three conceptual model diagrams illustrating
hypothesized cause-effect relationships leading from mine-related sources to endpoints of interest.
These diagrams illustrate potential effects of routine mine construction and operation on physical
habitat (Figure 6-12), potential effects of routine mine construction and operation on water chemistry
(Figure 6-13), and potential effects of unplanned events on physical habitat and water chemistry (Figure
6-14). These diagrams provide a framework for the analysis sections of the assessment, and the relevant
portions of these diagrams evaluated in each analysis section are highlighted throughout the Risk
Analysis and Characterization portion of the assessment. Note that not all pathways included in each
conceptual diagram are necessarily evaluated in the assessment. For example, in some cases, we
hypothesized pathways that may be significant, but data were not sufficient for quantitative analysis.
We also developed three more general conceptual model diagrams for specific topics that were defined
as outside of the assessment's scope, but are of key importance to stakeholders (Chapters 12 and 13).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment g . „ APril 2013
-------
Chapter 6
Mine Scenarios
Figure 6 12. Conceptual model illustrating potential effects of routine mine construction and operation on physical habitat.
mine
[ transportation corridor ]
waste rock ] \ tailings storage
piles ) I facilities
ore processing
facilities
additional step in
causal path'
epm)
way' J
4< salmon
(abundance, productivity or dive rsity)
4, other fish
(abundance, productivity or diversity)
1 marine-derived
nutrients
4' ecosystem
productivity
/ater collection c.
storage systems
.'•.•ater treatment \
facilities J
T water
discharges
T channel erosion
entrenchment
I materia
transport
riparian
vegetation
4 ground water -
surface •.'•.• ater interactions
non-native species
f r e q u e n c y of h i gh f I o •/.' s
4 macroinvertebrate
prey
,1 production ancl export of
food & other resources
T competition
&predation
I inhibition of
fish passage
a riparian
vegetation
1 aquatic habitat
fragmentation
fl migration
patterns
I over winter ing habitat
(qualitv or quantity)
J rear ing habitat
quality or quantity!
4 spawning habitat
(quality or quantity:
ncubation habitat
(quality or quantity)
additional step in
causal pathw ay
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-43
April 2013
-------
Chapter 6
Mine Scenarios
Figure 6 13. Conceptual model illustrating potential effects of routine mine construction and operation on water chemistry.
transportation corridor |
natural gas| (return water |
| tailings storage I , openpit
facilities J L. j { cave
ore processing
facilities
-------
Chapter 6
Mine Scenarios
Figure 6 14. Conceptual model illustrating potential effects of unplanned events on physical habitat and water chemistry.
fj
v
"i r
culvert blockage
or perching
\/
channel erosion
& entrenchment
\/
A stream 4, f
geomorphology cot
\t V
[t inhibition of
fish passage
V
f transportation corridor
I I f diese ] ['natural gas | [ rel
) I pipeline 1 1 pipeline I 1
V V
roadbed chemical or fuel spill pipelit
wash out duringtransport or
V V
nectvity sediment
\[/
4 feeding
ability
V
I
| f mine J
i
urn water ] f slurry | 1 I ta
Jipeline ) [pipeline J k^
/ 1 '
V N
le breaks tailing
leaks I fail
X
V
-v, f fine hprl
sediment
1
\f
I macro invertebrate
prey
V
T physiological
stress
V
•|. rearing habitat ,, spawning habitat 4- overwintering habitat 4 incubation labitat
(quality or quantity) (quality or quantity) (quality or quantity) I (quality or quantity)
LEGEND
( source J
—
additional step in addi
causal pathway cau
roximate /- — bTotic^X
stressor I response }
doialstepn modifying
sal pathway1 /actor
linir^; d-r.rapp ] f block I fy^ ^••a«t'a rock 1 T , » 1 f nrf> prnr^ <:*;inp ]
facilities J kcaveJk F'iles H<>penp«l ( facilities J f IZZIZZ
| | | / \ c// mo f e ^ ^
V V V '
K-,ater collection &X| >[<', ate r treatment 8>1 I7 chemical storage^ [ hydrologicevent
'L7oo5''ic?e/L7|,'LV'L'.'''":^f
< v ^
l! ^ \/ \/ \/ "^ /\ f 7"T "]
5 dam " "'if "1 " ~\ engineering failure
ure failure of vvater collection failure of -.-vater chemical
J ^storage systems treatment facilities spill
k, ) \. J \. J hum "'•"' --nrpr
|
\/ \/ """" 1 [ /.re
f • >
contaminated water
discharges
H stage of mining
operations
(copper, cobalt) "^ /\ (copper, cobalt)
1
V V V
A nth^r .'.stvr "-., ^1 £ m^t^l ^pa"i^ti'"-n 1 '"•rpani'' "t norg^nir
^ rsspons? f-'^f
\/ \/ \,' \/ f^r ~~TZ~ , f
If f hinari-iimiilst-i.-in ft "X -| *-Ks'JJ 1 'sf t rhrnnir | | | a^ijt" 1 ^ l/OCi'or)Oerf evert
\
/ \/
,^*^*" 4,samon ~"^-x -^" ^ other fish ~^~^v
V,^ (abundance, productivity or diversity) *} v^ (abundance, productivity or diversit*,1) ^y
-T— | A
± 1 1
[^ 1 ~1
nutrients | productivity |
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
6-45
April 2013
-------
Chapter 6
Mine Scenarios
This page intentionally left blank
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
April 2013
-------
This chapter addresses the stream habitat and flow risks associated with routine operations of the mine
scenarios described in Chapter 6. It considers the unavoidable environmental effects associated with the
footprint of each mine scenario, in the absence of failures of water containment or treatment facilities,
tailings storage facilities (TSFs), transportation corridor, or pipelines. This is not meant to suggest that
the absence of failures is a realistic possibility, because accidents and failures always happen in complex
and long-lasting operations. The risk and potential impact of failure of these components are described
in Chapters 8, 9,10, and 11. This chapter serves to separate the inevitable effects of the mine scenarios
from those that are the result of unintended failures.
Potential pathways linking mine components, stream habitat and flow alterations, and biotic responses
are highlighted in Figure 7-1. Key stressors associated with routine mine development and operations
include elimination and modification of habitat (Section 7.2) and changes in downstream flow
(Section 7.3), both of which can affect fish populations. Routine effects of water treatment and discharge
and the transportation corridor are discussed in Chapters 8 and 10, respectively.
7.1 Abundance and Distribution of Fishes in the Mine
Scenario Watersheds
The potential effects of the mine footprint (addressed in this chapter) and of routine mine operations
and failures (Chapters 8 through 11) on the assessment endpoints depend on the abundance and
distribution of salmonid fishes in the potentially affected streams and rivers.
7.1.1 Fish Distribution
The three watersheds draining the mine scenario footprints—the South Fork Koktuli River, North Fork
Koktuli River, and Upper Talarik Creek watersheds (hereafter referred to as the mine scenario
watersheds)—have been sampled extensively for summer fish distribution over several years. These
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-1
April 2013
-------
Chapter 7 Mine Footprint
data, collected by the Alaska Department of Fish and Game (ADF&G) and various consultants and non-
profits, are captured in the Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes—Southwestern Region (also known as the Anadromous Waters Catalog [AWC])
(Johnson and Blanche 2012) and the Alaska Freshwater Fish Inventory (AFFI) (ADF&G 2012). The AWC
provides the State of Alaska's official record of anadromous fish distribution and, if available, life stages
present (categorized as spawning, rearing, or present but life stage unspecified) interpreted by
individual stream reaches. The AFFI includes all fish species, including resident fishes, found at specific
sampling points. The catalogued distributions of the five Pacific salmon species (sockeye, coho, Chinook,
chum, and pink), Dolly Varden, and resident rainbow trout in the mine scenario watersheds are shown
in Figures 7-2 through 7-8. In addition, Arctic grayling, slimy sculpin, northern pike, ninespine
stickleback, threespine stickleback, Alaskan or Arctic brook lamprey, burbot, round whitefish, humpback
whitefish, least cisco, and longnose sucker occur in these watersheds (ADF&G 2012). These species are
discussed in more detail in Appendix B of this assessment. AWC stream reach designations and AFFI
observation points should be interpreted with care because not all streams could be sampled, and there
are potential errors associated with fish identification or mapping; caveats and uncertainties concerning
interpretation of AWC and AFFI data are discussed in Section 7.2.5.
Sockeye salmon use mainstem reaches of the mine scenario watersheds for spawning and rearing,
including a portion of Upper Talarik Creek that is within the waste rock footprints of the Pebble 2.0 and
Pebble 6.5 scenarios (Figure 7-2). Coho salmon have the most widespread distribution of the five
salmon species in the mine scenario watersheds, making extensive use of mainstem and tributary
habitats (Figure 7-3). Coho rear in many of the headwater streams that would be eliminated, blocked, or
dewatered by TSF 1, TSF 2, and TSF 3 under the Pebble 2.0 and Pebble 6.5 scenarios (Figure 7-3).
Chinook salmon have been documented throughout mainstem reaches of the mine scenario watersheds
(Figure 7-4). Chinook are known to use small streams for rearing habitat, and juveniles have been
observed in streams that would be buried under the TSF 1 (North Fork Koktuli River), TSF 2 (South Fork
Koktuli River), and waste rock pile (Upper Talarik Creek) footprints (Figure 7-4). The distributions of
chum and pink salmon are generally restricted to mainstem reaches where spawning and migration
occur. Chum salmon have been found in all three mine scenario watersheds and in the stream under the
footprint of TSF 2 (Figure 7-5). Pink salmon have only been reported at very low numbers in the lowest
section of Upper Talarik Creek and in the Koktuli River below the confluence of the north and south
forks (Figure 7-6, Figure 5-5). Dolly Varden are found even further upstream than coho salmon, and fish
surveys indicate that they are commonly found in the smallest streams (i.e., first-order tributaries)
throughout all three mine scenario watersheds, including in streams under each of the TSF footprints
(Figure 7-7). Rainbow trout have been collected at many mainstem locations, especially in Upper Talarik
Creek, and their reported distribution extends upstream throughout the TSF 1 footprint and in the
portions of Upper Talarik Creek in the waste rock footprint (Figure 7-8).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, » April 2013
-------
Chapter 7
Mine Footprint
Figure 7 1. Conceptual model illustrating potential linkages between sources associated with the mine scenario footprints, changes in physical habitat, and fish endpoints.
open ] [ waste rock ] \ tailings storage |
pit ) { piles J I facilities J
:-re processing
facilities
.vater collection &.
storage systems
.'.ateI treatment
facilities
1 leakage
of leachate
T water
discharges
Agrounch.vater -
surface ^vater interactions
dovvnstre an •.'. ate r to-:.;
4 ao ivnstr e am wate r temperatuf es
A magnitude &
f r e q LI e n c y of h i gh f I o ^ s
T alteration of channel
morphology & floodplain
connectivity
4 production and export of
foodS. other resources
I aquatic habitat
fragmentation
i migration
patterns
oyer-A'intering habitat
(quality or quantity}
rearing habitat
(quality or quantity!
4 spav/ning habitat
!quality or quantity}
I incubation habitat
(quality or quantity)
i salmon
u n d an c e. p r o d u cti •; ity o r d i -.• e r sity)
I other fish
(abundance, p r o d u cti •; ity o r d i v e r sity)
marine-derived
nutrients
ecos'p-stem
productivity
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-3
April 2013
-------
Chapter 7
Mine Footprint
This page intentionally left blank
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, . April 2013
-------
Chapter 7
Mine Footprint
Figure 7 2. Reported sockeye salmon distribution in the mine scenario watersheds. "Present"
indicates species was present but life stage use was not determined; "spawning" indicates spawning
adults were observed; "rearing" indicates juveniles were observed. Present, spawning, and rearing
designations are based on the Anadromous Waters Catalog (Johnson and Blanche 2012). Life stage
specific reach designations are likely underestimates, given the challenges inherent in surveying all
streams that may support life stage use throughout the year. See Section?.2.5 for additional notes on
interpretation offish distribution data.
SOUTH FORK KOKTULI
Present
Spawning
Rearing
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-5
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 3. Reported coho salmon distribution in the mine scenario watersheds. "Present" indicates
species was present but life stage use was not determined; "spawning" indicates spawning adults
were observed; "rearing" indicates juveniles were observed. Present, spawning, and rearing
designations are based on the Anadromous Waters Catalog (Johnson and Blanche 2012). Life stage
specific reach designations are likely underestimates, given the challenges inherent in surveying all
streams that may support life stage use throughout the year. See Section 7.2.5 for additional notes
on interpretation offish distribution data.
Present
Spawning
Rearing
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-6
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 4. Reported Chinook salmon distribution in the mine scenario watersheds. "Present"
indicates species was present but life stage use was not determined; "spawning" indicates spawning
adults were observed; "rearing" indicates juveniles were observed. Present, spawning, and rearing
designations are based on the Anadromous Waters Catalog (Johnson and Blanche 2012). Life stage
specific reach designations are likely underestimates, given the challenges inherent in surveying all
streams that may support life stage use throughout the year. See Section 7.2.5 for additional notes
on interpretation offish distribution data.
Present
Spawning
Rearing
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-7
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 5. Reported chum salmon distribution in the mine scenario watersheds. "Present"
indicates species was present but life stage use was not determined; "spawning" indicates spawning
adults were observed; "rearing" indicates juveniles were observed. Present, spawning, and rearing
designations are based on the Anadromous Waters Catalog (Johnson and Blanche 2012). Life stage
specific reach designations are likely underestimates, given the challenges inherent in surveying all
streams that may support life stage use throughout the year. See Section 7.2.5 for additional notes
on interpretation offish distribution data.
Present
Spawning
Rearing
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
I I Watershed Boundary
N
A
0 2.5 5
Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-8
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 6. Reported pink salmon distribution in the mine scenario watersheds. "Present" indicates
species was present but life stage use was not determined; "spawning" indicates spawning adults
were observed. Present and spawning designations are based on the Anadromous Waters Catalog
(Johnson and Blanche 2012). Life stage specific reach designations are likely underestimates, given
the challenges inherent in surveying all streams that may support life stage use throughout the year.
See Section 7.2.5 for additional notes on interpretation of fish distribution data.
NORTH FORK KO
UPPER TALARIK
AGAK
,-J
,-
&-N'
r River
J
^
r
$-C
Present
Spawning
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
Watershed Boundary
Iliamna Lake
NUSHAGAK
KVICHAKd,
N
0 2.5 5
2.5 5
]Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-9
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 7. Reported Dolly Varden occurrence in the mine scenario watersheds. Designation of
species presence is based on the Alaska Freshwater Fish Inventory (ADF&G 2012). Absence cannot
be inferred from this map; see Section 7.2.5 for additional notes on interpretation of fish distribution
data.
SOUTH FORK KOKTULI
Present (AFFI)
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-10
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 8. Reported rainbow trout occurrence in the mine scenario watersheds. Designation of
species presence is based on the Alaska Freshwater Fish Inventory (ADF&G 2012). Absence cannot
be inferred from this map; see Section 7.2.5 for additional notes on interpretation of fish distribution
data.
Present (AFFI)
Pebble 0.25 Footprint
Pebble 2.0 Footprint
Pebble 6.5 Footprint
Mine Scenario Watersheds
Watershed Boundary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-11
April 2013
-------
Chapter 7 Mine Footprint
7.1.2 Spawning Salmon Abundance
Index estimates of relative spawner abundance are available for chum, sockeye, coho, and Chinook
salmon in the mine scenario watersheds. Aerial index counts of spawning salmon are available from the
ADF&G and the Pebble Limited Partnership (PLP). This type of survey is primarily used as an index to
track variation in run size over time. We recognize that survey values tend to underestimate true
abundance for two reasons: an observer in an aircraft is not able to count all fish in dense aggregations,
and only a fraction of the fish that spawn at a given site are present at any one time (Bue et al. 1988,
Jones et al. 2007). Weather, water clarity, and other factors influencing fish visibility can also contribute
to underestimates. Additionally, surveys intended to capture peak abundance may not always do so. For
example, aerial surveys counted, on average, only 44% of the pink salmon counted by surveyors walking
the same Prince William Sound spawning streams (n = 18) (Bue et al. 1988). Peak aerial counts of pink
salmon in southeastern Alaska are routinely multiplied by 2.5 to represent more accurately the number
offish present at the time of the survey (Jones etal. 2007). Helicopter surveys of Chinook salmon on the
Kenai Peninsula's Anchor River over 5 years counted only 5 to 10% of the fish counted by a concurrent
sonar/weir counting station (Szarzi et al. 2007).
ADF&G conducts aerial index counts that target peak spawning periods of sockeye salmon on Upper
Talarik Creek and Chinook salmon on the Koktuli River. Sockeye salmon counts have been conducted
most years since 1955 (Morstad 2003), and Chinook salmon counts most years since 1967 (Dye and
Schwanke 2009). Between 1955 and 2011, sockeye salmon counts in Upper Talarik Creek have ranged
from 0 to 70,600, with an average of 7,021 over 49 count periods (Morstad pers. comm.) Between 1967
and 2009, Chinook salmon counts in the Koktuli River ranged from 240 to 10,620, with an average of
3,828 over 29 count periods (Dye and Schwanke 2009).
The Environmental Baseline Document 2004 through 2008 (EBD) (PLP 2011) provides aerial index
counts for Chinook, chum, coho, and sockeye salmon in the mine scenario watersheds from 2004 to
2008. Surveys on the South and North Fork Koktuli Rivers began at the confluence and extended upward
to the intermittent reach or Frying Pan Lake on the South Fork Koktuli River and upward to Big Wiggly
Lake or river kilometer 56 on the North Fork Koktuli River. Surveys on Upper Talarik Creek ran from the
mouth and extended upstream to Tributary 1.350 (just east of Koktuli Mountain) or to the headwaters.
Multiple counts were usually made for each stream and species in a given year (Table 7-1). Repeat
surveys of this type can be used to estimate the size of spawning populations if estimates of stream life
(i.e., the number of days that salmon are present on the spawning grounds) and observer efficiency are
available; however, PLP was unable to make reliable estimates of stream life and observer efficiency
(PLP 2011: 15.1-14). The average of each year's index counts is reported as an abundance index for each
population. We instead report the highest of each year's index counts for each population (Table 7-1;
approximated from figures in PLP 2011), because only a portion of the spawning population is present
on the spawning grounds on any given day, and therefore the highest index count is mathematically
closer to the true abundance than is the average of multiple surveys. Further, the averaged estimates
reported in PLP 2011 are often pulled downward by counts outside of the spawning period when no fish
were counted (PLP 2011: Figure 15.1-93). Finally, using the highest index count more closely matches
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~±2 April 2013
-------
Chapter 7
Mine Footprint
ADF&G's index methods that are based on a single count targeting peak spawning. The highest peak
index counts for coho and sockeye salmon were in Upper Talarik Creek, and the highest counts for
Chinook and chum salmon were in the South and North Fork Koktuli Rivers (Table 7-1). The overall
highest count was for sockeye salmon in Upper Talarik Creek and Tributary 1.60 in 2008, when
approximately 82,000 fish were estimated (Table 7-1).
Table 7 1. Highest reported index spawner counts in the mine scenario watersheds for each year,
2004 to 2008.
Mine Scenario Watershed
South Fork Koktuli River
North Fork Koktuli River
Upper Talarik Creek
Salmon
Species
Chinook
Chum
Coho
Sockeye
Chinook
Chum
Coho
Sockeye
Chinook
Chum
Coho
Sockeye
Highest Index Spawner Count Per Year
(Number Of Counts)3
2004
2,750 (3)
0
250 (2)
1,400 (2)
2,800 (3)
400 (1)
300 (3)
550 (2)
275 (2)
0
3,000 (4)
33,000 (2)
2005
1,500 (4)
350 (4)
550 (4)
2,000 (5)
2,900 (4)
350 (4)
350 (1)
1,100 (5)
100 (3)
3(1)
0
15,000 (4)
2006
250 (5)
850 (7)
1,375 (3)
2,700 (8)
750 (4)
750 (4)
1,050 (4)
1,400 (7)
80(3)
13(2)
6,300 (3)
10,000 (6)
2007
300 (8)
200 (11)
250 (10)
4,000 (11)
600 (8)
800 (9)
125 (8)
2,200 (10)
150 (9)
8(8)
4,400 (9)
10,000 (14)
2008
500 (9)
950 (7)
1,875 (20)
6,000 (13)
500 (8)
1,400 (7)
1,700 (15)
2,000 (12)
100 (8)
18(5)
6,300 (14)b
82,000 (14)b
Notes:
a Values likely underestimate true spawner abundance.
b Tributary 1.60, a major tributary to Upper Talarik Creek, was included in this count.
Source: PLP 2011.
The spatial distribution of spawner counts in the study streams during the 2008 return year was
provided by PLP (2011). Spawner counts were summarized by individual stream reaches throughout
the mainstem of each of the mine scenario watersheds. Data were reported for three reaches in the
South Fork Koktuli River (A through C, extending from the confluence upstream to the intermittent
reach), five reaches in the North Fork Koktuli River (A through E, extending from the confluence
upstream to beyond Big Wiggly Lake), and seven reaches in Upper Talarik Creek (A through G, extending
from the mouth to the headwaters) (Figure 15.1-2 in PLP 2011 illustrates the stream reaches). Count
data (approximated from figures in PLP 2011) for each of these reaches are given in Table 7-2, along
with the location of each stream reach (in river kilometers), to demonstrate the relative spatial
distribution of salmon during the 2008 spawning period.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-13
April 2013
-------
Chapter 7
Mine Footprint
Table 7 2. Average 2008 index spawner counts by stream reach3.
Stream
South Fork
Koktuli
River
North Fork
Koktuli
River
Upper
Talarik
Creek
Salmon
Species
Reach
Boundaries
(river km)
Chinook
Chum
Coho
Sockeye
Reach
Boundaries
(river km)
Chinook
Chum
Coho
Sockeye
Reach
Boundaries
(river km)
Chinook
Chum
Coho
Sockeye
Stream Reach, Downstream to Upstream
A
0-24.9
200
90
200
800
0-13.7
110
50
100
530
0-5.9
<10
<10
100
10,000
B
24.9-34.3
70
190
250
1,510
13.7-21.1
40
50
70
<10
5.9-16.8
<10
<10
50
4,500
C
34.3-51.7
0
0
8
1
21.1-36.6
50
320
210
220
16.8-24.8
20
<10
40
3,000
D
-
-
36.6-48.4
0
0
30
60
24.8-36.3
<10
<10
180
3,000
E
-
-
48.4-52.5
0
0
60
0
36.3-45.1
20
<10
280
500
F
-
-
-
-
45.1-59.1
<10
10
180
47
G
-
-
-
-
59.1-62.4
0
0
<10
0
Notes:
Blank values (-) indicate no applicable stream reach.
3 Values likely underestimate true spawner abundance.
Source: PLP 2011.
7.1.3 Juvenile Salmon and Other Salmonid Abundance
PLP (2011) reports counts of juvenile salmon and other salmonids in the South and North Fork Koktuli
Rivers and Upper Talarik Creek based on extensive sampling efforts from 2004 through 2008. Snorkel
surveys were the primary data collection method, but electrofishing, minnow traps, beach seines, gill
nets, angling, and dip netting were used in certain situations. It is not always possible to determine
which survey methods generated which counts. Raw field counts were expressed as densities (count per
100-m reach was the only unit reported for all three streams). These counts should not be viewed as
quantitative abundance estimates; they are very likely underestimated because of the extreme difficulty
of observing or capturing all fish in complex habitats (Hillman et al. 1992). Other methods are available
to generate density estimates with confidence bounds (e.g., mark-recapture or depletion estimates) but
are much more time-consuming or labor-intensive.
Reported fish densities (averaged over the 4 years) vary widely by stream, sample reach, and habitat
type (PLP 2011: Figures 15.1-23,15.1-52, and 15.1-82). Species that attain densities of several hundred
per 100-m reach in one setting were often absent or sparse in other habitat types or reaches in the same
stream, which is typical for fish in heterogeneous stream environments. Table 7-3 presents maximum
fish densities in the mainstem of each mine scenario watershed, approximated from figures in the EBD
(PLP 2011), for species that rear for extended periods in the surveyed streams and for which data are
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-14
April 2013
-------
Chapter 7
Mine Footprint
available: Chinook and coho salmon, Arctic grayling, and Dolly Varden. We report maximum density to
give a sense of the magnitude attained in the surveyed streams, but it should be stressed that abundance
varied widely by stream reach and habitat type within a given stream (PLP 2011: Figures 15.1-23,15.1-
52, and 15.1-82). Highest reported densities were approximately 2,500 Arctic grayling and 1,600 coho
salmon per 100 m from adjacent reaches on Upper Talarik Creek, and 1,400 coho salmon per 100 m
from a reach on the North Fork Koktuli River.
Table 7 3. Highest index counts of selected stream rearing fish species from mainstem habitats.
Highest Reported Density (count per 100 m)a
Stream
South Fork Koktuli River
North Fork Koktuli River
Upper Talarik Creek
Chinook
Salmon
450
500
400
Coho
Salmon
600
1,400
1,600
Arctic
Grayling
275
40
2,500
Dolly
Varden
55
40
10
Source
Figure 15.1-52b
Figure 15.1-23b
Figure 15.1-82b
Notes:
a Values were approximated from figures listed in the source column.
b Source: PLP 2011.
7.2 Habitat Modification
The mine footprint would directly modify the amount of habitat available to salmon, trout, and Dolly
Varden by eliminating headwater streams and wetlands within and up-gradient of the mine footprint.
Potential effects of this habitat modification are described for all three mine scenarios (Sections 7.2.2,
7.2.3, and 7.2.4), and uncertainties and assumptions are described in Section 7.2.5.
7.2.1 Stream Segment Characteristics in the Mine Scenario Watersheds
The three mine scenario watersheds encompass an area of 925 km2 and contain 1,047 km of stream
channels mapped for this analysis (methods described in Box 3-1). In this section, we summarize stream
segment characteristics in the mine scenario watersheds to better characterize stream environments in
and downstream of the mine footprint areas. In Section 7.2.2, we summarize the characteristics of
stream segments that would be lost under the mine footprints themselves. Stream segments for the
entire Nushagak and Kvichak River watersheds are characterized in Chapter 3.
Similar to the larger Nushagak and Kvichak River watersheds, streams are generally low-gradient, with
extensive flat floodplains or terraces in the larger valleys (Figure 7-9; also see PLP 2011: Chapter 15 and
Appendix B). There are no large rivers (more than 28 m3/s mean annual flow) in the mine scenario
watersheds. Compared to the larger Nushagak and Kvichak River watersheds, streams in the mine
scenario watersheds have a higher median gradient (0.7% versus 0.4%) and a lower proportion of
stream length flowing through lowlands with more than 5% flatland (46% versus 64%) (Table 7-4,
Figure 7-9).
Broadly classified, streams and rivers in the Nushagak and Kvichak River watersheds likely to provide
habitats of high quality for salmonids include the streams with gradients less than 3%, and of medium
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-15
April 2013
-------
Chapter 7
Mine Footprint
stream size (0.15 to 2.8 m3/s mean annual flow) or greater. Such streams and rivers account for 36% of
the stream network in the larger Nushagak and Kvichak River watershed (Table 3-3), and account for
32% of the stream network in the mine scenario watersheds (Table 7-4). Smaller, steeper streams
provide seasonal (and some year-round) habitat, and provide important provisioning services to
downstream waters (Section 7.2.3). Although streams in the mine scenario watersheds are smaller and
slightly steeper than streams and rivers throughout the entire Nushagak and Kvichak River watersheds,
these results highlight the high proportion of stream channels in these basins with the broad
geomorphic and hydrologic characteristics that support stream and river habitats highly suitable for fish
species such as Pacific salmon, Dolly Varden, and rainbow trout.
Table 7 4. Distribution of stream channel length classified by channel size (based on mean annual
flow in m3/s), channel gradient (%), and potential floodplain influence for streams and rivers in the
mine scenario watersheds. Gray shading indicates proportions greater than 5%; bold indicates
proportions greater than 10%.
Channel Size
Small headwater streams3
Medium streams'5
Small rivers0
Large riversd
Gradient
<1%
FP
18%
14%
6%
0%
NFP
6%
5%
2%
0%
>l%and <3%
FP
6%
1%
0%
0%
NFP
26%
3%
1%
0%
>3% and <8%
FP
1%
0%
0%
0%
NFP
12%
0%
1%
0%
>8%
FP
0%
0%
0%
0%
NFP
6%
0%
0%
0%
Notes:
a 0-0.15 m3/s; most tributaries in the mine footprints.
b 0.15-2.8 m3/s; upper reaches and larger tributaries of the South and North Fork Koktuli Rivers and Upper Talarik Creek.
c 2.8-28 m3/s; middle to lower portions of the South and North Fork Koktuli Rivers and Upper Talarik Creek, including mainstem Koktuli River.
d >28 m3/s; the Mulchatna River below the Koktuli confluence, the Newhalen River, and other large rivers.
FP = floodplain influence; NFP = nofloodplain influence.
7.2.2 Exposure: Habitat Lost to the Mine Scenario Footprints
For each mine scenario, the total mine footprint consists of the area devoted to mining, including the
mine pit, waste rock piles, TSFs, ore processing facilities, and other mine-related constructs (see Chapter
6 for additional details for each mine scenario). Portions of the mine scenario watersheds would be
affected by mining activity in this footprint. Stream and wetland habitats would be lost within and
upstream of the footprint (Figures 7-10 through 7-12), and downstream habitat would be degraded by
the loss of the headwater streams and wetlands. Streams under or upstream of each mine footprint
would be inaccessible by fish from downstream reaches because of the following factors.
• Elimination of streams and wetlands within the mine footprints, either due to removal (e.g.,
excavation of streams or wetlands in the pit area) or burial under a TSF or waste rock pile.
• Dewatering by capture into a groundwater drawdown zone associated with the pit.
• Blockage due to either of the above or channel diversion in a manner that prevents fish passage
(e.g., via pipes or conveyances too steep for fish passage).
Streams and wetlands removed or altered via these various mechanisms are collectively referred to as
"lost" in this assessment. Methods used to estimate these losses are described in Box 7-1.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-16
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 9. Cumulative frequency of stream channel length classified by mean annual flow (m3/s),
reach gradient (%), and floodplain potential (measured as % flatland in lowland) for the mine
scenario watersheds (Scale 3) versus the Nushagak and Kvichak River watersheds (Scale 2).
100%
M
ji
E
re
\s>
!R
v
+j
JS
1
80%
40%
f
^ - -•• •••••-••• ••-"-
ScaleS
Scale 2
MAP Classification
100%
I 80%
E
re
01
'4-*
_ro
1
0%
2%
u
>
5
0%
0%
10
20 30
Mean Annual Flow mj/s)
40
SO
-Scales
Scale2
Gradient Classification
4%
6%
8% 10% 12%
Reach Gradient
14% 16% 18% 20%
ScaleS
Scale2
5% Classification
20%
40% 60%
Floodplain Potential
80%
100%
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-17
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 10. Streams and wetlands lost (eliminated, blocked, or dewatered) under the Pebble 0.25
scenario. Blue areas indicate streams and lakes from the National Hydrography Dataset (USGS
2012a) and wetlands from the National Wetlands Inventory (USFWS 2012). See Box 7 1 for
definitions and methods used for delineation.
A*
V)PPe
UP
l « .N-i
Pebble 0.25 Footprint
Groundwater Drawdown Zone
Eliminated, Blocked, or Dewatered Streams
Eliminated, Blocked, or Dewatered Wetlands
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-18
April 2013
-------
Chapter 7
Mine Footprint
Pebble 2.0 Footprint
Groundwater Drawdown Zone
Eliminated, Blocked, or Dewatered Streams
Eliminated, Blocked, or Dewatered Wetlands
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-19
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 12. Streams and wetlands lost (eliminated, blocked, or dewatered) under the Pebble 6.5
scenario. Blue areas indicate streams and lakes from the National Hydrography Dataset (USGS
2012a) and wetlands from the National Wetlands Inventory (USFWS 2012). See Box 7 1 for
definitions and methods used for delineation.
Pebble 6.5 Footprint
Groundwater Drawdown Zone
Eliminated, Blocked, or Dewatered Streams
Eliminated, Blocked, or Dewatered Wetlands
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-20
April 2013
-------
Chapter 7 Mine Footprint
BOX 7 1. CALCULATION OF STREAMS AND WETLANDS AFFECTED BY MINE SCENARIO FOOTPRINTS
To calculate stream kilometers eliminated, blocked, or altered in flow by the mine scenario footprints, we
used the Alaska National Hydrography Dataset(NHD) (USGS 2012a). The scale of this dataset is 1:63,360.
In this assessment, a stream segment is classified as eliminated if it falls within the boundaries of the mine
pit, waste rock pile, or tailings storage facility (TSF); blocked if it or a downstream segment it connects to
directly intersects the mine pit, waste rock pile, or TSF; and dewatered if it falls within the groundwater
drawdown zone associated with the mine pit. For calculation of stream kilometers either eliminated or
blocked that are inhabited by anadromous and resident fish species, we used the Anadromous Waters
Catalog AWC (Johnson and Blanche 2012) and the Alaska Freshwater Fish Inventory (AFFI) (ADF&G 2012).
Eliminated and blocked streams were defined as detailed above. Stream lengths blocked, eliminated, or
dewatered were summed across each classification for both NHD fish-inhabited stream segments (Table 7-
5).
Estimates of wetland areas eliminated, blocked, or dewatered by mine scenario footprints were derived from
the National Wetland Inventory (USFWS 2012). For the State of Alaska, the scale of this dataset is 1:63,360.
In this assessment, wetland area is classified as eliminated if it falls within the boundaries of the mine pit,
waste rock pile, or TSF; dewatered if it falls within the groundwater drawdown zone associated with the mine
pit; and blocked if it directly intersects a previously categorized blocked NHD stream. Wetland areas in each
blocked, eliminated, or dewatered were summed within each classification (Table 7-7).
The area covered by facilities associated with mine site development (e.g., housing, crushing plant, WWTP)
is not considered in the calculation of eliminated and blocked streams and wetlands due to lack of
knowledge about the specific size, orientation, or placement on the landscape of these structures. The
values reported in Tables 7-5 and 7-7 are conservative estimates as additional development on the
landscape would likely impact additional wetland area and stream length due to abundance of aquatic
habitat in this region.
It is important to note that these estimates of both stream length and wetland area affected reflect a lower
bound on the estimate. The NHD does not capture all stream courses and may underestimate channel
sinuosity, resulting in underestimates of affected stream length. Additionally, the AWC and the AFFI do not
necessarily characterize all potential fish-bearing streams because it is not possible to sample all streams,
and there may be errors in identification and mapping. ADF&G, in its on-line AWC database, states: "Based
upon thorough surveys of a few drainages it is believed that this number represents less than 50% of the
streams, rivers and lakes actually used by anadromous species" (ADF&G 2013). The characterization of
wetland area is limited by resolution of the available NWI data product. Furthermore, the mine scenario
components often bisected wetland features, and wetland area falling outside the boundary was assumed
to maintain its functionality. We were also unable to determine the effect that mine site development may
have on wetlands with no direct surface connection to a blocked NHD stream segment, but with a potential
connection via groundwater pathways. Given these limitations, these estimates could be enhanced with
improved, higher-resolution mapping, increased sampling of possible fish-bearing waters, and ground-
truthing.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7 2± April 2013
-------
Chapter 7 Mine Footprint
7.2.2.1 Stream Losses
Under the Pebble 0.25 scenario, over 38 km of streams would be eliminated, blocked, or dewatered by
the mine footprint (Table 7-5, Figure 7-10). Under the Pebble 2.0 scenario, over 90 km of streams would
be eliminated, blocked, or dewatered by the mine footprint (Table 7-5, Figure 7-11). Under the Pebble
6.5 scenario, an additional 19 km of streams in the pit and waste rock pile areas, and an additional 36
km of streams under TSF 2 and TSF 3 would be eliminated or blocked, for a total of 145 km of streams
lost to the mine footprint (Table 7-5, Figure 7-12). These scenarios represent 4, 9, and 14% of the total
stream length within the mine scenario watersheds. Of the streams lost to the mine footprint under the
Pebble 6.5 scenario, 115 km (82%) are headwater streams (less than 0.15 m3/s mean annual flow); 74%
have less than 3% gradient, and 26% have less than 1% gradient (Table 7-6, Figure 7-13). The majority
(79%) of smaller streams lost to the mine footprint under the Pebble 6.5 scenario flow through valleys
with limited flatland (Table 7-6).
Compared to the larger Nushagak and Kvichak River watersheds, streams lost to the mine scenario
footprints are smaller: 9% of stream length in the larger assessment area exceeds 2.8 m3/s mean annual
flow, whereas no streams lost to the mine scenario footprints exceed this size (Figure 7-13). Streams
under the mine scenario footprints also have fewer segments with less than 1% gradient (26% versus
66% of stream length in the Nushagak and Kvichak River watersheds), and fewer segments with more
than 5% flatland in their valleys (22% versus 64%) (Figure 7-13).
These results provide some indication of the relative size, steepness, and geomorphic setting of streams
that would be lost under the mine scenario footprints. The streams that would be lost include those that
directly providing habitat for salmonids, and streams that may not contain salmonids at all times of year
but provide important sources of water, macroinvertebrates, and other materials (Section 7.2.3). Of the
145 km of streams under the Pebble 6.5 footprint, 35 km are currently cataloged anadromous fish
streams listed in the AWC. Most of these AWC streams are in the medium stream size class (0.15 to
2.8 m3/s), with gradients less than 3%. These include the upper reaches and larger tributaries of the
South and North Fork Koktuli Rivers and Upper Talarik Creek, including smaller streams with
documented occurrence of coho salmon (Figure 7-3). Many of the smaller, steeper tributaries have been
documented to contain Dolly Varden (Figure 7-7).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 122 April 2013
-------
Chapter 7
Mine Footprint
Figure 7 13. Cumulative frequency of stream channel length classified by mean annual flow (m3/s),
reach gradient (%), and f loodplain potential (measured as % f latland in lowland) for the mine
0
100% T
J=
5 Rfl1-^ -
E
ro
ffi
ro
u
nod .
ve % Stream Length Cumulative % Stream Length
i-1 h-
*» cr> oo o M*»CIOOO c
OOOC OCCCOO J
K at V S? S 3? SS ?S. 5? S
0 -
= 20% -
u
0% -
0
Scale4
Scale2
MAP Classification
5 10 15 20 25 30 35 40 45 50
Mean Annual Flow (m'/s)
i
i
i
i
\J
-- '~~/~
<''' /
Scale 4
Scale2
r- ]• t/-l •!•
% 2% 4% 6% 8% 10% 12% 1£% 16% 18% 20%
Reach Gradient
/
f
I"
I
_^
^, — -'~~
***' Scale 2
n : I ft
% 20% 40% 60% 80% 100%
Ftoodplain Potential
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-23
April 2013
-------
Chapter 7
Mine Footprint
Table 7 5. Stream length (km) eliminated, blocked, or dewatered by the mine footprints under the Pebble 0.25, 2.0, and 6.5 scenarios.
Component
Stream Length3
Eliminated
by Footprint
Blocked by
Footprint
Dewatered
by Footprint
Blocked and
Dewatered
by Footprint
TOTAL
Stream
Length Lostc
to Footprint
AWC Stream Length
Eliminated
by Footprint
Blocked by
Footprint
Dewatered
by Footprint
Blocked and
Dewatered
by Footprint
Salmon
Species
Present in
Lostc Streams
TOTAL AWC
Stream
Length
Lostc to
Footprint
Pebble 0.25
Pit
Waste rock
TSFld
TOTAL
3
5
11
19
0
<1
4
4
13
0
0
13
2
0
0
2
18
5
15
38
0
0
6
6
0
0
0
0
2
0
0
2
0
0
0
0
coho, Chinook
coho, Chinook
2
0
6
8
Pebble 2.0
Pit + waste
rock
TSFld
TOTAL
47
15
62
20
<1
20
2
0
2
6
0
6
75
15
90
11
6
17
4
0
4
0
0
0
3
0
3
coho,
Chinook,
sockeye
coho, Chinook
18
6
24
Pebble 6.5
Pit + waste
rock
TSF1
TSF2
TSF3
TOTAL
77
15
25
9
126
6
<1
1
1
8
3
0
0
0
3
8
0
0
0
8
94
15
26
10
145
18
6
5
2
31
2
0
0
0
2
0
0
0
0
0
2
0
0
0
2
coho,
Chinook,
sockeye
coho, Chinook
chum, coho,
Chinook
coho
22
6
5
2
35
Notes:
3 From the National Hydrography Dataset (USGS 2012a).
b From the Anadromous Waters Catalog (Johnson and Blanche 2012)
c Lost = eliminated + blocked + dewatered.
d TSF 1 expands in size under Pebble 2.0.
TSF = tailings storage facility; AWC = Anadromous Waters Catalog.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-24
April 2013
-------
Chapter 7
Mine Footprint
Table 7 6. Distribution of stream channel length classified by channel size (based on mean annual
discharge in m3/s), channel gradient (%), and potential floodplain influence for streams under the
Pebble 6.5 mine footprint. Gray shading indicates proportions greater than 5%; bold indicates
proportions greater than 10%.
Channel Size
Small headwater streams3
Medium streams'5
Small rivers0
Large riversd
Gradient
<1%
FP
9%
7%
0%
0%
NFP
6%
4%
0%
0%
>l%and <3%
FP
5%
1%
0%
0%
NFP
35%
7%
0%
0%
>3% and <8%
FP
0%
0%
0%
0%
NFP
27%
0%
0%
0%
>8%
FP
0%
0%
0%
0%
NFP
6%
0%
0%
0%
Notes:
a 0-0.15 m3/s; most tributaries in the mine footprints.
b 0.15-2.8 m3/s; upper reaches and larger tributaries of the North and South Fork Koktuli Rivers and Upper Talarik Creek.
c 2.8-28 m3/s; middle to lower portions of the North and South Fork Koktuli Rivers and Upper Talarik Creek, including the mainstem Koktuli
River.
d >28 m3/s; the Mulchatna River below the Koktuli confluence, the Newhalen River, and other large rivers.
FP = floodplain influence; NFP = no floodplain influence.
7.2.2.2 Wetland losses
In addition to streams, 5 km2 of wetland habitat would be lost under the Pebble 0.25 footprint, 12.4 km2
by the Pebble 2.0 footprint, and 19.4 km2 by the Pebble 6.5 footprint (Table 7-7). Methods used to
estimate these losses are described in Box 7-1.
Table 7 7. Wetland areas3 (km2) eliminated, blocked, or dewatered by the mine footprints under the
Pebble 0.25, 2.0, and 6.5 scenarios.
Component
Eliminated by
Footprint
Blocked by
Footprint
Blocked and
Dewatered by
Footprint
Dewatered by
Footprint
TOTAL Lost to
Footprint
Pebble 0.25
Mine pit
Waste rock
TSFlb
TOTAL
0.28
0.40
1.92
2.60
0.00
0.00
0.70
0.70
0.06
0
0
0.06
1.63
0
0
1.63
1.97
0.40
2.62
4.99
Pebble 2.0
Mine pit and waste rock
TSF1»
TOTAL
6.65
3.49
10.14
1.71
0.00
1.71
0.31
0
0.31
0.28
0
0.28
8.95
3.49
12.44
Pebble 6.5
Mine pit and waste rock
TSF1
TSF2
TSF3
TOTAL
11.79
3.49
1.73
0.31
17.32
0.46
0.00
0.00
0.02
0.48
0.77
0
0
0
0.77
0.87
0
0
0
0.87
13.89
3.49
1.73
0.33
19.44
Notes:
a Based on the National Wetlands Inventory (USFWS 2012).
b TSF 1 expands in size under Pebble 2.0.
TSF = tailings storage facility.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-25
April 2013
-------
Chapter 7
Mine Footprint
Table 7-8 provides a summary of the total documented anadromous fish stream length in the mine
scenario watersheds included in the AWC (Johnson and Blanche 2012). Approximately 2, 7, and 11% of
the total anadromous fish stream length in the mine scenario watersheds would be eliminated,
dewatered, or blocked by the Pebble 0.25, 2.0 and 6.5 mine footprints, respectively (Table 7-5). In
addition to these direct losses, loss of these headwater habitats would have indirect impacts on fishes
and their habitats in downstream mainstem reaches of each watershed (Section 7.2.3).
Table 7 8. Total documented anadromous fish stream length and stream length documented to
contain different fish species in the mine scenario watersheds.
Total Mapped Streams3
Total Anadromous Fish Streams'1
South Fork Koktuli
River (km)
315
95
North Fork Koktuli
River (km)
343
104
Upper Talarik
Creek (km)
427
123
Total (km)
1,085
322
By species
Chinook salmon
Chum salmon
Coho salmon
Pink salmon
Sockeye salmon
Dolly Varden0
59
37
93
0
6,450
48
61
31
103
0
47
0
63
45
122
7
80
26
183
113
318
7
191
75
Notes:
a From the National Hydrography Dataset (USGS 2012a).
b From Anadromous Waters Catalog (Johnson and Blanche 2012).
c Listed as Arctic char in some cases, but assumed to be Dolly Varden (Appendix B).
7.2.3 Exposure-Response: Implications of Stream and Wetland Loss for
Fish
7.2.3.1 Fish Occurrence in Streams and Wetlands Lost to the Mine Scenario Footprints
Table 7-5 provides an estimate of salmon habitat directly affected by the mine footprint under the three
scenarios. A total of 8 km, 24 km, and 35 km of documented anadromous fish streams would be
eliminated, blocked, or dewatered by the mine footprints under the Pebble 0.25, 2.0, and 6.5 scenarios,
respectively. The distribution of anadromous Dolly Varden in the Nushagak and Kvichak River
watersheds is not known, making an estimate of the total anadromous fish habitat affected by the mine
scenarios impossible. Of the total wetland area eliminated or blocked by each footprint, the proportion
used by anadromous salmonids or resident fish species is unknown. Fish access to and use of wetlands
are likely to be extremely variable in the mine area, due to differences in the duration and timing of
surface water connectivity with stream habitats, distance from the main channel, and physical and
chemical conditions (e.g., dissolved oxygen concentrations) (King et al. 2012). Wetlands can provide
refuge habitats (Brown and Hartman 1988) and important rearing habitats for juvenile salmonids by
providing hydraulically and thermally diverse conditions. Wetlands can also provide enhanced foraging
opportunities (Sommer et al. 2001). Given our insufficient knowledge of how fish use wetlands in the
deposit area, it is not possible to calculate the effects of lost wetland connectivity and abundance on
stream fish populations.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-26
April 2013
-------
Chapter 7 Mine Footprint
Spawning habitat for coho salmon would be lost in the South and North Fork Koktuli River watersheds
as a result of TSF 1 and TSF 2, respectively (Figure 7-3); sockeye and coho salmon spawning habitat
would be lost in the Upper Talarik Creek watershed as a result of the waste rock pile footprint
(Figures 7-2 and 7-3) (Johnson and Blanche 2012). In other regions, anadromous and resident forms of
Dolly Varden have been observed in the most upstream and high-gradient habitats available for
spawning, indicating that headwaters may be important source areas for downstream populations
(Bryant etal. 2004). Under the Pebble 6.5 footprint, 98% of stream kilometers are estimated to be less
than 8% gradient, and 73% are under 3% gradient, well within the range of gradients used by these
species.
In addition to spawning, streams in each mine footprint provide rearing habitat for fishes of the mine
scenario watersheds. Species known to rear in habitats in and upstream of the mine footprints are
sockeye salmon (Figure 7-2), coho salmon (Figure 7-3), Chinook salmon (Figure 7-4), chum salmon
(Figure 7-5), Dolly Varden (Figure 7-7), rainbow trout (Figure 7-8), Arctic grayling, slimy sculpin,
northern pike, and ninespine stickleback (Johnson and Blanche 2012, ADF&G 2012).
7.2.3.2 Importance of Headwater Stream and Wetland Habitats
The majority of streams directly in the footprint of the Pebble 6.5 scenario are classified as small
headwater streams (less than 0.15 m3/s mean annual flow) (Table 7-6). Because of their narrow width,
headwater streams receive proportionally larger inputs of organic material from the surrounding
terrestrial vegetation than do larger stream channels (Vannote et al. 1980). This material is either used
in the headwater environment (Tank et al. 2010) or transported downstream as a subsidy to larger
streams in the network (Wipfli et al. 2007). Consumers in headwater stream food webs, such as
invertebrates and juvenile salmon, have been shown to rely heavily on the terrestrial inputs that enter
the stream (Doucett et al. 1996, Dekar etal. 2012). Because of their shallow depths and propensity to
freeze, headwater streams may be largely uninhabitable in the winter (but see discussion of
overwintering below), and fish distribution in headwater systems in southwestern Alaska is likely most
extensive in late summer and early fall (Elliott and Finn 1983). This coincides with maximum growth
periods for rearing juvenile salmon, when both stream temperatures and food availability increase
(Quinn 2005:195-196).
Data on riparian vegetation communities specific to the mine footprints were not available, but
vegetation in the mine area is generally described by PLP (2011). Shrub vegetation communities
account for 81% of the total area, with four dominant vegetation types: dwarf ericaceous shrub tundra,
dwarf ericaceous shrub lichen tundra, open willow low shrub, and closed alder tall shrub (PLP 2011:
Chapter 13:10). Riparian areas are dominated by willow and alder shrub communities (PLP 2011:
Chapter 13:11). Deciduous shrub species such as alder and willow provide abundant and nutrient-rich
leaf-litter inputs, which are used more rapidly in stream food webs than coniferous plants or grasses
(Webster and Benfield 1986). In addition, alder is a nitrogen-fixing shrub known to increase headwater
stream nitrogen concentrations (Compton et al. 2003, Shaftel et al. 2012), which can result in more rapid
litter processing rates (Ferreira etal. 2006, Shaftel etal. 2011). The presence of both willow and alder in
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTI April 2013
-------
Chapter 7 Mine Footprint
headwater stream riparian zones implies high-quality basal food resources for stream fishes in the
deposit area.
In addition to providing summer rearing habitat, lower-gradient headwater streams and associated
wetlands may also provide important habitat for stream fishes during other seasons. Loss of wetlands is
a common result of land development (Pess etal. 2005), and in more developed regions has been
associated with reductions in habitat quality and salmon abundance, particularly for coho salmon
(Beechie etal. 1994, Pess etal. 2002). Thermally diverse habitats in off-channel wetlands can provide
rearing and foraging conditions that may be unavailable in the main stream channel (Sommer et al.
2001, Henning et al. 2006), increasing capacity for juvenile salmon rearing (Brown and Hartman 1988).
Winter habitat availability for juvenile rearing has been shown to limit salmonid productivity in streams
of the Pacific Northwest (Nickelson etal. 1992, Solazzi etal. 2000, Pollock etal. 2004) and may be
limiting for fishes in the mine scenario watersheds because of the relatively cold temperatures and long
winters in the region. Overwintering habitats for stream fishes must provide suitable instream cover,
dissolved oxygen, and protection from freezing (Cunjak 1996). Beaver ponds and groundwater sources
in headwater streams and wetlands in the mine footprints likely meet these requirements.
In winter, beaver ponds typically retain liquid water below the frozen surface, which makes them
important winter refugia for stream fishes (Nickelson et al. 1992, Cunjak 1996). Beavers preferentially
colonize headwater streams because of their shallow depths and narrow widths, and several studies
have indicated that dam densities are reduced significantly at stream gradients above 6 to 9% (Collen
and Gibson 2001, Pollock etal. 2003). Beaver ponds provide excellent habitat for rearing salmon by
trapping organic materials and nutrients and creating structurally complex habitat, large capacity pool
habitats with potentially high macrophyte cover, low flow velocity, and/or moderate temperatures
(Nickelson etal. 1992, Collen and Gibson 2001, Lang etal. 2006). Additionally, beaver dams, including
ponds at a variety of successional stages, provide a mosaic of habitats for not just salmon but other fish
and wildlife species.
An aerial survey of active beaver dams in the mine area, conducted in October 2005 (PLP 2011:
Chapter 16:16.2-8), mapped 113 active beaver colonies. The area surveyed did not include streams
draining the TSF 1 footprint (PLP 2011: Figure 16.2-20). Several active beaver colonies were mapped in
streams that would be eliminated or blocked by the mine pit and waste rock piles. These are lower-
gradient habitats than the headwater streams draining the TSF 1, TSF 2, and TSF 3 footprints. Loss of
beaver pond habitats in the headwaters of the South Fork Koktuli River and Upper Talarik Creek
watersheds would reduce both summer and winter rearing opportunities for anadromous and resident
fish species.
Inputs of groundwater-influenced streamflow from headwater tributaries likely benefit fish by
moderating mainstem temperatures (Cunjak 1996, Power et al. 1999, Huusko et al. 2007, Armstrong et
al. 2010, Brown et al. 2011). PLP collected temperature data from stream sampling sites using in-situ
field meters according to the procedures outlined in their Quality Assurance Project Plan (PLP 2011:
Figure 9.1-8). Maximum summer (June through August) water temperatures recorded at gage NK119A,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~28 April 2013
-------
Chapter 7 Mine Footprint
which drains the TSF 1 footprint, were approximately 5°C colder than the mainstem reach that it flows
into (PLP 2011: Tables 15.1 through 15.4). This difference was not as pronounced for the maximum
summer water temperatures recorded at gage SK119A, which drains the TSF 2 footprint and was
approximately 2°C colder than the mainstem reach that it flows into (PLP 2011: Tables 15.1 through
15.21).
Longitudinal temperature profiles for the South and North Fork Koktuli River watersheds from August
and October indicate that the mainstem reaches just downstream of the tributaries draining TSF 1 and
TSF 2 experience significant cooling in the summer and warming in the winter compared to the adjacent
upstream reaches (PLP 2011: Figures 15.1-11 and 15.1-41). Headwater streams in the South and North
Fork Koktuli River watersheds may provide a temperature-moderating effect, providing temperatures
beneficial to fishes in summer and possibly winter as well. Two aerial surveys of the mine scenario
watersheds provided additional information on groundwater inputs to headwater streams draining the
mine scenario footprints (PLP 2011, Woody and Higman 2011). PLP conducted aerial and foot surveys
during late winter low-flow conditions in 2006, 2007, and 2008 to determine the extent of open water
and ice cover (PLP 2011; Appendix 7.2B). Open-water reaches were consistently observed in strongly
gaining reaches in the South and North Fork Koktuli Rivers and Upper Talarik Creek. Open-water
reaches corresponded with areas of high upwelling potential modeled by Wobus etal. (2012)
(Figure 7-14). Aerial surveys documented by Woody and Higman (2011) in March 2011 showed broadly
similar patterns of open water, suggesting that the general patterns reflect consistent areas of strong
groundwater-surface water interaction.
In addition to providing habitat for stream fishes, headwater streams and wetlands serve an important
role in the stream network by contributing nutrients, water, organic material, and macroinvertebrates
downstream to higher-order streams in the watershed. In the northeastern United States, headwaters
contribute approximately 70% of the water volume and 65% of the nitrogen flux to second-order
streams, and 55% of the volume and 40% of the nitrogen flux to fourth- and higher-order rivers
(Alexander et al. 2007). Headwater contributions to downstream systems result from the high density of
those systems in the dendritic stream network. Headwater streams also have high rates of instream
nutrient processing and storage due to extensive hyporheic zone interactions resulting from a large bed
surface area compared to the volume of the overlying water (Alexander et al. 2007).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, -g April 2013
-------
Chapter 7
Mine Footprint
Figure 7 14. Comparison of MIKE SHE modeled groundwater upwelling areas and inferred
upwelling areas based on PLP (2011) aerial surveys. Colored dots represent the magnitude of
groundwater upwelling generated by MIKE SHE, with red and orange colors representing the strongest
upwelling. Black outlined areas show documented upwelling locations from PLP (2011) during three
winter open water surveys. Figure modified from Wobus et al. (2012); see Wobus et al. (2012) for
details on MIKE SHE modeling methods.
Mine Scenario Watersheds
I I Watershed Boundary
| | MIKE-SHE Model Domain
\////\ Observed Open Water During Any Year
Groundwater - Surface Water Exchange
• Greater than 200 m3/day
111 - 200 m3/day
58 -110 m3/day
13 - 57 m3/day
• Less than 13 m3/day
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-30
April 2013
-------
Chapter 7 Mine Footprint
In addition to nutrients, both invertebrates and detritus are exported from headwaters to downstream
reaches and provide an important energy subsidy for juvenile salmonids (Wipfli and Gregovich 2002).
This effect can be mediated by the surrounding vegetation: for example, in southeastern Alaskan
streams, riparian alder (a nitrogen-fixing shrub) was positively related to aquatic invertebrate densities
and the export rates of invertebrates and detritus (Piccolo and Wipfli 2002, Wipfli and Musslewhite
2004). Headwater wetlands and associated wetland vegetation can also be important sources of
dissolved organic matter, particulate organic matter, and macroinvertebrate diversity (King et al. 2012),
contributing to the chemical, physical, and biological condition of downstream waters (Shaftel et al.
2011, Dekar etal. 2012, Walker etal. 2012). Loss of headwater streams and wetlands from the mine
footprints would reduce inputs of organic material, nutrients, water, and macroinvertebrates to
downstream reaches, but the effect on fish cannot be quantified.
7.2.4 Risk Characterization
Direct loss of stream habitat under the mine footprints would be unavoidable for a project of the sizes
described for our mine scenarios, due to the density of National Hydrography Dataset (NHD)-mapped
streams in the project area (average 1 km/km2). Stream blockage is not necessarily unavoidable, but
would require appropriate engineering and maintenance.
Direct loss or blockage of these streams would remove these streams from available fish habitat, and
alter the important ecological functions they provide for downstream waters. Loss of headwater stream
contributions to downstream waters would reduce the capacity and productivity of stream habitats.
Together, these reductions in habitat capacity and productivity would result in adverse impacts on fish
populations (Figure 7-1). These streams provide known spawning and rearing habitats for anadromous
and resident fish species, and their watersheds support some of region's highest diversity of salmonid
species (Figure 5-3). Stream habitat losses leading to losses of local, unique populations would erode the
population diversity that is key to the stability of the overall Bristol Bay salmon fishery (Schindler et al.
2010).
Compensatory mitigation measures could offset some of the stream and wetland losses described here
(Box 7-2), although there are substantial challenges regarding the potential efficacy of these measures to
successfully offset adverse impacts. Appendix J presents a more detailed discussion of compensatory
mitigation issues.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, oj_ April 2013
-------
Chapter 7 Mine Footprint
BOX 7 2. COMPENSATORY MITIGATION
Compensatory mitigation refers to the restoration, establishment, enhancement, and/or preservation of
wetlands, streams, or other aquatic resources. Compensatory mitigation regulations jointly promulgated by
the U.S. Environmental Protection Agency (USEPA) and the U.S. Army Corps of Engineers (USAGE) state that
"the fundamental objective of compensatory mitigation is to offset environmental losses resulting from
unavoidable impacts to waters of the United States authorized by [Clean Water Act Section 404 permits
issued by the USAGE]" (40 Code of Federal Regulations [CFR] 230.93(a)(l)). Compensatory mitigation
enters the analysis only after a proposed project design has incorporated all appropriate and practicable
means to avoid and minimize adverse impacts on aquatic resources (40 CFR 230.91(c)). Compensatory
mitigation measures are usually not part of project design but are considered necessary to maintain the
integrity of the nation's waters.
The mine scenarios evaluated in this assessment identify that the mine footprints alone will result in the
loss (i.e., filling, blocking or otherwise eliminating) of high-functioning wetlands and tens of kilometers of
salmon-supporting streams. Such extensive habitat losses could also result in the loss of unique salmon
populations, potentially eroding the genetic diversity that is essential to the stability of the overall Bristol Bay
salmon fishery (i.e., reduction in the portfolio effect discussed in Section 5.2.4).
The public and peer review comments on the first external review draft of this assessment identified an
array of compensation measures that commenters believed could potentially offset these impacts on
wetlands, streams, and fish. Appendix J provides an overview of Clean Water Act (CWA) Section 404
compensatory mitigation requirements for unavoidable impacts on aquatic resources and discusses the
likely efficacy of these potential compensation measures at offsetting potential adverse impacts. Note that
any formal determinations regarding compensatory mitigation can only take place in the context of a
regulatory action. This assessment is not a regulatory action, and thus a complete evaluation of
compensatory mitigation is considered outside the scope of the assessment.
Potential compensatory mitigation measures identified by commenters and discussed in Appendix J include
mitigation bank credits, in-lieu fee program credits, and permittee-responsible compensatory mitigation
projects such as aquatic resource restoration and enhancement within the South and North Fork Koktuli
Rivers and Upper Talarik Creek watersheds as well as more distant portions of the Nushagak and Kvichak
River watersheds. The following additional measures are identified in Appendix J:
• Beaver dam removal
• Flow management
• Spawning channel construction
• Aquatic resource preservation
• Old mine site remediation
• Road removal
• Road stream crossing retrofits
• Hatchery construction
• Fish stocking
• Commercial fishery harvest reductions
As discussed in Appendix J, there are significant challenges regarding the potential efficacy of compensation
measures proposed by commenters for use in the Bristol Bay region, raising questions as to whether
compensation measures could address impacts of the type and magnitude identified for the mine scenarios.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, APril 2013
-------
Chapter 7 Mine Footprint
7.2.5 Uncertainties
Losses of anadromous fish-bearing streams (Table 7-5) in the mine scenario watersheds are likely
underestimated because of the difficulty of accurately capturing data on all streams that may support
fish use throughout the year. We rely on the AWC and AFFI for documentation of species distributions,
but these records are incomplete—not all stream reaches have been surveyed—and may be subject to
errors in fish identification. Additionally, depictions of species and life history distributions in the AWC
reflect a wide range of mapping policies, and it is difficult to interpret under which policies a particular
water body was mapped. That said, the fish sampling documented by PLP (2011) is one of the highest-
density efforts conducted to date in this portion of Alaska, such that estimates of anadromous fish
distributions are likely better represented here than elsewhere in Alaska.
Losses of headwater streams and anadromous fish-bearing streams (Table 7-5) in the mine scenario
watersheds may also be underestimated because of challenges associated with stream network
mapping. Estimates of headwater stream extent were derived from the Alaska NHD (USGS 2012a),
which does not capture all stream courses and may underestimate channel sinuosity, resulting in
underestimates of stream length. A LiDAR-derived stream network map would likely yield substantially
different results than those presented here. Similarly, actual wetland loss or blockage because of the
mine footprint (Table 7-7) would likely be higher than estimated here, as the NWI is based on remotely-
sensed imagery and generally underestimates wetland area. See Box 7-1 for additional discussion of
uncertainties associated with stream and wetland mapping.
7.3 Streamflow Modification
7.3.1 Exposure: Streamflow
In this section, we describe projected changes in the hydrology of the mine scenario watersheds and
associated effects on downstream flows that would result from mine development and operation. We
assume that streams in and downstream of the mine footprints would experience Streamflow alterations
due to water collection, treatment, and discharge to streams via WWTP outfalls; leakage from TSFs; and
leachate from waste rock piles. For a full description of water flows through the mine facilities, see
Chapter 6.
Streamflow alterations resulting from mine operations were estimated by reducing the flows recorded
at existing stream gages (Table 7-9, Figures 7-15 through 7-17) for the mine scenario watersheds by the
percentage of the expected surface area lost to each mine footprint. Reductions also included losses to
the drawdown zone, caused by the cone of depression of the mine pit, or other locations of dewatering
(Table 7-9, Section 6.2.2). Additions to Streamflow occurred in response to discharges through the
WWTP. The net effect on resulting streamflows were mapped and summarized for individual stream
and river segments (Figure 7-15 through 7-17, Table 7-9).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, «o April 2013
-------
Chapter 7
Mine Footprint
Figure 7 15. Stream segments in the mine scenario watersheds showing flow changes (%) associated with the Pebble 0.25 footprint. Flow
modification class is shown for each stream segment to indicate degree and direction of change. Flow modification classes are assigned at a
gage and extend upstream to the next gage, confluence point, or mine footprint. Channels and tributaries not classified are shown for
informational purposes. Data for the USGS gages based on Water Resources of Alaska (USGS 2012b); data for the PLP gages based on the
Environmental Baseline Document 2004 through 2008 (PLP 2011).
NK119CP10YT
\ "NK100C
B
<119CP
£ SK100C
V
-
I
LZ
PLP/USGS Gages
Confluence Points
i
J Pebble 0.25 Footprint
J Drawdown Zone
5-10% Decrease
^^— 10-20% Decrease ^
^^^ >20% Decrease —
^^— ' 0-5% Decrease or Increase
5-10% Increase
— 10-20% Increase
— >20% Increase
N
A
0 1
^•H
0 1
^^•ZZ
2
3 Kilometers
2
IMiles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-34
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 16. Stream segments in the mine scenario watersheds showing flow changes (%) associated with the Pebble 2.0 footprint. Flow
modification class is shown for each stream segment to indicate degree and direction of change. Flow modification classes are assigned at a
gage and extend upstream to the next gage, confluence point, or mine footprint. Channels and tributaries not classified are shown for
informational purposes. Data for the USGS gages based on Water Resources of Alaska (USGS 2012b); data for the PLP gages based on the
Environmental Baseline Document 2004 through 2008 (PLP 2011).
.SK119A
SK100C
© PLP/USGS Gages
© Confluence Points
1 1
| | Drawdown Zone
| Pebble 2.0 Footprint
5-10% Decrease
10-20% Decrease —
^-^ >20% Decrease —
•^— 0-5% Decrease or Increase
5-10% Increase
— 10-20% Increase
— >20% Increase
N
A
0 1
^•=
0 1
••=
2
3 Kilometers
2
(Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-35
April 2013
-------
Chapter 7
Mine Footprint
modification class is shown for each stream segment to indicate degree and direction of change. Flow modification classes are assigned at a
gage and extend upstream to the next gage, confluence point, or mine footprint. Channels and tributaries not classified are shown for
informational purposes. Data for the USGS gages based on Water Resources of Alaska (USGS 2012b); data for the PLP gages based on the
Environmental Baseline Document 2004 through 2008 (PLP 2011).
NK100B
© PLP/USGS Gages
© Confluence Points
\// , \ Drawdown Zone
| | Pebble 6.5 Footprint
5-10% Decrease
^^— 10-20% Decrease ^
^^— >20% Decrease —
^^— 0-5% Decrease or Increase
5-10% Increase
— 10-20% Increase
— >20% Increase
N
A
0 1
I^E=
0 1
^^MH
2
3 Kilometers
2
(Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-36
April 2013
-------
Chapter 7
Mine Footprint
Table 7 9. Stream gages and related characteristics for the South and North Fork Koktuli Rivers
and Upper Talarik Creek.
Stream and
Gage
Drainage Area
(km2)
Mean Annual Flow3
(mVs)
Mean Annual Unit Runoff
(m3/s*km2)
Mean Annual Runoff
(mVyr)
South Fork Koktuli River
SK100G
SK100F
SK124A
SK100C
SK119A
SK100B1
SKIOOB"
14
31
22
99
28
141
179
0.4
0.8
0.5
1.3
1.0
3.7
5.1
0.026
0.026
0.024
0.013
0.036
0.026
0.029
11,618,000
25,842,000
16,811,000
41,858,000
31,268,000
115,110,000
162,122,000
North Fork Koktuli River
NK119A
NK119B
NK100C
NK100B
NK100A1
NK100AC
20
10
64
97
219
277
0.7
0.1
1.3
2.4
5.8
7.0
0.035
0.013
0.021
0.025
0.026
0.025
21,515,000
4,081,000
41,853,000
76,408,000
182,297,000
220,715,000
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119A
UTlOOBd
7
30
131
157
183
10
221
0.3
0.8
2.9
3.4
4.5
0.8
6.2
0.036
0.025
0.023
0.022
0.024
0.084
0.028
7,996,000
24,201,000
92,734,000
107,971,000
141,213,000
25,549,000
196,182,000
Notes:
3 Calculated from stream gage data from PLP 2011.
b USGS 15302200.
c USGS 15302250.
d USGS 15300250.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-37
April 2013
-------
Chapter 7 Mine Footprint
Daily flow data were obtained using data from seven gages in the South Fork Koktuli River, six gages in
the North Fork Koktuli River, and seven gages in Upper Talarik Creek (Table 7-9) (PLP 2011). We
calculated mean and minimum monthly flows for each gage under pre-mining baseline conditions
(Tables 7-10 through 7-15, Figure 7-18). The periods of record varied for gages in the three mine
scenario watersheds, but generally covered the period from 2004 to 2010.
In addition, we estimated flow at six confluence points where mining-related flow impacts were
expected, but where established stream gage records were lacking. This allowed for more discrete
estimation of baseline streamflow as well as expected flow modification under each mine scenario due
to withdrawal, addition, or footprint loss. The tributary area to each stream gage or confluence point
was calculated based on the National Elevation Dataset digital elevation model (Gesch et al. 2002, Gesch
2007) in a geographic information system (CIS). The spatial orientation and area of each major mine
component (e.g., pit, TSFs, waste rock piles) in each drainage basin were determined (Table 7-16
through 7-18, Figures 6-1 through 6-3). The percentage of watershed area covered by mine components
was calculated for each gage and confluence subwatershed. Using the calculated percentage of
watershed area covered by mine components, mean annual flow records for each of the gages and
confluence watershed were adjusted downward. Next, the annual volume of return flow expected to
reach each gage was added back to the adjusted flow calculations based on the mine scenarios.
Expected changes to surface water flows were assessed for the three mine scenarios (Tables 7-10
through 7-15). We also considered water balance issues for the post-closure period, but flow estimates
were not assessed for this period. The Pebble 0.25 mine footprint consists of the mine pit, one waste
rock pile, and TSF 1. The Pebble 2.0 footprint would add a second or expanded waste rock pile, an
expanded TSF 1, and increased drawdown from groundwater flow to the pit (Section 6.2.2). The Pebble
6.5 footprint would add effects associated with the fully expanded mine footprint (including TSF 2 and
TSF 3) to accommodate expanded mine operations. We assume that during the post-closure period,
active dewatering of the pit would cease as the pit fills. Once the pit is filled, the water level would be
slightly lowered by pumping to maintain a gradient toward the pit. The pumped water would be treated
for as long as it did not meet water quality standards. When treatment is no longer necessary, the pit
would be allowed to have a natural outlet if the water level required one.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, oo April 2013
-------
Chapter 7
Mine Footprint
Table 7 10. Measured mean monthly pre mining flow rates (m3/s) and estimated mean monthly flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for gages along the South Fork Koktuli River.
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
SK100G
Pre
0.23
0.14
0.11
0.18
0.72
0.50
0.29
0.42
0.55
0.64
0.35
0.28
0.25
0.10
0.06
0.05
0.08
0.32
0.22
0.13
0.19
0.24
0.28
0.15
0.12
2.0
0.06
0.04
0.03
0.05
0.19
0.13
0.08
0.11
0.14
0.17
0.09
0.07
6.5
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
SK100F
Pre
0.44
0.25
0.19
0.24
1.95
1.38
0.59
0.83
1.20
1.47
0.75
0.53
0.25
0.32
0.18
0.14
0.18
1.42
1.01
0.43
0.61
0.88
1.07
0.54
0.39
2.0
0.28
0.16
0.12
0.16
1.24
0.88
0.38
0.53
0.77
0.94
0.48
0.34
6.5
0.17
0.09
0.07
0.09
0.74
0.52
0.22
0.32
0.46
0.56
0.28
0.20
SK124A
Pre
0.12
0.02
0.01
0.05
1.91
1.08
0.29
0.59
0.83
0.98
0.33
0.18
0.25
0.16
0.03
0.01
0.06
2.50
1.41
0.38
0.77
1.09
1.27
0.43
0.23
2.0
0.17
0.03
0.01
0.06
2.60
1.47
0.40
0.81
1.13
1.33
0.45
0.24
6.5
0.26
0.05
0.01
0.10
4.06
2.29
0.62
1.26
1.77
2.07
0.70
0.38
SK100C
Pre
0.37
0.03
<0.01
0.13
4.30
2.77
0.73
1.17
2.05
2.80
1.04
0.54
0.25
0.39
0.03
<0.01
0.13
4.47
2.87
0.76
1.22
2.13
2.90
1.08
0.56
2.0
0.38
0.03
<0.01
0.13
4.45
2.86
0.76
1.21
2.12
2.89
1.08
0.56
6.5
0.50
0.04
<0.01
0.17
5.76
3.70
0.98
1.57
2.74
3.74
1.39
0.73
SK119A
Pre
0.30
0.16
0.11
0.22
3.02
1.71
0.74
1.15
1.75
1.61
0.72
0.40
0.25
0.30
0.16
0.11
0.22
3.02
1.71
0.74
1.15
1.75
1.61
0.72
0.40
2.0
0.30
0.15
0.11
0.22
2.97
1.68
0.73
1.13
1.72
1.58
0.70
0.39
6.5
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
SK100B1
Pre
1.54
0.79
0.57
0.80
10.75
6.67
2.56
4.05
5.18
6.12
2.84
1.92
0.25
1.52
0.78
0.56
0.79
10.57
6.55
2.52
3.98
5.09
6.02
2.79
1.88
2.0
1.49
0.77
0.55
0.77
10.38
6.43
2.48
3.91
5.00
5.91
2.74
1.85
6.5
1.38
0.71
0.51
0.72
9.64
5.98
2.30
3.63
4.64
5.49
2.55
1.72
SKIOOB"
Pre
2.47
1.40
1.09
1.41
12.70
8.56
3.85
5.92
7.75
9.08
4.44
3.02
0.25
2.43
1.38
1.07
1.38
12.47
8.41
3.79
5.81
7.61
8.92
4.36
2.97
2.0
2.39
1.36
1.05
1.36
12.28
8.28
3.73
5.72
7.49
8.79
4.30
2.92
6.5
2.22
1.26
0.98
1.26
11.38
7.68
3.45
5.30
6.95
8.14
3.98
2.71
Notes:
a USGS 15302200.
NA = not applicable: SK100G would be eliminated by TSF 2, and SK119A would be eliminated by TSF 3 under the Pebble 6.5 scenario.
I lift]
liilrt
gages along the North Fork Koktuli River.
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
NK119A
Pre
0.15
0.10
0.08
0.21
2.28
1.15
0.55
0.71
1.10
1.10
0.52
0.24
0.25
0.11
0.07
0.06
0.15
1.72
0.86
0.41
0.53
0.83
0.82
0.39
0.18
2.0
0.05
0.03
0.03
0.07
0.80
0.40
0.19
0.25
0.39
0.39
0.18
0.09
6.5
0.05
0.04
0.03
0.07
0.82
0.41
0.20
0.25
0.40
0.39
0.19
0.09
NK119B
Pre
0.03
0.01
<0.01
0.03
0.54
0.20
0.05
0.10
0.20
0.26
0.09
0.04
0.25
0.03
0.01
<0.01
0.03
0.50
0.19
0.05
0.10
0.19
0.24
0.09
0.04
2.0
0.02
0.01
<0.01
0.03
0.45
0.17
0.04
0.08
0.16
0.21
0.08
0.04
6.5
0.02
<0.01
<0.01
0.02
0.35
0.13
0.03
0.07
0.13
0.17
0.06
0.03
NK100C
Pre
0.71
0.48
0.39
0.54
3.48
1.91
1.11
1.24
1.75
2.20
1.24
0.88
0.25
0.79
0.53
0.43
0.61
3.90
2.15
1.24
1.39
1.96
2.47
1.39
0.98
2.0
0.81
0.54
0.44
0.62
3.97
2.19
1.27
1.41
2.00
2.52
1.42
1.00
6.5
1.12
0.75
0.61
0.86
5.50
3.03
1.75
1.95
2.77
3.49
1.96
1.39
NK100B
Pre
1.04
0.67
0.54
0.88
7.03
3.64
2.04
2.44
3.31
4.01
2.12
1.35
0.25
1.06
0.68
0.55
0.89
7.12
3.69
2.07
2.48
3.35
4.06
2.15
1.37
2.0
0.97
0.63
0.50
0.82
6.56
3.40
1.90
2.28
3.09
3.74
1.98
1.26
6.5
1.21
0.78
0.62
1.02
8.14
4.21
2.36
2.83
3.83
4.64
2.46
1.57
NK100A1
Pre
2.08
1.44
1.23
2.17
16.57
9.48
5.13
6.21
7.98
9.40
4.79
2.89
0.25
2.09
1.45
1.23
2.18
16.63
9.51
5.15
6.23
8.01
9.44
4.81
2.90
2.0
2.01
1.39
1.19
2.10
16.01
9.15
4.96
5.99
7.71
9.08
4.62
2.79
6.5
2.20
1.53
1.30
2.30
17.55
10.04
5.44
6.57
8.46
9.96
5.07
3.06
NK100A03
Pre
2.85
1.88
1.55
2.66
20.10
11.39
5.88
7.40
9.35
11.14
5.95
3.84
0.25
2.86
1.89
1.56
2.68
20.18
11.44
5.91
7.43
9.39
11.18
5.97
3.85
2.0
2.78
1.83
1.52
2.60
19.59
11.10
5.73
7.21
9.11
10.85
5.80
3.74
6.5
3.00
1.98
1.64
2.80
21.15
11.98
6.19
7.78
9.84
11.72
6.26
4.04
Notes:
a USGS 15302250.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-39
April 2013
-------
Chapter 7
Mine Footprint
Table 7 12. Measured mean monthly pre mining flow rates (m3/s) and estimated mean monthly flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for gages along Upper Talarik Creek.
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
UT100E
Pre
0.15
0.13
0.12
0.18
0.61
0.30
0.21
0.23
0.31
0.36
0.25
0.20
0.25
0.14
0.13
0.11
0.18
0.59
0.29
0.20
0.23
0.30
0.36
0.24
0.20
2.0
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
6.5
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
UT100D
Pre
0.32
0.28
0.22
0.55
1.95
1.02
0.62
0.78
1.03
1.18
0.74
0.52
0.25
0.29
0.25
0.20
0.50
1.77
0.93
0.56
0.71
0.94
1.07
0.68
0.47
2.0
0.18
0.15
0.12
0.30
1.06
0.56
0.34
0.42
0.56
0.64
0.40
0.28
6.5
0.04
0.03
0.03
0.07
0.23
0.12
0.07
0.09
0.12
0.14
0.09
0.06
UT100C2
Pre
1.32
1.15
0.93
2.06
6.64
4.04
2.40
2.81
4.21
4.69
2.98
2.05
0.25
1.29
1.13
0.91
2.02
6.50
3.96
2.35
2.75
4.12
4.59
2.91
2.01
2.0
1.18
1.03
0.83
1.85
5.94
3.62
2.15
2.52
3.77
4.20
2.66
1.84
6.5
1.04
0.91
0.74
1.63
5.26
3.20
1.90
2.23
3.33
3.71
2.36
1.63
UT100C1
Pre
1.74
1.55
1.28
2.51
7.43
4.29
2.76
3.30
4.67
5.26
3.67
2.61
0.25
1.71
1.52
1.26
2.47
7.30
4.22
2.72
3.25
4.59
5.17
3.60
2.57
2.0
1.59
1.41
1.17
2.30
6.78
3.92
2.52
3.02
4.27
4.81
3.35
2.39
6.5
1.44
1.28
1.06
2.08
6.15
3.55
2.29
2.74
3.87
4.35
3.04
2.16
UT100C
Pre
2.45
2.25
1.98
3.44
9.11
5.63
3.77
4.38
6.09
6.67
4.59
3.37
0.25
2.42
2.22
1.95
3.39
8.98
5.55
3.72
4.32
6.00
6.57
4.52
3.32
2.0
2.27
2.08
1.83
3.18
8.43
5.21
3.49
4.05
5.63
6.17
4.24
3.11
6.5
2.09
1.92
1.68
2.93
7.76
4.79
3.21
3.73
5.18
5.68
3.90
2.87
UT119A
Pre
0.76
0.75
0.74
0.78
0.88
0.82
0.80
0.81
0.86
0.88
0.84
0.80
0.25
0.69
0.68
0.67
0.71
0.80
0.74
0.72
0.74
0.78
0.80
0.76
0.73
2.0
0.67
0.66
0.65
0.69
0.77
0.72
0.70
0.71
0.75
0.78
0.73
0.70
6.5
0.60
0.60
0.58
0.62
0.69
0.65
0.63
0.64
0.68
0.70
0.66
0.63
UTIOOB"
Pre
3.62
3.31
2.88
4.79
12.80
7.40
5.13
6.48
7.82
9.08
6.33
5.00
0.25
3.53
3.23
2.81
4.68
12.49
7.22
5.00
6.32
7.63
8.86
6.18
4.88
2.0
3.34
3.05
2.65
4.42
11.80
6.82
4.72
5.97
7.21
8.37
5.83
4.61
6.5
3.07
2.81
2.44
4.06
10.86
6.28
4.35
5.50
6.63
7.70
5.37
4.24
Notes:
a USGS 15300250.
NA = not applicable: UT100E would be blocked by waste rock pile under Pebble 2.0 (Figure 7-16), and by mine pit under Pebble 6.5 (Figure 7-17).
Table 7 13. Measured minimum monthly pre mining
BWl
lillililliltil
IBW1
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
SK100G
Pre
0.11
0.08
0.07
0.04
0.08
0.20
0.08
0.08
0.06
0.22
0.18
0.12
0.25
0.05
0.04
0.03
0.02
0.03
0.09
0.04
0.04
0.02
0.10
0.08
0.05
2.0
0.03
0.02
0.02
0.01
0.02
0.05
0.02
0.02
0.01
0.06
0.05
0.03
6.5
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
SK100F
Pre
0.20
0.15
0.11
0.11
0.14
0.46
0.22
0.16
0.08
0.63
0.34
0.21
0.25
0.14
0.11
0.08
0.08
0.10
0.34
0.16
0.12
0.06
0.46
0.25
0.15
2.0
0.13
0.09
0.07
0.07
0.09
0.29
0.14
0.10
0.05
0.40
0.22
0.14
6.5
0.08
0.06
0.04
0.04
0.05
0.17
0.08
0.06
0.03
0.24
0.13
0.08
SK124A
Pre
0.00
0.00
0.00
0.00
0.00
0.09
0.00
0.00
0.00
0.23
0.04
0.00
0.25
0.00
0.00
0.00
0.00
0.00
0.11
0.00
0.00
0.00
0.30
0.05
0.00
2.0
0.00
0.00
0.00
0.00
0.00
0.12
0.00
0.00
0.00
0.31
0.05
0.00
6.5
0.00
0.00
0.00
0.00
0.00
0.19
0.00
0.00
0.00
0.48
0.08
0.00
SK100C
Pre
0.00
0.00
0.00
0.00
0.00
0.12
0.00
0.00
0.00
0.71
0.12
0.00
0.25
0.00
0.00
0.00
0.00
0.00
0.12
0.00
0.00
0.00
0.73
0.12
0.00
2.0
0.00
0.00
0.00
0.00
0.00
0.12
0.00
0.00
0.00
0.73
0.12
0.00
6.5
0.00
0.00
0.00
0.00
0.00
0.16
0.00
0.00
0.00
0.95
0.16
0.00
SK119A
Pre
0.12
0.08
0.05
0.05
0.07
0.45
0.23
0.13
0.09
0.45
0.23
0.13
0.25
0.12
0.08
0.05
0.05
0.07
0.45
0.23
0.13
0.09
0.45
0.23
0.13
2.0
0.11
0.08
0.05
0.05
0.07
0.44
0.23
0.13
0.09
0.44
0.23
0.13
6.5
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
SK100B1
Pre
0.60
0.40
0.27
0.27
0.38
1.51
1.12
0.67
0.51
2.10
1.16
0.66
0.25
0.59
0.40
0.26
0.26
0.37
1.48
1.10
0.66
0.50
2.06
1.14
0.65
2.0
0.58
0.39
0.26
0.26
0.36
1.46
1.08
0.65
0.49
2.02
1.12
0.64
6.5
0.54
0.36
0.24
0.24
0.34
1.35
1.00
0.60
0.45
1.88
1.04
0.59
SKIOOB"
Pre
1.13
0.85
0.65
0.65
0.79
2.49
1.64
1.25
1.02
3.54
1.93
1.22
0.25
1.11
0.83
0.64
0.64
0.78
2.45
1.61
1.22
1.00
3.48
1.89
1.20
2.0
1.10
0.82
0.63
0.63
0.77
2.41
1.59
1.21
0.99
3.42
1.86
1.18
6.5
1.02
0.76
0.58
0.58
0.71
2.23
1.47
1.12
0.91
3.17
1.73
1.09
Notes:
= USGS 15302200.
NA = not applicable: SK100G would be eliminated by TSF 2, and SK119A would be eliminated by TSF 3 under the Pebble 6.5 scenario.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-40
April 2013
-------
Chapter 7
Mine Footprint
Table 7 14. Measured minimum monthly pre mining flow rates (m3/s) and estimated minimum monthly flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for gages along the North Fork Koktuli River.
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
NK119A
Pre
0.08
0.07
0.06
0.04
0.01
0.30
0.21
0.14
0.12
0.20
0.12
0.10
0.25
0.06
0.05
0.05
0.03
0.01
0.23
0.16
0.10
0.09
0.15
0.09
0.07
2.0
0.03
0.02
0.02
0.01
0.00
0.11
0.07
0.05
0.04
0.07
0.04
0.03
6.5
0.03
0.03
0.02
0.02
0.00
0.11
0.08
0.05
0.04
0.07
0.04
0.04
NK119B
Pre
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.25
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
2.0
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
6.5
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
NK100C
Pre
0.33
0.34
0.23
0.13
0.38
0.71
0.54
0.46
0.37
0.99
0.50
0.41
0.25
0.37
0.38
0.26
0.15
0.43
0.80
0.60
0.52
0.41
1.10
0.56
0.46
2.0
0.38
0.38
0.26
0.15
0.43
0.81
0.61
0.53
0.42
1.13
0.57
0.47
6.5
0.52
0.53
0.36
0.21
0.60
1.12
0.85
0.73
0.58
1.56
0.79
0.65
NK100B
Pre
0.43
0.44
0.33
0.18
0.54
1.31
1.04
0.96
0.91
1.53
0.71
0.57
0.25
0.44
0.45
0.34
0.19
0.55
1.33
1.05
0.97
0.92
1.55
0.72
0.58
2.0
0.40
0.41
0.31
0.17
0.51
1.22
0.97
0.89
0.85
1.43
0.67
0.53
6.5
0.50
0.51
0.38
0.21
0.63
1.52
1.20
1.11
1.05
1.77
0.83
0.66
NK100A1
Pre
0.93
0.95
0.80
0.84
1.16
3.75
2.57
2.02
1.89
3.19
1.51
1.21
0.25
0.93
0.95
0.80
0.84
1.16
3.76
2.58
2.03
1.90
3.20
1.51
1.21
2.0
0.90
0.92
0.77
0.81
1.12
3.62
2.48
1.96
1.83
3.08
1.46
1.17
6.5
0.98
1.00
0.85
0.88
1.22
3.97
2.72
2.14
2.00
3.38
1.60
1.28
NKIOOA"
Pre
1.10
1.13
0.91
0.96
1.44
4.27
2.35
1.93
1.76
4.39
1.98
1.53
0.25
1.11
1.14
0.91
0.97
1.45
4.29
2.36
1.93
1.76
4.41
1.99
1.54
2.0
1.08
1.10
0.88
0.94
1.41
4.17
2.29
1.88
1.71
4.28
1.93
1.49
6.5
1.16
1.19
0.95
1.01
1.52
4.50
2.47
2.03
1.85
4.62
2.08
1.61
Notes:
a USGS 15302250.
Table 7 15. Measured minimum monthly pre mining flow rates (m3/s) and estimated minimum monthly flow rates (m3/s) under the Pebble 0.25, 2.0, and 6.5 mine scenarios, for gages along Upper Talarik Creek.
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
UT100E
Pre
0.09
0.09
0.08
0.07
0.10
0.15
0.14
0.12
0.11
0.17
0.16
0.12
0.25
0.09
0.09
0.07
0.06
0.10
0.14
0.14
0.11
0.11
0.16
0.16
0.12
2.0
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
6.5
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
UT100D
Pre
0.12
0.10
0.12
0.11
0.22
0.23
0.21
0.22
0.20
0.33
0.31
0.22
0.25
0.11
0.10
0.11
0.10
0.20
0.21
0.19
0.20
0.18
0.30
0.28
0.20
2.0
0.07
0.06
0.07
0.06
0.12
0.13
0.11
0.12
0.11
0.18
0.17
0.12
6.5
0.01
0.01
0.01
0.01
0.03
0.03
0.02
0.03
0.02
0.04
0.04
0.03
UT100C2
Pre
0.53
0.47
0.53
0.50
0.91
1.46
1.46
1.35
1.29
1.71
1.34
0.91
0.25
0.52
0.46
0.52
0.49
0.89
1.43
1.43
1.32
1.26
1.67
1.32
0.89
2.0
0.47
0.42
0.47
0.44
0.81
1.31
1.31
1.21
1.15
1.53
1.20
0.81
6.5
0.42
0.37
0.42
0.39
0.72
1.16
1.16
1.07
1.02
1.35
1.07
0.72
UT100C1
Pre
0.80
0.73
0.80
0.76
1.25
1.57
1.37
1.58
1.52
2.24
2.04
1.25
0.25
0.78
0.71
0.78
0.75
1.23
1.54
1.35
1.55
1.49
2.20
2.00
1.23
2.0
0.73
0.66
0.73
0.70
1.14
1.43
1.25
1.44
1.39
2.04
1.86
1.14
6.5
0.66
0.60
0.66
0.63
1.04
1.30
1.14
1.31
1.26
1.85
1.68
1.04
UT100C
Pre
1.55
1.48
1.37
1.42
2.02
2.85
2.50
2.40
2.37
3.03
2.36
1.83
0.25
1.53
1.46
1.35
1.40
1.99
2.81
2.46
2.36
2.33
2.98
2.32
1.80
2.0
1.44
1.37
1.27
1.32
1.87
2.64
2.31
2.22
2.19
2.80
2.18
1.69
6.5
1.32
1.26
1.17
1.21
1.72
2.43
2.13
2.04
2.01
2.58
2.01
1.56
UT119A
Pre
0.72
0.71
0.72
0.71
0.63
0.62
0.65
0.62
0.67
0.69
0.78
0.74
0.25
0.65
0.65
0.65
0.65
0.58
0.56
0.59
0.56
0.60
0.62
0.70
0.67
2.0
0.63
0.63
0.63
0.63
0.56
0.54
0.57
0.54
0.58
0.60
0.68
0.65
6.5
0.57
0.56
0.57
0.56
0.50
0.49
0.51
0.49
0.53
0.54
0.61
0.58
UTIOOB"
Pre
2.09
1.98
2.09
2.04
2.83
2.58
2.55
2.97
2.83
3.82
3.68
2.83
0.25
2.04
1.93
2.04
1.99
2.76
2.51
2.49
2.90
2.76
3.73
3.59
2.76
2.0
1.93
1.83
1.93
1.88
2.61
2.37
2.35
2.74
2.61
3.52
3.39
2.61
6.5
1.78
1.68
1.78
1.73
2.40
2.19
2.16
2.52
2.40
3.24
3.12
2.40
Notes:
° USGS 15300250.
NA = not applicable: UT100E would be blocked by waste rock pile under Pebble 2.0 (Figure 7-16), and by mine pit under Pebble 6.5 (Figure 7-17).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-41
April 2013
-------
Chapter 7
Mine Footprint
Table 7 16. Pre mining watershed areas, mine footprint areas, and flows in the mine scenario watersheds, for the Pebble 0.25 mine scenario.
Stream and Gage
Pre-Mining
Watershed Area (km2)
Mean Annual Unit Runoff
(m3/s*km2)
Runoff per Unit Area
(m3/yr*m2)
Mean Annual Runoff
(m3/yr)
Captured Flow in Discrete Pathway (%)
Volume from Water Balance (m3/yr)
Total Mine Footprint
Drainage Area (km2)
Mine Footprint other
than TSF, NAG, or PAG
(km2)
TSF 1 Footprint (km2)
NAG Waste Rock
Footprint (km2)
PAG Waste Rock
Footprint (km2)
86.0
10,226,468
Flow Volume Returned
through WWTP (m3/yr)a
9.3
1,102,582
Flow Volume Returned
as TSF Leakage (m3/yr)
4.7
556,145
Flow Volume Returned
as NAG Waste Rock
Leachate (m3/yr)
0.0
0
Flow Volume Returned
as PAG Waste Rock
Leachate (m3/yr)
Operational Flows
Flow Volume Remaining
Due to Mine Footprint
with No Returns (m3/yr)
Captured Flow Volume
Returned from Footprint
(m3/yr)
Total Flow Volume in
Stream During
Operations (m3/yr)
Change in Average
Annual Runoff (%)
South Fork Koktuli River
SK100G
SK100F
SK100CP2b (local runoff)
SK100CP2b (losses to UTC)C
SK100CP2b (net flow at gage)
SK124A
SK124CP"
SK100C
SKIOOCPI"
SK119A
SK119CPb
SK100B1
SKIOOB"
14
31
54
54
54
22
24
99
99
28
30
141
179
0.026
0.026
0.017
0.0009
0.026
0.024
0.024
0.013
0.013
0.036
0.036
0.026
0.029
0.81
0.83
0.55
0.28
0.83
0.75
0.75
0.42
0.42
1.12
1.12
0.82
0.91
11,618,000
25,842,000
29,470,000
-14,735,000
44,206,000
16,811,000
17,981,000
41,858,000
41,987,000
31,268,000
33,314,000
115,110,000
162,122,000
8.4
9.2
9.2
9.2
9.2
-
9.2
9.2
-
9.2
9.2
7.9
<0.01
-
-
-
-
-
-
-
-
-
-
-
0.5
0.8
-
-
-
-
-
-
-
-
-
-
-
-
-
5,113,000
-
-
-
-
-
-
-
-
-
211,000
345,000
-185,000
-
-
-
-
-
-
-
-
-
-
4,882,000
18,274,000
24,425,000
-12,212,000
36,637,000
16,811,000
17,981,000
37,970,000
38,099,000
31,268,000
33,314,000
107,635,000
153,801,000
211,000
556,000
371,000
-185,000
556,000
5,113,000
5,113,000
5,484,000
5,484,000
-
5,484,000
5,484,000
5,092,000
18,830,000
24,796,000
-12,398,000
37,193,000
21,924,000
23,094,000
43,454,000
43,583,000
31,268,000
33,314,000
113,119,000
159,285,000
-56
-27
-16
-16
-16
30
28
4
4
0
0
-2
-2
North Fork Koktuli River
NK119A
NK119CP2b
NK119B
NK119CPlb
NK100C
NK100B
NK100A1
NK100A6
20
22
10
33
64
97
219
277
0.035
0.035
0.013
0.028
0.021
0.025
0.026
0.025
1.09
1.09
0.40
0.88
0.66
0.78
0.83
0.80
21,515,000
24,452,000
4,081,000
28,996,000
41,853,000
76,408,000
182,297,000
220,715,000
5.9
5.9
0.6
6.5
0.1
6.7
6.7
6.7
<0.1
0.6
0.1
0.1
-
5.9
-
-
-
-
-
-
-
-
-
-
-
-
-
5,113,000
-
1,103,000
-
-
-
-
-
-
-
-
-
-
-
15,065,000
18,003,000
3,837,000
23,289,000
41,818,000
71,193,000
176,773,000
215,420,000
1,103,000
1,103,000
-
1,103,000
5,113,000
6,216,000
6,216,000
6,216,000
16,168,000
19,105,000
3,837,000
24,392,000
46,932,000
77,409,000
182,988,000
221,636,000
-25
-22
-6
-16
12
1
<1
<1
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119A (local runoff)
UT119A (gains from SFK)C
UT119A (net flow at gage)
UT100B'
7
30
131
157
183
10
10
10
221
0.036
0.025
0.023
0.022
0.024
0.035
0.048
0.084
0.028
1.12
0.80
0.71
0.69
0.77
1.11
1.52
2.63
0.89
7,996,000
24,201,000
92,734,000
107,971,000
141,213,000
10,813,000
14,735,000
25,549,000
196,182,000
0.1
2.7
2.7
2.7
2.7
-
2.7
0.1
2.6
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
185,000
-
-
-
-
-
-
7,838,000
22,008,000
90,788,000
106,088,000
139,104,000
10,813,000
12,212,000
23,026,000
191,225,000
-
-
-
185,000
185,000
185,000
7,838,000
22,008,000
90,788,000
106,088,000
139,104,000
10,813,000
12,398,000
23,211,000
191,411,000
-2
-9
-2
-2
-1
0
-16
-9
-2
Notes:
Blank values (-) indicate that values are either not applicable or are equal to zero.
3 WWTP discharges 50% of flow to South Fork Koktuli River, 50% of flow to North Fork Koktuli River (no WWTP flows are directed to Upper Talarik Creek).
b Confluence point where virtual gage was created because physical gage does not exist.
c 1/3 of total return flow from is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location. Interbasin transfer flows are represented by negative flow values from SK100CP2 (losses transferred to UTC) and equivalent positive flow values for UT119A (gains transferred from SFK).
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating; WWTP = wastewater treatment plant.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-42
April 2013
-------
Chapter 7
Mine Footprint
Table 7 17. Pre mining watershed areas, mine footprint areas, and flows in the mine scenario watersheds, for the Pebble 2.0 mine scenario.
Stream and Gage
Pre-Mining
Watershed Area
(km2)
Mean Annual Unit Runoff
(m3/s*km2)
Runoff per Unit Area
(m3/yr*m2)
Mean Annual Runoff
(myyr)
Captured Flow in Discrete Pathway (%)
Volume from Water Balance (m3/yr)
Total Mine Footprint
Drainage Area
(km2)
Mine Footprint other than
TSF, NAG, or PAG
(km2)
TSF 1 Footprint
(km2)
NAG Waste Rock
Footprint (km2)
PAG Waste Rock Footprint
(km2)
73.7
12,176,060
Flow Volume Returned
through WWTP (m3/yr)a
14.2
2,350,061
Flow Volume Returned as
TSF Leakage (m3/yr)
10.8
1,782,486
Flow Volume Returned as
NAG Waste Rock
Leachate (m3/yr)
1.3
216,082
Flow Volume Returned as
PAG Waste Rock
Leachate (m3/yr)
Operational Flows
Flow Volume Remaining
Due to Mine Footprint with
No Returns (m3/yr)
Captured Flow Volume
Returned from Footprint
(m3/yr)
Total Flow Volume in
Stream During Operations
(m3/yr)
Change in Average Annual
Runoff (%)
South Fork Koktuli River
SK100G
SK100F
SK100CP2b (local runoff)
SK100CP2b (losses to UTC)C
SK100CP2b (net flow at gage)
SK124A
SK124CP"
SK100C
SK100CPlb
SK119A
SK119CPb
SK100B1
SK100B"
14
31
54
54
54
22
24
99
99
28
30
141
179
0.026
0.026
0.017
0.009
0.026
0.024
0.024
0.013
0.013
0.036
0.036
0.026
0.029
0.81
0.83
0.55
0.28
0.83
0.75
0.75
0.42
0.42
1.12
1.12
0.82
0.91
11,618,000
25,842,000
29,470,000
-14,735,000
44,206,000
16,811,000
17,981,000
41,858,000
41,987,000
31,268,000
33,314,000
115,110,000
162,122,000
11.7
13.0
13.0
13.0
13.0
0.1
0.1
13.1
13.1
0.6
0.6
13.6
13.6
9.8
-
-
-
-
-
-
-
-
-
0.1
-
0.6
-
-
1.4
1.3
-
-
-
-
-
0.5
0.0
-
-
-
-
-
-
-
-
6,088,000
-
-
-
-
-
-
-
13,000
-
86,000
-
-
610,000
540,000
-
-383,000
-
-
-
-
215,000
1,000
-
-72,000
-
-
-
-
2,184,000
15,133,000
22,331,000
-11,165,000
33,496,000
16,746,000
17,916,000
36,320,000
36,448,000
30,620,000
32,666,000
103,992,000
149,745,000
825,000
1,366,000
911,000
-455,000
1,366,000
6,101,000
6,101,000
7,012,000
7,012,000
86,000
86,000
7,097,000
7,097,000
3,009,000
16,499,000
23,242,000
-11,621,000
34,862,000
22,846,000
24,017,000
43,332,000
43,460,000
30,706,000
32,752,000
111,089,000
156,842,000
-74
-36
-21
-21
-21
36
34
4
4
-2
-2
-3
-3
North Fork Koktuli River
NK119A
NK119CP2b
NK119B
NK119CPlb
NK100C
NK100B
NK100A1
NK100A6
20
22
10
33
64
97
219
277
0.035
0.035
0.013
0.028
0.021
0.025
0.026
0.025
1.09
1.09
0.40
0.88
0.66
0.78
0.83
0.80
21,515,000
24,452,000
4,081,000
28,996,000
41,853,000
76,408,000
182,297,000
220,715,000
14.8
14.8
1.7
16.5
0.2
17.0
17.5
17.5
0.1
1.7
-
0.2
0.3
14.7
-
-
-
-
0.5
-
-
-
-
-
-
-
-
-
-
6,088,000
2,177,000
-
-
-
-
75,000
-
-
-
-
-
-
-
-
5,406,000
8,343,000
3,396,000
14,576,000
41,716,000
63,071,000
167,751,000
206,775,000
2,177,000
2,177,000
-
2,177,000
6,088,000
8,265,000
8,340,000
8,340,000
7,583,000
10,520,000
3,396,000
16,753,000
47,804,000
71,336,000
176,090,000
215,115,000
-65
-57
-17
-42
14
-7
-3
-3
Upper Talarik Creek
UT100D
UT100C2
UT100C1
UT100C
UT119A (local runoff)
UT119A (gains from SFK)C
UT119A (combined)
UT100B'
30
131
157
183
10
10
10
221
0.025
0.023
0.022
0.024
0.035
0.048
0.084
0.028
0.80
0.71
0.69
0.77
1.11
1.52
2.63
0.89
24,201,000
92,734,000
107,971,000
141,213,000
10,813,000
14,735,000
25,549,000
196,182,000
14.6
14.6
14.6
14.6
-
-
14.6
11.5
-
-
-
-
-
-
-
-
-
-
1.5
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
632,000
-
-
-
-
383,000
-
-
-
-
72,000
12,554,000
82,399,000
97,968,000
130,010,000
10,813,000
11,165,000
21,979,000
179,682,000
632,000
632,000
632,000
632,000
-
455,000
455,000
1,088,000
13,186,000
83,032,000
98,600,000
130,643,000
10,813,000
11,621,000
22,434,000
180,770,000
-46
-10
-9
-7
0
-21
-12
-8
Notes:
Blank values (-) indicate that values are either not applicable or are equal to zero. UT100E is blocked by the mine footprint in this scenario.
a WWTP discharges 50% of flow to South Fork Koktuli River, 50% of flow to North Fork Koktuli River (no WWTP flows are directed to Upper Talarik Creek).
b Confluence point where virtual gage was created because physical gage does not exist.
c 1/3 of total return flow from is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location. Interbasin transfer flows are represented by negative flow values from SK100CP2 (losses transferred to UTC) and equivalent positive flow values for UT119A (gains transferred from SFK).
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating; WWTP = wastewater treatment plant.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-43
April 2013
-------
Chapter 7
Mine Footprint
Table 7 18. Pre mining watershed areas, mine footprint areas, and flows in the mine scenario watersheds, for the Pebble 6.5 mine scenario.
Stream and Gage
Pre-Mining
Watershed Area
(km2)
Mean Annual Unit Runoff
(m3/s*km2)
Runoff per Unit Area
(m3/yr*m2)
Mean Annual Runoff
(m3/yr)
Captured Flow in Discrete Pathway (%)
Volume from Water Balance (m3/yr)
Total Mine Footprint
Drainage Area
(km2)
Mine Footprint other than
TSF, NAG, or PAG (km2)
E
Q.
s
£
•H sr
u- p
£ I
E
Q.
s
£
CM sr
u. p
£ I
TSF 3 Footprint
(km2)
+->
_c
a.
+->
o
£
J£
U
0
cc.
«
!~
« p
et =
Z £.
PAG Waste Rock Footprint
(km2)
82.4
49,409,253
Flow Volume Returned
through WWTP (m3/yr)a
12.0
7,187,867
Flow Volume Returned as
TSF Leakage (m3/yr)
3.9
2,367,714
Flow Volume Returned as
NAG Waste Rock Leachate
(m3/yr)
1.7
1,032,075
Flow Volume Returned as
PAG Waste Rock Leachate
(mVyr)
Operational Flows
Flow Volume Remaining
Due to Mine Footprint with
No Returns (m3/yr)
Captured Flow Volume
Returned from Footprint
(myyr)
Total Flow Volume in
Stream During Operations
(mVyr)
Change in Average Annual
Runoff (%)
South Fork Koktuli River
SK100F
SK100CP2b (local runoff)
SK100CP2" (losses to UTC)=
SK100CP2b (combined)
SK124A
SK124CPb
SK100C
SKIOOCPI"
SK119CPb
SK100B1
SK100B"
31
54
54
54
22
24
99
99
30
141
179
0.026
0.017
0.009
0.026
0.024
0.024
0.013
0.013
0.036
0.026
0.029
0.83
0.55
0.28
0.83
0.75
0.75
0.42
0.42
1.12
0.82
0.91
25,842,000
29,470,000
-14,735,000
44,206,000
16,811,000
17,981,000
41,858,000
41,987,000
33,314,000
115,110,000
162,122,000
22.0
22.0
22.0
22.0
9.8
9.8
31.8
31.8
19.1
52.5
52.5
2.6
0.1
-
-
-
-
0.0
"
0.1
-
-
-
0.6
1.4
-
-
-
18.5
1.6
0.0
8.2
-
-
-
-
2.6
-
-
-
-
-
2.4
-
-
-
-
-
24,705,000
-
-
-
-
2,000
-1,000
1,530,000
-
-
-
3,009,000
249,000
1,122,000
-374,000
-
-
-
-
-
1,032,000
-344,000
-
-
-
-
-
7,635,000
17,332,000
-8,666,000
25,999,000
9,417,000
10,587,000
28,349,000
28,477,000
11,891,000
72,284,000
114,449,000
2,156,000
1,437,000
-719,000
2,156,000
26,235,000
26,235,000
27,672,000
27,672,000
3,009,000
30,930,000
30,930,000
9,790,000
18,769,000
-9,385,000
28,154,000
35,651,000
36,822,000
56,020,000
56,149,000
14,900,000
103,214,000
145,379,000
-62
-36
-36
-36
112
105
34
34
-55
-10
-10
North Fork Koktuli River
NK119A
NK119CP2b
NK119B
NK119CPlb
NK100C
NK100B
NK100A1
NK100A6
20
22
10
33
64
97
219
277
0.035
0.035
0.013
0.028
0.021
0.025
0.026
0.025
1.09
1.09
0.40
0.88
0.66
0.78
0.83
0.80
21,515,000
24,452,000
4,081,000
28,996,000
41,853,000
76,408,000
182,297,000
220,715,000
14.8
14.8
3.5
18.2
0.5
19.1
19.6
19.6
0.1
-
3.5
-
0.5
0.3
14.7
-
-
-
-
0.5
-
-
-
-
-
0.0
-
-
-
-
-
-
-
-
-
-
-
-
-
24,705,000
2,314,000
-
5,000
-
-
79,000
-
-
-
-
-
-
-
-
5,406,000
8,343,000
2,682,000
13,017,000
41,502,000
61,419,000
166,001,000
205,099,000
2,314,000
2,314,000
5,000
2,319,000
24,705,000
27,023,000
27,103,000
27,103,000
7,719,000
10,657,000
2,688,000
15,336,000
66,207,000
88,443,000
193,104,000
232,201,000
-64
-56
-34
-47
58
16
6
5
Upper Talarik Creek
UT100D
UT100C2
UT100C1
UT100C
UT119A (local runoff)c
UT119A (gains from SFK)
UT119A (combined, at gage)
UT100B'
30
131
157
183
10
10
10
221
0.025
0.023
0.022
0.024
0.035
0.048
0.084
0.028
0.80
0.71
0.69
0.77
1.11
1.52
2.63
0.89
24,201,000
92,734,000
107,971,000
141,213,000
10,813,000
14,735,000
25,549,000
196,182,000
28.0
28.9
28.9
28.9
28.9
20.7
0.5
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1.8
0.4
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1,000
774,000
151,000
-
-
374,000
-
-
-
-
344,000
1,792,000
72,215,000
88,110,000
118,970,000
10,813,000
8,666,000
19,480,000
164,441,000
1,095,000
1,246,000
1,246,000
1,246,000
719,000
719,000
1,965,000
2,887,000
73,461,000
89,356,000
120,217,000
10,813,000
9,385,000
20,198,000
166,406,000
-88
-21
-17
-15
-0
-36
-21
-15
Notes:
Blank values (-) indicate that values are either not applicable or are equal to zero. UT100E is blocked and SK100G and SKllQAare eliminated by the mine footprint in this scenario.
a WWTP discharges 50% of flow to South Fork Koktuli River, 50% of flow to North Fork Koktuli River (no WWTP flows are directed to Upper Talarik Creek).
b Confluence point where virtual gage was created because physical gage does not exist.
c 1/3 of total return flow from is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location. Interbasin transfer flows are represented by negative flow values from SK100CP2 (losses transferred to UTC) and equivalent positive flow values for UT119A (gains transferred from SFK).
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating; WWTP = wastewater treatment plant.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-44
April 2013
-------
Chapter 7
Mine Footprint
Figure 7 18. Monthly mean streamf lows for stream gages in the (A) South Fork Koktuli River
(B) North Fork Koktuli River, and (C) Upper Talarik Creek watersheds, based on water years 2004
through 2010.
SK100G
•SK100F
•SK100C
•SK100B1
•SK100B
•SK119A
SK124A
B
•NK119A
•MK100B
NK100A1
•HK100A
•NK119B
•NK100C
•UT100D
•LIT100C1
•UT100C
-UT100B
-UT100E
LIT100C2
LIT119A
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-45
April 2013
-------
Chapter 7 Mine Footprint
For the three mine scenarios, it was assumed that some flows would be recovered from the mine
footprint. These recovered flows could be treated and returned as surface flow to downstream areas.
From the Pebble 0.25, 2.0, and 6.5 scenarios (Section 6.2), we estimated that recovery rates would be
73.5, 39.7, and 69.7% of the total water captured, respectively (Table 6-2). Figures 6-8 through 6-10
illustrate the various flowpaths expected under the three mine scenarios. For each of the watersheds,
the percentage of recovered flow was returned to the appropriate gage based on the expected flowpath
as defined by the mine scenarios. While some upper tributaries would experience reduced flows from
losses of watershed area, others would experience increased annual runoff from mining operation
discharges. The location of flow returns would be dictated by the facility that was expected to intercept
precipitation, the resultant runoff flow from those facilities, the processing of water for mine operations
(e.g., water use), and the proximity of the capture point relative to the drawdown area.
Precipitation on and runoff from facilities in the drawdown area would be captured and directed to the
WWTP. Much of the flow from facilities outside the drawdown area, such as leachate from TSFs and
waste rock piles, would be captured and directed to the WWTP but some would escape the collection
systems and flow back to the downstream receiving waters (Tables 7-16 through 7-18, Figures 6-8
through 6-10). It is important to note that the WWTP is designed to discharge to the South and North
Fork Koktuli River watersheds via the WWTP outfalls, so no treated flow from the Upper Talarik Creek
watershed would return to source streams in that watershed. The only exception is an area of interbasin
groundwater transfer that has been observed between the South Fork Koktuli River and Upper Talarik
Creek (PLP 2011: Chapter 7, Wobus et al. 2012). This transfer was included by allowing one third of the
return flows at gage SK100F to transfer to gage UT119A (Figures 6-8 through 6-10). It was assumed that
the baseflow transfer was accounted for in the gage record, and only the return flow volume was split
and transferred in this location. The spatial extent of these projected changes in streamflow and
implications for fish and aquatic habitat are discussed in Section 7.3.2.
7.3.1.1 Pebble 0.25 Scenario
Water balance estimates for the Pebble 0.25 scenario considered an operational facility that intercepts
precipitation from a footprint encompassing portions of the mine scenario watersheds (Table 7-16,
Figure 7-15). Based on these conditions, we estimate that in each watershed the uppermost gages
located in proximity to the mine footprint would experience the most significant reductions in
streamflow for this scenario. Overall, it is projected that 73.5% of captured watershed flows would be
returned (Table 6-3), but the location of return would vary depending on the mine needs for process
water and the location of mine facilities and water treatment (Table 7-16). In the Upper Talarik Creek
watershed under the Pebble 0.25 scenario, streamflow would be reduced by 2% at gage UT100E and 9%
at gage UT100D due to capture in the mine footprint. The most significant flow reductions in the South
Fork Koktuli River would be expected at gages SK100G (56%) and SK100F (27%) (Table 7-16). In the
North Fork Koktuli River, the greatest changes would be expected at gage NK119A (25% reduction)
(Table 7-16) as most of the watershed would be occupied by TSF 1 (Figure 7-15).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~46 April 2013
-------
Chapter 7 Mine Footprint
Reductions in streamflow due to capture of water in the mine footprint would be partially offset by
water returned via the WWTP, leakage through TSFs, and leaching through waste rock piles. Water
balance calculations for these components of the water budget are described in Chapter 6. Water
captured from headwater streams of the three watersheds would be treated at the WWTP and
discharged to gage SK124A in the South Fork Koktuli River and gage NK100C in the North Fork Koktuli
River (Figure 7-15). It is assumed that the WWTP would discharge equally to both outfalls, creating a
50/50 volume split for treated flows on an annual basis, but that on-site storage would allow
management of environmental flows to match seasonal hydrographs to the degree possible. Flows from
the WWTP outfalls would be projected to increase flows by 30% at gage SK124A, in a tributary to the
South Fork Koktuli River. In the North Fork Koktuli River watershed, flows would be projected to
increase by 12% at gage NK100C, downstream of the WWTP outfall. In the mainstem South and North
Fork Koktuli Rivers downstream of these points, WWTP outfall flows (approximately 5 million m3/year
from each outfall) (Table 7-16), leakage from the TSF and waste rock leaching would slightly offset
streamflow reductions expected from water capture within the mine footprint. Projected streamflow
changes for gages downstream of the WWTP outfalls are within 5% of pre-project streamflows
(Tables 7-16 and 7-19, Figure 7-15).
Because of the natural interbasin transfer of streamflow from the South Fork Koktuli River watershed to
the Upper Talarik Creek watershed (described above), decreased streamflows in the South Fork Koktuli
River resulting from capture by the mine footprint would translate to decreased rates of interbasin
transfer. As a result, there would be a projected 9% decrease in flow to the tributary of Upper Talarik
Creek where the interbasin transfer flows emerge (gage UT119A) (Tables 7-16 and 7-19, Figure 7-15).
7.3.1.2 Pebble 2.0 Scenario
Under the Pebble 2.0 scenario, area lost to the mine footprint would increase from the addition of a
second or expanded waste rock pile that would occupy much of the Upper Talarik Creek valley between
gages UT100E and UT100D (Figure 7-16). An expanded groundwater cone of depression would develop
around the larger excavated mine pit and further reduce water flowing to surrounding streams, and
TSF 1 would expand in size (Figure 7-16). Approximately 39.7% of the total water captured would be
returned to the three watersheds (Table 6-3). However, as for the Pebble 0.25 scenario described above,
flow returns in the upper watersheds would not necessarily be returned to their source stream reaches
but would be returned via the WWTP outfalls.
After accounting for water captured in the footprint, leakage, leachate, and returned water, flow
reductions in Upper Talarik Creek would be most severe for gage UT100D (46% reduction) (Table 7-
19). In the South Fork Koktuli River, gages SK100G, SK100F, and confluence point SK100CP2 would
experience reductions of 74, 36, and 21%, respectively. In the North Fork Koktuli River, the most severe
effects would be seen in the watershed occupied by TSF 1, with gages on this tributary predicted to
experience reductions in flow ranging from 42 to 65% (Tables 7-16 and 7-19, Figure 7-15). Factoring in
the flow return to streams from mining operations results in less severe reductions than if considering
only the percentage of surface area lost to the mine footprint (Table 7-17). Projections for the WWTP
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, .-, April 2013
-------
Chapter 7 Mine Footprint
flow contributions to the South Fork Koktuli River watershed would cause an increase in flow at gage
SK124A (36%) and the associated confluence point SK124CP (34%). WWTP contributions to the North
Fork Koktuli River watershed would cause a 14% flow increase at gage NK100C. At the lowermost gages
in each watershed, projected reductions in streamflow would be 8% (Upper Talarik Creek), 3% (South
Fork Koktuli River), and 3% (North Fork Koktuli River) (Table 7-19).
7.3.1.3 Pebble 6.5 Scenario
Under the Pebble 6.5 scenario, area lost to the mine footprint would increase with inclusion of a larger
pit and its associated drawdown zone, a substantially larger waste rock pile, and the development of TSF
2 on a tributary of South Fork Koktuli River upstream of gage SK100B1 and TSF 3 on a tributary
upstream of gage SK124A (Table 7-18, Figure 7-17). Gage SK100G would be eliminated under the
Pebble 6.5 waste rock piles, gage UT100E would be isolated above the mine footprint, and gage SK119A
would be buried under the TSF 2 dam (Figure 7-17). Efficiency of water recapture is estimated to be
69.7%, which would allow higher proportions of water captured in the footprint to be returned to
streams relative to the Pebble 2.0 scenario (Table 6-6). The net effects of lost effective watershed area
and recapture and release of water would result in streamflow reductions that would be most severe for
gages UT100D (88% reduction), SK100F (62% reduction), and NFK119A (64% reduction).
WWTP flows would be increased greatly over the Pebble 2.0 scenario and are projected to create an
increase in flow at SK124A (112%) and SK124CP (105%). This increase would continue to influence
flows downstream to gage SK100C (34% increase), but the large reduction attributed to the TSF on the
tributary measured by gage SK119A again creates a deficit, compared to pre-mining conditions, for
flows downstream at gages SK100B1 and SK100B (11% reductions) (Tables 7-18 and 7-19, Figure 7-
17). In the North Fork Koktuli River watershed, WWTP contributions would lead to flow increases of
58% at gage NK100C and increased flows at all downstream gages (Table 7-18). Upper Talarik Creek
would experience flow reductions of 15% or more at all mainstem gages. Upper Talarik Creek tributary
gage UT119A is projected to experience a 21% decrease in flow due to interbasin transfer of losses in
the South Fork Koktuli River watershed. At the lowermost gages in each watershed, projected changes in
streamflow are a 15% reduction for Upper Talarik Creek, 11% reduction in the South Fork Koktuli River
watershed, and an increase of 5% in the North Fork Koktuli River watershed.
7.3.1.4 Post-Closure
After the mine closes, pit dewatering would cease, leading to pit filling. As the pit fills, water from the pit
that had been returned to streams via pumping to the WWTP would no longer be available for
streamflow. This period is projected to last from about 20 years for the Pebble 0.25 scenario to about
300 years for the Pebble 6.5 scenario, after which the pit would reach equilibrium with surrounding
groundwater and pit water would flow into the groundwater system where the pressure gradient
allows. The pit water level could be lowered by pumping to maintain a hydraulic gradient toward the pit
for as long as water needed treatment. Absent active pumping, much of this groundwater would
eventually discharge to down-gradient streams, ponds, and wetlands (Section 6.3).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~48 April 2013
-------
Chapter 7
Mine Footprint
Post-closure streamflows would be a function of the pit cone of depression, and, as necessary, the
capture, treatment, and release of water through the WWTP. Temporary augmentation of streamflows
via TSF drawdown (Section 6.3) could be possible for a short time during this period. Given
uncertainties in the post-closure water balance, we have not attempted to estimate streamflows during
that period.
7.3.1.5 Uncertainties and Assumptions
Our assessment of streamflow changes distributes losses according to the percentage of the area lost to
the mine footprint in a given watershed. The analysis uses flow per unit area of measured data, and
allocates water routing through the three mine scenarios based on decisions about mine processes and
evaporation that will consume, treat, and and/or return water to the watersheds. We assume that
reduced flows would follow the same spatial patterns of gaining or losing groundwater reaches as would
initial (pre-mine) conditions. We acknowledge, however, that mine operations could alter the relative
importance of groundwater flowpaths, and therefore result in a different spatial distribution of
streamflow changes than we have reported.
Table 7 19. Estimated change in streamflow (%) and subsequent stream lengths affected (km) in
the mine scenario watersheds under the Pebble 0.25, Pebble 2.0, and Pebble 6.5 scenarios. Italics
indicates changes greater than 10% (moderate effects on salmon populations expected); bold
indicates changes greater than 20% (major effects on salmon populations expected).
Stream and
Gage
Pebble 0.25
Estimated
Change in
Streamflow
Stream Length
Affected
Pebble 2.0
Estimated
Change in
Streamflow
Stream Length
Affected
Pebble 6.5
Estimated
Change in
Streamflow
Stream
Length
Affected
South Fork Koktuli River— Mainstem
SK100G
SK100F
SK100CP2
SK100C
SK100CP1
SK100B1
SK100B3
-56
-27
-16
4
4
-2
-2
1.9
3.3
10.7
6.3
1.2
4.3
4.5
-74
-36
-21
4
4
-3
-3
0.5
3.3
10.7
6.3
1.2
4.3
4.5
NA
-62
-36
34
34
-10
-10
NA
0.8
10.7
6.3
1.2
4.3
4.5
South Fork Koktuli River— Tributaries
SK119A
SK119CP
SK124A
SK124CP
0
0
30
28
7.0
1.6
5.0
2.6
-2
-2
36
34
6.8
1.6
5.0
2.6
NA
-55
112
105
NA
1.5
5.0
2.6
North Fork Koktuli River— Mainstem
NKIOOC"
NK100B
NK100A1
NK100AC
12
1
< 1
< 1
4.5
0.8
20.4
8.4
14
-7
-3
-3
4.5
0.8
20.4
8.4
58
16
6
5
4.5
0.8
20.4
8.4
North Fork Koktuli River-Tributaries
NK119A
NK119CP2
NK119B
-25
-22
-6
0.6
1.3
6.8
-65
-57
-17
0.6
1.3
6.8
-64
-56
-34
0.6
1.3
6.6
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-49
April 2013
-------
Chapter 7
Mine Footprint
Table 7 19. Estimated change in streamflow (%) and subsequent stream lengths affected (km) in
the mine scenario watersheds under the Pebble 0.25, Pebble 2.0, and Pebble 6.5 scenarios. Italics
indicates changes greater than 10% (moderate effects on salmon populations expected); bold
indicates changes greater than 20% (major effects on salmon populations expected).
Stream and
Gage
NK119CP1
Pebble 0.25
Estimated
Change in
Streamflow
-16
Stream Length
Affected
0.4
Pebble 2.0
Estimated
Change in
Streamflow
-42
Stream Length
Affected
0.4
Pebble 6.5
Estimated
Change in
Streamflow
-47
Stream
Length
Affected
0.4
Upper Talarik Creek— Mainstem
UT100E
UT100D
UT100C2
UT100C1
UT100C
UTIOOB"
-2
-9
-2
-2
-1
-2
2.3
7.1
6.1
6.9
7.5
4.3
NA
-46
-10
-9
-7
-8
NA
2.1
6.1
6.9
7.5
4.3
NA
-88
-21
-17
-15
-15
NA
0.2
6.1
6.9
7.5
4.3
Upper Talarik Creek Tributary— Tributaries
UT119A
-9
6.5
-12
6.5
-21
6.5
Notes:
Stream lengths are typically calculated from the gage upstream to the next gage or the mine footprint (but see below); stream lengths affected
do not include portions of stream lost in the pit drawdown zone.
For gages UT100D, SK100G, SK100F, SK119A, SK124A, and NK119A, stream lengths include mainstem length upstream to edge of mine
footprint only, and do not include upstream lengths, including tributaries, that would be blocked or eliminated by the mine footprint.
a USGS 15302200.
b Upstream to wastewater treatment plant outfall point.
0 USGS 15302250.
d USGS 15300250.
NA = not applicable; the stream at the gage would be eliminated or blocked by the mine footprint
7.3.2 Exposure-Response: Streamflow
Water from streams originating upstream of the mine footprints (i.e., blocked streams) could be
captured at the footprint for use, or stored on site for eventual treatment and return to the stream
downstream of the footprint directly, or via the WWTP. Water from blocked streams would be returned
to downstream stream segments via diversion channels or pipes. Habitat upstream of the footprint (in
blocked streams) would no longer be accessible to fish downstream because of the inability offish to
move upstream through diversion channels or pipes.
7.3.2.1 Altered Streamflow Regimes
Altered streamflows can have various effects on aquatic life. Short-term effects include reduced habitat
availability resulting from water withdrawal (effects on winter habitat reviewed by West et al. 1992,
Cunjak 1996) and reduced habitat quality resulting from extreme and rapid fluctuations in flow if
withdrawals are intermittent (Curry et al. 1994, Cunjak 1996). Temporal variability in flows is a natural
feature of stream ecosystems (Poff et al. 1997), although the degree of variability differs depending on
hydrologic controls such as climate, geology, landform, human land use, and relative groundwater
contributions (Poff etal. 2006). Fish populations maybe adapted to periodic disturbances such as
droughts and may quickly recover under improved hydrologic conditions, but this is contingent on many
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-50
April 2013
-------
Chapter 7 Mine Footprint
factors (Matthews and Marsh-Matthews 2003). Longer-term effects of prolonged changes in streamflow
regime can have lasting impacts on fish populations (Lytle and Poff 2004).
The natural flow paradigm is widely supported and based on the premise that natural flow variability,
including the magnitude, frequency, timing, duration, rate of change and predictability of flow events
and the sequence of flow conditions, is crucial to maintaining healthy aquatic ecosystems (Postel and
Richter 2003, Arthington etal. 2006, Poff etal. 2009). However, numerous human demands can directly
alter natural flows, potentially affecting ecosystem function and structure. Guidelines for minimizing
impacts of altered hydrologic regimes have been offered by several researchers (Poff etal. 1997, 2009;
Richter 2010). Determining the natural flowregime is a data-intensive process, butitis crucial to
understanding how to manage flows within a system (Arthington et al. 2006).
Given the high likelihood of complex groundwater-surface water connectivity in the deposit area,
predicting and regulating flows to maintain key ecosystem functions associated with groundwater-
surface water exchange would be particularly challenging. PLP has invested in a relatively intensive
network of stream gages, water temperature monitoring sites, fish assemblage sampling sites,
groundwater monitoring wells, and geomorphic cross-section locations. The integration of information
gathered by this process will help identify relationships among surface water flow, groundwater and
surface water temperatures, and instream fish habitat (Bartholow 2010). However, until linkages
between biology, groundwater, surface water, and potential mining activities can be better evaluated,
predicted, and understood, a protective approach such as the sustainability boundary approach
described below is warranted to maintain surface-water and groundwater flows within natural flow
regimes across the mine scenario watersheds.
The sustainability boundary approach is a way to balance the maintenance of aquatic ecosystems with
human demands on the system (Richter et al. 2012). Under this approach, percentage-based deviations
from natural conditions are used to set flow alteration limits. These percentages are based on the
natural flow hydrologic regime and do not focus on the more simplistic approach of setting a percentage
based on a high-flow or low-flow event. Rather than a salmon-specific instream flow habitat model, this
is a system-based approach targeting the entire aquatic ecosystem. Numerous case studies have tested
this type of approach, and the percentage bounds of flow alteration around natural daily flow that
caused measurable ecological harm were determined to be similar regardless of geographic location
(Richter et al. 2012). Based on these studies, Richter et al. (2012) proposed that flow alteration be
managed based on the following thresholds of daily percentage flow alteration:
• Flow alteration below 10% would cause minor impacts on the ecosystem with a relatively high level
of ecosystem protection.
• Flow alteration of 11 to 20% would cause measurable changes in ecosystem structure and minor
impacts on ecosystem functions.
• Flow alteration greater than 20% would result in moderate to major changes in ecosystem
structures and functions. Increasing alteration beyond 20% would cause significant losses of
ecosystem structures and functions.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~5± April 2013
-------
Chapter 7
Mine Footprint
Losses could include reduced habitat availability for salmon and other stream fishes particularly during
low-flow periods (West et al. 1992, Cunjak 1996), reductions in macroinvertebrate production
(Chadwick and Huryn 2007), and increased fragmentation of stream habitats through increased
frequency and duration of stream drying. Increases in streamflow above background levels could result
in altered sediment transport dynamics with increased scour and transport of gravels. Increased
streamflows could also be associated with altered distributions of water velocities favorable for various
fish life stages. These alterations, depending on magnitude, could significantly decrease salmon habitat
quantity and quality in these watersheds (Figure 7-1).
We compared predicted flows for the Pebble 0.25, 2.0, and 6.5 scenarios (Tables 7-10 through 7-15)
with the sustainability boundary limits of 10 and 20% flow alterations around mean monthly flow. As an
example, mean monthly flows for the South Fork Koktuli River at gage SK100F during the pre-mining
period, projected flows under the Pebble 0.25 scenario, and the 10 and 20% sustainability boundaries
for the baseline flow are shown in Figure 7-19.
Figure 7 19. Monthly mean pre mining streamflow for South Fork Koktuli River gage SK100F (bold
solid line) illustrating 10 and 20% sustainability boundaries (gray lines) and projected monthly
mean streamflows under the Pebble 0.25 scenario (dashed line).
2.5
2.0
l.L.
1.0
0.5
0.0
We used this sustainability boundary approach to evaluate risks associated with potential flow
alterations throughout the mine scenario watersheds. To estimate the spatial extent of potential
deleterious streamflow alterations, we calculated the length of stream network upstream of the
uppermost stream gage to the edge of each mine footprint, and the length of each segment between
stream gages in each mine scenario watershed. This stream length is in addition to the length of stream
that would be eliminated or blocked by the mine footprint. That is, this and all subsequent references to
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-52
April 2013
-------
Chapter 7 Mine Footprint
stream lengths affected by flow modification reflect stream lengths downstream of the mine footprints
for each scenario, and so do not include stream lengths eliminated, blocked or dewatered by the
footprint described in Section 7.2. Table 7-19 summarizes estimated percent changes in streamflow at
each gage location, and the length of stream affected by each streamflow alteration under each mine
scenario. Figures 7-15 through 7-17 illustrate the spatial extent and location of flow alterations in
relation to gage sites. These estimates are for direct effects only. Stream sections throughout the stream
network could be affected indirectly, via reductions in flow downstream that could preclude use of
downstream habitats by fishes that move seasonally between headwater and mainstem habitats.
Similarly, these stream sections could be isolated by downstream flow reductions that reduce or
eliminate the potential for fish movement into those areas from downstream.
Pebble 0.25 Scenario
During operation of the Pebble 0.25 scenario, streamflow reductions exceeding 20% sustainability
boundaries would occur in 7 km of streams beyond the mine footprint. For these streams, substantial
reductions in fish habitat capacity and productivity could be expected. Streamflow increases greater
than 20% are expected for 8 km of stream downstream of the WWTP outfall, and would likely lead to
substantial changes in sediment dynamics and habitat suitability for fish. An additional 16 km of stream
would experience flow alterations exceeding 10%, with anticipated moderate effects on ecosystem
structure and function.
In the upper South Fork Koktuli River, gages SK100G and SK100F would experience 56 and 27%
reductions in flow, respectively, affecting 5 km of stream (Table 7-19). The tributary to the South Fork
Koktuli River receiving outfall from the WWTP would experience increased flows (28 to 30%), affecting
8 km of stream. In the North Fork Koktuli River, the tributary downstream of TSF 1 would experience
16 to 25% reductions in streamflow, affecting 2 km of stream (Table 7-19).
Several sections of the South Fork Koktuli River and tributaries below Frying Pan Lake are losing
reaches (i.e., discharge decreases in a downstream direction), which, under pre-mine conditions
experience periods of zero minimum monthly discharge (e.g., gage SK100C and WWTP outfall receiving
stream gage SK124A) (Table 7-13). We assumed that flow increases due to the WWTP would follow the
natural hydrograph, reflecting the amount of precipitation and runoff that must be captured and treated;
as a result, WWTP outfall flows would be lowest during periods when these streams typically go dry
based on pre-mine baseline data, and would be highest during period of snowmelt runoff and fall
storms.
Pebble 2.0 Scenario
Under the Pebble 2.0 scenario, streamflow reductions exceeding 20% sustainability boundaries would
occur in 19 km of stream downstream of the mine footprint. For these streams, substantial reductions in
fish habitat capacity and productivity would be expected. Increases in streamflow of 34 to 36% would
be expected for 8 km of stream downstream of the WWTP in the South Fork Koktuli River, and increases
of 14% would be expected for 4 km of the WWTP-receiving tributary to the North Fork Koktuli River,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~53 April 2013
-------
Chapter 7 Mine Footprint
and would likely lead to substantial changes in sediment dynamics and habitat suitability for fish. An
additional 6 km of stream in Upper Talarik Creek and 6.8 km of stream in the North Fork Koktuli River
would experience flow reductions of 10 to 20%, with anticipated moderate effects on ecosystem
structure and function.
Under the Pebble 2.0 scenario, the mine footprint captures 48% of the Upper Talarik watershed above
gage UT100D (Table 7-17). As a result, most of the total stream length in the upstream reaches of Upper
Talarik Creek watershed, including the mainstem and all tributaries above gage UT100D, would
experience either total loss of habitat from the mine footprint or indirect effects of fragmentation
(Section 7.2, Figure 7-10). Of this stream length, 2 km of mainstem downstream of the footprint would
experience a significant loss of habitat and decline in habitat quality from the predicted 46% reduction
in streamflow at gage UT100D (Figure 7-16). Downstream of gage UT100D in Upper Talarik Creek, flow
reductions would range from 8 to 11% (Table 7-19). Impacts on salmon habitat from flow reduction
would be moderated by inputs of tributary flow and groundwater that may help ameliorate flow losses
originating upstream, assuming that groundwater sources and flowpaths are not also altered by the
mine footprint. This assumption is questionable (Section 7.3.2.3). For instance, the groundwater-
dominated Upper Talarik Creek tributary monitored at gage UT119A would experience a 12% reduction
in flow due to reduced flow in portions of the South Fork Koktuli River resulting from losses to the mine
footprint. This was the only case of interbasin hydrologic connectivity explicitly modeled, but other
undocumented connections are likely to occur.
In the South Fork Koktuli River, reductions in streamflow would exceed the 20% sustainability
threshold at gages SK100G, SK100F, and SK100V2 (Table 7-19, Figure 7-16). In the South Fork Koktuli
River mainstem and tributaries upstream of gage SK100G, the majority of stream length would be
eliminated by the mine footprint (Figure 7-16), resulting in severe flow reductions at gages SK100G
(74%) and SK100F (36%) (Table 7-19). Streamflows in the South Fork Koktuli River at gage SK100C
would increase by 4% because of releases from the WWTP discharging into tributary gage SK124A,
which would experience a 34% increase in flow at the confluence with the South Fork Koktuli River
(Table 7-9, Figure 7-17).
In the North Fork Koktuli River, the majority of stream length above gage NK119A would be eliminated
by construction of TSF 1 (Figure 7-17), resulting in substantial loss in flow (65% reduction at gage
NK119A) for approximately 2 km of stream between TSF 1 and the North Fork Koktuli River (Table 7-9,
Figure 7-17). Approximately 7 km of stream length in the tributary measured by gage NK199B would be
expected to experience 17% reductions in flow. Increases in streamflow downstream of the WWTP
discharge point are projected to increase flows by 14% in 4 km of the North Fork Koktuli River
upstream of gage NK100C (Table 7-9, Figure 7-17).
Pebble 6.5 Scenario
The Pebble 6.5 scenario would capture an even larger portion of the South and North Fork Koktuli
Rivers and Upper Talarik Creek watersheds in its footprint. During operation of the Pebble 6.5 scenario,
streamflow reductions exceeding 20% sustainability boundaries would occur in 35 km of stream. For
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~54 April 2013
-------
Chapter 7 Mine Footprint
these streams, substantial reductions in fish habitat capacity and productivity could be expected. An
additional 28 km of stream in Upper Talarik Creek would experience flow reductions exceeding 10%,
with anticipated moderate effects on ecosystem structure and function. Increases in streamflow
exceeding 20% are expected for 15 km of stream downstream of the WWTP in the South Fork Koktuli
River and for 4 km of the WWTP-receiving tributary to the North Fork Koktuli River, and would likely
lead to substantial changes in sediment dynamics and habitat suitability for fish.
In the Upper Talarik Creek watershed, substantial flow reductions are projected at gages UT100D (88%)
and UT100C2 (21%), affecting 6 km of stream. Moderate streamflow alterations exceeding 10% would
occur in an additional 19 km of stream at gages UT100C1, UT100C, and UT100B (Table 7-19). In the
South Fork Koktuli River, gages SK100G and SK119A would be buried under the expanded mine
footprint. A 65% reduction in flow would be expected for 1 km of the upper South Fork Koktuli River
downstream from the edge of the waste rock to gage SK100F (Table 7-19, Figure 7-17).
Under the Pebble 6.5 scenario, the WWTP is estimated to discharge over 50 million m3 of water per year
(Table 7-18). This discharge would result in a 34% increase in flow for 8 km in the South Fork Koktuli
River above gage SK100CP1, and a 105% increase in flow for 8 km of stream above gage SK124CP
(Table 7-1, Figure 7-17). In the North Fork Koktuli River, WWTP outfalls would result in a 58% increase
in flows for 4 km of tributary above gage NK100C, and a 16% increase in flows for 1 km of stream
upstream of gage NK100B.
Flow reductions and stream habitat losses of the magnitudes estimated under the Pebble 0.25, 2.0, and
6.5 scenarios represent substantial risks to spawning and rearing habitat for populations of coho,
sockeye, and Chinook salmon, Dolly Varden, and rainbow trout in the upper portions of these
watersheds. Habitat quantity and quality would be significantly diminished by the loss of flow from the
mine footprint, resulting from multiple mechanisms such as direct reduction in habitat area and volume,
the loss of channel to off-channel habitat connectivity, increased periods of zero flow, and reduced food
production. Increases in flow could alter channel morphologies, induce higher rates of sediment
transport and erosion, and change the distribution of water velocities within habitats used by spawning
and rearing salmon and other fishes. Although the loss of salmonid production cannot be estimated, flow
alterations greater than 20% would be expected to have substantial effects based on those mechanisms
and on the substantial effects on stream structure and function (Figure 7-1) (Richter etal. 2012).
7.3.2.2 Connectivity and Timing and Duration of Off-Channel Habitats
Loss of streamflow resulting from the mine footprints and potential water withdrawals described above
would affect connectivity between the main channel and off-channel habitats important to juvenile
salmonids. Loss of flood peaks could alter groundwater recharge rates, influencing characteristics of
floodplain percolation channels, seeps, or other expressions of the hyporheic zone (Hancock 2002).
Rapid reductions in streamflow that exceed recession rates typically experienced by fishes in these
systems could result in stranding or isolation in off-channel habitats (Bradford et al. 1995). Off-channel
habitats, particularly those with groundwater connectivity, are critical rearing habitats for several
species of juvenile salmonids and can be important spawning habitats for sockeye salmon (Quinn 2005).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~55 April 2013
-------
Chapter 7 Mine Footprint
Maintaining connectivity and the physical and chemical attributes of these habitats in conditions similar
to baseline conditions will be important for minimizing risks to salmon and other native fishes.
Wetlands that are hydrologically connected to affected streams would also respond to alterations in
streamflow and groundwater. Fish access to and use of wetlands are likely to be extremely variable in
the footprint areas because of differences in the duration and timing of surface water connectivity with
stream habitats, distance from the main channel, or physical and chemical conditions (e.g., dissolved
oxygen concentrations) (Kingetal. 2012). Projecting the effects of lost wetland connectivity and
abundance on stream fish populations is beyond the scope of this assessment, but could be a significant
unknown.
Flow regulation through the WWTP could be designed to somewhat approximate natural hydrologic
regimes during periods when sufficient water and water storage capacity was available, which could
provide appropriate timing and duration of connectivity with off-channel habitats. Channel cross-
section data and gage data (PLP 2011) would provide useful insights into flow-connectivity
relationships and could help guide a flow management plan.
7.3.2.3 Changes in Groundwater Inputs and Importance to Fish
There is limited information describing potential surface water-groundwater interaction in the mine
scenario watersheds, but groundwater is likely the dominant source of streamflow in these streams
(Rains 2011), and locally can be very important (Figure 7-14). High baseflow levels in the monthly
hydrographs of the mine scenario watersheds illustrate groundwater's important influence on these
streams (Figure 3-10).
Aerial winter open-water surveys (Figure 7-14) consistently suggest the presence of upwelling
groundwater, which maintains ice-free conditions in portions of area streams and rivers. Highly
permeable glacial outwash deposits create a complex mosaic within less permeable, clay to silt
dominated Pleistocene lake deposits and bedrock outcrops, which can control surface water-
groundwater interactions in landscapes like this one (Power et al. 1999). Mine operations that reduce
surface water contributions in the natural drainage course, or that lower groundwater tables, may
influence groundwater paths and connections within and among streams in the mine area in ways that
are not predicted in this assessment, but that could have significant impacts on fish. In our analyses of
the water management regimes for the mine scenarios, we projected increasing proportions of
streamflow derived from water released from the WWTP as the mine develops. These increased releases
would result from increased interception of groundwater associated with the mine pit cone of
depression, rainwater, and surface runoff collection. Water treated and discharged would be replacing a
portion of the groundwater that would otherwise be feeding stream systems, and could have
substantially different chemical characteristics (Chapter 8).
Fish in the region are highly attuned to groundwater signals in the hydrologic and thermal regimes
(Power et al. 1999). Spatial heterogeneity in flow and temperature, largely mediated by groundwater-
surface water exchange, provides a template for diverse sockeye salmon life histories and migration
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~56 April 2013
-------
Chapter 7 Mine Footprint
timing (Hodgson and Quinn 2002, Rogers and Schindler 2008, Ruff etal. 2011). For example,
groundwater moderates winter temperatures, which strongly control egg development and hatch and
emergence timing (Brannon 1987, Hendry et al. 1998). Spatial thermal heterogeneity allows diverse
foraging strategies for consumers of sockeye salmon and their eggs such as brown bear and rainbow
trout, thereby benefitting not only sockeye salmon populations, but also the larger food web (Armstrong
etal. 2010, Ruff etal. 2011).
Altered groundwater contributions to surface waters in the mine area could have profound effects on
the thermal regimes and cued life histories of aquatic biota. Curry et al. (1994) examined the influence of
altered hydrologic regimes on groundwater-surface water interchange at spawning locations for brook
trout in an Ontario stream. Responses of groundwater-surface water exchange to changes in river
discharge varied among sites, precluding predictable responses. The complexity that can be inherent in
groundwater-surface water interactions can make regulating or controlling such interactions during
large-scale landscape development very difficult (Hancock 2002). Adequately protecting the critical
services that groundwater provides to fish is complicated by the fact that flowpaths vary at multiple
scales, and connections between distant recharge areas and local groundwater discharge areas are
difficult to predict (Power et al. 1999).
7.3.2.4 Stream Temperature
Projecting specific mining-associated changes to groundwater and surface water interactions and
corresponding effects on surface water temperature in the mine area is not feasible at this time.
Disruptions or changes to groundwater flowpaths could have significant adverse effects on winter
habitat suitability for fish, particularly if groundwater-dominated stream reaches are converted to
surface water-dominated systems. Irons etal. (1989 as cited in Reynolds 1997) reported that
groundwater-mediated unfrozen refugia were dependent on fall rains maintaining groundwater, but
that during a dry year, groundwater levels declined and allowed full freezing of stream surface waters
and the streambed. This suggests that the threshold between completely frozen and partially frozen
streams can be a narrow one, particularly for small streams with low winter discharge. The duration of
freezing and the extent and type of ice formation, including anchor ice, frazil, or surface ice (Slaughter
1990) can severely limit habitat availability during the winter and spring months. Maintaining winter
groundwater connectivity may be critical for fish in such streams (Cunjak 1996, Huusko et al. 2007,
Brown etal. 2011).
7.3.3 Risk Characterization
The water budget predicted for our mine scenarios would require large volumes of water from surface
streams or groundwater, necessitating alterations to streamflows. Streamflow alterations exceeding
20% will occur in 15, 26, and 54 km of streams under the Pebble 0.25, 2.0, and 6.5 scenarios,
respectively, leading to significant adverse effects on fish and other aquatic life. The seasonal timing and
magnitude of streamflow alterations would be contingent on water-storage and management systems
and strategies, but would be constrained by the fundamental needs for water at specific times and
locations within the mining process (Chapter 6). Impacts on fish habitat and fish populations would
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment -, „ April 2013
-------
Chapter 7 Mine Footprint
likewise depend on the magnitude and timing of flow changes, but would be most severe for streams
close to the mine footprint, particularly during phases of the mine life cycle when water processing
demands would be high and storage and distribution options would be limited.
The volume of water that would require treatment by the mine WWTP would range from
10 million m3/yr for Pebble 0.25 (Table 7-16) to over 50 million m3/yr for Pebble 6.5 (Table 7-18). To
avoid or minimize risks associated with altered streamflows in downstream effluent-receiving areas,
capacity for water storage and release would be required to maintain natural flow regimes or any
minimum flows required by regulatory agencies. Application of the Instream Flow Incremental
Methodology (IFIM) Physical Habitat Simulation (PHABSIM) system modeling approach (Bovee 1982,
1998) is being used by PLP to assess streamflow-habitat relationships (PLP2011: Chapter 15), and
could provide additional guidance for establishing flow requirements beyond those identified in this
document (Estes 1998).
Maintenance of mine discharges in terms of water quality, quantity, and timing, to avoid adverse impacts
would require long-term commitments for monitoring and facility maintenance. As with other long-term
maintenance and monitoring programs, the financial and technological requirements could be very
large, and the cumulative risks (and likely instantaneous consequences) of facility accidents, failures,
and human error would increase with time. Additionally, climate change and projected changes in
temperature and precipitation for the region (Section 3.7) would result in potential changes in
streamflow magnitude and seasonality and would interact with mining-related flow impacts (Box 14-2),
requiring adaptation to potentially new flow regimes. We know of no precedent for the long-term
management of water quality and quantity on this scale at an inactive mine.
7.3.4 Uncertainties and Assumptions
Projecting changes to groundwater-surface water interactions in the footprint area with any specificity
is not feasible at this time. Local geology and stream hydrographs are indicative of systems that are
largely driven by groundwater. Disruptions or changes to groundwater flowpaths in the footprint area
could have significant adverse effects on winter habitat suitability for fish, particularly if ground water-
dominated stream reaches are converted to stream reaches dominated by effluent from a WWTP. Given
the high likelihood of complex groundwater-surface water connectivity in the mine area, predicting and
regulating flows to maintain key ecosystem functions associated with groundwater-surface water
exchange would be particularly challenging.
Our approach for assessing potential risks of flow alteration rests on simplifying assumptions regarding
changes to the natural streamflow regime under the mine scenarios (Section 7.3.2). The natural flow
regime consists of multiple components, including flow magnitude, frequency, duration, timing, and rate
of change, all of which can have important implications for fish and other aquatic life (Poff etal. 1997).
We were unable to anticipate changes to the streamflow regime beyond simplistic alterations in flow
magnitude, yet it is very likely that other aspects of the flow regime would be modified as well,
depending on how flows respond to water management at the mine site. Our analysis does not account
for these possibilities.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~58 April 2013
-------
Chapter 7 Mine Footprint
We assumed that flow modifications would follow the natural hydrograph, reflecting the amount of
precipitation and runoff that was intercepted, and must be captured and treated. As a result, WWTP
outfall flows are lowest during periods when these streams typically go dry based upon pre-mine
baseline data, and are highest during period of snowmelt runoff and fall storms. Alternative flow
management strategies may be feasible, depending on the capacity to store and release flows to meet
environmental flow objectives.
Additionally, we assume that larger deviations from the natural flow regime pose greater risks of
ecological change. The scientific literature supports this assumption as a general trend (Poff et al. 2009,
Poff and Zimmerman 2010, Richter etal. 2012); however, as pointed out by Poff and Zimmerman
(2010), specific responses to changes in streamflow vary. While all stream studies reviewed by Poff and
Zimmerman (2010) showed declines in fish abundance, diversity, and demographic rates with any level
of flow modification, other ecological responses (e.g., macroinvertebrate abundance, riparian vegetation
metrics) sometimes increased. Responses offish populations and other ecological metrics to flow
modification would be dependent on a suite of interacting factors, including but not limited to stream
structural complexity, trophic interactions, and the ability of fish to move seasonally (Anderson et al.
2006).
One way to estimate the magnitude of uncertainty in an assessment is to compare it to an independently
produced estimate. An assessment of hydrologic and water quality issues at the Pebble deposit was
independently performed by Wobus etal. (2012). They used the same set of available data (primarily
PLP 2011) and based their modeling on the same preliminary mine plan (Ghaffari etal. 2011) as this
assessment. Where assumptions were similar between this assessment and Wobus et al. (2012)
modeling efforts, streamflow modification projections were similar (e.g., gages UT100C2, SK100G, and
NK119A; Table 7-20). However, model results were very sensitive to the location of WWTP discharges.
For example, in this assessment we estimated reductions in streamflow of 46% in Upper Talarik Creek
at gage UT100D (Tables 7-19 and 7-20). Wobus etal. (2012) estimated much less severe reductions of
less than 10%, largely, because their assessment placed one of the two WWTP outfall points atthis
location (Table 7-20) and ours did not. Other significant divergences between streamflow alteration
estimates in this assessment and Wobus et al. (2012) also are most likely due to differences in the
location of the WWTP outfalls (Table 7-20).
The potential impacts of the mine footprints discussed in this chapter do not explicitly take into account
the effects of climate change. Over the time scale at which large-scale mining will potentially affect the
assessment area, projected increases in temperature and precipitation are likely to substantially change
the physical environment (Section 3.8 and Box 14-2). Such changes could significantly alter the
variability and magnitude of stream flows. Increases in rain-on-snow events are likely, but the
consequence to large flooding is unclear. Nevertheless, these changes in streamflow regime will likely
lead to changes in sediment transport, bed stability, and channel morphology with potential adverse
impacts to fish habitat and population genetic diversity and resiliency.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7~59 April 2013
-------
Chapter 7
Mine Footprint
Table 7 20. Estimated change in streamf low (%) at selected stream gages. Estimates made for this
assessment are based on the Pebble 2.0 scenario; estimates made by Wobus et al. (2012) are for a
similar mine size, although details of water management and spatial arrangement of mine
infrastructure may differ.
Stream and Gage
This Assessment
Wobus etal. 2012
Potential Source of Discrepancy Between
Estimates
South Fork Koktuli River
SK100G
SK100F
SK100C
SK100B1
-74
-36
4
-3
-50 or greater reduction
-20 to -50
-10 to -20
-10 to -20
NA
NA
Water surplus in this assessment is discharged
from WWTP above this point
Water surplus in this assessment is discharged
from WWTP above this point
North Fork Koktuli River
NK119A
NK100C
-65
14
-50 or greater reduction
0 to -10
NA
Water surplus in this assessment is discharged
from WWTP above this point
Upper Talarik Creek
UT100D
UT100C2
-46
-11
0 to -10
-10 to -20
Water surplus in Wobus et al. (2012) modeling
is discharged from WWTP at this site
NA
7.4 Summary of Footprint Effects
Streams eliminated, blocked, or dewatered by the mine footprints under the Pebble 0.25, 2.0, and 6.5
scenarios would result in the loss of 8, 24, or 35 km, respectively, of documented anadromous waters as
defined in the AWC (Johnson and Blanche 2012). An additional 30 to 110 km of headwater streams
supporting habitat for non-anadromous fish species would be lost to the mine footprint under these
scenarios. Loss of headwater streams to the footprints would alter groundwater-surface water
hydrology, nutrient processing, and export rates of resources and materials for aquatic ecosystems
downstream. Losses of wetland habitat would be 5 km2 of wetlands under the Pebble 0.25 footprint,
12.4 km2 under the Pebble 2.0 footprint, and more than 19 km2 of wetland habitat under the Pebble 6.5
footprint. An unquantified area of riparian floodplain wetland habitat would either be lost or suffer
substantial changes in hydrologic connectivity with streams because of reduced flow from the mine
footprint.
Reduced flow resulting from water retention for use in mine operations, ore processing, transport, and
other processes, would further reduce the amount and quality offish habitat. Reductions in streamflow
exceeding 20% would adversely affect habitat in an additional 15, 26, and 54 km of streams under the
Pebble 0.25, 2.0, and 6.5 scenarios, respectively, further reducing production of coho salmon, sockeye
salmon, Chinook salmon, rainbow trout, and Dolly Varden. Losses of stream habitat leading to losses of
local, unique populations will erode the population diversity that is essential to the stability of the
overall Bristol Bay salmon fishery (Schindler et al. 2010).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
7-60
April 2013
-------
The water collection, treatment, and discharge scenarios presume that, under routine operations, runoff
water, leachate, and wastewater would be collected and properly treated before release to meet state
and federal criteria and permit requirements. However, some leachate would escape collection, and
some treatment failures would occur. This chapter begins with a description of the potential sources of
contaminants (Section 8.1) and then describes potential routes and magnitudes of exposure to
contaminated water (Section 8.2). It describes the exposure-response relationships used to screen
leachate constituents and considers toxicology of the major contaminant of concern, copper, in greater
detail (Section 8.3). The chapter ends with a characterization of the potential risks from aqueous
effluents (Section 8.4) and discussions of potential additional remediation and uncertainties
(Sections 8.5 and 8.6). Figure 8-1 illustrates potential linkages between sources, stressors, and
responses associated with water treatment, discharge, fate, and effects that are evaluated in this chapter.
8.1 Water Discharge Sources
Discharges were calculated for routine operations and wastewater treatment plant (WWTP) failure
under the Pebble 0.25, 2.0, and 6.5 mine scenarios; post-closure discharges are discussed qualitatively.
Sources of water discharge under routine operations associated with each mine scenario include
effluents discharged from the WWTP and uncollected leachates from the tailings storage facilities (TSFs)
and waste rock piles. Other routine sources, including site runoff and domestic wastewater, are
considered to be minor and are not analyzed here. In addition, we evaluate a WWTP failure scenario, in
which the system releases untreated wastewater. This failure represents one potential failure among
many accidents and failures that could occur. We specify that under routine operations, the WWTP
would meet permit limits; in the event of a complete treatment failure, flows would pass through the
WWTP at the estimated influent concentrations. These two water collection, treatment, and discharge
scenarios bound the likely degrees of water treatment failure, but do not encompass the worst case. For
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-1
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
example, treatment might fail when waste water composition is worse than average or an extreme
accident like dumping reverse-osmosis brine could occur.
Folio wing the termination of mine operations, it is expected that water collection and treatment would
continue for waste rock and tailings leachates. If the water is nontoxic, in compliance with all criteria
and standards, and its composition is stable or improving, the collection and treatment system may be
shut down under permit. Otherwise, treatment would continue in perpetuity—that is, until untreated
water quality was acceptable or institutional failures ultimately resulted in abandonment of the system.
If the mine operator abandons the site, the State of Alaska should assume operation of the treatment
system; if both the mine operator and the State of Alaska abandon the site, untreated leachate would
flow to streams draining the site.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ » APril 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Figure 8 1. Conceptual model illustrating the pathways linking water treatment, discharge, fate,
and effects.
effluent dilution
& transport
>.• ate r treatment
facilities
vaste rock ^1 | tailings storage
piles ) \ facilities
M/
I water I I
treatment ]
V
V
failure of vvater
treatment facilities
V
leakage of
I each ate
V
contaminated vvater
discharges
4 salmonicl fish
(ah LI n d an c e. p r o d u cti vity o r cl i v e r sit;)
untreated vvater
discharges
additional step in
causal pathway
factor
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-3
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.1.1 Routine Operations
Under the mine scenarios (Section 6.1.2.5), water in contact with tailings, waste rock, ore, product
concentrate, or mine walls would leach minerals from those materials. In addition, chemicals would be
added to the water used in ore processing. Most of the water used to transport tailings or concentrate or
used in ore processing would be reused. Leachates collected from TSFs or waste rock piles would be
stored in the TSF or treated for use or discharge, but leachate that escaped collection would flow to
streams (Figure 6-5). Waste rock used in the construction of dams, berms, roads and other mine
structures would be leached by rain and snowmelt, but that source is small relative to the waste rock
piles. Waste rock leachates are assumed to have the mean concentrations of reported humidity cell tests
(Appendix H and PLP 2011). Mine pit water would also be used or treated for disposal. Surplus water on
the site would be treated to meet applicable standards and discharged under permit. Based on Alaskan
Water Quality Standards, defined in the Alaska Administrative Code, Title 18, Section 70, no mixing
zones would be authorized for anadromous streams or spawning habitat for most game or subsistence
fish species, so it is expected that effluents would be required to meet national criteria, state standards,
and other permit limits (i.e., no exemptions would be granted).
During mine operations, water available on the site would exceed operational needs, and 10 to
49 million m3 of treated water would be discharged per year (Table 6-3). The mine scenarios specify
that effluent would be discharged to the South and North Fork Koktuli Rivers (Tables 8-1 through 8-3).
The effluent could contain treated tailings leachate, waste rock leachate, mine pit water, and excess
transport or process waters. Tailings leachate would come from the TSFs as either excess water in the
impoundment or leakage captured below the dams. The primary concern during routine operations
would be waste rock leachate. Captured waste rock leachate would become more voluminous as the
waste rock piles increased during operation. After mine closure, it would be a major source of routinely
generated wastewater, along with water pumped from the TSFs and the pit (after it has filled). In
addition, because the waste rock piles and TSFs would not be lined, some leachates from both would not
be captured and would flow to the three receiving streams.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ . April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 1. Effluent and receiving water flows at each gage under the Pebble 0.25 scenario. All values are presented in m3/yr.
Stream and Gage
Flow Returned
Through WWTP"
Flow Returned as
TSF Leakage
Flow Returned as NAG
Waste Rock Leachate
Flow Returned as PAG
Waste Rock Leachate
Flow of Interbasin
Transfer
TOTAL FLOW
South Fork Koktuli River
SK100G
SK100F
SK100CP2b'c
SK124A
SK124CPb
SK100C
SKIOOCPI"
SK119A
SK119CPb
SK100B1
SKIOOB"
-
-
5,113,000
-
-
-
-
-
-
-
-
-
211,000
345,000
-185,000
-
-
-
-
-
-
-
-
-
-
-
-12,212,000
-
-
-
-
5,092,000
18,830,000
24,796,000
21,924,000
23,094,000
43,454,000
43,583,000
31,268,000
33,314,000
113,119,000
159,285,000
North Fork Koktuli River
NK119A
NK119CP2b
NK119B
NK119CP1"
NK100C
NK100B
NK100A1
NK100A6
-
-
5,113,000
-
1,103,000
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
16,168,000
19,105,000
3,837,000
24,392,000
46,932,000
77,409,000
182,988,000
221,636,000
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119AC
UT100B'
-
-
-
-
-
-
-
-
-
-
-
185,000
-
-
-
-
-
-
-
-
12,212,000
-
7,838,000
22,008,000
90,788,000
106,088,000
139,104,000
23,211,000
191,411,000
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-5
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 1. Effluent and receiving water flows at each gage under the Pebble 0.25 scenario. All values are presented in m3/yr.
Stream and Gage
Flow Returned
Through WWTP"
Flow Returned as
TSF Leakage
Flow Returned as NAG
Waste Rock Leachate
Flow Returned as PAG
Waste Rock Leachate
Flow of Interbasin
Transfer
TOTAL FLOW
Notes:
Blank values (-) indicate that values are either not applicable or are equal to zero.
3 WWTP discharges 50% of flow to South Fork Koktuli River, 50% of flow to North Fork Koktuli River (no WWTP flows are directed to Upper Talarik Creek).
b Confluence point where virtual gage was created because physical gage does not exist.
c 1/3 of total return flow is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location. Interbasin transfer flows are represented by negative flow values for SK100CP2
(losses transfer to UTC) and equivalent positive flow values for UT119A (gains transferred from SFK).
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
WWTP = wastewater treatment plant; TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-6
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 2. Effluent and receiving water flows at each gage under the Pebble 2.0 scenario. All values are presented in m3/yr.
Stream and Gage
Flow Returned
Through WWTPa
Flow Returned as
TSF Leakage
Flow Returned as NAG
Waste Rock Leachate
Flow Returned as PAG
Waste Rock Leachate
Flow of Interbasin
Transfer
TOTAL FLOW
South Fork Koktuli River
SK100G
SK100F
SK100CP2b'c
SK124A
SK124CPb
SK100C
SKIOOCPI"
SK119A
SK119CPb
SK100B1
SK100B"
6,088,000
13,000
86,000
610,000
540,000
-383,000
215,000
1,000
-72,000
-11,165,000
3,009,000
16,499,000
23,242,000
22,846,000
24,017,000
43,332,000
43,460,000
30,706,000
32,752,000
111,089,000
156,842,000
North Fork Koktuli River
NK119A
NK119CP2b
NK119B
NK119CPlb
NK100C
NK100B
NK100A1
NK100A6
-
-
-
-
6,088,000
-
-
-
2,177,000
-
-
-
-
-
75,000
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
7,583,000
10,520,000
3,396,000
16,753,000
47,804,000
71,336,000
176,090,000
215,115,000
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119AC
UT100B'
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
632,000
-
-
-
383,000
-
-
-
-
-
-
72,000
-
-
-
-
-
-
11,200,000
-
6,190,000
13,186,000
83,032,000
98,600,000
130,643,000
22,434,000
180,770,000
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-7
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 2. Effluent and receiving water flows at each gage under the Pebble 2.0 scenario. All values are presented in m3/yr.
Stream and Gage
Flow Returned
Through WWTPa
Flow Returned as
TSF Leakage
Flow Returned as NAG
Waste Rock Leachate
Flow Returned as PAG
Waste Rock Leachate
Flow of Interbasin
Transfer
TOTAL FLOW
Notes:
Blank values (-) indicate that values are either not applicable or are equal to zero.
3 WWTP discharges 50% of flow to South Fork Koktuli River, 50% of flow to North Fork Koktuli River (no WWTP flows are directed to Upper Talarik Creek).
b Confluence point where virtual gage was created because physical gage does not exist.
c 1/3 of total return flow is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location. Interbasin transfer flows are represented by negative flow values for SK100CP2
(losses transfer to UTC) and equivalent positive flow values for UT119A (gains transferred from SFK).
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
WWTP = wastewater treatment plant; TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-8
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 3. Effluent and receiving water flows at each gage under the Pebble 6.5 scenario. All values are presented in m3/yr.
Stream and Gage
Flow Returned
Through WWTPa
South Fork Koktuli River
SK100F
SK100CP2b'c
SK124A
SK124CPb
SK100C
SKIOOCPI"
SK119CPb
SK100B1
SK100BC
-
24,705,000
-
-
-
-
-
-
North Fork Koktuli River
NK119A
NK119CP2b
NK119B
NK119CPlb
NK100C
NK100B
NK100A1
NK100A6
-
-
-
-
24,705,000
-
-
-
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119AC
UT100B'
-
-
-
-
-
-
-
Flow Returned as TSF
Leakage
2,000
-1,000
1,530,000
-
-
-
3,009,000
249,000
-
2,314,000
-
5,000
-
-
-
79,200
-
-
-
-
-
-
624
-
Flow Returned as NAG
Waste Rock Leachate
1,122,000
-374,000
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
321,000
774,000
151,000
-
-
374,000
-
Flow Returned as PAG
Waste Rock Leachate
1,032,000
-344,000
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
344,000
-
Flow of Interbasin
Transfer
-8,666,000
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
8,666,000
-
TOTAL
9,790,000
18,769,000
35,651,000
36,822,000
56,020,000
56,149,000
14,900,000
103,214,000
145,379,000
7,719,000
10,657,000
2,688,000
15,336,000
66,207,000
88,443,000
193,104,000
232,201,000
2,190,000
2,887,000
73,461,000
89,356,000
120,217,000
20,198,000
166,406,000
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-9
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 3. Effluent and receiving water flows at each gage under the Pebble 6.5 scenario. All values are presented in m3/yr.
Stream and Gage
Flow Returned
Through WWTPa
Flow Returned as TSF
Leakage
Flow Returned as NAG
Waste Rock Leachate
Flow Returned as PAG
Waste Rock Leachate
Flow of Interbasin
Transfer
TOTAL
Notes:
Blank values (-) indicate that values are either not applicable or are equal to zero. SK100G and SKllQAare eliminated by the mine footprint in this scenario.
3 WWTP discharges 50% of flow to South Fork Koktuli River, 50% of flow to North Fork Koktuli River (no WWTP flows are directed to Upper Talarik Creek).
b Confluence point where virtual gage was created because physical gage does not exist.
c 1/3 of total return flow is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location. Interbasin transfer flows are represented by negative flow values for SK100CP2
(losses transfer to UTC) and equivalent positive flow values for UT119A (gains transferred from SFK).
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
WWTP = wastewater treatment plant; TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-10
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.1.1.1 Tailings Leachate
Estimation of potential flow through the substrate located under and around proposed TSFs requires
estimation of hydraulic conductivity. The hydraulic conductivity of the substrate material located near
possible dam sites varies greatly with depth and location. Ghaffari et al. (2011) report a range from 10-6
to 10'5 m/s in the upper bedrock, with a general decrease with depth and a range on the order of 10-7 to
10~9 m/s in the lower portions of bedrock with some zones of higher hydraulic conductivity
(Figure 6-11). In addition, the presence of fractured bedrock allows for localized discontinuities in the
rate of groundwater movement that can greatly influence overall groundwater conveyance (Ghaffari et
al. 2011).
We estimated the leachate flow from the TSFs using a hydraulic conductivity of 10-6 m/s in the upper
100 m of overburden and bedrock with no flow below that depth. We allowed vertical downward flow in
the tailings and radial flow outward in all directions from the TSF, with the excess head dissipating over
a horizontal distance of 1,200 m, comparable to the distance of the mine pit drawdown beyond the pit
rim. The interior surface area of TSF 1 would be 5.7 km2 for the Pebble 0.25 scenario and 14.2 km2 for
the Pebble 2.0 scenario. The Pebble 6.5 scenario would include two additional impoundments, TSF 2 (20
km2 area) and TSF 3 (7.7 km2 area).
Total leakage amounts for the three mine scenarios are 1.1 x 106 m3/yr (Pebble 0.25), 2.4 x 106 m3/yr
(Pebble 2.0), and 7.2 x 106 m3/yr (Pebble 6.5) (Tables 8-1 through 8-3). These estimates are based on a
simple assessment of seepage from the TSFs. Actual hydraulic conductivity would likely span several
orders of magnitude, from rapid flow in large fractures to essentially no flow in tight formations. Even a
small number of flowpaths with higher than expected hydraulic conductivity could significantly affect
the direction and quantity of flow.
Two potential estimates of tailings leachate composition are presented in Tables 8-4 and 8-5. Tailings
leachate from the humidity cell tests is judged to better represent effluent from a tailings impoundment
than the supernatant. It is used to represent leachate from the bottom of the TSFs and excess water from
the TSFs routed to the WWTP.
The tailings slurry would also contain ore-processing chemicals. We use an estimated concentration of
sodium ethyl xanthate, the primary ore-processing contaminant of concern, of 1.5 mg/L in the tailings
slurry (NICNAS 1995). Process chemicals could enter the environment in leaking tailings slurry water or
WWTP effluent. The potential for process chemicals in product concentrate slurry is considered in
Chapter 11.
8.1.1.2 Waste Rock Leachate
Tertiary rock would be used for construction of tailings dams and berms, and potentially other
structures requiring fill, but most would be piled near the pit. It is classified as non-acid-generating
(NAG) and its leachate is neutral (Table 8-6). Pre-Tertiary rock is classified as potentially acid-
generating (PAG) and its leachate is acidic (Tables 8-7 and 8-8). PAG rock would be piled separately and
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TT7April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
blended with ore, as needed, to maintain consistent composition in the processing plant feed.
Incomplete collection of pre-Tertiary waste rockleachate could result in acid rock drainage.
The mine scenarios (and the plan put forth for Northern Dynasty Minerals in Ghaffari et al. 2011) do not
include liners for the waste rock piles. Instead, leachate within the pit's drawdown zone would be
captured and pumped to the WWTP. Outside the drawdown zone, half the leachate would be captured
by extraction wells or other means and the rest would flow to surface waters. This is considered
reasonable given the likelihood that water would flow between wells and below their zones of
interception in the relatively permeable overburden materials and upper bedrock. Wells would not
catch all flows from the mine site given its geological complexity and the permeability of surficial layers.
As a result, 84% of PAG leachate and 82% of total waste rockleachate would be captured by the pit and
the wells for the Pebble 2.0 mine.
8.1.1.3 Mine Pit Water
Water pumped from the mine pit would consist of captured waste rock leachate and leachate from the
walls of the pit as precipitation passes over it and groundwater flows through it. The estimated pit wall
leachate compares to mean Tertiary (NAG) waste rockleachate as follows: 101% copper, 46%
aluminum, 22% cadmium, 16% cobalt, 45% manganese, 72% nickel, 744% lead, 16% selenium, 50%
zinc, 1,126% total dissolved solids, and 78% pH. This means that the mine pit water is much cleaner
than PAG pre-Tertiary leachate (copper in estimated pit wall leachate is only 0.2% of PAG waste rock
leachate).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ , „ APril 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 4. Aquatic toxicological screening of tailings supernatant against acute (criterion
maximum concentration) and chronic (criterion continuous concentration) water quality criteria or
benchmark values. Values are presented in M&/L unless otherwise indicated. Average leachate values
are from Appendix H.
Analyte
pH (S.U.)
Alkalinity (mg/LCaCOa)
Hardness (mg/LCaCOa)
S04
Ag
Al
As
Ca
Cd
Co
Cr
Cua
Cub
Fe
Hg
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Tl
Zn
Sum of metals
Average Value
7.9
75
320
320,000
0.018
72
17
116,000
<0.1
<0.1
<1.0
7.8
7.8
17
<0.037
26,000
8,000
72
70
44,000
<0.8
0.2
6.0
7.6
0.0
4.3
-
CMC or
equivalent
-
-
24
750
340
-
6.3
89
1,500
40
7.2
350
1.4
-
760
32,000
1,300
220
14,000
-
316
-
CCC or
equivalent
-
-
-
87
150
-
0.55
2.5
190
24
4.4
0.77
-
690
72
140
8.8
1,600
5
316
-
Acute
Quotients
-
-
0.0007
0.096
0.051
-
<0.012
<0.0011
<0.0007
0.19
1.1
0.048
<0.026
-
0.095
0.0022
<0.0006
0.0010
0.0004
-
0.014
0.50a : 1.4b
Chronic
Quotients
-
-
-
0.82
0.11
-
<0.14
<0.040
<0.0051
0.32
1.8
<0.048
-
0.10
0.97
<0.0056
0.026
0.0038
1.5
0.014
3.9:5.8b
Blank values (-) indicate that criteria are not available.
3 Acute and chronic criteria from Alaska's hardness-based standard.
b Acute and chronic criteria from the national ambient water quality criteria based on the bioticligand model (BLM).
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-13
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 5. Aquatic toxicological screening of tailings humidity cell leachates against acute
(criterion maximum concentration) and chronic (criterion continuous concentration) water quality
criteria or benchmark values. Values are MS/I- unless otherwise indicated. Average concentrations
are from Appendix H.
Analyte
pH (S.U.)
Alkalinity (mg/LCaCOa)
Hardness (mg/LCaCOa)
Cl
F
S04
Ag
Al
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cua
Cub
Fe
Hg
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Sn
Tl
V
Zn
Sum of metals
Average Value
7.8
60
67
520
450
17,000
0.01
24
5.5
11
9.2
0.20
0.49
23,000
0.05
0.19
0.50
5.3
5.3
30
0.01
4,000
2,500
44
33
2,100
0.54
0.06
1.8
1.5
2.9
0.05
0.78
3.2
CMC or
equivalent
-
-
-
-
1.6
750
340
29,000
46,000
-
-
1.5
89
445
10
2.5
350
1.4
-
-
760
32,000
360
46
14,000
-
3,600
1,370
91
CCC or
equivalent
6.5-9
-
-
-
87
150
1,500
8,900
-
-
0.20
2.5
58
6.9
1.6
-
0.77
-
-
693
72
40
1.8
1,600
5.0
75
0.8
120
91
Acute
Quotients
-
-
-
-
0.0062
0.031
0.016
0.0004
0.0002
-
-
0.038
0.0021
0.0012
0.58
1.1
-
0.0071
-
-
0.058
0.0010
0.0016
0.0015
-
0.0008
0.0006
0.038
0.85a: 1.4b
Chronic
Quotients
-
-
-
-
0.27
0.036
0.0071
0.0010
-
-
0.28
0.076
0.0094
0.84
1.8
-
0.013
-
-
0.064
0.46
0.014
0.039
0.30
0.039
0.0065
0.038
2.5a;3.4»
Notes:
Blank values (-) indicate that criteria are not available.
a Acute and chronic criteria from Alaska's hardness-based standard.
b Acute and chronic criteria from the national ambientwater quality criteria based on the biotic ligand model (BLM).
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-14
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 6. Composition of test leachate from Tertiary waste rock in the Pebble deposit and
quotients relative to the acute (criterion maximum concentration) and chronic (criterion continuous
concentration) water quality criteria or benchmark values. Values are presented in micrograms per
liter (Mg/L) unless indicated otherwise. Average leachate values are from Appendix H.
Parameter
PH
Alkalinity (mg/LCaCOa)
Hardness (mg/LCaCOa)
Cl
F
S04
Ag
Al
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cua
Cub
Fe
Hg
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Sn
Tl
V
Zn
Sum of metals
Average Value
7.2
66
74
530
62
28,000
0.011
80
2.7
18
57
0.31
0.54
21,000
0.22
3.9
0.55
3.2
3.2
140
0.010
1,900
5,100
100
6.3
7,200
4.4
0.12
2.1
1.9
1.3
0.068
1.8
16
-
CMC or
equivalent
-
-
-
1.9
750
340
29,000
46,000
-
1.5
89
445
10
2.5
350
1.4
-
760
32,000
360
46
14,000
-
3,600
1,370
91
-
CCC or
equivalent
6.5-9
-
-
87
150
1,500
8,900
-
0.20
2.5
58
6.9
1.6
0.77
-
693
72
40
1.8
1,600
5.0
75
0.8
120
91
-
Acute
Quotients
-
-
-
0.0059
0.11
0.0081
0.0006
0.0012
-
0.15
0.044
0.0012
0.32
1.3
0.40
0.0073
-
0.13
0.0002
0.012
0.0025
0.0002
-
0.0003
0.0013
0.17
1.4a : 2.4b
Chronic
Quotients
-
-
-
0.92
0.018
0.012
0.0064
-
1.1
1.6
0.0094
0.46
2.0
0.013
-
0.15
0.087
0.11
0.06
0.0013
0.38
0.017
0.085
0.15
0.17
5.3a : 6.8b
Notes:
Blank values (-) indicate that criteria are not available.
a Acute and chronic criteria from Alaska's hardness-based standard.
b Acute and chronic criteria from the national ambient water quality criteria based on the biotic ligand model (BLM).
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-15
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 7. Composition of test leachate from Pebble East pre Tertiary waste rock and quotients
relative to acute (criterion maximum concentration) and chronic (criterion continuous
concentration) water quality criteria. Values are presented in micrograms per liter (Mg/L) unless
indicated otherwise. Average leachate values are from Appendix H.
Parameter
pH (S.U.)
Alkalinity (mg/LCaCOa)
Hardness (mg/LCaCOa)
Cl
F
S04
Ag
Al
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cua
Cub
Fe
Hg
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Sn
Tl
V
Zn
Sum of metals
Average Value
4.8
9.9
22
910
110
52,000
0.019
380
8.0
13
4.5
0.55
0.63
6,300
3.2
9.7
1.6
1,400
1,400
10,000
0.010
960
1,500
340
4.3
2,100
10
0.35
0.78
3.2
1.9
0.088
2.4
480
-
CMC or
equivalent
-
-
-
0.24
750
340
29,000
46,000
-
0.46
89
160
3.20
0.043
350
1.4
-
760
32,000
130
12
14,000
-
3,600
1,370
32
-
CCC or
equivalent
6.5-9
-
-
87
150
1,500
8,900
-
0.085
2.5
21
2.4
0.027
0.77
-
693
72
14
0.47
1,600
5.0
75
0.8
120
32
-
Acute Quotients
-
-
-
0.082
0.51
0.023
0.0004
0.0001
-
7.0
0.11
0.0096
440
33,000
0.0072
-
0.44
0.0001
0.081
0.029
0.0001
-
0.0005
0.0018
15
460a : 33,000b
Chronic
Quotients
-
-
-
4.4
0.053
0.0084
0.0005
-
38
3.9
0.073
580
52,000
0.013
-
0.49
0.059
0.73
0.75
0.0005
0.65
0.024
0.110
0.020
15
640a : 52,000b
Notes:
Blank values (-) indicate that criteria are not available.
a Acute and chronic criteria from Alaska's hardness-based standard.
b Acute and chronic criteria from the national ambient water quality criteria based on the biotic ligand model (BLM).
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-16
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 8. Composition of test leachate from Pebble West pre Tertiary waste rock and quotients
relative to acute (criterion maximum concentration) and chronic (criterion continuous
concentration) water quality criteria. Values are presented in micrograms per liter (Mg/L) unless
indicated otherwise. Average leachate values are from Appendix H.
Parameter
pH (S.U.)
Alkalinity (mg/LCaCOa)
Hardness (mg/LCaCOa)
Cl
F
S04
Ag
Al
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cua
Cub
Fe
Hg
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Sn
Tl
V
Zn
Sum of metals
Average Value
6.6
18
59
520
120
61,000
0.027
320
1.5
16
14
0.33
0.69
13,000
0.40
7.0
0.69
1,600
1,600
1,700
0.011
1,400
6,700
730
1.8
2,100
6.8
0.17
3.1
3.8
0.14
0.41
0.68
56
CMC or
equivalent
-
-
-
-
1.3
750
340
29,000
46,000
-
-
1.2
89
370
8.2
0.88
350
1.4
-
-
760
32,000
300
36
14,000
-
3,600
1,370
75
CCC or
equivalent
6.5-9
-
-
-
87
150
1,500
8,900
-
-
0.17
2.5
48
5.7
0.55
-
0.77
-
-
690
72
33
1.4
1,600
5.0
75
0.8
120
75
Acute Quotients
-
-
-
-
0.021
0.42
0.0044
0.0005
0.0003
-
-
0.33
0.079
0.0019
190
1,800
4.8
0.0076
-
-
0.96
0.0001
0.023
0.0047
-
0.00004
0.0005
0.74
200a : l,800b
Chronic
Quotients
-
-
-
-
3.7
0.0100
0.011
0.0015
-
-
2.3
2.8
0.014
280
2,900
-
0.014
-
-
1.1
0.025
0.20
0.12
0.76
0.0019
0.52
0.0057
0.74
290" : 2,900»
Notes:
Blank values (-) indicate that criteria are not available.
a Acute and chronic criteria from Alaska's hardness-based standard.
b Acute and chronic criteria from the national ambientwater quality criteria based on the biotic ligand model (BLM).
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-17
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
8.1.1.4 Wastewater Discharge
Under all three mine scenarios, the WWTP would be designed and sized to treat the expected volume
and composition of inflow water based on estimated groundwater flow from the mine pit and runoff
from other site areas (waste rock piles, TSFs, and support facilities). The WWTP would be fed by
pipelines that pump water to the plant from the mine pit, crusher area, waste rock and TSF leachate
collection systems, and other operating areas of the site. However, the mine pit water represents the
largest component of flow into the WWTP. The flow volume and composition contributed by each mine
component has been estimated for each scenario. If the volume or composition of untreated water
exceeded plant specifications, it could be stored temporarily in a TSF or process pond and fed into the
plant as needed to balance flows and meet permit effluent quality requirements.
We specify that the WWTP would operate under a permit that would require meeting all national
criteria and Alaskan standards. We also assume that the Alaskan Pollutant Discharge Elimination System
wastewater discharge permit for a mine would include requirements that all other potentially toxic
contaminants be kept below concentrations equivalent to national chronic criteria. The equivalent
benchmark values used in this assessment for metals with no criteria or standards appear in Table 6-9.
Assumed discharge concentrations are the minimum of the input water concentration and the chronic
criterion, standard or benchmark value. Influent and effluent concentrations of contaminants of concern
are presented in Table 8-9. WWTP discharge rates for the Pebble 0.25, 2.0, and 6.5 scenarios are
estimated to be 10,12, and 49 million m3/year, respectively, equally distributed to the South and North
Fork Koktuli Rivers.
Table 8 9. Estimated concentration of contaminants of concern in effluents from the wastewater
treatment plant, tailings, non acid generating waste rock, and potentially acid generating waste
rock. Values are MS/I- unless otherwise indicated.
Contaminant
IDS (mg/L)
Zn
Se
Pb
Ni
Mg
Co
Cd
Al
Cu
WWTP Influent and
Failure Effluent3
323/338/590
17/26/34
1.7/1.4/1.4
0.23/0.28/0.37
2.4/3.2/2.3
65/92/100
0.97/2.2/2.0
0.13/0.22/0.26
43/66/73
72/100/150
WWTP Effluent
280
17/23/23
1.7/1.4/1.4
0.23/0.28/0.29
2.4/3.2/3.3
65/92/100
0.97/2.2/2.0
0.064
43/66/73
1.1"
Tailings
Leachate
600
3.2
1.5
0.064
0.54
44
0.19
0.052
24
5.3
NAG Waste Rock
Leachate
140
16
1.9
0.12
4.4
100
3.9
0.22
80
3.2
PAG Waste Rock
Leachate
85
27
3.5
0.26
8.6
530
8.4
1.9
350
1,500
Notes:
a Concentrations for the Pebble 0.25/Pebble 2.0/Pebble 6.5 scenarios.
b Chronic water quality criterion based on the biotic ligand model (BLM) using mean North Fork Koktuli River water.
IDS = total dissolved solids; TSF = tailings storage facility; PAG = potentially acid-generating; NAG = non-acid-generating.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-18
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.1.1.5 Sources of Total Dissolved Solids
Neither total dissolved solids (TDS) nor specific conductance data are available for waste rock or tailings
leachates from the Pebble deposit. However, TDS can be estimated by summing the concentrations of
leachate analytes after converting alkalinity to bicarbonate. Estimated TDS concentrations for the
tailings leachates, waste rock leachates, and WWTP effluents are summarized in Table 8-9.
8.1.2 Wastewater Treatment Plant Failure
There are innumerable ways in which wastewater treatment could fail under the mine scenarios in
terms of failure type (e.g., breakdown of treatment equipment, ineffective leachate collection,
wastewater pipeline failure), location, duration, and magnitude (e.g., partial vs. no treatment). Box 8-1
presents an example wastewater collection failure, and mechanisms of treatment failure are discussed
in Box 8-2. To bound the range of reasonable possibilities, we assess a serious failure in which the
WWTP allows untreated water to discharge directly to streams. This type of failure could result from a
lack of storage or treatment capacity or treatment efficacy problems. Chronic releases would occur
during operation if a lengthy process were required to repair a failure. We evaluate potential effects of
this type of failure using the following assumptions.
• The effluent is untreated water that is released to discharge points on tributaries to the South and
North Fork Koktuli Rivers.
• Untreated water composition is a flow-weighted average of concentrations from multiple
wastewater sources, including mine pit dewatering, waste rock leachates, runoff from crusher and
ancillary areas, and TSF leachates.
• Discharge rates are based on the sum of component flow volumes from the wastewater sources,
developed as part of the water balance (Section 6.2.2).
• Discharge rates and concentrations were calculated for each of the three mine scenarios
(Pebble 0.25, 2.0 and 6.5) and accountfor shifts in the relative contribution and concentration of
different wastewater sources for different mine sizes.
• Duration of a release could range from a few days to several months, depending on the nature of the
failure and difficulty of repair and replacement.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ , „ April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
BOX 8 1. AN ACCIDENTAL TAILINGS WATER RELEASE: NIXON FORK MINE, ALASKA, WINTER 2012
The Nixon Fork Mine is an underground gold mine that was intermittently mined between 1917 and 1950.
The modern mine opened in 1995, then closed in 1999 (ADNR 2012) and reopened under new ownership
again in 2007. The current operation is mining two ore bodies with a defined resource of 241,966 metric
tons of ore containing an estimated 4.6 million grams of gold (ADED 2012). An additional 856,156 grams of
gold are estimated to be recovered by reprocessing tailings on site. The mine is located on federal lands
managed by the Bureau of Land Management. The mine operates under authorizations from the Bureau of
Land Management, Alaska Department of Natural Resources (ADNR), and Alaska Department of
Environmental Conservation (ADEC).
In January and February 2012, the tailings impoundment at the Nixon Fork Mine overtopped. Below is the
chronology of events described by the mine operator that led to this event, based on a March 15, 2012
memo to Alaska State Mine Safety Engineer from Mystery Creek Resources, Inc.
• Prior to October 25, 2011, mine staff monitored the freeboard in the tailings impoundment per
requirements of agency authorizations.
• After October 25, 2011, staff decided to waive gage observation until spring melt because the gage was
frozen in ice.
• During a mid-January trip to the site, the Mystery Creek Resources, Inc., president noticed insufficient
freeboard in the tailings pond. He notified the BLM, ADNR, and ADEC.
• Corrective action was taken and the pond level began to drop.
• In late February 2012, mill operations that had been completed in batches were switched to continuous
operation without recognizing the implications for water balance (i.e., more water would be flowing to
the tailings impoundment).
• On March 9, 2012, mine personnel noticed evidence of dam overtopping. The BLM, ADNR, and ADEC
were notified and action was taken to draw down the pond and stop the overtopping.
• On March 10, 2012, agency inspections began. It was found that water from the tailings impoundment
was not likely to have reached nearby streams. An estimated 32,400 gallons of tailings water were
discharged from the impoundment.
On inspection of the dam it was found that the engineered spillway for the dam had been frozen over by a
previously undiscovered tailings water release. The ice prevented the spillway from operating as designed,
such that the later spill overtopped the dam at another location not designed for overflow. This case
illustrates the diversity of potential failures that can happen and suggests the practical impossibility of
predicting all possible failure modes.
BOX 8 2. POTENTIAL FAILURES OF REVERSE OSMOSIS WASTEWATER TREATMENT PLANTS
Because the high-quality receiving waters in the mine scenario watersheds would require extremely low
copper criteria and standards, reverse osmosis has been discussed as a potential treatment technology for
wastewater at the Pebble site. Studies of wastewater treatment plant (WWTP) efficiency and design
considerations show that reverse osmosis water treatment systems can be compromised by fouling and
scaling from calcium, iron, barium, strontium, silica, microbial growth, and silt (Mortazavi 2008). The
Bingham Canyon WWTP in Utah treats groundwater contaminated with sulphate and total dissolved solids
from copper mining by reverse osmosis. Pilot tests and optimization studies have shown that the structural
integrity of its reverse osmosis membranes can be damaged by abrasive materials (e.g., silt) or chlorine
(ITRC 2010). Changes in water composition could increase the concentration of chlorine if the mine pit
encounters a large flow of brine transmitted to the pit through deep fracture systems, or from localized
areas of mineralized rock with anomalous water quality. An example of WWTP failure due to highly variable
chemical composition of inflow wastewater has been documented at a copper mine in Chile: when silica
concentrations exceeded the design range, the whole reverse osmosis system could not be operated and
was therefore shut down until feed water quality improved (Shaoetal. 2009).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-20
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.1.3 Post-Closure Wastewater Sources
The post-closure period (Section 6.3.4) includes two distinct phases with respect to water management.
The first phase would be from the time the mine ceased operations until the mine pit was effectively full
of water. The second phase would be after the mine pit filled until treatment was no longer necessary.
During both phases, the quality of the water captured at the mine site would be substantially better than
the water captured during operations. During operations, leachate from the PAG waste piles would
account for between 80% and 94% of the total copper load in the captured water, depending on the
scenario. Since the mine scenarios specify that all of the PAG waste rock would be processed by the close
of operations and the PAG areas rehabilitated during site closure, the remaining flows would carry a
much lower concentration of copper. The expected reduction in copper concentration in the loading to
the WWTP would be greater than 90%, with substantial reductions also expected for other metals.
During pit filling, the mine operator would potentially need to treat water captured from the surface of
the TSFs, captured leachate from the TSFs, captured leachate from the NAG waste rock piles outside the
drawdown zone, and runoff from remaining facility areas that support the ongoing water treatment. The
drawdown zone would not begin to shrink until pit water level was within about 100 m of its final level,
based on our drawdown model. As the remainder of the pit filled, the drawdown zone would shrink until
the pit reached its final level. If water in the pit required treatment, the final pit level would be
maintained below the level that allowed natural outflow by pumping water to the WWTP. We assume
that this drawdown would result in drainage toward the pit for about 100m beyond the pit perimeter. If
or when the pit water and other sources met the discharge criteria, all flows could be discharged
without treatment and the pit water level would be allowed to rise until natural discharge was
established at the low point of its perimeter.
Because post-closure water quality is expected to be better than water quality during operation, the
assessment does not model or evaluate water quality during this period. In addition, post-closure
conditions are much more uncertain than conditions during operations, so it is more difficult to defend a
particular set of conditions and assumptions. It is important to note, however, that although post-
closure water treatment failures would be less consequential, they also would be less likely to be
promptly detected and corrected. In addition, because site hydrology and chemistry would change over
time, particularly as the pit filled, treatment requirements would change and the responses might be
slow.
The pit lake is a novel feature of the post-closure period, and, because it has been a subject of
stakeholder and reviewer concern, it requires more specific consideration. After closure under the
Pebble 0.25, 2.0, and 6.5 scenarios, the mine pit would fill with water for approximately 20, 80, and
300 years, respectively. Eventually, the pit water would be a source of leached minerals to streams, if it
were not collected and treated. Precipitation on the pit walls, groundwater entering the pit, and water
collected in the pit would dissolve metals and anions from the rock walls and any waste rock returned to
the pit, resulting in leachate.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TT7April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
Leachate composition would be approximated by some mixture of the waste rock test leachates
(Section 8.1.1.2), with some dilution by ambient water. These tests were run in oxidizing conditions, so
they maximize leaching rates. Oxygen would be provided in the pit by atmospheric diffusion from the
surface, precipitation, shallow groundwater, and vertical mixing of water in the pit during turnover, but
oxygen levels are expected to be lower in the pit than in the leachate tests. Flow of waste rock leachate
to the pit after closure would contribute to the mixture in the pit. However, as the drawdown zone
shrinks, most waste rock would be outside of the drawdown zone and much would be downgradient of
the pit, so its leachate would flow away from the pit. Pit water composition cannot be predicted with any
confidence, but some degree of leaching is inevitable. The experience with closed pit mines is quite
variable, but some mines, such as the Berkeley Pit in Montana, are acidic and have high metal
concentrations. Water flowing out of the full pit would be expected to flow to Upper Talarik Creek where
it would mix with waste rock leachate and water diverted from upstream.
In sum, failure to collect and treat waters from the waste rock piles, TSFs, or mine pit could expose biota
in the streams draining the post-closure mine site to contaminated water. There is little information on
failure rates for post-closure wastewater management at mines. If the closure occurs as described in this
assessment, toxic effects could occur but they are unlikely to be severe. However, premature closures of
mines do occur and such closures are likely to leave acid-generating materials on the surface. Further, it
is much too soon to know whether mines that are permitted for perpetual water collection and
treatment (e.g., the Red Dog Mine in Alaska) can actually carry out those functions in perpetuity.
8.1.4 Probability of Contaminant Releases
Failures of water collection and treatment are a common feature of mines. A review of the 14 porphyry
copper mines that have operated for at least 5 years in the United States found that all but one (92%)
had experienced reportable aqueous releases, with the number of events ranging from three to 54
(Earthworks 2012). Mine water releases range from chronic releases of uncaptured leachate to acute
events caused by equipment malfunctions, heavy rains, or power failures. The USEPA has observed that
some operators continue to operate when they know that treatment is ineffective and not meeting
standards. Hence, the record of analogous mines indicates that releases of water contaminated beyond
permit limits would be likely over the life of any mine at the Pebble deposit.
The probability of the specific WWTP failure analyzed here cannot be estimated. It is improbable in that
it requires that wastewater not be treated and not be diverted to storage. However, it is plausible that
such an event would result from equipment failures, inadequate storage or human errors. It is more
likely that a partial failure (e.g., incomplete treatment) would occur, but any one of the innumerable
incomplete treatment scenarios is also unlikely. Hence, the WWTP failure scenario analyzed here is a
reasonable bounding case.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o »o April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.2 Chemical Contaminants
8.2.1 Exposure
8.2.1.1 Effluent Dilution and Transport
Under the mine scenarios, treated wastewater discharges would be divided between the South and
North Fork Koktuli Rivers. South Fork Koktuli River flows include interbasin transfer to Upper Talarik
Creek. Tailings water leakage would discharge to the North Fork Koktuli River from TSF1 and to the
South Fork Koktuli River from TSF2 and TSF3. Waste rock leachate that is not captured and treated
would flow to Upper Talarik Creek and South Fork Koktuli River. NAG waste rock leachate would be the
only direct source of wastewater to Upper Talarik Creek during routine operations.
WWTP effluents would be released at the surface, entering receiving waters as a plume and gradually
being diluted. Input of contaminated groundwater from waste rock or tailings leachates would be
introduced via upwelling through the cobble and gravel substrates (i.e., via hyporheic input). In either
case, a gradient would occur between full-strength effluents and fully mixed ambient waters.
Fully mixed ambient concentrations for each scenario are calculated by diluting the estimated discharge
(i.e., contributing loads) in the background receiving waters using ambient flows and concentrations
from the Environmental Baseline Document2004 through 2008 (EBD) (PLP 2011) after adjusting the
base flows for the reductions in watershed areas due to the mine footprint (Tables 8-1 through 8-3).
(Note that constituent flows at a gage are less then total flows because mine-related flows from
upstream are carried forward in the model.) Concentrations of contaminants of concern in wastewater
discharges, waste rock leachates, and tailings leachates are presented in Table 8-9. Discharge flow rates
are based on the water balance described in Chapter 6, including reduced stream flows due to water use
under the mine scenarios and interbasin transfers. Contaminant flows were blended with adjusted
ambient water flows and tracked downstream from one stream gage to the next.
Working from the upstream-most point in each mine scenario watershed, ambient contaminant mass
flows were added to discharge contaminant mass flows and divided by total flow at each stream gage to
determine the diluted concentration. Moving to the next downstream stream gage the process was
repeated, each time adding the mining process flows at their expected concentrations, assuming that
background concentration at each stream gage would be capturing all concentration inputs other than
mining inputs. This implies that mining processes cause no other degradation or metal contributions
through other mechanisms such as surface erosion or mobilization of metals from in situ minerals by
acidic leachates.
Because the streams draining a mine site are the receptors for wastewaters, their water quality
constitutes the dilution water quality. The water quality of streams in the Pebble deposit area has been
extensively characterized (PLP 2011, Zamzow 2011). The mine scenario watersheds are neutral to
slightly acidic with low conductivity, hardness, dissolved solids, suspended solids, and dissolved organic
carbon (Table 8-10). In this respect, they are characteristic of undisturbed streams. However, as would
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTZ April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
be expected for a metalliferous site, levels of sulfate and some metals (copper, molybdenum, nickel, and
zinc) are elevated, particularly in the South Fork Koktuli River. PLP (2011) found that copper levels in
some samples from the South Fork Koktuli River exceeded Alaska's chronic water quality standard.
However, most of the exceedances were "in sampling locations within or in proximity to, the general
deposit location" and the number and magnitude of exceedances decreased with distance downstream
(PLP 2011: Figure 9.1-35). Therefore, the stream reaches with significantly elevated copper
concentrations would be largely destroyed by the mine scenario footprints and by water diversions.
The chemical fate of metals in receiving streams may be complex. Aluminum, iron, and manganese are
commonly precipitated in streams receiving acid mine drainage, diminishing or destroying stream
habitats with deposited floes but also reducing the aqueous toxicity of those metals. Acidic leachates
would form from the PAG waste rocks, but concentrations of precipitating metals in waste rock
leachates are not particularly high (Tables 8-7 and 8-8). Other metals do not precipitate to a significant
degree but may have reduced bioavailability due to receiving water chemistry. That issue is largely dealt
with by the use of the biotic ligand model (BLM) for copper, which includes a metal speciation submodel.
It is treated less accurately, by the use of hardness-corrected criteria to screen other metals.
8.2.1.2 Biological Exposures
Aquatic biota would be directly exposed to contaminants in discharged waters. Fish embryos and larvae
(e.g., salmon eggs and alevins) would be exposed to benthic pore water, which would be provided by
groundwater in areas of upwelling and otherwise by surface water. In this chapter we assume that
sediments would not be contaminated by tailings, waste rock, or other mine-derived particles. Juvenile
fish (e.g., salmon fry and parr) would be exposed to surface water. Adult resident salmonids would also
be exposed to surface water, but, unlike the early life stages, they would occur in the smallest streams
only during spawning. Adult anadromous salmonids would have brief exposures to waters near the site.
Aquatic insects would be exposed in all juvenile stages, which constitute most of their life cycles. They
would be exposed to benthic pore water or surface water depending on their habits.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -4 April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 10. Mean and coefficient of variation of background surface water characteristics of the
mine scenario watersheds, 2004 2008.
Analyte
IDS (mg/L)
pH (field)
DO (mg/L)
Temperature (°C)
Specific Conductivity (pS/cm)
TSS (mg/L)
Ca (mg/L)
Mg(mg/L)
Na (mg/L)
K (mg/L)
Alkalinity (mg/L)
S04 (mg/L)
Cl (mg/L)
F (mg/L)
Hardness (mg/L)
Al (Mg/L)
As (Mg/L)
Ba (Mg/L)
Cd (Mg/L)
Cu (Mg/L)
Fe (Mg/L)
Mn(Mg/L)
Mo (Mg/L)
Ni(Mg/L)
Pb (Mg/L)
Zn (Mg/L)
CN (Mg/L)
DOC (mg/L)
South Fork
Koktuli River
44 (0.41)
7.0 (0.045)
10.2 (0.21)
4.77 (1.02)
ND
2.21(2.13)
6.34 (0.42)
1.41 (0.56)
2.35 (0.35)
0.38 (0.47)
17.4 (0.43)
8.78 (0.87)
0.69 (0.26)
ND
19.6 (0.47)
11 (0.68)
ND
4.1(0.48)
ND
1.3 (0.88)
120 (1.01)
20 (1.14)
0.66 (0.98)
0.41(0.61)
ND
2.7(1.02)
ND
1.36 (0.62)
North Fork
Koktuli River
37 (0.035)
6.74(0.10)
10.2 (0.2)
4.39(1.12)
ND
1.39 (1.9)
5.09 (0.35)
1.32 (0.44)
2.38 (0.23)
0.41 (0.39)
20.5 (0.38)
2.26 (0.56)
0.66 (0.25)
ND
14.4 (0.36)
13 (0.82)
ND
3.1(0.35)
ND
0.39 (0.84)
110 (0.68)
10(1.67)
0.19(1.2)
0.30(1.14)
ND
1.8 (0.64)
ND
1.5 (0.51)
Upper
Talarik Creek
51.2(0.37)
6.99 (0.091)
10.5(0.19)
4.04 (0.98)
73.4 (0.34)
2.52 (1.58)
8.77 (0.3)
2.12(0.46)
2.82 (0.32)
0.44 (0.43)
31.8 (0.34)
5.48 (1.43)
0.7 (0.29)
ND
26.5 (0.42)
13 (1.1)
ND
5.5(0.41)
ND
0.42 (0.89)
110 (0.83)
21 (1.09)
0.2(0.51)
0.63 (1.04)
ND
2.0(1.09)
ND
1.57 (0.82)
Notes:
Filtered concentrations are used for hardness and trace elements.
ND = analytes detected in less than half of samples.
Source: PLP 2011.
8.2.2 Exposure-Response
We screened potential contaminants against ecotoxicological benchmarks to identify the most
potentially toxic constituents and indicated the degree of treatment that would be required and what
sorts of effects might occur due to mining emissions. Criteria and equivalent screening benchmarks are
presented in Tables 8-4 through 8-8, and the sources of non-criteria screening benchmarks are
presented in Table 6-10. Benchmarks were derived from the literature to be as similar to criteria as
possible, given the available data (Section 6.4.2.3). Criteria for many of the metals are functions of
hardness, and copper criteria are a function of multiple water properties. For those metals, criteria are
calculated for each leachate based on its chemistry and for each receiving stream based on its
background chemistry.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-25
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.2.2.1 Copper
Although the ore and waste rock from porphyry copper mines contain a mixture of metals, copper is the
major resource metal and is particularly toxic to aquatic organisms. Hence, it is the most likely to cause
toxic effects, and actions taken to prevent copper's effects are likely to mitigate, to some extent, effects
from co-occurring metals. For these reasons, we focus on copper criteria, standards, and toxicity in this
assessment.
Copper Standards and Criteria
The State of Alaska's copper standard is a function of hardness and is based on a prior national criterion
(USEPA 1985a). The formulas for the state's acute value (criterion maximum concentration or CMC) and
chronic value (criterion continuous concentration or CCC), in micrograms per liter and based on
hardness in milligrams per liter, are:
Cu acute criterion = e°-9422xlnhardness-1-700 x 0.96
Cu chronic criterion = e°-8545xlnhardness-1-702 x 0.96
Note that the formulae are similar and yield similar values—that is, when copper causes toxic effects,
the effects occur relatively quickly. At 20 mg/L hardness (i.e., soft water typical of the Bristol Bay
region), Alaska's acute and chronic values for copper are 2.95 and 2.26 ug/L, respectively.
The federal government has developed new National Ambient Water Quality Criteria for Protection of
Aquatic Life (criteria) for copper (USEPA 2007). They are calculated using the BLM, which derives the
effects of copper as a function of the amount of metal bound to biotic ligands on gills or other receptor
sites on an aquatic organism. The ligands bind free copper ions and, to a lesser degree, copper hydroxide
ions (Figure 8-2). Copper competes for ligands with calcium and other cations. The competitive binding
model for the biotic ligand requires a metal speciation model and estimates of basic water chemistry
parameters. The BLM is an advance over hardness normalization, because it more fully accounts for the
mechanisms controlling variance in toxicity. In practice, its most important consequence is to estimate
the often large reduction in toxicity resulting from binding of copper by dissolved organic matter. The
BLM is freely available from USEPA (http://water.epa.gov) and from the model's developer Hydroqual
Inc. (http://www.hydroqual.com/blm).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment o 2g April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Figure 8 2. Processes involved in copper uptake as defined in the biotic ligand model (USEPA
2007).
Inorganic
Complexes
Gill Surface
(biotic ligand)
Active Metal
Sites
e.g. : Cu - Hydroxides
Cu - Carbonates
The results of applying the BLM to mean water chemistries of the South and North Fork Koktuli Rivers
and Upper Talarik Creek are presented in Table 8-11. These values are lower than the state's hardness-
based values and the variance among streams is potentially significant.
Table 8 11. Results of applying the biotic ligand model to mean water chemistries in the mine
scenario watersheds to derive copper criteria specific to receiving waters. Values presented in
Mg/L.
Stream
South Fork Koktuli River
North Fork Koktuli River
Upper Talarik Creek
CMC
2.4
1.7
2.7
CCC
1.5
1.1
1.7
Notes:
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
Biotic ligand model (BLM) source: USEPA 2007.
The results of applying the BLM to mean chemistries of the waste rock leachates are presented in
Table 8-12. The model runs used mean water chemistries from the PLP tests (Appendix H). These
effluent-specific values differ from each other and from the values for ambient waters due to differences
in water chemistries.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-27
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 12. Results of applying the biotic ligand model to mean water chemistries in waste rock
leachates to derive effluent specific copper criteria. Values presented in Mg/L-
Leachates
Pebble Tertiary
Pebble West pre-Tertiary
Pebble East pre-Tertiary
CMC
2.5
0.88
0.043
CCC
1.6
0.55
0.027
Notes:
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
Biotic ligand model (BLM) source: USEPA 2007.
For both the background waters and the leachates, temperature was set to the mean from three streams
on the site (4.5°C—PLP 2011). For the leachates, dissolved organic carbon was set to 1 mg/L (the lowest
level accepted by the model) andhumic acid was set to the default value (10% of dissolved organic
carbon).
Both the state standards and the national criteria for copper are derived from the 5th centile of the
aquatic genera sensitivity distribution. The most sensitive 33% of genera in acute tests and 42% of
genera in chronic tests are all invertebrates (USEPA 2007). Hence, the regulatory benchmarks are
determined by invertebrate sensitivities. However, the most sensitive vertebrates in both types of tests
are fish in the genus Oncorhynchus, which includes rainbow trout and the five Pacific salmon species.
Rainbow trout is a standard test species that is at least as sensitive to copper as Chinook and coho
salmon, brook trout, and brown trout in acute tests (CH2M Hill and The Clark Fork and Blackfoot, LLC
2004). Acute and chronic values for rainbow trout can be derived for background water quality using
the BLM (Table 8-13). BLM-estimated acute values could also be calculated for three cladoceran species
in the three streams draining the mine scenario footprints: Daphnia magna (8.68-13.02 ug/L), Daphnia
pulex (4.28-6.63 ug/L), and Ceriodaphnia dubia (5.99 to 9.13 ug/L). These zooplankters are less directly
relevant to the receiving streams, but they are relevant to ponds and Iliamna Lake and they illustrate the
great sensitivity of aquatic arthropods to copper.
Table 8 13. Rainbow trout site specific acute and chronic copper toxicity derived by applying the
biotic ligand model to mean water chemistries in the mine scenario watersheds.
Stream
South Fork Koktuli River
North Fork Koktuli River
Upper Talarik Creek
Acute Toxicity3
(LC5o in Mg/L)
63
59
75
Chronic Toxicity
(CV in Mg/L)
22
21
26
Notes:
a Acute toxicity: median lethal concentration (LCfeo).
CV = chronic value, calculated using the species-specific acute to chronic ratio of 2.88.
Biotic ligand model (BLM) source: USEPA 2007.
Alternative Copper Endpoints
Copper standards and criteria are based on conventional test endpoints of survival, growth, and
reproduction. However, research has shown that the olfactory sensitivity of salmon is diminished at
copper concentrations lower than those that reduce conventional endpoints in salmon (Hecht et al.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-28
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
2007). Salmon use olfaction to find their spawning streams, detect and avoid predators, find food, detect
reproductive and alarm pheromones, and perform other life processes. Although effects on fish olfaction
have not been shown to affect the viability of field populations, it is reasonable to expect that
interference with these essential processes would have population-level consequences (DeForest et al.
2011b).
Meyer and Adams (2010) applied the hardness-corrected criteria and the BLM to data from multiple
laboratory tests for olfactory effects and found that the BLM accounted well for variance among tests,
and that BLM-based criteria were consistently protective of olfactory effects in the test systems. In
contrast, hardness-corrected criteria were not consistently protective. DeForest et al. (2 01 la) extended
those results by applying the same models to 133 ambient waters in the western United States,
including Alaska, which exhibited a wide range of water chemistries. Usingthe 20% inhibitory
concentration (IC2o) for coho salmon olfaction from Mclntyre et al. (2008a, 2008b) as the endpoint, they
found that the hardness-corrected criteria were not consistently protective, but the BLM-based chronic
criteria were protective of this chronic effect in 100% of the waters. Even the acute BLM-based criteria
were protective of this chronic effect in 98% of waters, since the criteria are determined by sensitive
invertebrates that experience diminished survival, growth, or reproduction at even lower levels than
those that inhibit fish olfactory receptors.
Although the criteria are protective screening values for sensory effects of copper on salmonids, it is
necessary to consider potential effects when criteria are exceeded. Meyer and Adams (2010) adapted
the BLM to sensory data and derived IC2o:BLM factors that can be used to convert site-specific criteria
into estimates of the copper concentration at which 20% of rainbow trout avoid the contaminated water
or at which they experience 20% inhibition of their olfactory senses (Table 8-14).
Table 8 14. Rainbow trout site specific benchmarks for sensory effects. Derived by applying
IC2o:BLM ratios in Myer and Adams (2010) to the acute values in Table 8 8.
Stream
Avoidance
(IC20 in Mg/L)
Sensory Inhibition
(IC2o in
South Fork Koktuli River
5.2
26
North Fork Koktuli River
3.8
19
Upper Talarik Creek
5.9
30
IC2o = 20% inhibitory concentration.
Dietary Copper Exposure-Response
Dietary exposure to metals, particularly at mine sites, has become a topic of investigation in recent years
(Meyer et al. 2005). Studies of the tailings-contaminated Clark Fork River in Montana and the Coeur
d'Alene River in Idaho have shown thatmacroinvertebrates can accumulate metals at levels that result
in toxicity to fish that consume them (Farag et al. 1994, Woodward et al. 1994, Woodward et al. 1995,
Farag et al. 1999). Participants in a recent Pellston Workshop (a workshop series convened by the
Society for Environmental Toxicology and Chemistry to examine environmental toxicological issues)
reviewed the literature and developed an estimate of the degree to which aqueous toxicity thresholds
should be adjusted to account for dietary exposures in rainbow trout (Borgmann et al. 2005). The
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-29
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
estimate is based on an average bioconcentration factor of 2,000 L/kg and an average dietary chronic
value for rainbow trout of 646 ug/g. The resulting factor is 0.95, so the adjustment is small. If the factor
is applied to the lowest chronic value for rainbow trout (11.3 ug/L) (USEPA 2007), the result
(10.7 ug/L) is still much higher than the national ambient water quality criteria and state standards due
to the relative insensitivity offish. This result applies to aqueous-only exposures (i.e., it does not include
contaminated sediment or allochthonous material). Because dietary exposure factor has little influence
on risks to fish from direct aqueous exposure and adds another source of uncertainty, it is not applied to
the risk estimates in this chapter. However, dietary exposure offish to copper in sediments, where
direct aqueous exposures of post-larval fish maybe minor, is considered in Section 9.4.2.1.
Copper Exposure-Response Data from Analogous Sites
Evidence concerning exposure-response relationships for copper and other metals in streams at metal
mines also comes from field studies. Because the mine scenarios presume that water quality criteria
would be met during routine operations, the critical question is whether effects are observed at those
levels. The most relevant high-quality studies are those performed in the Colorado metal belt,
particularly near the Animas and Arkansas Rivers. These sites are contaminated predominately by mine
drainage and mine waste leachates, and field and laboratory experiments have confirmed that aqueous
metals, not tailings or other particles, cause the observed effects (Courtney and Clements 2002). These
studies have identified effects on aquatic insect populations and invertebrate communities at
concentrations below water quality criteria for the dominant metals (cadmium, copper, and zinc)
(Buchwalter et al. 2008, Schmidt et al. 2010). Application of the BLM and an additive combined effects
model reduced the discrepancy but did not eliminate it, suggesting that chronic criteria for metals are
not protective against effects on invertebrates (Schmidt et al. 2010). In particular, while the combined
criteria approximated thresholds for taxa richness, abundances of sensitive taxa were reduced at
exposures below the combined criteria (Griffith et al. 2004, Schmidt et al. 2010). This result was
supported by a study funded by Rio Tinto, which concluded that "aquatic insects are indeed very
sensitive to some metals and in some cases may not be protected by existing WQC [water quality
criteria]" (Brix et al. 2011). Potential reasons for the discrepancy are the absence of sensitive species or
life stages from the criteria, less-than-life-cycle exposures, and the absence of dietary exposures.
Unexpected field effects might be caused by an unknown factor that is correlated with both the
concentration of metals and the biological effects (i.e., a confounding variable). However, no such factor
is known for this assessment, and the hypothesized mechanisms for the greater sensitivity of field
communities are supported by evidence from laboratory and field experiments (Brix et al. 2011). It also
must be noted that the occurrence of biological effects below criterion concentrations does not
necessarily indicate that criteria are not adequately protective. By design, the criteria allow acute or
chronic effects on as much as 5% of species (USEPA 1985b).
Copper Exposure-Response Uncertainties
The copper criteria are based on a large body of data and a mechanistic model of exposure and effects.
Hence, it is one of the best-supported criteria. However, it is always possible that it would not be
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTI April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
protective in particular cases due to unstudied conditions or responses. Because the most sensitive taxa
are aquatic invertebrates, unknown aspects of invertebrates are most likely to be influential. In
particular, field studies, including studies of streams draining metal mine sites, show that
Ephemeroptera (mayflies) are often the most sensitive species and smaller instars are particularly
sensitive (Kiffney and Clements 1996, Clements etal. 2000). However, the copper criteria do not include
any Ephemeroptera in the sensitivity distribution (USEPA 2007). If the ephemeropteran, plecopteran,
trichopteran, or other invertebrate species in the streams draining the mine scenario footprints are
more sensitive than cladocerans (the most sensitive tested species), then they may not be protected by
the criteria. On the other hand, copper concentrations are naturally elevated in the highest reaches of
the South Fork Koktuli River so biota in those reaches maybe somewhat resistant to copper additions.
In addition, the chronic copper criterion is derived by applying an acute:chronic ratio to the BLM-
derived final acute value (USEPA 2007). Because of the complex dynamics of chronic uptake,
distribution, and sequestration of metals in aquatic insects, the BLM, which focuses on binding to a
surface ligand, may not adequately adjust chronic toxicity (Luoma and Rainbow 2005, Buchwalter et al.
2008). Brix et al. (2011) reviewed the toxicity testing literature and found that aquatic insects are highly
sensitive to copper and some other metals in chronic relative to acute exposures and may not be
protected by current criteria. Thus, the protectiveness of the chronic criterion is more uncertain than
the acute criterion.
Based on the literature cited above and the author's experience, resolution of this uncertainty by
additional research and testing is likely to lower the chronic criterion by approximately a factor of 2 and
not as much as a factor of 10. Therefore, this uncertainty biases the estimated length of streams
experiencing toxic effects and could change the conclusions with respect to relatively low toxicity
materials such as tailings and NAG waste rock.
8.2.2.2 Other Metals
Chronic national ambient water quality criteria, state standards, or equivalent benchmarks were used to
screen the constituents of tailings, waste rock, and product concentrate leachates (Section 8.1.1.4).
Those that were retained in the screening were carried forward to release, transport, and dilution
modeling. For hardness-dependent criteria, screening values were calculated for each receiving stream
using mean hardness (Table 8-15). Finally, those potential contaminants that exceeded screening values
in streams after dilution, or that were otherwise of concern, are discussed in more detail below.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ OH April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 15. Hardness dependent acute water quality criteria (criterion maximum concentration) and
chronic water quality criteria (criterion continuous concentration) for the three potential receiving
streams under the mine scenarios. All values are presented in Mg/L-
Criteria
CdCMC
CdCCC
PbCMC
PbCCC
NiCMC
NiCCC
ZnCMC
ZnCCC
South Fork Koktuli River
0.45
0.085
12
0.46
130
14
32
32
North Fork Koktuli River
0.30
0.064
7.4
0.29
91
10
23
23
Upper Talarik Creek
0.50
0.089
15
0.58
150
17
38
38
Notes:
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
Aluminum
The environmental chemistry and resulting toxicity of aluminum are complex (Gensemer and Playle
1999). Aluminum is amphoteric, being more soluble at acidic and alkaline pHs and less soluble at
circum-neutral pH. It appears in a variety of forms, including soluble complexes with common anions
and humic and fulvic acids, but in most streams soluble and insoluble hydroxide compounds dominate.
Free ionic aluminum is expected to be a small component of dissolved aluminum at the circum-neutral
pHs found in the streams draining the mine scenario footprints, unless mine drainage acidifies them.
Aluminum is most toxic in mixing zones where acidic waters mix with neutral or basic ambient waters,
apparently due to precipitation at the surface of gills. In general, fish are more sensitive than
invertebrates to aluminum.
Cadmium
Cadmium is an uncommon but highly toxic divalent metal (Mebane 2010). Rainbow trout acute median
lethal concentration (LCso) values for cadmium from tests at 7.5 and 21 mg/L hardness are 0.477 and
0.84 ug/L, respectively (WE 2002, Mebane 2010). An early-life-stage test of rainbow trout at 21 mg/L
hardness gave a chronic value for survival and growth of 0.36 ug/L, but it had quality control issues
(WE 2002). A later test without those issues, but without reported hardness, gave a higher rainbow
trout chronic value of 1.56 ug/L (WE 2002). Acute tests with mayflies, caddisflies, and stoneflies all gave
values that were much higher than the trout values (WE 2002). The tests by Windward Environmental
(WE 2002) were conducted for the State of Idaho to support the derivation of site-specific criteria for
the Coeur d'Alene River. BLM-derived acute values for Ceriodaphnia dubia were 37 to 51 ug/L for the
three streams draining the mine scenario footprints. This is consistent with the relative insensitivity of
invertebrates to acute lethality. Although these tests and other tests in the literature show fish to be
more sensitive to cadmium than invertebrates in acute exposures, in chronic exposures invertebrates
were more sensitive (Mebane 2010). In particular, mortality of the amphipod Hyalella azteca increased
at 0.16 ug/L cadmium at relevant hardness (17 mg/L) (Mebane 2010).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-32
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
Cobalt
Current studies of the aquatic toxicity of cobalt can be found in a recent Canadian review (Environment
Canada and Health Canada 2011). Acutely lethal concentrations range from 89 to 585,800 ug/L.
Chronically toxic concentrations for invertebrates range from 2.9 to 155 ug/L (with the exception of a
1972 4-day test for rotifers, which resulted in 59,000 ug/L). Only three fish species have been tested for
chronic effects, yielding values of 340 to 2,171 ug/L; the least sensitive of these three species was
rainbow trout. In experimental studies, Chinook salmon avoided waters with cobalt concentrations of
24 ug/L, but rainbow trout were less sensitive, avoiding concentrations of 188 ug/L (Hansen et al.
1999). It is expected that the same water quality parameters that modify copper toxicity also affect
cobalt toxicity, but existing data are insufficient to perform adjustments.
Lead
Lead is a divalent metal with national criteria and state standards based on water hardness (USEPA
1986, Eisler 2000, WE 2002). A BLM is available that estimates acute LC5o values for fathead minnows in
the South and North Fork Koktuli Rivers as 382 and 383 ug/L, respectively. In comparison, a rainbow
trout test at hardness similar to the South and North Fork Koktuli Rivers (20 mg/L) resulted in an LCso
of 120 ug/L (WE 2002). Tests at similar hardness levels for mayflies, caddisflies, stoneflies, and
chironomid midges gave higher LCsos (429 to greater than 1,255 ug/L) (WE 2002). This indicates that an
endpointfish species is more sensitive to lead than aquatic insect larvae, which is consistent with BLM-
derived acute values of 523 to 748 ug/L for Daphnia magna. Chronic tests gave values for reduced
rainbow trout weight and length of 36.0 and 12.1 ug/L at 21 and 26 mg/L hardness and for the midge
Chironomus tentans of 65.4 ug/L at 32 mg/L hardness (WE 2002). Note that we use tests performed for
the State of Idaho (WE 2002) for cadmium, lead, and zinc, because they are high-quality tests that use
species and water chemistries relevant to the Bristol Bay environment.
Manganese
The toxicity of manganese is strongly related to hardness. Acutely lethal concentrations for manganese
in soft water range from 0.8 to 4.83 mg/L for invertebrates and 2.4 to 3,350 mg/L for fish. The most
sensitive acutely tested fish was coho salmon. In four soft-water tests of rainbow and brown trout,
chronic values ranged from 0.79 to 14.6 mg/L. More details can be found in recent reviews (Reimer
1999, IPCS 2004).
Selenium
Selenium is a bioaccumulative and moderately biomagnifying element. Dissolved oxyanions of selenate
(Se+4) and selenite (Se+6) are taken up by microbes, algae, and plants and converted to organic forms. In
streams, periphyton growing on rocks and woody debris are the primary community that performs this
conversion and the conversion rates are relatively low. Selenium causes deformities and death in
embryos and larvae offish, which are exposed to selenium accumulated by their mothers. Therefore,
potential selenium toxicity is of concern for resident but not anadromous fish. Effects of selenium on
salmonids have been studied below mines in British Columbia. For example, cutthroat trout embryos
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -o April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
from a pond at a coal mine in British Columbia with selenium concentrations of 93 ug/L showed effects
ranging from deformities of larvae to mortality (Rudolph et al. 2008). The probability of mortality was
correlated with selenium concentrations in the embryos. Invertebrates are less sensitive to selenium
than fish.
The complex dynamics of selenium and its various forms have led to complex water quality criteria
(USEPA 2004). The acute national criteria, based on the proportions of selenite and selenate,
are!85.9 ug/L selenite and 12.82 ug/L selenate. The chronic criterion is 5.0 ug/L total selenium.
However, because the transformations and bioaccumulative processes are so complex, a chronic
criterion for fish tissue concentrations (7.91 ug/g whole body dry weight) has been proposed based on
juvenile mortality of bluegill sunfish. The genus mean chronic value for rainbow trout is a little higher
(9.32 ug/g dry weight). These tissue-based values are believed to be more accurate than benchmarks
based on water concentrations. However, implementing the criterion or using the dietary toxicity test
and field data that were used in its derivation would require a model of selenium bioaccumulation that
is applicable to streams and lakes in the Bristol Bay watershed. No such model is currently available.
Zinc
Zinc, like copper, is a divalent metal and trace nutrient that is a common aquatic toxicant. The national
criteria and state standard are based on water hardness (USEPA 1987), but a BLM is available that
provides more accurate predictions of acute toxicity, at least for some test species (DeForest and Van
Genderen 2012). The BLM-based LCso estimates for rainbow trout in the South and North Fork Koktuli
Rivers are 64 and 63 ug/L, respectively. In comparison, a rainbow trout test at similar hardness
(16 mg/L) resulted in an LCso of 117 ug/L (WE 2002). Acute tests at 14 mg/L hardness for two mayfly
species and a caddisfly species resulted in values greater than 2,926 ug/L, and a stonefly species test
resulted in values greater than 1,526 ug/L (WE 2002). These results suggest that an endpointfish
species is considerably more sensitive to zinc than relevant stream invertebrates in acute exposures.
BLM-derived acute values for Daphnia magna were 407 to 502 ug/L for the three receiving streams,
which is consistent with the relative insensitivity for invertebrates. The chronic value from a 69-day,
early-life-stage test of rainbow trout in 21 mg/L hardness water was 15 ug/L (USEPA 1986, Eisler 2000,
WE 2002).
8.2.2.3 Total Dissolved Solids
The Alaskan Water Quality Standard for Growth and Propagation of Fish, Shellfish, Other Aquatic Life
and Wildlife states: "TDS may not exceed 1,000 mg/L. A concentration of TDS may not be present in
water if that concentration causes or reasonably could be expected to cause an adverse effect on aquatic
life" (ADEC 2011). Meeting the state standard for TDS proved difficult at the Red Dog zinc and lead mine
(USEPA 1998, 2008). Laboratory tests of synthetic TDS for effluents from Red Dog and Kensington
Mines caused no statistically significant effects on rainbow trout embryo viability or fry survival or
weight, but did show statistically significant effects on chironomid larvae at 2,089 mg/L (Red Dog) and
1,750 mg/L (Kensington) (Chapman et al. 2000). However, the toxicity of TDS depends on the specific
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 77,7April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
mixture composition; chironomids are relatively tolerant of major ion mixtures (USEPA 2011). Also, the
rainbow trout tests did not include the State of Alaska's endpoint of concern, egg fertilization.
8.2.2.4 Whole Leachates and Effluents
Metals and other aqueous contaminants have combined toxic effects that may be concentration additive,
effects additive, or more or less than additive. The assumption of concentration additivity is considered
to provide the best general approximation of combined metals effects. If, as in this assessment, the
number of metals potentially discharged is large, less than and more than additive interactions may
roughly average out. However, pairwise laboratory tests of defined metal mixtures indicate the
complexity of potential interactions. For example, Chinook salmon avoided a mixture of 0.9 ug/L cobalt
and 1.0 ug/L copper, which suggests that cobalt has no effect on copper avoidance at low levels, but, at
overtly toxic levels (43 ug/L copper), copper does increase avoidance (Hansen et al. 1999).
As discussed above with respect to copper (Section 8.2.2.1), field studies of streams contaminated by
copper and other metals indicate that the laboratory-based criteria are not fully protective of aquatic
communities. Hence, the screening of metal mixtures by applying an additive model to criteria and
equivalent benchmarks does not overestimate and probably underestimates effects on aquatic
communities.
8.2.2.5 Ore-Processing Chemicals
Of the proposed ore-processing chemicals, sodium ethyl xanthate is the primary contaminant of concern
(Section 6.4.2.3). An assessment by Environment Australia generated a predicted no effect concentration
of 1 ug/L (NICNAS 2000). Australia and New Zealand have established a trigger value of 0.05 ug/L to
protect aquatic life (ANZECC 2000). However, because relatively little testing has been done, this is a
"low reliability" value that "may not protect the most sensitive species." Rainbow trout appear to be
relatively tolerant, with a range of lethal concentrations of 1 to 50 mg/L depending on test conditions
(Fuerstenau et al. 1974, Webb et al. 1976). Other fish gave median lethal concentrations of 0.01 to
10 mg/L (emerald shiner) and 0.32 to 3.2 mg/L (fathead minnow) (NICNAS 1995). Aquatic
invertebrates are represented by only Daphnia magna, which has a median effective concentration
(EC50) of 0.35 mg/L (Xu etal. 1988).
8.2.3 Risk Characterization
Risk characterization was performed in stages. First, screening was performed against mean
concentrations of tailings and waste rock leachates to determine whether leachates pose a potential risk
and which constituents contribute to risks. Second, contaminants of concern from the initial screening
were screened against estimated ambient concentrations for routine operations and for WWTP failure.
The implications of potential toxic effects are discussed in terms of their spatial distribution.
Contaminants were screened for risks to aquatic biota by comparing exposure levels to criteria or other
ecotoxicological benchmarks using a risk quotient (Box 8-3). This conventional approach was used to
determine which contaminants and materials are likely to be toxic (Tables 8-4 through 8-8).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -c April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
BOX 8 3. USE OF RISK QUOTIENTS TO ASSESS TOXICOLOGICAL EFFECTS
A risk quotient (Q) equals the exposure level divided by the ecotoxicological benchmark. If the quotient
exceeds 1, the effect implied by the benchmark is expected to occur, but with some uncertainty (see below).
Quotients much larger than 1 suggest larger effects than those that define the benchmark, with greater
confidence that an adverse effect would occur. Quotients much smaller than 1 suggest that even small
effects are unlikely. The acute criterion, or criterion maximum concentration (CMC), estimates a
concentration at which 5% of aquatic species experience some mortality among later life stages in short-
term exposures. The chronic criterion, or criterion continuous concentration (CCC), estimates a
concentration at which 5% of aquatic species experience decreased survival, growth, or reproduction in
longer-term exposures. The criteria, or equivalent numbers when criteria are not available, are relatively well-
accepted as approximate thresholds for significant effects. Thus, these values are the ecotoxicological
benchmarks used as the divisor for calculating quotients in the screening portion of this assessment.
To describe the results of screening using chronic criteria or equivalent benchmarks (acute criteria and less
protective benchmarks would be interpreted differently) in a consistent manner, the following scale was
developed:
• Q < 1 = not overtly toxic
• 2 > Q > 1 = marginally toxic
• 10 > Q > 2 = moderately toxic
• 100 > Q > 10 = highly toxic
• Q > 100 = extremely toxic
8.2.3.1 Screening Leachate Constituents
The results of screening inorganic leachate constituents from tailings and waste rock tests are presented
in Tables 8-4 through 8-8. All have at least moderate chronic toxicity based on their estimated total
toxicity (sums of chronic toxic units), and all are predicted to be acutely toxic if the BLM-based copper
criterion is used instead of the state standard. In all cases, copper is the dominant source of toxicity. The
acidic pre-Tertiary leachate is estimated to be extremely toxic, with copper concentrations thousands of
times higher than the chronic criterion. Figure 8-3 shows the copper concentrations of leachates and
ambient water in relation to state standards.
Tailings slurry would also contain processing chemicals, particularly sodium ethyl xanthate. The
predicted concentration in the slurry (1.5 mg/L) is above or within the range of acute lethality for fish
and well above the level for Daphnia. Therefore, the aqueous phase of the slurry delivered to the TSF
would be moderately toxic due to xanthate alone.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-36
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Figure 8 3. Comparison of copper concentrations in leachates and background water to state
hardness based acute (CMC) and chronic (CCC) water quality criteria for copper. North Fork
Koktuli River background water; Tails HCT leachate from humidity tests of tailings; Supernatant
leachate from column tests of tailings; PWZ Pebble West pre Tertiary; and PEZ Pebble East
pre Tertiary. Copper concentrations in tailings leachate in the field would be expected to lie in the
lower blue triangle. Copper concentrations in waste rock leachate would be expected to lie in the
upper blue triangle. Data sources: Appendix H and PLP 2011.
10,000
1,000
=L 100
y
01
"o
I/)
VI
10
0.1
PEZ
Waste Rock
Supernatant
Mean
North Fork Koktuli
50 100 150 200
Hardness mg/L CaCO3
250
300
350
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-37
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.2.3.2 Screening Contaminants in Receiving Waters
Concentrations of contaminants of concern were calculated at the gages on all three receiving streams
for each mine scenario (Pebble 0.25, 2.0, and 6.5), and those concentrations were screened against
chronic criteria and benchmarks. Because the Pebble 6.5 scenario resulted in the highest concentrations,
screening results for this mine size are presented in Tables 8-16 and 8-17 to show which contaminants
remain of concern after dilution. In the Pebble 6.5 scenario under routine operations, copper is
estimated to exceed chronic water quality criteria at all stations on the South Fork Koktuli River, two of
six stations on the North Fork Koktuli River, and three of seven stations on Upper Talarik Creek. The
pattern of exceedance is the same for the Pebble 6.5 scenario with WWTP failure, except that all stations
on the North Fork Koktuli River exceed the copper criterion. Cadmium and zinc also exceed chronic
criteria, but at fewer stations and by much smaller magnitudes. No other metal exceeded a criterion or
benchmark.
The concentrations of major ions are a particular concern at mine sites because of the leaching of large
volumes of crushed rock. However, the estimates of TDS, both with and without WWTP failure, are
within state standards (Tables 8-16 and 8-17). Without toxicity information on the dissolved solids
mixtures that would occur at the site, we must assume that the standard is protective.
The concentration of sodium ethyl xanthate was not estimated in the receiving streams. Although the
aqueous phase of tailings slurry would be toxic due to xanthate, we expect that the xanthate would occur
at non-toxic levels in ambient waters below TSFs, due to degradation and dilution (Xu et al. 1988).
8.2.3.3 Screening Total Metal Toxicity in Receiving Waters
Table 8-18 presents the sums of quotients across all of the nine metals of concern (excluding selenium,
which has a different mode of exposure and toxicity), for all three mine scenarios and gage locations. In
addition, the sums of quotients for background water are presented. As is expected for streams draining
a surficial ore body, background metal concentrations are elevated. Although total metal toxicities are
estimated to be significantly higher than any individual metal, copper is responsible for most of the
estimated toxicity. Therefore, copper concentrations in contributing loads and ambient waters, as well
as quotients with respect to chronic criteria for the receiving waters, are presented for all three mine
scenarios and gage locations in Table 8-19. The same information but with WWTP failure is presented in
Table 8-20.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ o April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 16. Estimated concentrations of contaminants of concern and associated risk quotients for the Pebble 6.5 mine scenario, at locations in the mine scenario watersheds. See Box 8 3 for a description of how
calculated.
Stream and
Gage
Copper
Mg/L
Quotient
Aluminum
Mg/L
Quotient
Cadmium
Mg/L
Quotient
Cobalt
Mg/L
Quotient
Manganese
Mg/L
Quotient
Nickel
Mg/L
Quotient
Lead
Mg/L
Quotient
Selenium
Mg/L
Quotient
Zinc
Mg/L
Quotient
risk quotients were
Total Dissolved Solids
mg/L
Quotient
South Fork Koktuli River
SK100G
SK100F
SK100CP2a'b
SK124A
SK124CP"
SK100C
SK100CP13
SK119A
SK119CPa
SK100B1
SKIOOB"
NA
160
57
1.3
1.3
20
20
NA
1.4
11
7.9
NA
150
53
1.2
1.2
19
19
NA
1.3
10
7.4
NA
56
28
54
53
44
44
NA
15
26
21
NA
0.64
0.32
0.62
0.60
0.51
0.51
NA
0.18
0.30
0.24
NA
0.23
0.09
0.05
0.05
0.07
0.07
NA
0.02
0.04
0.03
NA
3.5
1.4
0.84
0.83
1.0
1.0
NA
0.31
0.63
0.49
NA
1.4
0.54
1.4
1.4
0.60
0.43
NA
0.43
0.60
0.43
NA
0.56
0.22
0.58
0.56
0.24
0.17
NA
0.17
0.24
0.17
NA
97
58
79
77
60
60
NA
10
33
24
NA
0.14
0.08
0.11
0.11
0.09
0.09
NA
0.02
0.05
0.04
NA
1.7
0.88
2.4
2.4
1.8
1.8
NA
0.31
1.1
0.83
NA
0.17
0.09
0.24
0.24
0.18
0.18
NA
0.03
0.11
0.08
NA
0.11
0.10
0.23
0.22
0.19
0.19
NA
0.08
0.11
0.10
NA
0.38
0.33
0.78
0.77
0.64
0.64
NA
0.28
0.39
0.34
NA
0.70
0.34
1.1
1.1
0.82
0.82
NA
0.42
0.56
0.43
NA
0.14
0.07
0.22
0.22
0.16
0.16
NA
0.08
0.11
0.09
NA
32
13
18
18
15
15
NA
2.7
9.2
7.4
NA
1.4
0.58
0.78
0.76
0.66
0.66
NA
0.12
0.40
0.32
NA
60
49
430
420
290
280
NA
53
180
130
NA
0.06
0.05
0.43
0.42
0.29
0.28
NA
0.05
0.18
0.13
North Fork Koktuli River
NK119A
NK119CP23
NK119B
NK119CP13
NK100O
NK100B
NK100A1
NK100A6
1.8
1.4
0.42
1.1
0.62
0.72
0.73
0.53
1.7
1.3
0.40
1.0
0.58
0.67
0.68
0.50
20
19
17
19
33
29
17
20
0.23
0.22
0.19
0.22
0.38
0.33
0.20
0.23
0.02
0.02
0.01
0.02
0.03
0.03
0.02
0.02
0.37
0.32
0.15
0.26
0.53
0.42
0.28
0.28
0.12
0.11
0.03
0.08
0.80
0.60
0.29
0.25
0.05
0.04
0.01
0.03
0.32
0.24
0.11
0.10
18
15
2.7
12
49
37
17
20
0.03
0.02
0.00
0.02
0.07
0.05
0.03
0.03
0.39
0.37
0.26
0.35
1.4
1.2
0.63
0.61
0.04
0.04
0.03
0.04
0.14
0.12
0.06
0.06
0.06
0.06
0.05
0.06
0.16
0.16
0.07
0.09
0.22
0.22
0.19
0.21
0.54
0.55
0.24
0.29
0.54
0.43
0.14
0.34
0.63
0.54
0.34
0.30
0.11
0.09
0.03
0.07
0.13
0.11
0.07
0.06
2.1
1.9
1.5
1.8
9.8
7.8
4.8
4.1
0.09
0.08
0.06
0.08
0.43
0.34
0.21
0.18
58
51
30
44
250
200
110
97
0.06
0.05
0.03
0.05
0.25
0.20
0.11
0.10
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119Ab
UT100B'
0.76
1.5
0.39
0.34
0.41
27
3.6
0.71
1.4
0.36
0.32
0.38
25
3.3
18
40
13
9.7
11
17
10
0.21
0.46
0.15
0.11
0.13
0.20
0.12
0.04
0.09
0.01
0.01
0.01
0.05
0.02
0.67
1.4
0.19
0.18
0.18
0.71
0.29
0.61
1.6
0.14
0.11
0.09
0.26
0.11
0.24
0.63
0.06
0.04
0.03
0.11
0.05
19
64
24
14
10
27
17
0.03
0.09
0.03
0.02
0.02
0.04
0.02
1.0
2.0
0.51
0.46
0.44
0.60
0.46
0.10
0.20
0.05
0.05
0.04
0.06
0.05
0.06
0.08
0.04
0.04
0.04
0.08
0.07
0.20
0.26
0.14
0.13
0.14
0.26
0.24
0.40
0.81
0.19
0.18
0.18
0.23
0.17
0.08
0.16
0.04
0.04
0.04
0.05
0.03
4.1
7.2
2.0
1.3
1.5
6.9
2.4
0.18
0.31
0.09
0.06
0.06
0.30
0.10
73
94
48
49
48
49
47
0.07
0.09
0.05
0.05
0.05
0.05
0.05
Notes:
3 Confluence point where virtual gage was created because physical gage does not exist.
b 1/3 of total return flow is transferred from SK100CP2to UTllQAto represent interbasin transfer at this location.
c Wastewater treatment plant discharges 50% of its flow at this site.
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
NA = not applicable; the stream at the gage would be destroyed.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-39
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
K!nS:»F«^iTiffl!S!i5!i!!SiTRffl!!iH!T^
Talarik Creek would be unchanged from Table 8 16. See Box 8 3 for a description of how risk quotients were calculated.
Stream and
Gage
Copper
Mg/L
Quotient
Aluminum
Mg/L
Quotient
Cadmium
Mg/L
Quotient
Cobalt
Mg/L
Quotient
Manganese
Mg/L
Quotient
Nickel
Mg/L
Quotient
Lead
Mg/L
Quotient
Selenium
Mg/L
Quotient
Zinc
Mg/L
Quotient
Total Dissolved Solids
mg/L
Quotient
South Fork Koktuli River
SK100G
SK100F
SK100CP2a'b
SK124A
SK124CP"
SK100C
SK100CP13
SK119A
SK119CPa
SK100B1
SKIOOB"
NA
160
57
110
100
88
87
NA
1.4
48
34
NA
150
53
100
97
82
82
NA
1.3
45
32
NA
56
28
54
53
44
44
NA
15
26
21
NA
0.64
0.32
0.62
0.60
0.51
0.51
NA
0.18
0.30
0.24
NA
0.23
0.09
0.19
0.19
0.15
0.15
NA
0.02
0.09
0.07
NA
3.5
1.4
3.0
2.9
2.4
2.4
NA
0.31
1.4
1.0
NA
1.4
0.54
1.4
1.4
1.1
1.1
NA
0.12
0.60
0.43
NA
0.56
0.22
0.58
0.56
0.44
0.43
NA
0.05
0.24
0.17
NA
97
58
79
77
60
60
NA
10
33
24
NA
0.14
0.08
0.11
0.11
0.09
0.09
NA
0.02
0.05
0.04
NA
1.7
0.88
2.4
2.4
1.8
1.8
NA
0.31
1.1
0.83
NA
0.17
0.09
0.24
0.24
0.18
0.18
NA
0.03
0.11
0.08
NA
0.11
0.10
0.28
0.27
0.22
0.22
NA
0.08
0.13
0.11
NA
0.378
0.33
0.96
0.94
0.76
0.75
NA
0.28
0.45
0.39
NA
0.70
0.34
1.1
1.1
0.82
0.82
NA
0.42
0.56
0.43
NA
0.14
0.07
0.22
0.22
0.16
0.16
NA
0.08
0.11
0.09
NA
32
13
26
25
20
20
NA
2.7
12
9.3
NA
1.4
0.58
1.1
1.1
0.87
0.87
NA
0.12
0.52
0.40
NA
60
49
430
420
280
280
NA
53
180
130
NA
0.06
0.05
0.43
0.42
0.28
0.28
NA
0.05
0.18
0.13
North Fork Koktuli River
NK119A
NK119CP23
NK119B
NK119CP13
NK100O
NK100B
NK100A1
NK100A6
1.8
1.4
0.42
1.1
58
44
20
17
1.7
1.3
0.40
1.0
54
41
19
16
20
19
17
19
33
29
17
20
0.23
0.22
0.19
0.22
0.38
0.33
0.20
0.23
0.02
0.02
0.01
0.02
0.11
0.08
0.04
0.04
0.37
0.32
0.15
0.26
1.7
1.3
0.67
0.61
0.12
0.11
0.03
0.08
0.80
0.60
0.29
0.25
0.05
0.04
0.01
0.03
0.32
0.24
0.11
0.10
18
15
2.7
12
49
37
17
20
0.03
0.02
0.00
0.02
0.07
0.05
0.03
0.03
0.39
0.37
0.26
0.35
1.4
1.2
0.63
0.61
0.04
0.04
0.03
0.04
0.14
0.12
0.06
0.06
0.06
0.06
0.05
0.06
0.18
0.18
0.08
0.09
0.22
0.22
0.19
0.21
0.63
0.62
0.27
0.32
0.54
0.43
0.14
0.34
0.63
0.54
0.34
0.30
0.11
0.09
0.03
0.07
0.13
0.11
0.07
0.06
2.1
1.9
1.5
1.8
14
11
6.2
5.3
0.09
0.08
0.06
0.08
0.60
0.48
0.27
0.23
59
51
31
45
250
200
110
97
0.06
0.05
0.03
0.05
0.25
0.20
0.11
0.10
Notes:
a Confluence point where virtual gage was created because physical gage does not exist.
b 1/3 of total return flow is transferred from SK100CP2to UTllQAto represent interbasin transfer at this location.
c Wastewater treatment plant discharges 50% of its flow at this site.
d USGS 15302200.
e USGS 15302250.
NA = not applicable; the stream at the gage would be destroyed.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-40
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 18. Estimated total toxicity of metals of concern for each mine scenario, under routine
operations and with wastewater treatment plant failure, at locations in the mine scenario
watersheds. Values are the sums of the toxic quotients for the metals of concern.
Stream and
Gage
Background
Pebble 0.25
Routine
Operations
WWTP
Failure
Pebble 2.0
Routine
Operations
WWTP
Failure
Pebble 6.5
Routine
Operations
WWTP
Failure
South Fork Koktuli River
SK100G
SK100F
SK100CP2a'b
SK124A
SK124CP"
SK100C
SK100CP13
SK119A
SK119CPa
SK100B1
SKIDDED
3.1
2.4
-
2.5
2.3
1.2
-
1.2
1.1
3.4
2.7
2.6
3.1
3.1
2.7
2.7
1.2
1.2
1.3
1.2
3.4
2.7
2.6
19
18
11
11
1.2
1.2
4.3
3.4
110
22
12
3.6
3.5
7.9
7.9
1.2
1.2
3.4
2.7
110
22
12
29
28
21
21
1.2
1.2
8.6
6.4
NA
160
56
5.4
5.3
22
22
NA
2.4
12
9.1
NA
160
56
110
100
87
87
NA
2.4
48
34
North Fork Koktuli River
NK119A
NK119CP23
NK119B
NK119CP13
NKIOOC"
NK100B
NK100A1
NK100A6
1.0
-
1.0
1.1
1.2
1.1
1.1
1.5
1.4
1.0
1.3
1.6
1.6
1.3
1.2
1.5
1.4
1.0
1.3
8.9
6.0
3.1
2.8
2.8
2.3
1.0
1.8
1.8
1.8
1.4
1.3
2.8
2.3
1.0
1.8
14
10
4.7
4.0
2.8
2.3
1.1
1.9
3.1
2.8
1.9
1.7
2.8
2.3
1.1
1.9
58
44
21
17
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119Ab
UT100B'
0.93
1.3
0.90
0.76
0.89
0.75
0.99
0.96
1.3
0.93
0.80
0.92
1.7
1.1
0.96
1.3
0.93
0.80
0.92
1.7
1.1
0.96
1.8
1.0
0.86
0.97
6.5
1.7
0.96
1.8
1.0
0.86
0.97
6.5
1.7
2.4
5.0
1.1
0.94
1.0
27
4.2
2.4
5.0
1.1
0.94
1.0
27
4.2
Notes:
3 Confluence point where virtual gage was created because physical gage does not exist; no background values available.
b 1/3 of total return flow is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location.
c Wastewater treatment plant discharges 50% of its flow at this site.
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
WWTP = wastewater treatment plant; NA = not applicable, because the stream at the gage would be destroyed.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-41
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Stream and
Gage
Background
Pebble 0.25
CL
AW
Quotient
Pebble 2.0
CL
AW
Quotient
Pebble 6.5
CL
AW
Quotient
South Fork Koktuli River
SK100G
SK100F
SK100CP2a'b
SK124A
SK124CP"
SK100C
SK100CP13
SK119A
SK119CPa
SK100B1
SK100Bd
2.4
1.6
1.6
1.4
1.4
1.4
1.4
0.42
0.42
0.62
0.54
3.2
3.2
1.7
1.1
-
-
-
-
2.4
1.7
1.7
1.3
1.3
1.4
1.4
0.42
0.42
0.54
0.47
2.2
1.6
1.5
1.2
1.2
1.3
1.3
0.39
0.39
0.51
0.44
390
6.8
11
1.1
-
5.3
-
-
110
21
11
1.3
1.3
6.4
6.4
0.43
0.43
2.5
1.9
100
20
10
1.2
1.2
6.0
5.9
0.40
0.40
2.4
1.8
-
720
57
1.3
-
-
5.3
5.3
-
NA
160
57
1.3
1.3
20
20
NA
1.4
11
7.9
NA
150
53
1.2
1.2
19
19
NA
1.3
10
7.4
North Fork Koktuli River
NK119A
NK119CP23
NK119B
NK119CP13
NK1000
NK100B
NK100A1
NK100A6
0.31
0.31
0.41
0.33
0.35
0.40
0.61
0.41
5.3
-
-
1.1
-
0.66
0.60
0.41
0.56
0.43
0.51
0.65
0.45
0.61
0.56
0.39
0.52
0.40
0.48
0.61
0.42
5.3
-
-
1.1
-
5.3
1.8
1.4
0.41
0.99
0.44
0.61
0.69
0.48
1.6
1.3
0.39
0.92
0.41
0.57
0.64
0.45
5.3
-
5.3
1.1
-
5.3
1.8
1.4
0.42
1.1
0.62
0.72
0.73
0.53
1.7
1.3
0.40
1.0
0.58
0.67
0.68
0.50
Upper Talarik Creek
UT100E
UT100D
UT100C2
UT100C1
UT100C
UT119Ab
UT100B'
0.34
0.50
0.34
0.30
0.38
0.21
0.34
-
-
1.7
-
0.34
0.50
0.34
0.30
0.38
0.98
0.42
0.32
0.47
0.31
0.28
0.35
0.92
0.40
3.2
-
-
11
-
0.34
0.63
0.36
0.32
0.39
5.8
1.0
0.32
0.59
0.33
0.30
0.37
5.4
1.0
3.2
3.2
3.2
-
57
-
0.76
1.5
0.39
0.34
0.41
27
3.6
0.71
1.4
0.36
0.32
0.38
25
3.3
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-42
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Stream and
Gage
Background
Pebble 0.25
CL
AW
Quotient
Pebble 2.0
CL
AW
Quotient
Pebble 6.5
CL
AW
Quotient
Notes:
NA = not applicable, because stream at gage location would be destroyed. Blank values (-) indicate there are no contributing loads at that gage under that scenario.
a Confluence point where virtual gage was created because physical gage does not exist.
b 1/3 of total return flow is transferred from SK100CP2 to UTllQAto represent interbasin transfer at this location.
c Wastewater treatment plant discharges 50% of its flow at this site.
d USGS 15302200.
e USGS 15302250.
' USGS 15300250.
CL = contributing loads; AW = ambient waters; quotient = predicted/criterion.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-43
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 20.
waters and associated risk quotients under wastewater treatment plant failure. Upper Talarik Creek would be unchanged from Table 8 17.
See Box 8 3 for a description of how risk quotients were calculated. All concentrations are in M&/L-
Stream and
Gage
Background
Pebble 0.25
CL
AW
Quotient
Pebble 2.0
CL
AW
Quotient
Pebble 6.5
CL
AW
Quotient
South Fork Koktuli River
SK100G
SK100F
SK100CP2a."
SK124A
SK124CPa'c
SK100C
SK100CP13
SK119A
SK119CP3
SK100B1
SKIDDED
2.4
1.6
1.6
1.4
1.4
1.4
1.4
0.42
0.42
0.62
0.54
3.2
3.2
1.7
72
-
-
-
-
2.4
1.7
1.7
18
17
10
10
0.42
0.42
3.7
2.7
2.2
1.6
1.5
17
16
9.0
9.0
0.39
0.39
3.5
2.6
390
6.8
11.
100
-
-
5.3
-
110
21
11
28
26
20
20
0.43
0.43
8.0
5.7
100
20
10
26
25
19
19
0.40
0.40
7.4
5.4
-
720
57
150
-
-
-
5.3
5.3
NA
160
57
110
100
88
87
NA
1.4
48
34
NA
150
53
100
97
82
82
NA
1.3
45
32
North Fork Koktuli River
NK119A
NK119CP23
NK119B
NK119CP13
NK100O
NK100B
NK100A1
NK100A6
0.31
0.31
0.41
0.33
0.35
0.40
0.61
0.41
5.3
-
-
72
-
-
0.66
0.60
0.41
0.56
8.1
5.2
2.6
2.1
0.61
0.56
0.39
0.52
7.6
4.8
2.5
1.9
5.3
-
-
100
5.3
-
1.8
1.4
0.41
1.0
13
9.1
4.1
3.3
1.6
1.3
0.39
0.92
12
8.5
3.8
3.1
5.3
-
5.3
150
5.3
-
1.8
1.4
0.42
1.1
58
44
20
17
1.7
1.3
0.40
1.0
54
41
19
16
Notes:
NA = not applicable, because stream at gage location would be destroyed. Blank values (-) indicate there are no contributing loads at that gage under that scenario.
a Confluence point where virtual gage was created because physical gage does not exist.
b 1/3 of total return flow is transferred from SK100CP2to UTllQAto represent interbasin transfer at this location.
c Wastewater treatment plant discharges 50% of its flow at this site.
d USGS 15302200.
e USGS 15302250.
CL = contributing loads; AW = ambient waters; quotient = predicted/criterion.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-44
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
8.2.3.4 Screening for Severity of Effects
Tables 8-19 and 8-20 indicate that water management assuming both routine operations and WWTP
failure results in exceedance of the BLM-based chronic copper criteria at several locations in the mine
scenario watersheds. However, they do not provide an indication of the severity of effects. For that
purpose, we estimated copper concentrations in stream reaches. The reaches are defined by flow gages
and major confluences for each of the receiving streams (Table 8-21). Those fully mixed copper
concentrations were screened against a series of benchmarks of increasing severity, beginning with the
national criteria, as follows.
• Invertebrate Chronic (1C). The BLM-derived chronic ambient water quality criterion (Table 8-11),
based on toxicity to sensitive aquatic invertebrates in extended exposures. It implies reduced
survival, growth, or reproduction of copper-sensitive invertebrates.
• Invertebrate Acute (IA). The BLM-derived acute ambient water quality criterion (Table 8-11), based
on lethality to sensitive aquatic invertebrates in short-term exposures. It implies greatly reduced
survival of copper-sensitive invertebrates.
• Fish Avoidance (FA). The BLM-derived concentration at which 20% of rainbow trout avoid the
contaminated water (Table 8-14). It implies loss of habitat due to aversion.
• Fish Sensory (FS). The BLM-derived concentration at which the olfactory sensitivity of rainbow
trout is reduced by 20% (Table 8-14). It implies an inability to identify natal streams, reduced
predatory avoidance, and other behavioral effects.
• Fish Reproduction (FR). The BLM-derived chronic value for rainbow trout, the concentration at
which their fecundity or the survival and growth of their larvae are reduced (Table 8-13). It implies
partial or complete reproductive failure of salmonids.
• Fish Kill (FK). The BLM-derived rainbow trout LCso, the concentration at which half of adults and
juveniles are killed in short-term exposures (Table 8-13). It implies a fish kill and, in the long term,
local extirpation offish populations.
These copper benchmarks were applied to streams reaches rather than the point values in the prior
screening assessment (Table 8-21). The combinations of stream reaches and mine scenarios at which
these copper benchmarks are exceeded are shown in Tables 8-22 and 8-23. The range of severities of
effects extends from no overt effects expected (-) to the full range of effects up to numerous dead post-
larval salmonids (IC/IA/FA/FS/FL/FR/FK).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ . g April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 21. Description of stream reaches affected in the mine scenarios and sources of the concentration estimates applied to the
stream reaches.
Reach Designation3
Reach Description11
Length (km)c
Concentration Assigned and Qualifiers'1
South Fork Koktuli River— Mainstem
SK100B
SK100B1
SK100CP1/
SK119CP
SK100C
SK100CP2/
SK124CP
SK100F
SK100G
SK Rock
SK Halo/Rock
SK100B to confluence of the South and
North Fork Koktuli Rivers
SKlOOBltoSKlOOB
SK100CP1/ SK119CP confluence to
SK100B1
SK100C to SK100CP1
SK100CP2/SK124CP confluence to
SK100C
SK100F to SK100CP2
SK100G to SK100F (not Pebble 6.5)
Waste rock to SK100F (Pebble 6.5 only)
Dewatering halo and rock pile to SK100G
(Pebble 0.25 and 2.0)
22.5
4.53
4.32
1.23
6.35
10.7
3.25/3.25/NA
NA/NA/0.83
1.87/0.54/NA
SK100B, overestimates lower end due to dilution
SK100B1, small overestimate of lower end due to dilution
Mixed SK100CP1 and SK119CP, little dilution downstream
SK100C, negligible further dilution in short reach
Mixed SK100CP2 and SK124CP, little dilution downstream
Mean SK100F and SK100CP2 due to significant dilution
Mean SK100G and SK100F due to significant dilution
SK 100F, assuming input near base of rock pile and short reach
SK 100G, assuming input near base of rock pile and short reach
South Fork Koktuli River— Tributaries
SK Headwaters
SKTSF1
SK119A
SK124A
SK WWTP
Headwaters to SK119A (Pebble 0.25)
TSF1 to SK119A (Pebble 2.0)
SK119AtoSK119CP
SK124A to SK124CP
WWTP to SK124A
7.03
6.75
1.59/1.59/1.54
2.60
5.01
Background for Pebble 0.25 scenario
SK119A, significant dilution so underestimate
SK119A, for Pebble 0.25 and Pebble 2.0 scenarios, dilution is minimal;
SK119CP for remnant reach in Pebble 6.5 scenario, when SK119A destroyed
SK124A, no dilution in this reach within precision
SK124A, underestimate of upper end from dilution of WWTP and, in Pebble 6.5 scenario,
TSF 3 leachate
North Fork Koktuli River— Mainstem
NK100A
NK100A1
NK100B
NK100CP1/NK100C
NK100C
NKWWTP
NK100A to confluence of the South and
North Fork Koktuli Rivers
NKlOOAltoNKlOOA
NKlOOBtoNKlOOAl
NK100CP1/NK100C confluence to
NK100B
NK100C to confluence NK119A stream
WWTP Discharge to NK100C
4.69
8.35
20.4
0.79
0.19
4.34
NK100A, overestimates lower end due to dilution
N100A1, which has a small contributing load in the Pebble 2.0 and Pebble 6.5
scenarios, so small overestimate
NK100B, approximately two times dilution over long reach in Pebble 0.25 scenario but
no change in Pebble 2.0 and Pebble 6.5 scenarios due to balance of dilution by tailings
leachate
Mixed NK100CP1 and NK100C, little dilution downstream
NK100C, negligible further dilution in tiny reach
NK100C, underestimate of upper end, but assuming negligible dilution
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-46
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 21. Description of stream reaches affected in the mine scenarios and sources of the concentration estimates applied to the
stream reaches.
Reach Designation3
Reach Description11
Length (km)c
Concentration Assigned and Qualifiers'1
North Fork Koktuli River-Tributaries
NK119B/NK100CP2
NK119A
NKTSF1
NK Headwaters
NK119B/NK100CP2 confluence to
NK119CP1
NK119AtoNK119CP2
TSFltoNK119A
Headwaters or dewatering halo to
NK119B
0.43
1.31
0.64
6.84/6.84/6.57
Mixed NK119B and NK100CP2, little dilution downstream
NK119A, little dilution downstream
NK119A, assuming input near toe of dam and short reach
NK119B, which has a small contributing load in the Pebble 6.5 scenario from tailings
leachate at its upper end, so small underestimate
Upper Talarik Creek— Mainstem
UT100B
UT100C
UT100C1
UT100C2
UT100D
UT100E
UT Rock
UTIOOBto Iliamna Lake
UTIOOCto UT119 confluence
UT100C1 to UT100C
UT100C1 to UT100C2
UT100D to UT100C2
UTIOOEto UT100D (Pebble 0.25 only)
Waste Rock to UT100D (not for Pebble
0.25)
23.3
4.34
7.55
6.92
6.13
7.08/NA/NA
NA/2.08/0.15
UT100B, considerable dilution would occur in this long reach, so only the upper end
would not be overestimate
UT100C, some unquantified dilution at lower end so overestimate there
UT100C1, minimal change in concentration
UT100C2, dilution and loading balance in reach
Mean of UT100D and UT100C2 because of significant dilution in the reach
UT100E flows at background concentrations for the Pebble 0.25 scenario
UT100D, assuming input near base of rock pile and short reach
Upper Talarik Creek— Tributaries
UT Headwaters
Headwaters to UT119A
6.47
UT119A receives interbasin transfer; assumed along nearly all of length but
overestimates at upper end
Notes:
3 Reaches are designated by the gage or other feature at their heads. Designations in the form G1/G2 indicate the confluence of a stream and tributary with gages Gl and G2 above the confluence.
b Upper and lower bounds of the reach.
c Lengths that differ among mine sizes are presented as Pebble 0.25/Pebble 2.0/Pebble 6.5.
d Concentrations are point estimates at upstream gages from Table 8-19, flow-weighted mixtures of concentrations at upstream gages, or means of upstream and downstream gages. Qualifiers
explain the possibility of over or underestimation.
WWTP = wastewater treatment plant
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-47
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 22. Copper concentration benchmarks exceeded in ambient waters in each reach and for
each mine scenario during routine operations. Reaches are described in Table 8 21.
Reach Designation3
Pebble 0.25
Copper
(Mg/L)
Effects
Pebble 2.0
Copper
(Mg/L)
Effects
Pebble 6.5
Copper
(Mg/L)
Effects
South Fork Koktuli River— Mainstem
SK100B
SK100B1
SK100CP1/SK119CP
SK100C
SK100CP2/SK124CP
SK100F
SK100G
SK Rock
SK Halo/Rock
<0.47
0.54
0.95
1.4
1.5
1.7
2.0
NA
2.4
-
-
1C
1C
1C
NA
IC/IA
<1.9
2.5
3.8
6.4
6.1
16
66
NA
110
1C
IC/IA
IC/IA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA/FI/FR/FK
NA
IC/IA/FA/FI/FR/FK
<7.9
11
16
20
17
110
NA
160
NA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA/FS/FR/FK
NA
IC/IA/FA/FS/FR/FK
NA
South Fork Koktuli River— Tributaries
SK Headwaters
SKTSF1
SK119A
SK124A
SK WWTP
0.42
NA
0.42
1.3
1.3
NA
-
NA
>0,.43
0.43
1.3
1.3
NA
-
-
NA
NA
1.5
1.3
1.3
NA
NA
-
North Fork Koktuli River— Mainstem
NK100A
NK100A1
NK100B
NK100CP1/NK100C
NK100C
NKWWTP
<0.45
0.65
0.51
0.56
0.43
>0.43
-
-
-
<0.48
0.69
0.61
0.76
0.44
>0.44
-
-
-
<0.53
0.73
0.72
0.86
0.62
>0.62
-
-
-
North Fork Koktuli River-Tributaries
NK119B/NK100CP2
NK119A
NKTSF1
NK Headwaters
0.57
0.66
0.66
0.41
-
-
1.1
1.8
1.8
0.41
1C
IC/IA
IC/IA
-
1.2
1.8
1.8
0.42
1C
IC/IA
IC/IA
-
Upper Talarik Creek— Mainstem
UT100B
UT100C
UT100C1
UT100C2
UT100D
UT100E
UT Rock
<0.42
0.38
0.30
0.34
0.42
<0.34
NA
-
-
-
NA
<1.0
0.39
0.32
0.36
0.49
NA
0.63
-
-
NA
<3.6
0.41
0.34
0.38
0.95
NA
1.5
IC/IA
-
-
NA
Upper Talarik Creek— Tributaries
UT Headwaters
0.98
5.8
IC/IA/FA
27
IC/IA/FA
Notes:
3 Reaches are designated by the gage or other feature at their heads. Designations in the form G1/G2 indicate the confluence of a stream and
tributary with gages Gl and G2 above the confluence.
1C = invertebrate chronic; IA = invertebrate acute; FA = fish avoidance; FS = fish sensory; FR = fish reproduction; FK = fish kill;
NA = not applicable.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-48
April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Table 8 23. Copper concentration benchmarks exceeded in ambient waters in each reach and for
each mine scenario during a wastewater treatment plant failure. Reaches are described in Table 8
21.
Reach Designation3
Pebble 0.25
Copper
(Mg/L)
Effects
Pebble 2.0
Copper
(Mg/L)
Effects
Pebble 6.5
Copper
(Mg/L)
Effects
South Fork Koktuli River— Mainstem
SK100B
SK100B1
SK100CP1/SK119CP
SK100C
SK100CP2/SK124CP
SK100F
SK100G
SK Rock
SK Halo/Rock
<2.7
3.7
5.6
9.7
9.0
1.7
2.0
NA
2.4
IC/IA
IC/IA
IC/IA/FA
IC/IA/FA
IC/IA/FA
1C
1C
NA
IC/IA
<57
8.0
12
20
19
16
66
NA
110
IC/IA/FA/FS/FR
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA/FI/FR/FK
NA
IC/IA/FA/FI/FR/FK
<34
48
70
88
73
110
NA
160
NA
IC/IA/FA/FS/FR
IC/IA/FA/FS/FR
IC/IA/FA/FS/FR/FK
IC/IA/FA/FS/FR/FK
IC/IA/FA/FS/FR/FK
IC/IA/FA/FS/FR/FK
NA
IC/IA/FA/FS/FR/FK
NA
South Fork Koktuli River— Tributaries
SK Headwaters
SKTSF1
SK119A
SK124A
SK WWTP
0.42
NA
0.42
18
>18
NA
-
IC/IA/FA
IC/IA/FA
NA
>0.43
0.43
28
>28
NA
-
IC/IA/FA/FS/FR
IC/IA/FA/FS/FR
NA
NA
1.4
110
>110
NA
NA
-
IC/IA/FA/FS/FR/FK
IC/IA/FA/FS/FR/FK
North Fork Koktuli River— Mainstem
NK100A
NK100A1
NK100B
NK100CP1/NK100C
NK100C
NKWWTP
<2.1
2.6
5.2
5.7
8.1
>8.1
IC/IA
IC/IA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA
<3.3
4.1
9.1
10
13
>13
IC/IA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA
IC/IA/FA
<17
20
44
47
58
>58
IC/IA/FA
IC/IA/FA/FS
IC/IA/FA/FS/FR
IC/IA/FA/FS/FR
IC/IA/FA/FS/FR
IC/IA/FA/FS/FR
North Fork Koktuli River— Tributaries
NK119B/NK100CP2
NK119A
NKTSF1
NK Headwaters
0.57
0.66
0.66
0.41
-
-
1.1
1.8
1.8
0.41
1C
IC/IA
IC/IA
1.2
1.8
1.8
0.42
1C
IC/IA
IC/IA
Notes:
a Reaches are designated by the gage or other feature at their heads. Designations in the form G1/G2 indicate the confluence of a stream and
tributary with gages Gl and G2 above the confluence.
1C = invertebrate chronic; IA = invertebrate acute; FA = fish avoidance; FS = fish sensory; FR = fish reproduction; FK = fish kill;
NA = not applicable; WWTP = wastewater treatment plant.
8.2.3.5 Dilution Zones
Analyses in the preceding sections (Sections 8.2.3.3 through 8.2.3.4) have dealt with risks from
concentrations in fully mixed locations or reaches. Prior to achieving full mixing, the effluent plume
would create a gradient from undiluted to fully diluted, within which exposures would be higher than
the fully mixed concentrations. This should not be an issue for a plume of properly treated wastewater
plume, but could result in locally high exposures under WWTP failure. The untreated wastewater
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-49
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
concentrations of copper alone (Table 8-9) would be sufficient to cause lethality in trout, and all other
toxic endpoint effects among survivors, in both receiving streams under all three mine scenarios.
For waste rock and tailings leachates, effluent can enter a stream just below the pile or dam. Leachate
that drains to shallow aquifers would enter a stream through its cobble and gravel substrate. Where
leachates enter, benthic invertebrates and fish eggs and larvae could be exposed to a range of
concentrations, from undiluted to highly diluted leachate (the scenarios include significant dilution by
groundwater). Undiluted concentrations of metals of concern are listed in Table 8-9. NAG leachate,
which would enter Upper Talarik Creek in the Pebble 2.0 and Pebble 6.5 scenarios at 3.2 ug/L copper,
would be sufficient to cause mortality in invertebrates unless it was significantly diluted by
groundwater first. The NAG and PAG leachate, which would enter South Fork Koktuli River in the
Pebble 2.0 scenario at 395 ug/L copper, would be more than six times the acute lethal concentration for
trout. The Pebble 6.5 scenario, at 735 ug/L copper, would require dilution by more than a factor of 10 to
avoid acute lethality of trout. At the more sensitive end of the spectrum, invertebrates would require
dilution of the Pebble 6.5 scenario NAG+PAG leachate by a factor of 490 to avoid chronic toxicity.
Tailings leachates would enter tributaries of the South and North Fork Koktuli Rivers at an undiluted
copper concentration of 5.3 ug/L, unless they were significantly diluted by groundwater first. This is
sufficient to kill invertebrates and to cause avoidance by trout. It would require dilution by factors of
3.5 to 5 to avoid chronic toxicity to invertebrates.
8.2.3.6 Spatial Distribution of Estimated Effects
The results of screening for total metals and copper and the analysis of severity are presented in
Tables 8-18 to 8-23 and summarized below. They are best understood by consulting the maps of the
three mine scenario footprints showing streams and gages in Figures 7-15 through 7-17.
Pebble 0.25 Scenario—Routine Operations
• South Fork Koktuli River. Copper loading from NAG waste rock at SK100G and SK100F would
slightly increase the naturally high levels of copper and other metals and would increase estimated
concentrations to chronically toxic levels for invertebrates in the first 5 km. WWTP effluent would
enter the SK124 tributary. That effluent would slightly decrease copper concentration due to
treatment to achieve criteria. Concentrations would decline downstream to SK100B due to dilution.
• North Fork Koktuli River. Input of TSF1 leachate to the tributary at NK199A would increase copper
levels by more than five-fold from background, so that the acute and chronic copper criteria would
be exceeded (i.e., invertebrates would be depleted). Input of water treatment effluent at NK100C
would increase metal concentrations over background so that, although copper would meet the
criterion, the total metal risk quotient would rise to 1.7. At the confluence of the TSF- and WWTP-
influenced streams (NK100B), copper concentrations would be below the criterion and total metal
toxicity would be marginal and decline downstream.
• Upper Talarik Creek. Copper loading would come entirely from interbasin transfer to the UT119
tributary and copper would not reach toxic levels.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ „ April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
Pebble 2.0 Scenario—Routine Operations
• South Fork Koktuli River. Input of NAG and PAG waste rock leachate entering below the waste rock
pile and for 3.3 km downstream would raise copper concentrations to levels sufficient to kill trout
and other salmonids. Levels at SK100F and for 18 km downstream would be sufficient to cause
avoidance by trout and severely deplete invertebrates. Levels sufficient to cause acute lethality to
invertebrates would extend another 8.9 km and chronic toxicity would extend for some distance
beyond that. However, levels would be relatively low in the SK124 tributary due to dilution by the
water treatment outfall.
• North Fork Koktuli River. The pattern of input would be the same as for the Pebble 0.25 scenario,
but copper and total metals would be highly toxic to invertebrates in the NK119A tributary, because
the larger TSF would release more leachate. Concentrations would decrease below the confluence of
the tributary and the mainstem below NK100C due to WWTP effluent and background water, so that
by NK100A concentrations would be close to background.
• Upper Talarik Creek. Metals from NAG waste rock leachate would enter at UT100D and raise the
naturally marginally toxic total metal levels but not copper. Concentrations would decline
downstream to non-toxic levels in the mainstem. Interbasin transfers would raise copper and total
metal concentrations to levels that would be highly toxic to invertebrates in the UT119 tributary.
Pebble 6.5 Scenario—Routine Operations
• South Fork Koktuli River. SK100G would be buried by waste rock and SK119A would be buried by
tailings. SK100F would exceed the copper criterion by more than 100-fold due to NAG and PAG
waste rock leachate and TSF leakage, achieving levels sufficient to kill juvenile and adult trout and
other salmonids for 12 km. For another 45 km, aversion would occur, and for 26 km more
invertebrates would be killed or have inhibited reproduction. On the SK119 and SK124 tributaries
toxicity would be low despite TSF leakage and WWTP effluent.
• North Fork Koktuli River. TSF leakage would enter both the NK119A and NK119B tributaries,
resulting in copper and total metal toxicity to invertebrates for 2.4 km. Due to the WWTP, no copper
toxicity would occur at or below NK100C but total metal toxicity to invertebrates would occur.
• Upper Talarik Creek. NAG waste rock leachate would enter the creek from the base of the expanded
waste rock pile, resulting in copper toxicity at and below UT100D (just below the pile), and total
metal toxicity at all three gages, but particularly gage UT100D. Due to interbasin transfer from the
South Fork Koktuli River, copper in the UT119 tributary would be highly toxic to invertebrates and
aversive to trout. Below the confluence of that tributary, the mainstem would be toxic to
invertebrates.
Wastewater Treatment Plant Failure
The WWTP failure scenario would turn the WWTP effluent from a diluent for tailings leachates to a toxic
input that would be diluted by tailings leachate. The effects of releasing untreated wastewater would, of
course, be greatest at the points of release (on the SK124 tributary of the South Fork Koktuli River below
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ITT April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
the tailings dam location and at the head of the North Fork Koktuli River above gage NK100C)
(Table 8-17). Under Pebble 6.5, the copper quotient at SK124A would increase from 1.3 (marginal
toxicity) with routine operation to greater than 100 (high toxicity) with the WWTP failure (Table 8-20)
resulting in levels sufficient to cause a fish kill (Table 8-23). Untreated wastewater input above gage
NK100C would increase the copper quotient from 0.76 to 55 (Table 8-20), resulting in early-life-stage
toxicity to trout. The effects would increase as mine size increases. The most severe effects on trout in
the SK124 tributary are estimated to be aversion (FA), early-life-stage toxicity (FR), and lethality to all
life stages (FK) for the Pebble 0.25, 2.0, and 6.5 mine scenarios, respectively. The most severe effects on
trout below the North Fork Koktuli outfall are estimated to be aversion for the Pebble 0.25 and Pebble
2.0 scenarios and early-life-stage toxicity for the Pebble 6.5 scenario. That implies a shift in severity
from a depleted invertebrate community, which would reduce fish production, and fish aversion to loss
offish reproduction and death (Table 8-23). Toxicities for total metals are slightly higher (Table 8-18).
Toxic effects are functions of both duration of exposure and concentration. Because concentrations
would be so high, toxic effects on salmonids for the Pebble 2.0 and Pebble 6.5 scenarios with WWTP
failure would be severe in the South Fork Koktuli River, even if the failure was of short duration.
However, in the North Fork Koktuli, for the Pebble 0.25 scenario, or downstream of the area analyzed,
the effects of WWTP failure would depend on the duration of exposure. The WWTP failure described in
this chapter could range from hours to months depending on the mechanics of the failure and whether
replacement of parts or components would be required. Alternatively, WWTP failure could be a result of
an inadequately designed water treatment system, which could result in the release of inadequately
treated water as at the Red Dog Mine, Alaska (Ott and Weber Scannell 1994, USEPA 1998, 2008). In that
case, the failure could continue for years until a new or upgraded treatment system is designed,
approved, and constructed.
8.2.3.7 Analogous Mines
Water quality degradation has been commonly associated with mining in the United States and
elsewhere. In particular, the phenomenon known as acid mine or acid rock drainage has severely
damaged many streams due to high acidity and dissolved metals and, as the effluent is neutralized, the
formation of aluminum, iron, and manganese oxide precipitates. Pre-Tertiary waste rock at the Pebble
deposit could produce such effluents (Table 8-8). Although published studies have emphasized the
severe effects of acidic waters, it should not be assumed that neutral or alkaline leachates, such as would
be expected from the Pebble deposit Tertiary rock (Table 8-6), would have no effects.
Water quality degradation at metal mines in the United States have been reviewed and summarized in
recent reports (Kuipers etal. 2006, Earthworks 2012). These reports document that such degradation
has not been uncommon at mines due to various factors, including inadequate pre-mining data, poor
predictions of mitigation needs, inadequate design, improper operation, and equipment failure.
Although past frequencies of water quality degradation are not predictive of future frequencies due to
changes in engineering practices, they do provide a reasonable upper bound. Kuipers et al. (2006)
reviewed the available record for 183 U.S. hard rock metal mines operating since 1975 and selected
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ITIApril 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
25 case study mines for which both quantitative predictions and information on operational water
quality were available. These case study mines represent a range of conditions, locations and
operations; although these mines are not a random sample, the sample is not apparently biased. Of the
case study mines, 15 had reported exceedances of surface water standards and 13 had reported
exceedances of groundwater standards. This is likely an underestimate of exceedance rates, since only
drinking water standards were used as the benchmark for comparison, and chronic water quality
criteria for aquatic life are much lower for many contaminants, particularly metals such as copper.
Earthworks (2012) reviewed the 14 porphyry copper mines operating in the United States and found
that all but one had reported failures to collect and treat seepage resulting in water quality degradation.
Unfortunately, biological or ecological monitoring has not been routinely conducted at operating mines,
so ecological consequences are not reported in either of these reviews.
Where biological monitoring has occurred, acid drainage has been shown to eliminate fish and
invertebrates from streams and, after dilution, to reduce abundance, production, and diversity of stream
and river ecosystems (Marchand 2002, Jennings et al. 2008). For example, acid drainage from an
abandoned copper mine in Britannia Creek, British Columbia, resulted in pH levels below 6 and spring
copper concentrations greater than 1,000 ug/L (Barry etal. 2000). The abundance of chum salmon fry
was lower in the creek than in reference areas, and 100% of Chinook salmon smolts died when placed in
cages in the creek. In addition, sustained discharges have resulted in the loss of habitat through
precipitation of metal hydroxides. This case illustrates the sensitivity of salmon to acid drainage from a
copper mine.
The Fraser River watershed in British Columbia has been recommended as an example of how salmon
can coexist with metal mining, and therefore suggested as a model for potential mining in the Bristol Bay
watershed. However, a long and dramatic decline in Fraser River sockeye salmon led to an official
investigation of causes, with inconclusive results (Box 8-4). In any case, it is clear that the Fraser River is
not a good analogue because, unlike potential Bristol Bay development, Fraser River mines are located
away from salmon spawning and rearing habitat and because the many other activities occurring in the
Fraser River watershed confound efforts to pinpoint specific causes of salmon population decline.
Further, the dramatic variability in Fraser River sockeye abundance is not an example that would
reassure Alaskans accustomed to the more productive and much more stable Bristol Bay sockeye
salmon fishery.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ co April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
BOX 8 4. THE FRASER RIVER
The Fraser River watershed, which supports sockeye and other salmon and contains multiple copper mines,
could serve as an analogue for proposed mining development in the Bristol Bay watershed. Mining
proponents have argued that the Fraser River fishery demonstrates that mining and fishing can co-exist
(Joling2011). However, the Fraser River is much less productive per unit of habitat than the Bristol Bay
rivers, the fishery has been closed in some recent years, and most of the salmon runs are listed as
threatened or endangered (Cohen 2010, O'Neal and Woody 2011).
The Cohen Commission for Inquiry into the Decline of Sockeye Salmon in the Fraser River commissioned
scientific projects to investigate potential causes of decline. The report on freshwater ecological factors
considered mining as one issue (Nelitz et al. 2011). The authors concluded that metal miningwas a minor
issue for sockeye habitat relative to other development in the watershed, because there are only five active
metal mines and only one (Endako) was near sockeye rearing habitat. Other developments that potentially
affect habitat include logging; pulp, paper, and other wood products manufacturing; coal, placer, and gravel
mining; urbanization; hydroelectricity; oil and gas drilling; agriculture; and water withdrawal. Although the
authors argued that acid and metal drainage from closed mines pose a risk to salmon, they did not analyze
that exposure. They concluded that, based on sedimentation of stream habitats, miningwas a plausible
contributor but not the major contributor to declines in sockeye salmon.
Another Cohen Commission report that addressed contaminants listed mine-related contaminants, but
could not specifically quantify the effects of mines (MacDonald et al. 2011). The authors concluded that
concentrations of six metals (including copper) and phenols were sufficient to reduce survival, growth, or
reproduction of sockeye salmon in the Fraser River. The final report concluded that contaminants could be a
secondary contributor, but data were insufficient (Cohen 2012).
In light of this information, Cohen Commission reports on the Fraser River do not provide evidence that
mining and salmon co-exist. The fishery declined from 1990 to 2007 and has fluctuated widely since.
Recent fluctuations have been associated with marine conditions, but available evidence is insufficient to
conclude whether harvesting, habitat degradation, or contaminants are significant contributors.
Neither the Cohen Commission nor USEPA's contractor, ICF International, was able to assess the effects of
metal mines in the Fraser River watershed, because compliance documents are not readily available and
monitoring data are insufficient. Some raw monitoring data show episodes of low pH and frequently elevated
dissolved copper in waters at the Gibraltar and Mount Polly Mines. Other effects have been associated with
closed mines. In particular, a tailings impoundment failed at the Pinchi Lake Mine in 2004, during
reclamation activities, releasing tailings and leachateto Pinchi Lake. This accident, along with prior releases,
resulted in the imposition of a fish consumption advisory related to mercury bioaccumulation.
In sum, the mines in the Fraser River watershed are not located in salmon habitat (Cohen 2012, Gustafson
2012), and other development activities in the watershed obscure any effects of mines at the watershed
scale. This diverse and relatively intensive development and the spatial discontinuity between mining and
salmon habitat make the Fraser River watershed a poor analogue for potential mine development in the
Bristol Bay watershed.
8.2.3.8 Summary
The risks to salmon, rainbow trout, Arctic grayling, and Dolly Varden can be summarized in terms of the
total stream kilometers that would experience different types of effects (Table 8-24). Based on toxicity
to rainbow trout, the endpoint salmonids are estimated to be at risk of mortality at all life stages in
12 km under the Pebble 6.5 scenario and 3.8 km under the Pebble 2.0 scenario, assuming routine
operations. The waters would be aversive for a much greater length. It is not clear how much the
resident fish might acclimate to the copper, but newly arriving salmon would not be acclimated and
would lose spawning habitat. Hence, salmon could lose 29 and 57 km of spawning habitat in the Pebble
2.0 and Pebble 6.5 scenarios, respectively, due to copper contamination, assuming that they are as
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ITT April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
sensitive as rainbow trout. Additional habitat would be lost in tributaries that would not be accessed
due to aversion.
The effects of a WWTP failure would depend on its timing and duration. If it occurred during the period
of salmon return, more than 100 km of habitat could be lost due to aversion alone. Mortality of all life
stages offish would occur in 31 km (Pebble 6.5) and 3.8 km (Pebble 2.0). Mortality or inhibited
development of early life stages offish would occur in 84km (Pebble 6.5) and 34km (Pebble 2.0).
Toxic effects from copper on aquatic invertebrates would occur in 15, 62, and 83 km of streams in the
Pebble 0.25, 2.0, and 6.5 scenarios, respectively, under routine operations (Table 8-24). These effects
are highly relevant to protecting salmon and other valued fish. Immature salmon rely on invertebrates
as food and all post-larval life stages of resident rainbow trout and Dolly Varden feed on invertebrates.
In streams, these invertebrates are primarily aquatic insects, but immature sockeye salmon in lakes are
dependent on zooplankton. Hence, protection offish requires protection of sensitive invertebrates.
These estimated effects are based on metal concentrations in fully mixed reaches. Locally, in mixing
zones below outfalls or in areas of upwelling of contaminated water, effects would be more severe.
Because available data do not quantify fish production in the potentially affected reaches, it is not
possible to estimate the lost production of salmon, trout, Arctic grayling, or Dolly Varden. However, the
semi-quantitative surveys performed for the EBD (PLP 2011) and summarized in Section 7.1 provide
some indication of the relative amounts offish potentially affected. The focal species, Chinook salmon,
coho salmon, Arctic grayling, and Dolly Varden, are those that rear for extended periods in the receiving
streams.
The South Fork Koktuli River, which would be the most severely affected stream, has the lowest
reported density of focal species that rear for extended periods in the receiving streams and for which
data are available (14,000 fish/km for Chinook and coho salmon, arctic grayling and Dolly Varden)
(Table 7-3), as well as chum and sockeye salmon. Because 40 to 50 km of the South Fork Koktuli River
would have copper levels sufficient to directly affect fish in the Pebble 6.5 scenario, more than a half
million individuals of the focal species would be exposed to copper levels sufficient to cause aversion,
sensory inhibition, inhibited development, or death. The lower length estimate is to approximately
account for dilution below gage SK100B. In the Pebble 2.0 scenario, copper levels in 22 km of the South
Fork Koktuli River would have direct effects on more than 300,000 individuals of the focal species.
Direct effects on fish are not expected in the Pebble 0.25 scenario.
The North Fork Koktuli River has a focal species density of 20,000 fish/km plus unenumerated rainbow
trout and chum and sockeye salmon. Since 2.4 km of the North Fork Koktuli River would have copper
levels sufficient to be toxic to invertebrates in the Pebble 2.0 and Pebble 6.5 scenarios, more than 47,000
individuals of the focal species would experience reduced food resources.
Upper Talarik Creek has the highest density of the focal species at 45,000 fish/km plus occurrence of
unenumerated rainbow trout and sockeye and chum salmon. The 6.5 km of the tributary receiving South
Fork Koktuli River interbasin transfers would be expected to have direct toxic effects on fish in both the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ „ April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
Pebble 2.0 and Pebble 6.5 scenarios. In the mainstem below the confluence of that tributary, less than
23 km of stream would experience effects on invertebrates.
For the WWTP failure in the Pebble 2.0 and Pebble 6.5 scenarios, as under routine operations, 40 to
50 km of the South Fork Koktuli River would have copper levels sufficient to directly affect more than a
half million of the focal fish. However, the effects would be more severe than under routine operations,
including acute lethality to all life stages in most of the reaches. For the Pebble 0.25 scenario, 15 km
would experience aversive effects on fish and, in 40 to 50 km, toxicity to invertebrates would result in
reduced food resources for more than a half million of the focal fish species.
Due to the uncertainties in the density data and the compounding uncertainties in exposure and toxicity,
these effects estimates are rough. However, it appears that the number offish experiencing death or an
equivalent effect such as loss of habitat would be between 10,000 and one million across scenarios.
For the WWTP failure in the Pebble 0.25, 2.0, and 6.5 scenarios, 45,100, and 100 km of stream,
respectively, would have copper concentrations sufficient to directly affect fish (Table 8-24). Toxicity
would result in reduced survival or inhibited development in early life stages of salmonids in 84 km
under the Pebble 6.5 scenario for more than a half million fish, depending on the season. Sensory
inhibition or aversion would affect 600,000 to 1.4 million focal fish species in the three mine scenarios.
Table 8 24. Length of stream (km) in which copper concentrations would exceed levels sufficient to
cause toxic effects, under routine operations and wastewater treatment plant failure, for each of
the three mine scenarios.
Toxic Effect
Invertebrate chronic
Invertebrate acute
Fish avoidance
Fish sensory
Fish reproduction
Fish kill
Length of Stream Potentially Affected (km)
Pebble 0.25
Routine
Operations
15
1.9
-
-
-
WWTP
Failure
100
87
45
-
-
Pebble 2.0
Routine
Operations
62
39
29
3.8
3.8
3.8
WWTP
Failure
110
110
100
34
34
3.8
Pebble 6.5
Routine
Operations
83
82
57
12
12
12
WWTP Failure
130
130
100
92
84
31
Notes:
Blank values (-) indicate that no stream lengths are likely to be affected.
Effects are defined in Section 8.2.3.4.
WWTP = wastewater treatment plant.
8.2.4 Additional Mitigation of Leachates
The high metal concentrations in the South Fork Koktuli River due to PAG waste rock leachate suggests
that mitigation measures should be considered beyond those described in this scenario or the Northern
Dynasty mining case (Ghaffari et al. 2011). That design may be sufficient for a typical porphyry copper
mine, equivalent to Pebble 0.25, butnotthe massive Pebble 2.0 and 6.5 mine sizes. To avoid exceeding
the copper criteria, a leachate barrier or collection system for the Pebble 6.5 scenario would require
more than 99% effectiveness. Wells, trenches, or walls are not likely to achieve that. Lining the PAG
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-56
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
waste rock piles might be effective but liners have some leakage due to imperfect installation, punctures,
and deterioration. An alternative mitigation measure would be to ensure that all PAG waste rock is
stored within the drawdown zone for the mine pit. In that way, most acidic and high-metal leachate
would be collected and treated before discharge. If the PAG waste rock was processed before or at
closure, the risk of an acidic pit lake would be minimized (Section 8.1.3). Moving all PAG waste rock near
the pit would mean an increase in NAG waste rock leachate leakage to streams as NAG waste rock is
moved out of the drawdown zone. If all of the leakage of waste rock leachate for the maximum mine
were NAG and if mining did not affect the background copper concentration, the copper concentration
would be approximately 1.5 ug/L. That would be a great improvement but would equal the chronic
criterion for the stream and affect sensitive invertebrates. Hence, it would also be necessary to improve
the 50% efficiency of leachate capture. The magnitude and extent of these predicted effects suggest the
need for additional mitigation measures beyond the conventional practices assumed in the scenario to
reduce the input of copper and other metals. Simply improving the efficiency of the capture wells,
making the cutoff walls more extensive, or adding a trench is unlikely to achieve water quality criteria
under those scenarios. Additional measures might include lining the waste rock piles, reconfiguring the
piles or processing more of the waste rock as it is produced.
8.2.5 Uncertainties
Although it is certain that operating a mine would adversely affect water quality at the mine site, risks to
fish from water collection, treatment, and release would be highly uncertain. This uncertainty is
demonstrated by the record of inaccurate water quality predictions contained in environmental impact
statements for major hard rock metal mines in the United States (Kuipers et al. 2006). As described
above, that review found that most of the hard rock mines considered had violated water quality
standards. These cases represent prediction failures, because mine permits included mitigation
measures to prevent such exceedances. The primary causes of these prediction failures were described
as inadequate geochemical and hydrological characterization and optimistic characterization of
mitigation.
These results cannot be considered to be quantitatively predictive of the likelihood of water quality
modeling failures in this or future assessments; however, they do indicate that predicting the effects of
mining on water quality is difficult, and results are uncertain. Further, the effects of water quality
changes on aquatic communities are uncertain. The following factors contribute to these uncertainties.
• The range of potential failures is wide and the probability of occurrence for any of them cannot be
estimated from available data. Therefore, we can only state that, based on the record of the mining
industry, treatment failures of some sort are likely to occur.
• The waste rock leachate concentrations used in the assessment are from humidity cell tests. Because
these tests involve repeated flushing of rock under oxic conditions, they may reasonably represent
waste rock piles or pit walls leached intermittently by precipitation and snowmelt. However,
laboratory tests of relatively small samples are imperfect models of large rock piles in the field. This
uncertainty may be minimally estimated by comparing the humidity cell tests with barrel tests
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ „ April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
conducted in the field with more realistic rock sizes and test conditions (PLP 2011: Section 11.7.1).
These tests give qualitatively similar results, but the initial flush of high leachate concentrations in
the barrel tests seems to the PLP authors to be persisting past the date of the EBD data compilation.
Therefore, the concentrations reported for humidity cells are used, because they are likely to be
closer to long-term values, and the magnitude of uncertainty cannot be estimated.
• The tailings leachate concentrations are also from laboratory tests. Such tests of relatively small
samples are imperfect models of the processes in tailings slurries, TSF surface water, near surface-
deposited tailings, deeply buried tailings, and leakage into groundwater in the field. It is not clear
whether these tests tend to over or underestimate leachates in the field or how large the
discrepancy might be.
• The tailings test data do not include pyritic tailings, which are strongly acid-generating. This would
tend to underestimate the metal content of tailings leachate, but the effects on leachates from a TSF
are likely to be small due to the relatively small proportion of pyritic tailings.
• The surface-water and groundwater hydrology of the potential mine site is complex and the
hydrological models used to estimate exposures are inevitably simplifications. This is one of the
greatest sources of uncertainty for the water quality risks. More information is needed concerning
the movement of water from precipitation to groundwater and surface water, including seasonality
and storm and melt events.
• The water quality models assume that mining would not affect background water quality. That is
unlikely, but the change could not be estimated. This assumption is expected to result in
overestimation of copper levels, particularly in the South Fork Koktuli River. However, as mining
reduces background levels, it would be increasing levels from leachate input even more. It could
change the expected effects at the margins of toxicity but would not significantly affect the
conclusions.
• The use of average receiving water flows neglects the potential consequences of low dilution during
low-flow periods. This would be difficult to model, because low stream flows would be associated
with low leachate formation and low groundwater levels. This consideration suggests that the low
dilution and low leaching rates might balance to some extent, but the degree of balance is unknown.
• Chemical criteria and other single chemical benchmarks do not address interactions or combined
effects of individual constituents. The additivity model used here is a reasonable default, but the lack
of test data for the actual mixture adds uncertainty. This is a concern of some reviewers but is
judged to be a relatively small contributor to uncertainty. Strong interactions tend to occur when all
constituents are at or near toxic levels and only cadmium and zinc reach toxic levels and only in the
WWTP failure scenario. Given the overwhelming dominance of copper toxicity, this uncertainty
appears to be relatively minor.
• Studies of streams receiving mine effluents and laboratory studies suggest that the abundance of
important insect taxa could be reduced even if criteria are met. The implications of this uncertainty
are explained at the end of Section 8.2.2.1.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ gg April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
• Criteria for chemicals other than copper do not address site water chemistry, or they address it in a
simple way (e.g., via hardness normalization). Hence, they may be inaccurate estimates of threshold
concentrations for toxic effects in these highly pure waters. The example of copper suggests that the
criteria and screening benchmarks could be too high by a factor of 2 (see discussion below).
• Some leachate and process water constituents have water quality criteria, standards or benchmarks
for aquatic life that are based on old or sparse literature. Additional data are likely to reveal more
sensitive species or responses. This would result in lower benchmarks and criteria and higher risks
for the poorly studied chemicals. However, this is unlikely to affect conclusions, because the well-
studied metal copper so dominates the toxicity.
• If the State of Alaska uses its standards in effluent permitting rather than the national criteria, and if
the toxicity of chemicals with no state standards is not considered in the permit, the toxicity of the
effluents would be significantly higher than estimated in this assessment. This would result in an
underestimate of effects by more than a factor of 2, primarily due to using the hardness adjusted
copper standard.
• The concentrations of xanthate and other ore-processing chemicals in ambient waters are roughly
estimated to be below toxic levels, but studies in the laboratory or at mine sites are insufficient to
determine whether that would actually be the case. If xanthate does not degrade rapidly in the
tailings, the estimate that it would not leach into streams at toxic concentrations could be incorrect.
• If the tested rock and tailings samples are not representative, other waste water constituents may be
of concern. Some waste rocks or tailings may have high levels of elements other than those
identified in the screening analysis for mean concentrations. For example, selenium concentrations
are not high on average but are well above criteria in some individual leachate samples. This
uncertainty might be estimated from statistical analyses of sampling results and modeling of waste
rock piles with variance in concentrations among locations, but that is beyond the scope of this
assessment.
• The separation of PAG and NAG into separate waste rock piles will inevitably be imperfect. Ghaffari
et al. (2011) estimates that 5% of rock in the NAG piles would be PAG. The humidity cell tests were
not reported to include any PAG, so humidity cell tests on NAG mixed with 5% PAG would be
expected to have higher leachate concentrations. This causes underestimation of risk, as discussed
below.
• Although the state has a standard for TDS from any source, the toxicity of mixtures of major ions
depends on the constituents. Most studies of TDS are based on sodium chloride, which is less toxic
than mining leachates that have been studied. Hence, the degree of protection provided by the
standard is uncertain. The toxicity of different salt mixtures may vary by at least a factor of 3.
However, estimated TDS levels are not high enough for this to be a major uncertainty.
One means to estimate the magnitude of uncertainty in an assessment is to compare it to an
independently produced assessment. An assessment of hydrologic and water quality issues at the Pebble
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ „ April 2013
-------
Chapter 8
Water Collection, Treatment, and Discharge
deposit was independently performed by Wobus etal. (2012). Wobus etal. (2012) used the same set of
available data (primarily the EBD [PLP 2011]) as this assessment and based their modeling on the same
mining plan (Ghaffari et al. 2011). However, those authors made somewhat different assumptions in
model implementation that gave different results. In particular, their estimates of copper concentration
in waste rock leachates are higher (Table 8-25) than the estimates used in this assessment. Wobus et al.
(2012) added one standard deviation to mean concentrations of waste rock humidity cell tests to
represent actual leachate exposures, because high values and potential episodic exposures would be
underrepresented by the mean. In addition, they assumed, based on a statement in Ghaffari etal. (2011),
that separation of NAG and PAG waste rock would be imperfect and NAG rock piles would contain 5%
PAG rock. In this assessment, we used the mean leachate concentration and assumed complete
separation of NAG and PAG rock. Hence, if concentrations reflect initial weathering rates, concentration
peaks with snowmelt and storm events, or the high sulfur and copper samples, and if 5% of material in
the NAG rock pile would be PAG rock, estimates of leachate copper concentrations in this assessment
could be too low by factors of 2.8 (PAG leachate) and 66 (NAG leachate) (Table 8-25).
Table 8 25. Copper concentrations (mg/L) in waste rock leachates for two water quality models.
Leachate
Non-acid-generating (NAG)
Potentially acid-generating (PAG)
Water Quality Model
This Assessment
0.0032
1.5
Wobus etal. 2012
0.21
4.2
These differences in estimated leachate concentration between this assessment and Wobus et al. (2012)
do not translate into proportionate differences in ambient exposure. We compared the Wobus et al.
(2012) results for the Ghaffari etal. (2011) 25-year scenario to our results for the equivalent Pebble 2.0
scenario. Due to differences in the models used and in hydrologic assumptions, estimates may be higher
or lower in this assessment than in Wobus et al. (2012). For example, our estimates of copper at gages
SK100G and SK100C are 4.4 and 2.3 times those from Wobus et al. (2012), respectively. However, the
Wobus et al. (2012) estimates of copper concentrations at gages UT100D and UT100B are 79 and 7.5
times as high, respectively, as ours, apparently due to the 5% PAG rock in the NAG waste rock pile. This
difference is not due to differences in the assumed mining plan. Rather it is primarily a difference in
assumptions concerning how well the plan to segregate NAG and PAG can be carried out. These
differences illustrate how interpretation of data can be an important source of uncertainty in
environmental modeling.
Another quantification of uncertainty is provided by comparing the benchmarks for copper toxicity that
might be used as thresholds for minimum risk (Section 8.2.2). The national ambient water quality
criteria for copper are based on the BLM, which better accounts for the influence of water quality on
bioavailability than the hardness-derived state standard. These two benchmarks differ by a factor of 1.7
for acute values and 2.1 for chronic values. Four other metals of concern (cadmium, lead, nickel, and
zinc) have hardness-dependent standards but no BLM-based criteria. If they also are too high by a factor
of approximately 2, then those metals and total metal toxicity are more of a concern than suggested by
the screening assessment The same applies to relevant conventional thresholds for acute and chronic
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
8-60
April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
toxicity in salmonids (the LCso and chronic value for rainbow trout), which are BLM-corrected for
copper (Table 8-13) but not for other metals. However, the threshold for avoidance of copper exposures
(IC2o) is 12 to 28 times lower than the LCso and 4.2 to 5.4 times lower than the chronic value. Hence, the
concentration at which a stream would no longer be suitable for trout is considerably underestimated
by conventional endpoints. The effects thresholds for less well-studied metals are likely to be
equivalently underestimated. Even copper toxicity is likely to be underestimated, due to the absence of
tests for sensitive insect species—much less tests of the most sensitive responses of those species
(Section 8.2.2.1).
8.3 Temperature
Changes in water temperature associated with mine development activities are a concern given the
importance of suitable water temperatures for Pacific salmon. This section begins with a description of
current thermal regimes in the mine scenario watersheds and potential alterations due to WWTP
discharges under routine operations (Section 8.3.1). It then describes exposure-response relationships
for temperature (Section 8.3.2). It ends with a characterization of potential risks associated with the
thermal regime of water treatment effluents (Section 8.3.3) and a discussion of uncertainties
(Section 8.3.4).
8.3.1 Exposure
8.3.1.1 Thermal Regimes in the Mine Scenario Watersheds
Water temperature data collected by PLP and published in the EBD (PLP 2011: Appendix 15.IE,
Attachment 1) indicate significant spatial variability in thermal regimes. Average monthly stream
temperatures in the Pebble deposit area in July or August can range from 6°C to 16°C. Extensive glacially
reworked deposits with high hydraulic conductivity allow for extensive connectivity between
groundwater and surface waters in the region (Power et al. 1999). This groundwater-surface water
connectivity has a strong influence on the hydrologic and thermal regimes of streams in the Nushagak
and Kvichak River watersheds, providing a moderating influence against both summer heat and winter
cold extremes in stream reaches where this influence is sufficiently strong. The range of spatial and
temporal variability in temperatures provided in the EBD is consistent with streams influenced by a
variety of thermal modifiers including upstream lakes, groundwater, or tributary contributions (Mellina
et al. 2002, Armstrong et al. 2010). Longitudinal profiles of temperature in the EBD indicate that
summertime stream temperatures in the Pebble deposit area do not uniformly increase with decreasing
elevation. This is often due to substantial inputs of cooler water from tributaries or groundwater inputs
(PLP 2011). An example of combined tributary and groundwater inflow contributing to significant
cooling in summer mainstem temperatures is the South Fork Koktuli River downstream of gage SK100C
(Figure 7-15). This is the section of the South Fork Koktuli River fed by the tributary gaged by SK119A,
on which a WWTP outfall would be located and to which a portion of the WWTP flows would be
directed. As reported in EBD Appendix 15.IE (PLP 2011), combined groundwater and tributary
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TT7 April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
contributions between gages SK100C and SK100B1, including contributions from the tributary gaged by
SK119A, contributed to a cooling of 5.4oC, with a gain in flow of 1.36 m3/s on August 24, 2007. Other
examples of spatial variability in summer temperatures are detailed in the EBD (PLP 2011:
Appendix 15.IE, Attachment 1).
Winter water temperatures are also spatially variable, as indicated by instream temperature monitoring
data (PLP 2011). The same reach of the South ForkKoktuli River that was cooled in August by
groundwater and tributary inflows experienced warming in October. Contributions of relatively warmer
groundwater were observed to maintain ice-free conditions in some areas, as revealed by patchiness in
ice cover seen in aerial surveys (PLP 2011, Woody and Higman 2011).
8.3.1.2 Thermal Regime Alterations
Mine development and operation would result in alteration of surface-water and groundwater flows and
water collection and treatment, all of which would affect water temperatures. Some streams would
experience increased streamflows, whereas others would experience significant reductions (Table 7-
19). Increased streamflows due to additions of effluent from the WWTP would alter temperatures
significantly, depending on the temperature and quantity of the effluent. Changes in the source of water
supplying streams would also influence thermal responses. For example, reductions in the proportion of
thermally moderated groundwater inputs would result in surface-water temperatures that would be
warmer in summer and colder in winter. Conversely, active thermal management (i.e., heating or cooling
of effluent) and timed releases from the WWTP could be used to attempt to compensate for mine-related
thermal modifications. However, the plan for a Pebble mine outlined in Ghaffari et al. (2011) does not
include temperature control by the planned WWTP. The mine scenarios include temperature control to
meet state standards, but not to match natural water temperature regimes.
Treated water would be released to tributaries of the South and North Fork Koktuli Rivers and would
influence flows and water temperatures in downstream reaches. Thermal effects of WWTP effluent
would be greatest in the receiving tributaries. Effects would moderate with distance from the WWTP
outfall, due to mixing with surface-water and groundwater inputs and heat exchange. Due to the
substantial increases in discharge over baseline levels associated with the WWTP (12 to 112% increases
in monthly mean flow depending on mine scenario and location; Table 7-19), the thermal loads
attributable to WWTP discharges would potentially influence temperatures downstream in the South
and North Fork Koktuli Rivers. For example, WWTP discharge is expected to comprise 12 to 60% of
mean annual flow in the North Fork Koktuli River at gage NK100C (calculated from Tables 8-1 through
8-3). Sensitivities of downstream reaches to WWTP outfall temperatures were not evaluated in this
assessment due to uncertainties in the timing and temperature of WWTP discharges and heat exchange
processes in downstream reaches. Managing treated water temperatures to maintain baseline thermal
regimes would be most protective offish populations adapted to local thermal regimes, but would
require temperature and hydrologic modeling informed by baseline monitoring, and the ability to
control temperatures and quantities of discharged flows to meet temperature targets. Baseline data
collected by PLP contractors (PLP 2011) for the purposes of developing and applying surface-water
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTIApril 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
temperature models would be useful for managing flows and temperatures to minimize impacts on
aquatic life.
8.3.2 Exposure-Response
Water temperature controls the metabolism and behavior of salmon, and, if temperatures are stressful,
fish can be more vulnerable to disease, competition, predation, or death (McCullough etal. 2009).
Recognizing the importance of water temperature to healthy salmon populations, the State of Alaska
requires that maximum water temperatures not exceed 20°C at any time, with specific maximum
temperatures for migration routes and rearing areas (15°C) and spawning areas and egg and fry
incubation (13°C). For all other waters, the weekly average temperature may not exceed site-specific
requirements needed to preserve normal species diversity or to prevent the appearance of nuisance
organisms (ADEC 2012).
This standard is designed to protect against increases in summer temperature, a serious concern for
salmon populations particularly in light of projected climate change effects on streamflow and
temperatures (Bryant 2009). Elevated summer temperatures are a management concern due to
potential adverse effects including increased risk of direct mortality, disease, elevated metabolic costs,
and altered community interactions. Sockeye salmon are particularly sensitive to high temperature
during spawning, being limited to temperatures between 2°C and 7°C (Weber Scannell 1991). Summer,
however, is not the only period during which salmon are sensitive to temperatures (Poole et al. 2004).
Salmon and other native fishes in the mine scenario watersheds rely on suitable temperature regimes to
successfully complete their life cycles (Quinn 2005). The period of salmon egg incubation in gravels can
be particularly sensitive to temperature changes, and changes of just a few degrees Celsius in winter
mean temperature can change emergence timing of young salmon by months (Brannon 1987, Beacham
and Murray 1990, Quinn 2005). For locally adapted populations, timing of key life-history events (i.e.,
spawning, incubation, and out-migration) can be closely tied to the timing of other ecosystem functions
that provide critical resources for salmon (Brannon 1987, Quinn and Adams 1996). Thus, changes to
thermal and hydrologic regimes that disrupt life-history timing cues can result in mismatches between
fish and their environments or food resources, adversely affecting survival (Jensen and Johnsen 1999,
Angilletta etal. 2008).
8.3.3 Risk Characterization
Stream temperatures in the mine scenario watersheds could be substantially altered due to changes in
streamflow, sources of streamflow (e.g., relative importance of groundwater versus WWTP
contributions), or other changes to the heat balance of WWTP discharges. We expect treated water
returned to streams would have different thermal characteristics than water derived from groundwater
sources (the dominant water source prior to mining). The extent and duration of temperature effects
would depend not only on source water temperatures, but also on the quantity and timing of water
contributed from various additional sources, such as tributaries and groundwater inputs. Simple mixing
models can be used to estimate stream temperatures below the confluence of multiple sources with
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment T7I April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
known temperature and discharge. However, we do not use such models here, because we cannot
account for all sources of heat transfer. In the absence of models, we have relied on available literature
to identify the most likely risks to fish associated with deviations from current thermal regimes in the
Pebble deposit area.
Interception of groundwater that is collected then released as a point-source through a WWTP would
alter the ways in which groundwater feeds stream channels through dispersed and complex pathways.
Groundwater-surface water interaction in streams can create thermal heterogeneity, enhancing the
diversity of habitats available to fish (Power et al. 1999). Migration, spawning, and incubation timing are
closely tied to seasonal water temperatures. Diversity of thermal habitats can allow a diversity of
spawning migration timing to persist (Hodgson and Quinn 2002). For the Bristol Bay region, this
asynchrony in spawning timing helps buffer Bristol Bay salmon populations from climatic events or
other environmental changes that may adversely affect a particular run (Schindler et al. 2010). An
additional benefit of staggered spawner return timing is the extended availability of spawning sockeye
salmon to mobile consumers like brown bear (Schindler et al. 2010). Depending on the degree to which
adaptation and compensatory strategies may mitigate thermal effects on life-history development and
spawning timing, deviations from the thermal regime to which local populations of salmon may be
adapted could have serious population-level consequences (Angilletta etal. 2008).
The volume of water that would require treatment ranges from 10 to 49 million m3/year across the
three mine scenarios (Tables 8-1 through 8-3). To avoid or minimize risks associated with altered
thermal regimes in downstream effluent-receiving areas, capacity for thermal control of effluent would
be required to maintain natural thermal regimes or temperatures required by regulatory agencies.
Water temperature modeling is being used by PLP to assess thermal characteristics of streams in the
Pebble deposit area (PLP 2011: Chapter 15, Appendix 15.IE) and could provide additional guidance for
establishing a temperature management plan for the WWTP.
8.3.4 Uncertainties
The temperature of waters discharged from the mine, whether directly from the WWTP or indirectly
through groundwater or surface-water runoff, would be influenced by a number of factors controlling
heat exchange that cannot be known with confidence at this point. Likewise, the influence of these
discharges on stream temperatures downstream of the mine site is unknown. Because exchange with
groundwater is so important to surface-water properties in the mine area, simple models that assume
primarily surface-water heat exchange would be incomplete and inaccurate.
Projecting changes to temperature due to changes in groundwater-surface water interactions in the
mine area was not attempted for this assessment. Local geology and stream hydrographs are indicative
of systems that are largely driven by groundwater. Disruptions or changes to groundwater flowpaths
and mechanisms of thermal exchange in the mine area could have significant adverse effects on winter
habitat suitability for fish, particularly if groundwater-dominated stream reaches are converted to
stream reaches dominated by effluent from a WWTP with a novel thermal regime. Given the high
likelihood of complex groundwater-surface water connectivity in the mine area, predicting and
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTT April 2013
-------
Chapter 8 Water Collection, Treatment, and Discharge
regulating temperatures to maintain key ecosystem functions associated with groundwater-surface
water exchange would be particularly challenging.
Maintenance of mine discharges, in terms of water quality, quantity, and timing, to avoid adverse
impacts would require long-term commitments for monitoring and facility maintenance. As with other
long-term maintenance and monitoring programs, the financial and technological requirements could be
large, and the cumulative risks (and likely instantaneous consequences) of facility accidents, failures,
and human error would increase with time. Additionally, climate change and the predicted increases in
water surplus for the region (Chapter 3) will result in potential changes in streamflow magnitude and
seasonality, requiring adaptation to potentially new water management regimes for the water
processing facilities. We know of no precedent for the long-term management of water temperature on
this scale at a mine.
Finally, responses of the endpoint fish species to alterations in thermal regimes are not well understood.
The existing information consists largely of field studies of salmonid distributions with respect to
temperatures supplemented by a few laboratory studies of development, growth, and survival at
controlled temperatures. Monitoring studies of responses to temperature alterations of various
magnitudes and durations are desirable.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ April 2013
-------
9.1 Overview
In this chapter, we describe risks to stream habitats and salmonid populations from potential failures of
tailings storage facility (TSF) dams. Specifically, we consider tailings dam failures at TSF 1 and potential
consequences from physical and toxicological effects to fish and fish habitat (Figure 9-1). Similar types
of effects would occur following tailings dam failures at TSF 2 or TSF 3, or at TSF locations in other parts
of the Bristol Bay watersheds, although the specifics of a failure at these locations would differ.
A breach of a TSF 1 dam would result in a flood wave and subsequent tailings deposition that would
greatly alter the downstream channel and floodplain (Figure 9-1). The initial flood wave for the tailings
dam failure scenarios modeled here would far exceed the typical flood event currently experienced in
these watersheds. The flood itself would have the capacity to scour the channel and floodplain and alter
the landscape. In addition to the hydraulic flow, the quantity of tailings that could discharge from the
TSF could bury the existing channel and floodplain system with meters of fine-grained tailings material.
The existing channel and floodplain would be eliminated and a new channel form would develop in the
resulting topography. Given the size of these new fine-grain deposits, sediment would be highly mobile
under typical flow events and channel form would remain unstable. Sediment deposited on floodplains
and remaining behind the breached dam would create a concentrated source of highly mobile material
that does not currently exist in the study watersheds. Although a sediment transport study would be
required to quantify the temporal and spatial extent of effects, it is likely that the sediment regime of the
affected stream and downstream waters would be greatly altered and the existing and well-defined
gravel bed stream would be transformed to an unstable, silt- and sand-dominated channel.
Remediation is possible following a tailings spill, but the occurrence and effectiveness of these measures
would be uncertain. A tailings spill would flow into a roadless area with streams and rivers that are too
small to float a dredge, so the proper course of remediation is not obvious. The remediation process
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-1
April 2013
-------
Chapter 9 Tailings Dam Failure
could be delayed by planning, litigation, and negotiation, particularly concerning the proper removal
and disposal of excavated tailings. If the operator was no longer present at the site or was no longer in
existence, the response would, at best, be delayed further. Once started, the building of a road and
support facilities and the excavation, hauling, and disposal of tailings could take years, particularly given
the long winter season. Therefore, the extent to which tailings exposure would be diminished by
remediation cannot be estimated. Given this uncertainty, the assessment assumes that significant
amounts of tailings would remain in the receiving watersheds for some time, and that remediation may
not occur at all.
9.1.1 Causes of Tailings Dam Failures
A tailings dam failure occurs when a tailings dam loses its structural integrity and releases tailings
material from the impoundment. The released tailings flow under the force of gravity as a fast-moving
flood containing a dense mixture of solids and liquids, often with catastrophic results. This flood can
contain several million cubic meters of material traveling at speeds in excess of 60 km/hour. At dam
heights ranging from 5 to 50 m—substantially less than the 90-m and 209-m tailings dam failures
considered here—the flood wave can travel many kilometers over land and more than 100 km along
waterways (Rico etal. 2008). There are many international examples of such failures (Box 9-1), and we
note that dams in these failure examples were significantly smaller than the dams included in our mine
scenarios.
Causes of tailings dam failure are similar to those for earthfill and rockfill water retention dams, and
include the following circumstances.
• Overtopping. Overtopping occurs when insufficient freeboard is maintained and the water level
behind the dam rises due to heavy rainfall, rapid snowmelt, flooding, or operator error.
• Slope instability. These failures occur when shear stresses in the dam exceed the shear resistance of
the dam material, most frequently resulting in a rotational or sliding failure of a portion of the
downstream slope, leading to overtopping or breaching of the dam.
• Earthquake. Shaking resulting from earthquakes (Box 9-2; Section 3.6) causes additional shear
forces on the dam that can lead to a slope instability failure.
• Foundation failure. Weak soil or rock layers and high pore pressures below the base of the dam can
lead to shear failures in the foundation, causing entire dams to slide forward or rotate out of
position.
• Seepage. Seepage through an earthfill embankment increases interstitial pore pressures and
reduces the intergranular effective stresses and shear resistance, potentially leading to a slope
instability failure. Seepage can also cause internal erosion and piping within a dam leading to a
hydraulic failure.
• Structural failure. Tailings dams often contain structural components such as drainage systems or
spillways that, if they fail to function properly, can cause overtopping or slope instability failure.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment [TI April 2013
-------
Chapter 9 Tailings Dam Failure
• Erosion. Erosion, especially along the toe of a dam, can reduce slope stability to the point of failure.
Erosion near the crest can reduce freeboard (the distance between the top of the dam and the
impounded water level) and increase the chance of overtopping.
• Subsidence. If a tailings dam is built on compressible soils or overlies cavities such as underground
mining works, subsidence can cause displacement or cracking of the dam. Cracking can lead to a
direct hydraulic breach or to slope instability. Settlement can reduce freeboard and increase the
chance of overtopping.
BOX 9 1. EXAMPLES OF HISTORICAL TAILINGS DAM FAILURES
The tailings dam failures below illustrate the characteristics and potential consequences of a tailings dam
failure. The details of the design, construction, or operation of any tailings dams constructed for mines in the
Bristol Bay watershed would not be the same as these mine tailings dams, but these examples demonstrate
that tailings dam failures can occur, and illustrate how these failures may affect downstream areas. In
addition, the dams in these failure examples were significantly smaller than the dams in our mine scenarios.
Aznalcollar Tailings Dam, Los Frailes Mine, Seville, Spain, 1998. A foundation failure resulted in a 45-m-
long breach in the 27-m-high, 600-m-long tailings dam, releasing up to 6.8 million m3 of acidic tailings that
traveled 40 km and covered 2.6 million ha of farmland (ICOLD 2001).
Stava, Italy, 1985. Two tailings impoundments were built, one upslope from the other, in the mountains of
northern Italy. The upslope dam had a height of 29 m; the downslope dam had a height of 26 m. A stability
failure of the upper dam released tailings, which then caused the lower dam to fail. The 190,000 m3 of
tailings, traveling at up to 60 km/hour, reached the village of Tesero (4 km downslope from the point of
release) in 5 or 6 minutes. The failure killed 269 people (ICOLD 2001).
Aurul S.A. Mine, Baia Mare, Romania, 2000. A5-km-long, 7-m-high embankment on flat land enclosed a
tailings impoundment containing a slurry with high concentrations of cyanide and heavy metals. Heavy rains
and a sudden thaw caused overtopping of the embankment, cut a 20- to 25-m breach, and released
100,000 m3 of contaminated water into the Somes and Tisza Rivers. Flow continued into the Danube River
and eventually reached the Black Sea. The contamination caused an extensive fish kill and the destruction
of aquatic species over 1,900 km of the river system (ICOLD 2001).
Tennessee Valley Authority Kingston Fossil Plant, Roane County, Tennessee, USA, 2008. After receiving
nearly 20 cm of rain in less than 4 weeks, an engineered 18-m-high earthen embankment of a 34-ha
storage impoundment failed, producing a 14-m-high surge wave and releasing 4.1 million m3 of coal fly ash
slurry. The release covered over 121 ha with slurry containing arsenic, cobalt, iron, and thallium. Over 2.7
million m3 of coal ash and sediment were dredged from the Emory River to prevent further downstream
contamination (AECOM 2009).
A number of studies have attempted to analyze the historical record to determine proximate causes and
probabilities of tailings dam failures (Davies et al. 2000, ICOLD 2001, Davies 2002, Rico et al. 2008,
Chambers and Higman 2011). These efforts have been hindered by the lack of a worldwide inventory of
tailings dams, incomplete reporting of tailings dam failures, and incomplete data for known failures. The
National Inventory of Dams (2005) lists 1,448 tailings dams in the United States, and the worldwide
total is estimated at over 3,500 (Davies et al. 2000). The International Commission on Large Dams
compiled a database of 221 tailings dam incidents and failures that occurred from 1917 through 2000
(ICOLD 2001). Causes of incidents and failures were reported for 220 of these, comprising 85 incidents
and 135 failures; Table 9-1 summarizes causes of the 135 reported failures (ICOLD 2001).
Perhaps most noteworthy is the relatively high number of accidents or failures for active compared to
inactive tailings dams, primarily resulting from slope instability and failure (Table 9-1). This suggests
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment [TI April 2013
-------
Chapter 9
Tailings Dam Failure
that the stability of tailings dams and impoundments may increase with time, as dewatering and
consolidation of the tailings occurs and additional loads are no longer applied.
However, failures do occur after operation. In December 2012, the tailings dam at the closed Gullbridge
Mine, Newfoundland, failed leaving a gap 50 m wide and the height of the dam (Fitzpatrick 2012). The
mine opened in 1967, rehabilitation of the site occurred in 1999, and an inspection in 2010 found that
the dam was deteriorating (Stantec Consulting 2011). The primary cause of failure for inactive tailings
dams is overtopping, accounting for 80% of the recorded failures for which the cause is known
(Table 9-1).
Table 9 1. Number and causes of tailings dam failures at active and inactive tailings dams.
Failure
Failure cause
Overtopping
Slope instability
Earthquake
Foundation
Seepage
Structural
Erosion
Mine subsidence
Unknown
TOTALS
Number of Tailings Dam Failures3
Active Dams
20
30
18
11
10
12
3
3
15
122
Inactive Dams
8
1
0
1
0
0
0
0
3
13
Total
28
31
18
12
10
12
3
3
18
135
Notes:
a Data are presented for 135 tailings dam accidents and failures for which causes were reported, from 1917 to 2000.
Source: ICOLD 2001.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-4
April 2013
-------
Chapter 9
Tailings Dam Failure
Figure 9 1. Conceptual model illustrating potential pathways linking tailings storage facility failure and effects on fish endpoints.
[ tailings storage
L facilities
T dissolved metals
adsorbed or precipitated metals
A other water
quality parameters
A metal spe elation
& b io aval I ability
4- macroinvertebrate
4 rearing habitat
(quality or quantity!
4 spanning habitat
(quality or quantity!
4- ov e r '/•.• i nte ring h al:• itat
(quality or quantity!
T taioaccutnulation &
biomagnification
1 tissue
concentration
"f chronic
toxicity
LEGEND
f source J
additional step in
caLisal pathw ay
proximate /-
stressor (^
additional step in
causal pathway
response^
mollifying
factor
V
"f acute
toxicity
4- salmon
(abundance, productivity or diversity!
4 other fish
(abundance, productivity or diversity)
V
4 marine-derived
nutrients
4- ecosystem
productivity
V
A
operations
volcanic activity
hvdrologic event
la r> dstidf/o vo la i? ch e
engineering failure
human error
response time
seasonal timing of
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-5
April 2013
-------
Chapter 9
Tailings Dam Failure
This page intentionally left blank
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p g April 2013
-------
Chapter 9 Tailings Dam Failure
9.1.2 Probability of Tailings Dam Failures
It is difficult to estimate the probability of low-frequency events such as tailings dam failures, especially
when every tailings dam is a unique structure made of natural materials and subject to its individual
loading conditions. Several studies have calculated the frequency of past tailings dam failures, resulting
in the following failure frequencies:
• An estimated 0.00050 failures per dam year (where dam year is the existence of one dam for one
year), or 1 tailings dam failure every 2,000 mine years, based on 88 failures from 1960 to 2010
(Chambers and Higman 2011).
• An estimated 0.00049 failures per dam year, or 1 tailings dam failure every 2,041 mine years, based
on 3,500 appreciable tailings dams that experienced an average 1.7 failures per year from 1987 to
2007 (Peck 2007).
• An estimated 0.00057 to 0.0014 failures per dam year, or 1 tailings dam failure every 714 to
1,754 mine years, based on a database including many unpublished failures that showed 2 to
5 major tailings dam failures annually from 1970 to 2001 (Davies 2002, Davies etal. 2000).
Available data do not permit reliable estimation of failure rates for different causes of failure or stages of
activity. Most failures have occurred while the tailings dams were actively receiving tailings (Table 9-1),
but the dam inventories do not indicate whether the thousands of dams in the inventory are active or
inactive and do not include years of operation. This prevents estimation of the proportion in each
category and makes it impossible to calculate the number of active dam-years.
Low failure frequencies and incomplete datasets also make any meaningful correlations between failure
probabilities and dam height or other characteristics questionable. Very few existing rockfill dams
approach the size of the structures in our mine scenarios, and none of these large dams have failed. For
example, although the 1,448 tailings dams listed in the National Inventory of Dams create a statistically
large and fairly complete database that includes dam heights, the International Commission on Large
Dams failure database includes only 49 U.S. tailings dam failures—too small a datasetto develop a
meaningful correlation between dam height and failure probability.
The historical frequencies of tailings dam failures presented above may be interpreted as an upper
bound on the failure probability of a modern tailings dam. Morgenstern (2011), in reviewing data
(Davies and Martin 2009), did not observe a substantial downward trend in failure rates over time.
However, improvements in the understanding of dam behavior, dam design, construction techniques,
construction quality control, dam monitoring, and dam safety assessment would be expected to reduce
the probability of failure for dams designed, constructed, and operating using more modern or advanced
engineering techniques. Silva et al. (2008) reported on over 75 earth dams, tailings dams, natural and
cut slopes, and some earth-retaining structures to illustrate the relationship between the level of
engineering, the annual probability of slope failure in earth structures, and factors of safety. They
grouped projects into the following four categories based on the level of engineering applied to the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~~ZApril 2013
-------
Chapter 9
Tailings Dam Failure
design, site investigation, materials testing, analysis, construction control, operation, and monitoring of
each project.
• Category I: Facilities designed, built, and operated with state-of-the-practice engineering. Generally,
these facilities are constructed to higher standards because they have high failure consequences.
• Category II: Facilities designed, built, and operated using standard engineering practice. Many
ordinary facilities fall into this category.
• Category III: Facilities without site-specific design and substandard construction or operation.
Temporary facilities and those with low failure consequences often fall into this category.
• Category IV: Facilities with little or no engineering.
The State of Alaska regulates its dams, including tailings dams, under Alaska Administrative Code (AAC)
Title 11, Chapter 93, Article 3, Dam Safety (11 AAC 93). Each dam is assigned to a class based on the
potential hazards of a tailings dam failure (Table 9-2). The tailings dams in our mine scenarios would be
classified as either Hazard Class I or II, both of which require a detailed computer stability analysis with
verification by other methods, and may require more sophisticated finite element analyses in special
circumstances. This analysis considers the effects of earthquakes based on a site-specific evaluation of
seismicity in the area (Section 3.6). Box 9-2 describes the selection of earthquake characteristics for
design criteria.
Table 9 2. Summary of Alaska's classification of potential hazards of dam failure.
Hazard Class
1 (High)
II (Significant)
III (Low)
Effect on Human Life
Probable loss of one or more lives
No loss of life expected, although a
significant danger to public health may
exist
Insignificant danger to public health
Effect on Property
Irrelevant for classification, but may include the same losses
indicated in Class II or III
Probable loss of or significant damage to homes, occupied
structures, commercial or high-value property, major highways,
primary roads, railroads, or public utilities, or other significant
property losses or damage not limited to the owner of the barrier
Probable loss of or significant damage to waters identified under
11 AAC 195.010(a) as important for spawning, rearing, or
migration of anadromous fish
Limited impact on rural or undeveloped land, rural or secondary
roads, and structures
Loss or damage of property limited to the owner of the barrier
Notes:
TSF dams in mine scenarios would be classified as Hazard Class 1 or II.
AAC = Alaska Administration Code.
Source: ADNR 2005.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-8
April 2013
-------
Chapter 9
Tailings Dam Failure
BOX 9 2. SELECTING EARTHQUAKE CHARACTERISTICS FOR DESIGN CRITERIA
Design criteria for dams specify that an evaluation be conducted to determine the effect of seismicity on
stability and performance of the dam. The seismic evaluation must establish the operating basis earthquake
(QBE) and maximum design earthquake (MDE). One important characteristic of determining earthquake
sizes is the return period (recurrence period) over which the event is likely to occur. If long return periods are
used in the analysis of earthquake size, the likelihood for a larger earthquake increases and the resulting
design basis earthquake will have a greater margin of safety.
The QBE represents the characteristic earthquake with a reasonable probability of occurring during the
functional lifetime of a project. Critical structures should be designed to withstand the effects of the QBE
and remain functional, with little, easily repairable damage. The QBE can be defined using a probabilistic
approach based on the likelihood that an earthquake of a certain magnitude and ground motion will be
exceeded during a particular period. Fora dam in Alaska with a Class II hazard potential, the return period
that must be considered for the QBE is 70 to 200 years—that is, the OBE represents the largest earthquake
likely to occur in 70 to 200 years.
The MDE represents the most severe earthquake considered at the site for which acceptable consequences
of damage will result. All critical structures such as tailings dams must be designed to resist the effects of
the MDE. The MDE can be determined based on historical earthquake patterns or through a rigorous
probabilistic analysis. For a Class II dam, the return period considered appropriate for the MDE is 1,000 to
2,500 years. Underestimating the MDE could increase the risk of a catastrophic tailings dam failure.
A third category of earthquake design level is the maximum credible earthquake (MCE). The term is not
defined in the Alaska dam safety regulations, but supporting guidance defines it as the greatest earthquake
that reasonably could be generated by a specific seismic source, based on seismological and geologic
evidence and interpretations. Design engineers sometimes use the MCE to represent a floating earthquake
(i.e., not associated with a known fault) located directly under a critical structure.
The return periods stated in Alaska dam safety guidance are inconsistent with the expected lifetime for
tailings dams for porphyry copper mines developed in the Bristol Bay watersheds, and represent a minimal
margin of safety. The mine scenarios represent approximately 25 to 78 years of mineral extraction, with
additional long-term operations likely required for closeout and maintenance of the mine. This period is
barely within the minimum OBE return period for Class II dams. The MDE analysis presents a potentially
greater chance of underestimating the size of a characteristic earthquake. Tailings storage facilities would
operate during the active mining period and could have a life expectancy of 10,000 years after operations
cease. Because the return period for the MDE is 1,000 to 2,500 years, this could lead to significantly
underestimating the largest earthquake that is likely to occur.
The Initial Application Package for Approval to Construct a Dam submitted by Northern Dynasty Minerals
(NDM) to the Alaska Department of Natural Resources (NDM 2006) included a seismic safety and design
analysis prepared by Knight Piesold Consulting that identified the following design criteria for the tailings
dams at the storage facility.
• OBE return period of 200 years, magnitude 7.5.
• MDE return period of 2,500 years, magnitude 7.8, with maximum ground acceleration of 0.3g, based on
Castle Mountain Fault data.
Northern Dynasty Minerals used a deterministic evaluation to select the MDE and MCE, which were deemed
equivalent for the preliminary safety design. In the application, NDM reports that the preliminary design
incorporates additional safety factors, including design of storage facility embankments to withstand the
effects of the MDE and a distant magnitude 9.2 event. Ghaffari etal. (2011) state that an MCE of
magnitude 7.5 with 0.44gto 0.47g maximum ground acceleration was used in the stability calculations for
the tailings dam design. Although the design specifications proposed in Ghaffari et al. (2011) exceed the
minimum requirements for dams in Alaska, the deterministic dataset used is small and contains
considerable uncertainties, which could lead to an underestimate of the potential seismic risk.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-9
April 2013
-------
Chapter 9 Tailings Dam Failure
The Guidelines for Cooperation with the Alaska Dam Safety Program (ADNR 2005) do not specify a
minimum safety factor for dams, but rather allow the applicant to propose one. Guidelines suggest that
the applicant follow accepted industry design practices such as those provided by U.S. Army Corps of
Engineers (USAGE), Bureau of Reclamation (Reclamation), Federal Energy Regulatory Commission
(FERC), and other agencies. Both USAGE and FERC require a minimum factor of safety of 1.5 for the
loading condition, corresponding to steady seepage with the maximum storage pool (FERC 1991, USAGE
2003).
Combining the required factor of safety with the correlations among level of engineering, factor of
safety, and slope failure probability (Figure 9-2) derived from Silva etal. (2008) yields an expected
annual probability of slope failure between 0.0001 (Category II) and 0.000001 (Category I). This
translates to one tailings dam failure due to slope failure every 10,000 to 1 million dam years. The upper
bound of this range is lower than the historical average of 0.00050 (1 failure every 2,000 dam years) for
tailings dams. This is partly because slope failure is only one of several possible failure mechanisms, but
it also suggests that past tailings dams may have been designed for lower safety factors or designed,
constructed, operated, or monitored to lower engineering standards. As shown in Table 9-1, slope
failures only account for about 25% of all tailings dam failures with known causes. Thus, the probability
of failure from all causes may be about four times higher than dam failures from slope instability alone
(an expected annual probability of failure between 0.0004 and 0.000004 or one tailings dam failure
every 2,500 to 250,000 dam years), when all potential failure causes are considered, albeit recognizing
that the small dataset may not be representative. Because 90% of tailings dam failures have occurred in
active dams (Table 9-1), the probability of a tailings dam failure after TSF closure would be expected to
be lower than the historical average for all tailings dams. These extremely low probabilities are
aspirational, based on engineering judgment, and cannot be confirmed by data. Modern high earthen
dams do not exist in large numbers and have not stood for very long. The frequencies and time courses
of failures may differ from the historical record or the design goals. In particular, the failure rates of
large earthen dams that are hundreds of years old are not known.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p , „ April 2013
-------
Chapter 9
Tailings Dam Failure
Figure 9 2. Annual probability of dam failure due to slope failure vs. factor of safety (modified from
Silva et al. 2008).
pF, Annual Probability of Failure
0 0 0 0 0 0 —
a, L> i. u r^ -^ O 1
*• .
*- ,
^
^^
•\
^
^^
^
^
^
^
^^
^\~
Category 1 Projects
-•-Category II Projects
.0 1.1 1.2 1.3 1.4 1.5
FS, Factor of Safety
Given an annual probability of failure per dam year, we can calculate the probability of the failure of any
project dam over any number of years. The three mine scenarios have different numbers of dams and
different operating lives. The Pebble 0.25 scenario would have a single dam and an operating life of
20 years. The Pebble 2.0 scenario would have three dams and an operating life of 25 years. The
Pebble 6.5 scenario would have eight tailings dams and an operating life of 78 years. For an upper bound
annual probability of failure of 0.0004, there would be a 0.8% probability of dam failure over the
operating life of the Pebble 0.25 scenario, a 3% of dam failure over the operating life of the Pebble 2.0
scenario, and a 22% probability of dam failure over the operating life of the Pebble 6.5 scenario. Using
the lower bound annual probability of failure of 0.000004 per dam, the failure probabilities during
operations would decrease to 0.008, 0.03, and 0.25% for the three scenarios, respectively.
If the tailings in the TSFs remain saturated, in part because of the need to keep the pyritic tailings
covered with water, the potential for failure of the dams over a longer period needs to be considered.
The probabilities that any of the dams would fail during a post-closure period of 1,000 years range from
upper bounds of 33, 70, and 96% to lower bounds of 0.4,1.2, and 3% for the three scenarios,
respectively, using the same range of annual failure probabilities as during operations.
9.1.3 Uncertainties
The variability in published probabilities of tailings dam failure reflects the uncertainty inherent in these
estimates. Much of this uncertainty is due to incomplete data. TSFs may remain in place for long periods.
Most dams are created as water holding dams with a limited expected lifespan (generally about
50 years). TSFs can be drained after mine closure to reduce the probability and consequences of tailings
dam failure, but draining a thick layer of fine-grained material can be difficult. Only 17 to 28% of the net
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-11
April 2013
-------
Chapter 9 Tailings Dam Failure
precipitation (depending on the TSF) would need to infiltrate into the tailings to maintain full saturation
with steady-state downward flow, so draining the TSF would require maintaining a very high runoff
percentage. Furthermore, if a tailings pond needs to be maintained to keep the pyritic tailings hydrated
and isolated from oxidation, the tailings dams must retain solid and liquid materials behind them in
perpetuity. This requires that the dams be maintained in perpetuity, in the face of uncertain seismic and
weather events that may have cumulative effects and occur over thousands of years.
9.2 Material Properties
9.2.1 Tailings Dam Rockfill
In our mine scenarios, TSFs would be enclosed by rockfill dams constructed primarily of well-graded,
non-acid-generating (NAG) waste rock obtained from the mine pit during operations. The starter dike
would contain material excavated from the upstream toe trench and local quarry. Waste rock from the
mine pit would be used as it became available. The size of the rock used to construct the dam would
depend on the rock's fracture characteristics, the methods used to blast and remove it from the mine pit,
and the lift thickness specified for adequate compaction. Particle sizes typically range from sand to large
boulders (Blight 2010). For a large rockfill dam with a high or significant hazard potential, the lift
thickness would be expected to be limited to 1.5 m to guarantee adequate compaction, limiting the
maximum particle size to about 1 m (Breitenbach 2007).
Well-graded rock would have a coefficient of uniformity, Deo/Dio, greater than 4 and would have a
coefficient of curvature, Dso/ (D6o*Dio), between 1 and 3. Combining these coefficients with Dawson and
Morin's (1996) report of a D50 particle size greater than 200 mm for waste rock, one can generate a
representative particle size distribution curve for the bulk of the tailings dam material (Figure 9-3).
9.2.2 Tailings Solids and Liquids
The tailings solids would include both bulk and pyritic tailings (Figure 9-3). The bulk tailings would
consist largely of sand and silt-sized particles (Dso = 200 urn), and have a dry density of 1.36 metric
tons/m3. The pyritic tailings would consist of predominantly silt-sized particles (Dso = 30 urn), and
would have a dry density of 1.76 metric tons/m3. The mass of the bulk tailings and the pyritic tailings
would equal 85 and 14% of the mass of the ore mass, respectively (Ghaffari et al. 2011). Representative
particle size distribution curves for the bulk, pyritic, and combined tailings are shown in Figure 9-3.
Given the bulk tailings dry densities reported above and using the specific gravity reported for the ore
(2.61 for the solids), the bulk tailings would be 52% solids and 48% liquid by volume. The pyritic
tailings, with a dry density of 1.76 metric tons/m3 and the same specific gravity, would be 68% solids
and 32% water. Based on the proportions of bulk and pyritic tailings, the combined material in the TSF
would be 55% solids and 45% water by volume, exclusive of any ponded water above the settled
tailings. As the tailings consolidate, the bulk density of the deeper tailings would be expected to increase,
although this consolidation may be limited (Section 6.3.2).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TI April 2013
-------
Chapter 9
Tailings Dam Failure
Figure 9 3. Representative particle size distributions for tailings solids (bulk and pyritic tailings)
and tailings dam rockfill. Tailings distributions are based on particle sizes specified by Ghaffari et al.
(2011).
-*- bulk tailings
-•-pyritic tailings
-A-combined tailings
— rockfill
,01
.10 1 10
Particle Size (mm)
100
1,000
9.3 Tailings Dam Failure via Flooding and Overtopping
Although a tailings dam failure is a low-probability event, the probability is not zero. Should such an
unlikely event occur, it is important to understand the potential impacts on the Bristol Bay watershed. In
this assessment, we consider the effects of a potential dam failure of TSF 1. Two cases of dam failure are
modeled: a volume approximating the complete Pebble 0.25 scenario (dam height = 90 m, maximum
capacity = 177 million m3) and a volume approximating the complete Pebble 2.0 scenario (dam height =
209 m, maximum capacity = 1,380 million m3). In each case, there is a maximum capacity and a
maximum expected capacity. The maximum capacity is the volume measured at the top of the dam, and
maximum expected capacity is the maximum volume expected to be stored, which is less than the
maximum capacity to provide freeboard in the TSF. In both cases, we assumed 20% of the impounded
tailings would be mobilized (Azam and Li 2010, Dalpatram 2011). While a variety of failure mechanisms
could cause a failure, this assessment used an overtopping scenario. We used a hydrologic model to
simulate a maximum flood hydrograph (Box 9-3) for both dam size failures and then modeled resulting
hydrologic conditions in the stream channel and floodplains, for a 30-km reach downstream of the TSF
(Box 9-4). Compared to the volume of released material during a failure, the hydrologic event that is
applied to create the overtopping is 2.5% of the total peak flow in the Pebble 0.25 failure and 0.1% of the
total peak flow in the Pebble 2.0 failure. The overtopping flow would not contribute significantly to the
released tailings volume, but would provide the failure trigger in the hydraulic modeling exercise.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-13
April 2013
-------
Chapter 9
Tailings Dam Failure
BOX 9 3. MODELING THE PROBABLE MAXIMUM FLOOD HYDROGRAPH AT TSF 1
We used the U.S. Army Corps of Engineers Hydrologic Engineering Center's Hydrologic Model ing System
(HEC-HMS) to generate a reasonable runoff hydrograph based on a 24-hour probable maximum
precipitation (PMP) event of 356 mm (14 inches) (Miller 1963). Application of the PMP to calculate the
probable maximum flood (PMF) is the accepted method for design and study of dams (ADNR 2005). The
PMF is used to determine appropriate spillway/bypass facilities, or to predict the greatest flood that can
cause failure. This conservative approach allows the full assessment of potential damage and impacts on
the facilities and downstream reaches. However, this PMP value is extrapolated from limited precipitation
gage data and has not been updated since 1963. It could be refined based on expanded gage data or
climate change projections (Box 14-2) and may ultimately change the predicted flood peak, but no update is
currently available. The HEC-HMS performs one-dimensional steady- and unsteady-state hydraulic
calculations for river systems. Inputs of combined watershed parameters are used to model stormwater
runoff characteristics for discrete watersheds. Basin characteristics for the TSF 1 site and the PMP were
applied to the SCS Type 1 hydrograph methodology to model data for the probable PMF hydrograph.
Modeled Precipitation and Flow Data for the Probable Maximum Flood (PMF) at TSF 1
Time (hour)
0:00
1:00
2:00
3:00
4:00
5:00
6:00
7:00
8:00
9:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
23:00
0:00
1:00
Notes:
Data are shown for a 24-hour period.
Precipitation (mm)
0.0
6.1
6.4
6.9
7.9
8.6
8.9
10.9
13.5
21.3
92.7
38.4
21.6
17.0
13.5
11.4
10.7
9.9
9.1
8.6
7.9
7.1
6.4
5.6
5.1
0.0
Total Flow (m3/s)
0.1
18.2
23.5
26.0
29.7
32.8
34.6
41.3
50.4
75.6
291.5
192.3
110.6
77.0
58.3
47.9
43.4
40.2
37.4
34.6
31.8
29.0
26.2
23.4
20.6
5.3
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-14
April 2013
-------
Chapter 9 Tailings Dam Failure
BOX 9 4. MODELING HYDROLOGIC CHARACTERISTICS OF TAILINGS DAM FAILURES
We used the U.S. Army Corps of Engineers Hydrologic Engineering Center's River Analysis System (HEC-RAS)
to model hydraulic characteristics of Pebble 0.25 and Pebble 2.0 tailings dam failures caused by flooding
and subsequent dam overtopping at TSF 1. HEC-RAS inputs included geometry of an inline structure to
simulate the dam cross-section and stream channel geometry data, both derived from a 30-m digital
elevation model, as well as hydrograph data to simulate the probable maximum flood (PMF) (Box 9-3). Under
both dam failure scenarios, results were modeled for 30 km (18.6 miles) downstream, from the face of the
TSF 1 dam down the North Fork Koktuli River valley to the confluence of the South and North Fork Koktuli
Rivers. The extension of the simulation beyond this point would have introduced significant error and
uncertainty associated with the contribution of the South Fork Koktuli River flows. The entire modeled flood
wave hydrograph includes the PMF inflow, excess water on top of the tailings, and 20% of the total tailings
volume. Channel roughness (i.e., Manning's n coefficient) was increased over typical values used in "clean
water" models to better reflect the influence of sediment-rich water during tailings dam failure.
The headwater location of TSF 1 (and of other likely TSF locations in the Nushagak River and Kvichak River
watersheds) would help to reduce the total volume of expected stormwater runoff into the TSF. If sufficient
freeboard were maintained, it would be possible to capture and retain the expected volume of the PMF in
the TSF. However, to examine potential downstream effects in the event of a tailings dam failure, we
assume that sufficient freeboard would not exist and overtopping would occur. This may be less likely when
the TSF would be actively monitored and maintained, but barring human error in the near term, may be
more representative of post-closure conditions in the future.
Tailings dam failure via overtopping is expected to have similar effects as failures resulting from other
causes (e.g., slope failure, earthquakes). We did not include a "dry weather" failure in our assessment but it
is judged that this kind of a failure (one that does not depend on a large precipitation event) would result in
similar liquefaction of stored tailings; however, transport of tailings downstream may be reduced in a dry
weather failure, as there is no precipitation generating additional flow. Available dry weather failure data
indicate that sediment distribution varied greatly from site to site. Our results are well within the ranges of
reported case histories.
9.3.1 Hydrologic Characteristics
Model results for hydrologic characteristics of the Pebble 0.25 and Pebble 2.0 dam failures are shown in
Tables 9-3 and 9-4. In both cases, estimated peak flows during a TSF dam failure would be much larger
than, and atypical, for flows experienced in this watershed. This is because the probable maximum flood
(PMF) and impounded tailings would create a flood wave far larger than any that could result from a
precipitation event alone. For example, under the probable maximum precipitation (PMP) event, the
14-km2 watershed above the TSF 1 would generate an instantaneous peak flow of 291 m3/s. This peak
flow would be compounded by the TSF failure and release of massive quantities of impounded tailings,
producing a peak flood immediately downstream of the dam of 11,637 m3/s for the Pebble 0.25 dam
failure and 149,263 m3/s for the Pebble 2.0 dam failure. A flood of this magnitude would dwarf the peak
flows of even the largest rivers in the region. For example, a local flood event measured by a
U.S. Geological Survey (USGS) gage on the Nushagak River located near the village of Ekwok, Alaska,
experienced a record peak flood of 3,313 m3/s. Despite this gage measuring the runoff from a 2,551-km2
watershed, the peak flow was well below the predicted release from a breach in the Pebble 0.25 TSF,
which would drain an area of only 14 km2.
It is important to acknowledge that use of the PMP and PMF to simulate an overtopping event is only
one potential failure mechanism that could be applied to a dam failure. Here it was used to acknowledge
a possible event and to review the overall influence of the flood wave generated from extreme
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTZ April 2013
-------
Chapter 9 Tailings Dam Failure
precipitation versus that of the dam failure release flow. The precipitation provides little contribution to
the overall failure flow rate and volume, but serves as the failure mode within the hydraulic model.
Specific information about the hydrologic and hydraulic modeling is provided below. However, this does
not provide a comprehensive review of debris flow and predictive sediment deposition. It merely
considers the magnitude of flood occurring following the rapid failure of a large tailings dam and
informs the assessment as to the potential for initial sediment distribution. This assessment recognizes
that a variety of scenarios could occur that would influence tailings and debris transport potential.
Included here is only one hydrology failure scenario where impoundment capacity is exceeded, due to
either lack of freeboard or bypass infrastructure failure. It should be noted that a scenario involving
failure during fair weather could also occur and cause similar down-valley flows.
Maximum instantaneous flood peak discharge would generally decrease with increasing distance
downstream from the dam, as the downstream topography becomes flatter and the flood wave spreads
out into the wider floodplain. When the flood wave recedes, water velocities would be expected to
decrease similarly under both dam failure scenarios (as reflected in the same minimum flow velocities
in Tables 9-3 and 9-4) and the potential for tailings deposition would be expected to increase.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p , g April 2013
-------
Chapter 9
Tailings Dam Failure
Table 9 3. HEC RAS model results for the Pebble 0.25 TSF dam failure analysis.3 Values were modeled for more than 80 river stations along
a 30 km length of the North Fork Koktuli River; representative river stations along that length are shown here, listed by distance upstream
from the confluence of the South and North Fork Koktuli Rivers (River Station 30 km = foot of the dam for TSF1).
River Station
(km)
30
28
25
20
15
10
5
1
Maximum Values from Dam Failure
Discharge
(mVs)
11,600
7,100
5,900
4,200
2,800
1,800
700
500
Depth
(m)
30
19
20
11
7
11
11
6
Velocity (m/s)
LFP
3
1
1
<1
<1
<1
<1
<1
CH
12
3
4
2
2
1
1
1
RFP
3
1
<1
<1
<1
<1
<1
<1
Minimum Values Post-Flood11
Discharge
(mVs)
<1
<1
<1
<1
<1
<1
<1
<1
Depth
(m)
<1
<1
<1
<1
<1
<1
<1
<1
Velocity (m/s)
LFP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
CH
1
<1
<1
<1
<1
<1
<1
<1
RFP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Notes:
a Dam height = 90 m, maximum volume of tailings and water expected to be stored = 158 million m3.
b Calculated by steady-state flow, discharge equal to mean monthly minimum volume runoff.
HEC-RAS = Hydrologic Engineering Center's River Analysis System; LFP = leftfloodplain; CH = channel; RFP = right floodplain.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-17
April 2013
-------
Chapter 9
Tailings Dam Failure
Table 9 4. HEC RAS model results for the Pebble 2.0 TSF dam failure analysis.3 Values were modeled for more than 80 river stations along a
30 km length of the North Fork Koktuli River; representative river stations along that length are shown here, listed by distance upstream from
the confluence of the South and North Fork Koktuli Rivers (River Station 30 km = foot of the dam forTSFl).
River Station
(km)
30
28
25
20
15
10
5
1
Maximum Values from Dam Failure
Discharge
(mVs)
149,300
64,500
55,900
42,300
30,000
20,400
7,200
7,200
Depth
(m)
99
100
97
28
25
35
53
44
Velocity (m/s)
LFP
4
1
1
1
1
1
<1
<1
CH
22
2
4
4
3
3
1
1
RFP
5
1
1
1
1
1
<1
<1
Minimum Values Post-Flood11
Discharge
(mVs)
<1
<1
<1
<1
<1
<1
<1
<1
Depth
(m)
<1
<1
<1
<1
<1
<1
<1
<1
Velocity (m/s)
LFP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
CH
1
<1
<1
<1
<1
<1
<1
<1
RFP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Notes:
a Dam height = 209 m, maximum volume of tailings and water expected to be stored = 1,270 million m3.
b Calculated by steady-state flow, discharge equal to mean monthly minimum volume runoff.
HEC-RAS = Hydrologic Engineering Center's River Analysis System; LFP = leftfloodplain; CH = channel; RFP = right floodplain.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-18
April 2013
-------
Chapter 9 Tailings Dam Failure
9.3.2 Sediment Transport and Deposition
Dam failure flood waves and post-failure low flows under both the Pebble 0.25 and Pebble 2.0 failure
conditions (Tables 9-3 and 9-4) suggest that transport and deposition of tailings material would occur
throughout (and beyond) the 30-km modeled reach. After the initial deposition event, concentrated
channel flows and floodplain conveyance areas would continue to transport sediment, as channel and
valley morphology re-established in the newly deposited substrate.
The flood wave and tailings deposition that would result from a tailings dam failure under either dam
failure scenario could significantly alter the downstream channel and floodplain, even with only 20% of
impounded tailings mobilized. The initial flood itself would have the capacity to scour the channel and
floodplain as the high-velocity wave of tailings slurry traveled down the valley. The quantity of
mobilized sediments that could be released from the TSF would bury the existing channel and floodplain
under meters of fine-grained sediment as sections of debris slowed and as the flood receded. Tables 9-3
and 9-4 describe the full range of velocities predicted for both failure volumes. The sediment regime of
the affected stream and downstream waters would be greatly altered. Based on review of historical
failures, it was determined that an initial "wedge" of sediment could be deposited rapidly in the near
downstream vicinity of the dam. Given the mobility potential of the fine grains that make up the majority
of the expected tailings, initial modeled slope was held to 15:1 (H:V). Extending this slope from the
lowest elevation in the dam breach, the calculated sediment depths ranged from 45 to 10 m extending
downstream 1.3 and 3.3 km for the Pebble 0.25 and Pebble 2.0 dams, respectively. This represented a
compromise of not describing a complete sediment runout (i.e., emptying of the TSF), and recognizing
that depths near the dam would likely be larger than depths further downstream. The choice of slope
was determined to be reasonable by comparison to a simple tailings flow slide calculator publicly
available (World Information on Service Energy 2012), which predicts flow depths for tailings with a
variety of viscosities.
Downstream of the initial sediment wedge, deposition could occur in the channel and the floodplain of
each section as the flood and debris flow receded. To illustrate one reasonable outcome of sediment
deposition and distribution, we used the one-dimensional Hydrologic Engineering Center's River
Analysis System (HEC-PvAS) hydraulic model (Box 9-4) to estimate tailings deposition and quantities of
sediment available for downstream transport along the stream network (Box 9-5). Downstream of the
initial depositional wedge, it was assumed that depths would continue to vary. For the purposes of
illustrating the volume of sediment available and one possible scenario of deposition and transport, we
modeled that, on average, approximately 0.3 m of deposited tailings would remain on valley surfaces
following the failure. Given the volume of available sediment released, we also assumed that, on average,
0.3 m of sediment would initially deposit in the stream, or the stream would be buried and a new stream
would form in the sediment deposited on the floodplain.
Most of the deposition would be very fine material that is susceptible to re-suspension and deposition
with each subsequent natural flow event. Following the dam failure, the stream channel would seek
equilibrium and may remain unstable over several flow events, potentially creating a braided system in
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ , „April 2013
-------
Chapter 9 Tailings Dam Failure
the post-failure depositional zone. As the new valley fluvial geomorphology develops with time, newly
deposited materials on the floodplain, at the base of the dam, and sediments that remained behind the
breached dam of the TSF (if not contained by corrective action) would serve as concentrated sources of
easily transportable, potentially toxic material (Section 9.4).
In both failure calculations, substantial sediment remained available for subsequent transport and
deposition. For seven of the eight scenarios considered in Table 9-5,17 to 84% of the released tailings
(3.8 million to over 303 million metric tons) remained available at the 30 km endpoint (Table 9-5). We
made a simple assumption for modeling purposes that deposition depths averaged 0.3 m. We emphasize
that our scenarios reflect a range of possible outcomes, but are not exhaustive. Depths of sediment on
the floodplain could be much greater than 0.3 m, in which case more sediment could be captured in the
valley closer to the TSF dam. Alternatively, average deposition depths could be lower and effects could
actually extend far beyond the end of the 30-km reach. Based on historical tailings dam failure data,
potential runout distances can range from hundreds to thousands of kilometers (Box 9-1). When the
parameters for the Pebble 0.25 and Pebble 2.0 dam failures were applied to runout distance equations
from Rico et al. (2008), the expected runout distance under the Pebble 0.25 dam failure was 35 km,
reaching the mainstem Koktuli River. Under the Pebble 2.0 dam failure, this distance increased to
307 km (190 miles), reaching the marine waters of Bristol Bay.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p 2Q April 2013
-------
Chapter 9 Tailings Dam Failure
BOX 9 5. USING HYDROLOGIC MODELS TO ESTIMATE TAILINGS DEPOSITION AFTER
A TAILINGS DAM FAILURE
We used outputs from the one-dimensional Hydrologic Engineering Center's River Analysis System (HEC-
RAS) hydraulic model (Box 9-4) to estimate tailings deposition along the stream network, based on
calculated water depths and the assumption that tailings would settle as the velocity of sediment-rich water
decreased across the floodplain. HEC-RAS is most often used to simulate clear water flows. The flow
calculation is completed between successive cross-sections in the model, balancing the hydraulic energy to
determine the water surface elevations and flow velocity, and then moving to the next cross-section in the
sequence and repeating the process. When applied to tailings dam failure events, it is appropriate to
increase channel roughness coefficients to better emulate flow characteristics of concentrated sediment
flows. Manning's n = 0.2 for the channel and 0.6 for the floodplain were selected. The dam failure was set
to fail from overtopping 1 hour after the peak flow of the potential maximum flood (PMF), and the breach
reached maximum size over 30 minutes, releasing a flood hydrograph equal to 20% of the available storage
capacity for each scenario. The 20% released volume included both solids and pore water. We assumed that
sediment deposition would be greatest near the dam, forming a "wedge" from the lowest elevation of the
breach and extending downstream. The calculated sediment depths ranged from 45 to 10 m and extended
1.3 and 3.3 km for the 90-m (Pebble 0.25) and 209-m (Pebble 2.0) dam failures, respectively. It was also
predicted that deposition could occur in the channel and the floodplain of each section following the
maximum predicted flow depth during the peak of the flood wave as the flood and debris flow receded.
Using this maximum width of inundation, a 0.3-m depth of sediment was deposited on the floodplain and
channel. This created a very conservatively uniform estimate of sediment deposition. Deposition at each
cross-section at this 0.3-m meter depth was used to calculate the volume between modeled river sections,
and this volume was subtracted from the volume released from the tailings dam failure. It was assumed that
the remaining sediment in the tailings dam failure flow was available to deposit at the next downstream
section. This logic was carried downstream until the end of the modeled river length was reached.
We did not extend the analysis beyond the 30-km reach of the North Fork Koktuli River near its confluence
with the South Fork Koktuli River. At some point downstream of the tailings dam failure, the gross deposition
of sediment would cease and the flow dynamics of a typical sediment transport analysis would govern. We
assumed that the confluence is where a more traditional sediment transport analysis would be appropriate.
Given the scope of the current analysis, a traditional sediment transport analysis was not feasible. This
discussion is limited to the estimation of probable sediment distribution immediately after the tailings dam
failure and the total volume of sediment available to accommodate these assumptions.
We assumed a particle size distribution of 0.001- to 1.0-m diameter for the dam construction material, and
less than 0.01-to just over 1.0-mm diameter for the impounded tailings material (Figure 9-3). Based on the
Hjulstrom curve—which estimates when a stream or river will erode, transport, or deposit sediment based on
flow speed and sediment grain size—all of the mobilized tailings would remain in suspension at water
velocities greater than 0.05 m/s. This indicates that the channel would transport tailings under typical
stormflow conditions and deposited tailings from floodplain terraces could be suspended and transported,
altering channel planform in the newly altered valley during typical storm events subsequent to the failure
occurrence.
Based on the particle size curves in Figure 9-3, it is reasonable to assume that all construction material from
the dam breach and from 30 to 66% of the impounded tailings material could contribute to debris flow
following a tailings dam failure (Browne 2011). However, the volume of material mobilized, the distance it
travels downstream, and the amount of deposition can vary greatly based on factors such as dam height,
material size distribution, and material water content at the time of failure (Rico etal. 2008). Thus, we used
conservative estimates for the percentages of impounded tailings material mobilized (5 to 20%) (Table 9-5).
Using a value less than measured historical release volumes ensured that we were not overestimating
available sediment in the tailings dam failure calculations, and that volumes up to 20% would be considered
reasonable at this level of detail. We focus on transport and deposition of the fine-grained (less than
1.0 mm) tailings material, given the assumption that larger dam construction material would deposit within
the first few kilometers downstream of the failure.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment Q 2± April 2013
-------
Chapter 9
Tailings Dam Failure
Table 9 5. Tailings potentially mobilized and deposited during Pebble 0.25 and Pebble 2.0 dam failures at TSF 1. The total of mobilized
tailings includes material within the dam cross section that has failed, plus a percentage (5 to 20%) of stored tailings material. The weight of
tailings dam.
Failure
Scenario
Pebble 0.25
Pebble 2.0
Maximum
Capacity3
(million m3)
177
1,380
Volume of Tailings Solidsb
(million m3)
158
1,270
% Mobilized'
20
15
10
5
20
15
10
5
Mobilized Solids
(metric tons)
44,872,000
33,654,000
22,436,000
11,218,000
360,396,000
270,297,000
180,198,000
90,099,000
Solids in Transport
at Downstream
Extent of Model
(metric tons)
26,264,000
15,046,000
3,828,000
0
303,322,000
213,223,000
123,124,000
33,025,000
Downstream
Extent of
Wedge
(km)
1.25
1.25
1.25
1.00
3.27
3.27
3.27
3.27
Downstream
Extent of
Expected
Deposition11
(km)
30(+)
30(+)
30(+)
1
30(+)
30(+)
30(+)
30(+)
Notes:
a Volume measured at top of dam. This volume was used to model failure via overtopping.
b Maximum volume of tailings and water expected to be stored, allowing for freeboard in TSF. This volume was used to estimate metric tons of stored tailings released in a TSF dam failure, using an
average tailings density of 1.42 tons/m3.
c 20% value was used in model; values less than 20% are shown to illustrate how weight of mobilized solids changes with % mobilized.
d Measured downstream from face of dam.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-22
April 2013
-------
Chapter 9 Tailings Dam Failure
9.3.3 Remediation
The above description of a tailings dam failure and the habitat impacts described below do not consider
the mitigating effects of remediation efforts on the part of the mine operator or other parties. Past
tailings dam failures have been substantially remediated, perhaps most notably the 1998 failure of the
Aznalcollar Tailings Dam at Los Frailes Mine in Seville, Spain.
The remediation of the Aznalcollar Tailings Dam failure was completed in the 6 months between the
April failure and the onset of heavy rains in October. The Aznalcollar area, near Seville, has a much drier
and warmer climate than the Bristol Bay area. It is also a heavily farmed area with flatter topography,
better access, and more readily available equipment and labor. The success at removing the tailings
from the Aznalcollar failure may be difficult to replicate in the Bristol Bay area. The potential releases
from TSF 1 would be much larger than the release at Aznalcollar, road access would be extremely
limited, and it is unlikely that an accessible, ready supply of labor or equipment to perform the
remediation would be available for a rapid response. Remediation might be more successful in the
winter when vehicle and equipment access might be less damaging to the environment and if the
deposited tailings sediment on the floodplains froze, thereby limiting re-suspension during removal.
Despite remediation efforts, it is likely that significant amounts of tailings would be re-suspended before
remediation efforts were implemented and completed and that significant amounts of tailings would
remain in the North Fork Koktuli River watershed for some time after remediation was completed. It
may be difficult to source the needed quantity of cobbles and gravel to restore the streambed substrate.
It may be technically impractical to recover tailings fines that would have been transported past the
point of confluence with larger rivers.
9.3.4 Uncertainties
The uncertainty of sediment deposition depths and downstream distribution is great, because valley
topography, rate of the dam failure, and ultimate make-up of the flood wave sediment concentration and
viscosity can alter outcomes and upset the best predictive efforts. Despite the uncertainty associated
with the massive quantities of sediment available and the complexities of hydraulic forces that would act
on this sediment, we present one reasonable post-failure sediment deposition outcome. Other outcomes
are possible, but all share the common reality that massive quantities (millions of cubic meters) of
tailings fines would be available for transport and deposition (Table 9-5).
Use of a traditional sediment transport model would likely improve estimates of sediment movement
and deposition, especially as the model is extended further downstream. As more sediment is deposited,
flow would be expected to become less saturated. In addition, tributary streams would input clean water
at each confluence. Because of the site-specific data required to implement a sediment transport model,
we limited our model to the 3 0 km above the confluence of the South and North Fork Koktuli Rivers.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p -3 APril 2013
-------
Chapter 9 Tailings Dam Failure
9.4 Scour, Sediment Deposition, and Turbidity
The Pebble 0.25 and Pebble 2.0 tailings dam failures described above could have devastating effects on
aquatic life and habitat. We identified several processes associated with a tailings dam failure that would
pose risks to aquatic habitat. These include exposure to hydraulic scour and bed mobilization,
deposition of tailings fines, and mobilization and suspension of tailings fines affecting downstream
water and habitat quality. Effects of suspended sediments are discussed in Section 9.5.1, and those
associated with toxicity of spilled water and deposited sediments are discussed in Section 9.5.2.
Figure 9-1 illustrates potential pathways linking a tailings storage facility failure to habitat and water
quality alterations, and projected effects on fish endpoints.
Natural background conditions provide an indication of the sediment levels that support the current
productivity of salmonid populations. Two available sources provide data on substrate size distribution
and fine sediment concentrations in the study area. Pebble Limited Partnership's (PLP's) Environmental
Baseline Document2004 through 2008 (EBD) (PLP 2011: Appendix 15.1F, Fluvial Geomorphology
Studies) reports concentrations of fine sediments from sieved bulk gravel samples collected at three
known salmon spawning site in the South and North Fork Koktuli Rivers and Upper Talarik Creek (PLP
2011: Figure 4 in Appendix 15.IF shows sample locations). Average concentration of fines (less than
0.84 mm) was less than 6% for all streams and dates, except for the August sample from the uppermost
South Fork Koktuli River site (gage SGSK3) (PLP 2011: Figure 27 in Appendix 15.IF), which had nearly
8% fines. The geometric mean grain size was greater than 15 mm at all sites for both sampling periods,
except the uppermost Upper Talarik Creek site (gage SGUT3), where the mean grain size for both
seasons was between 10 and 15 mm (PLP 2011: Figure 26 in Appendix 15.IF). These data led the
authors to conclude that gravel quality was generally high and that, based on published criteria (Shirazi
etal. 1981, Chapman and McLeod 1987, Kondolf 2000), salmonid survival to emergence would be high
(presumably above 80%) at all sites except the uppermost Upper Talarik Creek site, where criteria
predicted survival between 50 and 80% (PLP 2011).
Areal coverage of substrate sizes is available for 77 wadeable stream sites around the Nushagak and
Kvichak River watersheds, including one site each on the South and North Fork Koktuli Rivers and
Upper Talarik Creek (Table 9-6). Substrate sampling at these study sites followed U.S. Environmental
Protection Agency (USEPA) methodology (Peck et al. 2006), in which five particles were systematically
selected across each of 21 evenly spaced transects (from each wetted margin and three locations in
between). These data indicate that a mix of substrate sizes occurs in these streambeds, with cobble and
gravel generally abundant (Table 9-6).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p -4 April 2013
-------
Chapter 9
Tailings Dam Failure
Table 9 6. Sediment size distributions surveyed at the South and North Fork Koktuli Rivers, Upper Talarik Creek, and 77 wadeable stream
sites in the Nushagak and Kvichak River watersheds. Figures represent percentage areal coverage based on 105 systematically selected
particles at each site, following USEPA methods. All data were collected during June.
River or Stream(s)
South Fork Koktuli River
North Fork Koktuli River
Upper Talarik Creek
77 streams
Date
6/8/2010
6/6/2009
6/13/2011
2008 to 2011
Latitude
59.83047
59.84033
59.9182
Longitude
-155.27719
-155.71272
-155.27771
% Large
Boulder
(>1000 mm)
-
0(±1)
% Small Boulder
(250-1000
mm)
3
-
2
2 (±4)
% Cobble
(64-250
mm)
3
17
30
13 (±13)
% Coarse
Gravel
(16-64 mm)
55
49
29
40 (±15)
% Fine
Gravel
(2-16 mm)
16
24
13
17 (±12)
% Sand
(0.06-2
mm)
0
10
24
17 (±11)
% Fines
(<0.06
mm)
23
-
2
Notes:
Blank values (-) indicate values equal to zero.
Sources: Rinella pers. comm., Pecketal. 2006.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-25
April 2013
-------
Chapter 9 Tailings Dam Failure
9.4.1 Exposure through Sediment Transport and Deposition
The tailings dam failure scenarios described above would result in intense scour and deposition in the
North Fork Koktuli River valley, from the tailings dam downstream to at least the confluence with the
South Fork Koktuli River, a distance of approximately 30 km. The volume of available fine tailings
material that could be mobilized could result in many meters of deposition in a sediment wedge of
tailings fines across the entire valley near the TSF dam, with lesser quantities of fines deposited to at
least the confluence with the South Fork Koktuli River. Substantial volumes of sediment would remain
available for transport and deposition beyond this point for failures that mobilize at least 10 to 15% of
the tailings facility content (Table 9-5). Erosion and transport of fines would be expected to continue as
the channel adjusts to the vastly increased fine sediment supply.
To translate this tailings dam failure into effects on aquatic habitat and biota, we assumed that the
velocities calculated during the tailings dam failure flood event (Tables 9-3 and 9-4) would result in a
nearly complete reworking and mobilization of the existing North Fork Koktuli River channel bed and
banks, and much of the valley. Given the volume of material that would be exported from the TSF, we
assume that portions of the new valley floor would be predominately tailings material, with particle
sizes ranging from less than 0.01 mm to just over 1.0 mm, of which 70% by mass would be finer than
0.1 mm. Following recession of the tailings dam failure flood event, we assume that the bed, margins and
floodplain would be primarily tailings material, with incorporated coarser dam fill and valley fill
material accounting for less than 20%.
Immediately following either a Pebble 0.25 or Pebble 2.0 tailings dam failure, suitable spawning and
rearing habitat for salmon and other native fishes in the North Fork Koktuli River downstream of the
tailings dam would be eliminated. Tributaries of the North Fork Koktuli River, including portions of the
watershed upstream of the confluence of North Fork Koktuli River tributary containing the TSF, could
also be adversely affected. Temporary flooding of tributary junctions during the tailings dam failure
event, and subsequent sediment deposition at confluence zones, causing local aggradation, steepening,
or shallowing of tributary confluences, could make movement of resident and anadromous fish between
tributaries and the mainstem difficult Recovery of channel dimensions and substrate size distributions
suitable for salmonid spawning and rearing habitat would be contingent upon rates of fine sediment
export and recruitment of gravels and larger substrates from tributaries or pre-failure valley fill.
The type, magnitude, and frequency of channel adjustments that would occur in the North Fork Koktuli
River valley following a tailings dam failure would depend on available sediment, channel slope, and
discharge. Post-failure streams flowing across the depositional zone would have tremendous supplies of
fine-grained sediments in the channel bed and banks available for transport. Channels would be
expected to experience rapid channel incision with frequent bank failure, followed by periods of channel
widening and aggradation interspersed with episodic channel avulsion. Given the volume and depth of
deposition, stream channels would be expected to remain unstable and continue to contribute
sediments to downstream reaches until equilibrium conditions were met. Recovery of suitable
structural habitat in the North Fork Koktuli River watershed would likely take decades, given the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -g April 2013
-------
Chapter 9 Tailings Dam Failure
volume of sediment that could be delivered under the tailings dam failures considered here. Whether
the benefits of removal of spilled tailings fines outweigh the risks of additional adverse impacts resulting
from dredging and removal operations would be contingent upon the nature and distribution of tailings
spill, duration of risks, and existing technologies (e.g., Wenning et al. 2006).
The TSF 1 tailings dam failure scenarios considered above would have the potential to fill the North Fork
Koktuli River valley with extensive deposits of tailings fines less than 0.1 mm in size and still carry a
substantial volume of fine sediments farther downstream. The mass of material remaining in transport
at the confluence of the South and North Fork Koktuli Rivers and thus available for deposition in the
mainstem Koktuli, Mulchatna, and Nushagak Rivers following a Pebble 0.25 tailings dam failure could be
as high as 26 million metric tons, depending on the proportion of TSF fill material mobilized in the spill
(Table 9-5). The volume of sediment remaining in transport at the confluence following a Pebble 2.0
tailings dam failure could exceed 303 million metric tons (Table 9-5). The depth and distribution of fines
in the mainstem Koktuli, Mulchatna, and Nushagak Rivers cannot be estimated at this time, but it is
reasonable to expect that continued pulses of fine sediments would be transported through and
transiently stored in these mainstem rivers during spring snow melt and fall rain events for many years
(Major etal. 2000).
9.4.2 Exposure-Response
9.4.2.1 Fish
The State of Alaska standard for accumulation of fine sediment (0.1 to 4.0 mm) is "no more than 5%
increase by weight above natural conditions (as shown by a grain size accumulation graph) with a
maximum of 30% fines in waters used by fish for spawning" (ADEC 2011). Bryce et al. (2010) found that
even small amounts of fines (exceeding 5% fines or 13% sands and fines) in streambed fine sediments
were associated with declines in sediment-sensitive aquatic vertebrates, including salmonids. The
tailings dam failures evaluated here would completely scour and transport or bury existing substrates in
the North Fork Koktuli River valley under tailings fines, greatly exceeding all sediment criteria for
salmonid spawning. Continued erosion and transport of fines deposited on bars, floodplains, and
terraces would provide a chronic source of additional fine sediments during precipitation events,
providing new inputs of fines during spawning and egg incubation. Thus, exceedance of fine sediment
standards in the entire North Fork Koktuli River would be a likely outcome for years to decades.
Interstitial spaces used by juvenile salmonids for overwintering and concealment are a critical habitat
resource, particularly in northern ice-bound rivers and streams (Bustard and Narver 1975, Cunjak 1996,
Huusko etal. 2007, Brown etal. 2011). Interstitial habitat would initially be eliminated by the tailings
dam failure, and then subject to continued high levels of embeddedness as new channels erode into the
new valley fill composed of tailing fines. The new sediment regime in the North Fork Koktuli River
watershed and associated transport and storage of massive quantities of fine sediments would
essentially eliminate interstitial habitat from the watershed for years to decades, if not longer. Altered
valley morphology and substrate composition would also very likely lead to changes in groundwater
flowpaths and interactions with surface waters. Infiltration and burial of coarse valley fill by fine
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -7 April 2013
-------
Chapter 9 Tailings Dam Failure
sediments could greatly reduce hydraulic conductivity and result in decreased rates of exchange
between surface water and groundwater (Hancock 2002). Because of these habitat changes, suitable
spawning environments and overwintering habitats for salmonids would be greatly diminished in this
watershed, likely leading to severe declines in salmonid spawning success and juvenile survival (Wood
andArmitage 1997).
9.4.2.2 Invertebrates
Aquatic macroinvertebrates are an important food source for Chinook and coho salmon, rainbow trout,
Dolly Varden, Arctic grayling, and other fishes that rear in the study area's streams (Nielsen 1992,
Scheuerell etal. 2007). Two available data sources describe the existing macroinvertebrate communities
for streams in the study area: PLP (2011: Chapter 15.2) and Bogan et al. (2012). Both documents
describe broadly similar communities that are consistent with those reported from other regions of
Alaska (Oswood 1989). Communities are reasonably diverse: Bogan etal. (2012) reported 137 taxa from
38 families, with 9 to 40 taxa occurring at a given site (Chironomidae were lumped at the family level).
Communities are dominated by Diptera (true flies), primarily Chironomidae (midges), with lesser
numbers of Ephemeroptera (mayflies) and Plecoptera (stoneflies) and relatively few Trichoptera
(caddisflies). Macroinvertebrate densities were characteristically variable, ranging two orders of
magnitude (102 to 11,371 organisms per m2) (Bogan et al. 2012).
In addition to the direct impacts on fish described in Section 9.4.2.1, catastrophic sedimentation
associated with tailings dam failure also would likely affect fish populations through habitat-related
reductions in macroinvertebrate food resources (see Section 9.5 for discussion of toxic effects).
Sedimentation can affect benthic macroinvertebrates through abrasion, burial, or reductions in living
space, oxygen supply, and food availability (Jones et al. 2011). Deleterious effects of sedimentation have
been reviewed thoroughly (Wood and Armitage 1997, Jones et al. 2011). Sedimentation typically leads
to reductions in density and taxonomic diversity (Wagener and LaPerriere 1985, Gulp et al. 1986, Quinn
et al. 1992, Milner and Piorkowski 2004), even at sediment loads substantially lower than those
modeled under the tailings dam failure (Wood and Armitage 1997, Jones et al. 2011). The conversion of
a stable streambed dominated by gravel and cobble to a highly unstable one composed entirely of fine
sediments, as would occur in the tailings dam failures considered here, would certainly lead to
reductions in the biomass and diversity of macroinvertebrate prey available to fish populations.
9.4.3 Risk Characterization
The complete loss of suitable salmonid habitat in the North Fork Koktuli River in the short term (less
thanlO years), along with the likelihood of very low-quality spawning and rearing habitat in the long
term (decades), would likely result in near-complete loss of North Fork Koktuli River fish populations
downstream of the tailings dam. These impacts would persist for multiple salmon life cycles, so salmon
cohorts that are at sea during the tailings dam failure would eventually return to find degraded
spawning and rearing habitat. The North Fork Koktuli River provides complex, low-gradient, high-
quality habitats that currently support spawning and rearing populations of sockeye, Chinook, and coho
salmon, and spawning populations of chum salmon (Johnson and Blanche 2012). The North Fork Koktuli
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -gApril 2013
-------
Chapter 9 Tailings Dam Failure
River (ADF&G 2012) also supports rearing Dolly Varden and rainbow trout. The Koktuli River
watershed has been recognized as an important producer of Chinook salmon for the greater Nushagak
River Management Zone (Dye and Schwanke 2009, ADF&G 2011). Total Chinook salmon run-size
estimates for the Nushagak River include estimates of harvest plus escapement of spawners. Estimates,
based on a variety of techniques including sonar, averaged over 190,000 Chinook salmon from 2002
through 2011 (Buck et al. 2012), making the Nushagak the largest producer of Chinook salmon for the
Bristol Bay region. Of all the Chinook salmon tallied during annual aerial index counts (Section 7.1.2 for
a discussion of limitations of aerial counts for abundance estimates) in the Nushagak River watershed
between 1969 and 1985 (years that all reported spawning areas were surveyed), on average 29%
(range 21 to 37%) were counted in the Koktuli River (Dye and Schwanke 2009). The Mulchatna River
accounts for another 12% (range 9 to 17%) of the Nushagak Chinook salmon count, and the Stuyahok
River (which drains to the Mulchatna downstream of the Koktuli River) represents another 18% (range
10 to 27%). Hence, Chinook salmon production could be significantly degraded by loss of habitat
downstream of the tailings dam, particularly if effects extended downstream into the Koktuli and
Mulchatna Rivers and beyond.
Sockeye are the most abundant salmon returning to the Nushagak River watershed, with annual runs
averaging more than 1.9 million fish between 2001 and 2010 (Jones etal. 2012: 86). Spatially extensive
sockeye salmon spawner data are not available for the Nushagak River watershed, so it is impossible to
estimate what proportion of the population spawns in the Koktuli River. The Nushagak River watershed
supports two genetically and ecologically distinct groups of sockeye salmon (Dann etal. 2011: 26)—
those that rear in, and spawn in and near large lakes (lake-type, as in Semko 1954), and those that
spawn and rear, at least briefly, in rivers and streams (sea-type and river-type, as in Semko 1954,
collectively called here riverine-type}. Sockeye salmon in much of the Mulchatna River system, including
the Koktuli River and adjacent Stuyahok River, are riverine-type, and are more closely related to
riverine-type sockeye salmon of the Kuskokwim River drainage than to Nushagak River watershed lake-
type sockeye salmon (Dann etal. 2011: 26). It is likely these population groups share a similar life
history pattern. Riverine-type sockeye in Kuskokwim River tributaries preferentially rear in off- and
side-channel habitats within floodplain-prone stream reaches (Ruggerone etal. 2011: 56). From 1995 to
2006, an estimated 528,000 adult sockeye salmon annually migrated to spawning areas in the Nushagak
and Mulchatna river systems upstream of the Wood River system (Jones etal. 2012: 85). Of these,
approximately 70% (an annual average of 363,000) appear to be riverine-type sockeye salmon based on
the proportion of sockeye that escaped to the Nushagak/Mulchatna portion of the basin.
Spawning and rearing riverine-type sockeye salmon habitats occur repeatedly in the South and North
Fork Koktuli Rivers downstream to beyond the confluence of the Mulchatna and Nushagak Rivers
(ADF&G 2012). The tailings dam failure considered here would likely affect sockeye salmon production
throughout the Koktuli River system, but the proportion of the total Nushagak River sockeye salmon
production that would potentially be affected is unknown. See Section 7.1 for more information on fish
abundance.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -g April 2013
-------
Chapter 9 Tailings Dam Failure
Populations of resident and anadromous fishes present in North Fork Koktuli River headwaters
upstream of TSF 1 or in tributaries at the time of the tailings dam failure would not immediately suffer
habitat losses, but would suffer indirect effects resulting from alteration of the North Fork Koktuli River
valley. Many species in the region's rivers, including resident non-anadromous species, undergo
extensive seasonal migrations (West et al. 1992). Such movements are important for juveniles moving
from natal areas to overwintering habitats, for smolts emigrating to the sea, for adult spawning
migrations, or, in the case of resident species, for migration between different areas for spawning,
foraging, and overwintering. Sediment deposition at tributary mouths in the North Fork Koktuli River
valley could adversely affect passage of juvenile and adult fish into and out of these tributaries. For
several years, mainstem river habitats upon which many tributary fish depend upon for portions of their
life history could be gone, inaccessible due to blocked access, or severely degraded.
Successful re-colonization of the North Fork Koktuli River by resident fish would depend on whether
unimpaired tributary habitats or downstream areas could function as suitable refugia and source areas
for re-colonization of the North Fork Koktuli River. Resident fishes would require sufficient tributary
habitat to complete their entire life history, as it is likely that downstream habitat would be unusable for
multiple generations. Re-colonization of salmon from tributary refugia or downstream areas would
require suitable passage at tributary junctions and suitable migratory corridors throughout the
mainstem. Aquatic macroinvertebrate food resources would also likely be adversely affected in the main
river channel, limiting rearing potential for insectivorous fish such as juvenile salmonids. Given
estimates of fine-sediment deposition and the unstable, silt and sandbed channels that would likely form
across the valley floor, and likely metal concentrations in these tailing substrates that could inhibit
migratory behavior (Section 9.5.2.1), successful migratory conditions seem unlikely for at least several
years after a tailings dam failure.
The near-complete loss of North Fork Koktuli River fish populations and long-term transport of fine
sediment to downstream locations would have significant adverse effects on the Koktuli and Nushagak
River salmon, Dolly Varden, and rainbow trout populations, affecting downstream fisheries, including
subsistence users (Figure 5-2). Spawning and rearing habitat would be eliminated or impaired by
deposition of transported sediment and/or reductions in invertebrate prey base. Direct loss of habitat in
the North Fork Koktuli River, and impairments further downstream because of transport and deposition
of sediment could adversely affect a substantial portion of Chinook salmon returning to the Nushagak
watershed. Assuming that Alaska Department of Fish and Game (ADF&G) aerial survey counts reflect the
proportional distribution of Chinook salmon in the Nushagak River watershed, habitat destruction of the
North Fork Koktuli River valley, downstream transport of sediment to the Koktuli River mainstem, and
the subsequent loss of access to the South Fork Koktuli River would affect the entire Koktuli River
component of the Nushagak Chinook run. If the deposited tailings material is of sufficient quantity and
toxicity (Section 9.5.2) to have effects on aquatic life and migratory behavior offish in the lower Koktuli,
Mulchatna, and Stuyahok Rivers, much greater proportions of the Nushagak Chinook populations and
other resident and anadromous fish populations could be affected. Adult salmon returning to these
rivers could potentially seek other tributaries for spawning, but successful recruitment of displaced
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -Q April 2013
-------
Chapter 9 Tailings Dam Failure
spawners would require access to and comparable use of spawning and rearing capacity elsewhere in
the Nushagak watershed.
9.4.4 Uncertainties
It is certain that a tailings dam failure would have devastating effects on aquatic habitat and biota, but
the distribution and magnitude of effects is uncertain. Uncertainties associated with the initial events,
including the likelihood of dam failure, and sediment transport and deposition processes, are discussed
in Sections 9.1.3 and 9.3.4. Other uncertainties related to the timeframe for geomorphic recovery, the
longitudinal extent and magnitude of habitat impacts downstream of the end of our modeled reach of
the North Fork Koktuli River, and the fish populations affected, are discussed in this section.
We estimate that recovery of suitable structural habitat in the North Fork Koktuli River and off-channel
areas would likely take years to decades, given the scouring action of the flood wave and the volume of
fine-grain sediment that would potentially be delivered under the tailings dam failure. However, the
period for recovery could be substantially longer. Recovery of suitable gravel substrates and
development of channel morphology suitable for salmon habitat could be delayed even further if the
flood wave were to scour much of the North Fork Koktuli River valley to bedrock, which would then be
buried under massive deposits of tailings fines. Recruitment of gravels and coarser substrates to the
North Fork Koktuli River valley could be delayed by low supplies and/or low rates of transport from
tributaries or unaffected upstream sources. Recovery may also be delayed if riparian vegetation does
not recover because the tailings are toxic to plants. However, that causal pathway is not assessed.
The tailings dam failure simulation (Section 9.3) was restricted to approximately 3 0 km of the North
Fork Koktuli River, from the face of the TSF 1 dam downstream to the confluence of the South and North
Fork Koktuli Rivers. Extension of the simulation beyond this confluence would introduce significant
error and uncertainty associated with the contribution of South Fork Koktuli River flows, and would
require a more sophisticated sediment transport model. As a result, we were unable to quantify
sediment transport and deposition in the mainstem Koktuli, Mulchatna, and Nushagak Rivers. However,
given the high volume of tailings fines that could be transported beyond the confluence of the South and
North Fork Koktuli Rivers (Table 9-5), it is highly likely that impacts on fish habitat estimated for the
North Fork Koktuli River would extend for some significant distance down the mainstem Koktuli River.
We estimate that the combined effects of direct losses of habitat in the North Fork Koktuli River,
downstream in the mainstem Koktuli River and beyond, and impacts on macroinvertebrate prey for
salmon could adversely affect 25% or more of Chinook salmon returning to spawn in the Nushagak
River watershed. If the Koktuli River, Stuyahok, and Mulchatna portion of the Nushagak runs are
impacted via downstream transport of tailings fines, the TSF failure may affect nearly 60% of the
Chinook run (59% on average of the aerial survey counts were from these three watersheds, range 48%
to 75%). Uncertainty around this estimate is associated with the downstream extent of habitat impacts
(described above) and the variable and imprecise estimate of the relative abundance of Chinook salmon
in the Nushagak, Mulchatna, and Koktuli Rivers. While aerial counts can substantially underestimate
true abundance (Jones etal. 1998), we based our estimate of proportions on long-term (1969 to 1970
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~April 2013
-------
Chapter 9 Tailings Dam Failure
and 1974 to 1985) aerial counts of Chinook salmon collected and interpreted by ADF&G (Dye and
Schwanke 2009).
Because long-term abundance data are lacking for most other fish species and locations in the mine
scenario watersheds, losses caused by a tailings dam failure are not quantified for other species. Our
analysis focuses on a few endpoint species, and does not incorporate considerations of metacommunity
dynamics, which are poorly understood for the region yet may be critical to understanding species
responses to environmental change (Westley et al. 2010). Information documenting known occurrence
offish species in the region's rivers and major streams is available (Johnson and Blanche 2012, ADF&G
2012), but information on abundances, productivities, or limiting factors is not currently available.
9.5 Post-Spill Water Quality
9.5.1 Suspended Tailings Particles
9.5.1.1 Exposure
During a tailings dam failure, aquatic biota would be exposed to a slurry of suspended tailings moving at
up to 22 m/s (Table 9-4). Thirty km downstream, at the mouth of the North ForkKoktuli River (the limit
of the model), much of this material would still be flowing (Table 9-5).
For years after a tailings dam failure, settled tailings would be resuspended and carried downstream. At
first, this process would be frequent if not continuous (except when and where the substrate is frozen),
as a channel and floodplain structure is established by erosional processes resuspending the tailings
(Section 9.4.1). Gradually, as the tailings flow downstream, a substrate consisting of gravel and cobble
embedded in tailings fines would become established, and the flow velocities necessary to suspend
sediment would increase until they resembled those of an undisturbed stream.
Studies at other tailings-contaminated sites do not usefully address suspended tailings, as they have
been carried out long after the spills occurred, are based on events that differ from the one large spill
that would result from a tailings dam failure, and focus on toxic properties of the tailings
(Section 9.5.2.3).
9.5.1.2 Exposure-Response
Suspended sediment has a variety of effects on fish that are similar to effects of toxic chemicals. Like
chemical effects, the severity of effects increases with concentration and duration of exposure
(Newcombe and Jensen 1996). At low levels, suspended sediment causes physiological and behavioral
effects; at higher levels it causes death. Salmonids avoid turbid waters when possible, which may result
in loss or underutilization of traditional spawning habitats (Bisson and Bilby 1982, Newcombe and
Jensen 1996). However, salmonids must withstand brief periods of high suspended sediment
concentrations associated with spring floods (Rowe et al. 2003). Empirically derived effective exposures
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ~ April 2013
-------
Chapter 9 Tailings Dam Failure
for lethal and sublethal effects (i.e., reduced abundance or growth or delayed hatching) on juvenile and
adult salmonids may be summarized as follows (derived from Newcombe and Jensen 1996):
• 22,026mg/Lforlhour
• 2,981 mg/L for 3 hours
• 1,097 mg/L for 7 hours
• 148 mg/L for 1 to 2 days
• 55 mg/L for 6 days
• 7 mg/L for 2 weeks
• 3 mg/L for 7 weeks to 11 months
However, salmon may adapt to migrate through high levels of suspended sediment. During mid-May to
early August, when adult salmon migrate upstream through the lower Copper River (El Mejjati et al.
2010), suspended sediment concentrations range from 750 to 1780 mg/L (Brabets 1992).
9.5.1.3 Risk Characterization
During and immediately after a tailings spill, exposure to suspended sediment would be far higher than
any of the effects thresholds listed above. Fishes could be literally smothered and buried in the slurry.
Because the standard of 1,000 mg/L of suspended sediment is exceeded by ordinary events such as
erosion of construction sites and tilled fields, erosion of tailings from the re-formation of the channel
and floodplain would likely to exceed that standard for days at a time for a period of years. Fishes would
be likely to avoid these streams or experience lethality, reduced growth, or reduced abundance.
Avoidance could also block migrating salmon and other fish from their spawning areas in upstream
tributaries during these periods, but salmon have adapted to migration corridors with high suspended
sediment levels. The potential for tailings to be more aversive or toxic than natural suspended sediment
is unknown. Exposure levels would gradually decline over time as tailings are carried downstream,
channel stability increases, and the floodplain becomes revegetated. Rates of these processes are
unknown, but it is reasonable to assume that decades would be required for suspended sediment loads
in the Koktuli and Mulchatna Rivers to drop to levels that occur with normal high flows in stable
channels of the Bristol Bay watershed.
9.5.1.4 Uncertainties
There can be little doubt that, during and in the years immediately following a tailings dam failure,
suspended sediment concentrations would be sufficient to cause the loss offish populations for many
kilometers downstream of a failed tailings dam. A major uncertainty, however, is the number of years
required to reduce suspended sediment concentrations to levels that are not adverse. Another major
uncertainty is the downstream extent of the effects. The data and modeling effort required to determine
how far the initial slurry deposition would extend, how far re-suspended sediments would travel, and
how long erosional processes would continue were not feasible for this assessment.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ -o April 2013
-------
Chapter 9 Tailings Dam Failure
9.5.2 Tailings Constituents
Although the most dramatic effect of a tailings dam failure would be habitat destruction and
modification due to the flow of tailings slurry downstream, exposures to potentially toxic materials in
the slurry would also occur. The effects of a tailings dam failure can be assessed using the composition of
the tailings and of experimental tailings leachates, as well as experience with tailings spills at other sites.
Descriptions of these cases are presented in Box 9-6.
9.5.2.1 Exposure
Aqueous Exposures to Impoundment Waters
During a tailings dam failure, aquatic biota would be exposed to water that had been in contact with
tailings during processing and in the TSF. This water includes pore water associated with the deposited
tailings and water overlying the tailings. If the spill was caused by flow through a fault in the dam or by a
seismically induced tailings dam failure, undiluted pore water and supernatant water would be released.
If the dam was eroded or overtopped by a flooding event, as in the tailings dam failures evaluated in this
assessment, the pore and surface water could be diluted by fresh water. However, the dilution would be
trivial relative to the volume of pore water in the tailings.
A spill would have two phases in our tailings dam failure scenarios; other scenarios could differ in
timing and magnitude. At first, tailings slurry would pour through the breach for approximately 3 hours
based on the specified rates of dam erosion and slurry flow. Pore water would then drain from the
tailings that are not released but are above the elevation of the breach. The latter process is slow and
could continue until the dam was repaired. If a tailings dam failure occurred after the mine site was
abandoned, equilibrium would be achieved in which rain, snow, and upstream flows were balanced by
outflow of leachate through the breach.
Once in the stream, potentially toxic constituents dissolved in the water would not settle out. Because
the potentially toxic constituents are not degradable or volatile, they would eventually flow to Bristol
Bay, although they would be diluted along the way. In the failure of the Pebble 2.0 tailings dam (TSF 1),
the peak flow of spilled water at the bottom of the North Fork Koktuli River is estimated to be
7,169 m3/s (Table 9-4). The Nushagak River at Ekwok would be the first gaging station downstream
where most of the tailings would have settled out and dilution could be estimated. Using the annual
average and highest monthly average flows (668 and 1,215 m3/s, respectively), concentrations of
dissolved chemicals in the Nushagak River would be 91 and 86% of those in the spill. The highest
monthly average flow is a reasonable comparison. Because the hypothesized flood that would cause the
dam to fail is an intense local storm, a flood at the scale of the Nushagak River watershed is not implied,
but relatively high flows would occur. Minimum flow is not considered, because the scenario does not
occur in winter (although winter overtopping did occur at Nixon Fork Mine as a result of human error;
Box 8-1).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p 34 April 2013
-------
Chapter 9 Tailings Dam Failure
BOX 9 6. BACKGROUND ON RELEVANT ANALOGOUS TAILINGS SPILL SITES
Past deliberate or accidental spills of metal mine tailings into salmonid streams and rivers have occurred by
mechanisms and mining practices other than those evaluated in this assessment. However, they provide evidence
concerning the fate of tailings and the nature of exposures to aquatic biota once the tailings are in streams and
floodplains. In the United States, some of these sites are relatively well studied because the observed effects of
such spills have led to their classification as Superfund sites. Other tailings spills caused extensive fish kills and
other significant effects, but have not generated useful long-term monitoring data. These brief descriptions provide
background information and support the use of evidence from these cases in analyzing risks from potential tailings
dam failures in the Bristol Bay watershed.
Clark Fork River, Montana. The Clark Fork River Operable Unit of the Milltown Reservoir/Clark Fork River
Superfund Site includes 120 river miles (193 km), extending from the river's headwaters to the Milltown Reservoir,
just east of Missoula, Montana. Mining for gold, silver, copper, lead, and zinc began in the Clark Fork watershed in
the late 1800s. Most of the wastes released were tailings from copper mines in Butte and Anaconda, but aqueous
mine discharges and aerial smelter emissions also contributed wastes. Two sedimentation ponds were constructed
by 1918, with a third constructed by 1959. Mine water treatment was initiated between 1972 and 1975. By the
mid-1970s, waste inputs to the Clark Fork River were largely limited to movement of previously released solids. It
became a Superfund site in 1983. Contaminants of concern are arsenic, cadmium, copper, lead, and zinc, but
copper was the focus of assessment and planning because of its high toxicity.
The primary source of exposure is tailings deposited on the floodplains, resulting in aquatic pollution through
erosion and leaching. Large areas with acidic tailings (both acidic and neutral tailings were deposited) are barren of
plant life due to metal toxicity, which contributes to erosion and leaching. The river was fishless from the late
1800s to the 1950s, but has begun to recover. Trout and other fish continue to exhibit low growth and abundance,
and intermittent fish kills have followed metal pulses from rainstorms or rapid snow melt. However, sedimentation
was also thought to contribute to effects on fish populations through habitat degradation. Detailed information can
be found in the responsible party's remedial investigation (ARCO 1998) and in U.S. Environmental Protection
Agency (USEPA) documents (2012a).
Coeur d'Alene River, Idaho. The Coeur d'Alene River in northern Idaho flows from the Bitterroot Mountains to
Lake Coeur d'Alene. From the late 19th to late 20th century, the upper basin was mined for silver, lead, zinc, and
other metals, and much of the ore was smelted locally. Tailings were dumped into gullies, streams, and the river
until dams and tailings impoundments were built beginning in 1901. Plank tailings dams failed in the 1917 and
1933 floods; direct discharge of tailings did not end until 1968. According to USEPA's remedial investigation,
approximately 56 million metric tons (62 million tons) of tailings were discharged to the Coeur d'Alene River. In
1983, the area of the Bunker Hill smelter was added to the Superfund national priority list, and in 1998 the
contaminated river watershed, Lake Coeur d'Alene, and part of the Spokane River were explicitly included.
Metals concentrations above ambient water quality criteria, lethality in tests of ambient waters, and the absence of
some fish species from reaches with high metal concentrations were all attributed to leachates from tailings and
other mine wastes in floodplains and tributary watersheds. In addition, toxicity of bed sediments, which include
tailings, was found in the Coeur d'Alene and Spokane Rivers and tributaries. Aquatic effects were attributed
primarily to zinc, but cadmium, lead, and copper also reached toxic levels.
More detailed background information can be found in the Ecological Risk Assessment for the Coeur d'Alene Basin
Remedial Investigation/Feasibility Study (USEPA 2001), other USEPA documents (USEPA 2012b) and the National
Research Council's review of USEPA's assessment and management documents (NRC 2005).
Soda Butte Creek, Montana and Wyoming. The headwaters of Soda Butte Creek drain the New World mining
district in Montana before entering Yellowstone National Park. From 1870 to 1953, porphyry deposits were mined
for gold and copper with some arsenic, lead, silver, and zinc. In June 1950, the earthen tailings dam at the
McLaren mine failed, releasing approximately 41 million m3 of water and an unknown mass of tailings into Soda
Butte Creek (Marcus et al. 2011). In 1969, the creek was rerouted around the tailings pile, which was covered and
seeded. In 1989, a Superfund emergency response re-created and riprapped the creek channel to accommodate
a 100-year flood. Despite these actions, metal levels remain high in the creek and floodplain sediments and the
biota are impaired. The lack of any decrease in sediment copper despite floods in 1995, 1996, and 1997 and the
lack of macroinvertebrate recovery following remediation of acid drainage in 1992 indicate that the tailings are
persistent and are the primary cause of biological impairments. The primary sources of information on effects of
the tailings spill are academic studies (Nimmo et al. 1998, Marcus et al. 2001).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p 05 April 2013
-------
Chapter 9 Tailings Dam Failure
The composition of the aqueous phase is uncertain. None of the tests performed by PLP represents the
leaching conditions in a tailings impoundment, and no model exists to mathematically simulate the
leaching process. However, some mixture of tailings supernatant, which represents the source water for
the impoundment (Table 9-7); humidity cell leachate, which represents aqueous leaching from tailings
under oxidizing conditions (Table 9-8); and local water (Table 8-10) can be used to approximate
aqueous phase composition.
During mine operation, tailings impoundment surface waters would consist of water used to transport
the tailings (supernatant) and any other waters stored in the impoundments prior to reuse or treatment
and discharge. Hence, the surface water is expected to resemble the PLP's test supernatant (Table 9-7)
with some dilution by precipitation. However, those results do not include process chemicals that would
be associated with the supernatant and that are not quantified in this assessment, such as xanthates.
Supernatant water would be slightly diluted by rain and snow onto the surface of the impoundment, but
peripheral berms should prevent dilution by runoff except during flood events.
The waters released from a tailings spill during mine operation could consist of surface water, surficial
pore water, and a much larger volume of deep pore water. The surficial tailings pore water would be
generated by leaching tailings in the presence of some oxygen. The composition and concentrations of
constituents in that water may be roughly similar to a mixture of those observed in the supernatant and
humidity cell tests (Tables 9-7 and 9-8). Pore water from deep anoxic tailings would have begun
primarily as supernatant, but may have lower metal content due to chemical precipitation under anoxic
conditions. Leachate flowing from an abandoned and failed impoundment would be more oxidized
because the cover water and much of the pore water would have drained away.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p og April 2013
-------
Chapter 9
Tailings Dam Failure
Table 9 7. Aquatic toxicological screening of tailings supernatant against acute water quality
criteria (CMC) and chronic water quality criteria (CCC). Values are ng/L unless otherwise indicated.
Average leachate values are from Appendix H.
Analyte
pH (S.U.)
Alkalinity
(mg/L CaC03)
Hardness
(mg/L CaC03)
S04
Ag
Al
As
Ca
Cd
Co
Cr
Cua
Cub
Fe
Hg
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Tl
Zn
Sum of metals
Average Value
7.9
74.8
323
319,000
0. 018
71.8
17.2
116,000
<0.1
<0.1
<1.0
7.8
7.8
16.8
0.037
26,000
8,000
71.9
69.7
43,800
<0.8
0.2
6.0
7.6
0.022
4.3
-
CMC
-
-
24
750
340
-
6.3
89
1500
40
7.2
350
1.4
-
760
32,000
1300
220
14,400
-
316
-
CCC
-
-
-
87
150
-
0.55
2.5
190
24
4.4
0.77
-
693
72
140
8.8
1600
5
316
-
CMC Quotients
-
-
0.0007
0.096
0.051
-
<0.012
<0.001
<0.0007
0.19
1.1
0.048
0.026
-
0.095
0.0022
<0.0006
0.0010
0.0004
-
0.014
0.31a : 1.7"
CCC Quotients
-
-
-
0.82
0.11
-
<0.14
<0.04
<0.0051
0.32
1.8
0.048
-
0.10
0.97
<0.0056
0.026
0.0038
1.5
0.014
3.3a : 4.8b
Notes:
Blank values (-) indicate that criteria are not available.
"" Acute and chronic criteria from Alaska's hardness-based standard.
b Acute and chronic criteria from the national ambientwater quality criteria based on the biotic ligand model.
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-37
April 2013
-------
Chapter 9
Tailings Dam Failure
Table 9 8. Aquatic toxicological screening of tailings humidity cell leachates against acute water
quality criteria (CMC) and chronic water quality criteria (CCC).
Analyte
pH (S.U.)
Alkalinity (mg/LCaCOa)
Hardness (mg/LCaCOa)
Cl
F
S04
Ag
Al
As
B
Ba
Be
Bi
Ca
Cd
Co
Cr
Cua
Cub
Fe
Hg
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
Sn
Tl
V
Zn
Sum of metals
Average Value
7.8
59.7
66.8
515.5
450.9
17,448
0.01
23.64
5.46
10.67
9.25
0.20
0.49
22,551
0.05
0.19
0.50
5.33
5.33
29.66
0.01
4,015
2,547
44.15
33.46
2,099
0.54
0.06
1.80
1.48
2.93
0.05
0.78
3.16
-
CMC
-
-
-
1.6
750
340
29,000
46,000
-
1.4
89
410
9.2
4.8
350
1.4
760
32,000
330
41
14,400
-
-
83
-
CCC
6.5-9
-
-
-
i
87
150
1500
8900
-
0.19
2.5
53
6.4
3.0
0.77
693
72
37
1.6
1600
5
-
83
-
CMC
Quotients
-
-
-
0.0062
0.031
0.016
0.0004
0.0002
-
0.038
0.0021
0.0012
0.58
1.1
0.085
0.0071
0.058
0.0010
0.0016
0.0015
0.0001
-
-
0.038
0.72a: 1.2"
CCC Quotients
-
-
-
0.27
0.036
0.071
0.0010
-
0.28
0.076
0.0094
0.84
1.8
0.013
0.064
0.46
0.014
0.039
0.0011
0.30
-
0.038
2.3a: 3.3b
Notes:
Blank values (-) indicate that criteria are not available.
Values are presented in micrograms per liter (ug/L) unless indicated otherwise. Average leachate values are from Appendix H.
a Acute and chronic criteria from Alaska's hardness-based standard.
b Acute and chronic criteria from the national ambient water quality criteria based on the biotic ligand model.
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-38
April 2013
-------
Chapter 9 Tailings Dam Failure
Aqueous Exposures from Deposited Tailings
After a tailings dam failure, aquatic biota would be exposed to potentially toxic tailings that covered the
substrate of streams or rivers. Benthic organisms, or aquatic insects and other invertebrates that
burrow into or crawl upon substrates, would be the most exposed. In addition, eggs and larvae (fry) of
any salmon, trout, or char that spawned in the contaminated substrate would be exposed. In either case,
the bioavailable contaminants would be those that are dissolved in the pore water of the deposited
tailings. Hence, exposure is determined by the rate of leaching of the tailings and the rate of dilution of
the leachate, which depend on hydrological conditions. Unlike the lakes and estuaries that are the usual
sites of sediment pollution studies, streams have a high level of interaction between substrates and
surface waters. Shallow, turbulent water is typically near oxygen saturation. Bedload sediment bounces
and slides downstream during high flows, and during higher flows, sediment is suspended, exposing it to
oxygen. Water also flows longitudinally and laterally through bed and floodplain sediments and
vertically between groundwater and surface water.
Because the biologically active zone is oxidized, the tailings leachate to which biota would be exposed
could resemble leachates from the supernatants and humidity cells. Ideally, a leaching test would be
performed that simulated conditions in a streambed, but no such test results are available. In theory,
leachate composition could be estimated using a mechanistic model, but no such model is available.
Dilution of the leachate would be minimal in low-flow areas such as pools and backwaters and during
low-flow periods. Dilution would be greatest in high-flow and turbulent locations such as riffles, in
groundwater upwellings or downwellings, and during high-flow periods such as spring runoff and
floods. However, high flows would be expected to increase leaching rates, resulting in complex dynamics
(Nagorskietal. 2003).
Although we assume that spilled tailings would be mixed and would have average metal compositions
(Table 9-9), stream processes would be expected to sort them. In Soda Butte Creek (Box 9-6), copper
concentrations in riffles and glides gradually decreased downstream from the tailings spill site.
However, fine sediments in pools had higher copper concentrations than the high-energy segments, and
some of the highest copper concentrations were found in fine pool sediments more than 10 km
downstream (Nimmo etal. 1998).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p on April 2013
-------
Chapter 9
Tailings Dam Failure
Table 9 9. Comparison of mean metal concentrations of tailings (Appendix H) to threshold effect
concentration and probable effect concentration values for fresh water sediments and sums of the
quotients (ETU).
Tailings Constituents
Ag
As
Ba
Be
Bi
Cd
Co
Cr
Cu
Hg
Mn
Mo
Ni
Pb
Sb
Se
Tl
U
V
Zn
Sum
Mean
0.7
25.2
30.0
0.3
0.6
0.1
8.1
149.9
682.9
0.1
359.9
51.9
67.7
15.0
1.0
1.8
0.3
0.4
87.3
87.4
-
TEC"
-
9.8
-
0.99
-
43
32
0.18
630
23
36
-
-
-
120
-
TEC Quotient
-
2.6
-
0.10
-
3.5
21
0.56
0.57
2.9
0.41
-
-
-
0.72
32
PECa
-
33
-
5.0
-
110
150
1.1
1200
49
130
-
-
-
459
-
PEC Quotient
-
0.76
-
0.020
-
1.3
4.5
0.091
0.30
1.4
0.12
-
-
-
0.19
8.7
Notes:
Blank values (-) indicate that criteria are not available.
a TECsand PECs are consensus values from (MacDonald etal. 2000), except for Mn which are the TEL and PEL for Hyalella azteca 28-day
tests from (Ingersoll etal. 1996).
TEC = threshold effect concentration; PEC = probable effect concentration; TEL = threshold effect level; PEL = probable effect level.
All concentrations are mg/kg dry weight.
After a spill, aquatic biota would also be indirectly exposed to tailings deposited on land, primarily in the
floodplains. Erosion of these tailings would result in deposition in streams, potentially replacing tailings
lost through streambed erosion (Marcus etal. 2001). In addition, rain and snowmelt would run across
and percolate through tailings deposited on floodplains, leaching metals and carrying them into the
stream. Leachate would also form during lateral groundwater movement through tailings, particularly
where tailings deposited in wetlands. Floodplain-deposited tailings are leached in the presence of
oxygen with episodes of saturation and drainage (ARCO 1998). Hence, humidity cell leachates would be
more relevant to this exposure route than to others, and leachate concentrations in Table 9-8 may
roughly estimate leachate composition from floodplain-deposited tailings. This leachate could have
three fates: it could move upward during dry periods and deposit on the surface as soluble salts (e.g.,
hydrated metal sulfates); it could move down into buried soils and deposit as weak acid-extractable
compounds (e.g., metal sulfides); it could sorb to organic matter or it could move laterally to the surface
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-40
April 2013
-------
Chapter 9 Tailings Dam Failure
channel as dissolved metal ions (Nimik and Moore 1991, ARCO 1998). Runoff from tailings-
contaminated floodplains of the Clark Fork River had high copper levels (67.8 to 8,380 ug/L) (Nimik and
Moore 1991, ARCO 1998). Concentrations from spills in the Bristol Bay watershed would probably be
lower than for the acidic Clark Fork tailings and salt accumulation on the surface would be less because
of greater precipitation, but the same processes would occur. Dilution of the leachate that moves into
the stream would be highly location- and condition-specific. Once in a stream, leached metals are likely
to remain dissolved because of the highly dilute water chemistry in the region, but precipitation or
sorption to clays or organic matter would occur to some extent, depending on the conditions of the
event that moved the leachate into the stream.
Remobilization of deposited tailings during high flows could result in acute exposures to suspended
tailings and extend the downstream range of exposure to deposited tailings. In the Coeur d'Alene River,
floods occurring in 1995,1996, and 1997—more than 30 years after the last release of tailings—carried
metal-enriched sediment from both the floodplain and streambed more than 210 km downstream (the
furthest extent of the study) (USGS 2005).
Less dramatic increases in flow would cause bedload transport (movement of sediment without
suspension in the water column), which could release sediment pore water (leachate) into the water
column. First, copper could leach from the tailings and accumulate in sediment pore water during low-
flow periods. Then, when flows increase sufficiently to mobilize the sediment, pore water would mix
with surface water, resulting in exposure of aquatic biota and downstream copper transport. Studies in
the tailings-contaminated Clark Fork River found that copper concentrations in interstitial water were
3 to 36 ug/L in depositional areas and 3 to 22 ug/L in riffles (ARCO 1998). Concentrations would differ
with Bristol Bay tailings, but this result demonstrates that deposited tailings can have significant
interstitial water concentrations, even in a hydrologically active stream where leaching has proceeded
for decades. If sediment movement was sufficient to mobilize deep anoxic sediments, precipitated or
complexed metals may be mobilized and, depending on local water chemistry, dissolved.
Solid Phase Exposure to Deposited Tailings
Although the mostbioavailable metals in sediment are those dissolved in pore water, it is useful to
consider the whole sediment as a source of exposure. This approach avoids uncertainties associated
with using leaching tests to represent field processes. It is reasonable to consider the average tailings
composition to represent stream sediment to which biota downstream of a spill would be exposed
(Table 9-9). During and after the spill, there may be some sorting of the tailings by size or density that
would result in locally higher metal compositions, but that cannot be quantified. Although the material
in the failed dam would dilute the tailings initially, particles in the dam would be larger than the tailings
and would settle out in the first few kilometers downstream. Some soil would be scoured from the
receiving stream, but that would be associated with the first wave of the slurry. Hence, given the volume
spilled, tailings in most of the initial depositional area would be effectively undiluted. After the spill, the
tailings sediment would be diluted by clean sediment from tributaries, but that process would be slow
because the watershed is nearly undisturbed except for potential mine facilities, and the volume of
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment [T77April 2013
-------
Chapter 9 Tailings Dam Failure
tailings deposited in the watershed would be so large. The background sediment load (1.4 to 2.5 mg/L
total suspended solids (Table 3-4) is miniscule compared to the multiple meters of tailings that would be
deposited (Table 9-5). The washing of tailings from floodplains into streams and rivers would be more
important for many years, so the sediments in streams and rivers below a tailings spill would resemble
average tailings.
Dietary Exposures
As discussed in Section 8.2.2, dietary exposures offish to metals have been an issue of concern at mine
sites. An adjustment factor for rainbow trout to account for a dietary component to aqueous exposures
(0.95) is presented in that section. It may be applied to cases in which both direct aqueous and dietary
exposures may occur, such as flow into a stream through floodplain tailings or from upwelling through
tailings. Dietary exposures with respect to sediment levels may also be estimated. In such cases, direct
aqueous exposures offish maybe negligible, but invertebrates, particularly metal-tolerant insects such
as chironomids, may accumulate metals, carry them out of the sediment, and then serve as sources of
dietary exposure. This phenomenon has been documented in both the Clark Fork and Coeur d'Alene
River basins (Kemble et al. 1994, Farag et al. 1999).
A review of metal bioaccumulation by freshwater invertebrates (mostly Ephemeroptera and Diptera)
derived models for two relevant feeding guilds:
Collector-Gatherer Copper = 0.294x
Scraper-Grazer Copper = 1.73 x
where x is sediment concentration and copper is tissue concentration, both in ug/g dry weight
(Goodyear and McNeill 1999). Studies of the Soda Butte Creek tailings spill found that copper
concentrations in mixed invertebrates were slightly lower than sediment concentrations (Marcus et al.
2001). Studies of the Clark Fork River give bioaccumulation factors for copper and river invertebrates
ranging from 0.18 to 1.62, with factors generally rising as sediment concentrations declined (calculated
from Brumbaugh et al. 1994, Ingersoll et al. 1994). Equivalent studies in the Coeur d'Alene River give
very similar factors (0.15 to 1.77) (calculated from Farag et al. 1998). These results support the use of
the average bioaccumulation factor of 1.0 (Goodyear and McNeill 1999). This implies copper
concentrations in invertebrates are equal to those in sediments, which in this case are tailings with an
average copper concentration of 683 mg/kg (Table 9-9).
Another method used to estimate the bioaccumulation and toxicity of divalent metals in sediment is the
acid volatile sulfides (AVS)/simultaneously extracted metals (SEM) approach (Ankley et al. 1996).
However, this method requires measurements of SEM and AVS within the sediment of concern. The
source of copper in the tailings is sulfide ores, so one might assume that there is adequate sulfide for the
copper, but experience with tailings spills refutes that assumption. The availability process of concern is
oxidation of the sulfides, not binding of added copper by sediment sulfides. Studies in the Clark Fork
River found that, contrary to expectations of that model, invertebrates accumulated metals at locations
with AVS greater than SEM (Ingersoll et al. 1994). This discrepancy may be due to spatial variability, the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ . „ April 2013
-------
Chapter 9 Tailings Dam Failure
high oxidizing conditions in riffles where most invertebrates are found, and the fact that much of the
metals in these sediments are in a form (metal sulfide particles of the tailings) that is very different from
the lake and estuary sediments for which the model was developed. Hence, for practical and empirical
reasons, the AVS/SEM model is not appropriate to estimate bioaccumulation or toxicity in this system.
Persistence of Exposures
Evidence that tailings persist in streams as sources of metal exposures is provided by prior tailings
releases. A review by Miller (1997) found persistence of high metal content sediment in streams after
10 to 100 years. One well-documented case is provided by a tailings dam failure in Soda Butte Creek,
Montana, in 1950 (Box 9-6) (Marcus et al. 2001). Sediment was still characterized by high copper
concentrations after 48 years despite two 100-year floods, indicating that tailings are retained by
streams and maintain high metal levels even after decades of leaching. Similarly, the Coeur d'Alene River
basin was contaminated by direct discharge of tailings to floodplains, tailings dam failures, and mine
drainage, causing extensive damage to the watershed (Box 9-6) (NRC 2005). Treatment of the mine
drainage improved biotic communities, but they were still impaired, apparently as a result of metals
leaching from deposited tailings that entered the river until 1968 (Holland et al. 1994, NRC 2005). At
least as late as 2000, metal (cadmium, lead, and zinc) concentrations were elevated in caddisflies and
were more highly correlated with sediment concentrations than with surface water concentrations,
suggesting that deposited tailings were the primary source of exposure (Maretetal. 2003).
A new study has modeled future decline in sediment metals concentrations for the Clark Fork River
(Box 9-6), assuming an exponential decay in concentrations over time due to loss and dilution (Moore
and Langner 2012). Although there was no significant change over time (1991 to 2009) in downstream
concentration declines (which one would expect as tailings wash downstream), concentrations did
decline over time at three individual sites. Based on regression for each of those sites, Moore and
Langner (2012) estimated that average copper concentrations would decline below the probable effect
concentration (PEC) in less than 85 years. At the most contaminated of the three sites, copper is
predicted to reach the threshold effect concentration (TEC) in 163 years. In the Bristol Bay watershed,
dilution with clean sediment would likely be slowed by denser vegetation and less land disturbance.
Lower gradients in the Bristol Bay receiving streams relative to Montana would also tend to slow
recovery, as recovery is primarily achieved by tailings transport downstream. It should be noted that
these time estimates are not from the date of a spill, but rather from a date decades later, when channel
structure stabilized and much of the tailings had been carried downstream.
9.5.2.2 Exposure-Response
Aqueous Chemicals
The toxic effects of exposure to a tailings spill can be estimated from aquatic toxicity data. Ambient
water quality criteria and equivalent benchmarks are used to screen the metals in the two types of
tailings leachates (Tables 9-7 and 9-8). Copper is the dominant contaminant in tailings leachates, and
criteria values based on the biotic ligand model (BLM), described in Section 8.2.2.1 are used as
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ .„ April 2013
-------
Chapter 9
Tailings Dam Failure
benchmarks (Table 9-10). Acutely lethal levels for rainbow trout exposed to the humidity cell leachate
and supernatant are estimated to be 93 and 188 ug/L, respectively, based on the BLM.
Note that these criteria are calculated for the water chemistry of the supernatant and leachate. This is
clearly appropriate for the acute exposures immediately following a tailings dam failure, when the slurry
volume would greatly exceed natural flows. It would also be appropriate for situations like sediment
pore water, where dilution is minimal. However, for situations in which significant dilution occurs, these
criteria would be too high because uncontaminated water has lower hardness and lower levels of
chemicals that reduce copper binding to biotic ligands. Hence, dilution by a factor of 2 would not quite
reduce toxicity of metal-contaminated water by a factor of 2.
Table 9 10. Results of applying the biotic ligand model to mean water chemistries in tailings
leachates and supernatants to derive effluent specific copper criteria.
Stream
Acute Criterion
(CMC in
Chronic Criterion (CCC in
Tailings humidity cell leachates
4.8
3.0
Tailings supernatants
7.2
4.5
Notes:
CMC = criterion maximum concentration; CCC = criterion continuous concentration.
Source: USEPA2007.
Sediment Chemicals
The toxicity of settled tailings may also be estimated from tailings metal concentrations. Various
approaches have been employed to derive sediment quality guidelines, but the most common are the
threshold effect level (TEL) and the probable effect level (PEL). TELs and PELs have been used in
assessments of sites contaminated by mine wastes (USEPA 2001; USGS 2004, 2007). These levels are
derived from distributions of sediment concentrations that do or do not exhibit apparent toxicity in
laboratory or field studies. MacDonald et al. (2000) performed a meta-analysis of published values,
proposed consensus threshold effect concentrations (TECs) and probable effect concentrations (PECs),
and then tested them using additional sediment studies. One of the sites in the test data set was the
tailings-contaminated Clark Fork River. Out of 347 total sediments from 17 rivers and lakes, that
validation study found toxic effects in 17.7% of sediments with copper concentrations less than the TEC,
in 64% of sediments with copper concentrations between the TEC and PEC, and in 91.8% of sediments
with copper concentrations above the PEC (MacDonald et al. 2000). The consensus TECs and PECs are
used to evaluate the potential toxicity of tailings should they become sediment following a spill, because
they have the best scientific support.
Dietary Chemicals
Effects may also be estimated from dietary exposures. If the primary source of exposure is dissolved
copper in the water column (e.g., if significant upstream and floodplain leaching occurs), then the
0.95 adjustment factor (Section 8.2.2.1) is applicable. However, if sediment is the primary source of
exposure, a dietary value is needed for consumption of benthic invertebrates. A dietary chronic value for
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-44
April 2013
-------
Chapter 9 Tailings Dam Failure
rainbow trout derived from multiple studies is 646 ug/g (micrograms of copper per gram of dry diet)
(Borgmann etal. 2005), at which survival and growth are observed to decline in multiple studies.
Analogous Sites
The effects of exposure to leachate from tailings can also be estimated from effects at analogous sites. In
the Clark Fork River, Coeur d'Alene River, and Soda Butte Creek (Box 9-6), both field effects on fish and
invertebrates and toxicity have been associated with deposited tailings. However, the magnitude and
nature of effects are so site-specific that quantitative empirical exposure-response models from these
sites cannot be reasonably applied to the tailings dam failure analyzed here. Nevertheless, the
qualitative relationships are applicable.
9.5.2.3 Risk Characterization
Toxicological risks are usually judged by comparing exposure levels to a criterion or other
ecotoxicological benchmark using a risk quotient (Box 8-3). A risk quotient equals the exposure level
divided by the ecotoxicological benchmark. If the quotient exceeds 1, the effect implied by the
benchmark is expected to occur, but with some uncertainty (Box 8-3). Quotients much larger than 1
suggest larger effects than those that define the benchmark and there is greater confidence that an
adverse effect would occur. Quotients much smaller than 1 suggest that even small effects are unlikely.
The criterion maximum concentration (CMC or acute criterion) and criterion continuous concentration
(CCC or chronic criterion), and equivalent numbers when they are not available, are the primary
ecotoxicological benchmarks used in this assessment for aqueous exposures, because they are relatively
well accepted as thresholds for significant effects. The CMC estimates a concentration at which 5% of
aquatic species experience some mortality among developed life stages in short-term exposures. The
CCC estimates a concentration at which 5% of aquatic species experience decreased survival, growth, or
reproduction in longer-term exposures. Other benchmarks are used to indicate direct toxicity to
salmonids (Tables 8-13 and 8-14).
Acute Toxic Risks
At sites closest to a failed TSF, acutely toxic effects of a tailings spill would be indistinguishable from the
concurrent effects of being smothered by tailings particles. Aquatic life in the range of the tailings slurry
would be devastated by its physical effects. Dissolved components of the spill would continue to flow to
Bristol Bay, beyond the extent of significant particle deposition. Undiluted leachates of both types would
be expected to exceed the acute national criterion for copper, which suggests that they would kill
invertebrates (Tables 9-7 and 9-8). However, even the minimal dilution in the Nushagak River at Ekwok
would dilute leachate from the maximum spill to the national criterion or below. Even copper in
undiluted tailings leachates (5.3 and 7.8 ug/L for the humidity cell and supernatant, respectively) would
be well below levels required to kill post-larval trout in acute exposures (93 and 188 ug/L for the
humidity cell and supernatant, respectively). Hence, in the tailings dam failures, acute exposure to
dissolved copper in the near field would be sufficient to kill sensitive invertebrates but not salmonids,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ . g April 2013
-------
Chapter 9 Tailings Dam Failure
but those effects would be eclipsed by the physical effects. Downstream, where physical effects would be
minimal, toxic effects would be reduced or eliminated by dilution.
Chronic Toxic Risks for Aqueous Exposure
Potential effects of chemicals leaching from tailings in streambed and riverbed sediments and associated
floodplains are addressed by dividing the leachate concentrations by chronic water quality criteria and
standards to derive hazard quotients (exposure concentrations divided by effects concentrations).
These hazard quotients can be interpreted as relative degrees of toxicity of leachate constituents or as
indicators of the degree of dilution required to avoid significant toxic effects. The two estimates of
tailings leachate composition give similar results (Tables 9-7 and 9-8). Undiluted leachate of both types
would be expected to exceed the chronic national criterion, but not the Alaskan standard, for copper. If
combined toxic effects of metals are considered (see the Sum of Metals line in Tables 9-7 and 9-8),
chronic toxicity would be expected with both the hardness-based and BLM-based copper criteria, and
acute lethality would be expected with the BLM-based copper criterion. However, direct aqueous
exposures offish to copper are unlikely to be toxic unless concentrations in the actual field leachates are
much higher than the tailings test leachate concentrations.
The quotients with respect to chronic criteria (CCCs) imply that dilution by a factor of 2 to 4 would be
sufficient to render leachate nontoxic. Low dilutions would be expected in the years immediately after a
spill, when flows would pass through large volumes of tailings. After tailings have eroded and a more
normal channel and floodplain are established, low dilution of tailings could occur in sediments during
normal flows and in locations where water contaminated by floodplain tailings feeds a stream. In those
situations, sensitive invertebrates could be reduced or eliminated.
Chronic Toxic Risks from Sediment Chemicals
Sediment quality guidelines provide another line of evidence to assess risks from tailings after a tailings
dam failure. Table 9-9 shows that tailings would be expected to cause severe toxic effects on the
organisms that live in or on them. Notably, copper concentration would be 4.5 times the PEC; chromium
and nickel concentrations would also exceed their PECs. The sum of TEC quotients of 32 implies that
tailings would need to be diluted by 32 parts clean sediment to one part tailings before toxic effects
would be unlikely (below the TEC). Because the Bristol Bay watershed is relatively undisturbed,
background levels of total suspended solids are low (Table 8-10), so the time required to achieve that
degree of dilution would be very long.
Chronic Toxic Risks from Dietary Chemicals
The most relevant estimate offish dietary exposure to tailings is provided by bioaccumulation factors
with respect to sediment. The best estimate bioaccumulation factor of 1 implies copper concentrations
in invertebrates of 683 mg/kg (Section 8.2.2.1). Dividing this concentration by a consensus dietary
chronic value for rainbow trout of 646 ug/g (micrograms of copper per gram of dry diet) (Borgmann et
al. 2005) results in a quotient of 1.1. This implies thatthe undiluted tailings would produce toxic prey
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ . g April 2013
-------
Chapter 9 Tailings Dam Failure
for fish, but the result is marginal and certainly within the range of uncertainty. As discussed above,
dilution of tailings with clean sediment is likely to be a slow process.
Chronic Toxic Risks—Analogous Sites
Some well-documented cases indicate that adverse effects of chronic toxicity on aquatic communities in
general, and on salmonids in particular, can occur in streams and rivers that receive tailings spills. These
cases have shown that effects continue indefinitely, but that the nature and magnitude of those effects
vary among sites. In every case in which the ecological consequences of a major spill of metal ore tailings
to a stream or river was studied, extensive and long-lasting toxic effects were observed. However, these
tailings are likely to be more toxic than future tailings due to more efficient metal extraction in modern
ore processing.
The most relevant case appears to be Soda Butte Creek in Montana, where a tailings spill from a
porphyry gold and copper mine occurred in 1950 (Box 9-6). In the Soda Butte Creek case, the copper
content of macroinvertebrates was positively correlated (r2 = 0.80) and their taxa richness was
inversely correlated (r2 = 0.48) with sediment copper (Marcus etal. 2001). Although copper
concentrations generally decreased downstream, sediments and sediment pore waters were toxic to the
amphipod Hyalella azteca for the full 28-km length of the study area (Nimmo et al. 1998).
Macroinvertebrate community effects persisted for at least 40 years after the spill. These effects were
attributed to sediment toxicity (Nimmo et al. 1998), but habitat effects of deposited tailings also may
have contributed. Although they were less well studied, it was clear that trout were also affected. Only
two trout were found in the 300-m reach downstream of the spill site in 1993, although prior to mining,
Soda Butte Creek was known for "fast fishing and large trout" (Nimmo et al. 1998).
In the Coeur d'Alene River and its tributaries, elevated metals concentrations and effects on both benthic
invertebrates and fish persisted more than 30 years after tailings releases ended and after treatment of
mine drainage. Some fish species were absent; others were reduced in abundance and experienced toxic
effects from both aqueous and dietary exposures (Farag et al. 1999, Maret and Maccoy 2002, Maret et al.
2003). Returning Chinook salmon avoided the more contaminated South Fork in favor of the North Fork
(Goldstein etal. 1999). Macroinvertebrate communities and taxa were also impaired (Holland etal.
1994, Maret etal. 2003).
In the Clark Fork River, a sediment quality triad approach demonstrated that tailings-containing
sediments had high metal levels, were toxic to the amphipod Hyalella azteca, and shifted the
macroinvertebrate community to generally metal-tolerant Oligochaeta (worms) and Chironomidae
(midges) (Canfield et al. 1994). Rainbow and brown trout abundances were low in contaminated
reaches of the Clark Fork, fish kills occurred apparently due to metals washing from floodplain tailings
deposits, and metals in invertebrates were sufficient to cause toxic effects in laboratory tests of trout
(Kemble etal. 1994, Pascoe etal. 1994, ARCO 1998).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p .-, April 2013
-------
Chapter 9 Tailings Dam Failure
9.5.2.4 Uncertainties
All of the lines of evidence concerning risks to aquatic communities from the toxic properties of spilled
tailings have notable uncertainties.
Toxic Risks from Aqueous Exposures
The use of leachate and supernatant concentrations to estimate effects of a tailings spill is uncertain
primarily because of uncertainty concerning test relevance to leaching in the field. Leaching of tailings in
the impoundment, streambeds, and floodplains would occur under very different conditions than in
humidity cell tests. In addition, it is possible that tailings could become more acidic over time as their
acid-neutralizing capacity is consumed or as acid-neutralizing chemicals are dissolved, resulting in
increased metal concentrations. Test leachates are available for bulk tailings but not pyritic tailings.
Finally, the degree of leachate dilution in the field would be highly variable and could be roughly
estimated, at best. These uncertainties concerning exposure are significant in terms of both their
potential size (at least an order of magnitude uncertainty) and in terms of their implications (leachates
from the spilled tailings may be non-toxic or severely toxic given this uncertainty in exposure).
The exposure-response relationships for this line of evidence are also uncertain. As noted above
(Section 8.2.2), the water quality criteria and standards used in this assessment may not be protective of
all macroinvertebrate taxa that are important prey for fish. However, direct aqueous exposures of fish to
copper are unlikely to be toxic unless field concentrations are much higher than test leachate and
supernatant concentrations, so fish toxicity is not an important uncertainty. The uncertainty concerning
exposure-response relationships is smaller than the uncertainty concerning exposure.
Toxic Risks from Sediment
Although the consensus TECs and PECs are the best available effects benchmarks for sediment, their
applicability to tailings in streams and rivers of Bristol Bay is uncertain. The studies from which the
values are derived include lakes, reservoirs, and other systems that differ from the rivers and streams in
the Bristol Bay watershed. However, the Clark Fork River (a tailings-contaminated salmonid stream)
was one of the confirmation sites for the TECs and PECs, which suggests that they are relevant to this
type of situation.
Because the TECs and PECs are geometric means of prior sediment guidelines, the range of guidelines
provides an estimate of uncertainty. Alternate threshold values for copper range from 16 to 70 mg/kg
and probable effect values range from 86 to 390 mg/kg (MacDonald etal. 2000). The average copper
concentration of tailings (683 mg/kg) is well above all of these values, so this uncertainty is immaterial
(Table 9-9).
Some evidence suggests that these sediment guidelines may not be fully protective. When quotients of
sediment concentrations/TELs (one of the sources of the TECs and a numerically similar value) were
summed to address the combined toxicity of cadmium, copper, lead, and zinc, that value was not a
threshold for effects on stream invertebrates in the Colorado mining belt (Griffith et al. 2004), and
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ . „ April 2013
-------
Chapter 9 Tailings Dam Failure
reductions in four different community metrics occurred below the sum of TEL values. This result
suggests that toxicity would be even more severe than the TECs and PECs suggest, but it may be
somewhat confounded by mine drainage.
Dietary Risks
Effects of dietary exposures would depend on the tailings composition, the copper bioaccumulation
factor for aquatic invertebrates, and the chronic toxic threshold for dietary exposures of rainbow trout.
Tailings composition may differ in practice from the PLP's tests, but that uncertainty is unknown.
Ecological uncertainties are likely to be larger. Bioaccumulation factors for invertebrates and sediment
range from 0.15 to 1.77, even in a single river (above), which translates to invertebrate body burdens of
102 to 1,210 ug/g. That range encompasses the seven available estimates for the copper toxic dietary
threshold in rainbow trout, which range from 458 to 895 ug/g (Borgmann et al. 2005). This range of
bioaccumulation factors is not surprising given the differences in feeding habits, morphology, and
physiology among invertebrates.
Analogous Sites
The analogous sites for a potential tailings spill are all salmonid streams or rivers that received large
deposits of tailings from metal mines and that were well studied over an extended period (Box 9-6). A
large source of uncertainty when evaluating effects at those sites is the composition of the tailings. The
Pebble test tailings are, in general, less acidic and contain less copper. On that basis, the nature and
magnitude of effects are likely to be less. However, the setting is different in ways that might increase
effects. For example, low hardness and low levels of dissolved materials in the mainstem Koktuli River
receiving waters would make biota of the receiving streams more susceptible to metals than in the
analogous sites. Although these cases are highly uncertain sources of information concerning the
potential toxicity of spilled tailings, they can be used with confidence to identify or confirm important
modes of exposure and the processes leading to exposure. They also confidently demonstrate the
persistence of tailings and the leaching of their metals for multiple decades.
9.5.3 Weighing Lines of Evidence
This risk assessment is based on weighing multiple lines of evidence, and evidence for the various
routes of exposure is complex, as summarized in Table 9-11. For each route, sources of the exposure
estimate and the exposure-response relationship are indicated. All evidence is qualitatively weighed
based on three attributes: its logical implication, its strength, and its quality (Suter and Cormier 2011).
• The logical implication is the same for all lines of evidence: they all suggest that a spill from a
tailings dam failure would have adverse effects.
• The strength of the evidence is based primarily on the magnitudes of the hazard quotients. For
example, if the predicted concentration of copper is twice the median lethal concentration (LCso) for
rainbow trout, that is evidence of acute lethality, but if it is 10 times as high, that is stronger
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment „ . „ April 2013
-------
Chapter 9
Tailings Dam Failure
evidence. In Table 9-11, zero signifies a low quotient, + a moderate quotient, and ++ a high quotient
In this case, there are no moderate quotients.
• Quality is a complex concept that includes conventional data quality issues, but in this case, the
primary determinate is the relevance of the evidence to the mine scenario. Because this is a
predictive assessment, none of the evidence is based on observations of an actual spill. Hence, the
evidence is based on the tailings dam failure scenarios, laboratory studies, or field studies at other
sites where tailings have spilled into streams or rivers or where biota were exposed to other
sediments with high copper levels. Separate quality scores are provided for the exposure estimate
and for the exposure-response relationship.
The scores indicated in Table 9-11 are not a substitute for the actual evidence, but rather are intended to
remind the reader what evidence is available and summarize the pattern of strength and quality of the
several lines of evidence.
Table 9 11. Summary of evidence concerning risks to fish from a tailings dam failure. The risk
characterization is based on weighing multiple lines of evidence for different routes of exposure. All
evidence is qualitatively weighed (using one or more +, 0, symbols) on three attributes: logical
implication, strength, and quality. Here, all lines of evidence have the same logical implication that is,
all suggest a tailings dam failure would have adverse effects. Strength refers to the overall strength of
the line of evidence, and quality refers to the quality of the evidence sources in terms of data quality
and relevance of evidence to mine scenarios.
Route of Exposure
Source of Evidence (Exposure/ E-R)
Suspended sediment
Assumption/synthesis of laboratory and
field studies
Acute aqueous exposure
Leachate measurements/laboratory-
based criteria
Chronic aqueous exposure
Leachate measurements/laboratory-
based criteria
Chronic sediment exposure
Tailings measurements/sediment
guidelines
Chronic dietary exposure
Tailings measurements and
bioaccumulation factors/mean of
laboratory-based effects levels
All routes in the field
Exposure and effects at analogous sites
Summary Weight of Evidence
Logical
Implication
+
+
+
+
+
+
+
Strength
++
0
0
++
0
++
+
Quality
Exposure
0
0
0
+
+
+
0
E-R
+
+
+
0
0
0
Results
Adverse effects on fish are certain.
Although the exposure level is unknown,
it would clearly be at effective levels.
Acute lethality to invertebrates in near
field but not downstream and not to fish.
Chronic toxic effects on invertebrates
due to in situ leachate but would end
after some years if diluted by clean
sediment.
High likelihood of toxic effects on
invertebrates or fish based on a
summary of field studies.
Marginal dietary copper toxicity to trout
eliminated by minimal dilution.
Tailings spilled to streams have persisted
and caused severe effects, but the
toxicity of the tailings is likely to be higher
in those cases.
All lines of evidence are consistent with
toxic effects of tailings. Despite the
ambiguous quality and marginal strength
of some lines of evidence, the overall
strength is positive.
Notes:
E-R = exposure-response relationship.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
9-50
April 2013
-------
Chapter 9 Tailings Dam Failure
9.6 Summary of Risks
9.6.1 Tailings Spill
Following a tailings spill, fishes in the receiving stream and the invertebrates on which they depend
would be exposed to deposited tailings, suspended tailings, and tailings leachates. The fine texture of
deposited tailings would make them unsuitable for salmonid spawning and development, and a poor
substrate for the invertebrates that serve as food for developing salmon and resident trout and char.
Suspended tailings would have lethal and sublethal physical effects on fish and invertebrates
immediately following the spill, which are likely to continue with gradually diminishing intensity for
years thereafter. The most toxic constituent of the leachate and tailings would be copper, and exposures
would be both direct and through diet. Copper in leachate and in food is mildly toxic for fish, but copper
and other constituents in the tailings themselves would be moderately toxic to benthic invertebrates
and potentially toxic to fish eggs and larvae spawned in tailings-contaminated streams.
The physical and chemical effects of tailings on fish and invertebrates would be extensive in both space
and time. Elevated levels of suspended tailings would last for years. Deposited tailings and their leachate
would persist at toxic levels for decades. The acute effects of a tailings spill would extend far beyond the
modeled distance, which resulted in modeled tailings deposition of 3 to 14 m approximately 30 km
downstream (Section 9.3.1). Based on data from other sites, tailings deposition from a spill would
extend for more than 100 km downstream, resulting in chronic exposures and effects. From the
confluence of the South and North Fork Koktuli Rivers, the mouth of the Koktuli River is 63.6 km; from
there, the mouth of the Mulchatna River is another 66.5 km, and the mouth of the Nushagak River at
Dillingham is another 170.5 km so contamination of the entire downstream system would be likely.
We did not explicitly model failures of TSF 2 or TSF 3. The types of and risks of effects that would occur
with a failure of TSF 2 or TSF 3 would be generally similar to those described for a failure of TSF 1. The
content and toxicity of TSF 2 and 3 are assumed to be similar to that of TSF 1, and the magnitude and
extent of risks would be largely dependent on the volume of material released. One important
distinction, however, is recognized. The South Fork Koktuli River and Upper Talarik Creek are
hydrologically connected via groundwater transfer. In the event of a failure of TSF 3, transfer of
contaminated water leached from TSF tailings fines deposited in the South Fork Koktuli River valley to
the Upper Talarik Creek watershed and Iliamna Lake would be expected.
9.6.2 Remediation of a Tailings Spill
Although streams typically recover from aqueous effluents in less than a decade, the effects of tailings
deposition in streams and floodplains persist for as long as they have been monitored at analogous sites.
For that reason, many tailings-contaminated streams, rivers, and lakes in the United States have been or
will be dredged, riprapped, or redirected under the federal Superfund or state cleanup programs.
Although such remedial actions have net benefits, they create long-term impacts on aquatic habitats. For
example, riprapping reduces downstream exposure to tailings and associated metals by reducing
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTj April 2013
-------
Chapter 9 Tailings Dam Failure
erosion of floodplain tailings, but it also reduces habitat complexity and quality for fish by channelizing
the stream or river (Schmetterling et al. 2001).
Remediation in this case would be particularly difficult and damaging because the streams and their
floodplains are physically pristine and because a road would need to be built into a roadless area to
bring in equipment and haul out the tailings. At the upper end of the affected area, the process of
removing tailings would do little additional damage because the structure of the watershed would have
been destroyed. If tailings removal extended to streams that were not scoured in the initial release,
removal would destroy those streams and associated wetlands. If removal was not undertaken, the
substrate of the streams would still consist of tailings until high flows carried them downstream. It may
be technically impractical to recover tailings fines that have been transported past the point of
confluence with larger rivers.
Emergency plans for metal mines in Alaska that have been provided to USEPA do not address
remediation or restoration after a tailings spill has occurred. In fact, no tailings spill has been reported in
Alaska, so it is not clear what remediation or restoration might be required. Given the uncertain toxicity
of the tailings, the difficulty and expense of remediation and restoration, and the damage that would be
done by remediation, it is possible that a spill would be left to be restored by natural processes.
Remediation of prior tailings dam failures might serve as case studies, but, although failures are
numerous, the degree to which remediation results in restoration of natural resources has not been well
documented. The 1998 failure of the Aznalcollar Tailings Dam at Los Frailes Mine in Seville, Spain, has
been described as a case of substantial remediation (Section 9.3.3). The remediation was completed
within the 6 months between the April failure and the onset of heavy rains in October. However, the
Aznalcollar area, near Seville, has a much drier and warmer climate than the Bristol Bay area. It is also a
heavily farmed area with flatter topography, good access from existing roads, and more readily available
equipment and labor. Also, the potential releases from TSF1 are much larger than the release at
Aznalcollar. Finally, the goal of the remediation was restoration of land use, which would not be the
primary goal in the Bristol Bay watershed.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment p g» APril 2013
-------
10.1 Introduction
Because the Bristol Bay watershed is located in one of the last remaining virtually roadless areas in the
United States, development of any mine in the Bristol Bay watershed would require substantial
expansion and improvement of the region's transportation infrastructure. There are few existing
roadways, no improved federal or state highways, and no railroads, pipelines, or other major industrial
transportation infrastructure (Figure 6-6). As described in Section 6.1.4, the mine scenarios evaluated in
this assessment include a 138-km gravel surface, all-weather permanent access road (Figure 6-6)
connecting the mine site to a new deep-water port on Cook Inlet (Ghaffari et al. 2011). This length does
not include road sections within the mine site itself. Approximately 113 km of this corridor would fall
within the Kvichak River watershed.
The transportation corridor area considered in the assessment comprises 27 subwatersheds draining to
Iliamna Lake (Figure 2-7). These subwatersheds encompass approximately 1,760 km2 and contain
nearly 1,400 km of stream channels mapped for this analysis (see Box 3-1 for a description of these
methods). The seven largest subwatersheds are, from west to east, the headwaters of Upper Talarik
Creek, the headwaters of the Newhalen River, Chekok Creek, Canyon Creek, Knutson Creek, Pile River,
and the Iliamna River. The Newhalen River is the largest river crossed by the corridor, draining Sixmile
Lake and Lake Clark. Sockeye return to spawn in the Newhalen River and tributaries to Sixmile Lake and
Lake Clark. The transportation corridor would cross the Newhalen River and parallel the north shore of
Iliamna Lake (Figure 6-6). It would traverse rolling, glaciated terrain for approximately 60.5 road km
until reaching steeper hillsides northwest of the village of Pedro Bay and the shoreline of Knutson Bay.
After crossing gentler terrain around the northeast end of Iliamna Lake (Pedro Bay and Pile Bay), the
corridor would cross the Chigmit Mountains (the highest source of runoff in the Bristol Bay watershed)
along the route of the existing Pile Bay Road to tidewater at Williamsport. From there it would cross
Iliamna Bay and follow the coastline to the port site on Iniskin Bay, off Cook Inlet. Highly variable terrain
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-1
April 2013
-------
Chapter 10 Transportation Corridor
and variable subsurface soil conditions, including extensive areas of rock excavation in steep
mountainous terrain, are expected over this proposed route.
Although this route is not necessarily the only option for corridor placement, the assessment of potential
environmental risks would not be expected to change substantially with minor shifts in road alignment.
Along any feasible route, the proposed transportation corridor would cross many streams, rivers,
wetlands, and extensive areas with shallow groundwater, including numerous mapped (and likely more
unmapped) tributary streams to Iliamna Lake (Figures 10-1 and 10-2).
In this chapter, we consider the risks to fish habitats and populations associated with the transportation
corridor, as illustrated in a conceptual model showing potential linkages among the corridor, associated
stressor, and assessment endpoints (Figure 10-3). We begin with a discussion offish habitats and
populations along the corridor. We then consider potential impacts on these habitats and populations
resulting from its construction and operation. Although the transportation corridor would include the
road and adjacent pipelines (Section 6.1.4), we focus primarily on the road component; potential
pipeline failures are considered in Chapter 11.
Best management practices (BMPs) or mitigation measures would be used along the transportation
corridor to minimize potential risks to salmonids and the ecosystems that support them. Relevant BMPs,
and their likely effectiveness, are discussed in text boxes throughout the chapter.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ „ APril 2013
-------
Chapter 10
Transportation Corridor
Figure 10 1. Streams, wetlands, ponds, and lakes along the transportation corridor. Blue areas indicate streams and lakes from the National
Hydrography Dataset (USGS 2012) and wetlands from the National Wetlands Inventory (USFWS 2012).
WETLAND INVENTORY
DATA NOT AVAILABLE
Approximate Location of Pebble Deposit
Transportation Corridor
Transportation Corridor (Outside Assessment Area)
Transportation Corridor Area
Subwatersheds within Area
Rivers and Streams
Freshwater Wetland
Freshwater Pond
Lake
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-3
April 2013
-------
Chapter 10
Transportation Corridor
A, Newna en River
B. Knutson Bay
C. Iliamna River
"X" Approximate Location of Pebble Deposit
Transportation Corridor
Rivers and Streams
Freshwater Wetland
Freshwater Pond
Lake
Kilometers
4
(Miles
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-4
April 2013
-------
Chapter 10
Transportation Corridor
Figure 10 3. Conceptual model showing potential pathways linking the transportation corridor and related sources to stressors and assessment endpoints.
V
pipelines
v
pipeline breaks
or leaks
1,
See Lhoptfr 11
\/
1 material
transport
\/ \
introduction or eiimu
non-native species bocked
••
\
1 transportation corridor I
\
[culvert
1 •'-'<• pe
/ V
ated or t'flow
'.•etlands velocity
•
[4 production ancl export of
food & other resources
^ V
•T" competition A r i p ar i an
& preclation vegetation
LEGEND
f source j
\/ \
v
f stream \^r ( ^ \ -»j roadbed cuts, ]
1 crossings )<, ld 1 ->l tills S. ditches J
nr
V V
1 runoff 1 dust
rching i ^\ &. entrenchment 1 \/
J I J ( -\
4 riparian
1 vegetation I
-
? 7
1
\/ \/ V V V V V V
A^tr^an 4 flr'ridpNn ^suspended S ~-y 1 fin^ I'^d AdofT'n^trearn '^ater flofl|r "^> A clovvnstream -'^fr
geomorphologv conn^rtivitv sediment sediment temperatures
t chemical contaminants AmaimitiiHi> a ^ fir..;..
(metals, salts, other chemicals; frequency of high flovvs intemittency
v
prev , [t alteration of channel]
morphology & floodplain
/ V V V ( connectK*'' J v
•T inhibition of 1 .1 feeding 1 f |:hysblogical T aquatic habitat
fish passage 1 ability | stress fragmentation
V V V
4 rearing habitat 4 . spa-.vning habitat 4 overwintering habitat 4 incubation habitat
[additional step in | f additional step in ]
causalpathway J [ causal pathway1 j
proximate Abiotic
l>
V V
s*' 4 salmon "--. /• — ^ 1 other fish ^^~~x
\^^ (abundance, productivity or diversity! J x^^ (abundance, productivity or diversity) j/
^*~~~~\ — — , . — — — lT >S — . _ - — — ' —
t t
V 1 1
[J. marine-derived "s J. •ari"isif'5tpiri
nutrie nts | productivity
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-5
April 2013
-------
Chapter 10
Transportation Corridor
This page left intentionally blank
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ g April 2013
-------
Chapter 10 Transportation Corridor
10.2 Fish Habitats and Populations along the
Transportation Corridor
Streams along the transportation corridor have not been sampled as extensively as streams near the
Pebble deposit. Small to large rivers (2.8 m3/s mean annual flow and larger) that would be crossed by
the corridor provide spawning and rearing habitat, and are important routes for adult salmonid
migration to upstream spawning areas and juvenile salmonid migration downstream to Iliamna Lake
(Table 10-1). Large and small streams with low to moderate gradients (3% or less) provide important
high-quality spawning habitats, primarily for sockeye salmon. These streams also likely provide high-
quality seasonal and some year-round habitats for resident Dolly Varden and rainbow trout. Dolly
Varden are distributed across a much wider range of stream gradients (ADF&G 2012). The majority of
stream length in the subwatersheds intersected by the corridor consists of small headwater (62%) and
medium (27%) streams, whereas small and large rivers made up 9 and 3% of stream length,
respectively (Table 10-1). A majority (72%) of the stream length is classified as low to moderate
gradient (42% at less than 1% gradient, and 30% at 1 to 3% gradient) (see Box 3-1 for discussion on
how gradient was calculated). However, streams in the transportation corridor subwatersheds are
generally steeper and the extent of flatland in valley lowlands, or the extent to which these
subwatersheds are floodplain-prone, is less than in streams across the entire Nushagak and Kvichak
River watersheds (Figure 10-4). Floodplain-prone streams and rivers are more likely to be
unconstrained and to develop complex off-channel habitats and provide a diversity of channel habitat
types and hydraulic conditions that create favorable conditions, particularly for salmonid rearing.
However, the corridor streams are unique within the Nushagak and Kvichak River watersheds in that
many of them are short and originate from within the subwatersheds intersected by the corridor. In
addition, all of them flow into Iliamna Lake, which provides a high-quality habitat suitable for rearing
and migration among streams.
At the scale of the Nushagak and Kvichak River watersheds, 87% of stream length is classified as less
than 3% gradient (66% at less than 1% gradient and 21% at 1 to 3% gradient). In the subwatersheds of
the transportation corridor, 40% of total stream length is classified as floodplain prone, versus 64%
across the Nushagak and Kvichak River watersheds (Figure 10-4). These differences stem in large part
from the large portions of the unconfined, low-gradient lower Nushagak River watershed. Percent of
stream length less than 3% gradient is 76 and 92% in the Kvichak and Nushagak River watersheds,
respectively; the percent of stream length classified as floodplain prone is 53% across the Kvichak River
watershed and 69% across the Nushagak River watershed. Thus, stream characteristics in the
transportation corridor area are generally more similar to those in the Kvichak River watershed.
Characterization of stream segments for the entire Nushagak and Kvichak River watersheds, as well as
the methods used, are described in Chapter 3.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , » -, April 2013
-------
Chapter 10
Transportation Corridor
Table 10 1. Proportion of stream channel length in stream watersheds intersected by the
transportation corridor (Scale 5) classified according to stream size (based on mean annual
discharge in m3/s), channel gradient (%), and potential floodplain influence. Gray shading indicates
proportions greater than 5%; bold indicates proportions greater than 10%.
Mean annual discharge
Small headwater and Iliamna
Lake tributary streams3
Medium streams'5
Small rivers0
Large riversd
Gradient
<1%
FP
15%
11%
4%
3%
NFP
4%
3%
2%
0%
>1 % and <3 %
FP
5%
1%
1%
0%
NFP
14%
8%
1%
0%
>3 % and <8 %
FP
1%
0%
0%
0%
NFP
17%
4%
1%
0%
>8%
FP
0%
0%
0%
0%
NFP
6%
0%
0%
0%
Notes:
3 0-0.15 m3/s; headwater tributaries of streams crossing the transportation corridor and small streams flowing directly to Iliamna Lake (e.g.,
Eagle Bay and Chekok Creeks).
b 0.15-2.8 m3/s; upper reaches and larger tributaries of streams crossing the transportation corridor, and medium streams flowing directly
into Iliamna Lake (e.g., Chinkelyes and Knutson Creeks).
c 2.8-28 m3/s; middle to lower portions of the Iliamna and Pile Rivers.
d >28 m3/s; the Newhalen River.
FP = floodplain influence; NFP = no floodplain influence.
These low- to moderate-gradient streams provide important spawning habitat for sockeye. The Kvichak
River watershed includes over 100 separate sockeye salmon spawning locations (Demory et al. 1964,
Morstad 2003), including small tributary streams, rivers, mainland beaches, island beaches, and spring-
fed ponds. The spatial separation and diverse spawning habitat features within the watershed have
influenced genetic divergence among spawning populations of sockeye salmon at multiple spatial scales
(Gomez-Uchida etal. 2011). These distinct populations can occur atvery fine spatial scales. Sockeye
salmon that use spring-fed ponds and streams approximately 1 km apart exhibit differences in traits
such as spawn timing, spawn site fidelity, and productivity consistent with discrete populations (Quinn
etal. 2012).
Sockeye spawning has been observed at 30 locations along the transportation corridor (Demory et al.
1964). The Alaska Department of Fish and Game (ADF&G) has conducted aerial index counts of sockeye
salmon spawning abundance at these locations in most years since 1955 (Morstad 2003). Aerial survey-
based indices of sockeye salmon spawning abundance vary considerably. Sockeye spawners are most
abundant in the Iliamna River (averaging over 100,000 spawners), the Newhalen River (averaging over
80,000 spawners), and on beaches in Knutson Bay (averaging over 70,000 spawners) (Table 10-2,
Figure 10-5). In some years, these populations can be very large; for example, the 1960 survey for
Knutson Bay reported 1 million adults (Demory et al. 1964). Sockeye spawning is associated with
upwelling groundwater areas on beaches along the north and east shores of Knutson Bay, adjacent to
the transportation corridor. In addition, sockeye use of spring-fed ponds has been observed at eight
locations along the corridor. These locations tend to have fewer spawners (approximately 2,700 on
average), but fish using these locations may be adapted to the unique abiotic features of ponds (Quinn et
al.2012).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-8
April 2013
-------
Chapter 10
Transportation Corridor
Figure 10 4. Cumulative frequency of stream channel length classified by mean annual flow (m3/s),
reach gradient (%), and f loodplain potential (measured as % f latland in lowland) for watersheds
intersected by the transportation corridor (Scale 5) versus the Nushagak and Kvichak River
watersheds (Scale 2).
100%
%
E
fu
_ro
|
U
Scale2
MAP Classification
10
20 30
Mean Annual Flow (m5/s)
40
50
Scalc2
Gradient Classification
0%
4%
6%
8% 10% 12%
Reach Gradient
14%
16% 18%
20%
Scale2
5% Classification
0%
0%
20%
40% 60%
Floodplain Potential
80%
100%
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-9
April 2013
-------
Chapter 10
Transportation Corridor
Table 10 2. Average number of spawning adult sockeye salmon at locations near the transportation corridor.
Map
Point
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Area
Upper Talarik
Newhalen River System
Newhalen River System
Newhalen River System
Newhalen River System
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
North East
Iliamna River System
Iliamna River System
Iliamna River System
Iliamna River System
Iliamna River System
Area Name
Upper Talarik Creek
Newhalen River
Little Bear Creek/Ponds
Alexi Creek
Alexi Lakes
Roadhouse Creek
N.W. Eagle Bay Creek
N.E. Eagle Bay Creek/Ponds
NE Eagle BayCr. Ponds
Youngs Creek
Chekok Creek/Ponds
Tomkok Creek
Canyon Creek
Wolf Creek Ponds
Mink Creek
Canyon Springs
Prince Creek Ponds
Knutson Bay
Knutson Creek
Knutson Ponds
Pedro Creek & Ponds
Russian Creek
Lonesome Bay Creek
Pile River
Swamp Creek
Iliamna River
Bear Creek & Ponds
False Creek
Old Williams Creek
Chinkelyes Creek
Type
Stream
River
Ponds
Stream
Lake
Stream
Stream
Stream
Ponds
Stream
Stream
Stream
Stream
Ponds
Stream
Ponds
Ponds
Lake
Stream
Ponds
Ponds
Stream
Stream
River
Stream
River
Ponds
Stream
Stream
Stream
Average Number of Sockeye
Salmon Spawners (1955-2011)
7,021
84,933
527
1,176
7,121
1,052
1,649
3,416
4,766
3,532
1,840
10,882
8,015
4,469
1,144
884
3,797
72,845
1,548
1,200
4,259
2,263
1,026
6,431
1,091
101,306
1,748
1,317
3,726
9,128
Number of Years Spawners
were Counted (Max = 57)
49
34
20
27
33
28
32
38
5
38
32
38
38
26
35
20
34
47
41
39
48
17
6
38
18
53
30
21
27
46
Range
0-70,600
97-730,900
0-1,860
0-13,200
11-38,000
0-4,950
0-17,562
0-18,175
200-11,700
0-26,500
0-8,700
300-56,600
200-48,000
0-28,000
0-6,000
0-5,000
5-34,800
1,000-1,000,000
1-6,600
0-6,350
0-38,150
0-20,000
32-2,675
0-39,200
25-7,700
3,000-399,300
40-10,300
0-13,300
0-38,000
50-44,905
Notes:
Locations are organized from west to east along the corridor.
Sources: Morstad 2003, Morstad pers. comm.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-10
April 2013
-------
Chapter 10
Transportation Corridor
Figure 10 5. Location of sockeye salmon surveys and number of spawners observed along the transportation corridor. Numbers refer t<
map points listed in Table 10 2.
N
A
10
Lake Clark
Average Number of Sockeye Spawners
] Kilometers
10
] Miles
< 1,000
21,000 to <2,000
>2,000 to <10,000
>10,000to <50,000
>50,000
W Approximate Location of Pebble Deposit
^^^ Transportation Corridor
• • m Transportation Corridor (Outside Assessment Area)
| Transportation Corridor Area
Sub watersheds within Area
Cook Inlet
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-11
April 2013
-------
Chapter 10 Transportation Corridor
Less is known about the occurrence or abundance of other salmon species in streams and rivers
crossing or adjacent to the transportation corridor. Chinook, coho, pink, and chum salmon are present in
the Kvichak River watershed, but data for spatial occurrence are for isolated points in the system
(Johnson and Blanche 2012). In streams intersected by the transportation corridor, sockeye salmon are
in all streams included in the Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes—Southwestern Region (also known as the Anadromous Waters Catalog [AWC])
(Johnson and Blanche 2012) (Figure 10-6). Working from west to east along the corridor, streams with
salmon species in addition to sockeye are as follows: Upper Talarik Creek (Chinook, coho, chum, and
pink salmon), Newhalen River (Chinook and coho salmon), Youngs Creek (East and West Branches),
Chekok and Tomkok Creeks (coho salmon), Swamp Creek (a tributary to Pile Bay) (Chinook salmon),
and Iliamna River (Chinook, coho, chum, and pink salmon).
Dolly Varden and rainbow trout distribution has not been surveyed as extensively as salmon
distribution along the transportation corridor (ADF&G 2012). Dolly Varden have been documented in
nearly every sockeye salmon-bearing stream that would be crossed by or adjacent to the corridor, as
well as in locations upstream of reported anadromous salmon use (Figure 10-6). Rainbow trout
presence along the corridor is reported for only a few streams, including Upper Talarik Creek, the
Newhalen River, an unnamed tributary to Eagle Bay, Youngs Creek, Tomkok Creek, and Swamp Creek
(ADF&G 2012). Rainbow trout have also been documented in Chinkelyes Creek (Berejikian 1992).
The distribution of both Dolly Varden and rainbow trout along the transportation corridor is likely much
more extensive than reported in the Alaska Freshwater Fish Inventory (AFFI) resident fish database,
which does not account for seasonal movements and low sampling effort. Sockeye salmon provide an
important food subsidy to Dolly Varden and rainbow trout. For example, Denton et al. (2009) reported
Dolly Varden movement into multiple ponds used by spawning sockeye next to the Pedro Bay village, to
feed on sockeye salmon fry, eggs, and carcass-associated blowflies. Information on rainbow trout
movement between Iliamna Lake and streams intersected by the corridor is not available, but these
movements are likely to occur. Movements between lakes and tributary streams in response to feeding
and spawning opportunities have been documented elsewhere in Iliamna Lake (Russell 1977) the
Alagnak River system (Meka et al. 2003), and in the Wood River lake system (Ruff et al. 2011).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ , „ APril 2013
-------
Chapter 10
Transportation Corridor
Figure 10 6. Reported salmon, Dolly Varden, and rainbow trout distribution along the transportation corridor. Salmon presence data from
the Anadromous Waters Catalog (AWC) (Johnson and Blanche 2012); Dolly Varden and rainbow trout presence data from the Alaska
Freshwater Fish Inventory (AFFI) (ADF&G 2012).
IN
A
10
] Kilometers
10
1 Miles
Lake Clark
Note: Sampling intensity is greatly reduced away from the Pebble deposit area.
Streams without data points may not have been surveyed; thus, it is unknown
whether or not they provide suitable habitat for this species.
^ Approximate Location of Pebble Deposit
• • • • Transportation Corridor (Outside Assessment Area)
* Dolly Varden (AFFI)
• Rainbow Trout (AFFI)
Transportation Corridor
Transportation Corridor Area
Subwatersheds within Area
Salmon (AWC)
Cook Inlet
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-13
April 2013
-------
Chapter 10 Transportation Corridor
10.3 Potential Risks to Fish Habitats and Populations
Only rarely has it been possible to build roads that have no negative effects on streams (Furniss et al.
1991). Roads modify natural drainage networks and accelerate erosion processes, which can lead to
changes in streamflow regimes, sediment transport and storage, channel bank and bed configurations,
substrate composition, and the stability of slopes adjacent to streams. Road construction can increase
the frequency of slope failures by orders of magnitude, depending on variables such as soil type, slope
steepness, bedrock type and structure, and presence of subsurface water. These slope failures can result
in episodic sediment delivery to streams and rivers, potentially for decades after the roads are built
(Furniss et al. 1991). All of these potential changes can have important biological consequences for
anadromous and resident fishes by negatively affecting food, shelter, spawning habitat, water quality,
and access for upstream and downstream migration (Appendix G) (Furniss et al. 1991).
In the Bristol Bay region, risks to fish from construction and operation of the transportation corridor are
complex and potentially significant largely because of hydrological issues. Field observations in the mine
area (Hamilton 2007, Woody and O'Neal 2010) are descriptive of terrain with abundant near-surface
groundwater and a high incidence of seeps and springs associated with complex glaciolacustrine,
alluvial, and slope till deposits (Appendix G). The abundance of mapped wetlands further demonstrates
the pervasiveness of shallow subsurface flows and high connectivity between groundwater and surface
water systems in the areas traversed by the transportation corridor (Appendix G). As noted in Section
3.3, the strong connection between groundwater and surface waters helps to moderate water
temperatures and streamflows, and this moderation can be critical for fish populations. The
construction and operation of the transportation corridor could fundamentally alter connections
between shallow aquifers and surface channels and ponds by intercepting shallow groundwater
flowpaths, leading to impacts on surface water hydrology, water quality, and fish habitat (Darnell et al.
1976, Stanford and Ward 1993, Forman and Alexander 1998, Hancock 2002).
The lengths of the transportation corridor and their proximities to National Hydrography Dataset (NHD)
streams (USGS 2012) and National Wetlands Inventory (NWI) wetlands (USFWS 2012) are shown in
Tables 10-3 and 10-4, respectively (see Box 10-1 for a description of methods used to estimate these
values). In sum, the length of road within 200 m of NHD streams or NWI wetlands would be
approximately 67 km (Table 10-5). These lengths do not encompass the section of corridor outside of
the Kvichak River watershed (i.e., watersheds flowing into Cook Inlet). The 200-m road buffer was
derived from an estimate of the road-effect zone for secondary roads (Forman 2000). The largest impact
on sockeye salmon would likely occur where the road would run parallel to the Iliamna River and
Chinkelyes Creek, sites at which many sockeye salmon spawn (Figure 10-2, Inset C). Other high-impact
areas include where the road would run parallel to Knutson Bay, intersecting many small streams and
where groundwater supports spawning for hundreds of thousands of salmon (Figure 10-2, Inset B), and
where the road crosses wetlands north of Iliamna Lake (Figure 10-2, Inset A).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~14 April 2013
-------
Chapter 10 Transportation Corridor
In the following sections, we consider potential risks to fish habitats and populations resulting from
construction and operation of the transportation corridor. We focus on risks related to filling and
alteration of wetlands, stream crossings, fine sediments, dust deposition, runoff contaminants, and
invasive species.
BOX 10 1. CALCULATION OF STREAM LENGTHS AND WETLAND AREAS AFFECTED BY
TRANSPORTATION CORRIDOR DEVELOPMENT
The National Hydrography Dataset(NHD) (USGS 2012), Alaska Anadromous Waters Catalog (AWC) (Johnson
and Blanche 2012), and Alaska Freshwater Fish Inventory (AFFI) (ADF&G 2012) were used to evaluate the
effects of the transportation corridor on hydrologic features and fish populations.
The length of stream downstream of each crossingwas estimated from the NHD flowlines. Stream length by
subwatershed, based on 12-digit hydrologic unit codes, was calculated as the total distance from each
crossing to Iliamna Lake. In the multiple instances where stream crossings were tributaries to a single main
channel, the mainstem length was only counted once (Table 10-3). Downstream length reported in Table 10-
6 includes mainstem length downstream of tributary crossings. In cases that the corridor crossed tributaries
of a mainstem channel the mainstem length is included in both crossings.
Mean annual flow of NHD streams upstream of the transportation corridor was estimated using methods
described in Box 3-3.
The valley gradient of NHD stream segments intersected by the corridor and upstream of the corridor was
estimated using a 30-m National Elevation Dataset digital elevation model (DEM) (Gesch 2007). A drainage
network was developed from a flow analysis using the DEM and slope estimated using this drainage
network. The DEM based drainage network paralleled the NHD stream flowlines and therefore, usingthe
toolset in the spatial analyst extension in ArcGIS, slope from the drainage network was transferred to NHD
reach segments. A 12% slope was used to calculate stream length likely to support fish (Table 10-6). Stream
length upstream of the corridor that was less than 12% was based on the NHD stream length to the first
instance that slope was greater than 12%. The analysis of upstream fish habitat was extended to include
streams in subwatersheds in the Headwaters Newhalen River, Tomkok Creek, Pile River, and Iliamna River.
For the analysis of road length intersecting and within 100 m or 200 m of either a stream or wetland
(Tables 10-3, 10-4, and 10-5), each stream (NHD) or wetland (NWI) was buffered to a distance of 100 m
and 200 m and the length of corridor within these ranges was summed. Similarly, for the area of wetlands
within 100 m and 200 m of the road corridor, the road corridor was buffered and the area of wetlands within
that buffered area was summed across the length of road. For the area of wetlands directly filled by the road
corridor, a road width of 9.1 m was used.
The characterization of both stream length and wetland area affected is likely a conservative estimate. The
NHD may not capture all stream courses and may underestimate channel sinuosity resulting in
underestimates of affected stream length. Additionally, the AWC and the AFFI do not necessarily
characterize all potential fish-bearing streams because of limited sampling along the corridor. The
characterization of wetland area is limited by the resolution of the available NWI data product. In this
analysis, the transportation corridor often bisects wetland features and the wetland area falling outside the
200-m boundary was assumed to maintain its functionality. We were also unable to determine the effect
that the transportation corridor may have on wetlands that had no direct surface connection, but may be
connected via groundwater pathways. Together, these limitations likely make our calculations an
underestimate of the effect that development of the transportation corridor would have on hydrologic
features in this region. These estimates could be enhanced with improved, higher-resolution mapping,
increased sampling of possible fish-bearing waters, and ground-truthingof surface-water and groundwater
connections.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ . c April 2013
-------
Chapter 10
Transportation Corridor
Table 10 3. Proximity of transportation corridor to National Hydrography Dataset streams.
HUC-12 Name or Description
Headwater, Upper Talarik Creek
Upper tributary stream to Upper Talarik Creek
Tributary to Newhalen River portion of corridor
Headwaters, Newhalen River
Outlet, Newhalen River
Roadhouse Creek
Iliamna Lake
Eagle Bay Creek
Youngs Creek Mainstem (Roadhouse Mountain HUC)
Youngs Creek East Branch
Chekok Creek
Canyon Creek
Knutson Creek
Outlet, Pile River
Middle Iliamna River
Chinkelyes Creek
HUC-12 Digit
190302060702
190302060701
190302051404
190302051405
190302051406
190302060907
190302060914
190302060905
190302060903
190302060904
190302060302
190302060902
190302060901
190302060104
190302060205
190302060206
Total length across all HUCs
Percentage across all HUCs
Proximity to Streams
Not nearby
(km)
5.4
4.3
7.8
2.6
4.2
0.8
29.3
3.1
3.0
1.4
1.8
1.1
1.2
2.1
4.5
9.6
82.1
72.9%
<100 m
(km)
0.8
0.2
1.9
0.4
1.5
1.2
4.3
0.5
0.1
1.0
0.3
0.1
0.3
0.6
1.1
0.8
15.3
13.6%
100-200 m
(km)
1.2
0.1
1.2
0.4
0.8
1.3
4.1
0.8
0.2
0.6
0.3
0.2
0.4
0.7
0.7
2.1
15.2
13.5%
Total
(km)
7.4
4.6
10.9
3.4
6.5
3.3
37.7
4.4
3.4
3.0
2.5
1.4
2.0
3.4
6.4
12.5
113
100%
Notes:
HUC = hydrologic unit code.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-16
April 2013
-------
Chapter 10
Transportation Corridor
Table 10 4. Proximity of transportation corridor to National Wetlands Inventory wetlands.
HUC-12 Name or Description
Headwater, Upper Talarik Creek
Upper tributary stream to Upper Talarik Creek
Tributary to Newhalen River portion upstream of corridor
Headwaters, Newhalen River
Outlet, Newhalen River
Roadhouse Creek
Iliamna Lake
Eagle Bay Creek
Youngs Creek Mainstem (Roadhouse Mountain HUC)
Youngs Creek East Branch
Chekok Creek
Canyon Creek
Knutson Creek
Outlet, Pile River
Middle Iliamna River
Chinkelyes Creek
HUC-12 Digit
190302060702
190302060701
190302051404
190302051405
190302051406
190302060907
190302060914
190302060905
190302060903
190302060904
190302060302
190302060902
190302060901
190302060104
190302060205
190302060206
Total length across all HUCs
Percentage across all HUCs
Proximity to Wetlands
Not nearby
(km)
0.2
1.7
4.0
2.3
1.1
0.7
28.3
1.3
0.9
0.3
1.8
0.8
1.0
0.3
2.7
7.7
55.0
48.8%
Intersects
(km)
1.9
0.3
0.4
0.1
2.4
0.3
1.8
0.7
0.2
0.5
0.2
0.0
0.1
1.2
0.6
1.4
12.2
10.8%
<100 m
(km)
4.0
1.4
3.9
0.4
1.7
1.8
3.9
1.7
1.1
0.8
0.3
0.2
0.6
1.5
1.7
1.9
27.0
23.9%
100-200 m
(km)
1.2
1.2
2.6
0.5
1.4
0.5
3.7
0.8
1.2
1.5
0.2
0.3
0.3
0.5
1.3
1.5
18.5
16.4%
Total
(km)
7.4
4.6
10.9
3.4
6.5
3.3
37.7
4.4
3.4
3.0
2.5
1.4
2.0
3.4
6.4
12.5
113
100%
Notes:
HUC = hydrologic unit code.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-17
April 2013
-------
Chapter 10
Transportation Corridor
Table 10 5. Proximity of transportation corridor to water (within 200 m of National Hydrology Dataset streams or National Wetland
Inventory wetlands).
HUC-12 Name or Description
Headwater, Upper Talarik Creek
Upper tributary stream to Upper Talarik Creek
Tributary to Newhalen River portion upstream of corridor
Headwaters, Newhalen River
Outlet, Newhalen River
Roadhouse Creek
Iliamna Lake
Eagle Bay Creek
Youngs Creek Mainstem (Roadhouse Mountain HUC)
Youngs Creek East Branch
Chekok Creek
Canyon Creek
Knutson Creek
Outlet, Pile River
Middle Iliamna River
Chinkelyes Creek
HUC-12 Digit
190302060702
190302060701
190302051404
190302051405
190302051406
190302060907
190302060914
190302060905
190302060903
190302060904
190302060302
190302060902
190302060901
190302060104
190302060205
190302060206
Total length across all HUCs
Percentage across all HUCs
Proximity to Streams Or Wetlands
Not nearby
(km)
0.1
1.5
3.8
2.3
1.1
0.0
22.1
0.9
0.7
0.3
1.5
0.8
0.7
0.3
1.9
7.3
45.4
40.3%
Within 200 m
(km)
7.3
3.1
7.0
1.1
5.4
3.3
15.5
3.5
2.7
2.8
1.0
0.5
1.2
3.1
4.5
5.2
67.3"
59.7%
Total
(km)
7.4
4.6
10.9
3.4
6.5
3.3
37.7
4.4
3.4
3.0
2.5
1.4
2.0
3.4
6.4
12.5
113
100%
Notes:
HUC = hydrologic unit code.
3 Reported length is the sum of the road length within 200 m of a National Hydrology Dataset stream or National Wetland Inventory wetland reported in Tables 10-3 and 10-4, respectively. In cases
where the same section of road is near both types of water bodies, section is only reported once. Therefore total length is less than sum of lengths in Tables 10-3 and 10-4.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-18
April 2013
-------
Chapter 10 Transportation Corridor
10.3.1 Wetland Filling and Alteration
10.3.1.1 Exposure
Approximately 11% (12 km) of the transportation corridor would intersect mapped wetlands, an
additional 24% (27 km) would be located within 100 m of wetlands, and another 16% (19 km) would be
located within 100 to 200 m of wetlands (Table 10-4). In total, approximately 51% (58 km) of the
corridor would fill or otherwise alter wetlands. Wetlands constitute nearly 11% of the total area within
200 m of the transportation corridor. Wetland area within 100 m and 200 m of the corridor would be
2.3 km2 and 4.6 km2, respectively. These areas do not include wetlands that would be covered by the
mine footprint under the mine scenarios (Chapter 7). The area of wetlands filled by the roadbed would
be 0.11 km2 (i.e., approximately 12 km of road, assuming a road width of 9 m).
10.3.1.2 Exposure-Response
The distribution of salmonids in wetlands along the transportation corridor is not known. However,
wetland loss can result in the loss of resting habitat for adult salmonids and of spawning and rearing
habitat in ponds and riparian side channels. Wetlands can provide refuge habitats (Brown and Hartman
1988) and important rearing habitats for juvenile salmonids by providing hydraulically and thermally
diverse conditions. In addition, by damming and diverting surface flow and inhibiting subsurface flow,
road construction could block or limit access by fish to important habitats. Beaver ponds associated with
small streams and wetlands can be important winter refugia for coho salmon (Nickelson et al. 1992,
Cunjak 1996). Beaver ponds provide high-quality habitat for salmon rearing, because they provide
macrophyte cover, low-flow velocity, and increased temperatures and trap organic materials and
nutrients (Nickelson etal. 1992, Collen and Gibson 2001, Langetal. 2006).
Wetlands can also provide enhanced foraging opportunities (Sommer et al. 2001). Floodplain wetlands
can be an important contribution to the abundance and diversity of food (and food webs) upon which
salmon depend (Opperman et al. 2010). Within wetlands that are not blocked and are still accessible, the
road prism could alter hydrology and flow pathways from wetlands to the stream network. These
alterations could mobilize minerals and stored organic carbon, and expose soils to new wetting and
drying and leaching regimes, and thus lead to changes in vegetation, nutrient and salt concentrations,
and water quality (Ehrenfeld and Schneider 1991). These changes in wetland dynamics and structure
could affect the availability of wetlands to fish and the contribution from headland wetlands of nutrients,
organic material, and a diverse array of macroinvertebrates to higher order streams in the watershed
(i.e., streams receiving wetland drainage) and downstream waters (King et al. 2012, Shaftel et al. 2011,
Dekar et al. 2012, Walker et al. 2012).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ . p April 2013
-------
Chapter 10 Transportation Corridor
10.3.1.3 Risk Characterization
Filling wetlands would eliminate habitat for salmonids and would indirectly alter wetlands in ways that
could reduce the quality, quantity, and accessibility of habitat for fish. Effects on fish production cannot
be estimated given available data; however, the loss of long riparian side channels to culvert or bridge
crossings that do not span the entire floodplain could be locally significant. These wetlands provide
important spawning and rearing habitats and resting areas for migrating adults. Other wetlands such as
shallow ponds may also provide habitat, but all wetlands serve to moderate variation in flow and
maintain water quality.
10.3.2 Stream Crossings
The transportation corridor would cross approximately 64 streams in the Kvichak River watershed. Of
these streams, 20 are listed as supporting anadromous fishes in AWC (Johnson and Blanche 2012), and
an additional 33 are likely to support salmonids (Table 10-6). Thus, the transportation corridor would
cross 53 streams known or likely to support salmonids. The physical effects of roads on streams and
rivers often propagate long distances from actual stream crossings, because of the energy associated
with moving water (Richardson et al. 1975).
Alteration of hydrology and sediment deposition by road crossings can change channels or shorelines
many kilometers away. The transportation corridor could affect 290 stream km between its road
crossings and Iliamna Lake (Table 10-7). Fish may also be affected in the approximately 830 km of
stream upstream of the transportation corridor that are likely to support salmonids (based on surveys
and stream gradients less than 12%, Table 10-8). In this assessment, we assume streams with segment
gradients less than 12% both downstream and upstream of the corridor-stream crossing are likely to
support salmonids (i.e., salmon, rainbow trout, or Dolly Varden). The amount of upstream length that
may be salmonid habitat is stream length to the first reach segment with a gradient greater than 12 %.
This criterion is used as an upstream limit for salmonid habitat, as Dolly Varden can be dispersed across
a wide range of channel gradients (Wissmar etal. 2010) and have been observed in higher-gradient
reaches (average 12.9% gradient) throughout the year in southeastern Alaska (Bryant et al. 2004).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1020 April 2013
-------
Chapter 10
Transportation Corridor
Table 10 6. Road stream crossings along the transportation corridor, upstream lengths of streams of different sizes likely to support
salmonids (based on stream gradients of less than 12%), and downstream length to Iliamna Lake. Bold reach codes are those assumed to
be bridged.
HUC-12 Name or Description
Headwaters Upper Talarik Creek
Upper Tributary to Upper Talarik
Creek"
Tributary to Newhalen River0
Headwaters Newhalen River
Outlet Newhalen River
Roadhouse Creek
Iliamna Lake-Eagle Bay
Eagle Bay Creek
Youngs Creek Mainstem
(Roadhouse Mountain HUC)
NHD Reach Code at Road-
Stream Crossing
19030206007354
19030206007015
19030206007159
19030206007175
19030205007587
19030205007593
19030205007598
19030205007606
19030205007602
19030205007615
19030205000002
19030205013069
19030205013055
19030205013057
19030205013041
19030206010623
19030206010628
19030206010629
19030206006712
19030206006678
19030206006677
19030206006644
19030206006671
19030206006663
19030206006654
19030206006598
AWC
(*Salmonid
Potential)
Y*
Y*
Y*
N *
N*
N*
N *
Y*
Y*
N *
Y*
N
N*
N*
N *
N *
N*
N *
N
Y*
N
N *
N *
Y*
Y*
Y*
Upstream Fish Habitat Length (km)
Small
Headwater
Streams3
3.4
112
1.4
2.9
3.9
4.0
3.9
5.6
5.1
3.2
71.2
0.0
7.3
1.8
3.4
0.7
0.2
0.7
0.0
4.1
0.0
1.5
0.8
9.4
4.0
27.5
Medium
Streams3
0.0
39.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
49.7
0.0
2.6
0.0
0.0
0.0
0.0
0.0
0.0
1.5
0.0
0.0
5.5
1.7
0.0
12.8
Small
Rivers3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Large
Rivers3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
13.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Total
3.4
152
1.4
2.9
3.9
4.0
3.9
5.6
5.1
3.2
135
0.0
9.8
1.8
3.4
0.7
0.2
0.7
0.0
5.7
0.0
1.5
6.4
11.1
4.0
40.3
Downstream Length
to Iliamna Lake
(km)
57.6
57.0
55.6
66.0
45.9
41.7
44.5
37.2
34.8
29.4
26.4
1.1
1.3
3.7
3.7
2.4
3.6
2.2
15.7
9.6
10.3
11.1
6.4
6.3
6.4
10.4
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-21
April 2013
-------
Chapter 10
Transportation Corridor
Table 10 6. Road stream crossings along the transportation corridor, upstream lengths of streams of different sizes likely to support
salmonids (based on stream gradients of less than 12%), and downstream length to Iliamna Lake. Bold reach codes are those assumed to
be bridged.
HUC-12 Name or Description
Youngs Creek East Branchd
Chekok Creek
Canyon Creek
Iliamna Lake-Knutson Bay
Knutson Creek
Iliamna Lake-Pedro Bay
Iliamna Lake-Pile Bay
NHD Reach Code at Road-
Stream Crossing
19030206006553
19030206006533
19030206032854
19030206006359
19030206006336
19030206006337
19030206006236
19030206006331
19030206006329
19030206006327
19030206006325
19030206006322
19030206006320
19030206006321
19030206006318
19030206006317
19030206006316
19030206006315
19030206006314
19030206006251
19030206006255
19030206006280
19030206006239
19030206006248
19030206006231
19030206006230
19030206006228
19030206006227
AWC
(*Salmonid
Potential)
Y*
Y*
Y*
Y*
N *
N *
N *
N*
N
N *
N*
N
N
N
N
N*
N*
N *
N*
N *
Y*
N*
N
N *
N
N
Y*
N*
Upstream Fish Habitat Length (km)
Small
Headwater
Streams3
34.8
6.2
50.7
0.3
4.9
0.3
1.1
0.7
0.0
0.4
1.0
0.0
0.0
0.0
0.0
0.3
0.5
0.7
0.7
0.6
0.6
0.7
0.0
0.5
0.0
0.0
0.3
0.0
Medium
Streams3
13.5
0.0
34.5
7.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
5.2
0.0
0.0
0.0
0.0
0.0
0.0
1.8
Small
Rivers3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Large
Rivers3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Total
48.3
6.2
85.2
7.9
4.9
0.3
1.1
0.7
0.0
0.4
1.0
0.0
0.0
0.0
0.0
0.3
0.5
0.7
0.7
0.6
5.8
0.7
0.0
0.5
0.0
0.0
0.3
1.8
Downstream Length
to Iliamna Lake
(km)
9.0
5.0
8.4
12.1
3.8
3.6
3.4
4.2
3.9
1.9
2.6
0.1
0.7
0.7
0.8
0.9
0.5
0.6
0.7
1.7
4.4
4.4
2.5
4.7
0.6
0.4
1.5
3.0
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-22
April 2013
-------
Chapter 10
Transportation Corridor
Table 10 6. Road stream crossings along the transportation corridor, upstream lengths of streams of different sizes likely to support
salmonids (based on stream gradients of less than 12%), and downstream length to Iliamna Lake. Bold reach codes are those assumed to
be bridged.
HUC-12 Name or Description
Outlet Pile River
Middle Iliamna River
Chinkelyes Creek
NHD Reach Code at Road-
Stream Crossing
19030206006222
19030206000474
19030206010632
324-10-10150-2343-
3006e
19030206000032
19030206005773
19030206005761
19030206005759
19030206005754
19030206005737
AWC
(*Salmonid
Potential)
N *
Y*
Y*
Y*
Y*
N*
N *
N*
N *
N *
Upstream Fish Habitat Length (km)
Small
Headwater
Streams3
3.2
34.9
2.0
Medium
Streams3
0.2
26.5
0.0
Small
Rivers3
0.0
48.7
0.0
Large
Rivers3
0.0
0.0
0.0
Total
3.4
110
2.0
NO NHD DATA
34.8
0.1
0.5
0.6
0.5
0.7
45.0
0.0
3.5
0.0
1.0
8.0
45.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
125.1
0.1
4.1
0.6
1.5
8.7
Downstream Length
to Iliamna Lake
(km)
6.3
5.7
0.9
1.0
10.2
13.4
14.5
18.0
21.6
22.1
Notes:
Values (lengths) are arranged by 12-digit HUG from west (top) to east (bottom) along the transportation corridor. Each upstream value is a sum of NHD stream segment lengths in the HUCs between the
crossing and upper extent of salmonid habitat potential based on 12% gradient. Each downstream value is a sum of stream segment lengths in the HUCs between the crossing and Iliamna Lake.
Because the lengths at each crossing represent contiguous lengths, a portion of stream may be included in more than one crossing.
a Small headwater streams = 0-0.15 m3/s; medium streams = 0.15-2.8 m3/s; small rivers = 2.8-28 m3/s; large rivers = >28 m3/s.
b 190302060701.
c 190302051404.
d 190302060904.
e Anadromous Waters Catalog (Johnson and Blanche 2012) stream code used, because no corresponding NHD (USGS 2012) stream code (and no upstream habitat data) available.
NHD = National Hydrography Dataset; AWC = Anadromous Waters Catalog; HUC = hydrologic unit code.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-23
April 2013
-------
Chapter 10
Transportation Corridor
Table 10 7. Lengths downstream of road stream crossings, by stream size. Stream size was based on mean annual flow; downstream length
was measured from road stream intersection to Iliamna Lake.
HUC-12 Name or Description
Headwaters Upper Talarik Creek
Upper Tributary to Upper Talarik Creekb
Tributary to Newhalen River0
Headwaters Newhalen River
Outlet Newhalen River
Roadhouse Creek
Iliamna Lake-Eagle Bay
Eagle Bay Creek
Youngs Creek Mainstem
(Roadhouse Mountain HUC)
Youngs Creek East Branchd
Chekok Creek
Canyon Creek
Iliamna Lake-Knutson Bay
Knutson Creek
Iliamna Lake-Pedro Bay
Iliamna Lake-Pile Bay
Outlet Pile River
Middle Iliamna River
Chinkelyes Creek
Total length across all HUCS
Percentage across all HUCS
Downstream Length (km)
Small Headwater
Streams3
12.3
0.8
4.3
0.8
2.9
12.9
4.4
0.6
0.0
0.8
4.8
6.0
15.9
1.8
8.1
5.0
3.3
0.0
0.9
85.5
29%
Medium Streams3
8.7
8.3
14.5
0.0
1.3
10.7
11.9
10.3
4.2
8.4
6.6
6.4
3.0
2.9
4.3
3.0
1.0
0.0
15.3
120.7
42%
Small Rivers3
37.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
4.1
10.2
0.4
51.7
18%
Large Rivers3
0.0
0.0
0.0
8.3
23.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
31.9
11%
Total
57.9
9.1
18.8
9.1
27.8
23.6
16.3
10.9
4.2
9.1
11.5
12.4
18.9
4.6
12.3
8.0
8.4
10.2
16.6
290
100%
Notes:
Values (lengths) are arranged by 12-digit HUC, from west (top) to east (bottom) along the transportation corridor. Downstream values are the sum of National Hydrography Dataset stream segment
lengths in the HUCs between the crossing and Iliamna Lake.
a Small headwater streams = 0-0.15 m3/s; medium streams = 0.15-2.8 m3/s; small rivers = 2.8-28 m3/s; large rivers = >28 m3/s.
b 190302060701
c 190302051404
d 190302060904
HUC = hydrologic unit code
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-24
April 2013
-------
Chapter 10
Transportation Corridor
Table 10 8. Lengths of different stream sizes that occur upstream of road stream crossings and are likely to support salmonids (based on
stream gradients of less than 12%).
HUC-12 Name or Description
Headwaters Upper Talarik Creek
Upper Tributary to Upper Talarik Creekb
Tributary to Newhalen River0
Headwaters Newhalen River
Outlet Newhalen River
Roadhouse Creek
Iliamna Lake-Eagle Bay
Eagle Bay Creek
Youngs Creek Mainstem (Roadhouse Mountain HUC)
Youngs Creek East Branchd
Chekok Creek
Canyon Creek
Iliamna Lake-Knutson Bay
Knutson Creek
Iliamna Lake-Pedro Bay
Iliamna Lake-Pile Bay
Outlet Pile River
Middle Iliamna River
Chinkelyes Creek
Total length across all HUCS
Percentage across all HUCS
Upstream length of fish habitat (km)
Small Headwater
Streams3
77.8
42.1
35.8
61.0
12.5
1.6
5.6
14.1
27.5
34.8
56.9
0.3
11.2
1.3
0.5
0.3
40.1
34.8
2.4
461
56%
Medium Streams3
19.0
20.6
19.1
30.6
2.6
0.0
1.5
7.3
12.8
13.5
34.5
7.6
0.0
5.2
0.0
1.8
26.8
45.0
12.6
260
31%
Small Rivers3
0.0
0.0
0.0
0.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
48.7
45.2
0.0
94.8
11%
Large Rivers3
0.0
0.0
0.0
13.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
13.4
2%
Total
96.8
62.7
54.9
106
15.0
1.6
7.1
21.4
40.3
48.3
91.3
7.9
11.2
6.5
0.5
2.1
115
125
15.0
829
100%
Notes:
Values (lengths) are arranged by 12-digit HUC, from west (top) to east (bottom) along the transportation corridor. Each upstream value is a sum of National Hydrography Dataset stream segment lengths
in the HUCs between the crossing and upper extent of salmonid habitat potential based on 12% gradient.
a Small headwater streams = 0-0.15 m3/s; medium streams = 0.15-2.8 m3/s; small rivers = 2.8-28 m3/s; large rivers = >28 m3/s.
b 190302060701.
c 190302051404.
d 190302060904.
HUC = hydrologic unit code.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
10-25
April 2013
-------
Chapter 10 Transportation Corridor
10.3.2.1 Exposure
Based on the assumption that crossings over streams with mean annual flows greater than 0.15 m3/s
would be bridged (Section 6.1.3), the transportation corridor would include 18 bridges (11 over known
anadromous streams and 7 over streams likely to support salmonids) (Table 10-6). All other stream
crossings would be culverted. Thus, given that the transportation corridor would cross a total of
53 streams and rivers known or likely to support, migrating or resident salmonids, culverts would be
constructed on 35 presumed salmonid streams.
Where flow restrictions such as culverts are placed in stream channels, stream power increases. This
can lead to increased channel scouring and down-cutting, streambank erosion, and undermining of the
road. Salmonids and other riverine fishes actively move into seasonal floodplain wetlands and small
valley floor tributaries to escape the stresses of main-channel flood flows (Copp 1989). Culverts can
reduce flow to these habitats by funneling flow from the entire floodplain through the culvert and into
the main channel. High water velocities in a stream channel may result from storm and snowmelt flows
being forced through a culvert rather than spreading across the floodplain. Higher velocities cause
scouring and down-cutting of the channel downstream of the culvert, hydrologically isolating the
floodplain from the channel and blocking fish access to floodplain habitat. Entrenchment of the channel
also prevents fish from reaching slow-water refugia during high-flow events and reduces nutrient and
sediment cycling processes between the stream channel and the floodplain. Lastly, channel
entrenchment may cause a change in the water table and the extent of the hyporheic zone, with
consequences for floodplain water-body connectivity and water temperatures in the floodplain habitat.
Culverts are deemed to have failed if fish passage is blocked (e.g., by debris, ice, or beaver activity) or if
stream flow exceeds culvert capacity, resulting in overtopping and road washout. Reported culvert
failure frequencies vary in the literature but are generally high. Values of 30% (Price et al. 2010), 53%
(Gibson et al. 2005), and 58% (Langill and Zamora 2002) have been reported, for an average culvert
failure estimate of 47% (i.e., culvert surveys indicate that, on average, 47% block or inhibit fish passage
at any given time).
When culverts are plugged by debris or overtopped by high flows, road damage, channel realignment,
and severe sedimentation often result (Furniss et al. 1991). Changes in sediment load due to culvert
failures can change stream hydraulics and geomorphic pressures. Generally, habitat value in the stream
is diminished as the channel becomes wider and shallower and silt is deposited in the streambed.
Stream crossing failures that divert stream flow outside of stream channels are particularly damaging
and persistent (Weaver et al. 1987).
Free access to spawning and early rearing habitat in headwater streams is critical for a number offish
species, and culverts are common migration barriers. Culvert blockages are usually caused by woody
debris and sometimes by woody material used by beavers to block a culvert and create a pond. In
addition, aufeis—an ice feature that forms when water in or adjacent to a stream channel rises above
the level of an existing ice cover and gradually freezes to produce a thickened ice cover (Slaughter
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~26 April 2013
-------
Chapter 10 Transportation Corridor
1982)—can completely fill culverts and, when this occurs, unless flow is initiated through the culvert,
water will run over the roadway (Kane and Wellan 1985). The ice also reduces the cross-sectional area
of flow so that high headwater conditions (and higher velocities than indicated by the culvert design)
are produced during periods of peak flow. In some cases, considerable ice remains after the breakup
period, particularly upstream of the culvert in the channel and floodplain (Kane and Wellan 1985).
Blockages could persist for as long as the intervals between culvert inspections. We assume that the
transportation corridor would receive daily inspection and maintenance during the operation of the
mine. Under such surveillance, a single erosional failure of a culvert that damaged the road would likely
be identified soon after it occurred and temporary repairs would be made to protect the road and
possibly provide fish passage. However, long-term fixes may not be possible until conditions are suitable
to replace a culvert or bridge crossing. Further, multiple failures such as might occur during an extreme
precipitation event would likely take longer to repair. Inspections would be likely to identify debris
blockages sufficient to cause water to pool above the road, and such blockages would be cleared to
prevent overflow of the road. These fixes may not fully address fish passage, which may be reduced or
blocked for longer periods. Also, some failures that would reduce or block fish passage (e.g., gradual
downstream channel erosion resulting in a perched culvert) might not be noticed by a driving
inspection. Thus, blockage of migration could persist for an extended period.
After mine operations end, traffic would be reduced to that which is necessary to maintain any residual
operations on the site, and inspections and maintenance would likely decrease. If the road was adopted
by the state or local governmental entity, the frequency of inspections and quality of maintenance would
likely decline to those provided for other roads. Either of these possibilities could result in a proportion
of failed culverts similar to those described in the literature.
10.3.2.2 Exposure-Response
Blockage of a culvert by debris or downstream erosion would inhibit the upstream and downstream
migration of salmon and the movement of other fish among seasonal habitats. The effects of a blockage
would depend on its timing and duration. A blockage would result in the loss of spawning and rearing
habitat if it occurred during migration periods and persisted for several days. It could cause the loss of a
year class of salmon from a stream if it occurred during migration periods and persisted for several days
or more.
Erosional failure of a road resulting from failure of a culvert would create suspended sediment that
would be carried and deposited downstream. Relationships between the concentration and duration of
elevated sediment concentrations and effects on fish and invertebrates are presented in Section 9.4.2.1.
10.3.2.3 Risk Characterization
The mine scenarios specify that culverts would be installed along the transportation corridor with
adequate size for normal flows of the streams crossed, and that the roadway would be monitored daily
to ensure that failures could be rapidly identified and repaired. Even with these assumptions, inhibition
of fish passage and reductions in habitat still could occur. Although culverts would be designed to
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~27 April 2013
-------
Chapter 10 Transportation Corridor
certain specifications (Box 10-2), they are not always installed correctly or do not stand up to the rigors
of a harsh environment, as indicated by the failure frequencies cited in Section 10.3.2.1. The
transportation corridor would traverse varied terrain and subsurface soil conditions, including
extensive areas of rock excavation in steep, mountainous terrain where storm runoff can rapidly
accumulate and result in intense local runoff conditions (Ghaffari etal. 2011). Although the road design,
including placement and sizing of culverts, would take into account seasonal drainage and spring runoff
requirements, culvert failures would still be expected. For example, heavy rains in late September 2003
washed out sections of the Williamsport-Pile Bay Road (Lake and Peninsula Borough 2009), and
culverts on this road have been washed out on numerous occasions (PLP 2011, Appendix 7.3A).
Culverts are not always built to specifications and the behavioral responses of migrating salmonid life
stages to culvert-induced changes in flow are not always anticipated correctly. Standards for culvert
installation on fish-bearing streams in Alaska consider road safety and fish passage, but not the physical
structure of the stream or habitat quality (ADF&G and ADOT&PF 2001). Culvert capacities are allowed
to be less than channel capacity (ADF&G and ADOT 2001). In most cases, culvert width must be greater
than 90% of the ordinary high-water channel width, but where channel slope is less than 1.0%, culvert
width must only be greater than 75% of the ordinary high-water channel width. During flood flows, this
reduced channel width results in slower than normal velocities upstream of the culvert and higher
water velocities exiting the culvert This could result in scoured downstream channel beds, scoured,
altered channel dynamics, and disassociated channels and floodplains disassociated. These processes
would reduce the capacity of downstream reaches to support salmonids. High flows in and immediately
downstream of the culvert, as well as the structure of the culvert itself, could inhibit fish passage even if
movement is not blocked. Downstream erosion could result in perched culverts that, if they were not
inspected and maintained, would inhibit and ultimately block fish passage. Floodplain habitat and
floodplain/channel ecosystem processes could be disrupted by channel entrenchment resulting from
culvert-induced erosion. These potential reductions in downstream habitat quality and inhibited fish
passage could occur in any of the 35 culverted streams that support salmonids.
Culverts and other road crossings that do not provide free passage between upstream and downstream
reaches can fragment populations into small demographic isolates vulnerable to extinction (Hilderbrand
and Kershner 2000, Young et al. 2005). In a study of natural long-term isolates of coastal cutthroat trout
and Dolly Varden in southeastern Alaska, Hastings (2005) found that about 5.5 km of perennial
headwater stream habitat, supporting a census population size of greater than 2,000 adults, is required
for a high likelihood of long-term population persistence. Table 10-8 shows that, of the 53 known or
likely salmonid-supporting streams that would be crossed by the transportation corridor, 36 contain
less than 5.5 km of habitat (stream length) upstream of the proposed road crossings. These 36 stream
crossings contain a total of 66 km of upstream habitat and 521 km of downstream habitat. Four of these
crossings would be bridged, leaving 32 with culverts. Assuming typical maintenance practices after mine
operations, roughly 47% of these streams, or 15 streams, would be entirely or partially blocked at any
one time. As a result, these streams would likely not be able to support long-term populations of
resident species such as rainbow trout or Dolly Varden.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~28 April 2013
-------
Chapter 10 Transportation Corridor
BOX 10 2. CULVERT MITIGATION
Bridge or culvert installation and maintenance activities in fish-bearing water bodies require a fish habitat
permit. Permit application information requirements for culvert installations in fish streams are detailed in a
memorandum of agreement (MOA) between the Alaska Department of Fish and Game (ADF&G) and the Alaska
Department of Transportation and Public Facilities (ADOT) (ADF&G and ADOT 2001). The MOA provides
guidance to project designers and permitting staff to ensure that culverts are designed and installed to provide
efficient fish passage and to ensure statewide consistency in Title 16 permitting of culvert related work. Title
16 is the statute by which the ADF&G performs Fish Habitat and Special Area permitting.
Fish habitat regulations under Title 16 include the Anadromous Fish Act and the Fishway (or Fish Passage) Act.
• The Anadromous Fish Act (AS 16.05.871-.901) requires that an individual or government agency
provide prior notification and obtain permit approval from ADF&G before altering or affecting "the
natural flow or bed" of a specified water body or fish stream. All activities within or across a specified
anadromous water body—including construction; road crossings; gravel removal; mining; water
withdrawals; the use of vehicles or equipment in the waterway; stream realignment or diversion; bank
stabilization; blasting; and the placement, excavation, deposition, or removal of any material—require
approval from ADF&G's Division of Habitat.
• The Fishway (or Fish Passage) Act (AS 16.05.841), requires that an individual or government agency
notify and obtain authorization from the ADF&G, Division of Habitat for activities within or across a
stream used by fish if it is determined that such uses or activities could represent an impediment to
the efficient passage of resident or anadromous fish.
The MOA describes the procedures, criteria and guidelines used for permitting culvert related work in fish-
bearing waters; these criteria augment but do not replace ADOT's standard design criteria presented in the
Alaska Highway Drainage Manual (ADOT 1995). Culverts are designed and permitted using one of the following
design approaches.
• Tier I—Stream Simulation Design (developed by the USDA Forest Service [FSSSWG 2008]). The Tier 1
approach most clearly replicates natural stream conditions, and is applicable in stream gradients less
than 6%. Using this design, culverts are sized larger than culverts sized hydraulically for floodwater
conveyance alone. The culvert width at the ordinary high water (OHW) stage waterline must be greater
than 90% of the OHW width. The culvert grade should approximate the channel slope, but in no
instance should it deviate more than 1% from the natural grade. In stream channels with slopes less
than 1%, culverts may be installed at slopes less than 0.5% with culvert widths greater than 75% of
the OHW width.
• Tier II—FISHPASS Program Design. Under this approach, culverts are designed using a combination
of traditional hydraulic engineering methods and the Alaska Interagency Fish Passage Task Force's
1991 "FISHPASS" computer modeling program (Behlke et al. 1991). The FISHPASS program
evaluates component hydraulic forces in a culvert against a fish's available power and energy
capabilities.
• Tier III—Hydraulic Engineering Design. The Tier III approach is used where site-specific conditions
preclude use of Tier I and Tier II designs. Under this approach, professionally recognized hydraulic
engineering methods are used to ensure appropriate fish passage characteristics in the culvert.
10.3.3 Chemical Contaminants in Stormwater Runoff
In this section we address three sources of potentially toxic chemicals that could run off the road: traffic
residues, road construction, and road treatment and chemical cargos.
During runoff events, traffic residues produce a contaminant mixture of metals (e.g., lead, zinc, copper,
chromium, and cadmium), oil, and grease that can get washed into streams and accumulate in sediments
(Van Hassel et al. 1980) or disperse into groundwater (Van Bohemen and Van de Laak 2003). It is
unclear if the transportation corridor would have sufficient traffic to contaminate runoff with significant
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~29 April 2013
-------
Chapter 10 Transportation Corridor
amounts of metals or oil (although storm water runoff from roads at the mine site itself are more likely
to contain metal concentrations sufficient to affect stream water quality). Therefore, this risk is not
considered further.
Road construction involves the crushing of minerals for the road fill and bed and the exposure of rock
surfaces at road cuts. That leads to leaching of minerals and increased dissolved solids. Fish mortality in
streams, with effects on populations recorded as far as 8 km downstream, has been related to high
concentrations of aluminum, manganese, copper, iron, or zinc from highway construction activities in
geological formations containing pyritic materials (Morgan et al. 1983). Because it is not clear where the
materials for the road will come from or their composition, this risk is not considered further.
Two potentially significant components of stormwater runoff may contaminate aquatic habitats along
the transportation corridor: chemicals released during spills along the corridor, and salts or other
materials used for winter road treatment. It should also be noted that increased runoff associated with
roads may increase rates and extent of erosion, reduce percolation and aquifer recharge rates, alter
channel morphology, and increase stream discharge rates (Forman and Alexander 1998). These effects
of stormwater runoff are not assessed, however, because they are highly location-specific and not
quantifiable given available data. Increases in sediment associated with stormwater runoff are
addressed in Section 10.3.4.
10.3.3.1 Exposure
Many chemical reagents would be used to process ore (Box 4-5), and these chemicals would need to be
transported to the mine site. Assuming these chemicals are trucked in, accidents along the
transportation corridor could spill reagents into wetlands or streams. To estimate how much reagent
and thus how many transport trucks would be needed for the mine scenarios, we extrapolated from the
number of trucks required to transport reagents at a smaller gold mine (175 trucks per year at Pogo
Mine) to the mine scenarios based on the relative annual ore production at the two mines. Assuming 20
tons of reagent per truck and expected annual production rates of 3,000 tons per day at Pogo Mine
(USEPA 2003a) and 200,000 tons per day in the mine scenarios (Ghaffari et al. 2011), we estimate that
transport of reagents would require approximately 11,725 truck trips per year.
The length of the transportation corridor within the Kvichak River watershed would be 113 km. The
probability of truck accidents and releases was reported as 1.9 x 10-7 spills per mile of travel for a rural
two-lane road (Harwood and Russell 1990). Based on this rate, the number of spills over the roughly 25-
year life Pebble 2.0 scenario would be 3.9—that is, 4 spills from truck accidents would be expected
during mine operations. Over the roughly 78-year life of the Pebble 6.5 scenario, 12 spills would be
expected. Only one-way travel is considered, because return trips from the mine would be with empty
trucks. Because the conditions on the mine road would be different from those for which the statistics
were developed (e.g., more difficult driving and road conditions), this calculation provides an order of
magnitude estimate. The reasonableness of this estimate is suggested by an assessment of the Cowal
Gold Project in Australia, which estimated that a truck wreck would occur every 1 to 2 years, resulting in
a spill every 3 to 6 years (NICNAS 2000).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~30 April 2013
-------
Chapter 10 Transportation Corridor
For 14% of its length (15 km), the transportation corridor would be within 100 m of a stream or river
(Table 10-3), and for 35% of its length it would be within 100 m of a mapped wetland (Table 10-4). If
the probability of a chemical spill is independent of location, and if it is assumed that spills within 100 m
of a stream could flow to that stream, a spill would have a 14% probability of entering a stream within
the Kvichak River watershed. This would result in roughly 0.6 stream-contaminating spills over the 25-
year life of Pebble 2.0 scenario or up to 2 stream-contaminating spills over the 78-year life of the Pebble
6.5 scenario. Similarly, a spill would have a 35% probability of entering a wetland, resulting in an
estimate of 1.4 wetland-contaminating spills under the Pebble 2.0 scenario, or 4 wetland-contaminating
spills under the Pebble 6.5 scenario. A portion of those wetlands would be ponds or backwaters that
support fish. It should be noted that the risk of spills could be somewhat mitigated by using tanks that
are spill resistant.
Roads are treated with salts and other materials to reduce dust and improve winter traction. In Alaska,
calcium chloride is commonly used for dust control and is mixed with sand for winter application.
During periods of rain and snowmelt, these materials are washed off roads and into streams, rivers, and
wetlands, where fish and their invertebrate prey can be directly exposed. We found no relevant data for
sodium chloride levels in runoff or streams from roads treated in this way.
10.3.3.2 Exposure-Response
The primary chemical of concern would be sodium ethyl xanthate (Section 6.4.2). A risk assessment by
Environment Australia estimated that a spill of as little as 10% of a 25-metric-ton-capacity truck of
sodium ethyl xanthate into a stream would require a "650000:1 dilution before the potential hazard is
considered acceptable" and that the spill could not be mitigated (NICNAS 2000).
Compounds used to control ice and dust (Hoover 1981) have been shown to cause toxic effects when
they run off and enter surface waters. Dissolved calcium, like sodium, has little influence on the toxicity
of dissolved chloride salts (Mount etal. 1997). Based on that study, the toxicity of the calcium chloride
commonly used in Alaska would be expected to be a little greater than the more studied sodium
chloride, based on chlorine concentrations. Alaska acute and chronic water quality standards for
chloride are 860 and 230 mg/L, respectively (ADEC 2003). However, these values may not provide
adequate protection from calcium salts. In addition, exceedances of the acute criterion could affect many
species, because freshwater biota have a narrow range of acute susceptibilities to chloride (ADEC 2003).
These standards and the associated federal criteria also may not be adequately protective due to the
absence of tests of critical life stages such as egg fertilization.
Rainwater tends to leach out the highly soluble chlorides (Withycombe and Dulla 2006), which can
degrade nearby vegetation, surface water, groundwater, and aquatic species (Environment Canada
2005). Salmonids are sensitive to salinity, particularly at fertilization (Weber-Scannell and Duffy 2007).
According to the Bolander and Yamada (1999), application of chloride salts should be avoided within at
least 8 m of water bodies (including shallow groundwater, if significant migration of chloride would
reach the groundwater table), and restricted if low salt-tolerant vegetation occurs within 8 m of the
treated area. Adverse biological effects are likely to be particularly discernible in naturally low-
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~31 April 2013
-------
Chapter 10 Transportation Corridor
conductivity waters, such as those of the Bristol Bay watershed, but research is needed to substantiate
this (Appendix G).
10.3.3.3 Risk Characterization
Given the toxicity of sodium ethyl xanthate (Section 8.2.2.5), it is expected that a spill of this compound
into a stream along the transportation corridor would cause a fish kill. Runoff or groundwater transport
from a more distant spill would cause effects that would depend on the amount of dilution or
degradation occurring before the spilled material entered a stream.
Risks to salmonids from de-icing salts and dust suppressants could be locally significant, but would
depend on the amount and frequency of application. The transportation corridor would intersect
53 streams and rivers known or likely to support salmonids, and there would be approximately
290 stream km between road crossings and Iliamna Lake (Table 10-8). Additionally, approximately
12 km of roadway would intersect wetlands within and beyond those mapped by NWI. Runoff from
these road segments could have significant effects on fish and the invertebrates that they consume,
particularly if sensitive life stages are present.
10.3.4 Fine Sediment
10.3.4.1 Exposure
During rain and snowmelt, soil eroded from road cuts, borrow areas, road surfaces, shoulders, cut-and-
fill surfaces, and drainage ditches (as well as road dust deposited on vegetation; see Section 10.3.5),
would be washed into streams and other water bodies. Erosion and siltation are likely to be greatest
during road construction. The main variables determining surface erosion are the inherent credibility of
the soil, slope steepness, surface runoff, slope length, and ground cover. Mitigation measures for fine
sediments are discussed in Box 10-3. It is worth noting that improvements have been proposed for the
road between Iliamna and Nondalton, in part to alleviate erosion and sedimentation problems at some
areas along the road (ADOT 2001).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1032 April 2013
-------
Chapter 10 Transportation Corridor
BOX 10 3. STORMWATER RUNOFF AND FINE SEDIMENT MITIGATION
The Alaska Department of Environmental Conservation (ADEC) administers Alaska Pollutant Discharge Elimination
System (APDES) stormwater general permits for construction activities and multi-sector general permits for
industrial operation activities. ADEC also approves stormwater pollution prevention plans (SWPPPs) that include
stormwater best management practices (BMPs).
A permittee covered under the APDES stormwater general permit for construction activities (ADEC 2011a) must
comply with control measures that are determined by site-specific conditions. ADEC developed the Alaska Storm
Water Guide (ADEC 2011b) to assist permittees with selecting, installing, and maintaining control measures that
may be used for projects in Alaska. Erosion and sediment control measures covered under the stormwater
general permit for construction activities (ADEC 2011a) are summarized below.
Erosion Control Measures
• Delineate the site; specifically, the location of all areas where land disturbing activities will occur and
areas that will be left undisturbed (e.g., boundaries of sensitive areas or established buffers).
• Minimize the amount of soil exposed during construction activity by preserving areas of native topsoil on
the site where feasible and sequencing or phasing construction activities to minimize the extent and
duration of exposed soils.
• Maintain natural buffer areas.
• Control stormwater discharges and flow rates, via the following mechanisms:
Diversion of stormwater around the site.
Slow down or containment of stormwater that collects and concentrates at the site.
Avoidance of structural control measure placement in active floodplains, to the degree practicable and
achievable.
Placement of velocity dissipation devices (e.g., check dams, sediment traps, or riprap) along conveyance
channels and where discharges from conveyance channels join water courses.
• Protect steep slopes, via the following mechanisms:
Design and construction of cut-and-fill slopes to minimize erosion.
Diversion of concentrated storm water flows away from and around the disturbed slopes, using
interceptor dikes, swales, grass-lined channels, pipe slope drains, surface drains, and check dams.
Stabilization of exposed slope areas.
Sediment Control Measures
Sediment control measures (e.g., sediment ponds, traps, filters) should be functional before other land-disturbing
activities take place. These measures may include:
• Storm drain inlet protection measures (e.g., filter berms, perimeter controls, temporary diversion dikes),
that minimize the discharge of sediment prior to entry into the inlet for storm drain inlets located on site
or immediately downstream.
• Water body protection measures (e.g., velocity dissipation devices) that minimize the discharge of
sediment prior to its entry into water bodies located on site or immediately downstream.
• Down-slope sediment controls (e.g., silt fences, temporary diversion dikes) for any portion of the down-
and side-slope perimeters where stormwater would be discharged from disturbed areas of the site.
• Establishment and stabilization of construction vehicle access and exit points, limited to one route if
possible.
• Minimization of dust generation through the application of water or other dust suppression techniques
prior to vehicle exit.
Stabilization or coverage of soil stockpiles, protection with sediment trapping measures, and, where
possible, located away from storm drain inlets, water bodies, and conveyance channels.
• Design sediment detention basins to capture runoff or conveyed storm water and reduce water velocity
to allow sediments to settle out before they can enter streams or other water bodies. Storm flows
eventually pass through an outflow structure leaving the sediment (i.e., solids that can settle) in the
basin. There are important design and management considerations for sediment detention basins for
hard rock mining (USEPA 2003b, Section 6.1.6).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ 03 April 2013
-------
Chapter 10 Transportation Corridor
Soil Stabilization
All disturbed areas of the site should be stabilized to minimize on-site erosion and sedimentation and the
resulting discharge of pollutants according to the requirements in ADEC 2011a. Existing vegetation should be
preserved wherever possible.
Many of the BMPsfor industrial operations associated with metal mining focus on sediment and erosion control
and are similar to BMPs used in the construction industry (USEPA 2006). Some of these BMPs pertain specifically
to haul and/or access roads (USEPA 2006).
• Construction of haul roads should be supplemented by BMPs that divert runoff from road surfaces,
minimize erosion, and direct flow to appropriate channels for discharge to treatment areas. Examples of
these BMPs include:
Dikes, curbs, and berms for discharge diversions.
Conveyance systems such as channels, gutters, culverts, rolling dips and road sloping, and/or roadway
water deflectors.
Check dams, rock outlet protection, level spreaders, stream alternation, and drop structures for runoff
dispersion.
Gabions, riprap, native rock retaining walls, straw bale barriers, sediment traps/catch basins, and
vegetated buffer strips for sediment control and collection.
Vegetation to stabilize soils.
• Roads should be placed as far as possible from natural drainage areas, lakes, ponds, wetlands, or
floodplains.
• Width and grade of roads should be as small as possible to meet regulatory requirements and designed
to match the area's natural contours.
All stabilization and structural erosion control measures should be inspected frequently and all necessary
maintenance and repairs should be performed.
10.3.4.2 Exposure-Response
Sediment loading from roads can severely affect streams downstream of the roadbed (Furniss et al.
1991). Salmonids are adapted to episodic exposures to suspended sediment, but as concentrations or
durations of exposure increase, survival and growth can be affected (Section 9.4.2.1). Increased
deposition of fine sediment decreases the abundance and production offish and benthic invertebrates
(Section 9.4.2.2). Fine sediments have been linked to decreased fry emergence, decreased juvenile
densities, loss of winter carrying capacity, increased predation on fishes, and reduced benthic organism
populations and algal production (Newcombe and MacDonald 1991, Gucinski etal. 2001, Angermeier et
al. 2004, Suttle et al. 2004). In low-velocity stream reaches, an excess of fine sediment can completely
cover suitable spawning gravel, rendering it useless for spawning, and sediment deposited after
spawning may smother eggs and alevins. Excessive stream sediment loading can also result in channel
braiding, increased width-depth ratios, increased incidence and severity of bank erosion, reduced pool
volume and frequency, and increased subsurface flow. These changes can result in reduced quality and
quantity of available spawning habitat (Furniss etal. 1991).
Increased runoff associated with roads may increase rates and extent of erosion, reduce percolation and
aquifer recharge rates, alter channel morphology, and increase stream discharge rates (Forman and
Alexander 1998). During high-discharge events and in high velocity streams, accumulated sediment
tends to be flushed out and re-deposited in larger water bodies (Forman and Alexander 1998). Because
streams crossed by the transportation corridor connect downstream to Iliamna Lake and ponds,
accelerated sedimentation could have an impact on the concentrated spawning populations of sockeye
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±034 April 2013
-------
Chapter 10 Transportation Corridor
salmon in these habitats. Accelerated sedimentation could also have a localized impact on the clarity and
chemistry of Iliamna Lake affecting the photic zone (depth of light penetration sufficient for
photosynthesis), thereby affecting primary production in the lake and zooplankton abundance critical to
juvenile sockeye salmon.
10.3.4.3 Risk Characterization
Suspended and deposited sediment washed from roads, shoulders, ditches, cuts, and fills would likely
diminish habitat quality in the streams below road crossings. The magnitude of effects cannot be
estimated in this assessment. However, published studies of the influence of silt on salmonid streams
(Chapter 9) indicate that even relatively small amounts of additional sediment could have locally
significant effects on reproductive success of salmonids and production of aquatic invertebrates.
Potential mitigation measures for storm water runoff, erosion, and sedimentation are discussed in Box
10-3.
10.3.5 Dust
Dust results from traffic operating on unpaved roads in dry weather, grinding and breaking down road
materials into fine particles (Reid and Dunne 1984). These fines are either transport aerially in the dry
season or mobilized by water in the wet season. Dust particles may also include trace contaminants,
including de-icing salts, hydrocarbons, and metals. Following initial suspension by vehicle traffic, aerial
transport by wind spreads dust over long distances, so that it can reach surface waters that are
otherwise buffered from sediment delivery via aqueous overland flow (Appendix G). Dust control agents
such as calcium chloride have been shown to reduce the generation of road dust by 50 to 70% (Bader
1997), but these agents may cause toxic effects when they run off and enter surface waters (Section
10.3.3).
10.3.5.1 Exposure
The amount of dust derived from a road surface is a function of many variables, including composition
and moisture state of the surface, amount and type of vehicle traffic, and speed. An Iowa Highway
Research Board project (Hoover et al. 1973) that quantified dust sources and emissions created by
traffic on unpaved roads found that one vehicle, traveling 1 mile of unpaved road once a day every day
for 1 year, would result in the deposition of 1 ton of dust within a 1,000-foot corridor centered on the
road (i.e., traffic would annually deposit 1 ton of dust per mile per vehicle).
To estimate truck traffic required by the mine scenarios, we extrapolated from vehicle use at a smaller
gold mine (Pogo Mine) based on the relative rate of ore production atPogo relative to the scenarios.
Estimated production rate at Pogo is 3,000 tons per day (USEPA2003a), versus 200,000 tons per day in
the mine scenarios (Ghaffari etal. 2011). Overall mine-related vehicle use at Pogo averages between 10
and 20 round trips per day (USEPA 2003a). Approximately 175 truck trips per year (0.5 round trip per
day) are required at Pogo to transport reagents, leaving 19.5 round trips per day for other purposes. The
number of truck trips required for transport of reagents is assumed to be roughly proportional to ore
production, resulting in an estimate of 33 round trips per day to transport reagents in the mine
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±035 April 2013
-------
Chapter 10 Transportation Corridor
scenarios. The number of daily round trips for purposes other than reagent transport was estimated at
19.5 round trips per day, for a total daily traffic estimate of 52.5 round trips in the mine scenarios. This
value is likely an underestimate, as it does not account for potential effects of size differences between
Pogo Mine and the mine scenarios or the number of trips for purposes other than reagent transport.
The length of the transportation corridor within the Kvichak River watershed would be 113 km. Based
on the estimate from Hoover et al. (1973), the average amount of dust (in tons) generated per mile of
road per year along the transportation corridor within the Kvichak River watershed would be
equivalent to the daily average number of vehicles passing along the corridor (one vehicle making a
round-trip constituting two passings). Using this method, the mine scenarios would generate
approximately 105 tons of dust per mile (59 metric tons per km) annually or approximately
6,700 metric tons annually for the entire length of road within the Kvichak River watershed. This value
may be underestimated because smaller vehicles use typical rural roads in Iowa, or overestimated if
roads in Iowa are dryer or if dust suppression is effective, but it indicates that dust production along the
transportation corridor could be substantial.
10.3.5.2 Exposure-Response
Walker and Everett (1987) evaluated the effects of road dust generated by traffic on the Dalton Highway
and Prudhoe Bay Spine Road in northern Alaska. Dust deposition altered the albedo of snow cover,
causing earlier (and presumably more rapid) snowmelt up to 100 m from the road margin, as well as
increased depth of thaw in roadside soils. Dust was also associated with loss of lichens, sphagnum, and
other mosses, and reduced plant cover (Walker and Everett 1987). Loss of near-roadway vegetation has
important implications for water quality, as that vegetation helps to filter sediment from road runoff.
Thus, dust deposition can contribute to stored sediment that can mobilize in wet weather, and
deposition can reduce the capacity of roadside landscapes to filter that sediment.
In a study of road effects in Arctic tundra at acidic (soil pH less than 5.0) and less acidic (soil pH at least
5.0) sites, Auerbach et al. (1997) found that effects on vegetation were more pronounced at the acidic
site. Permafrost thaw was deeper next to the road than away from the road at both sites, and could
affect road structure detrimentally. Vegetation biomass of most taxa was reduced near the road at both
sites. Species richness in acidic tundra next to the road was less than half the richness at 100 m away
from the road. Sphagnum mosses, dominant in acidic low arctic tussock tundra, were virtually
eliminated near the road. According to PLP (2011: Chapter 5), approximately 72% of the mine area is
composed of well-drained acidic soils (58% strongly acidic). Approximately 34% of the transportation
corridor is composed of well-drained acidic soils (3.5% strongly acidic).
10.3.5.3 Risk Characterization
The main impact of dust from the transportation corridor on salmonids would likely be reduced habitat
quality due to a reduction in riparian vegetation and subsequent increase in suspended sediment and
fine bed sediment, especially during road construction. Potential effects of increased sediment loading
are discussed in Section 10.3.4. Loss of riparian vegetation would also occur atthe mine site, but there
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±036 April 2013
-------
Chapter 10 Transportation Corridor
the main impact of dust would be a direct increase in fine bed sediment due to mine construction and
operation.
10.3.6 Invasive Species
10.3.6.1 Exposure
Construction and operation of the transportation corridor would increase the probability that new
terrestrial and aquatic species would be transported to and could establish themselves in the Bristol Bay
region. Roads can facilitate introductions via contaminated soil or gravel used in road construction and
maintenance, or via contaminated vehicles, equipment, cargo, and people that travel those roads. For
example, road fill appears to be the mode of introduction and spread for invasive sweetclover (Melilotus
alba) in central and southeast Alaska (Wurtz et al. 2010). Elsewhere, road maintenance further spreads
invasive plants along suitable roadside habitat (Christen and Matlack 2009). Vehicles can carry
contaminated equipment and cargo. Over the 2-year construction of a research station in Antarctica, an
estimated 5,000 seeds from 14 different plant families were introduced on almost 15,000 m3 of cargo
(Lee and Chown 2009). Once docked, seeds on cargo can disperse at almost any location along the
transportation corridor. Finally, people unintentionally introduce and spread invasive species in Alaska
and other Arctic environments on their shoes (Bella 2011, Ware et al. 2012).
Once established along or near the transportation corridor, terrestrial species that thrive in riparian and
floodplain areas can spread to salmon-bearing habitat at any of the points where the road crosses a
river, stream, or wetland. In a survey of 2,865 km (1,780 miles) of major highways in interior and south-
central Alaska, 64 of 192 sampled bridge crossings (over 30%) were found to have sweetclover adjacent
to them, and sweetclover had spread to downstream floodplains at 17 of these bridge crossings (Wurtz
et al. 2010). This survey likely underestimates the number of floodplain invasions, because it did not
sample numerous stream crossings serviced by culverts or other locations along streams where fill had
been placed.
Aquatic invasive species, including macrophytes, shellfish, and salmonid pathogen and parasites like
those that cause whirling disease, can also be introduced along the transportation corridor on
equipment that has come into contact with contaminated waters. Most literature emphasizes recreation
equipment (Arsan and Bartholomew 2008, Johnson et al. 2001); little or no information exists about the
incidence of aquatic or riparian species introductions specifically on construction or mining equipment.
Transported equipment contaminated with aquatic invaders could spread those species to salmon-
bearing habitat via direct contact with anadromous waters during construction of stream crossings, or
during mining activity. Aquatic invaders could also be carried by water to other salmon-bearing habitats
downstream of the initial introduction locations, including into Iliamna Lake and other parts of the
Kvichak River watershed.
The likelihood that an aquatic invasive species will establish and spread successfully can depend heavily
on environmental requirements. For instance, Myxobolus cerebralis, a cnidarian parasite that causes
whirling disease, has already been detected in an Anchorage, Alaska trout hatchery. This parasite has
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±037 April 2013
-------
Chapter 10 Transportation Corridor
very specific abiotic and biotic conditions under which it infects salmonids. If the pathogen is introduced
to a new area, susceptible genetic variants of the secondary host (an oligochaete worm called Tubifex
tubifex) must be present, seasonal water temperatures must exceed 10°C with approximately
1,500 degree-days, and susceptible species and life-stages of salmonids must co-occur with the
secondary host (Arsan and Bartholomew 2008). In addition to the hatchery location where whirling
disease has already been found, favorable conditions exist for parasite establishment in two tributaries
of Cook Inlet near Anchorage (Arsan and Bartholomew 2008). However, conditions for whirling disease
establishment are not known for the Bristol Bay region.
10.3.6.2 Exposure-Response
Invasive species can drastically alter the composition of riparian and floodplain vegetation adjacent to
salmon habitats. Invasive sweetclover, purple loosestrife (Lythrum salicarid), and giant knotweed
[Polygonum sachalinense)—all current invaders in Alaska—can replace native riparian species (Blossey
etal. 2001, Spellman and Wurtz 2011, Urgenson etal. 2009). In general, it has been difficult to show a
direct effect of riparian vegetation alteration on fish diversity, abundance, or biomass (Smokorowski
and Pratt 2007), but indirect effects on salmon via the aquatic food webs have been documented (Wipfli
and Baxter 2010). Giant knotweed was shown to release nitrogen-poor litter into a tributary of the
salmon-bearing Skagit River in Washington, which can have cascading, negative effects on fish by
altering their invertebrate food sources (Urgenson et al. 2009). Purple loosestrife was found to
decompose four times faster than native sedge in the Fraser River, making detritus available in fall
rather than winter and spring, when it was usually used by invertebrates that support salmon
production (Groutetal. 1997).
Links between aquatic invaders, particularly macrophytes, and fish performance have been made in
lentic, but rarely in lotic, habitats (Smokorowski and Pratt 2 007). Effects of invasive macrophytes range
from increased native fish abundance, to no effect, to detrimental effects on fish and their food sources
via exuded toxic compounds, depending on the invasive species and fish species of interest (Schultz and
Dibble 2012). Streambed coverage of several species of aquatic macrophytes, both native and
introduced (including the recent Alaskan invader Elodea canadensis), in northern California reduced the
number of Chinook salmon redds and the percentage of available spawners observed using infested
habitat (Merz etal. 2008). This is significant in the regulated, low-flow Mokelumne River in California,
where spawning habitat is considered a limiting resource.
Evidence of the effects of other aquatic invaders on salmonids also exists. Didymo (Didymosphenia
germinatd) is a colonial diatom capable of covering stream substrates with thick, slippery mats.
Documented effects of didymo on salmonids vary with location and fish species of interest. Documented
effects of didymo on the invertebrate communities that serve as fish food sources could ultimately affect
salmonid growth and abundance (Whitton et al. 2009). The aquatic invader that causes whirling disease
(M. cerebralis) has had devastating effects on several wild fisheries in the United States intermountain
west (Nehring and Walker 1996). The disease can cause lesions, neurological defects, skeletal
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±038 April 2013
-------
Chapter 10 Transportation Corridor
deformities, and death. Both sockeye salmon and rainbow trout fry are highly susceptible to whirling
disease, should conditions be right for infection.
10.3.6.3 Risk Characterization
The spread of aquatic, riparian, and floodplain invasive species along roads to stream crossings and into
salmon-bearing habitats could occur during construction and operation of the proposed transportation
corridor, although mitigation measures can lower the likelihood of invasion (Box 10-4). Invasion of
riparian and floodplain species is occurring in Alaska because of the use of contaminated gravel road fill.
In the case of invasive sweetclover, subsequent dispersal to almost 9% of floodplains downstream of
bridges along one major highway was observed (Wurtz et al. 2010). Assuming similar rates of invasion
along the transportation corridor, and similar rates of invasion along bridges and culverts, 9% of the
64 streams and rivers—5 to 6 streams—crossed by the corridor in the Kvichak River watershed would
experience invasion. Given that 53 of the 64 streams crossed by the transportation corridor are known
or likely to support salmonids, alteration of salmon habitats would be expected in 4 to 5 streams.
However, this is almost assuredly an underestimate because it is based on rate of invasion of only one
species and assumes that the spread of that species has reached equilibrium.
Should sweetclover, purple loosestrife, giant knotweed or other species invade riparian and floodplains
adjacent to salmon-bearing streams and wetlands in the Bristol Bay region, they could change organic
matter inputs into those streams and affect salmon food sources (Blossey et al. 2001, Spellman and
Wurtz 2011, Urgenson et al. 2009). To what extent salmon growth, diversity, or abundance would be
altered would depend on the extent and intensity of infestation. Once initiated, these invasions would be
difficult to reverse.
The use of contaminated gravel road fill in Alaska has fostered the invasion of nonnative riparian and
floodplain plant species. In some cases, the species are subsequently dispersed to floodplains downstream
of road-stream crossings. Introduction and invasion of nonnative riparian and floodplain species may also
occur via contaminated cargo, equipment, and boots. The following steps can help to mitigate the
introduction and spread of invasive species.
• Purchase of fill from existing or new gravel pits certified by the Alaska Department of Natural
Resources Division of Agriculture as weed-free (ADNR 2013).
• Proper and thorough inspection and de-contamination of cargo, equipment, and boots, at the port
and at the mine site.
• Use of new equipment, where possible.
• Use of a process for cleaning, draining, and drying equipment previously used at another site
(including personal gear worn by workers) that is advocated by the Alaska Department of Fish and
Game for recreational equipment.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1039 April 2013
-------
Chapter 10 Transportation Corridor
10.4 Overall Risk Characterization for the
Transportation Corridor
Risks to salmonids from filling of wetlands, hydrologic modifications, runoff of contaminants and fine
sediment, dust deposition, and introduction of invasive species are likely to diminish the production of
anadromous and resident salmonids in many of the 53 streams known or likely to support salmonids
that would be crossed by the transportation corridor. Salmonid spawning migrations and other
movements may be impeded by culverts in 35 streams, 32 of which contain restricted (less than 5.5 km)
upstream habitat. Assuming typical maintenance practices after mine operations, approximately 15 of
these 32 streams would be entirely or partly blocked at any time. As a result, salmonid passage—and
ultimately production—would be reduced in these streams, and they would likely not be able to support
long-term populations of resident species such as rainbow trout or Dolly Varden. Approximately 290 km
of stream downstream of road crossings also could be affected.
The exact magnitudes of changes in fish productivity, abundance, and diversity cannot be estimated at
this time, but the species, abundances, and distributions that could be affected are summarized below.
• Sockeye salmon spawning has been observed at 30 locations along the transportation corridor.
Highest average abundances are in the Iliamna River (100,000 spawners), the Newhalen River
(80,000 spawners), and Knutson Bay (70,000 spawners), although abundances can be much higher
(e.g., 1 million adults were reported in 1960 survey of Knutson Bay).
• Chinook, coho, pink and chum salmon have been reported at isolated points in the Kvichak River
watershed, and all four species have been observed in Upper Talarik Creek and the Iliamna River.
• Dolly Varden have been reported in nearly every sockeye salmon-bearing stream that would be
crossed by or adjacent to the corridor, as well as in locations upstream of sites with reported
anadromous salmon use.
• Rainbow trout have been reported in Upper Talarik Creek, the Newhalen River, an unnamed
tributary to Eagle Bay, Youngs, and Tomkok, Swamp, and Chinkelyes Creeks.
10.5 Uncertainties
In this chapter we evaluated the risks to salmonid habitats and populations associated with the
transportation corridor (Figure 10-3). A number of uncertainties are inherent in assessing these risks,
which are summarized below (uncertainties related to the effectiveness of mitigation measures are
discussed in Box 10-5).
• Characterization of streams and wetlands affected by the transportation corridor. The NWI, NHD,
AWC, and AFFI were used to evaluate the effects of the transportation corridor on hydrological
features and fish populations (Box 10-1). These datasets include the following limitations.
o Underestimation of the number of stream crossings and degree of channel sinuosity, resulting in
underestimates of affected stream lengths.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~40 April 2013
-------
Chapter 10 Transportation Corridor
o Underestimation offish-bearing streams due to limited sampling.
o Potential undercharacterization of wetland area due to limited resolution of available NWI data
product.
o Underestimation of potential impacts on wetlands bisected by the transportation corridor,
because wetland area outside the 200-m boundary was assumed to maintain functionality.
Overall, these uncertainties likely result in a moderate underestimation of risks to fish.
• Estimation of dust production from the transportation corridor. Our dust production estimate is
based on a study that quantified dust sources and omissions created by traffic on unpaved roads.
Extrapolating that study to the transportation corridor does not take into account variables such as
composition and moisture of the road surface, number of tires and their widths, and speed. In
addition, road dust generation may be reduced by 50 to 70% by the application of dust control
agents such as calcium chloride. Overall, these uncertainties likely have a negligible effect on risks to
fish, but a moderate effect on our dust production calculations.
• Estimation of chemical spill frequency due to truck accidents. Extrapolation of truck accident
probability from a study of rural two-lane roads does not take into account specific, generally more
difficult road and weather conditions prevalent in the area of the Pebble deposit. However, the risk
of spills could be at least partially mitigated by using spill-resistant tanks. Overall, these
uncertainties likely result in a moderate underestimation of risk to fish, due primarily to the toxicity
of sodium ethyl xanthate, but also because of effects on spill frequency calculations.
• Estimation of culvert failure frequencies. These frequencies, derived from the literature, assume
that culverts are designed to specifications but are not always installed correctly and/or do not
stand up to the rigors of a harsh environment. This uncertainty likely has a moderate effect on risk
to fish, with unclear direction. Nonetheless, this does not change overall conclusions reached with
respect to reduction of passage and ultimately production of salmonids or the viability of long-term
populations of resident species.
• Climate change effects. The potential impacts of road construction and operation discussed in this
chapter do not take into account the effects of climate change. Over the timeframe considered in this
assessment (approximately 80 years), the physical environment of the Bristol Bay watershed is
likely to change substantially as a result of increases in temperature and precipitation (Section 3.8).
Increases in rain-on-snow events are likely to increase flood frequency. Such changes could
undermine the structure of the transportation corridor and stream crossings. The variability and
magnitude of stream flows could also enhance the other impacts described in this chapter, including
channel entrenchment and a loss of water-body connectivity. Collectively, these impacts would
likely further reduce the diversity offish habitat, causing a loss of population genetic diversity over
time that would reduce the resiliency of salmon stocks to environmental fluctuations related to
climate change. Overall, these climate-related uncertainties result in a moderate underestimation of
risk to fish.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 10~41 April 2013
-------
Chapter 10 Transportation Corridor
BOX 10 5. LIKELY EFFECTIVENESS OF MITIGATION MEASURES
Environmental characteristics along the transportation corridor would likely render the effectiveness of
standard or even state-of-the-art mitigation measures highly uncertain.
• Subarctic extreme temperatures and frozen soil conditions could greatly reduce infiltration and
capture of runoff and sediment during thaws and could decrease the ability to capture accidental
releases of liquid substances in transport.
• Subarctic climatic conditions could limit the lushness and rapidity of vegetation growth or re-growth
following ground disturbance, reducing the effectiveness of vegetated areas as sediment and
nutrient filtration buffers.
• Widespread and extensive areas of near-surface groundwater and seasonally or permanently
saturated soils could limit the potential for absorption or trapping of road runoff, and increase
likelihood of its delivery to surface waters.
• The likelihood of ice flows and drives during thaws could make water crossing structures
problematic locations for jams and plugging.
• The region is seismically active, and even a small increment of ground deformation could easily
disturb engineered structures and alter patterns of surface and subsurface drainage in ways that
render engineered mitigations inoperative or harmful.
• Remote locations that are not frequented by humans mean that mitigation failures and accidents
could go undetected until substantial harm to waters has occurred unless regular inspections are
conducted.
Although many possible mitigation measures can be identified and listed in a mitigation plan, they cannot all
be ideally applied in every instance. Mitigation measures are often mutually limiting or offsetting when
applied in the field. As a salient example for the transportation corridor, choosing a road location that
minimizes crossings of streams, wetlands, and areas of shallow groundwater in a landscape that is rich in
those hydrologic features could result in a tortuous alignment that is excessively long and curved to
accommodate the upland terrain. This alignment would greatly increase the total ground area disturbed, and
increased road curvature in either horizontal and vertical dimensions may increase risk of traffic accidents
and consequent spills. It would also increase the length and structural complexity of the road-parallel
pipelines. Thus, avoidance of sensitive habitat features could elevate other environmental risks.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ . „ APril 2013
-------
As described in Section 6.1.4, the mine scenarios include four pipelines along the transportation
corridor—one each for natural gas, diesel, product concentrate, and return water—and various
pipelines on the mine site. Any of these pipelines could fail and release their contents to the
environment. The risks from failure of product concentrate (Figure 11-1), return water (Figure 11-2),
and diesel (Figure 11-3) pipelines are considered particularly high; these failure scenarios are evaluated
in the following sections. Other pipelines are discussed briefly below.
On the mine site, the largest pipelines would carry tailings slurry from the mill to the tailings storage
facilities (TSFs) and reclaimed water from the TSF to the mill (Table 6-5). Smaller pipelines would
convey water for processing, domestic uses, firefighting, and other uses and wastewater for treatment
or storage. Other pipelines would carry diesel and natural gas from storage tanks to points of use. On-
site pipeline spills have occurred at porphyry copper mines in the United States and some have resulted
in significant aquatic exposures (Earthworks 2012). Such spills are possible at a future mine and could
result in uncontrolled releases within the mine site; however, these spills are more likely to be
contained or controlled without significant environmental effects than pipeline spills along the
transportation corridor. In this assessment, we decided that leakage from on-site pipelines would be
captured and controlled by the mine's drainage system and either treated prior to discharge or pumped
to the process water pond or TSF.
Natural gas is lighter than air, so any release due to a natural gas pipeline failure would rise and
dissipate. If the gas cloud ignited, most of the heat would travel upward, but the initial blast and
subsequent radiation heating could affect the road and nearby environment. During dry periods, a
wildfire could result. Such failures were considered to pose relatively low risks to the assessment
endpoints and are not evaluated further in this assessment.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-1
April 2013
-------
Chapter 11
Pipeline Failures
Figure 111. Conceptual model illustrating potential stressors and effects resulting from a
concentrate pipeline failure.
product
reco'/en/-
additional step in
causal pathway'
modifying
foam '
| concentrate
I pipeline
concentrate pipeline
breaker leak
•? product
in aquatic habitats
T concentrate water
in aquatic habitats
T prod LI ct tr an sp o rt
to aquatic habitats
product re suspension
& transport
T product settling
T metal dissolution
'I chronictoxicity 'T acute toxicity
I invertebrate
abundance
4- salmon id fish
i ab u n d an c e, p r o cl u cti :• ity o r c) i •.• e r sit,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-2
April 2013
-------
Chapter 11
Pipeline Failures
Figure 112. Conceptual model illustrating potential stressors and effects resulting from a return
water pipeline failure.
LEGEND
additional step in
causal pathway
\
return water j
pipeline )
V
re turn water pipeline
breaker leak
V
'T return •/•.•ater
in aquatic habitats
V
f metals
V
I yanthates
V
'T acute toxicity
\/
4, invertebrate
abundance
\/
\/
4 salmonid fish
(abundance, productivity or diversity)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-3
April 2013
-------
Chapter 11
Pipeline Failures
Figure 113. Conceptual model illustrating potential stressors and effects resulting from a diesel
pipeline failure.
diesel
pipeline
V
diesel pipe line
breaker leal'.
t diesel
in aquatic habitats
V
-| diesfil on land
\
/
I subsurface transport
of oil
additional st^pin
causal pathway
1 chronictoxicity 1 acute toxicity
\/
4 invertebrate
abundance
V
•i- salrnonid fish
(abundance, productivity or diversity)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-4
April 2013
-------
Chapter 11 Pipeline Failures
11.1 Causes and Probabilities of Pipeline Failures
The U.S. transportation system includes more than 4 million km of pipeline, of which more than
3.8 million km are gas transmission or natural gas distribution mains and more than 280,000 km carry
hazardous liquids, primarily petroleum products (PHMSA 2012). The principal causes of failures along
these pipelines are external corrosion and mechanical damage such as impacts by excavating
equipment. Internal corrosion and material breakdown also may cause pipeline failures, but are less
common. The failure rate from impacts, such as can occur during road, pipeline, or bridge maintenance,
tends to be steady over the lifetime of a pipeline, whereas corrosion failures tend to increase with age of
the pipe.
Pipeline failures include both leaks and ruptures. Leaks are small holes and cracks that result in product
loss but do not immediately prevent the functioning of the pipeline. Ruptures are larger holes or breaks
that render the pipeline inoperable. A study of over 2 million km-yr of pipelines in Canada indicated that
leaks account for 87% of failures and ruptures account for 13% (EUB 1998). A rupture could result in
the immediate release of a significant amount of pipeline product. A leak would allow pipeline product
to escape more slowly than a rupture, but a leak could remain undetected for a much longer time,
ultimately releasing quantities comparable to or exceeding a rupture.
The most extensive pipeline failure statistics are derived from oil and gas industry data (Table 11-1).
The industry's record in terms of pipeline failures is directly relevant to the oil and gas pipelines
considered in the pipeline failure scenarios. The failure rate of metal concentrate slurry pipelines is
unknown, because few such pipelines are in operation and no published failure rates are available for
those that are in operation. The failure rates of oil pipelines are used as the best available estimate,
although it is possible that the erosive or corrosive nature of the concentrate would increase pipeline
failure rates.
Although the range of published annual failure rates for oil and gas pipelines spans more than 1 order of
magnitude (0.000046 to 0.0011 per km) (URS 2000), the range for pipelines most similar to the
assessment pipelines along the transportation corridor is much narrower. For example, the failure rate
is 0.0010 failure/km-yr for pipelines less than 20 cm in diameter (OGP 2010), 0.0015 failure/km-yr for
pipelines in a climate similar to Alaska (Alberta, Canada) (ERCB 2013), and 0.00062 failure/km-yr for
pipelines run by small operators (those operating total pipeline lengths less than 670 km) (URS 2000).
The geometric mean of these three values yields a probability of failure of 0.0010 failure/km-yr.
This overall estimate of annual failure probability, coupled with the 113-km length of each pipeline as it
runs along the transportation corridor within the Kvichak River watershed, results in an 11%
probability of a failure in each of the four pipelines each year. Thus, the probability of a pipeline failure
occurring over the duration of the Pebble 2.0 scenario (i.e., approximately 25 years) would be 95% for
each pipeline. The expected number of failures in each pipeline would be about 2.2, 2.8, and 8.6 over the
life of the mine in the Pebble 0.25, 2.0, and 6.5 scenarios, respectively. The chance of a large rupture in
each of the three pipelines over the life of the mine would exceed 25%, 30%, and 67% in the Pebble
0.25, 2.0, and 6.5 scenarios, respectively. In each of the three scenarios, there would be a greater than
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 777-April 2013
-------
Chapter 11
Pipeline Failures
99.9% chance that at least one of the three pipelines carrying liquid would fail during the project
lifetime.
Although data are insufficient to determine failure probabilities specific to the metal mining industry,
the record suggests that pipeline failures at mines are not uncommon. A review of 14 operating
porphyry copper mines in the United States (including all operating U.S. porphyry copper mines but two
that have been operating for less than 5 years) found that all had experienced pipeline spills or
accidental releases and thatpipeline failures have continued into 2012 (Earthworks 2012).
It may be argued that engineering can reduce pipeline failures rates below historical levels, but
improved engineering has little effect on the rate of human errors. Many pipeline failures, such as the
cyanide water spill at the Fort Knox mine (Fairbanks, Alaska) that resulted from a bulldozer ripper blade
hitting the pipeline (ADEC 2012), are due to human errors. Perhaps more important, human error can
negate safety systems. For example, on July 25 and 26, 2010, crude oil spilled into the Kalamazoo River,
Michigan, from a pipeline operated by Enbridge Energy. A series of in-line inspections had showed
multiple corrosion and crack-like anomalies at the river crossing, but no field inspection was performed
(Barrett 2012). When the pipeline failed, more than 3 million L (20,000 barrels) of oil spilled over 2 days
as operators repeatedly overrode the shut-down system and restarted the line (Barrett 2012). The spill
was finally reported by a local gas company employee who happened to witness the leak. The spill may
have been prevented if repairs had been made when defects were detected, and the release could have
been minimized if operators had promptly shut down the line.
Table 11 1. Studies that examined pipeline failure rates.
Study
OGP 2010
(oil pipelines)
OGP 2010
(gas pipelines)
Caleyo 2007
URS2000
(56 U.S. oil pipeline
operators)
ERCB 2013
Km-Years
Analyzed
667,000
2,770,000
34,595
28,270
1,268,370
285,000
380,331
395,479
386,930
398,253
406,974
Pipeline or Failure Parameter Assessed
Diameter <20 cm
Diameter 20-36 cm
Wall thickness <5 mm
Wall thickness 5-10 mm
1970-2004
2000-2004
Mexican gas pipelines
Mexican oil pipelines
Highest failure rate
Average failure rate
Minimum failure rate
10 smallest operators (<670 km)
10 largest operators (>6,900 km)
2000, Alberta, Canada
2007, Alberta, Canada
2008, Alberta, Canada
2009, Alberta, Canada
2010, Alberta, Canada
2011, Alberta, Canada
Annual Failure Rate
(per km-year)
0.0010
0.00080
0.00040
0.00017
0.00041
0.00017
0.0030
0.0052
0.0011
0.00028
0.000046
0.00062
0.00020
0.0033
0.0022
0.0021
0.0016
0.0015
0.0015
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-6
April 2013
-------
Chapter 11 Pipeline Failures
11.2 Potential Receiving Waters
The transportation corridor pipelines evaluated in the assessment would cross approximately
64 streams and rivers in the Kvichak River watershed, 53 of which are believed to support salmonids
and all of which could convey contaminants to Iliamna Lake. This number of crossings is much larger
than the number of hydrologic units presented in Tables 10-3 through 10-5, because hydrologic units
may contain multiple watersheds and each watershed may include crossings of multiple tributaries.
For approximately 14% of their length (15 km), these pipelines would be within 100 m of a stream or
river (Table 10-3), and for 24% of their length (27 km) they would be within 100 m of a mapped
wetland (Table 10-4). This proximity would create the potential for spilled slurry to flow into surface
waters either directly or via overland flow. Some of these wetlands include ponds that support
salmonids, but the number and distribution of salmonids in the area's wetlands are unknown.
Approximately 290 km of streams, as well as Iliamna Lake, are downstream of those crossings
(Table 10-6).
Although exposure pathways for all failure locations are considered, the quantitative analysis addressed
two stream crossings, Chinkelyes Creek and Knutson Creek, along the assessment's transportation
corridor. Channel velocities for these creeks were calculated to estimate the time it would take for a spill
to reach Iliamna Lake. Information from the Environmental Baseline Document 2004 through 2008 (PLP
2011: Chapter 15.3) was used to develop channel width and depths. Flows were calculated from
precipitation models used to determine mean annual runoff for the assessment's stream culvert analysis
(Section 10.3.2). These mean annual flows applied to the basic channel geometry yielded channel
velocities and thus travel times from the crossing to Iliamna Lake.
From the Chinkelyes Creek crossing, the creek flows 14 km to a confluence with the Iliamna River that
continues for 7.6 km to Iliamna Lake. Lake levels can be seasonally high and create a backwater effect in
the lower 3.5 km of the Iliamna River; however, most of the year the river flows freely the entire
distance to the lake shore. From the Knutson Creek crossing, the creek flows 2.6 km to Iliamna Lake. As
Knutson Creek approaches the lake, the creek is steeper than the Iliamna River and it flows freely into
the lake year-round. Total travel times to Iliamna Lake are estimated to be 240 minutes and 24 minutes
for a Chinkelyes Creek and a Knutson Creek spill, respectively. More details concerning these and other
stream crossings are presented in Section 10.3.2.
11.3 Concentrate Pipeline Failure Scenarios
11.3.1 Sources
A full pipeline break or a defect of equivalent size in the copper (+gold) concentrate pipeline (Table 6-4)
at the Chinkelyes Creek or Knutson Creek crossing would release slurry into these water bodies. This
kind of failure could result from mechanical failure of the pipe due to ground movement, vehicle impact,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment TTZ April 2013
-------
Chapter 11 Pipeline Failures
maintenance error, or material failure. Concentrate pipeline failure conditions are summarized in
Table 11-2.
In the concentrate pipeline failure scenarios, a single complete break of the pipeline would occur at the
edge of the stream, just upstream of an isolation valve. These valves would be placed on either side of
major crossings (Ghaffari et al. 2011) and could be remotely activated. Pumping would continue for
5 minutes until the alarm condition was assessed and an operator shut down the pumps. The estimated
total slurry volume draining to the stream would equal the pumped flow rate times 5 minutes, plus the
volume between the break and local high point in the pipeline (i.e., the nearest watershed boundary)
(Table 11-2). During the entire spill, gravity drainage governs the flow rate based on calculations for
free-flowing pipes.
The product concentrate slurry would have a density of 3.8 metric tons/m3 and would sink rapidly if
released into a water body at low flows. The slurry water would have a density near 1.0 metric ton/m3
and would readily mix with surface waters. No analyses of product concentrate or concentrate transport
water are available for the Pebble deposit or any other ore body in the region. To estimate the
concentration of metals and other constituents in the receiving environment, we used analyses from the
Aitik (Sweden) porphyry copper mine as described in Appendix H.
The fine particles of product concentrate would, like spilled tailings (Section 9.3), degrade habitat
quality for fish and benthic invertebrates. However, these potential physical effects would be much
lower in magnitude than for a tailings dam failure because of the much lower volume of material, and
would be less important than potential toxic effects. Thus, we focus on toxic effects rather than effects of
sediment deposition on habitat.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ,, „ April 2013
-------
Chapter 11
Pipeline Failures
Table 11 2. Conditions of concentrate pipeline spill to Chinkelyes Creek and Knutson Creek.
Condition
Spill into Chinkelyes Creek
Chinkelyes Creek
Iliamna River
Spill into Knutson
Creek
Knutson Creek
Water Flow
Discharge (m3/s)
Velocity (m/s)
Channel Length (km)
2.44
1.8
14
25.51
1.1
7.6
2.03
1.8
2.6
Pipeline Drainage and Dilution
Volume of slurry spilled (L)
Mass of concentrate solids spilled (metric tons)
Volume of aqueous phase spilled (L)
Maximum fully mixed dissolved copper
concentration (Mg/L)
Quotient15, acute copper criterion
Quotient15, chronic copper criterion
Travel time to confluence (minutes)3
76,000
67
58,000
35
13
21
130
-
3.5
1.3
2.1
110
27,000
24
21,000
22
8.1
13
24
Notes:
Blank values (-) indicate that spill is not directly into Iliamna River, would flow into it from Chinkelyes Creek.
a Confluence with Iliamna River for Chinkelyes Creek; confluence with Iliamna Lake for the Iliamna River and Knutson Creek.
b See Box 8-3 for a description of how risk quotients were calculated.
11.3.2 Exposure
Under these concentrate pipeline failure scenarios, 67 metric tons of product concentrate would be
released into Chinkelyes Creek or 24 metric tons into Knutson Creek. Based on its size and the well-
established relationship between particle size and particle mobilization and transport (commonly
represented by the Hjulstrb'm diagram), the concentrate would be transported in suspension by flows
greater than approximately 20 cm/s and would be transported as bedload between approximately 1 and
20 cm/s. Estimated mean velocities of the streams (1.8 m/s for Chinkelyes Creek and Knutson Creek and
1.1 m/s for the Iliamna River) are consistent with those described for these streams (PLP 2011), and are
well above the transport velocities. Therefore, the fine sand-sized concentrate would be carried
downstream during typical or high flows, even given that the concentrate is denser (3.8 metric tons/m3)
than typical rock (2.8 metric tons/m3 for granite) and would move less readily. Concentrate would be
deposited in any backwaters, pools, or other low-flow locations. If the spill occurred during a period of
high flow, it would be carried downstream immediately, potentially reaching Iliamna Lake within
4 hours (via Chinkelyes Creek and Iliamna River) or 0.5 hour (via Knutson Creek). Because flood flows
are a potential cause of pipeline failure at stream crossings, this is a reasonable possibility. If the spill
occurred during low flows, concentrate that is not collected would be spread downstream by erosion
during subsequent typical or high-flow periods, eventually entering Iliamna Lake. Concentrate that
entered the lake could mix into sand and gravel beaches used by spawning sockeye salmon. These
transport and deposition processes cannot be quantified with existing data and modeling resources.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-9
April 2013
-------
Chapter 11
Pipeline Failures
Table 113. Comparison of mean metal concentrations in copper concentrate from the Aitik
(Sweden) porphyry copper mine (Appendix H) to threshold effect concentration and probable effect
concentration values for fresh water. All concentrations are mg/kg dry weight.
Concentrate
Constituents
Ag
As
Ba
Bi
Cd
Co
Cu
Ga
In
Mn
Mo
Ni
Pb
Sb
Te
Th
Tl
U
V
Zn
Sum of metals
Concentrations
>10
12
59
44.9
2.4
53.9
>10,000
0.88
2.35
345
1,100
72.1
64.9
43.4
4.1
1.5
0.2
2.2
23
2,190
TEC"
9.8
-
0.99
-
32
-
630
-
23
36
-
-
-
-
120
TEC Quotient"
1.2
-
2.4
-
>310
-
0.55
-
3.1
1.8
-
-
-
-
18
>340
PECa
33
-
5.0
-
150
-
1,200
-
49
130
-
-
-
-
459
PEC Quotient"
0.36
-
0.48
-
>67
-
0.29
-
1.5
0.50
-
-
-
-
4.8
>75
Notes:
Blank values (-) indicate that values are not available.
a TECs and PECs are consensus values from MacDonald et al. (2000), except for Mn values, which are the TEL and PEL for Hyalella azteca 28 d
tests from Ingersoll etal. (1996).
b See Box 8-3 for a description of how risk quotients were calculated.
TEC = threshold effect concentration; PEC = probable effect concentration; TEL = threshold effect level; PEL = probable effect level.
The estimated annual failure rate of one per 1,000 km per year (Section 11.1) results in an estimated
failure rate of 0.11 per year for the 113 km of concentrate pipeline within the Kvichak River watershed.
If the probability of a pipeline failure is independent of location, and if it is assumed that spills within
100 m of a stream could flow to that stream, a spill would have a 14% probability of entering a stream
within the Kvichak River watershed. This would result in an estimate of 0.015 stream-contaminating
concentrate spills per year, or 1.2 stream-contaminating concentrate spills over the duration of the
Pebble 6.5 scenario (approximately 78 years). In other words, we expect 1 to maybe 2 such spills in the
Pebble 6.5 scenario. Similarly, a spill would have a 3 5% probability of entering a wetland, resulting in an
estimate of 0.038 wetland-contaminating spills per year or 2 wetland-contaminating spills in the
Pebble 6.5 scenario. A portion of those wetlands would be ponds or backwaters that support fish.
Spills from the pipeline failure would contaminate 2.6 km of Knutson Creek or 14 km of Chinkelyes
Creek and 7.6 km of Iliamna River with leachate and product concentrate before entering Iliamna Lake.
The potential extent of contamination of wetlands cannot be readily estimated.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-10
April 2013
-------
Chapter 11 Pipeline Failures
As with a tailings spill (Chapter 9), lexicologically relevant exposures could occur via multiple routes in
the event of a concentrate pipeline spill. During and immediately following a spill, organisms would be
acutely exposed to leachate (the slurry water that has leached ions from the product concentrate) and
suspended particles. After a spill, product concentrate deposited on a stream or lake bed would result in
chronic aqueous exposures to pore water and acute aqueous exposures during resuspension events.
Unlike the tailings spill, which would inevitably enter a stream and its floodplain, a slurry spill might
directly enter a stream, pond, or wetland; it might flow overland to a nearby water body; or it might flow
across the landscape without reaching water. Terrestrial slurry deposits are likely to be collected by the
operator, so rain and snowmelt are unlikely to leach those concentrate deposits and contaminate
streams. However, the spilled leachate from the pipeline slurry could enter a stream, wetland, pond, or
lake by overland or groundwater flow. Contaminated groundwater could upwell through the gravels and
cobbles of streams or deltaic gravels and sands in Iliamna Lake, exposing benthic invertebrates, fish
eggs, and larvae to toxic concentrations.
11.3.2.1 Aqueous Phase Chemical Constituents
The concentrate slurry is estimated to contain 77% water by volume, with dissolved constituents that
include dissolved salts of the product, and trace metals, as well as process chemicals. Copper is the
principle ecotoxicological concern, because it is the principal product and is highly toxic to aquatic life.
Analyses of aqueous filtrate from samples taken on 3 different days, from a concentrate pipeline at a
porphyry copper mine with a separation process similar to that considered in the mine scenarios
(Section 6.1.3), reported copper concentrations of 500, 664, and 800 ug/L (Adams pers. comm.). The
mean of these values (655 ug/L) was used as the estimated copper concentration in this assessment.
Sodium ethyl xanthate is the highest risk ore-processing chemical due to its relatively high toxicity. We
were unable to find an estimate of process chemical concentrations in the concentrate slurry, but
xanthate concentration would be 1.5 mg/L, if we assume that it occurs in the concentrate slurry at the
same concentration as in tailings slurry (NICNAS 1995). Unlike the metals, xanthate would degrade, but
its environmental half-life is approximately 260 hours (pH 7, 25°C) (NICNAS 2000), so it could persist
long enough to cause significant exposures until diluted in Iliamna Lake.
Flows in the potential receiving streams vary considerably. Measurements in streams along the
transportation corridor in 2004 and 2005 yielded a maximum observed flow of 58,000 L/s in the
Iliamna River and a minimum observed flow of 2.8 L/s in an unnamed stream (PLP 2011). Thus, full
mixing of spilled leachate could result in as much as a 33-fold dilution, but in smaller streams dilution
effectively would not occur. Of 12 monitored streams along the transportation corridor, only two had
observed flows in August 2004 (an estimate of summer low flow) that were greater than the estimated
leachate flow (PLP 2011: Table 7.3-10).
11.3.2.2 Solid Phase Chemical Constituents
If spilled product concentrate entered a stream, pond, or wetland directly or by overland flow or
erosion, it would flow for some distance, settle, and become the substrate for invertebrates and possibly
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1111 April 2013
-------
Chapter 11 Pipeline Failures
salmon eggs and fry. In streams, it would be carried downstream by the current and would collect in
pools, behind debris, and in other localized low-flow areas. Some would settle into the cobble substrate
until high flows mobilized the bed. Much of the product concentrate could wash into Iliamna Lake,
where it could contribute to the substrate for spawning sockeye salmon.
Metal concentrations in the solid phase are expected to be similar to those of the Aitik copper product
concentrate (Table 11-3). Settled concentrate would be leached, resulting in direct aqueous exposure of
benthic invertebrates and fish eggs and larvae that inhabit the substrate to concentrations similar to the
experimental leachate (Table 11-4). Local accumulation in streams could result in local exposures to
nearly pure concentrate and leachate. However, concentrate in Iliamna Lake would be distributed and
diluted to an extent that could not be estimated. Dietary exposure offish is not considered, because
invertebrate abundance would be greatly diminished due to toxicity in sediment formed of spilled
concentrate, even with considerable dilution by clean sediment.
11.3.3 Exposure-Response
Acute water quality criteria (criterion maximum concentrations [CMCs]), chronic criteria (criterion
continuous concentrations [CCCs]), and equivalent benchmark values are used as thresholds for
aqueous toxicity. Consensus sediment quality guidelines are used as thresholds for sediment solids
toxicity. These aqueous and sediment benchmark values are discussed in Sections 8.2.2 and 9.5.2,
respectively. The biotic ligand model (BLM) generates low acute and chronic water quality criteria and
other toxicity values because of the extreme water chemistry of the leachate and receiving waters
(Section 8.2.2). However, the parameters are all within calibration range of the model (exceptfor
alkalinity and dissolved organic carbon were set to minimum values, because they were absent from the
leachate, which slightly raises criteria values) (HydroQual 2007).
In addition to the product concentrate and its dissolved constituents, the slurry would contain process
chemicals. Sodium ethyl xanthate is sufficiently toxic that it has been used as a pesticide (NICNAS 2000).
Exposure-response information for xanthates is summarized in Section 8.2.2.
11.3.4 Risk Characterization
Toxicological risk characterization is performed primarily by calculating risk quotients, from the ratios
of exposure concentrations to aquatic toxicological benchmarks (Box 8-3). However, it also includes
consideration of actual concentrate spills, the potential for remediation, and site-specific factors.
11.3.4.1 Concentrate Pipeline Failure Scenarios
The concentrate pipeline failure scenarios and resulting spill would release 58,000 L of leachate to
Chinkelyes Creek or 21,000 L to Knutson Creek (Table 11-2). Risks to aquatic biota would result from
direct exposure to the aqueous phase of the slurry, the deposited concentrate, and in situ leachate from
the concentrate.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±±±2 April 2013
-------
Chapter 11 Pipeline Failures
The estimated dissolved copper concentration in the aqueous phase of the slurry is 655 ug/L, which is
243 times the acute water quality criterion and 390 times the chronic criterion for Upper Talarik Creek,
the nearest stream with complete water quality data (Table 8-10). Clearly, this would be sufficient to
cause severe toxic effects in small streams, large streams at low flow, and wetlands. The dilution
provided by the receiving waters considered here would not be enough to prevent acute, much less
chronic, toxicity based on the copper criteria. These criteria are based on toxicity to sensitive
invertebrates, so the food base for salmonids could be severely reduced.
In all three streams, the diluted values are below the BLM-derived acute lethal levels for rainbow trout,
so a fish kill would not be expected (Table 8-13). Therefore, copper is not predicted to cause a kill of
adult salmonids in the receiving streams once mixing has occurred, but localized mortality might occur
in the mixing zone in the absence of avoidance behavior. However, fully diluted concentrations are
above (Chinkelyes Creek) or equal to (Knutson Creek) the chronic toxicity value for rainbow trout,
suggesting that fry would be affected.
Sodium ethyl xanthate, after fully mixing in Chinkelyes and Knutson Creeks, would occur at
approximately 0.1 and 0.07 mg/L, which is at the low end of observed acutely lethal concentrations for
aquatic biota and below the observed median lethal concentrations for rainbow trout (Table 8-13).
Hence, the processing chemicals would contribute to acute toxicity in sensitive species.
The occurrence of acute toxicity depends on the exposure duration relative to the concentration. The
5.0- to 7.6-minute exposure duration may be sufficient to cause acute injury or lethality to invertebrates
or fish in receiving streams, given the high concentrations of copper (the rate of toxic response is a
function of the concentration). However, it would be more likely to cause acute effects in backwaters
and ponds that retained spilled water, and those areas are important rearing habitat for salmon
(Appendix A).
Where the 24 to 67 metric tons of concentrate settled, sediment and benthic invertebrates and fish eggs
and fry would be exposed. The Aitik concentrate exceeds the sediment probable effect concentration
(PEC) for copper by more than a factor of 67 (Table 11-3). Hence, based on experience with other high-
copper sediments, any Pebble concentrate would be certain to cause toxic effects on benthic organisms,
including invertebrates and fish eggs and larvae. Because copper is aversive to salmonids (Goldstein et
al. 1999, Meyer etal. 2010), the chronic leaching of copper from deposited product concentrate may
prevent returning salmon from using a contaminated stream or river.
Exposure to pore water in sediments consisting of spilled product concentrate would be chronic. The
screening assessment performed here on Aitik concentrate leachate suggests that spilled concentrate
would cause severe toxic effects (Table 11-4). The 8,400 ug/L of dissolved copper in leachate would be
sufficient to kill benthic or epibenthic invertebrates and fish eggs and larvae.
At mine closure, concentrate and return water pipelines would be removed. Therefore, these risks
would be limited to the approximately 78-year maximum operational life of the mine.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±±±3 April 2013
-------
Chapter 11
Pipeline Failures
Table 114. Aquatic toxicological screening of leachates from Aitik (Sweden) mine copper
concentrate (Appendix H) based on acute and chronic benchmarks (water quality criteria or
equivalent values) and quotients of concentrations divided by benchmark values. Units presented in
jjg/L unless specified otherwise.
Analyte
pH (S.U.)
Spec, conductivity (|jS/cm)
Alkalinity (mg/L)
Sulfate (mg/L)
Si02(mg/L)
Ag
Al
As
Ba
Ca
Cl
Cd
Co
Cr
Cu
F
Fe
K
Mg
Mn
Mo
Na
Ni
Pb
Sb
Se
U
Zn
Sum of metals
Concentrations
5.36
264
0
121
58.8
<1
844
<1
38.4
26,900
800
3.53
136
<1
8,400
1,600
210
3,980
4,450
644
<2
889
484
10.6
12.8
7.3
10.5
1,300
Acute/Chronic Benchmarks
6.5-9
-
-
0.90/-b
750/87
340/150
46,000/8,900
19/11
1.73/0.22b
89/2.5
500/65b
11. 61/7. 9b
0. 046/0.028=
-
350/-
-
760/693
32,000/72
-
410/46b
54/2. lb
14,400/1,600
-/5.0
33/15
100/100b
Quotients3
-
-
-
-------
Chapter 11 Pipeline Failures
reported that the 2004 spill was controlled in less than 2 hours and water for drinking and irrigation
was not contaminated (Minera Alumbrera 2004). They do not mention an earthquake, do not explain
why control required 2 hours, and attribute the failure to "an existing outer mark on the pipe" (Minera
Alumbrera 2004). They reported other pipeline failures with concentrate spills in 2006 and 2007, but
not in other years (Minera Alumbrera 2004, 2005, 2006, 2007, 2008, 2009, 2010). They claimed that
those releases were small due to automatic shutoff, that concentrate did not reach water, and that "no
hazard is involved in concentrate handling since it is a harmless product consisting of ground rock"
(Minera Alumbrera 2006). Composition of this ground rock included 28% copper and 32% sulfur
(Minera Alumbrera 2006).
Operators subsequently built collection pits at pumping stations, monitored streams at pipeline
crossings, and brought water into the community of Amanao in part to mitigate effects of "potential
pipeline failure" (Minera Alumbrera 2008, 2010). They stated based on monitoring that pipeline
crossings of streams have no adverse effects on biodiversity, but they do not report monitoring to
address the effects of or recovery from the 2004 spill (Minera Alumbrera 2010). Although the interval
during which Minera Alumbrera has provided sustainability reports is too short to reliably estimate an
annual failure probability, it is notable that, despite International Organization for Standardization
14001 certification of the pipeline, it failed and released concentrate in 3 of 7 years.
More recently (July 25, 2012), a joint broke on the product slurry pipeline for the Antamina copper and
zinc mine in Peru (Briceno and Bajak2012, Taj and Cespedes 2012). It released 45 metric tons of slurry
over 2 hours, of which 3 metric tons escaped the containment area. Local villagers intervened to stop the
flow of slurry to the nearby Rio Fortenza. A mine spokesman stated that the river showed no signs of
contamination and the material was only an irritant, although a company document called the
concentrate very toxic (Taj and Cespedes 2012). An Associated Press photo shows workers in white
suits apparently cleaning a channel. News reports and Minera Antamina's press releases on the event
emphasized human health effects: 210 people received medical treatment and 45 were hospitalized,
apparently due to inhalation of aerosolized slurry. People reported a strong pesticide odor, which
suggests significant concentrations of a xanthate collector chemical, but no analyses have been reported.
Ecological effects are unknown. Antamina is a modern mine (operation began October 1, 2001) where
sustainability is given a higher priority than cost or profitability (Caterpillar Global Mining 2009). As in
the mine scenarios evaluated here, the pipeline is buried, except at bridges, and is monitored using a
parallel fiber optic system.
Product concentrate spills from pipeline failures have also occurred at the Bingham Canyon mine in
Utah. Between May 31 and June 2, 2003, operators reported to the U.S. Coast Guard's National Response
Center a spill of 70 tons of product concentrate from a pipeline failure. On October 2, 2009, they
reported a pipeline leak that spilled 14,00 gallons (5,300 L) of copper concentrate.
Although the Alumbrera, Antamina, and Bingham Canyon cases do not provide evidence concerning the
ecological effects of a concentrate spill, they do support the plausibility of pipeline failures leading to
concentrate spills. Our estimated pipeline failure rate of one per 1,000 km per year (Section 11.1)
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 11~15 April 2013
-------
Chapter 11 Pipeline Failures
implies a failure rate of 0.32 per year for the 316-km Alumbrera pipeline, which is similar to the
0.43 observed rate at Alumbrera from 2004 to 2010.These cases indicate that concentrate pipeline
failures do occur at modern copper mines operated by large international mining companies, and that
they can result in spills that are potentially larger than our assumptions indicate.
11.3.4.3 Concentrate Spill Remediation
Remediation of a product concentrate spill would be less problematic than remediation of a tailings
impoundment spill. The concentrate is valuable, it would be spilled near a road, and the volume would
be much smaller than a potential tailings spill. Hence, remediation would likely occur relatively quickly
by excavation or dredging and trucking back to the mine if the spill occurred on land or in a wetland.
However, because concentrate would be carried downstream by high or typical flows in the receiving
streams, substantial recovery of material spilled into a stream is unlikely except possibly during low-
flow periods (less than one-ninth of mean flows). The proportion recovered by dredging would depend
on the circumstances, the rapidity of response, and the balance between the desire to minimize habitat
damage and to reduce potential toxic effects. If the spill was associated with high flows, it is likely that
little of the material would be recovered from a stream even if the entire stream was dredged. Dredging
in Iliamna Lake might be feasible if concentrate was not too dispersed or diluted by other sediment.
11.3.4.4 Weighing and Summarizing the Evidence
Past experience with pipelines, in general, and with the Alumbrera, Antamina, and Bingham Canyon
product concentrate pipeline failures, in particular, suggests that pipeline failures and product spills
would be likely under the Pebble 6.5 scenario. A concentrate spill into a stream is likely to kill
invertebrates and early life stages offish immediately. If it is not remediated (and remediation of
streams may not be possible), it would certainly cause long-term local loss offish and invertebrates. The
settled concentrate would become sediment, which would be toxic to fish and invertebrates in the
receiving streams for many years. Ultimately, it would reach Iliamna Lake, where it could be toxic to the
eggs and larvae of sockeye salmon until it was sufficiently mixed with or buried by clean sediment. The
length of streams affected in the scenarios would be 14km of Chinkelyes Creek and 7.6 km of Iliamna
River for a release to Chinkelyes Creek or 2.6 km of Knutson Creek for a release there. The area of the
lake that would experience toxic effects cannot be estimated at this time.
Overall, the available lines of evidence for effects of a concentrate spill are positive for the occurrence of
acute and chronic toxic effects (Table 11-5). The quality of the exposure-response information is good,
but the quality of the exposure information for the deposited concentrate and its leachate is uncertain
because of the uncertain potential for dispersal in streams. The analogous spills provide no information
on exposure or effects beyond confirming that concentrate spills do occur. However, this evidence
supplements the more extensive experience with oil pipelines (Section 11.1), which suggests that a spill
is likely.
If the spill could be remediated, some fraction of the concentrate (but none of the leachate) could be
recovered and the extent of chronic (but not acute) toxic effects would be diminished. The proportion of
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 11~16 April 2013
-------
Chapter 11 Pipeline Failures
concentrate recovered would depend on the location, time of year, diligence of the operator, and amount
of physical damage due to remediation that is considered acceptable. Concentrate spilled into streams is
unlikely to be recovered unless flows were particularly low. Recovery of the concentrate would require
excavation of stream beds, wetlands, or uplands, depending of the location of a spill. When determining
how thoroughly to excavate and, in particular, how far downstream to dredge the stream, reduction in
toxicity would need to be balanced against habitat destruction.
The effects of a spill on salmonid populations would depend on the receiving waters. Streams along the
transportation corridor that might receive a spill, as described in Section 10.1, are quite variable.
Chinkelyes Creek receives an average of more than 9,000 spawning sockeye salmon; it flows to the
Iliamna River, which receives an average of more than 100,000 sockeye spawners (Table 10-2). Knutson
Creek receives an average of 1,500 sockeye spawners and flows to Knutson Bay, which receives an
average of 73,000 beach spawning sockeye (Table 10-2). Not all of those salmon spawn below stream
crossings, but copper leaching from concentrate spills could be aversive to salmon and thereby reduce
spawning production along the entire stream lengths. Also, the concentrate deposited in Knutson Bay
would persist and could render a considerable area unsuitable for spawning and rearing for years. In
any case, these values indicate that a non-trivial number of spawners and potential salmon production
would be at risk.
Potential effects on those salmon and other fishes in the receiving waters would include the following.
• Reduced production of salmon fry and parr and all life stages of other salmonids from the loss of
invertebrate prey resulting from extensive acute lethality during the spill and persistent chronic
toxicity in areas where the concentrate deposited.
• Loss of a year-class of salmon and other salmonids due to direct acute toxicity during and
immediately following a concentrate spill.
• Loss of salmon spawning habitat due to avoidance of copper in areas of deposition and possibly in
the entire stream, if aqueous concentrations from leaching concentrate were sufficiently high.
• Persistent chronic toxicity to salmonid eggs and fry in areas of concentrate deposition, where it is
not aversive to spawning adults.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . *-, April 2013
-------
Chapter 11
Pipeline Failures
Table 115. Summary of evidence concerning risks to fish from a product concentrate spill. The risk
characterization is based on weighing four lines of evidence for different routes of exposure. All
evidence is qualitatively weighed (using +, 0, ) on three attributes: logical implication, strength, and
quality. Here, all lines of evidence have the same logical implication that is, all suggest a
concentrate spill would have adverse effects. Strength refers to the overall strength of the line of
evidence, and quality refers to the quality of the evidence sources in terms of data quality and
relevance of evidence to the spill scenario.
Route of Exposure
Source of Evidence (Exposure/E-R)
Dissolved copper
Measurements from analogous
mine and dilution
model/Laboratory-based
benchmarks
Concentrate particles
Undiluted concentration from
analogous mine/ Field-based
benchmarks
Concentrate leachate
Leachatefrom analogous mine/
Laboratory-based benchmarks
Actual spills
Amount spilled/None
Summary Weight of Evidence
Logical
Implication
+
+
+
0
+
Strength
++
++
++
0
++
Quality
Exposure
+
0
0
0
0
E-R
++
+
++
0
+
Results
Lethality to invertebrates is certain
and sensitive larval fish may also
be killed
The concentrate would clearly form
toxic sediment but its distribution
in unclear
Invertebrates and fish in sediment
would experience toxic effects
unless the concentrate was highly
diluted
The record indicates that
concentrate spills occur but
exposure and effects have not
been studied
A spill is likely to occur and toxicity
to aquatic biota is highly likely
Notes:
E-R = exposure-response relationship.
11.3.5 Uncertainties
Based on multiple lines of evidence, it is certain that a spill from a product concentrate pipeline into a
stream would cause toxic effects. However, there are uncertainties regarding individual pieces of
evidence, which are summarized below.
• The composition of the product concentrate and its leachate are uncertain, because they are based
on a surrogate material and because leaching test conditions are inevitably somewhat artificial.
Copper concentrations in North and South American copper concentrates generally fall in the 200 to
340 mg/kg range, so a variance of a factor of 2 is a reasonable estimate for potential variance in
Pebble deposit concentrate from Aitik concentrate. Hence, uncertainty concerning the major source
of toxicity is not large, and therefore, it is implausible that the concentrate and its leachate would be
nontoxic to aquatic biota. An informal internet search for copper concentrate compositions suggests
that minor metals differ by an order of magnitude among copper concentrates. Thus, it is possible
that metals other than copper may be significant contributors to toxicity.
• The copper concentration of the aqueous fraction of the slurry is also based on analyses from an
existing mine. However, the ore type and processing are believed to be very similar. Therefore, the
uncertainty is estimated to be at least a factor of 2 but no more than 5. Therefore, effects on
invertebrates are certain but effects on fish may not occur or may be more severe than estimated.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-18
April 2013
-------
Chapter 11 Pipeline Failures
• The composition of the aqueous fraction of the slurry is unknown for constituents other than
copper. While it is certain that copper is by far the most toxic metal in the slurry, the composition of
other constituents, particularly the xanthates, is unknown. Sodium ethyl xanthate is highly toxic and
might increase the toxicity of a spill. Combined metal toxicity would make some difference but is
unlikely to change the qualitative conclusions.
• The 5-minute spill duration is uncertain, and our estimate appears to be conservative. For example,
Trans Canada's risk assessment for the Keystone XL pipeline assumed that the time to detection
would range from 90 days for a small leak (1.5% of pumping volume) to 9 minutes for a large leak
(50% of pumping volume) and that an additional 2.5 minutes would be required for the shutdown
sequence (DNV Consulting 2006, O'Brien's Response Management 2009). Therefore, a large spill like
the one assessed here would leak for 11.5 minutes based on a state-of-practice design from an
experienced company. This is more than twice our assumed duration.
• The 5-minute spill duration depends on successful operation of a remote shutoff. The potential for a
larger spill if the shutoff failed (e.g., if an earthquake damaged the pipeline and the shutoff system)
or was overridden by the operators is unknown. There are precedents for large spills but not
enough data to quantify the risk.
• The frequency and location of spills are also uncertain. The extensive experience with oil and gas
pipelines provides probabilistic estimates, but these estimates vary considerably among studies.
The more directly relevant experiences with concentrate pipelines at Alumbrera, Antamina, and
Bingham Canyon mines suggest that estimates based on oil and gas pipeline failure rates are
consistent with mining-related pipeline failures.
11.4 Return Water Pipeline Failure Scenarios
A spill from a return water pipeline would result in an acute aqueous exposure (Table 11-6), as
discussed above for a concentrate spill. The return water is expected to be the same as the aqueous
phase of the concentrate slurry (i.e., it would not be treated at the port), although estimated flow rates
would differ. Hence, the copper concentration in the return water is assumed to be the mean of analyses
of the aqueous phase of slurry from a Rio Tinto mine (655 ug/L). Because of the short duration of the
spill and the absence of a persistent solid phase, toxicity is not expected to be severe, and effects are
most likely in low-flow habitats such as backwaters, ponds, and bays. We know of no analogous return
water pipeline failures that might be used to assess this risk; however, experience with pipelines in
general suggests that multiple failures and spills would occur over the life of the mine, and at least one
would be expected to occur at or near a stream (Section 11.1).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ,, . p April 2013
-------
Chapter 11
Pipeline Failures
Table 11 6. Conditions of return water pipeline spill to Chinkelyes and Knutson Creeks.
Condition
Spill into Chinkelyes Creek
Chinkelyes Creek
Iliamna River
Spill into Knutson Creek
Knutson Creek
Water Flow
Discharge (m3/s)
Velocity (m/s)
Channel Length (km)
2.44
1.8
14
25.51
1.1
7.6
2.03
1.8
2.6
Pipeline Drainage and Dilution
Volume spilled (L)
Maximum concentration dissolved copper (Mg/L)
Travel time (minutes)
57,000
36
130
-
3.6
110
20,000
21
24
11.5 Diesel Pipeline Failure Scenarios
As with the product concentrate pipeline, effects of a diesel pipeline failure would depend on many
factors, including pipeline design, location of the pipeline failure along the transportation corridor, and
time of year at which the pipeline failure occurred. Conditions of the diesel pipeline failure scenarios are
presented in Table 11-7.
11.5.1 Sources
11.5.1.1 Pipeline Failure
The volume of material released from a pipeline leak would depend on the type of failure, rate of loss
from the pipe, pumping rate, duration of the leak, diameter of the pipe, distance to the nearest shutoff
valves, and time until those valves are closed. For the purposes of this assessment, we evaluate a full
break or a defect of equivalent size in the diesel pipeline that occurs at a stream crossing, thereby
releasing fuel into that aquatic ecosystem. Characteristics of the pipeline are described in Table 6-4. This
could occur as a result of mechanical failure of the pipe from ground movement, vehicle impact, material
failure or other cause. We analyzed spills to two streams that would be crossed by the transportation
corridor, Chinkelyes Creek and Knutson Creek (Section 11.2).
11.5.1.2 Diesel Fuel Composition
Under the diesel pipeline failure scenarios, the pipeline would contain fuel from one of the Alaskan
refineries and would have a composition similar to those presented by Geosphere and CH2M Hill
(2006). Diesel fuel is a mixture of many hydrocarbon compounds, and the composition is a function of
the petroleum feedstock source and the refining process. The type and amount of water-soluble
hydrocarbons in the diesel determines the dissolved aqueous concentration when mixed with water.
The most soluble compounds in diesel are the volatile aromatic hydrocarbons benzene, toluene,
ethylbenzene and xylene (together, BTEX). Most diesel fuels have a low proportion of these soluble
compounds and, therefore, have low solubilities. The bulk of diesel fuel is made up of heavier
hydrocarbons that are essentially insoluble. A study of the composition of four diesel fuels from Alaskan
refineries shows less than 2% BTEX and resulting diesel solubilities from 1.89 to 7.81 mg/L.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-20
April 2013
-------
Chapter 11
Pipeline Failures
In the analysis of concentrations and solubilities, we incorporate all hydrocarbon compounds in the
diesel samples and calculated the solubility based on Raoult's Law to account for effects of the mixture
on the solubility of individual compounds.
Table 11 7. Conditions of diesel pipeline spill to Chinkelyes and Knutson Creeks.
Condition
Spill into Chinkelyes Creek
Chinkelyes Creek
Iliamna River
Spill into Knutson Creek
Knutson Creek
Water Flow
Discharge (m3/s)
Velocity (m/s)
Channel Length (km)
2.44
1.8
14
25.51
1.1
7.6
2.03
1.8
2.6
Pipeline Drainage and Dilution
Flow rate while draining (m3/s)
Flow rate while pumping (m3/s)
Release time— draining (minutes)
Release time— pumping (minutes)
Volume— draining (m3)
Volume % diesel to water in stream at spill
Mass diesel in stream at input (mg/L)
Maximum concentration dissolved diesel (mg/L)
Distance traveled during release (km)
Travel time (minutes)
0.044
0.005
11
5
30
2.0%
16,000
1.9-7.8
1.2
130
1,600
1.7-7.3
110
0.019
0.005
5.5
5
8
1.2%
8,500
1.9-7.8
0.7
24
11.5.2 Exposure
11.5.2.1 Background
A failure of the diesel pipeline under these scenarios could occur in the buried or above-ground
portions. An above-ground failure would occur at a bridged stream or river crossing. An underground
failure would result in diesel leaking into the soil and flowing down-gradient as in the Trans-Alaska
pipeline failure described in Section 11.5.3.3. If the underground failure occurred below a stream, it
would float upward and into the surface water. An above-ground failure would release diesel directly to
a river or stream, a wetland, or upland soil.
The behavior of diesel fuel in fresh water is less well-studied than the behavior of crude oil or diesel in
marine environments. Diesel fuel has a density of less than 1.0 metric ton/m3 and floats on water. It
typically dissolves or evaporates within a day. In turbulent stream reaches, diesel would form small
droplets suspended in the water column.
The soluble fraction would mix into the streamflow, be transported by advection and dispersion, and
flow with the water. Solubility decreases with temperature, so in colder temperatures a smaller amount
is dissolved in the stream. The soluble fraction is attenuated through dilution (advection and
dispersion), biological activity, photodegradation, and aeration in turbulent streams, but is renewed by
dissolution from the floating oil. The soluble compounds are also susceptible to evaporation from the
floating oil, which typically occurs at a faster rate than dissolution. The soluble fraction compounds have
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-21
April 2013
-------
Chapter 11 Pipeline Failures
relatively short residence times in water and sediments (Hayes et al. 1992) and can be reduced to below
detection levels in a few days or weeks, depending on site-specific conditions.
Diesel components that are lighter than water and have low solubility tend to spread on the surface until
forming a thin film or sheen less than 0.1 mm thick. As the diesel spreads, it is more susceptible to
destruction by evaporation, dissolution, and photodegradation but is also more likely to contact and
attach to suspended sediments and shorelines. Most of the spilled diesel would flow with the stream
until it reaches Iliamna Lake and dissipates. The pour point of diesel (the temperature below which the
oil will not flow) is approximately -7°C (20°F); thus, if the spill occurs during cold weather, the diesel
would be less likely to spread and would instead form globs or strings and become suspended within the
water column. For example, a 1999 cold-weather diesel spill in the Delaware River resulted in more than
90% of the diesel forming globules that were not visible from the surface (Overstreet and Gait 1995).
Oil dispersed in the water column can adhere to fine-grained suspended sediments that settle and
deposit on stream edges and bottoms in low-energy areas. Depending on the source of the diesel, there
may be a significant portion of compounds that are heavier than water and, therefore, sink, sorb to
sediments, and persist longer than the dissolved fraction. In wetlands or pools and slack water areas of
streams, a large percentage of spilled diesel can be deposited in the sediments.
When spilled on ice, diesel is viscous and forms tar-like accumulations on the surface. Lighter diesel
components can penetrate the ice, become trapped within the ice structure, and be released as the ice
melts. If, as is more likely with buried pipelines, the spill is trapped below the ice it would spread and
stick to the underside of the ice in thin layers. Because cold temperatures reduce the solubility of diesel
components, less would be dissolved in the stream water (NOAA and API 1994). As the ice breaks up
and melts, the diesel would be released from the ice and mix with the stream water.
Because of its low viscosity (except in cold weather), diesel spilled onto the land tends to be rapidly
absorbed by soil so that an above-ground spill on land could soon resemble an underground spill. In this
area, where the groundwater surface tends to be shallow, spilled diesel would flow on top of the
groundwater and a fraction would dissolve in that groundwater. It would then flow down-gradient to
any nearby stream, possibly passing through wetlands on the way. Upon reaching a stream, it could pass
into the channel through the gravels in which salmon, trout, and Dolly Varden spawn. In some locations,
it might flow to Iliamna Lake and pass through a deltaic spawning beach used by sockeye salmon. Diesel-
contaminated soil could episodically contaminate water when the water table rises following rain or
snow melt. The extent to which eggs or fry are exposed by this route following a spill would depend on
the specific structure of the site of the spill. Given the abundance of streams, wetlands, and shallow
groundwater in the area crossed by the diesel pipeline, some variant of this route of exposure is likely.
However, saturated soils and particularly those that are frozen could result in overland flow of diesel
fuel rather than groundwater flow.
The primary cause of toxicity to aquatic organisms in oil spills is direct exposure to the dissolved
fraction. Exposure via this route would occur immediately following a direct spill to a stream or wetland
as the oil dissolved, resulting in an acute exposure. Longer exposures to dissolved oil could result from
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±±22 April 2013
-------
Chapter 11 Pipeline Failures
slow releases of oil from terrestrial spills, flows from oiled wetlands, or the gradual dissolution of oil
sorbed to sediments or plant materials. Oil spills can indirectly expose aquatic organisms to low
dissolved oxygen as microbes decompose the oil.
Which of these transport and exposure processes occur in a spill depends on the spill location. The
number and nature of water body crossings are the same as for the potential product concentrate
pipeline (Section 11.3.1).
11.5.2.2 Transport and Fate
Under the diesel pipeline failure scenarios, a pipe failure would result in release of diesel directly into
either Chinkelyes or Knutson Creek at mean streamflows (Table 11-7). The spill at Knutson Creek would
release 8,000 L of diesel into approximately 800,000 L of stream water, resultingin a 1:100 dilution. At
Chinkelyes Creek, the spill would release approximately 30,000 L of diesel into 1.5 million L of stream
water, resultingin a 1:50 dilution. Ata typical diesel density of 850 g/L, this would result in 8,500 and
16,000 mg diesel/L water in Knutson and Chinkelyes Creeks, respectively. Both of these dilutions are
less than the minimum aqueous volume required to get below the saturation of the diesel, if the
dissolved hydrocarbons are well-mixed. This conclusion is based on calculation of the minimum volume
of water required for diluting each component to a concentration below saturation. For benzene, the
minimum volume of water required for dilution below saturation is 169 to 225 L benzene/L diesel; all
other components would require higher dilutions. Thus, it is reasonable to assume that at both spill
locations the diesel will be at saturation (i.e., at concentrations between 1.89 and 7.81 mg/L) in the
receiving waters. Concentrations in the Iliamna River would be lower due to depletion of benzene. The
benzene concentration would fall below its Raoult's saturation limit resulting in a diesel concentration
of 1.74 to 7.32 mg/1 and a saturation of 92 to 94%.
11.5.3 Exposure-Response
Diesel is considered to be one of the most acutely toxic petroleum products (NOAA 2006), but its
composition is variable. Although a model exists for estimating the acute aquatic toxicity of petroleum
products from their chemical composition (Redman etal. 2012), the composition of diesel that would be
piped to the mine is unknown. For example, the compositions of water-soluble fractions of two brands
of Alaskan diesel fuel were found to be C4-C6 non-aromatic hydrocarbons (0.4-1.2 mg/L), benzene
(0.03-0.2 mg/L), toluene (0.03-0.2 mg/L), and C2 benzenes (0.005-0.1 mg/L) (Guard et al. 1983). Given
this variance in composition, data from laboratory tests and field studies of various whole diesel oils are
used in this section to indicate the range of toxic effects observed in response to different exposures.
11.5.3.1 State Standards
According to Alaska water quality standards (ADEC 2011), total aqueous hydrocarbons in the water
column may not exceed 15 ug/L and total aromatic hydrocarbons in the water column may not exceed
10 ug/L. The standards state (ADEC 2011): "There maybe no concentrations of petroleum
hydrocarbons, animal fat, or vegetable oils in shoreline or bottom sediments that cause deleterious
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±±23 April 2013
-------
Chapter 11 Pipeline Failures
effects to aquatic life. Surface waters and adjoining shorelines must be virtually free from floating oil,
film, sheen, or discoloration."
11.5.3.2 Laboratory Tests
Laboratory tests of the toxicity of petroleum and its derivative fuels to aquatic organisms are performed
with either an oil-water dispersion or a dissolved solution, called the water-soluble fraction. Dispersions
are created by adding oil to water at prescribed ratios and mixing. The vigor and duration of the mixing
is variable, ranging from gentle mixing with a stirring rod to extended mixing with a magnetic stirrer.
The resulting dispersion may have an oil layer on the surface as well as suspended oil droplets, although
most tests attempt to avoid suspended material. The oil layer may be left in the test container, but more
often the aqueous material is drawn off for the test. Results may be expressed as mg diesel/L or volume
percent diesel. Water-soluble fractions are created by mixing oil and water to create a nominally
saturated solution. The aqueous solution is drawn off and should be filtered to remove any suspended
oil droplets. It is then diluted in water to create the test media. Results may be expressed as mg
hydrocarbons/L or percent water-soluble fraction. In theory, one could also use toxicity data for each of
the component chemicals in diesel fuel and estimate the combined effect based on individual effects, but
that approach was judged to be impractical given uncertainties about diesel fuel composition in the
scenarios and the paucity of toxicity data.
Potentially relevant results of tests of diesel dispersions and water-soluble fractions are summarized in
Table 11-8. Results range over 4 orders of magnitude, and are highly variable even within an individual
species or test type. This range results from differences in test procedures and diesel fuel compositions.
Tests with biodiesel, synthetic diesel, sub-organismal endpoints, saltwater, and dispersants were not
included.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1124 April 2013
-------
Chapter 11
Pipeline Failures
Table 11 8. Toxicity of diesel fuel to freshwater organisms in laboratory tests.
Species
Life Stage3
Test Endpoint
Concentration
Source— Notes
Water-Soluble Fraction
Rainbow trout
Rainbow trout
Dap/i nia magna
Microcystis
aeruginosa
Selenastrum
capricornutum
Pseudokirchneriella
subcapitata
Free-swimming
embryos
2 months after
yolk resorption
1st instar
Culture
Culture
Cultures
9-day LCso
48-hour LCso
48-hour ECso
4-hour C fixation
4-hour C fixation
96-hour ICso
8 mg/L
2.43 mg/L
6.7%
100%
100%
58.7%
Schein et al. 2009-total dissolved
hydrocarbon concentration
Lockhartetal. 1987— total hydrocarbon
concentration
Giddings et al. 1980— percent water
soluble fraction
Giddings etal. 1980-sign if leant
inhibition as percent water soluble
fraction
Giddings etal. 1980— significant
inhibition as percent water soluble
fraction
Pereira etal. 2012— inhibition of growth
as percent water soluble fraction
Aqueous Dispersion
Coho salmon
Coho salmon
Pink salmon
Rainbow trout
Rainbow trout
Rainbow trout
Rainbow trout
Fathead minnow
Daphnia magna
Daphnia magna
Daphnia magna
Chironomidae
Selenastrum
capricornutum
Juvenile
Fry
Juvenile
Juvenile
Fry
Swim-up fry
Juvenile
Juvenile
Juvenile
Unspecified
Juvenile
Larvae
Culture
96-hour LCso
96-hour TLm
96-hour LCso
96-hour LCso
14-day LCso
72-hour LCso
96-hour LCso
96-hour LCso
24-hour LCso
96-hour LCso
48-hour LCso
96-hour LCso
72-hour ELso
10,299 mg/L
2,870 mg/L
74 mg/L
3,017 mg/L
44.9 mg/L
133.52 mg/L
31 (6.6-65) mg/L
57 mg/L
1.78 mg/L
20.0 mg/L
36(2-210) mg/L
346 mg/L
20(1.8-78) mg/L
Wan et al. 1990-soft water
Hebertand Kussat 1972
Wan et al. 1990-soft water
Wan et al. 1990-soft water
Mos etal. 2008
Khan etal. 2007
API 2003— mean and range of three
tests
API 2003
Khan etal. 2007
Das and Konar 1988
API 2003-mean and range of 12 tests
Das and Konar 1988
API 2003— mean and range of seven
results from three endpoints (inhibition
of cell density, biomass, or growth) and
three tests
Notes:
3 As described by the authors.
LC5o =median lethal concentration; EC5o = median effective concentration; ICso = median inhibitory concentration; TLm = equivalent to LC5o;
EUo = median effective level.
11.5.3.3 Analogous Spills: Diesel in Streams
Diesel spills into streams and wetlands are not uncommon, but their biological effects are seldom
determined and published. Relevant diesel spill case studies are summarized in Table 11-9 and
discussed in the text below. None of these studies were conducted in the Bristol Bay region, so they
provide only a general indication of the nature and duration of effects expected from an instream diesel
spill. We found no publications describing biological effects of diesel spills in relevant wetland habitats.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-25
April 2013
-------
Chapter 11
Pipeline Failures
Table 11 9. Cases of diesel spills into streams and the diesel pipeline failure scenarios.
Case
Happy Valley Creek, AK
Camas Creek, MT
Hayfork Creek, CA
Mine Run Creek, VA
Reedy River, SC
Cayuga Inlet, NY
Westlea Brook, UK
Hemlock Creek, NY
Chinkelyes Creek
Knutson Creek
Diesel Released (m3)
3.7
Unknown
15
240
3,600
26
9.8
0.5
30
8
Receiving Stream Flow (m3/s)
14
0.42
4.1
1.2
6.4
1.8
1.34
0.76
2.44
2.03
Observed Effects
Significant declines in the abundance
and species richness of invertebrates
Low invertebrate abundance and
richness
Large kill of vertebrates and
invertebrates
Reduced invertebrate abundance and
diversity
Near-complete fish kill
Fish kill and reduced abundance,
reduced invertebrate abundance and
species composition
Fish kill, invertebrates severely affected
No significant effects on invertebrates
Diesel pipeline failure scenario
Diesel pipeline failure scenario
Notes:
a Mean flow from NHDPIus v2; others as reported by the authors.
Multiple diesel spills have been associated with construction of the Trans-Alaska Pipeline, but biological
effects were studied only in for a 1972 spill from a broken underground pipeline that released 3,750 L to
Happy Valley Creek (during spring streamflows of 14 m3/s); biological effects of the spill were studied
downstream in the Sagavanirktok River (Nauman and Kernodle 1975, Alexander and VanCleve 1983).
Invertebrate abundance declined by 89% after the spill (Nauman and Kernodle 1975), and stonefly and
caddisfly nymphs were eliminated from the stream (Alexander and VanCleve 1983). Recovery was not
reported.
A pipeline spill into Camas Creek, Montana, of an oil that "most strongly resembled diesel fuel" resulted
in low abundance and low richness of the invertebrate community with few mayfly, stonefly, and
caddisfly taxa (Van Derveer et al. 1995). After remediation, including stream diversion and extensive
removal of contaminated soil below the spill, and recovery for approximately 1 year, taxa richness and
abundance at the spill site were 60 to 70% of the upstream reference site, whereas at sites farther
downstream from the remediation activities taxa richness and abundance were less than 15% and less
than 10%, respectively, of the reference site levels.
A tanker truck wreck in Trinity County, California, resulted in the flow of approximately half of the
15,000-L tank of diesel fuel into Hayfork Creek, a tributary of the Trinity River (Bury 1972). The oil was
spilled on land and reached the stream after 36 hours. An area 1 to 2.5 miles below the spill was
surveyed, because it had been previously studied. Numerous dead organisms were collected, including
4,469 vertebrates (rainbow trout and other fish, tadpoles, snakes, turtles, and a bird) and uncounted
thousands of macroinvertebrates. Recovery was not monitored.
A 1980 pipeline break released 340 m3 of Number 2 fuel oil to a small tributary of Mine Run Creek,
which ultimately flows to the Rapidan River, Virginia (Bass et al. 1987). The operator reported collecting
240 m3 of oil. Monitoring was initiated 4 months after the spill, so acute effects were not observed.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-26
April 2013
-------
Chapter 11 Pipeline Failures
Standing crop, density, and diversity of macroinvertebrates were reduced in Mine Run Creek
downstream of the tributary. Trichoptera were particularly affected. Effects were still observed at
16 months when the study ended.
In 1996, a pipeline ruptured and released 22,800 barrels (3.6 x 106 L) of diesel into the Reedy River,
South Carolina (Kubach et al. 2011). That spill resulted in a severe fish kill for 37 km downstream to the
confluence with a reservoir. Recovery of the fish community, based on non-metric multidimensional
scaling, occurred after 52 months.
In 1997, a train wreck spilled an estimated 26,500 L of diesel into Cayuga Inlet, a tributary stream of
Cayuga Lake, New York (Lytle and Peckarsky 2001). Despite containment efforts, a kill occurred, which
reduced fish abundance (including rainbow trout abundance) by 92% and invertebrate abundance by
90%. Invertebrate density recovered within 1 year, but species composition had not recovered after
15 months.
In 2005, 9,800 L of diesel spilled into Westlea Brook in Wiltshire, UK (Smith et al. 2010). Due to its urban
location, response was rapid, and approximately 7,000 L were recovered. However, the spill killed
approximately 2,000 fish and a few frogs and birds. Invertebrate surveys showed that
macroinvertebrates were severely affected and impacts were discernable for 4 km. Recovery occurred
within the 13.5-month sampling period for all but the most affected site.
A tank of home heating oil (described as similar to diesel) leaked 500 liters, and an unknown amount
entered Hemlock Creek, New York (Coghlan and Lund 2005). Three days after the spill, a survey of
benthic invertebrates (Coghlan and Lund 2005) below the spill site found no significant reduction in the
Hilsenhoff index. The authors concluded that their techniques were sufficiently sensitive and no
significant effects resulted from this small spill.
11.5.3.4 Analogous Spills: Crude Oil in Salmon Spawning Streams
The Exxon Valdez oil spill infiltrated the beaches of tidal Alaskan streams that provide spawning habitat
for pink salmon (Rice et al. 2007). Water draining over the buried oil dissolved hydrocarbons, exposing
salmon eggs and resulting in embryo histopathology and mortality for at least 2 years after the spill. The
oil spilled and the circumstances of the spill are different from the diesel pipeline failure scenarios, but
the studies described by Rice et al. (2007) demonstrate that oil buried near spawning habitats can be a
source of potentially toxic exposures for years.
11.5.4 Risk Characterization
Toxicological risk characterization is performed primarily by calculating risk quotients, from the ratios
of exposure levels to aquatic toxicological benchmarks (Box 8-3). However, it also includes
consideration of actual diesel spills, the potential for remediation and recovery, site-specific factors, and
the overall weight of evidence.
To characterize risks from a potential diesel spill, we weighed four lines of evidence based on different
exposure estimates and sources of exposure-response relationships. The first two lines of evidence
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±±27 April 2013
-------
Chapter 11 Pipeline Failures
relate modeled estimates of dissolved hydrocarbon concentrations to laboratory test results for
dissolved fractions of diesel oil and to state water quality standards. Because the diesel pipeline failure
scenarios are sufficient to saturate the two potential receiving streams, we assume dissolved
concentrations equal the solubilities of the Alaskan diesels (1.9 and 7.8 mg/L). Estimated concentrations
in the Iliamna River are a little lower (1.7 and 7.3 mg/L) due to limited concentrations in diesel of
soluble chemicals. These exposure levels are similar to the two LCso toxicity values for rainbow trout
(2.43 and 8 mg/L) (Table 11-8) and far higher than the state standard (0.015 mg/L). Based on these
estimates of soluble hydrocarbon concentrations, invertebrate kills would be highly likely and some
salmonid mortality would be expected under the failure scenario at either location.
The next line of evidence relates exposure (expressed as the amount of oil added to the stream) to
laboratory test results for diesel dispersed in water. This line of evidence is based on the assumption
that diesel added to a flowing stream is equivalent to diesel added to water and stirred. Exposure levels
within the receiving water would be 16,000 mg/L for Chinkelyes Creek, 1,600 mg/L for Iliamna River,
and 8, 500 mg/L for Knutson Creek (Table 11-7). The laboratory LCso tests for diesel dispersions are
shown in Table 11-8, and strongly suggest that an oil spill would result in acute lethality offish and
invertebrates, even if turbulent mixing in a stream is not as efficient as stirring. In addition, tests of the
alga Selenastrum capricornutum found that multiple growth and production endpoints were reduced by
50% at 20 mg/L (API 2003) also well below the estimated exposure.
The published history of freshwater diesel spills provides the final line of evidence. Diesel spill volumes
at the two locations considered in these diesel pipeline failure scenarios—30 m3 at Chinkelyes Creek
and 8 m3 at Knutson Creek—fall within the range of the cases described in Table 11-9 that caused effects
on the biotic communities of rivers and streams. In addition, the sizes of the receiving streams, under
these failure scenarios and those in the case studies are similar. If we divide the amount of diesel spilled
by the streamflow as a crude index of exposure, the two scenarios (12.3 and 3.9) fall in the middle of the
range of cases (0.26 to 562).
Only the case of a very small spill (less than 500 Linto Hemlock Creek, NY) caused no significant
biological effects. Other diesel spills caused fish and invertebrate kills and reduced invertebrate
abundance and diversity. Invertebrate community effects persisted for several months to more than 3
years. Exposures and effects may be more persistent in Alaska's cold climate, but the only Alaskan study
did not monitor recovery. Based on past diesel spills in streams, the diesel spills evaluated in this
assessment, and any other spill that released more than a trivial amount of diesel to a stream, would be
expected to cause an immediate loss offish and invertebrates, and the community would be likely to
recover in 1 to 3 years.
11.5.4.1 Weighing and Summarizing the Lines of Evidence
The diesel pipeline failure probability used in this assessment is based on one line of evidence, the
record of actual oil pipelines. However, the predicted effects of a diesel spill are based on four lines of
evidence. All lines of evidence lead to the conclusion that a diesel spill into a stream would result in an
invertebrate and fish kill and reductions in abundance and diversity (Table 11-10). In the diesel pipeline
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±±28 April 2013
-------
Chapter 11 Pipeline Failures
failure scenarios evaluated here, the lengths of affected stream would be 21 km (Chinkelyes Creek and
the Iliamna River) or 7.3 km (Knutson Creek). Because these distances are short relative to oil
degradation rates, effects would be likely to extend to Iliamna Lake. Effects in the lake are not estimated
here, but are unlikely to extend far beyond the area of input due to dilution. In Knutson Creek, however,
flow to the Knutson Bay could result in mortality of congregated spawning salmon, their eggs, and other
fish attracted by salmon eggs as a food source (Appendices A and B). Based on the monitoring of diesel
spills in streams, effects on stream communities would be likely to persist for 1 to a several years.
Although all lines of evidence have associated uncertainties and weaknesses (Section 11.5.5), they all
support these general conclusions.
The weighing of these lines of evidence is summarized in Table 11-10. For each route of exposure,
sources of the exposure estimate and the exposure-response relationship are indicated. All evidence is
qualitatively weighed based on three attributes: its logical implication, its strength, and its quality
(Suter and Cormier 2011). For logical implication, possible scores are indicated as (+) for results
supportive of effects on the endpoint populations, (-) for results contrary to effects on assessment
endpoints, and (0) for neutral or ambiguous results. In this case, the logical implication is that the diesel
pipeline failure scenario evaluated here would have adverse effects. The strength of the evidence is
based primarily on the magnitudes of the hazard quotients (exposure concentrations divided by effects
concentrations): a low quotient is indicated as (0), a moderate quotient as (+), and a high quotient as
(++). Quality is a more complex concept. It includes conventional data quality issues, but in this case the
primary determinate is the relevance of the evidence to the mine scenario. Separate quality scores are
provided for the exposure estimate and for the exposure-response relationship. The scores are intended
to remind the reader what evidence is available and show the pattern of strength and quality of the
several lines of evidence that might not be apparent from reading the text, and to transparently present
our weighing process and results.
Overall, the available lines of evidence for effects of a diesel spill are positive for the occurrence of acute
toxic effects (Table 11-10). The quality of the exposure-response information is good (+) for all routes of
exposure based on reported observations in case studies, because the information is realistic; the
quality of information is considered good (+) for exposure via dissolved and hydrocarbons, based on
laboratory acute tests,, because of the information reflects multiple tests. The quality of the exposure
information for the dissolved and dispersed hydrocarbons is considered ambiguous (0), because of the
uncertain relationship between the laboratory preparations and modeled stream exposures. The quality
of the exposure-response information is considered very good (++), because it is based on the Alaska
water quality standard, an official standard. The analogous spills, as a whole, are considered very strong
(++) evidence that a diesel spill would cause toxic effects in streams.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1129 April 2013
-------
Chapter 11
Pipeline Failures
Table 1110. Summary of evidence concerning risks to fish from a diesel spill. The risk
characterization is based on weighing four lines of evidence for different routes of exposure. All
evidence is qualitatively weighed (using +, 0, ) on three attributes: logical implication, strength, and
quality. Here, all lines of evidence have the same logical implication that is, all suggest a diesel spill
would have adverse effects. Strength refers to how strongly the line of evidence indicates effects, and
quality refers to the quality of the evidence sources (i.e., data quality and relevance to the diesel
pipeline failure scenario).
Route of Exposure
Source of Evidence (Exposure/E-R)
Dissolved hydrocarbons
Model/laboratory acute tests
Dissolved hydrocarbons
Model/laboratory-based standard
Dispersed hydrocarbons
Diesel:water ratio/laboratory acute
tests
All routes in actual spills
Amount spilled/observed effects
Summary Weight of Evidence
Logical
Implication
+
+
+
+
+
Strength
+
++
++
++
++
Quality
Exposure
0
0
0
+
0
E-R
+
++
+
+
+
Result
Modeled dissolved diesel
concentrations are clearly lethal to
invertebrates and approximately
lethal to trout
Modeled dissolved diesel
concentrations greatly exceed State
standard
Diesel oil/water ratios in the spills
and in tests suggest lethality to
invertebrates and trout
Diesel spills in other streams cause
acute biological effects
The effects by four lines of evidence
are consistent and the observed
effects are strong. The greatest
uncertainty is the relation of
laboratory to field exposures.
Notes:
E-R = exposure-response relationship.
The specific effects of a diesel spill on salmonid populations would depend on the individual receiving
waters. Streams along the transportation corridor that could receive a spill are described in Section 10.1.
Chinkelyes Creek receives on average more than 9,000 spawning sockeye salmon and flows to the
Iliamna River, which receives on average more than 100,000 sockeye spawners (Table 10-2). Knutson
Creek receives 1,500 sockeye spawners and flows to Knutson Bay, which receives an average of
73,000 beach spawning sockeye (Table 10-2). Not all of those salmon spawn below the stream crossing,
but these values indicate that a non-trivial number of spawners and their potential production are at
risk. In this scenario, a spill would likely disrupt spawning, if it occurred during the spawning season,
would potentially kill adults and likely kill fry, and would certainly kill invertebrates on which salmon
fry and all stages of other salmonids depend.
11.5.4.2 Duration of Risks
Diesel and natural gas pipelines would be retained after mine closure as long as fuel was needed at the
mine site (e.g., for monitoring, water treatment, and site maintenance). Therefore, the diesel pipeline
risks would continue for indefinitely.
11.5.4.3 Remediation
Remediation of freshwater oil spills is discussed in detail in a review by NOAA and API (1994). For diesel
spills in small rivers and streams, remediation via booms, skimming, vacuum, berms and sorbents
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
11-30
April 2013
-------
Chapter 11 Pipeline Failures
results in the least environmental impact. Diesel is difficult to remediate by conventional techniques
because its components seep into soil, dissolve in water, or evaporate relatively quickly, making it is less
containable by berms, sorbents, or booms than typical crude oil. Also, booms, although useful, are
imperfect tools for containing floating oil. After the diesel spill in Cayuga Inlet, booms were deployed,
but within 24 hours a slick was reported on Cayuga Lake, 16 km downstream (Lytle and Peckarsky
2001). Even when recovery of diesel fuel was rapid and approximately 70% effective, as in the Westlea
Brook spill (Table 11-9), the rapidly dissolved component was sufficient to cause severe acute effects
(Smith etal. 2010).
Remediation of oil spills in freshwater wetlands has been relatively little studied. For diesel, the NOAA
and API (1994) review recommends natural recovery, sorbents, flooding, and low-pressure cold-water
flushing as least adverse. Wetlands also have been remediated by burning, which can remove floating oil
and destroy oiled vegetation that is likely to die from the effects of the oil. Burning can cause severe but
localized and short-term air pollution and, if improperly controlled, can result in fires that spread
beyond the oiled area. However, burning does not destroy the dissolved fraction, which is primarily
responsible for aquatic toxicity and which would move to streams or the lake.
Cold winter weather complicates remediation of diesel spills (NOAA and API 1994). Spills into water
below the oil's pour point can result in the formation of viscous tar-like particles that are difficult to
recover. Ideally, a spill onto ice could congeal on the surface where it might be relatively recovered if
action is prompt; however, diesel oil can penetrate ice, and solar absorption by the oil can result in
freeze-thaw cycles that create a complex material. Spills that flow under ice deposit on the undersurface.
Standard procedures for oil remediation do not address those conditions.
11.5.5 Uncertainties
Based on weighing multiple lines of evidence, it is certain that a diesel pipeline spill into a stream would
cause acute toxic effects. However, the following uncertainties apply to individual pieces of evidence.
• The composition of diesel oil is highly variable. As a result, the fate and toxicity of diesel spills are
inherently uncertain unless the specific source is known and analyzed; the source does not change
over time; and physical, chemical, and biological tests are performed with that specific oil. This
uncertainty cannot be resolved without case-specific studies of a sort that are not normally
performed. This and other uncertainties concerning test results could cause errors in the risks
estimated from laboratory toxicology of at least a factor of 10.
• Measurement of petroleum hydrocarbons in water is performed using a variety of methods. Because
the results of hydrocarbon analysis are method-specific, significant uncertainty can be introduced
when these results are compared to benchmarks generated using different analytical methods. This
contributes to the overall uncertainty of toxicity test results.
• If a spill occurs at a stream, losses of invertebrates and fish are likely, but the magnitude and nature
of these losses would be highly uncertain—some mortality would occur for some species, but the
species and number of organisms affected cannot be specified. This uncertainty would take a major
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 11~31 April 2013
-------
Chapter 11 Pipeline Failures
case-specific research program to resolve. The inability to exactly define the expected ecological
effects occurs in all risk assessments, but is worse for the diesel spill than for other contaminants
such as copper.
• The ability of the laboratory toxicity tests to predict responses to diesel in the field is highly
uncertain due to the variety of preparation methods, the simplicity of laboratory exposures relative
to the complexity of oil spills in streams, and the lack of field validation studies.
• Variation in sensitivity to diesel among species appears to be high relative to other aquatic
pollutants. Remarkably, even in the same test series, different salmon species range in sensitivity
over two orders of magnitude (Table 11-8).
• Spills into wetlands are likely to have severe and persistent effects due to low rates of flow, but no
relevant studies of diesel spills in freshwater wetlands are available to confirm even that very
general hypothesis.
• The applicability of previous diesel spills considered in Table 11-9 to streams in the Bristol Bay
region is uncertain, given that all of the spills occurred elsewhere. However, the effects observed in
the one Alaskan case are not dissimilar from those in temperate regions. The most likely differences
are slower loss of oil and longer recovery times. Therefore, the effects are likely to be more severe in
Alaska than in the temperate cases.
• Although diesel is a highly variable mixture, the principle uncertainty is the number and location of
spills into aquatic ecosystems given the probability of a pipeline failure. We can say with some
certainty that a spill of a non-trivial volume of diesel into a stream would have adverse ecological
effects. We can also say that a spill is likely, based on the record of oil pipelines in general and large
recent spills from oil pipelines into the Yellowstone and Kalamazoo Rivers.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1132 April 2013
-------
Large-scale mining, as described in the mine scenarios (Table 6-1), could have both direct and indirect
effects on wildlife and Alaska Native cultures (Figures 12-1 and 12-2). In this chapter, we primarily
consider indirect effects, focusing on how wildlife and Alaska Native cultures may be affected by any
mining-associated changes in salmon resources. Direct effects on these endpoints—defined here as
effects that are independent of impacts on fish populations—could be significant, and would need to be
fully evaluated as part of a comprehensive environmental impact statement for any proposed future
development. However, these direct effects are generally considered outside the scope of the current
assessment (Chapter 2) and are only mentioned briefly here. Potential cumulative effects that multiple
mines in the region may have on wildlife and Alaska Native cultures are discussed in Chapter 13.
12.1 Effects on Wildlife
As discussed in Chapters 7 through 10, a large-scale mine and its associated transportation corridor
would affect the abundance, productivity, and diversity of Pacific salmon. These changes in salmon
resources could stem from direct habitat losses and downstream flow alterations resulting from the
mine footprint, or from changes in the physical and chemical habitat characteristics resulting from mine
operations and potential accidents or failures. Wildlife species in the Nushagak and Kvichak River
watersheds that depend on salmon could be affected by decreases in salmon abundance. Interactions
between salmon and other fish and wildlife, and the potential for disruption of these interactions, are
complex. In this section, we qualitatively consider how a decrease in salmon abundance may affect
wildlife—that is, salmon-mediated effects on wildlife—via the direct loss of salmon as a food source and
the indirect loss of marine-derived nutrients (MDN) as a source of productivity.
Lower salmon production would likely reduce the abundance and production of wildlife in the mine
area and presumably in the range areas of the affected species, but the magnitude of those effects cannot
be quantified. The Bristol Bay region is home to a complex food web that includes salmon and salmon
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
12-1
April 2013
-------
Chapter 12 Fish-Mediated Effects
predators and scavengers (Figure 12-1) (Brna and Verbrugge 2013). Annual salmon runs provide food
for brown bear, wolf, bald eagle, other land birds, and water birds, and it is likely that these species
would be directly affected by a reduction in salmon abundance. Waterfowl prey on salmon eggs, parr,
and smolts, and scavenge on carcasses. Salmon carcasses are an important food source for bald eagles,
water birds, other land birds, other freshwater fish, and other terrestrial mammals. Aquatic invertebrate
larvae also benefit from carcasses and are an important food source for water birds and land birds.
Salmon predators and scavengers then deposit MDN on the landscape, as either carcasses or excreta.
These nutrients likely increase the plant production that supports caribou, birds, and other terrestrial
wildlife. Caribou are, in turn, prey species for wolves and brown bears. The link between increased
vegetation and MDN distributed by brown bears has been documented (Hilderbrand et al. 1999, Helfield
and Naiman 2006), but additional research is needed to confirm and quantify the links between moose,
caribou, and MDN.
Factors such as the magnitude, seasonality, duration, and location of the salmon loss would influence the
specific wildlife species affected and the magnitude of effects. Generally, a loss of salmon as food
resources in any area of the mine scenario watersheds would be expected to create displacement or loss
of wildlife species dependent on those food resources. If the loss were of a sufficient magnitude and
duration, there may be additional indirect effects, such as loss of vegetation from lack of MDN and
subsequent loss of food resources for species such as moose and caribou. Should riparian vegetation be
reduced by long-term loss of MDN, there would be decreased food resources for moose, particularly in
the Nushagak and Mulchatna River systems, which have large riparian zones (Brna and Verbrugge
2013).
Seasonality of salmon resources is also important for wildlife species. Brown bears, wolves, bald eagles,
and other species depend on salmon for a large fraction of their summer diet. Mine failures that reduced
or eliminated a salmon substock would be expected to reduce or displace those wildlife species that
depend on those particular salmon nutrients during important periods such as breeding, nesting, and
pre-winter feeding.
Alaska Natives have expressed concerns that wildlife may be affected by consuming contaminated fish.
Two potential contaminants of concern are copper and selenium. The primary aquatic contaminant of
concern from a porphyry copper mine is copper, which can cause both acute and chronic toxicity to
salmon and other fish (Chapter 8). However, copper is relatively weakly accumulated by fish in both
aqueous and dietary exposures and does not bioaccumulate. In fact, in the Clark Fork River, copper
concentrations were lower in fish than in invertebrates, and lower in invertebrates than in periphyton
(ARCO 1998).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ o April 2013
-------
Chapter 12
Fish-Mediated Effects
Figure 12 1. Conceptual model illustrating potential effects on wildlife resulting from ef
fects on fish.
/ \ / \
I transportation | 1 , ] [ power generation & , ', ancillary ] [ nirt !
1 corridor J \ / ^ transmission facilities ; ^ facilities ,' '^ ' f>
i
V
'f road-related 1s
mortality
LEGEND
[ sou
proxi
stre:
addition;
1 causal |:
(^^respi
clashed arro'/i
indicate topics
of asse:
i
i
\/ \y V V V V
migration "] conditioning 'f terrestrial 1" noise 4 air ' 'f environmental impacts
barriers on human food habitat loss S. light quality! on freshwater habitats
'
i i ill
i
i
V \/
;'' -fdirecteffects **\ X " 4- salmon ~~^
\ on 'Afildlifp -' V^^ (abundance, productivity or diversity) ^^
ce )
nate
;sor
1! step in
athwav I
jnsey'
s& outlines
outside scope
;sment
C 4 adults 3 C. 4- carcasses 3 ( 4 eggs ) C 4 juveniles )
(^
V V
^ ~ ;Jl.r "I
V V
I 4 terrestrial distribution of 4' aquatic distribution of I
| marine-derived nutrients | marine-derived nutrients
V - • V \/ \/
r ^ r i
4/ terrestrial I I 4 riparian I
vegetation vegetation -I aquatic food resources
V V
C 4- caribou ) f 4-rnoose ,)
V V V V
(\I brown bearsj) v/ C\|, bald eagles ij) . . j< 4, other ^\ /"^ 4- other ^^
x^*_^-_^^x \/ x^..— — s^^x >s/ ( landlords J ^^ Ifreshwaterfish ) \(
( 4- wolves J C 4- water birds } ' s*^~ ~~T^i " ' '/^~~~' i • ~^~^^
\^^"^f x^^-'^^^^x x- 4, marine fish \ f 4- other ^\
V. S.mammals ) \ terrestrial mammals )
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
12-3
April 2013
-------
Chapter 12
Fish-Mediated Effects
Figure 12 2. Conceptual model illustrating potential effects on Alaska Native cultures resulting from effects on fish.
4 salmon
(abundance, productivity or diversity)
4, other fish
i ab u n d an c e. p r o d u cti :• ity o r cl i v e r sity i
marine-derived
nutrients
-J, ecosystem
productivity
additional step in
causal pathway1
.1 food resources
4 A i Id life
(ab u n d an c e, p r o d u cti v ity o r cl i v e r sitv)
4 commercial
fishing
4 subsistence
equipment
village subsistence
(quality or quantity)
Q4- spirituality^) Q 1 mining jobs J)
social relations
4 family activities &
trad iti o n al
-------
Chapter 12 Fish-Mediated Effects
Data for copper toxicity to wildlife are not available, because direct toxicity has not been a problem—the
indirect effects of reduced aquatic prey are likely to be greater than direct toxic effects. However, the
dietary maximum for pigs and poultry of 200 mg/kg dry weight can be used as a surrogate benchmark
(Eisler 2000). If we use the highest reported fish bioconcentration factor (290 for fathead minnows,
from an unpublished manuscript cited in USEPA 1985) as a conservative value, we obtain a safe water
concentration of 690 ug/L. This safe level for wildlife is much higher than both the toxic levels for
aquatic biota (Section 8.2.2.1) and the estimated instream concentrations (Table 8-20). It is a little
higher than the estimated concentration in the concentrate transport water and return water
(655 ug/L), but is not a concern because of dilution and the short duration of exposures to spills. The
copper concentration in the product concentrate leachate is 12 times that value. However, a product
spill would be localized at the mine site or along the transportation corridor. Most important, if copper
concentrations were so high that the leachate was not diluted by a factor of 12, there would not be a
sufficient aquatic community remaining after a spill to attract species such as mink, river otters, or
belted kingfishers, which may forage in a particular stream or river for fish such as trout, char, or
salmon fry. Hence, copper toxicity to fish-eating wildlife is improbable.
Selenium is a well-characterized avian toxicant that has been a concern for waste rock leachate at other
mines (USEPA 2011). It does biomagnify and the primary route of exposure for fish is diet (Chapman et
al. 2010). However, selenium bioaccumulation depends on biogeochemical conditions that occur in
slowly flowing (lentic) ecosystems such as ponds and wetlands, and is much less prevalent in the
streams that are the likely receptors for effluents with elevated selenium levels. In addition, the
selenium concentrations in wastes identified in the mine scenarios would be relatively low. Mean
selenium (the appropriate measure for a biomagnifying chemical) in waste rock leachates is below the
water quality criterion and leachates from tailings and product concentrate are only 1.5 times the
criterion concentration (Section 8.2.2.2). Hence, minimal dilution would bring these concentrations
down to safe levels. Although fish and birds are sensitive to selenium, mammals are not. The body
burdens of adult salmon are almost entirely due to marine exposures; because salmonid eggs take up
contaminants relatively slowly, body burdens of eggs and larvae (early fry) would also be expected to
reflect marine sources. Local sources of selenium are therefore not relevant, and bioaccumulation of
selenium would only be a concern for wildlife consuming resident fish or older salmon fry. Fish-eating
birds feeding on resident fish also may forage in more than one stream or other water body, potentially
providing further dilution. For these reasons, aqueous selenium is unlikely to pose a risk to wildlife via
fish consumption.
These salmon-mediated effects on wildlife represent only one component of large-scale mining impacts
on wildlife. Wildlife species would experience direct impacts as well, including loss of terrestrial and
aquatic habitat, reduced habitat effectiveness (e.g., in otherwise suitable habitats adjacent to mine area),
habitat fragmentation, increased stress and avoidance due to noise pollution, and conditioning on
human food (Figure 12-1). Although a comprehensive evaluation of these direct effects is beyond the
scope of this assessment, it is clear that large-scale mining activities would likely result in numerous and
complex direct effects on wildlife.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 77T April 2013
-------
Chapter 12 Fish-Mediated Effects
12.2 Effects on Alaska Natives
As discussed in Chapters 7 through 11, routine development and operation of a large-scale mine, as well
as potential mine accidents or failures, would likely affect salmon resources in the Nushagak and
Kvichak River watersheds. The importance of salmon to Alaska Native cultures is well documented
(Appendix D). Because these cultures are so intimately related to the local landscape and the resources
it provides, any change to salmon or other subsistence resources would likely result in changes to the
cultures. The magnitude of these changes could be assumed to be dependent on the magnitude and
duration of both the loss of subsistence resources and the disruption to the landscape itself. Changes in
salmon resources may affect indigenous health, welfare, and cultural stability in several ways
(Appendix D).
• Because the traditional diet is heavily dependent on wild foods, particularly salmon, diets would
move from highly nutritious wild foods to increased reliance on purchased processed foods.
• Social networks are highly dependent on procuring and sharing salmon and wild food resources, so
the current social support system would be significantly degraded.
• The transmission of cultural values, language learning, and family cohesion would be affected
because meaningful family-based work takes place in fish camps or similar settings for traditional
ways of life.
• Values and belief systems are represented by interaction with the natural world through salmon
practices, clean water practices, and symbolic rituals; thus, core beliefs would be challenged by a
loss of salmon resources, potentially resulting in a breakdown of cultural values, mental health
degradation, and behavioral disorders.
• A shift from part-time to full-time wage employment in mining or mine-associated jobs would affect
subsistence-gathering capabilities by reducing the time available to harvest and process subsistence
resources.
• The region exhibits a high degree of cultural uniformity tied to shared traditional and customary
practices, so significant change could provoke increased tension and discord both between villages
and among villagers.
Human health and cultural effects related to potential decreases in salmon resources would depend on
the magnitude of these reductions. A small reduction in salmon quality or quantity may not have
significant impacts on subsistence food resources, human health, or cultural and social organization, but
a significant reduction in salmon quality or quantity would certainly have significant negative impacts
on these salmon-based cultures.
Salmon-mediated effects from potential accidents and failures associated with large-scale mining would
likely have much greater effects on human welfare and Alaska Native cultures than the effects from the
mine scenario footprints. It should be assumed that any negative impact on salmon quantity or quality
resulting from mine failures or accidents would affect human health and welfare, both directly from loss
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T7April 2013
-------
Chapter 12 Fish-Mediated Effects
of or change in food resources, and indirectly from cultural disruption. Because all aspects of these
cultures are closely tied to salmon and other fish, cultural vulnerability to long-term environmental
disruption is very high (Appendix D). A major failure or accident that resulted in long-term disruption of
salmon habitat and ongoing toxicity to salmon or their food would significantly affect both subsistence
resources and cultural identity. Potential causes of salmon-mediated effects on Alaska Native cultures
would differ across the two watersheds. For example, villages near the transportation corridor could be
negatively affected by pipeline spills or road and culvert failures (Chapter 10). Villages downstream of
the mine would be more affected by any water collection, treatment, and discharge failures (Chapter 8),
and impacts from these failures would likely be much greater than impacts from routine operations.
Although not a focus of this assessment, large-scale mining could also have multiple direct effects on
Alaska Native cultures. Direct effects could result from stressors that include noise, air emissions,
changes to water supply and quality, an influx of new residents, and ancillary development. Mine
construction and operation also would have direct (positive and negative) economic and social effects
on Alaska Native cultures. An influx of new residents in response to mine development could decrease
the local population percentage of Alaska Natives and have a corresponding effect on local culture. A
comprehensive evaluation of wildlife-mediated and direct effects is beyond the scope of the current
assessment, but it is clear that these effects could significantly contribute to changes in traditional
Alaska Native cultures.
This assessment focuses on potential effects on Alaska Native cultures, but other groups would also be
particularly vulnerable to mining-associated impacts on salmon. Many of the non-Alaska Natives that
reside in the area practice a subsistence way of life and have strong long-term cultural ties to the
landscape that go back generations (Box 12-1). In addition, many seasonal commercial anglers and
cannery workers depend on these resources and have strong, multi-generational cultural connections to
the region.
At this time, it is difficult to determine what, if any, effects routine operations at the Pebble deposit
would have on drinking water sources in the Nushagak and Kvichak River watersheds. Private wells are
a primary drinking water source for many residents of the Nushagak and Kvichak River watersheds.
Communities also rely on groundwater for their public water supply. The extent to which surface water
influences the quality or quantity of the groundwater source for these wells is unknown. There are also
communities in the area that rely on surface water sources, which may be more susceptible to
contamination.
In this section, we discuss the range of potential salmon-mediated effects on Alaska Native cultures from
large-scale mining in the Nushagak and Kvichak River watersheds. We also reference key impacts that
other mining, oil, and gas development activities in Alaska—including northwest Alaska's Red Dog Mine
and oil and gas development on Alaska's North Slope—have had on Alaska Native cultures. Although not
directly applicable to large-scale mining, information about oil and gas extraction activities can provide
insight into potential effects of large-scale mining in the Nushagak and Kvichak River watersheds.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7TZApril 2013
-------
Chapter 12 Fish-Mediated Effects
12.2.1 Subsistence Use
As discussed in Chapter 5, subsistence foods make up a substantial proportion of the human diet in the
Nushagak and Kvichak River watersheds and likely contribute a disproportionately high amount of
protein and certain nutrients. The percentage of salmon harvest in relation to all subsistence resources
ranges from 29 to 82% in the villages (Appendix D).
There would be some effects on subsistence resources from the mine scenario footprints. Although
there are no subsistence salmon fisheries documented directly in any of the mine scenario footprints,
other fish are harvested in these locations, and the areas are identified as being important for the health
and abundance of subsistence resources (PLP 2011: Chapter 23). Negative impacts on downstream
fisheries from headwater disturbance (Section 7.2) could affect subsistence salmon resources beyond
the mine scenario footprints. Those residents using the mine area and immediate areas downstream of
the mine pit and tailings storage facilities (TSFs) for subsistence harvests would be most affected
(Figure 5-2). Access to subsistence resources is also important. A reduction in downstream seasonal
water levels caused by mine-related withdrawals during and after mine operation could pose obstacles
for subsistence users who are dependent on water for transportation to fishing, hunting, or gathering
areas.
There could also be effects from the footprint of the transportation corridor. A review of ADF&G data
(Appendix D: Table 13) indicates that some residents use the area along the transportation corridor for
subsistence salmon harvest. Of the villages in these watersheds, reliance on salmon is highest in Pedro
Bay, where salmon provides 82% of per-capita subsistence harvest. The estimated per-capita
subsistence harvest for Pedro Bay, the village closest to the transportation corridor, was 306 pounds in
2004. Thus, this village is particularly vulnerable to losses of salmon resources.
The effects of the transportation corridor on subsistence resources would be complex and
unpredictable. Based on the analysis in Chapter 10, we anticipate that routine transportation operations
would have some negative effects on salmon habitat in streams along and downstream from the
transportation corridor. Some subsistence users in these areas could be displaced. The corridor also
could increase accessibility of the area, which could increase subsistence use of these streams but also
create greater competition for resources.
The initial effect of an accident or failure on Alaska Native cultures would be the loss or decrease of
subsistence salmon resources downstream. It is not possible to quantify the magnitude of subsistence
resources that would be lost, nor is it possible to evaluate the geographic extent of disruption to
subsistence resources. However, this assessment provides examples of the potential magnitude of
salmon impacts from failures. One such example is the potential effect of a TSF failure on Chinook
salmon in the Nushagak River. As described in Chapter 9, a failure of TSF 1 could significantly affect
Koktuli River Chinook runs—up to 28% of the larger Nushagak River Chinook runs. Stuyahok River and
Mulchatna River Chinook runs, which constitute up to 17 and 10% of the Nushagak River Chinook runs,
respectively, could also be affected. The Alaska Native villages on the Nushagak River (Koliganek, New
Stuyahok, Ekwok, and Dillingham) are culturally and nutritionally dependent on Chinook salmon. Thus,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T7April 2013
-------
Chapter 12 Fish-Mediated Effects
a water collection, treatment, or discharge failure would likely have negative, and potentially very
significant, effects on the ability of subsistence users to harvest salmon downstream of the mine area.
It is not possible to predict the magnitude of effects from the loss of salmon as a subsistence food, nor is
it possible to predict what level of subsistence resource loss would be necessary to overcome the
adaptive capacity of these cultures. On a physical level, the loss of salmon as a highly nutritious wild
food, and the substitution of purchased foods, would have a negative effect on individual and public
health (Appendix D). Salmon is especially valued around the world for nutrition and disease prevention.
Dietary transition away from subsistence foods in rural Alaska carries a high risk of excess consumption
of processed simple carbohydrates and saturated fats. This has occurred in urban communities that
have low availability and high cost of fresh produce, fruits, and whole grains (Kuhnlein et al. 2001,
Bersamin et al. 2006). Also, alternative food sources may not be economically viable and are certainly
not as healthy. Compounding the shift to a less healthy diet, the physical benefits of engaging in a
subsistence lifestyle would be reduced (Appendix D).
In addition to the salmon-mediated effects of large-scale mining considered here, there could be effects
from the loss of non-salmon subsistence resources, such as land mammals, birds, and other fish,
Subsistence use of the mine area is high and centers on hunting caribou and moose and trapping small
mammals (Braund and Associates 2011 in PLP 2011). Because no subsistence salmon fisheries are
documented in the mine scenario footprints, direct loss of non-salmon subsistence food resources likely
would represent a greater direct effect than loss of salmon harvest areas in the mine scenario footprints.
Tribal Elders have expressed concerns about ongoing mine exploration activities directly affecting
wildlife resources, especially the caribou herd range (Appendix D).
Experience with existing development in Alaska supports the contention that development of a large-
scale mine operation would directly affect wildlife subsistence resources within and around the mine
scenario footprint during routine operations and in perpetuity, both from loss of habitat and
disturbance from routine operations. For example, the supplemental environmental impact statement
for the Red Dog Mine (USEPA 2009) documented multiple subsistence impacts, including reduced
harvest of beluga by Kivalina harvesters, likely related to port activities. Related to transportation
corridors, traffic along the Delong Mountain Regional Transportation System road was found to cause
"limited, localized" effects on caribou movement and distribution, and nine caribou fatalities occurred
because of traffic collisions. Kivalina harvesters and harvest data also indicated that the traffic along the
road has likely resulted in fewer caribou harvested by Kivalina harvesters than would otherwise be the
case.
A study of the cumulative environmental effects of oil and gas activities on Alaska's North Slope reports
that subsistence hunting areas have been reduced, the behavior and migratory patterns of key
subsistence species have changed, and that there is increased incidence of cancer and diabetes and
disruption of traditional social systems (NRC 2003). Alaska Native residents report subtle changes in
species harvested by subsistence hunters, including changes in color, texture, and taste of the flesh and
skin of several subsistence species. Transportation corridors associated with resource extraction
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ 9 April 2013
-------
Chapter 12 Fish-Mediated Effects
activities can also increase competition for local subsistence resources. For example, hunting by non-
local residents along the Dalton highway has been reported to have increased after the development
(and later public opening) of the road (NRC 2003).
The experiences of subsistence users near Red Dog Mine and Alaska's North Slope indicate that localized
changes in resource movement can affect that resource's availability and predictability to subsistence
users, even when the overall pattern or abundance of the resource may not be affected by development
activities. From a biological standpoint, changes in caribou related to the Red Dog Mine may be viewed
as minimal. However, because residents rely on only a portion of the expansive range of the Western
Arctic caribou herd to harvest caribou, small and localized changes in caribou availability can have large
effects on subsistence uses. Subsistence users have observed changed or diverted migration routes,
reduced harvest of caribou, decreased size of caribou individuals and groups, and increased disease and
infection since mine operations began, and cite both mine-related and other causes (USEPA 2009).
The Exxon Valdez oil spill also resulted in reduced subsistence activities (Palinkas et al. 1993). These
reductions resulted from the closure of many areas to subsistence activities, local concerns over
subsistence food safety, voluntary abstinence from consumption after the spill, and reduced time for
subsistence activities by Alaska Natives who participated in cleanup efforts.
12.2.2 Perception of Food Security
Even a negligible reduction in salmon quantity or quality related to mining could decrease use of salmon
resources, based on the perception of subtle changes in the salmon resource. Interviews with tribal
Elders and culture bearers indicate that perceptions of subtle changes to salmon quality are important
to subsistence users, even if there are no measureable changes in the quality and quantity of salmon
(Appendix D). Aside from actual exposure to environmental contamination, the perception of exposure
to contamination is also linked to known health consequences, including stress and anxiety about the
safety of subsistence foods and avoidance of subsistence food sources (CEAA 2010, Joyce 2008, Loring et
al. 2010), with potential changes in nutrition-related diseases as a result. These health results arise
regardless of whether there is contamination at a level that could induce toxicological effects in humans;
the effects are linked to the perception of contamination (NMFS 2011).
Literature on impacts from oil and gas development on Alaska's North Slope and ongoing operations at
Red Dog Mine demonstrates that even perceived contamination could have a real effect on subsistence
harvesters. In a recent survey, 44% of Inupiat village residents reported concern that fish and animals
maybe unsafe to eat (Poppel etal. 2007, NMFS 2011). Residents of Kivalina and Noatak, the
communities closest to Red Dog Mine, also have expressed concerns about food safety and potential
contamination of subsistence resources and corresponding changes in subsistence foraging (USEPA
2009). Kivalina residents are concerned about potential contamination of the Wulik River, which is used
both for subsistence and as the drinking water source for the village. These concerns persist even
though studies by the Alaska Department of Health and Social Services found that heavy metal
concentrations in drinking water were low and did not pose a risk (USEPA 2009).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 12~10 April 2013
-------
Chapter 12 Fish-Mediated Effects
12.2.3 Economic Impacts
On an economic level, should there be a decreased availability of salmon, the necessity of purchasing
expensive foods from outside the region in conjunction with limited opportunities to obtain paid
employment in the region, could make it extremely difficult for families to survive. For those able to
benefit economically from mining and secondary development, increased full-time employment could
decrease subsistence activities and social relationships derived from these activities. Some residents
have expressed a desire for jobs and development related to large-scale mining and a market economy,
whereas other residents have expressed concerns that this type of economic shift would be detrimental
to their culture (Box 12-1, Appendix D).
BOX 12 1. TESTIMONY ON POTENTIAL EFFECTS OF MINING ON ALASKA NATIVE CULTURES
The U.S. Environmental Protection Agency (USEPA) held a series of public meetings to collect input on the
first draft of this assessment. Many Alaska Natives, including tribal Elders and other tribal leaders, provided
testimony on concerns about potential effects of large-scale mining in the Bristol Bay watershed. The
following a re selected quotes representative of this testimony. To view the full public meeting transcripts,
visit www.epa.gov/bristolbav.
• "Salmon has been part of our Native spiritual food; and without the food and waters we will die slowly,
we'll be here existing, but our spirit will be gone."
• "I urge you to pay especially close attention to the voices of our Elders across Bristol Bay. They have
instilled in them the deepest of our roots, and our God given way of life; our culture that has been slowly
fading away. It is the adaptation to modern civilization that we have embraced so far that is causing our
cultures to become lost."
• "We support the science of this document as it in turn supports what the Elders of this area and their
traditional knowledge have said all along. We are preparing our boats, we are mending our nets, we are
cleaning our smokehouses and we are sharpening our knives. However, by testifying at these meetings
and missing out on one day or maybe several days of preparation for those who attended multiple
meetings here in the region, we hope that this will prevent, with the help of the EPA, missing out on a
lifetime of salmon and missing out on a way of life that we have treasured for thousands of years."
• "Bristol Bay is much different. Everyone who lives here has a deep and strong sense of place. There is a
powerful connection to the lands and waters and resources of Bristol Bay. It is a connection that starts
before birth. It is genetic. It is handed down through the generations and it is also learned from a very
young age. A connection told in stories from parents and Elders and experienced firsthand. Toddlers
accompany parents and grandparents fishing, hunting, berry picking. They participate at home to store
that food and save it. It is part of the family experiencing for anyone who grows up in Bristol Bay, and as a
result, this land its water and its resources become a part of who you are. This is a connection without a
price tag and it cannot be replaced. If it lost, it is lost forever."
Although large-scale mining would inject some market-based economic benefits for a period, it would
likely have only modest direct employment benefits in the local region, based on resource extraction
experiences in other rural Alaska areas (Goldsmith 2007). For example, at the Red Dog Mine, ownership
of the resource empowered the NANA Regional Corporation, Inc. (NANA) to negotiate a development
agreement with strong protections and benefits to Northwest Alaska Natives (Storey and Hamilton
2004). NANA shareholders account for approximately 56% of the mine's 464 full-time employees and
91% of the 78 part-time employees. Although first preference in hiring goes to shareholders, as do most
of the training slots, shareholders disproportionately occupy the mine's lower-skilled positions (Storey
and Hamilton 2004). Additionally, the supplemental environmental impact statement showed that
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ -M April 2013
-------
Chapter 12 Fish-Mediated Effects
employment at the Red Dog Mine may have facilitated community residents to relocate to Anchorage for
lifestyle or economic reasons (USEPA 2009, Storey and Hamilton 2004).
For those benefiting from employment opportunities, there would likely be decreased participation in a
subsistence way of life. For the development of the Red Dog Mine supplemental environmental impact
statement, interviews were conducted that included questions asking Noatak and Kivalina residents
about their employment history related to the Red Dog Mine, including the company itself and its
subcontractors, and those companies' subsistence leave policies. Responses were mixed regarding
whether or not interviewees were aware of a subsistence leave policy and whether or not the policy
worked. Some of the companies did not have subsistence leave policies, so workers conducted
subsistence activities during their weeks off or would take personal time. Where the companies did have
policies for subsistence leave, an average of 46% of respondents were unsure whether or not the policy
worked.
The creation of mining-related jobs for local residents, and attendant increases in the region's cash
economy, are often mentioned as potential benefits of large-scale mining development. However,
increases in personal income may not be the best measure of benefits, and should be considered over
the long term, as oil and gas resources are exhausted and future opportunities—including subsistence
resources—are potentially damaged. These types of damages resulting from exploration and
development persist, even when resource extraction ceases (NRC 2003).
Another example of economic and employment issues is illustrated in the disproportionately low
number of Inupiat people employed by the oil and gas industry on Alaska's North Slope, although this
may partially result from the large percentage of young people in the population (NRC 2003). The Alaska
Department of Labor reported that, of the 7,432 people who reported working in the oil and gas sector
on the North Slope in 1999 (and worked for companies that collected and reported residency
information), only 64 lived in the state's Northern Region—the Nome, North Slope and the Northwest
Arctic boroughs. A variety of factors affected both male Inupiat willingness to work in the oil fields and
the desire of the companies in Prudhoe Bay to hire them (NRC 2003, Kruse et al. 1983).
• Inupiat often wish to participate both in the cash economy and in the subsistence harvest.
• Inupiat often prefer borough jobs because they are local and offer flexible time off for hunting.
• Inupiat at Prudhoe Bay find they are a small minority in the primarily white workforce that
sometimes expresses hostility towards Alaska Natives.
• Jobs often available to the Inupiat are often perceived as menial or token jobs.
• Employment by the oil companies can threaten participation in hunting the bowhead whale, which
provides high cultural status.
12.2.4 Social, Cultural, and Spiritual Impacts
The inability to harvest salmon from portions of these watersheds would result in some degree of
cultural disruption, which goes well beyond a loss of food supply. Boraas and Knott (Appendix D) state,
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 12~12 April 2013
-------
Chapter 12 Fish-Mediated Effects
"The people in this region not only rely on salmon for a large proportion of their highly nutritional food
resources; salmon is also integral to the language, spirituality, and social relationships of the culture."
On a cultural level, a significant loss of salmon would result in negative stress on a culture that is highly
reliant on this resource. Boraas and Knott (Appendix D) discuss and document several of the social
values and activities that are integrated with subsistence such as sharing and generalized reciprocity,
fish camp, steam baths, gender and age equity, and wealth. Likewise, they document how spirituality
and psychological health of the cultures are integrated with the natural world, and specifically with
salmon. Of particular importance is the sharing and passing along of traditional knowledge to future
generations. This knowledge transfer occurs in several ways but one critical component is fish camp.
According to Boraas and Knott (Appendix D):
Families typically view fish camp as a good time when they can renew bonds of togetherness by
engaging in the physical work of catching and processing salmon. Family members who don't live
in the villages often schedule vacation time to return home to fish camp, not just for the salmon
but for family. The importance of sharing in vigorous, meaningful work cannot be overestimated. It
creates cross-generational bonds between children, their parents, aunts, uncles, and/or
grandparents that, today are rare in Western culture because there are so few instances in which
meaningful, multi-generational work occurs.
Some interviewees expressed fear of the future that a traditional prophecy of "bad times" told by
Elders might be coming true due to economic development resulting in cultural loss characterized
as "anomie," the loss of meaningfulness, sense of belonging, and direction in life. Anomie
increases cultural and individual risk for social ills such as depression and suicide, alcoholism and
drug abuse, domestic violence, and aggressive behaviors. Healing practices can include those
used for trauma and post-traumatic stress disorders, includingtraditional practices that reconnect
the individual to society and the natural environment through meditative rituals. Culture camps
and other methods of cultural revitalization can be both preventative and healing for children and
adults of indigenous cultures.
Studies on disruption to Alaska Native cultures from resource extraction industries illustrate the
potential social and cultural effects of large-scale mining and effects on a key subsistence resource in the
Nushagak and Kvichak River watersheds. Land use by Alaska Natives on the North Slope has been for
the most part non-intensive, leaving few traces on the landscape outside the established villages. In
contrast, oil development has altered the landscape in ways that will persist long after resource
extraction activities have ceased. Testimony repeatedly cited "scars on the land" that result from
industrial development, and indicated that these scars have altered both the physical and spiritual
elements of the landscape, and thus the very basis of Alaska Native cultures on the North Slope (NRC
2003).
Alterations to the North Slope physical environment have had aesthetic, cultural, and spiritual effects on
human populations (NRC 2003). These alterations have resulted primarily from the construction of
roads, pipelines, buildings, and power lines and from off-road travel. Hunters report that they do not
hunt in the oil fields for aesthetic reasons. North Slope residents have reported that the imposition of a
huge industrial complex on the Arctic landscape was offensive to the people and an affront to the spirit
of the land.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 12~13 April 2013
-------
Chapter 12 Fish-Mediated Effects
North Slope residents report that there has been a vast increase in the time, effort, and funding
necessary to respond politically and administratively to the ever-multiplying number of projects
proposed in the region (NRC 2003). Local residents must attend industry-related meetings and hearings
and review documents, because they believe that decisions will be made that can significantly affect
both their daily lives and future generations. Additionally, Alaska Natives stated that increasing anxiety
about offshore and onshore development is widespread in North Slope communities. Hunters worry
about contamination of the food they consume and know that their health will suffer if they are unable
to eat as their ancestors did. They worry about not being able to provide for their families, or about the
added risk and expense if essential and traditional foods are harder to find. Elders who are no longer
able to provide for themselves worry about the challenges younger hunters face. Families worry about
the safety of hunters who must travel farther and more often if game is not easily accessible (NRC 2003).
According to NRC (2003), increased alcoholism, drug abuse, and child abuse have resulted from the
stresses inherent in integrating traditional and new ways of life. Health effects also are apparent, as the
incidence of diabetes has increased with higher consumption of non-subsistence foods (NRC 2003). The
North Slope Borough bears the costs of these social stresses and provides services such as counseling,
substance abuse treatment, public assistance, crisis lines, shelters, and other social service programs. It
also supports search and rescue services and the police force that responds to domestic violence and
other situations arising when communities are subjected to long-term and persistent stress. The
borough supports biologists, planners, and other specialists who review and offer recommendations on
lease sale, exploration, and development project documents that are produced each year, and bears the
expense of traveling to Fairbanks, Anchorage, Juneau, Seattle, and Washington, DC, where agencies with
permitting authority make decisions that affect their way of life (NRC 2003).
The goal of a more recent study on the effects of oil and gas development on subsistence harvesters on
the North Slope (Braund and Kruse 2009, Braund and Associates 2009) was to enhance benefits and
mitigate impacts of development. This study reported that, despite raising concerns about oil
development as early as 1975, the Inupiat have, until recently, been successful in maintaining their
subsistence lifestyle. Since 2003, North Slope active harvesters have been experiencing impacts of oil
development at higher rates and their wellbeing has declined. This has led to social problems, including
higher rates of drug and alcohol abuse and suicide.
A study that looked at the social, cultural, and psychological impacts of the Exxon Valdez oil spill
determined that the psychosocial impacts of contamination were as significant as the physical impacts
on the environment (Palinkas et al. 1993). Reported issues included declines in traditional social
relations with family members, friends, neighbors, and coworkers; perceived increases in the amount of
and problems associated with drinking, drug abuse, and domestic violence; a decline in perceived health
status and an increase in the number of physician-verified medical conditions; and increased post-spill
rates of generalized anxiety disorder, post-traumatic stress disorder, and depression (Palinkas et al.
1993).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 12~14 April 2013
-------
Chapter 12 Fish-Mediated Effects
Community-wide pre-occupation with the spill and cleanup affected traditional social relations and
resulted in conflicts and divisiveness, arguments about environmental effects of the spill, issues of fault
and responsibility, issues of whether to work on the cleanup or not, and related monetary and
employment issues (Palinkas etal. 1993). There were pervasive fears and increased fundamental
concerns about cultural survival for many residents in the affected Alaska Native villages.
This study documented the profound impact that exposure to the oil spill had on social relations,
traditional subsistence activities, the prevalence of psychiatric disorders, community perceptions of
alcohol and drug abuse and domestic violence, and the physical health of Alaskan Native and non-native
residents of the affected communities (Palinkas et al. 1993). Although the specifics of the Exxon Valdez
oil spill may be quite different, a large-scale or long-term failure of mine waste collection, treatment, and
containment systems would produce a similar reduction of subsistence activities, and similar social and
cultural effects may be expected.
12.2.5 Mitigation and Adaptation
It is not likely that any direct or indirect loss of subsistence use areas resulting from the mine scenario
footprints could be avoided. Under the mine scenarios, the mine pit, waste rock piles, and TSFs would
remain on the landscape in perpetuity and thus represent permanent habitat loss. Some measures could
be put in place to prevent and respond to accidents and spills. Small spills and releases that are
contained in a timely manner may not affect the salmon subsistence resource. However, large-scale
releases, even with active remediation, would have long-term effects on the salmon subsistence
resource and Alaska Native cultures. Because the Alaska Native cultures in this area have significant ties
to specific land and water resources that have evolved over thousands of years, it would not be possible
to replace the value of lost subsistence use areas elsewhere, or to relocate residents and their cultures,
making compensatory mitigation infeasible (Appendix J).
The ability of Alaska Native cultures to adapt to a loss of subsistence use areas, or to the larger impact of
a mine failure or accident, is unknown. Several studies have considered adaptation related to
subsistence resources. Holen (2009) studied the adaptations related to the Nondalton subsistence
fishery and identified two major socio-cultural factors that could potentially affect the long-term
resilience of the fishery: children and young adults are not actively participating in subsistence salmon
fishing as they have in the past; and summer is often when seasonal employment is available, and some
residents miss the subsistence fishing season because of work obligations. These factors decrease the
inter-generational transfer of existing knowledge and wisdom.
On Alaska's North Slope, the issue has not been a question of whether Alaska Natives adapt to oil and gas
development, butrather the consequences of that adaptation (NRC 2003). There are two potential
problems: the loss—sometimes quickly—of traditional languages, patterns of behavior, economic
activities, skills and capital improvements that are no longer relevant; and the use of human and
financial capital and non-renewable resources by the new development (NRC 2003).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 12~15 April 2013
-------
Chapter 12 Fish-Mediated Effects
As the cash economy develops and Alaska Natives become involved in discussions about how changes
associated with oil and gas development affect their cultures, they increasingly must use English as their
primary language. They lose fluency in their native language and the traditional ecological knowledge
embedded in that language. Many North Slope residents expressed concern about the loss of their
traditional way of life, while at the same time enjoying the benefits of the cash economy (NRC 2003).
However, over-adaptation can also occur, leaving communities less able to survive in their
environments when extraction activities decline or stop. The significant tax revenues that oil and gas
development have provided North Slope Borough residents are now declining, and the current standard
of living for North Slope residents will be impossible to maintain unless significant external sources of
local revenue are found. If borough revenues decline, residents may face lower standards of living, or be
forced to find other sources of economic activity or migrate to different areas (NRC 2003).
Offshore exploration and development and the announcement of offshore sales have resulted in
perceived risks to Inupiat culture that are widespread and intense. People of the North Slope have a
centuries-old nutritional and cultural relationship with the bowhead whale, and most view offshore
industrial activity as a threat to bowheads and thus their cultural survival. North Slope residents have
generally supported onshore development done in an environmentally responsible manner, but this
onshore development could pose a threat to the Gwich'in culture in Alaska and the Yukon Territory via
impacts on the Porcupine caribou herd.
12.3 Uncertainties
The preceding sections provide a qualitative overview of how wildlife and Alaska Natives may be
affected by mining-associated changes in salmon resources. Because we do not evaluate direct effects of
mining on wildlife and Alaska Natives, this assessment represents a conservative estimate of how these
endpoints could be affected by mine development and routine operations. We focused on a limited suite
of wildlife species (Brna and Verbrugge 2013), but additional species also could be affected by changes
in salmon resources. We also did not consider mining-related changes to all subsistence species.
In addition to these scope-related limitations, there are several uncertainties inherent in our
consideration offish-mediated effects on wildlife and Alaska Natives.
• The magnitude of salmon-mediated effects on wildlife, subsistence resources, and indigenous
cultures cannot be quantified at this time, and is uncertain. Ultimately, the magnitude of overall
impacts will depend on many factors, including the location of effects, the temporal scale of effects,
cultural resilience, the degree and consequences of cultural adaptation, and the availability of
alternative subsistence resources.
• Interactions between salmon and other wildlife species are complex and reciprocal, and the
assessment did not comprehensively evaluate all potential linkages between endpoints. Many of
these linkages have not been well documented or researched (e.g., potential relationships between
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 12~16 April 2013
-------
Chapter 12 Fish-Mediated Effects
MDN, riparian vegetation, and moose and caribou), but may be significant. Therefore, this
assessment likely underestimates salmon-mediated risks to wildlife.
• The magnitude of effects on Alaska Native cultures resulting from any mining-associated changes in
salmon resources is unknown, but other studies related to resource extraction industries (North
Slope, Red Dog Mine) or environmental contamination (Exxon Valdez) in Alaska confirm that there
certainly would be changes in human health and Alaska Native cultures.
• The cumulative effects of mining and climate change represent a significant uncertainty in the
region (Section 3.8, Box 14-2). Residents of the Kvichak River watershed have observed that social
and cultural changes are occurring in an environment where they are also seeing rapid climate
changes (Holen 2009). These changes, which include climate variability and unpredictable weather,
make it difficult to plan for subsistence activities (Appendix D). On Alaska's North Slope, climate
change and oil and gas development together result in greater cumulative effects on the
environment and Inupiat cultural traditions (Braund and Associates 2009). The cumulative effects
of climate change and potential effects on subsistence resources from large-scale mining are
unknown.
• There is little uncertainty about the inability to mitigate or replace subsistence resources or cultural
values lost from effects of large-scale mining because of the significant and long-standing ties Alaska
Native cultures have to the specific land and water resources in these watersheds.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , „ 17 APril 2013
-------
13.1 Overview
Thus far, this assessment has focused on the potential effects of a single mine, described by a range of
mine scenarios. Although the Pebble deposit represents the most imminent and likely site of mine
development in the Nushagak and Kvichak River watersheds, the development of a number of mines, of
varying sizes, is plausible in this region. Several known mineral deposits with potentially significant
resources are located in the two watersheds (Table 13-1), and active exploration of deposits is occurring
in a number of claim blocks (Figure 13-1). If the infrastructure for one mine is built, it would likely
facilitate the development of additional mines. Thus, the potential exists in these watersheds for the
development of a mining district with a number of mines, their associated infrastructure, and induced
development. In this chapter, we briefly consider potential cumulative effects of the establishment of
large-scale mining in the Nushagak and Kvichak River watersheds on Pacific salmon. In addition to
addressing the potential impacts of multiple mines, we briefly consider induced development and
potential increases in accessibility of currently roadless areas.
13.1.1 Definition of Cumulative and Induced Impacts
National Environmental Policy Act regulations define cumulative impacts as "the impact on the
environment [that] results from the incremental impact of the action when added to other past, present,
and reasonably foreseeable future actions regardless of what agency (Federal or non-Federal) or person
undertakes such other actions. Cumulative impacts can result from individually minor but collectively
significant actions taking place over a period of time." Assessing the cumulative impacts of multiple
mines requires considering the impacts of their combined footprints, as well as the cumulative risks of
leaks, spills, and other accidents and failures associated with each individual mine.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-1
April 2013
-------
Chapter 13 Cumulative Effects
Induced effects contribute to the cumulative effects of an action, and are those effects that are "caused
by the action and are later in time or farther removed in distance, but are still reasonably foreseeable"
(43 CFR 1508.8(b)).
Figure 13-2 illustrates how cumulative and induced effects could follow the initiation of large-scale
mining in the Nushagak and Kvichak River watersheds. The original mine—with its associated
transportation corridor, port, power generation facilities, and other infrastructure—likely would start
accumulation of impacts across the watersheds. Mineralized areas in the region (Figure 13-1) are
currently without development infrastructure (e.g., roads, utilities, and airports), which creates an
expensive barrier to development. Thus, it is reasonably foreseeable that development of infrastructure
for an initial mine could make mining cost-effective for other smaller mineral deposits, facilitating
further accumulation of impacts. In addition, the initial and subsequent mines would increase
accessibility of the region, causing other induced development and associated impacts.
As environmental effects on freshwater habitats accumulate, the magnitude of the total impact on the
fishery would increase. Increased spatial dispersion of effects, both within and across watersheds,
means that individual effects may go unnoticed, but still cumulatively affect the greater Bristol Bay
salmon fishery. Cumulative effects would potentially reduce biodiversity of the overall salmon
population (Section 5.2.4) and its resilience to natural and anthropogenic disturbances, thereby
exacerbating total effects on salmon.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . o 2 APril 2013
-------
Chapter 13
Cumulative Effects
Table 13 1. Mining prospects, in addition to the Pebble deposit, with more than minimal recent
exploration in the Nushagak and Kvichak River watersheds.
Prospect
AUDN
Big Chunk North
Big Chunk South
Fog Lake
Groundhog
Humble
Iliamna
Kamishak
Kaskanak
Kisa
Northern Bonanza
Shotgun
Sleitat
Pebble South/PEB (38/308 Zones/BOO)
Stuy
Resource
Porphyry copper
Porphyry copper
Porphyry copper
Gold, copper
Porphyry copper
Porphyry copper
Porphyry copper
Porphyry copper
Porphyry copper
Gold
Gold
Gold
Tin, tungsten
Porphyry copper
Porphyry copper
References
Millrock Resources 2011
Millrock Alaska 2012a
U5 Resources 2010
AHEA2012
Big Chunk Corp. 2012
Liberty Star 2012a
Alix Resources 2008
USGS 2008
Kennecott Exploration Co. 2011
Szumigala etal. 2011
Millrock Resources 2011
Millrock Alaska 2012b
Bristol Exploration Co., Inc. 2011
AERI 2008
Full Metal Minerals 2008, 2012
Golden Lynx 2009
Northern Bonanza Trust 2011
TNR Gold Corp. 2011, 2012a
ADNR 2012a
Thor Gold Alaska, Inc. 2011
Full Metal Minerals 2008, 2012
PLP 2011
Szumigala etal. 2011
Stuy Mines, LLC2010
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-3
April 2013
-------
Chapter 13
Cumulative Effects
Figure 13 1. Claim blocks with more than minimal recent exploration in the Nushagak and Kvichak
River watersheds.
Northern Bonanza
Shotgun
Fog Lake
Cook Inlet
Kamishak
Bristol Bay
IN
A
25 50
] Kilometers
25 50
] Miles
Approximate Pebble Deposit Location
Towns and Villages
Watershed Boundary
Existing Roads
Active Mining Claims
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-4
April 2013
-------
Chapter 13
Cumulative Effects
Figure 13 2. Conceptual model illustrating potential cumulative effects of multiple large scale
I initial mine
1
/ \
/ \
/ \
/
[' r.rt '} (^transportation 1 f povver generation & } \ other ancillary
\ / X corridor J I transmission facilities J L facilities
f accessibility
of region
V
I fuel, energy
5.freight costs
f e c o n o rn i c fe asi b i I ity of
additional mine development
V
v
[ additional | [ additional ancillary 1 | additional
I mines J I facilities J I infrastructure
induced ']
development J~
\/
'T effective
impermeable surface
accumulation of environmental impacts
on freshv.-'ater habitats
- spatial extent
of impacts
T magnitude of
impacts
effects on 'T' number of
intraspecificfish stocks
I fish population
resiliency
V
-t fish effects
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-5
April 2013
-------
Chapter 13 Cumulative Effects
13.1.2 Vulnerability of Salmonids to Cumulative Impacts
Throughout the range of Pacific salmon, most ecosystems outside of Bristol Bay face the cumulative
effects of multiple land and water uses within and across watersheds, resulting in a variety of stressors
that occur in combination. Anadromous and resident fish stocks in these watersheds are subject to
persistent disturbance-induced stresses, the effects of which accumulate through the river network. For
example, sedimentation of spawning beds from accelerated erosion, loss of rearing habitat from filling of
streamside wetlands, and reduced out-migration success from downstream channelization are separate
effects that together have a cumulative impact on fishes in a river system. The effect of each stressor
accumulates regardless of whether factors occur at the same time, or even in temporal proximity.
Because Pacific salmon, Dolly Varden, and rainbow trout are migratory, at least within a given stream
system, adverse impacts can even accumulate when fish are absent from a particular reach. The overall
result of these cumulative effects has been the reduction and even extinction of salmonid populations.
Historical salmon losses have resulted from the cumulative impacts of many land use activities over a
time span of 150 years (NRC 1996). Salmon depend on adequate supplies of clean, cool water
throughout the freshwater portions of their lifecycles. In addition, well aerated streambed gravels are
essential for spawning. These instream conditions depend on the overall health of the entire watershed
(NRC 1996). Human development in the watershed can degrade these conditions by increasing
sedimentation, raising water temperature, degrading water quality, changing water flow, and reducing
water depths (NRC 1996).
In the Pacific Northwest, the four principle factors responsible for the degradation of salmon stocks have
been referred to colloquially as the "four H's": habitat degradation and loss, hydroelectric dams and
other impoundments, harvest practices, and hatchery propagation (Ruckleshaus etal. 2002). Of these
factors, habitat degradation and loss is the most likely to affect salmon stocks in the Nushagak and
Kvichak River watersheds. In the Pacific Northwest, habitat degradation and loss related to human land
use have obviously been a major factor in salmon declines by reducing population productivity, adult
densities, and early-life-stage production over large geographic areas (Ruckleshaus et al. 2002).
13.1.3 Nature and Extent of Past, Present, and Future Impacts
In cumulative impact analyses, the U.S. Environmental Protection Agency (USEPA) generally evaluates
any past, present, and reasonably foreseeable future actions that are spatially and temporally linked
and, therefore, can act in combination on the resource(s) of interest. In the Bristol Bay watershed, the
contribution of past or present actions to degradation of salmon habitat and populations is minimal. To
date, the Nushagak and Kvichak River watersheds have experienced minimal cumulative stresses
associated with human activity, and their ecosystems are relatively undisturbed by significant human
development (Section 3.7). Large-scale, human-caused modification of the landscape—a factor
contributing to the extinction risk for many native salmon populations in the Pacific Northwest (Nehlsen
etal. 1991)—is absent from these watersheds, and development consists of only a small number of
towns, villages, and roads. There are no hydroelectric dams and reservoirs and no salmon hatcheries. In
fact, harvesting is the only one of the four H's that have devastated salmon stocks of the Pacific
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T7April 2013
-------
Chapter 13 Cumulative Effects
Northwest that has thus far been more than minimal in the Nushagak and Kvichak River watersheds.
There have been periods of poor harvesting practices and overfishing in Bristol Bay in the past, but
when Alaska achieved statehood, the Alaska Department of Fish and Game (ADF&G) assumed
management of the fishery and established the primary objective of restoring the runs to their former
abundance (Box 5-2, Appendix A). The ADF&G management strategy, based on maximum sustainable
yield, is considered a success in maintaining sustainable salmon harvests (Hilborn et al. 2003, Hilborn
2006). Indeed, since the late 1970s, sockeye salmon catch, spawning stock, and total return have been at
record levels. Although there has been some concern that harvest of returning salmon has reduced
ecosystem productivity in this region, Hilborn (2006) found that paleoecological analysis of returns does
not indicate decreased salmon production due to commercial fishing.
Reasonably foreseeable future development can be predicted based on project approvals, planning
documents, and data on local trends. The Bristol Bay Area Plan for State Lands (BBAP) (ADNR 2005)
provides information on reasonably foreseeable mining in the Nushagak and Kvichak River watersheds.
Although the Pebble deposit represents the most imminent and likely site of mine development in the
watersheds, the development of several mines of varying sizes is plausible in this region. Several known
mineral deposits with potentially significant resources are located in the two watersheds, and active
exploration of deposits is occurring in a number of claim blocks (Figure 13-1, Table 13-1). Reasonably
foreseeable non-mining development is discussed in Section 13.4.
13.2 Cumulative Impacts from Multiple Mines
Construction of mining and transportation infrastructure at and for the Pebble deposit would
substantially reduce development costs for surrounding prospects and could facilitate creation of a
mining district. Based on planning documents and current patterns of mineral exploration in the
Nushagak and Kvichak River watersheds, it is possible to identify a scenario for potential mine
development in the region over the next 50 to 100 years. Although this scenario is plausible given
available information, it should be kept in mind that it is impossible to predict with certainty what
mining activities would occur in the region in the future, the order in which mines would be developed,
or the specific impacts of those mines.
The BBAP assigns land use designations to discrete areas of state-owned or selected lands called
management units. These designations represent the uses and resources for which ADNR will manage
the units. In the Nushagak and Kvichak River watersheds, the BBAP assigns the land use designation
"Mi" (Minerals) to seven management units (Shotgun, Sleitat, Kemuk, Fog Lake, and units 06-23, 06-24,
and 10-02 in the Pebble deposit area), totaling more than 1,300 km2 (ADNR 2005). The Mi designation
applies to areas "associated with significant resources, either measured or inferred, that may experience
minerals exploration or development during the [BBAP's] planning period" of 20 years. The BBAP also
allows for mining on units with other designations (e.g., General Use, Public Recreation and Tourism-
Dispersed), which far exceed the area of those designated Mi although the BBAP does not describe them
as having the same known potential for mining.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7TZApril 2013
-------
Chapter 13 Cumulative Effects
Since 2008, there has been exploration in all of the BBAP Mi-designated management units, as well as in
claim blocks with several other designations. Table 13-1 lists mine prospects experiencing more than
minimal recent exploration activity. These target areas could be future mine sites, if exploration
identifies marketable quantities of metals. Other mineral claim blocks exist in the watersheds, but at the
time of this writing they had experienced limited exploration in recentyears (ADNR 2012b, 2012c,
2012d).
Any potential mine site would presumably include a mine pit and an adjacent waste rock disposal area
(Chapter 6). For analysis purposes, we assume other ore bodies in the area would be more typical of the
median size, based on the range of worldwide porphyry copper deposits (Table 4-3). Thus, we used the
Pebble 0.25 scenario, which is comparable to a median size porphyry copper mine to characterize the
footprint of additional mines. The Pebble 0.25 scenario represents 250 million tons (227 million metric
tons) or ore, resulting in a mine pit and waste rock disposal area of approximately 1.5 and 2.3 km2,
respectively (Table 6-2). Mines not affiliated with or more distant from a previously developed mine
would also require one or more TSFs (5.9 km2 in the Pebble 0.25 scenario) (Table 6-2), as well as a mill
and other operational infrastructure as described in Box 6-1. Any additional mines would also require
construction of transportation infrastructure, including access roads, pipelines, and possibly port
facilities.
To examine the potential scope of cumulative impacts from large-scale mining, we consider
development of additional mines at six potential sites where there was notable activity and/or
investment in drilling or other exploration in 2011-2012: Pebble South/PEB (PLP/NDM claim block),
Big Chunk North, Big Chunk South, Groundhog, Humble, and AUDN/Iliamna prospects (Figure 13-1).
This list does not include four other prospects designated Mi in the BBAP: Shotgun, Sleitat, Kemuk, and
Fog Lake. Kemuk is an older name for the Humble prospect. The other three prospects are not porphyry
copper deposits. Because exploration of the six selected prospects began approximately 15 to 25 years
later than at the Pebble deposit, mine development at these sites could be 20 years or more into the
future (Millrock Resources 2011, ADNR 2012e, 2012f, and 2012g, Liberty Star 2012a, TNR Gold Corp.
2012b). We describe the waters, fish, and subsistence resources that could be affected by mines at these
locations (Tables 13-2 through 13-7). The sources of information for these tables are the Alaska
Anadromous Waters Catalogue and Freshwater Fish Inventory (Johnson and Blanche 2012, ADF&G
2012) and a series of technical papers from ADF&G on subsistence harvest and use by villages in the
Nushagak and Kvichak River watersheds (Fall et al. 1986, Fall et al. 2006, Krieg et al. 2009, Schichnes
and Chythlook 1991). We also estimate the stream lengths and wetland areas that could be eliminated
by the mine footprints. Box 13-1 describes the methodology for estimating these impacts; results are
summarized in Table 13-8. It is important to note that we did not estimate the size of the hydrologic
drawdown zones around dewatered pits at the six additional mines as we did at the Pebble site, so our
estimates of habitat lost to the mine footprints are conservative. Inclusion of the drawdown area in the
Pebble 0.25 scenario increases the area of stream and wetland losses by 84%. Similar increases in lost
habitat estimates, subject to variations in the local geology, would be expected at each of the other mine
sites.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T7April 2013
-------
Chapter 13 Cumulative Effects
BOX 13 1. METHODS FOR ESTIMATING IMPACTS OF OTHER MINES
To estimate the extent of aquatic habitat that each mine would eliminate, we overlaid a typical footprint area
onto the stream, water body, and, where available, wetland densities for each entire claim block. In this way
the analysis is less affected by uncertainty about the precise location of a potential mine within the claim
block. Since the Pebble South/PEB prospects are associated with the PLP/NDM claims, we used the aquatic
area densities for the PLP/NDM claims to assess that potential mine.
We derived the boundaries of the prospects from ADNR's State Mining Claims dataset (ADNR 2012h). We
then determined stream and water body density usingthe National Hydrography Dataset (NHD) for Alaska
(USGS 2012). For wetland density, we used the National Wetlands Inventory (NWI) coverage for the
Groundhog, AUDN/lliamna, and Pebble South/PEB (part of the PLP/NDM claim block) prospects (USFWS
2012). The NWI covered 95, 69, and 58% of these prospects, respectively. For these three areas, we
compared NHD water body density to NWI wetland density to ascertain the efficacy of using the former as a
surrogate for wetland density in the other areas where NWI coverage is lacking. This analysis revealed that
the NHD water body dataset severely underestimates wetland density in area of overlap: NWI wetland
density was 9.7 to 13.7 times the NHD water body density. Thus, for the three prospects with no NWI
coverage (Big Chunk North, Big Chunk South, and Humble), we calculate a range of wetland impacts, using
NHD water body density as a minimum bound and 13.7 times that density as a maximum bound. We also
provide a range of wetland impacts for the three claim blocks with NWI coverage, because the NHD water
body density in the NWI-covered area of all three prospects was lower than for the claim block as a whole.
For those three claim blocks (Groundhog, AUDN/lliamna, and Pebble South/PEB), the higher estimate
applies the wetland-to-water body differential from the area of NWI/NHD overlap to the full claim block's
higher water body density.
For Pebble South/PEB, we used a direct impact footprint that represents only the typical mine pit and waste
rock disposal area, based on our assumption that any mine at that site would use the mill, tailings storage
facility, and other facilities at the initial mine at the Pebble deposit (Section 13.3.1.1). We applied a similar
assumption as a lower bound for Big Chunk South, Big Chunk North, and Groundhog; for all three, the upper
bound represents a stand-alone mine, with no shared facilities (Sections 13.3.2 through 13.3.4). We
assume no sharing of mine facilities for AUDN/lliamna or Humble, based on their more remote locations
(Sections 13.3.5 and 13.3.6).
13.2.1 Pebble South/PEB
13.2.1.1 Description
The Pebble South/PEB prospect, which is part of the PLP/NDM claim block, includes the 38 and
308 Zone prospects, approximately 15 km south west of the Pebble deposit (Ghaffari etal. 2011). The
BOO prospect is 4km south of 308 Zone on claims held by Full Metal Minerals (USA), Inc. (Ghaffari etal.
2011, Full Metal Minerals 2012). Full Metal entered into an option agreement with PLP/NDM; if
completed, the option will result in at least 60% PLP/NDM interest in the claims (Full Metal Minerals
2012).
Due to their proximity to the Pebble deposit, we assume that any future mines at Pebble South/PEB
would use the TSFs, mill, and other operational infrastructure initially built for potential mining at the
Pebble deposit. Thus, we anticipate that the primary additional development associated with this
prospect would be a mine pit(s), waste rock area(s), and a transportation corridor to existing
operational infrastructure.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . o g April 2013
-------
Chapter 13 Cumulative Effects
13.2.1.2 Potentially Affected Waters, Fish, and Subsistence Uses
Table 13-2 summarizes information on the waters, fish, and subsistence uses potentially affected by a
mine at the Pebble South/PEB prospect The 38 and 308 Zones occur near the south edge of the South
Fork Koktuli River watershed, within the Mulchatna River watershed of the Nushagak River watershed.
The upper reaches of some streams on Sharp Mountain likely are too steep to provide fish habitat, and
there have been few fish surveys in this area to date, even in the lower reaches of those streams. Drilling
on the BOO prospect has been just south of the divide between the Nushagak and Kvichak River
watersheds, in the uppermost portion of the Lower Talarik Creek watershed, which flows to Iliamna
Lake. BOO is located approximately 2 km southwest of the fish-bearing stream that drains the south side
of Sharp Mountain. No fish survey data are available for the immediate area of the BOO prospect, and the
streams may be spatially intermittent.
Connecting to infrastructure at an existing mine at the Pebble deposit likely would involve following the
South Fork Koktuli River upstream. This route would presumably involve crossing the river, as well as a
number of tributaries, water bodies, and wetlands.
Waters of both the Nushagak and Kvichak River watersheds could be affected at this site, though no
information is available on the fish resources in the tributaries of the Kvichak River that would be
affected by mining of the BOO prospect. The tributaries of the Nushagak River that would be affected by
mining of the 38 and 308 Zone prospects are known to contain four Pacific salmon species and Dolly
Varden. People from seven Alaska Native villages hunt for bear, moose, caribou, other mammals, and
birds in these areas, but no subsistence fishing has been reported (Table 13-2). Based on the average
stream density in the area of the prospect, the mine footprint would eliminate 4.1 km of streams and
between 0.71 and 1.2 km2 of wetlands (Table 13-8).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . o H Q April 2013
-------
Chapter 13
Cumulative Effects
Table 13 2. Waters, fish, and subsistence uses potentially affected by a mine at the Pebble South/PEB prospect.
Location
38/308 Zones and
BOO Prospect
Transportation
Corridor to Pebble
Sited
Downstream of
Mine(s) and
Transportation
Corridor
Affected Waters
Watersheds and Named Subwatersheds
Nushagak River
X
X
Mulchatna River
X
X
X
Koktuli River
X
X
X
South Fork Koktuli River
X
X
Kvichak River
X
X
(D
Ji
3
CB
c
E
CO
X
X
Lower Talarik Creek
X
X
Waters3
14 unnamed streams
2 unnamed lakes and
several ponds
Unknown extent of wetlands
4 unnamed first-order
streams
Numerous ponds
Unknown extent of wetlands
South Fork Koktuli River
>4 unnamed tributaries
Unknown extent of wetlands
South Fork Koktuli River
Lower Talarik Creek
Fish Species'1
Unknown Lamprey
Northern Pike
Round Whitefish
Unknown Whitefish
Coho Salmon
A-Sp,
J
Chinook Salmon
J
Sockeye Salmon
J
Chum Salmon
A
Pink Salmon
CD
.C
O
o
'•s
^
Rainbow Trout
Dolly Varden
A, J
Arctic Grayling
J
s
.Q
3
CQ
Ninespine Stickleback
A, J
Threespine Stickleback
Slimy Sculpin
A, J
Unknown
A, J
A, J
A
A, J
A, J
X
A-Sp,
J
A-Sp,
J
A-Sp,
J
J
A-Sp,
J
A-Sp,
J
A-Sp,
J
X
X
A, J
A, J
A, J
A, J
A, J
X
A, J
A, J
A
A, J
A, J
A, J
Subsistence Usec
Village
Dillingham
Ekwok
Igiugig
Iliamna
Newhalen
Nondalton
Port Alsworth
Dillingham
Ekwok
Igiugig
Iliamna
Newhalen
Nondalton
Port Alsworth
Igiugig
Iliamna
Kokhanok
Newhalen
NewStuyahok
Nondalton
Target Species/Group
Salmon
X
Other Salmonids
X
X
X
X
.c
lit
L_
CD
£
o
X
X
CB
(D
CO
|
CO
X
X
X
(D
8
O
X
X
X
X
X
X
X
Caribou
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Other Mammals
X
X
X
X
X
X
X
X
Waterfowl
X
X
X
X
in
•o
bo
CD
£
o
X
X
X
X
Notes:
a Use of the term "tributary" indicates the channel flows into the stream listed above it, while "stream" indicates the receiving water is off site, identified in the columns to the left. For waters downstream of the mine and transportation corridor, fish and subsistence are noted only where different from the waters' prior listing.
b A = adult; Sp = spawning; J = juvenile; x = unknown life stage. "Unknown" indicates apparent lack of surveys.
c Subsistence uses not separated by watershed. Uses noted only for areas of direct impacts (i.e., mine and/or transportation corridor), if it occurs there; otherwise, noted for areas downstream. "Downstream applies only to the drainage immediately downstream (e.g., Lower Talarik Creek, but not Iliamna Lake or the Kvichak River). Data
for Dillingham and Ekwok are less detailed and more dated than for other villages, so use patterns may have changed.
d Assumes existing mine infrastructure at the Pebble deposit; hypothetical routing of the new corridor minimizes distance, topographic gradients, and stream crossings and assumes water body crossings avoided.
Sources: Fall etal. 1986, Schichnesand Chythlook 1991, Fall etal. 2006, Kriegetal. 2009, ADF&G 2012, Johnson and Blanche 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-11
April 2013
-------
Chapter 13
Cumulative Effects
Table 13 3. Waters, fishes, and subsistence uses potentially affected by a mine at the Big Chunk South prospect.
Location
Claim Block
Transportation
Corridor to
Pebble Sited
Downstream
of Mine(s) and
Transportation
Corridor
Affected Waters
Watersheds and Named Subwatersheds
Nushagak River
X
X
Mulchatna River
X
X
1
(£
1
O
^
X
X
North Fork Koktuli River
X
X
Kvichak River
X
X
X
X
0
^£
2
ra
c
E
CO
X
X
X
X
Newhalen River
X
X
Ji
CD
O
£
3
X
X
Chulitna River
X
X
Upper Talarik Creek
X
X
Waters3
>66 unnamed streams
Nikabuna Lakes
130 other lakes and
ponds
Extensive wetlands
North Fork Koktuli
River
>7 unnamed
tributaries
Unknown extent of
wetlands
Upper Talarik Creek
x3 unnamed
tributaries
Chulitna River
North Fork Koktuli
River
Upper Talarik Creek
Fish Species'1
Northern Pike
A, J
A, J
Longnose Sucker
J
A
Humpback Whitefish
A
Least Cisco
A, J
A, J
Pygmy Whitefish
J
Round Whitefish
J
Round Whitefish
J
Coho Salmon
A
Sp,
J
A-
Sp,
J
Chinook Salmon
J
J
A-
Sp
A-
Sp,
J
Sockeye Salmon
A
X
X
A-
Sp,
J
A-
Sp,
J
Chum Salmon
A-
Sp
A-
Sp,
J
Pink Salmon
X
CD
6
O
«
^
X
X
Rainbow Trout
J
A
Dolly Varden
A, J
A, J
A, J
Arctic Grayling
A, J
J
A, J
A
1
3
to
A, J
Ninespine Stickleback
A, J
A
A, J
Threespine Stickleback
A, J
A
Slimy Sculpin
A, J
A, J
A, J
Subsistence Usec
Village
Dillingham
Ekwok
Newhalen
Nondalton
Port Alsworth
Dillingham
Ekwok
Iliamna
Newhalen
Nondalton
Port Alsworth
Igiugig
Iliamna
Newhalen
New
Stuyahok
Nondalton
Port Alsworth
Target Species/Group
Salmon
X
Other Salmonids
.c
lit
Li-
CD
.C
+->
O
X
X
X
CB
(D
CO
1
CO
X
X
X
(D
8
0
X
X
X
X
X
X
X
Caribou
X
X
X
X
X
X
X
X
X
X
X
Other Mammals
X
X
X
X
X
X
Waterfowl
X
X
X
X
(/)
•c
in
(D
.C
+->
O
X
X
X
X
Notes:
a Use of the term "tributary" indicates the channel flows into the stream listed above it, while "stream" indicates the receiving water is off site, identified in the columns to the left. For waters downstream of the mine and transportation corridor, fish and subsistence are noted only where different from the waters' prior listing; use of the
term "system" indicates the presence of tributaries in addition to the named water.
b A = adult; Sp = spawning; J = juvenile; x = unknown life stage.
c Subsistence uses not separated by watershed. Uses noted only for areas of direct impacts (i.e., mine and/or transportation corridor), if it occurs there; otherwise, noted for areas downstream. "Downstream" applies only to the drainage immediately downstream (e.g., the Chulitna River, but not Lake Clark or the Newhalen River). Data for
Dillingham and Ekwok are less detailed and more dated than for other villages, so use patterns may have changed (see Section 13.3.6.2).
d Assumes existing mine infrastructure at the Pebble deposit; hypothetical routing minimizes distance, topographic gradients, and stream crossings and assumes water body crossings avoided.
Sources: Fall etal. 1986, Schichnesand Chythlook 1991, Stickman etal. 2003, Fall etal. 2006, Woody and Young 2006, Kriegetal. 2009, ADF&G 2012, Johnson and Blanche 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-12
April 2013
-------
Chapter 13
Cumulative Effects
Table 13 4 Waters, fishes, and subsistence uses potentially affected by a mine at the Big Chunk North prospect.
Location
Claim Block
Transportation
Corridor to Big
Chunk South'1
Downstream
of Mine(s) and
Transportation
Corridor
Affected Waters
Watersheds and Named Subwatersheds
Nushagak River
X
X
X
X
Mulchatna River
X
X
X
X
Keefer Creek
X
X
(D
>
(£
1
0
X
X
(D
>
cc
c
CO
1
X
X
North Fork Swan
River
X
X
Kvichak River
X
X
X
£
5
CD
C
E
CO
X
X
X
Newhalen River
X
X
X
Ji
CD
o
(D
Ji
3
X
X
X
Chulitna River
X
X
X
Waters3
Chulitna River
>JO tributaries
xLOO lakes and ponds
Unknown extent of wetlands
Headwaters Keefer Creek
>7 tributaries
X35 lakes and ponds
Unknown extent of wetlands
Unnamed stream
Chulitna River
x4 tributaries
Unknown extent of wetlands
Chulitna River
Keefer Creek
North Fork Swan River
Fish Species'1
Arctic-Alaskan
Brook Lamprey
Northern Pike
Longnose Sucker
Humpback
Whitefish
Least Cisco
Pygmy Whitefish
Round Whitefish
Coho Salmon
Chinook Salmon
Sockeye Salmon
Chum Salmon
Arctic Char
Dolly Varden
Arctic Grayling
Burbot
Ninespine
Stickleback
Slimy Sculpin
Unknown
Unknown
Unknown
Unknown
A, J
A, J
A, J
A
A, J
J
J
A-
Sp
A-
Sp,
J
A-
Sp
J
X
X
Sp
A-
Sp
X
X
A, J
J
A, J
J
A, J
A, J
A, J
A, J
Subsistence Usec
Village
Dillingham
Ekwok
Nondalton
Port Alsworth
Dillingham
Ekwok
Nondalton
Port Alsworth
Iliamna
Newhalen
New
Stuyahok
Nondalton
Port Alsworth
Target Species/Group
Salmon
Other Salmonids
.G
(/)
Li-
CD
1
X
X
CO
c
o
m
X
X
Moose
X
X
X
Caribou
X
X
X
X
X
X
X
X
X
Other Mammals
X
X
X
X
X
X
X
Waterfowl
X
X
X
(/)
•c
in
(D
1
X
X
Notes:
a Use of the term "tributary" indicates the channel flows into the stream listed above it, while "stream" indicates the receiving water is off site, identified in the columns to the left. For waters downstream of the mine and transportation corridor, fish and subsistence are noted only where different from the waters' prior listing; use of the
term "system" indicates the presence of tributaries in add tion to the named water.
b A = adult; Sp = spawning; J = juvenile; x = unknown life stage. "Unknown" indicates an apparent lack of surveys.
c Subsistence uses not separated by watershed. Uses noted only for areas of direct impacts (i.e., mine and/or transportation corridor), if it occurs there; otherwise, noted for areas downstream. "Downstream applies only to the drainage immediately downstream (e.g., North Fork Swan River, but not the Swan or Koktuli Rivers). Data for
Dillingham and Ekwok are less detailed and more dated than for other villages, so use patterns may have changed (see Section 13.3.6.2).
d Waters, fish, and subsistence uses limited to those upstream of the Big Chunk South claim block. See Table 13-5 for resources potentially affected by connecting a transportation corridor through and beyond that block. Hypothetical routing minimizes distance, topographic gradients, and stream crossings and assumes water body
crossings avoided.
Sources: Fall etal. 1986, Schichnesand Chythlook 1991, Stickman etal. 2003, Fall etal. 2006, Woody and Young 2006, Kriegetal. 2009, ADF&G 2012, Johnson and Blanche 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-13
April 2013
-------
Chapter 13
Cumulative Effects
Table 13 5. Waters, fishes, and subsistence uses potentially affected by a mine at the Groundhog prospect.
Location
Claim Block
Transportation
Corridor to
Pebble Sited
Affected Waters
Watersheds and Named Subwatersheds
Nushagak River
X
Mulchatna River
X
(D
>
(£
1
0
^
X
(D
X .>
£*
€£
0 0
z ^
X
Kvichak River
X
X
X
X
X
X
X
X
X
Iliamna Lake
X
X
X
X
X
X
X
X
X
Upper Talarik
Creek
X
X
Newhalen River
X
X
X
X
X
X
X
X
^
CD
o
(D
^
3
X
X
X
X
X
X
X
X
Chulitna River
X
X
X
X
X
X
X
X
^
(D
0
o
^
o
o
cc
X
X
Groundhog Creek
X
Long Lake
X
Koksetna River
X
X
Black Creek
X
Waters3
Chulitna River
>72 unnamed tributaries
Joe Nort Lake
58 unnamed lakes and ponds
Extensive wetlands
Rock Creek
X39 unnamed tributaries
9 unnamed lakes and ponds
Extensive wetlands
Groundhog Creek
>20 unnamed tributaries
30 unnamed ponds and lakes
Unknown wetland extent
^5 unnamed tributaries
Unknown wetland extent
Unknown wetland extent
Black Creek
>7 unnamed tributaries
one unnamed lake
Unknown wetland extent
Unknown wetland extent
>2Q unnamed tributaries
42 unnamed ponds and lakes
Unknown wetland extent
^13 unnamed tributaries
7 unnamed ponds
Unknown wetland extent
>.i unnamed tributary
Unknown extent of wetlands
Fish Species'1
Northern Pike
Longnose Sucker
Humpback
Whitefish
Least Cisco
Round Whitefish
Coho Salmon
Chinook Salmon
Sockeye Salmon
Arctic Char
Rainbow Trout
Dolly Varden
Arctic Grayling
J
Burbot
A, J
Ninespine
Stickleback
Slimy Sculpin
A, J
A, J
Unknown
Unknown
Unknown
Unknown
Unknown
A-
Sp,
J
J
A, J
A, J
A, J
Subsistence Usec
Village
Dillingham
Ekwok
Kokhanok
Newhalen
Nondalton
Port Alsworth
Dillingham
Ekwok
Kokhanok
Newhalen
Nondalton
Port Alsworth
Target Species/Group
Salmon
Other Salmonids
.G
(/)
Li-
CD
*
CO
c
o
X
X
X
X
o
0
X
X
X
X
Caribou
X
X
X
X
X
X
X
X
X
X
X
X
Other Mammals
X
X
X
X
X
X
Waterfowl
X
X
(/)
•c
in
(D
1
X
X
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-14
April 2013
-------
Chapter 13
Cumulative Effects
Table 13 5. Waters, fishes, and subsistence uses potentially affected by a mine at the Groundhog prospect.
Location
Downstream
of Mine(s) and
Transportation
Corridor
Affected Waters
Watersheds and Named Subwatersheds
Nushagak River
X
Mulchatna River
X
(D
1
0
X
(D
^ >
0 0
z ^
X
Kvichak River
X
X
X
X
X
Iliamna Lake
X
X
X
X
X
Upper Talarik
Creek
X
Newhalen River
X
X
X
X
CB
o
(D
3
X
X
X
X
Chulitna River
X
X
X
X
(D
0
O
o
o
cc
X
Groundhog Creek
Long Lake
X
Koksetna River
X
Black Creek
Waters3
Chulitna River
Rock Creek
Unnamed tributaries
Koksetna River
Unnamed tributaries
Unnamed tributaries
Fish Species'1
Northern Pike
A, J
Longnose Sucker
A, J
J
J
A, J
Humpback
Whitefish
A
Least Cisco
A, J
Round Whitefish
J
A, J
A, J
Coho Salmon
A-
Sp,
J
Chinook Salmon
A
A, J
J
Sockeye Salmon
X
A-
Sp
Arctic Char
X
Rainbow Trout
A, J
Dolly Varden
Arctic Grayling
A, J
A
A, J
J
Burbot
A, J
J
Ninespine
Stickleback
A, J
Slimy Sculpin
A, J
A, J
Subsistence Usec
Village
Igiugig
Iliamna
Newhalen
Nondalton
Port Alsworth
Target Species/Group
Salmon
X
Other Salmonids
X
LI-
CD
.c
X
CO
c
o
o
0
X
X
Caribou
X
Other Mammals
X
X
Waterfowl
X
X
X
in
(D
1
X
X
Notes:
a Use of the term "tributary" indicates the channel flows into the stream listed above it, while "stream" indicates the receiving water is off site, identified in the columns to the left. For waters downstream of the mine and transportation corridor, fish and subsistence are noted only where different from the waters' prior listing; use of the
term "system" indicates the presence of tributaries in addition to the named water.
b A = adult; Sp = spawning; J = juvenile; x = unknown life stage. Unknown" indicates an apparent lack of surveys.
c Subsistence uses not separated by watershed. Uses noted only for areas of direct impacts (i.e., mine and/or transportation corridor), if it occurs there; otherwise, noted for areas downstream. "Downstream" applies only to the drainage immediately downstream (e.g., Lower Talarik Creek, but not Iliamna Lake or the Kvichak River). Data
for Dillingham and Ekwok are less detailed and more dated than for other villages, so use patterns may have changed (see Section 13.3.6.2).
d Assumes existing mine infrastructure at the Pebble deposit; hypothetical routing minimizes distance, topographic gradients and stream crossings and assumes water body crossings avoided.
Sources: Fall etal. 1986, Schichnesand Chythlook 1991, Stickman etal. 2003, Fall etal. 2006, Woody and Young 2006, Kriegetal. 2009, ADF&G 2012, Johnson and Blanche 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-15
April 2013
-------
Chapter 13
Cumulative Effects
Table 13 6. Waters, fishes, and subsistence uses potentially affected by a mine at the AUDN/lliamna prospect.
Location
^
o
CO
E
CD
0
.1 o „
s^%
o«|
8 5 3
ro O -1
£
Transportation Corridor to
Naknek6
Transportation Corridor
to Newhalen6
Affected Waters
Watersheds and Named Subwatersheds
Kvichak River
X
X
X
X
X
X
X
X
X
X
X
^
(D
(D
O
(D
£
8
Alagnak River
X
Levelock Creek
X
Jensen Creek
X
Ji
(D
(D
0
0
«
X
J£
(D
0
O
(D
o
X
J£
(D
0
O
2
U
(D
Q_
X
£
2
(0
c
E
CO
X
Lower Talarik Creek
Upper Talarik Creek
Pete Andrews Creek
Newhalen River
Waters3
Jensen Creek
>4 unnamed tributaries
Numerous unnamed lakes and
ponds
Unknown extent of wetlands
Ye How Creek
>70 unnamed tributaries
Numerous unnamed lakes and
ponds
Extensive wetlands
Unnamed stream
3 unnamed ponds
Unknown extent of wetlands
>1 unnamed tributary
Unknown extent of wetlands
Kvichak River
Unknown extent of wetlands
Alagnak River
>1 unnamed tributary
Unknown extent of wetlands
Coffee Creek
>4 unnamed tributaries
Unknown extent of wetlands
Kvichak River (>4 crossings)
>8 unnamed tributaries
Unknown extent of wetlands
Ole Creek
Unknown wetland extent
Pecks Creek
Unknown extent of wetlands
>15 unnamed tributaries
Unknown extent of wetlands
Fish Species'1
Arctic-Alaskan Brook
Lamprey
Longnose Sucker
Northern Pike
Alaska Blackfish
Rainbow Smelt
Humpback Whitefish
Round Whitefish
Unknown Whitefish
Unknown
}
A, J
Unknown
Coho Salmon
X
Chinook Salmon
A-Sp
A-Sp
Sockeye Salmon
A-Sp
A-Sp
Chum Salmon
A-Sp
A-Sp
A-Sp
Pink Salmon
Arctic Char
X
X
Rainbow Trout
Dolly Varden
Lake Trout
Arctic Grayling
1
=
CO
Ninespine Stickleback
Threespine Stickleback
Unknown Stickleback
Coastrange Sculpin
Slimy Sculpin
Unknown Sculpin
Unknown
J
A, J
A, J
Unknown
Unknown
A
A, J
A
A
X
X
A
A-Sp,
}
X
A-Sp
A
A
A
A-Sp
X
A
A
X
A
A
X
A
A
A
A, J
Unknown
A
A
A
A
A
A
X
X
X
A
X
X
X
A
X
A-Sp
A
A-Sp
X
A-Sp
A
X
A-Sp
A
X
X
X
X
A
A
A
A
Subsistence Usec
Village
Dillingham
Ekwok
Igiugig
Kokhanok
Koliganek
Levelock
New
Stuyahok
Ekwok
Levelock
New
Stuyahok
Ekwok
Igiugig
Koliganek
Levelock
Ekwok
Igiugig
Iliamna
Kokhanok
Koliganek
Target Species/Group
Salmon
X
X
X
Other Salmonids
X
X
X
X
.c
lit
Li-
CD
.c
+->
0
X
X
X
JB
CO
CO
uo
3
CD
CO
X
X
X
(0
(D
(/j
X
X
(0
(D
CO
1
CO
X
(D
S
0
X
X
X
X
X
X
Caribou
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Other Mammals
X
X
X
X
X
X
Waterfowl
X
X
X
X
X
X
X
U)
•c
bo
(D
.c
+->
O
X
X
X
X
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-16
April 2013
-------
Chapter 13
Cumulative Effects
Table 13 6. Waters, fishes, and subsistence uses potentially affected by a mine at the AUDN/lliamna prospect.
Location
rtation Corridor
lene (Continued)
M
8 1
E z
•-S
Downstream of Mine(s) and
Transportation Corridor
Affected Waters
Watersheds and Named Subwatersheds
Kvichak River
X
X
X
X
X
Ji
(D
(D
O
(D
£
8
Alagnak River
Levelock Creek
Jensen Creek
X
Ji
(D
(D
0
0
«
J£
(D
0
O
(D
o
^
(D
0
O
&
O
(D
Q_
s
3
(0
c
E
ro
X
X
Lower Talarik Creek
X
Upper Talarik Creek
X
Pete Andrews Creek
X
Newhalen River
X
Waters3
Lower Talarik Creek
>3 unnamed tributaries
Wetlands at least along stream
Upper Talarik Creek
Wetlands at least along streams
Pete Andrews Creek
>1 unnamed tributary
Wetlands at least along stream
Newhalen River
Unknown extent of wetlands
Jensen Creek
Fish Species'1
Arctic-Alaskan Brook
Lamprey
Longnose Sucker
\
Northern Pike
A
A
Alaska Blackfish
Rainbow Smelt
Humpback Whitefish
A
Round Whitefish
A, J
A
A, J
Unknown Whitefish
Coho Salmon
J
A-Sp,
J
X
A
Chinook Salmon
X
A-Sp
A-Sp
Sockeye Salmon
A-Sp,
J
A-Sp,
J
A-Sp
A-Sp,
J
Chum Salmon
A-Sp
A
Pink Salmon
X
Arctic Char
A
X
X
A
Rainbow Trout
A, J
A
A, J
Dolly Varden
A
A
Lake Trout
A
Arctic Grayling
A
A
A, J
1
=
CO
Ninespine Stickleback
A, J
Threespine Stickleback
A, J
Unknown Stickleback
A
A, J
Coastrange Sculpin
Slimy Sculpin
A, J
A, J
Unknown Sculpin
X
Subsistence Usec
Village
Levelock
Newhalen
New
Stuyahok
Nondalton
Port
Alsworth
Levelock
Target Species/Group
Salmon
X
X
X
X
Other Salmonids
X
X
.c
lit
Li-
CD
.c
+->
0
X
X
JB
(0
(0
uo
3
(D
CO
X
(0
(D
(/)
X
(0
(D
CO
1
CO
X
X
(D
8
0
X
X
Caribou
X
X
X
X
Other Mammals
X
X
Waterfowl
X
X
X
X
(/)
•o
bo
(D
.c
+->
O
X
X
Notes:
a Use of the term "tributary" indicates the channel flows into the stream listed above it, while "stream" indicates the receiving water is off site, identified in the columns to the left. For waters downstream of the mine and transportation corridor, fish and subsistence are noted only where different from the waters' prior listing; use of the term "system" indicates the
presence of tributaries in addition to the named water.
b A = adult; Sp = spawning; J = juvenile; x = unknown life stage. "Unknown" indicates an apparent lack of surveys.
c Subsistence uses not separated by watershed. Uses noted only for areas of direct impacts (i.e., mine and/or transportation corridor), if it occurs there; otherwise, noted for areas downstream. "Downstream" applies only to the drainage immediately downstream (e.g., Yellow Creek for the mine, but not the Kvichak River). Data for Dillingham and Ekwokare less
detailed and more dated than for other villages, so use patterns may have changed (see Section 13.3.6.2).
d Hypothetical routing minimizes distance, topographic gradients and stream crossings and assumes water body crossings avoided.
e Hypothetical routing follows that shown in Vne Southwest Alaska Transportation Plan (ADOT 2004), but avoids stream and water body crossings, where possible. Waters and fish for the transportation corridor to Naknek are limited to those in the Kvichak River watershed ( .e., Coffee Creek and north); subsistence use does not include that by villages outside
the watershed.
Sources: Fall etal. 2006, Levelock Village Council 2005, Kriegetal. 2009, ADF&G 2012, Johnson and Blanche 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-17
April 2013
-------
Chapter 13
Cumulative Effects
Table 13 7. Waters, fishes, and subsistence uses potentially affected by a mine at the Humble prospect.
Location
CO
E
o
2
5
•o
SS
Transportation C
Aleknagi
Affected Waters
Watersheds and Named Subwatersheds
Nushagak River
X
X
X
X
X
X
X
X
X
X
X
Nuyakuk River
X
X
Napotoli Creek
X
Klutuk Creek
X
Kokwok River
X
X
X
X
X
Kenakuchuk Creek
X
Nameless Creek
X
Koggilung Creek
lowithia River
(C
1
X
Muklung River
X
Kvichak River
Bear Creek
Waters3
Unnamed stream and >6 tributaries
17 unnamed lakes and ponds
Wetlands at least along main stem
Unnamed stream and >2 tributaries
8 unnamed lakes and ponds
Wetlands at least along streams
Unnamed stream and >3 tributaries
2 unnamed ponds
Unknown wetland extent
Napotoli Creek
>43 unnamed tributaries
48 unnamed lakes and ponds
Wetlands at least along main stem and
primary tributary
Klutuk Creek
>17 unnamed tributaries
Numerous unnamed lakes and ponds
Unknown wetland extent
Kenakuchuk Creek
>3 unnamed tributaries
7 unnamed lakes and ponds
Unknown wetland extent
Unnamed stream
+/- 1.3 km2 (300 ac) of wetlands
Unnamed stream and >17 tributaries
Numerous unnamed lakes and ponds
Kokwok River
Extensive wetlands
Nameless Creek
X3 tributaries
Extensive wetlands
Muklung River
>± tributary
Extensive wetlands
Fish Speciesb
Arctic-Alaskan
Brook Lamprey
J
Longnose Sucker
Northern Pike
Alaska Blackfish
Rainbow Smelt
Round Whitefish
A, J
Unknown Whitefish
Coho Salmon
J
J
J-R
A-Sp, J
A-Sp, J
Chinook Salmon
J
A-Sp, J
A-Sp
Sockeye Salmon
J
A-Sp
Chum Salmon
A-Sp
A-Sp
Pink Salmon
Arctic Char
X
X
Rainbow Trout
Dolly Varden
A, J
A, J
Arctic Grayling
A, J
1
3
CO
Ninespine
Stickleback
A, J
A, J
Threespine
Stickleback
Coastrange Sculpin
A
A, J
A
Slimy Sculpin
A
A, J
A, J
Unknown Sculpin
A, J
A, J
A, J
A, J
Unknown
Unknown
Unknown
J
A-Sp
J
A-Sp
J
A-Sp
A-Sp
A-Sp
A-Sp
A-Sp
A-Sp
X
X
A, J
A, J
A, J
Unknown
A, J
A, J
A, J
Subsistence Use0
Village
Dillingham
Ekwok
Koliganek
NewStuyahok
Aleknagik
Ekwok
NewStuyahok
Target Species/Group
Salmon
Other Salmonids
X
Other Fish
X
-------
Chapter 13
Cumulative Effects
Table 13 7. Waters, fishes, and subsistence uses potentially affected by a mine at the Humble prospect.
Location
Transportation Corridor to Levelock6
Downstream of Mine(s) and
Transportation Corridor
Affected Waters
Watersheds and Named Subwatersheds
Nushagak River
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Nuyakuk River
X
Napotoli Creek
X
Klutuk Creek
X
Kokwok River
X
X
X
X
X
Kenakuchuk Creek
X
Nameless Creek
Koggilung Creek
X
X
lowithia River
X
X
(C
1
Muklung River
Kvichak River
X
X
Bear Creek
X
X
Waters3
Kokwok River
>2 unnamed tributaries
Extensive wetlands
>10 unnamed streams
Unknown wetland extent
Unnamed stream
Unknown wetland extent
Nushagak River (>1 channel)
>2 unnamed tributaries
Unknown wetland extent
Koggilung Creek
X3 unnamed tributaries
Extensive wetlands
Bear Creek
X3 unnamed tributaries
Unknown wetland extent
Unnamed tributary
Napotoli Creek
Klutuk Creek
Kokwok River
Kenakuchuk Creek
Unnamed tributary
Unnamed tributary
Nushagak River
Koggilung Creek
Lowithia River
Bear Creek
Fish Speciesb
Arctic-Alaskan
Brook Lamprey
Longnose Sucker
Northern Pike
Alaska Blackfish
Rainbow Smelt
Round Whitefish
Unknown Whitefish
Unknown
A, J
A
J
J
J
J
A, J
A, J
A, J
J
A, J
A, J
A, J
A, J
A, J
J
X
Coho Salmon
A-Sp
A-Sp, J
J
A-Sp, J
Chinook Salmon
A-Sp
A-Sp
A-Sp, J
Sockeye Salmon
A-Sp
A-Sp
Chum Salmon
A-Sp
A-Sp
Pink Salmon
A-Sp
A-Sp
Unknown
Unknown
A-Sp
A-Sp
A-Sp
J
A-Sp
A-Sp
J
A-Sp
A-Sp
A-Sp
J
A-Sp
Arctic Char
X
X
X
Rainbow Trout
Dolly Varden
Arctic Grayling
1
3
CO
Ninespine
Stickleback
Threespine
Stickleback
Coastrange Sculpin
Slimy Sculpin
Unknown Sculpin
Unknown
A, J
J
J
J
A, J
A, J
J
J
A, J
J
A, J
A, J
A, J
A, J
A, J
A, J
J
A
A, J
J
Unknown
A, J
J
A
A, J
A
A, J
A, J
X
J
J
A, J
J
J
J
J
A-Sp
A-Sp
A, J
A, J
A, J
J
A
A, J
J
A, J
A, J
A, J
Subsistence Use0
Village
Dillingham
Ekwok
Koliganek
Levelock
NewStuyahok
Aleknagik
Dillingham
Koliganek
Newhalen
Portage Creek
Target Species/Group
Salmon
X
X
X
X
X
Other Salmonids
X
X
X
X
X
Other Fish
X
X
X
X
-------
Chapter 13
Cumulative Effects
This page intentionally left blank
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1320 April 2013
-------
Chapter 13
Cumulative Effects
Table 13 8. Streams, water bodies, and wetlands potentially eliminated by additional large scale mines in the Nushagak and Kvichak River
watersheds.
Mine
Pebble South/PEBd
Big Chunk South6
Big Chunk North
Groundhog
AUDN/lliamna
Humble
TOTALS
Claim Block
Size
(km2)
1,380
142
119
317
183
280
2,420
Mine
Footprint3
(km2)
3.87
3.87
9.75
3.87
9.75
3.87
9.75
9.75
9.75
35.0
52.6
Streams
Density
(km/km2)
1.07
1.18
1.45
1.23
1.19
1.07
Length Eliminated11
(km)
4.14
4.56
11.5
5.61
14.1
4.75
12.0
11.6
10.4
41.1
63.8
Water Bodies
Density
(%)
3.14
6.11
4.18
1.24
6.01
0.66
Area Eliminated
(km2)
0.12
0.24
0.60
0.16
0.41
0.05
0.12
0.59
0.06
1.22
1.90
Wetlands
Density0
(%)
18.3
30.5
6.11
83.5
6.11
83.5
4.18
57.2
4.18
57.2
15.8
17.0
15.8
17.0
57.3
75.3
0.66
9.06
Area Eliminated
(km2)
0.71
1.18
0.24
3.23
0.60
8.14
0.16
2.21
0.41
5.58
0.61
0.66
1.54
1.66
5.59
7.34
0.06
0.88
7.37
24.8
Notes:
a Mine footprint areas do not include hydrologic drawdown zones or ancillary facilities; where two values are presented fora mine, the small value represents the footprint assuming the mine uses
existing mill and facilities at the Pebble deposit, whereas the larger value represents the footprint assuming the mine uses its own mill and facilities.
b Length eliminated = mine footprint x stream density.
c For claim blocks with NWI coverage (i.e., Pebble South/PEB, Groundhog, and AUDN/lliamna), minimum density = NWI wetland density and maximum density = (differential between NWI wetland
density and NHD water body density in area of NWI wetland coverage) x NHD water body density for entire claim block. For claim blocks with no NWI coverage, minimum density = NHD water body
density and maximum density = (maximum differential between NWI wetland density and NHD water body density) x NHD water body density.
d Claim block size for entire PLP/NDM block; water body density includes portion of Iliamna Lake.
e Water body density includes portions of Nikabuna Lakes.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-21
April 2013
-------
Chapter 13 Cumulative Effects
13.2.2 Big Chunk South
13.2.2.1 Description
The Big Chunk South prospect may be of the same geologic origin as the Pebble deposit. The claim block
abuts the north edge of the PLP/NDM claim block approximately 20 km north of the potential Pebble
mill site (AREA 2012, Big Chunk Corp. 2012). It is approximately 24 km northwest of the village of
Nondalton.
Based on its proximity to the Pebble deposit, a future mine at Big Chunk South may use the TSFs, mill,
and other facilities built for potential mining at the Pebble deposit, under a joint venture or other
agreement with PLP/NDM. In late 2012, Big Chunk partner Liberty Star settled debt and terminated
joint-venture negotiations with NDM (Liberty Star 2012b). For the purposes of this assessment, we
consider Big Chunk South under two scenarios: one in which it shares some facilities built for potential
mining of the Pebble deposit and the other in which it operates as a fully separate, stand-alone mine,
with no shared facilities other than the transportation corridor connecting the Pebble deposit site to
Cook Inlet (Table 13-8), referred to in this cumulative effects discussion as the assessment corridor. A
mine at Big Chunk South presumably would connect to the roads and pipelines of such a corridor
somewhere near their western termini, currently estimated to be approximately 14 km south of the Big
Chunk South claim block.
13.2.2.2 Potentially Affected Waters, Fish, and Subsistence Uses
Table 13-3 summarizes information on the waters, fish, and subsistence uses potentially affected by a
mine at Big Chunk South prospect. The 142-km2 Big Chunk South claim block is entirely within the
drainage of the Chulitna River, which flows into Lake Clark National Park and Preserve and then into the
lake itself, at Turner Bay, 40 km northeast of the block. A segment of the strongly meandering river,
including the Nikabuna Lakes system, runs along the entire 27-km north boundary of the block. Current
National Wetlands Inventory (NWI) mapping does not include the Big Chunk South claim block;
however, based on aerial photography and U.S. Geological Survey (USGS) topographic mapping,
extensive wetlands appear to be associated with the river, extending upstream along several tributaries.
Stream density at Big Chunk South (1.2 km/km2) is higher than in most of the other claim blocks,
including PLP/NDM, and the majority of streams in the block are headwaters tributaries. Water body
density—more than 6% of the block—is the highest of all the potential mine sites, due in part to the
Nikabuna Lakes. The relatively flat valley of the Chulitna River occupies most of the claim block;
elevations range from approximately 90m along the river to 320 m in the more rugged south-central
part of the block.
To connect to Cook Inlet, a road and pipelines from the Big Chunk South claim block would likely ascend
the valley of an unnamed Chulitna River tributary and then cross the headwaters of the North Fork
Koktuli River to join the end of the assessment corridor, approximately 14 km south of the block.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~22 April 2013
-------
Chapter 13 Cumulative Effects
Very few known fish surveys have been conducted in either the Big Chunk South claim block or the
middle or upper Chulitna River to date (Table 13-3).Fall et al. (2006) describe the Chulitna River valley
as "one of the most important" subsistence areas for Nondalton. Among the salmonids included as
assessment endpoints, only Dolly Varden are found in the claim block. One village subsistence fishes and
hunts for bear, moose, caribou, other mammals, and waterfowl in the claim block; four other villages
hunt for caribou; and one other village hunts for moose. As an average across the claim block, stream
loss would range from 4.6 to 12 km, and wetlands eliminated would range from 0.24 to 8.1 km2,
depending on whether or not a TSF is constructed on site (Table 13-8).
13.2.3 Big Chunk North
13.2.3.1 Description
The Big Chunk North prospect is approximately 11 km northwest of Big Chunk South prospect, 34 km
northwest of the Pebble deposit, approximately 48 km northwest of Nondalton and 96 km northeast of
Koliganek. A mine at Big Chunk North would potentially use TSFs, mill, and other facilities at either Big
Chunk South or Pebble. We consider the impacts of both a mine with shared facilities and one without
(Table 13-8). In both cases, we anticipate that a mine at Big Chunk North would connect to the
assessment corridor.
13.2.3.2 Potentially Affected Waters, Fish, and Subsistence Uses
Table 13-4 summarizes information on the waters, fish, and subsistence uses potentially affected by a
mine at Big Chunk North. Like the Pebble deposit, the 119-km2 Big Chunk North claim block straddles
the drainage divide between the Nushagak and Kvichak River watersheds. Northwest of Buck Mountain,
in the northwest corner, the block contains ponds and streams that are part of the headwaters of Reefer
Creek, a tributary of the Mulchatna River, as well as the uppermost reaches of the North Fork Swan
River in the Koktuli River watershed. Nearly 90% of the block, though, is a high-density network of
streams, ponds, and wetlands that form the headwaters of the Chulitna River, which rises immediately
north of the block. The system departs the block along its south boundary. National Hydrography
Dataset (NHD) stream density in the Big Chunk North claim block (1.45 km/km2) is higher than in any of
the other areas of potential mines we consider, and more than one-third higher than in the PLP/NDM
claim block (USGS 2012). Water body density (>4 %) is also fairly high compared to other mine sites, but
lower, than at Big Chunk South and AUDN/Iliamna.
A transportation corridor to service a mine in the Big Chunk North claim block would presumably follow
the Chulitna River valley south and eastward to the Big Chunk South block, where it would link up to the
corridor described in Section 13.2.2.1.
To date, no known fish surveys have been conducted in the Big Chunk North claim block or along its
potential transportation corridor, nor any freshwater fish surveys anywhere in Reefer Creek. Table 13-4
summarizes what little information is available on fish presence in the waters potentially affected by a
mine in this claim block. As in the south block, we do not have data on population sizes. At least one of
the stream systems—the North Fork Swan River—has an abundance of beaver dams, indicating it may
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~23 April 2013
-------
Chapter 13 Cumulative Effects
provide important overwintering and rearing habitat (Johnson and Blanche 2012). Four villages hunt in
the claim block but no subsistence fishing is reported (Table 13-4). Across the claim block, stream loss
would range from 5.6 to 14 km, and wetland loss would range from 0.16 to 5.6 km2, depending on
whether a TSF is constructed on site (Table 13-8).
13.2.4 Groundhog
13.2.4.1 Description
The 317-km2 Groundhog claim block abuts the northeast corner of the PLP/NDM block, approximately
10 km west of Nondalton and 20 km north-north west of Iliamna. At present there does not appear to be
a relationship between the Groundhog claim holders and PLP/NDM. Nevertheless, given its proximity to
the potential Pebble facility (approximately 6 km), we consider it both as a separate, stand-alone mine
and as one that shares some facilities associated with potential mine development at the Pebble deposit
(Table 13-8), including the assessment corridor. The current route for such a corridor is approximately
4 km south of the claim block and 13 km from the recent target area for exploration drilling at
Groundhog (AREA 2011); it would presumably follow one of the Upper Talarik Creek tributaries down
to the road.
13.2.4.2 Potentially Affected Waters, Fish, and Subsistence Uses
Table 13-5 summarizes information on the waters, fish, and subsistence uses potentially affected by a
mine at Groundhog. Similar to Big Chunk, nearly 90% of the Groundhog prospect lies in the drainage of
the Chulitna River, which passes through the narrow, central part of the block. South of Groundhog
Mountain, the claim includes a number of headwater tributaries to Upper Talarik Creek and the North
Fork Koktuli River; a very small portion of the block—less than 1 km2 in the southeast corner—drains to
tributaries of the Newhalen River, which connects Sixmile Lake, below Lake Clark, to Iliamna Lake. A
road from the claim block to the assessment corridor would presumably follow one of the Upper Talarik
Creek tributaries.
Based on NHD mapping, stream density in the Groundhog block (1.23 km/km2) is the second highest of
those we consider; water body density (>1%) is much lower than at all other sites except Humble
(Table 13-8). The NWI maps extensive wetlands along the Chulitna River and Rock Creek, as well as
along lower Groundhog Creek, the drainage in which the most recent exploration of the prospect has
occurred (USFWS 2012, AREA 2011).
Surveys offish use in waters of the Chulitna River drainage have been rather limited to date (Table 13-
5). There is more information for the headwater tributaries in the Upper Talarik Creek and North Fork
Koktuli River watersheds in the southern part of the claim block, which both support salmonids. The
Upper Talarik Creek tributary system originates in the same series of lakes and ponds as Groundhog
Creek, in the Chulitna River watershed, at an elevation of approximately 460 m. Two Pacific salmon
species and Dolly Varden have been reported in the claim block and one village hunts there (Table 13-5).
On average across the claim block, stream loss would range from 4.8 to 12 km, and wetland loss would
range from 15.8 to 17.0 km2, depending on whether or not a TSF is constructed on site (Table 13-5).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~24 April 2013
-------
Chapter 13 Cumulative Effects
13.2.5 AUDN/lliamna
13.2.5.1 Description
The AUDN/lliamna prospect is approximately 3 5 km west of Iliamna Lake and 90 km south west of the
Pebble deposit. It is in the midst of the native villages of Levelock, New Stuyahok, Ekwok, and Igiugig
(Figure 13-1). The bulk of the claims associated with this prospect were newly established in 2012 by
Millrock Alaska, the same company that owns the Humble claims. Millrock began exploration of the
prospect in 2012 and describes it as being part of the porphyry copper-gold belt that includes the
Pebble deposit (Millrock Resources 2011). Their AUDN claims surround others staked earlier and still
held by a partnership that includes TNR Gold Corp., the owner of the Shotgun claims; TNR Gold Corp.
calls their project "Iliamna" (TNR Gold Corp. 2012b).
In light of the higher development costs associated with this prospect's distance from other potential
mines, we assume, for purposes of this assessment, that a future mine at this prospect would require a
joint venture involving both claim blocks and would be self-contained (i.e., it would not share any
facilities with other mines) (Table 13-8). The closest potential port site would be on the lower, tidally
influenced reach of the Kvichak River, at or near Levelock. Naknek, approximately 56 km south of the
prospect, already operates as a port and could be an alternative location, should establishing a port on
the Kvichak River prove infeasible.
Development of a mine at AUDN/lliamna could trigger the involvement of the Alaska Department of
Transportation and Public Facilities (ADOT) in building the Levelock-to-Naknek portion of the Cook
Inlet-to-Bristol Bay (CIBB) and Dillingham/Bristol Bay (DBB) corridors described in the Southwest
Alaska Transportation Plan (SWATP) (ADOT 2004). The SWATP anticipates that such construction
would not occur within its 20-year planning period; however, it noteds that "changing circumstances,"
such as "discovery of high-value resource that could potentially be accessed economically through
development of [such roads could] trigger consideration of an earlier implementation" schedule. A mine
at AUDN/lliamna could be such a trigger, and ADOT involvement in road construction could defray mine
development costs. ADOT is currently investigating such a project, the Ambler Mining District Access, in
northwestern Alaska, as part of the state's "Roads to Resources" program (ADOT 2011a). For similar
reasons, Cook Inlet could be another alternative location for a port, using the Levelock-to-Newhalen
portion of the CIBB and DBB corridors; shipping distances to Canada and the lower 48 states appear to
be shorter from Cook Inlet than from Naknek. For the purposes of this assessment, we consider all three
possible transportation corridors.
13.2.5.2 Potentially Affected Waters, Fish, and Subsistence Uses
Table 13-6 summarizes information on the waters, fish, and subsistence uses potentially affected by a
mine at AUDN/lliamna. The 183 km2 AUDN/lliamna claim block occupies a low, relatively flat area of the
Kvichak River watershed, including a portion of the glacial outwash plain at the western edge of the end
moraine that originally formed Iliamna Lake (Detterman and Reed 1973). The Yellow Creek system
drains approximately 90% of the claim block and flows into the Kvichak River approximately 15 km
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~25 April 2013
-------
Chapter 13 Cumulative Effects
upstream of Levelock. Jensen Creek drains the southern 11 km2 the block and enters the Kvichak River
10 km above Levelock; some maps incorrectly identify it as Yellow Creek (Levelock Village Council
2005). Both stream systems are strongly meandering. The remainder of the claim flows to a third
tributary system that enters the Kvichak River above Yellow Creek. Residents of Levelock use the
AUDN/Iliamna area most for subsistence, but other villages also fish or hunt in the area or along the
potential transportation corridors for the mine (Table 13-6).
Stream density at AUDN/Iliamna is higher than at PLP/NDM and Humble, but lower than at most of the
other blocks. However, AUDN/Iliamna has more water bodies than any of the other claim blocks except
PLP/NDM, and despite its lack of any large lakes (the largest is less than 0.5 km2), water body density at
AUDN/Iliamna (>6% of the block) is nearly twice that of the PLP/NDM claim block (which encompass
4.9 km2 of Iliamna Lake) and almost the same as at Big Chunk South (which includes 2.7 km2 of the
Nikabuna Lakes). NWI mapping, which covers almost 70% of the block, also shows extensive wetlands
encompassing approximately 57% of the block. Four species of Pacific salmon are cataloged in
numerous streams on the claim block and seven villages hunt there for a variety of mammals and birds
(Table 13-6). Across the claim block, stream loss would average 11.6 km, and wetland loss would range
from 5.6 to 7.3 km2 (Table 13-8).
The short transportation corridor from AUDN/Iliamna to Levelock would potentially involve only one
stream crossing, of an unnamed tributary to Levelock Creek. To connect onto Naknek or Cook Inlet (via
the conceptual CIBB route and the assessment corridor), the road would first have to cross the Kvichak
River at or near Levelock. To reach Naknek, it would also have to cross the Alagnak River and Coffee
Creek, as well as several tributaries. Instead of crossing the Alagnak River, the route to the assessment
corridor would follow it briefly, before crossing Ole and Pecks Creeks, as well as crossing the Kvichak
River at least one more time (at or near Igiugig). From there, it would traverse along the west and north
shores of Iliamna Lake, crossing Lower and Upper Talarik Creeks, at least 16 other streams, and the
Newhalen River, before joining the assessment corridor near Iliamna.
13.2.6 Humble
13.2.6.1 Description
Accounts characterize the Humble prospect as geologically and geochemically similar to the Pebble
deposit (Szumigala etal. 2011, Millrock Resources 2012b). It is approximately 135 km south west of the
Pebble deposit, 60 km north west of AUDN/Iliamna, and 20 to 30 km west to northwest of the villages of
Koliganek, New Stuyahok, and Ekwok. Wood-Tikchik State Park, the largest state park in the United
States, is approximately 13 km northwest and 29 km west of the claim block.
Due to Humble's distance from the Pebble deposit and the AUDN/Iliamna prospect, we do not anticipate
sharing of facilities between the mines (Table 13-8). The Dillingham-Aleknagik Road (46 km to the
southwest) is the closest link to existing road infrastructure and port facilities, the latter of which are
another 40 km further at Dillingham. ADOT is currently pursuing federal permits for construction of a
bridge over the Wood River, which presumably would be the southernmost link in a transportation
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~26 April 2013
-------
Chapter 13 Cumulative Effects
corridor from Humble to Aleknagik (ADOT 2011b, USCG 2012). Alternatively, development of a mine at
Humble could trigger the involvement of ADOT in building all or portions of the CIBB and DBB corridors,
which could serve to connect Humble to a port on Cook Inlet. A road from the Humble claim block to the
conceptual DBB route presented in the SWATP would be shorter than a mine road connecting to
Aleknagik (ADOT 2004).
13.2.6.2 Potentially Affected Waters, Fish, and Subsistence Uses
Table 13-7 summarizes information on the waters, fish, and subsistence uses potentially affected by a
mine at Humble. The 280-km2 claim block's east- and south-flowing streams are entirely within the
Nushagak River watershed (Figure 13-1), and the mainstem channels are all strongly meandering. More
than 40% of the block drains to Napotoli Creek, which enters the Nushagak River approximately 14 km
upstream of Koliganek. The northernmost 36 km2 of the block flow to one unnamed tributary of the
Nushagak River and two unnamed tributaries of the Nuyakuk River, which connects Tikchik Lake, in
Wood-Tikchik State Park, to the Nushagak River.
In the south, nearly one-quarter of the block is in the Klutuk Creek watershed, which flows into the
Nushagak River immediately downstream of Ekwok. The remainder of the block—approximately 20%—
drains to Kenakuchuk Creek and other tributaries of the Kokwok River, which enters the Nushagak
River 8 km downstream of Ekwok. Stream density in the claim block is approximately the same as at the
Pebble deposit (1.1 km/km2), which is lower than at all of the other sites (Table 13-8).
NHD water body density at Humble is less than 25% of that at PLP/NDM and only slightly more than
half of that at Groundhog, the next most similar prospect (USGS 2012). A long band of ponds occupies
the divide between Klutuk Creek and the Kokwok River tributaries. Current NWI mapping does not
extend to the Humble block. At a minimum, based on aerial photography and USGS topographic
mapping, there appear to be wetlands along most of the larger stream corridors (e.g., Naptoli, Klutuk,
and Kenakuchuk Creeks) and in an approximately 4-km2 area in the southwest corner of the block. Four
Pacific salmon species and Dolly Varden are present in many streams in the claim block, and four
villages huntfor a variety of mammals (Table 13-7). On average across the claim block, stream loss
would be 10.4 km, and wetland loss would range from 0.06 to 0.88 km2 (Table 13-8).
Minimizing distance and topographic gradient, a potential route for a transportation corridor from the
Humble claim block to Aleknagik would cross the Kokwok River, just downstream of Kenakuchuk Creek,
ascend the valley of Nameless Creek (a Kokwok tributary), skirt Wood-Tikchik State Park, and then
follow the Muklung River (a tributary of the Wood River), until turning north of Marsh Mountain for
Aleknagik. The overland route to Cook Inlet would presumably cross the Kokwok River further
downstream from the route to Aleknagik, connecting to the Dillingham/Bristol Bay corridor near the
lowithla River, another Nushagak River tributary. The SWATP's conceptual corridor would cross the
Nushagak River downstream of Ekwok. After crossing Koggilung Creek, the route would pass into the
Kvichak River watershed, reaching the river at or near Levelock. Section 13.2.5.2 describes the route
from Levelock to the assessment corridor.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~27 April 2013
-------
Chapter 13 Cumulative Effects
Despite its relatively low stream density, a large number offish-bearing streams traverse the Humble
claim block (Table 13-7). Information on local population sizes is not available. The Napotoli and Klutuk
Creek systems contain numerous beaver complexes, as well as frequent seeps and springs, which may
provide important overwintering habitat for juvenile salmonids (Johnson and Blanche 2012: nomination
forms 04-158, 04 160, 04-171, 04-890, 06 753, 06-754,11-369-11-372,11-381,11-382, and 11-384-
11-386). Residents of Aleknagik, Dillingham, Ekwok, Koliganek, and New Stuyahok have historically
used the claim block, potential transportation corridor west of Levelock, and/or downstream areas for
subsistence fishing, hunting, and/or gathering (Table 13-7). Information for Aleknagik, Dillingham, and
Ekwok is less detailed than for the other villages. The reports for Dillingham and Ekwok are more than
20 years old, so their subsistence-use patterns may have changed, particularly for caribou hunting, since
that species' population and migration routes have shifted (Brna and Verbrugge 2013).
13.2.7 Impacts of Multiple Mines
In the preceding sections, we examined the waters, fish, and other subsistence resources that could be
affected by a typical-sized mine footprint at six prospects that could proceed after initial development of
a large mine at the Pebble deposit. For the purposes of this assessment, we consider the cumulative
impacts of these six mines—that is, potential effects on assessment endpoints resulting from the
establishment of six additional mines and their associated transportation corridors. These influences
would likely accumulate over time and space, potentially having widespread and extensive effects on the
region's populations offish, wildlife, and human residents.
13.2.7.1 Habitat Eliminated
Table 13-8 summarizes direct losses of aquatic habitat to the footprints of the additional six potential
mines. Total stream length eliminated by these footprints would range from 41 to 64 km, and total water
body and wetland area lost would range from 1.2 to 1.9 km2 and 7.4 to 25 km2, respectively. Conversion
of these areas to mine footprints would result in extensive losses of floodplain, riparian habitat, and
wetland areas. Waters on these claim blocks include the Chulitna River and Rock, Jensen, Yellow,
Napotoli, Klutuk and Kenakuchuk Creeks, as well as, over 250 unnamed tributaries and over 50
unnamed lakes and ponds. Although not all support salmon, many do. Elimination of substantial habitat
across the watersheds would contribute to diminishing the biological complexity of salmon stocks and
lead to the portfolio effect discussed in Sections 5.2.4 and 13.4.1.
13.2.7.2 Flow Alteration
Water Withdrawal and Retention
Routine operations at additional mines would also likely degrade or destroy downstream habitat due to
water withdrawal and management of precipitation at the mine facilities (Chapters 6 and 7). Mines
require water for mill operation and transport of tailings and concentrate. The required withdrawal and
retention of surface water and groundwater would effectively reduce the size of the watershed
contributing to downstream flow. Mine pit dewatering would further reduce the contributing watershed
by creating a zone of depression (Section 6.2.2). Streams, wetlands, and ponds within this zone that
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~28 April 2013
-------
Chapter 13 Cumulative Effects
receive water through groundwater would dry up, discontinuing any contributions to downstream
waters. Groundwater flow down the valley would also be disrupted, potentially affecting spawning and
wintering habitat downstream (Chapter 7).
Increased Effective Impervious Surface
Additional mines and potential transportation corridors at each of the prospects would convert
additional land to impervious surface, which could increase total effective impervious surface area
above 10% in several subwatersheds in the region (e.g., Groundhog, Jensen, and Napotoli Creeks and the
Nikabuna Lakes). Mine and road operators and village governments may be able to limit or even
eliminate downstream damage from impervious surface. If operators could moderate discharges to
simulate natural flows and prevent higher than natural stormflows, downstream channels could be
maintained in relatively stable condition. Achieving this goal would require sufficient on-site water
retention capacities to allow a slow meting out of storm-generated flows, rather than contemporaneous
discharge of all surface runoff generated by storm events.
Road Crossings
In addition to increasing runoff rates due to their impervious surfaces, the transportation corridors
associated with additional mines and other induced development would increase the likelihood of flow
regime changes, including channel modification, wherever they crossed streams, wetlands, and other
water bodies (Chapter 10, Appendix G). Such alterations frequently affect salmonids and other fish by
blocking access to habitat and/or physically degrading the habitat itself. Cumulatively, these additional
transportation corridors would add at least five river crossings (South and North Fork Koktuli, Chulitna,
Kokwok, and Nushagak or Muklung) and a minimum of 27 smaller stream crossings (including at least
one of Upper Talarik Creek) (Tables 13-2 through 13-7). Overland access to AUDN/Iliamna from a port
at Naknek or Cook Inlet (rather than near Levelock) would add at least one and possibly as many as four
crossings of the Kvichak River, at least one of another river (Alagnak or Newhalen) and a minimum of six
at other streams. These crossings would substantially increase the potential for hydraulic alterations
resulting in upstream and downstream habitat degradation (Chapter 10).
13.2.7.3 Water Quality Degradation
Chapters 8, 9,10 and 11 discuss the water quality impacts of a single mine at the Pebble site from water
treatment and discharge, tailings storage failure, road construction and operation, and pipeline spills.
Additional mines and transportation corridors would have these same potential impacts.
Routine Operations
Routine operations at six additional mines would result in 35 to 53 km2 of ground disturbance
(Table 13-8) depending on whether or not TSFs are built at each of the mines. Rivers and streams in
which water quality could be affected include the Chulitna River and Rock, Jensen, Yellow, Napotoli,
Klutuk, and Kenakuchuk Creeks (Tables 13-2 through 13-7). The transportation corridors for these
mines would potentially span the width of the Nushagak and Kvichak River watersheds from Cook Inlet
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 13~29 April 2013
-------
Chapter 13 Cumulative Effects
to Dillingham and potentially affect the Alagnak, Kvichak, Kokwok, and Nushagak Rivers and Coffee and
Bear Creeks. Salmon are reported in streams in the areas of all of these prospects except Big Chunk
North and Big Chunk South.
Accidents and Failures
Chapters 8 through 11 describe the probabilities and consequences of a variety of accidents and failures
under the mine scenarios. Although the probability of such failures at an individual facility at any given
time is low, the cumulative probability of failures increases as the number of facilities increases. For
example, historical data suggest a 100% cumulative probability of failure in one of the four pipelines
over the life of the Pebble 2.0 scenario (Section 11.1). Additional pipelines at additional mines would
increase the overall probability of failure at some location in the Nushagak and Kvichak River
watersheds each year. Similarly, the chances of a road failure with significant consequences for
downstream waters would be substantial with development of a single mine, and would increase as
length of road in the watersheds increases (Section 10.3.2).
Although failure of any single TSF dam would be a low-probability event, it could be catastrophically
damaging to fisheries in the receiving waters if it were to occur. The presence of multiple large-scale
mines would increase the probability of at least one TSF dam failure occurring in the watersheds over
the mine lifetimes and post-closure periods roughly in proportion to the total number of dams, and thus
increase the chance of long-term adverse downstream effects. TSF dam failures at additional mines
would likely be similar in nature to the Pebble 0.25 scenario failure described in Section 9.3. The
magnitude of adverse impacts would vary with location, TSF size, and the degree of failure. Salmon-
bearing waters into which slurry from TSF dam failures at additional mine sites could flow would
include the North Fork and mainstem Koktuli Rivers; the Kvichak, Nushagak, Kokwok, and Chulitna
Rivers; Upper Talarik, Rock, Napotoli, Klutuk, and Yellow Creeks; Iliamna, Nikabuna, and Long Lakes;
and Lake Clark.
Another potential source of pollutant discharges results from human errors in characterizing the mining
environment (e.g., its geochemistry or hydrology) and/or anticipating long-term needs for pollutant
control (Box 13-2). Human error, as well as mechanical failure, can also result in water bypassing a
treatment system. Similar unintended failures in human judgment could result in unanticipated
discharge of pollutants from mine sites in the Nushagak and Kvichak River watersheds, particularly as
the number of additional sites increases. The cumulative effect of such incidents in the Nushagak and
Kvichak River watersheds would likely be a steady decline in productivity in these systems as the
affected reaches grow in length and number.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1330 April 2013
-------
Chapter 13 Cumulative Effects
BOX 13 2. EXAMPLES OF MINE CHARACTERIZATION ERRORS
Errors in mine site characterization or anticipation of long-term needs for pollutant control can contribute to
mine-related pollutant discharges. Examples at existing mines include the following incidents (see Figure 4-1
for mine locations).
• At the Red Dog Mine in northwest Alaska, treatment of waste rock runoff for metals elevated dissolved
solids in the runoff to the point that it had to be directed to the TSF, rather than discharged.
Compounding this problem, failure to implement planned surface-water diversions early in mine
development resulted in unpredicted rapid filling of the TSF. Unscheduled discharges from the TSF were
necessary to prevent overtopping of its dam.
• At the Greens Creek Mine in southeast Alaska, plans included reclamation of the dry stack tailings
storage facility to prevent acid drainage. However, mine life exceeded the anticipated timeframe, delaying
reclamation and resulting in acid drainage from the tailings. Anew understanding of the geochemistry at
the site indicates that perpetual water treatment will be necessary even after reclamation, a substantial
change from the original design. Moreover, mine operators have discovered that local wetland chemistry
resulted in a treatment system that redissolves metals before discharge, requiring construction of a new
water treatment facility to address this unanticipated source of pollution.
• Human error resulted in an uncontrolled discharge from a TSF at the Nixon Fork Mine, in interior Alaska,
in January 2012 (see Box 6-2 for a description of events). This unpermitted discharge does not appear to
have reached nearby streams at the time of this writing and may have caused no environmental harm.
As discussed in Section 14.1.2.5, common mode failures—that is, multiple failures with a common
cause—could result from incidents such as an earthquakes or severe storms. The potential for common
mode failures would be compounded if a mining district were created, increasing the chance that a
single severe event could result in multiple failures and adversely affect multiple salmon-bearing waters
at one time.
The passage of time would be another component influencing cumulative impacts of large-scale mining
in these watersheds. Although time could contribute to the recovery of streams affected by accidents or
failures, it could also increase the likelihood that a particular accident or failure would occur. We assume
that post-closure site management considerations (Section 6.3) would generally apply to each additional
mine, although the specifics would be based on design and operational assumptions of each mine and
thus would differ from site to site. Closure at each mine would typically require hundreds to thousands
of years of monitoring, maintenance, and treatment of any water flowing off site. Given the magnitude of
these timeframes, we would expect multiple and more frequent system failures in future years. In light
of the relatively ephemeral nature of human institutions over these timeframes, we would expect that
monitoring, maintenance, and treatment would eventually cease, leading to increased release of
contaminated waters downstream.
13.3 Cumulative Impacts from Induced Development
Induced development, or development resulting from the introduction of industry, roads, and
infrastructure associated with a specific activity to a region, is an iterative phenomenon. Opportunities
for employment at mines or in mine-related services would contribute to growth in nearby communities
and would increase demand for housing, community infrastructure, and amenities such as recreation in
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1331 April 2013
-------
Chapter 13 Cumulative Effects
and around those localities. Independent of growth due to mine-related employment, improved
accessibility due to road and port construction would reduce the cost of shipping fuel and freight to
areas near such infrastructure. Reduced shipping costs, in turn, would make construction, business
operation, recreation/tourism, and general cost of living more affordable, which would facilitate
increased growth.
Induced development following the advent of large scale mining would undoubtedly bring very welcome
economic opportunities to the region. The potential road systems described in Section 13.3 would be a
major driver of induced development, potentially extending completely across the Nushagak and
Kvichak River watersheds (approximately 250 miles). Currently, access to sites in the Nushagak and
Kvichak River watersheds is by air, boat, snow machine, or foot. Thus, access throughout the watersheds
is typically facilitated and regulated by the tourism industry. Transportation corridors associated with
large-scale mines likely would increase vehicle access throughout the two watersheds, thereby
increasing unregulated access, both lawful and unlawful, to currently remote sites. Access to all-terrain-
vehicles (ATVs) and snow machines would be greatly enhanced by a road system, and areas in the two
watersheds that are essentially never accessed by humans would become available. This increased
access would extend fishing and hunting pressure and make areas along any transportation corridor
more susceptible to trespassing, poaching, and illegal dumping (ADOT 2001).
13.4 Potential Effects on Assessment Endpoints
13.4.1 Fishes
Based on information regarding the general locations of likely development and the fish species present
in the waters in these areas (Tables 13-2 through 13-7), we can estimate some of the potential impacts
on fish, such as the extent of direct habitat losses to typical mine footprints (41 to 64 km of streams), the
approximate number of streams crossed by transportation corridors (5 rivers and 27 streams), and the
potential for additional habitat loss or degradation that would result from accidents or failures
(Section 13.3.7.3). By identifying general areas where development is reasonably foreseeable (Tables
13-2 through 13-7 and 13-9), we can also, to some extent, qualify those losses by habitat type (e.g.,
headwater streams) and fish species.
In addition to the effects associated with the sheer quantity of lost or degraded habitat, the impacts of
large-scale mining could cumulatively threaten biological complexity of the Nushagak-Kvichak salmonid
stock complex (Section 5.2.4). Impacts on genetically distinct populations of salmon across the
watersheds can reduce biological complexity (the portfolio effect) and lead to salmon population
declines. As described in this chapter, reasonably foreseeable development during an 80-year timeframe
could span the watersheds, encompassing many geographically and hydrologically distinct waters. For
anadromous fish, the potentially affected waters would include at least 10 different rivers and 20 feeder
stream systems. As described by Schindler et al. (2010), each river stock includes tens to hundreds of
locally adapted populations distributed among tributaries and lakes. Given the extent of stream loss and
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±332 April 2013
-------
Chapter 13 Cumulative Effects
habitat degradation, it is reasonable to assume that loss of genetic and life-history diversity would occur
as part of the development of multiple large-scale mines. Even non-anadromous fish would be subject to
genetic diversity losses wherever development blocked movement to different spawning, rearing, and
overwintering habitats within a stream and/or stream-lake system, and thus isolated portions of the
population from each other (Appendix B) (Charles et al. 2000).
Although we can estimate potential habitat losses and anticipate resulting losses of population diversity,
we cannot quantify specific fish losses or their significance in terms of overall populations, given the
current lack of data on abundance and uncertainties about the precise locations and magnitude of future
developments.
13.4.2 Wildlife and Alaska Native Cultures
As the extent of development in the region increases, so would development-related effects on wildlife
and Alaska Native culture. The six additional mines would affect a wide range of wildlife, including both
resident and highly migratory species (Tables 13-2 through 13-7). As with fish, data are insufficient to
project impacts in terms of populations.
As for Alaska Natives, 13 of the 14 villages in the watersheds would experience some impact on
traditional subsistence use areas from additional development (Figure 13-3). Levelock and Igiugig
would encounter the most impacts, insofar as the additional development would have direct and/or
indirect impacts on all categories of their subsistence resources due primarily to proximity of the
potential AUDN/Iliamna mine and a transportation corridor from that mine to Newhalen. Iliamna,
Koliganek, Newhalen, New Stuyahok, and Nondalton would also have a large number of subsistence
resource categories affected.
To the extent that cumulative development from large-scale mining would reduce the areas or fish and
wildlife populations available for subsistence activities, development of a mining district would have
broader cultural impacts related to the diminishing role of subsistence in village life (Chapter 12,
Appendix D). Employment opportunities and reduced cost of living resulting from large-scale mining
could address some current concerns associated with subsistence, chiefly the cost of fuel needed to
access these resources. At the same time, loss of subsistence areas or populations likely would be
coupled with increased westernization resulting from increased access, development, tourism, and
prosperity to erode the current subsistence cultures at least to some extent.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . o 33 April 2013
-------
Chapter 13
Cumulative Effects
Figure 13 3. Location of claim blocks in relation to subsistence use intensity for salmon, other
fishes, wildlife, and waterfowl in the Nushagak and Kvichak River watersheds. See Box 5 1 for
discussion of subsistence use methodology.
Cook Inlet
Clark's Point
South Naknek
Bristol Say
N
A
Approximate Pebble Deposit Location
Watershed Boundary
Existing Roads
Nonsurveyed Towns and Villages
• Surveyed Towns and Villages
Active Mining Claims
Subsistence Use Intensity
High
Low
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
13-34
April 2013
-------
Chapter 13 Cumulative Effects
13.5 Summary
The industrial complex and transportation corridor associated with potential mine development at the
Pebble deposit would constitute the second largest human population center in the Bristol Bay region
and its longest road system. Figure 13-2 illustrates how cumulative impacts from that first mine,
multiple subsequent mines, and induced development could result from large scale mining in the region.
Infrastructure in the form of roads, airports, ports, and utilities is virtually absent from the mineralized
areas of the Nushagak and Kvichak River watersheds. The transportation corridor, deep-water port,
power generation facility, and other infrastructure that would be required for mining the Pebble deposit
likely would increase the economic feasibility of developing and operating other, smaller mines. Over
the approximately 25-year life of the Pebble 2.0 scenario mine, and certainly over the approximately 78-
year life of the Pebble 6.5 scenario mine, it is reasonably foreseeable that a number of prospects recently
under active exploration (Pebble South/PEB, Big Chunk South, Big Chunk North, Groundhog,
AUDN/Iliamna, and Humble) could be developed. Mines at these sites would cause their own direct
impacts, which would accumulate over a much greater area of the two watersheds. The large spatial
extent of impacts would increase the number of distinct salmon populations affected, which could
cumulatively threaten the biological complexity of the Nushagak-Kvichak salmon stocks and the
portfolio effect, potentially contributing to salmon population declines. Each of the mines would need
road access to a port facility and airport. Some of the additional mines could use the assessment
corridor, with only relatively short extensions, but more distant sites such as AUDN/Iliamna and
Humble, located 90 km and 135 km southwest of Pebble, respectively, would require development of
extensive additional road systems.
It is reasonably foreseeable that infrastructure from large scale mining in the watershed, particularly the
transportation corridors could induce further development in the region. Existing communities, the
tourism industry, and the recreational housing market could benefit if large-scale mining expanded
through the watersheds. Unmanaged access to currently roadless wilderness areas also could expand.
The improved access would increase hunting and fishing pressure, as well as competition with existing
subsistence users; increase damage from off-road vehicle, boat, and foot traffic to currently inaccessible
areas; facilitate poaching, dumping, trespassing, and other illegal activities; and lead to scattered
development in the watersheds.
The mines and road systems described herein are not certain, but are part of state planning documents.
A large-scale mine could easily be the trigger that starts this pattern of development in motion.
Development in the Pacific Northwest has followed this pattern for over 100 years and has led to the
near complete loss of wild salmon. Even in the coastal population centers of Alaska, hatcheries are
supplementing the salmon returns.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1335 April 2013
-------
This chapter summarizes the risk analysis results, organized by assessment endpoint, for a potential
mine at the Pebble deposit. For each endpoint, it integrates the various sources of risk, including those
from routine operations and accidents and failures, different physical and chemical exposures, and
different path ways of exposure and mechanisms of effects. In addition, it combines multiple types of
evidence, including evidence from analysis of the mine scenarios and from knowledge of analogous
mining operations. Limitations and uncertainties in the risk characterization are also summarized.
Finally, these results are extrapolated to the cumulative effects of multiple mines. See Chapters 7
through 13 for the derivation of these conclusions.
14.1 Overall Risk to Salmon and Other Fish
14.1.1 Routine Operation
During routine operations, mining would be conducted according to modern conventional practices,
including common mitigation measures at the mine site and along the transportation corridor. Toxic
effects would be minimized by collection of nearly all water from the site and treatment of collected
water to meet state standards and national criteria before discharge. However, toxic effects would still
occur, primarily due to the inevitable leakage of leachates. In addition, habitat loss and modification
would occur due to destruction of streams and wetlands and water withdrawals. As a result, local
populations of salmonids would decline in abundance and production.
14.1.1.1 Mine Footprint
Even in the absence of accidents or failures, the development of a mine at the Pebble deposit would
result in the destruction or modification of streams, wetlands, and ponds. Local habitat loss is
significant, because losses of stream habitat leading to losses of local, unique populations will erode the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
14-1
April 2013
-------
Chapter 14 Integrated Risk Characterization
population diversity that is key to the stability of the overall Bristol Bay salmon fishery (Schindler et al.
2010).
• Loss of 38, 90, and 145 km of streams in the footprint of the mine pit, tailings storage facilities
(TSFs) and waste rock piles, under the Pebble 0.25, 2.0, and 6.5 scenarios, respectively, would result
in the loss of 8, 24, and 35 km of streams known to provide spawning or rearing habitats for coho
salmon, sockeye salmon, Chinook salmon, and Dolly Varden.
• Altered streamflow resulting from retention and discharge of water used in mine operations, ore
processing, transport, and other processes would reduce the amount and quality offish habitat.
Alterations in streamflow exceeding 20% would adversely affect habitat in 15, 26, and 54 km of
streams under the Pebble 0.25, 2.0, and 6.5 scenarios, respectively, reducing production of coho
salmon, sockeye salmon, Chinook salmon, rainbow trout, and Dolly Varden. Reduced flows would
also result in an unquantified area of riparian floodplain wetland habitat being lost or altered in
terms of hydrologic connectivity with streams.
• Loss of 5.0 km2,12.4 km2, and 19.4 km2 of wetlands under the Pebble 0.25, Pebble 2.0, and Pebble
6.5 footprints, respectively, would reduce off-channel habitat for salmon and other fishes,
diminishing availability of and access to hydraulically and thermally diverse habitats that can
provide foraging opportunities and important rearing habitats for juvenile salmon.
• Indirect effects of stream and wetland losses would include reductions in the quality of
downstream habitat in the three headwater streams draining the mine scenario footprints, affecting
the same species as the direct effects. Modes of action for these effects include the following.
o A reduction in food resources would result from the loss of organic material and drifting
invertebrates exported from the 38 to 145 km of streams lost to the mine footprint.
o The balance of surface-water and groundwater inputs to downstream reaches would change.
Shifting the source water flow from groundwater to surface water could reduce winter habitat
and make the streams less suitable for spawning and rearing.
o Water treatment and discharge, resulting in reduced passage through groundwater flowpaths,
could alter summer and winter water temperatures, making streams less suitable for Pacific
salmon and Dolly Varden.
These indirect effects cannot be quantified, but it is likely that one or more of these mechanisms would
diminish fish production downstream of the mine scenario footprints in each watershed.
14.1.1.2 Water Collection, Treatment, and Discharge
Water in contact with tailings, waste rock, or the pit walls would leach copper and other metals. Our
assessment evaluates, for the three mine sizes, the emissions of treated wastewater and the realistic
expectation that leachate would escape the collection systems for the waste rock piles and TSFs. Routine
emissions from the wastewater treatment plant (WWTP) to the South and North Fork Koktuli Rivers
should be non-toxic due to treatment to achieve permit requirements. However, it may be somewhat
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7TI April 2013
-------
Chapter 14 Integrated Risk Characterization
toxic due to combined effects of multiple chemicals, poorly known and unregulated contaminants, and
untested receptor species.
The retention and collection of leachates are inevitably incomplete. Under our routine operations
scenario, leakage under the Pebble 2.0 and Pebble 6.5 scenarios would be sufficient to cause toxic levels
of copper and, to a much lesser extent, other metals in the streams draining the mine scenario
footprints. The most severe effects, including death of salmonids, would occur in the South Fork Koktuli
River, which would receive leachate from the acid-generating waste rock. Upper Talarik Creek would
experience death of invertebrates only below the station at which it receives interbasin transfer from
the South Fork Koktuli River. The North Fork Koktuli River would experience death of invertebrates
below TSF1. Death or inhibited reproduction of aquatic invertebrates, which are food for fish, are
estimated to occur in 15, 62, and 83 km of streams in the Pebble 0.25, 2.0, and 6.5 mine scenarios,
respectively. Avoidance of streams by salmonids would occur in 29 and 57 km of streams in the Pebble
2.0 and Pebble 6.5 scenarios. Death and reduced reproduction of salmonids would occur in 3.8 and 12
km of streams in the Pebble 2.0 and Pebble 6.5 mine scenarios.
The magnitude and extent of these predicted effects suggest the need for additional mitigation
measures, beyond the conventional practices assumed in the routine operations scenario, to reduce the
input of copper and other metals. A design based on conventional practices may be sufficient for a
typical porphyry copper mine, equivalent to Pebble 0.25, but not the massive Pebble 2.0 and 6.5 mine
sizes. Simply improving the efficiency of the capture wells or adding a larger wall or trench is unlikely to
achieve water quality criteria for those scenarios. Additional measures might include lining the waste
rock piles, reconfiguring the piles, or processing more of the waste rock as it is produced.
14.1.1.3 Road Construction and Operation
The assessment transportation corridor, including a road and four pipelines, would cross more than
50 streams; nearly 290 km of streams between the road and Iliamna Lake could be affected. Risks to
salmonids from the construction and operation of the transportation corridor are as follows.
• Loss and alteration of habitat through filling of wetlands for the road.
• Suspended and deposited sediment washed from the road, shoulders, ditches, cuts, and fills.
• Increased storm water runoff leading to increased suspended sediment, fine-bed sediment, salts,
and—at the mine site—metals.
• Chemical spills, likely four to five over a 2 5-year mine life.
• Increased dust leading to a direct increase in fine-bed sediment in the mining area, and indirect
increase—through a reduction in riparian vegetation—along the entire transportation corridor.
• Possible introduction of invasive species.
All of the above sources and stressors would likely lead to degraded or reduced habitat for salmon and
other fish.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7TI April 2013
-------
Chapter 14 Integrated Risk Characterization
14.1.2 Accidents and Failures
Any complex activity such as the mine described in this assessment inevitably experiences accidents and
failures. The number of ways in which failures and accidents can occur—their magnitudes, their
locations, and the circumstances of their occurrence—are effectively infinite. Hence, a complete and
specific assessment of risks from potential accidents and failures is not possible. Rather, a few failure
scenarios are presented, which emphasize the consequences of failures rather than the means by which
they are initiated. These scenarios address potential failures that could occur during mine operations or
after mine closure in perpetuity: failure to treat contaminated water; tailings dam failure; failures of
roads and culverts; wreck of a truck carrying a process chemical; and failure of a diesel, product
concentrate, or return water pipeline. Many other potential failures are not analyzed, including failures
of the on-site pipelines, spills of ore-processing chemicals on site, failures of tailings dams on streams
other than the North Fork Koktuli River, wild fires, waste rock slides, or failures at the port.
The probabilities and consequences of the failures analyzed are summarized in Table 14-1, and the
derivation of these estimates is discussed in Box 14-1. Probabilities of occurrence were estimated using
the best available information. Some of them are qualitative, because no applicable data are available.
Those that are quantitative are somewhat uncertain and their interpretation is not straightforward. For
example, the range of annual probabilities of a tailings dam failure is based on design expectations
rather than performance data, which are unavailable for recently constructed large earthen dams. The
actual observed frequency of tailings dam failures is near the upper end of that range, which suggests
that the range is reasonable at that bound. However, the lower bound (1 in 250,000 per annum) is
purely aspirational, in that it has no empirical basis.
It is important to remember that this is an assessment of mine scenarios. It is based on modern
conventional mining practices, especially the plan proposed for the Pebble site by Northern Dynasty
Minerals (Ghaffari et al. 2011). However, like any predictive assessment, it is hypothetical. Although the
major features of the scenarios will undoubtedly be correct (e.g., a pit at the location of the ore body,
waste rock deposited near the pit, the generation of a large volume of tailings), some specifics would
inevitably differ. This would be true of any scenario including a mining plan submitted for permitting or
even a plan approved by the state. All plans are scenarios, and although each new plan is expected to be
closer to actual operations than the ones before, unforeseen circumstances and events and new
technologies inevitably compel changes in practice.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , ._. April 2013
-------
Chapter 14
Integrated Risk Characterization
Table 14 1. Summary of probability and consequences of potential failures under the mine
scenarios.
Failure Type
Tailings dam
Product concentrate pipeline
Concentrate spill into a stream
Concentrate spill into a wetland
Return water pipeline spill
Diesel pipeline spill
Culvert, operation
Culvert, post-operation
Truck accidents
Water collection and treatment,
operation
Water collection and treatment,
managed post-closure
Water collection and treatment,
after site abandonment
Probability3
4 x 1O4 to 4 x 1O6 per dam-year =
recurrence frequency of 2,500 to
250,000 yearsb
10'3 per km-year = 95% chance
per pipeline in 25 years
1.5 x 10'2 per year = 1 to 2
stream-contaminating spills in 78
years
3.8 x ID-2 per year = 2 wetland-
contaminating spills in 78 years
Same as product concentrate
pipeline
Same as product concentrate
pipeline
Low
3 x 10-1 to ~6 x 10-1 per culvert;
instantaneous = 11 to 21 culverts
1.9 x 10-7spills per mile of travel =
4 accidents in 25 years and 2
near streams in 78 years
0.60 to 0.93 = proportion of
recent U.S. mines with reportable
water collection and treatment
failures. Better practices might
reasonably be expected to reduce
this to 0.1.
Somewhat higher than operation
Certain
Consequences
More than 30 km of salmonid stream would be
destroyed and more streams and rivers would have
degraded habitat for decades.
Most failures would occur between stream or wetland
crossing and might have little effect on fish.
Fish and invertebrates would experience acute
exposure to toxic water and chronic exposure to toxic
sediment in a stream and potentially extending to
Iliamna Lake.
Invertebrates and potentially fish would experience
acute exposure to toxic water and chronic exposure to
toxic sediment in a pond or other wetland.
Fish and invertebrates would experience acute
exposure to toxic water if spilled to a stream or
wetland.
Acute toxicity would reduce the abundance and
diversity of invertebrates and possibly cause a fish kill
if spilled to a stream or wetland.
Frequent inspections and regular maintenance would
result in few impassable culverts, but for those few,
blockage of migration could persist for a migration
period, particularly for juvenile fish.
In surveys of road culverts, 30 to 58% are impassable
to fish at any one time. This would result in 11 to 21
salmonid streams blocked. In 10 to 19 of the 32
culverted streams with restricted upstream habitat,
salmon spawning may fail or be reduced and the
streams would likely not be able to support long-term
populations of resident species.
Accidents that spill processing chemicals into a stream
or wetland could cause a fish kill.
Water collection and treatment failures are very likely
to result in exceedance of standards potentially
including death offish and invertebrates, but not
necessarily as severe or extensive as estimated for the
failure scenario.
Collection and treatment failures are highly likely to
result in release of untreated or incompletely treated
leachates for days to months, but the water would be
less toxic due to elimination of PAG waste rock.
When water is no longer managed, untreated
leachates would flow to the streams. However, the
water would be less toxic due to elimination of PAG
waste rock.
3 Because of differences in derivation, the probabilities are not directly comparable.
b Based on expected state safety requirements. Observed failure rates for earthen dams are higher (about 5 x 1CH per year or a recurrence
frequency of 2,000 years).
PAG = potentially acid-generating.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
14-5
April 2013
-------
Chapter 14 Integrated Risk Characterization
BOX 14 1. FAILURE PROBABILITIES
Table 14-1 presents probability estimates and consequences of different kinds of failures. Here, we explain the
derivation of these estimates. As much as possible, multiple methods are used within a failure type to determine
how robust the estimates may be. The methods differ amongfailure types and the results are not strictly
equivalent, but they do convey the likelihood of occurrence. More details can be found in Chapters 6 through 11.
Tailings Dam Failures. The most straightforward method of estimating the annual probability of failure of a tailings
dam is to use the failure rates of existing dams. Three reviews of earthen dam failures produced an average rate
of 1 failure per 2,000 dam years (i.e., a recurrence frequency of 2,000 years), or 5 x 1Q-4 per year. The argument
against this approach is that it does not reflect current engineering practice. The State of Alaska's guidelines
suggest that an applicant follow accepted industry design practices such as those provided by U.S. Army Corps of
Engineers and the Federal Energy Regulatory Commission. Both regulatory agencies require a minimum factor of
safety of 1.5 for the loading condition corresponding to steady seepage at the maximum storage facility. An
assessment of the correlation of dam failure probabilities with safety factors against slope instability suggests an
annual probability of failure of 1 in 1,000,000 years for Category I Facilities (those designed, built, and operated
with state-of-the-practice engineering) and 1 in 10,000 years for Category II Facilities (those designed, built, and
operated using standard engineering practice). This corresponds to risks of 10'4to 10'6 per year. The advantage of
this approach is that it addresses current regulatory expectations and engineering practices. The disadvantage is
that we do not know whether standard practice or state-of-the-practice dams designed with safety factors would
perform as expected. Slope instability is only one type of failure; other failure modes, such as overtopping during a
flood, would increase overall failure rates. Slope stability failures account for about one-fourth of tailings dam
failures, so the probability of failure from all causes could be estimated to be 1 in 250,000 (Category I) to 1 in
2,500 (Category II). The mine scenarios include up to three TSFs, most with multiple dams, so the annual
probability of any dam failing would be approximately equal to the annual probability of a single dam failure times
the number of dams.
Pipeline Failures. A review of observed pipeline failure rates for oil and gas pipelines yields an average annual
probability of failure per kilometer of pipeline of 10'3 or a frequency of 1 failure per 1,000 km per annum. This
average risk comes very close to estimating the observed failure rate of the copper concentrate pipeline at the
Minera Alumbrera mine, Argentina. This annual failure probability, over the 113-km length of each pipeline within
the Kvichak River watershed, results in a 0.11 probability of a failure in each of the four pipelines each year, or a
recurrence frequency of 8.5 years. If the probability of a failure is independent of location, and if it is assumed
that spills within 100 m of a stream could flow to that stream, a spill would have a 0.14 probability of entering a
stream within the Kvichak River watershed. This would result in an estimate of 0.015 stream-contaminating spills
per annum, or 1 stream-contaminating spill over the duration of the Pebble 6.5 scenario (approximately 78 years).
Similarly, a spill would have a 0.35 probability of entering a wetland, resulting in an estimate of 0.038 wetland-
contaminating spills per annum or 3 wetland-contaminating spills over the duration of the Pebble 6.5 scenario.
Water Collection and Treatment Failures. During mine operation, collection or treatment of leachate from mine
tailings, pit walls, or waste rock piles would be incomplete and could fail in various ways. Under the routine
operations scenario, leachate from the unlined TSFs and waste rock piles would not be fully collected. Equipment
and operation failures and inadequate designs would also result in failures to avoid toxic emissions. Reviews of
U.S. mine records found that 60 to 93% of mines reported a water collection or treatment failure (Kuipers et al.
2006, Earthworks 2012). Improved design and practices should result in lower failure rates, but it is unlikely that
failure rates would be lower than 10% over the life of a mine, given this record. During operation, the failures
should be brief (less than 1 week) unless they involve a faulty system design or parts that are difficult to replace.
After a mine is abandoned (potentially many years after closure), water management would end and the discharge
of untreated water would become inevitable but may not be problematical.
Culvert Failures. Culvert failure is defined as a condition that blocks fish passage. Empirical data for culvert
failures are not based on rates of failure of culverts but rather on instantaneous frequencies of culverts that were
found to have failed in road surveys. The frequencies in recent surveys range from 0.30 to 0.58 (3 to 6 x 10'1) per
culvert. In the Kvichak River watershed, 35 streams that are believed to support salmonids (salmon, trout, or Dolly
Varden) have culverts, so at anytime 11 to 21 culverted streams would be expected to have blocked fish passage
at the published frequencies. The proportion of failed culverts during mine operation should be much lower.
14.1.2.1 Tailings Dam Failure
Failure of a tailings dam would have a one in 2,500 to one in 250,000 probability of occurrence per year
for each TSF. Probability of a tailings dam failure increases with an increase in the number of dams. The
Pebble 0.25 scenario includes one TSF and the Pebble 6.5 scenario includes three. Each TSF has multiple
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment *. g April 2013
-------
Chapter 14 Integrated Risk Characterization
dams, but the probability of a spill from a TSF would not increase in proportion to the number of dams
for an individual TSF, because failures would not be independent events. The failure of one dam on a TSF
would relieve pressure on others, reducing the probability of multiple failures; conversely, common
mode failures could occur, increasing the probability of multiple failures. The dam failure analyses in
this assessment simulated the release of 20% of the tailings (a conservative estimate) from a partial-
volume (90-m) and a full-volume (209-m) dam at TSF 1.
Failure of the primary TSF 1 dam would result in the release of a flood of tailings slurry into the North
Fork Koktuli River, scouring the valley and depositing tailings. The complete loss of suitable salmonid
habitat in the North Fork Koktuli River (at least 30 km of habitat, which was the extent of the model) in
the short-term (less than 10 years) and the high likelihood of very low-quality spawning and rearing
habitat in the long-term (for decades) would result in the nearly complete loss of mainstem North Fork
Koktuli River fish populations. Even salmon at sea during the failure would not find suitable spawning
habitat on their return to the North Fork Koktuli River as adults. The river currently supports spawning
and rearing populations of sockeye, Chinook, and coho salmon, spawning populations of chum salmon,
and rearing populations of Dolly Varden and rainbow trout. Suspended mine tailings sediments would
continue for an unknown (due to model and data limitations) distance further down the Koktuli River,
and probably into the Mulchatna and Nushagak Rivers, causing degraded spawning habitat and reduced
food resources. Fish anywhere in the flowpath below a tailings dam failure would be killed or forced
downstream. Fish migrating into tributaries of affected rivers would be inhibited from migration for
some period of time, which our model could not predict.
Following the slurry flood, deposited tailings would continue to erode from the North Fork Koktuli and
Koktuli River valleys. After many years, a new channel with gravel substrate and a natural floodplain
structure would become established. However, that recovery would come at the expense of the
downstream Mulchatna and Nushagak Rivers, as much of the spilled tailings initially deposited in the
North Fork Koktuli and Koktuli Rivers would be resuspended by erosion and transported down the
drainage. This process could not be modeled with existing data and resources, but would be inevitable if
a tailings spill occurred.
High concentrations of suspended tailings would occur following a tailings dam failure, but over time
they would decline as erosion progressed. For some years, periods of high flow would be expected to
suspend sufficient concentrations of tailings to cause avoidance, reduced growth and fecundity, and
possibly even death offish. Migration to and from any affected tributaries would be impeded, if flow
from the tributaries was not sufficient to adequately dilute suspended sediment concentrations,
meaning that fish would not reach spawning grounds, winter refugia, or seasonal feeding habitats.
Deposited tailings would degrade habitat quality for both fish and the invertebrates they eat. Pacific
salmon, Dolly Varden, and rainbow trout spawn in gravels, and their eggs and larvae require sufficient
space within the gravel for water to circulate. Juvenile salmonids require even larger clear spaces for
concealment from predators and for overwintering habitat. Tailings would fill those interstitial spaces.
An increase in fines of more than 5% causes unacceptable effects on salmonid reproduction. Until
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7TZ April 2013
-------
Chapter 14 Integrated Risk Characterization
considerable erosion occurred and a gravel-bedded channel was re-established, female salmonids would
be unable to clean the gravel to spawn. Even where gravel was available, high deposition from upstream
erosion of tailings could smother eggs and larvae. Recovery of suitable substrates via mobilization and
transport of tailings fines would take decades, and would affect much of the watershed downstream of
the failed dam.
In addition to degrading fish habitat, deposited tailings are potentially toxic. Based largely on their
copper content, deposited tailings would be toxic to benthic macroinvertebrates, although existing data
concerning toxicity to fish is less clear. Estimated pore water concentrations are less than published
thresholds for chronic effects in fish, but directly relevant tests of salmonid early life stages have not
been conducted. The combined effects of copper toxicity and poor habitat quality (particularly low
dissolved oxygen concentrations) caused by fine sediment are unknown. Dietary exposures of salmonids
via invertebrate prey exposed to tailings are estimated to be marginally toxic.
In sum, a TSF 1 dam failure would have severe direct and indirect effects on aquatic resources, and
specifically on salmonids. In the short-term (less than 10 years), certainly the North Fork Koktuli River
below the TSF 1 dam failure location and very likely much of the Koktuli River would not support
salmonids. For a period of decades, those waters would provide very low-quality spawning and rearing
habitat, likely resulting in the nearly complete loss of North Fork Koktuli fish populations. Deposition,
resuspension, and redeposition of tailings would likely cause serious habitat degradation in the Koktuli
River and downstream into the Mulchatna River. Ultimately, spring floods and stormflows would carry
some portion of the tailings into the Nushagak River. Effects would be qualitatively the same for both the
partial-volume and full-volume dam failures, although effects from the full-volume failure would extend
up to 272 km further and last longer.
The Koktuli River watershed is an important producer of Chinook salmon for the larger Nushagak
Management Zone. The Nushagak River watershed is the largest producer of Chinook salmon in the
Bristol Bay region, with an annual escapement of nearly 160,000 fish (1966 to 2010) (Buckpers.
comm.). Assuming ADF&G aerial survey counts (Dye and Schwanke 2009) reflect the proportional
distribution of Chinook salmon within the Nushagak River watershed, the tailings dam failure would
eliminate 29% of that run due to loss of the Koktuli River salmon population; an additional 10 to 20%
could be lost because tailings deposited in the Mulchatna River would affect its tributaries. Sockeye
salmon are the most abundant salmon returning to the Nushagak River watershed, with spawning
escapement averaging more than 1.3 million fish. However, the proportion of sockeye and other salmon
species that originates in the Koktuli and Mulchatna River watersheds is unknown. Similarly,
populations of rainbow trout and Dolly Varden of unknown size would be lost for decades.
The dam failures evaluated in the assessment used TSF 1 as a plausible location. Failure of the other
tailings dams at TSF 2 and TSF 3 were not modeled, but would have similar effects in the South Fork
Koktuli River and downstream.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment , ._„ April 2013
-------
Chapter 14 Integrated Risk Characterization
14.1.2.2 Wastewater Treatment Plant Failure
Under the WWTP failure scenario, untreated wastewater would be discharged. The most severe effects,
including lethality to invertebrates and fish, would occur in the South Fork Koktuli River where the
untreated effluent would mix with toxic waste rock leachate. The North Fork Koktuli River, where the
untreated waste would mix with tailings leachate, would experience lethality to invertebrates and,
depending on the season, reduced growth or survival of early life stages offish. Upper Talarik Creek
would receive no wastewater discharge and would experience no additional effects. The WWTP failure
is estimated to result in lethality or reduced reproduction of invertebrates in more than 100 km of
stream under all three mine sizes. For salmonids, it is estimated to cause avoidance of 103 km of stream,
sensory inhibition in 92 km, reduced reproduction in 84 km, and mortality in 31 km under the Pebble
6.5 scenario. Direct effects on fish would be less extensive under the Pebble 2.0 scenario avoidance in
102 km, sensory inhibition in 34 km, and reduced reproduction in 34 km—and would be limited to
avoidance in 45 km under the Pebble 0.25 scenario.
14.1.2.3 Culvert Failure
The most likely serious failure associated with the potential access road would be blockage or failure of
culverts. Culverts can commonly become blocked by debris that may not stop water flow, but that could
block fish passage. Culverts can also fail to convey water because of landslides or, more commonly,
flooding that washes out the culvert. In such failures, the stream may temporarily be impassable to fish
until the culvert is repaired or until erosion re-establishes the channel. If either of these failures
occurred during adult salmon immigration or juvenile salmon outmigration and the blockage was not
cleared for several days, production of a year-class could be lost or diminished. The most common
problems for fish from culverts are excessive gradient, insufficient water depth, excessive velocity, and
excessive inlet hydraulic jumps. These problems could also inhibit fish passage and may not be detected
by casual inspections.
Culvert failures could also result in the downstream transport and deposition of sediment. This could
cause returning salmonids to avoid the stream, if they arrived during or immediately following the
failure. More likely, the deposition of fine sediment from the washed-out culvert would smother
salmonid eggs and larvae, if they were present, and would degrade the downstream habitat for
salmonids and the invertebrates that they eat. It would also change stream hydraulics and channel
morphology, generally diminishing habitat value.
Blockage of fish passage at road crossings would be infrequent during operation, because our scenarios
assume daily inspection and maintenance. However, after mine operations end, the road may be
maintained less carefully or maintenance may be transferred to a state or a local governmental entity. In
that case, the proportion of culverts that are impassable would be expected to revert to the levels found
in published surveys (30 to 60% inhibit fish passage at any time) (Langill and Zamora 2002, Gibson etal.
2005, Price etal. 2010). Of the many culverts that would be required, 35 would be on streams that are
believed to support salmonids. Hence, 11 to 21 streams would be expected to lose passage of salmon or
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment *. „ April 2013
-------
Chapter 14 Integrated Risk Characterization
resident trout or Dolly Varden and some proportion of those would have degraded downstream habitat
resulting from sedimentation caused by road washout
Of the 35 culverted salmonid streams, 32 contain restricted (less than 5.5 km) upstream habitat.
Assuming typical maintenance practices after mine operations, approximately 10 to 19 of the 32
streams would be entirely or in part blocked at any time. As a result, isolation of resident species such as
rainbow trout or Dolly Varden in such short stream segments would likely result in failure of the
populations, if that isolation was sustained.
It should be noted that high flows in and immediately downstream of a culvert and the structure of the
culvert may inhibit fish passage even if movement is not blocked. Culvert-induced erosion could cause
channel entrenchment, disrupting floodplain habitat and floodplain/channel ecosystem processes.
14.1.2.4 Pipeline Failure
The primary product of the mine would be a concentrate of copper and other metals that would be
pumped as a slurry in a pipeline to a shipping facility on Cook Inlet. Water that carried the sand-like
concentrate would be returned to the mine site in a second pipeline. Based on the record of pipelines, in
general, and metal concentrate pipelines, in particular, one to two near-stream failures of each of these
pipelines would be expected to occur over the duration of the Pebble 6.5 scenario (approximately
78 years). In either case, metal-contaminated water would be released, potentially killing fish and
invertebrates in the affected stream over a relatively brief period. The aqueous phase of the concentrate
slurry would be lethal to sensitive invertebrates and potentially to larval fish, but a kill of adult fish is
not expected. If the concentrate pipeline spilled into a stream, concentrate would, depending on flows,
settle and form bed sediment, be carried downstream and deposited in low areas, or be carried to
Iliamna Lake and deposited near the shore. Deposited concentrate is predicted to be highly toxic based
on its high copper content and the acidity of its leachate. Unless the receiving stream was dredged,
causing physical damage, this sediment would persist for decades before ultimately being washed into
Iliamna Lake. Potential concentrations in the lake could not be predicted; however, near the pipeline
route, Iliamna Lake contains important beach spawning areas for sockeye salmon that could be exposed
to a spill. Sockeye also spawn in the lower reaches of streams that could be directly contaminated by a
spill.
Spills from a diesel pipeline are estimated to have the same probability of occurrence as concentrate
spills. Based on multiple lines of evidence, a spill under the diesel pipeline failure scenario would be
sufficient to kill invertebrates and possibly fish. Remediation is expected to have little success, and
recovery would likely occur in 1 to 3 years.
14.1.2.5 Common Mode Failures
Multiple failures could result from a common event, such as an earthquake or a severe storm with heavy
precipitation (particularly heavy rain on snow). Failures resulting from such an event could include one
to three tailings dam failures that spill tailings slurry to streams and rivers, road culvert washouts that
send fine sediment downstream and potentially block fish passage, and product slurry and return water
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14-10April 2013
-------
Chapter 14 Integrated Risk Characterization
pipeline failures resulting from a culvert washout and scouring of the streambed or a slide of the
roadbed. The effects of these accidents individually would be the same as discussed previously, but the
co-occurrence of these failures would cause cumulative effects on salmonid populations and would
make any mitigative response more difficult.
Over the perpetual timeframe that tailings, mine pit, road, and waste rock would be in place, the
likelihood of multiple extreme precipitation events, earthquakes, or combinations of these events
becomes much greater. Multiple events further increase the chances of weakening and eventual failure
of facilities that are still in place.
14.2 Overall Loss of Wetlands
Wetlands are a dominant feature of the landscape in the Pebble deposit area and are important habitats
for salmon and other fish. Ponds and riparian wetlands provide spawning, rearing, and refuge habitat
for both anadromous and resident fish. Other wetlands moderate flows and water quality, and can
influence downstream delivery of dissolved organic matter, particulate organic matter, and aquatic
macroinvertebrates that supply energy sources to fish. 5.0 km2,12.4 km2, and 19.4 km2 of wetlands
would be filled or excavated under the Pebble 0.25, 2.5, and 6.5 scenarios, respectively. In addition, an
unquantifiable area of riparian floodplain would be lost or would suffer substantial changes in
hydrologic connectivity with streams, due to reduced flow from the mine footprint. Another 0.11 km2 of
wetlands would be filled in the Kvichak River watershed by the roadbed of the transportation corridor.
By interrupting flow and adding silt and salts, the roadbed would also affect approximately 2.4 to 4.9
km2 of wetlands. Finally, a diesel or product concentrate spill could damage wetlands and eliminate or
degrade their capacity to support fish.
14.3 Overall Fish-Mediated Risk to Wildlife
The interactions between salmon and wildlife and the potential for disruption of these interactions are
complex. Annual salmon runs provide food for brown bears, bald eagles, other land birds, and wolves. In
addition, the abundance and production of wildlife is enhanced by the marine nutrients that salmon
carry on their spawning migration. Those nutrients are released into streams when the salmon die,
enhancing the production of other aquatic species that feed wildlife. Salmon predators deposit nutrients
on the landscape, fertilizing the vegetation and increasing the abundance and production of moose,
caribou, and other wildlife.
The effects of reduced Pacific salmon, Dolly Varden, and rainbow trout production on wildlife would be
complex, cannot be quantified at this time, and may not be linearly proportional. Factors such as the
magnitude of salmon loss, seasonality and duration of the loss, and location of the loss would contribute
to the specific species affected and the magnitude of the effect. However, some degree of reduction in
wildlife would be expected from both the mine scenario footprint and routine operations under each
scenario. Because salmon provide a food source for brown bear, wolf, bald eagle, and other birds, it is
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14-11 April 2013
-------
Chapter 14 Integrated Risk Characterization
likely these species would be directly affected by a reduction in salmon abundance. Indirect effects on
water birds and land birds through a loss of aquatic invertebrates, and on moose and caribou through a
loss of marine-derived nutrients to vegetation are likely, but research is needed to document those
linkages.
Because of their importance in the ecosystem of these watersheds, the loss of salmon and brown bears
in a given area would result in "significant changes in the productivity, diversity and physical structure
of their communities, far beyond just their 'food chain' interactions" (Brna and Verbrugge 2013).
Fish-eating wildlife species are also potentially exposed to contaminants bioaccumulated by fish.
However, analyses based on the concentrations of metals in waste rock, tailings, and product
concentrate leachates suggest that toxic effects to wildlife via this route of exposure are unlikely.
14.4 Overall Fish-Mediated Risk to Alaska Native Cultures
Because of the high reliance on salmon for subsistence, and the close connection between salmon and
the indigenous culture, Alaska Natives are particularly vulnerable to any changes in the quantity or
quality of wild salmon resources. Any change in salmon resources would likely change the diet, social
networks, cultural cohesion, and spiritual well-being of the Alaska Native cultures in the watershed.
These changes could, in turn, result in the following.
• Effects on human health from loss of a highly nutritious subsistence food and the physical and
mental benefits of a subsistence way of life.
• Degradation of a social support system based on food sharing.
• Decrease in family cohesion and cultural continuity from a loss of family-based subsistence work.
• Mental health degradation from a disruption of spiritual practices and beliefs centered on salmon
and clean water.
Human health and cultural effects related to decreases in salmon resources would vary with the
magnitude of these reductions and cannot be predicted quantitatively. Some fish-mediated effects on
Alaska Native culture are likely from the footprint or operation under any of the scenarios considered.
At a minimum, there would be a loss of subsistence use areas, potential loss of access by boat due to
water level fluctuations, and the risk of decreased use offish because of a perception of a change in
quality of the fish from mine operations. In the transportation corridor, complex and unpredictable
changes to subsistence use would result from increased access (by both Alaska Natives and others) and
possible habitat changes. If significant failures of water treatment or other infrastructure during or after
mine operation, which greatly affect salmon resources, large-scale impacts on both subsistence food
resources and the cultural, social, and spiritual cohesion of the local indigenous cultures could occur.
Because the Alaska Native cultures in the Bristol Bay watershed have significant ties to specific land and
water resources, which have evolved over thousands of years, it is not possible to replace elsewhere the
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14^12 April 2013
-------
Chapter 14 Integrated Risk Characterization
value of these subsistence use areas lost to mine operations. As a result, compensatory mitigation,
restoration, or replacement in the case of a failure would be particularly difficult, if not impossible.
It should be noted that, although this assessment focuses on potential effects on Alaska Native cultures,
many of the non-Alaska Natives that reside in the area practice a subsistence way of life and have strong
long-term cultural ties to the landscape that go back generations. In addition, a large group of seasonal
commercial fisherman and cannery workers depend on these resources and have strong, multi-
generational cultural connections to the region as well. These groups are also vulnerable to negative
impacts on salmon.
14.5 Summary of Uncertainties and Limitations in the
Assessment
This assessment makes various reasonable assumptions about mining, processing, and transporting of
the porphyry copper resources in the Pebble deposit and elsewhere in the Nushagak and Kvichak River
watersheds. If those resources are mined in the future, actual events would not be identical to the mine
scenarios. This is not treated as a source of uncertainty, because it is an inherent aspect of any predictive
assessment. Even an environmental assessment of a mining company's proposed plan would be an
assessment of a scenario that undoubtedly would differ from actual events.
As discussed in the individual chapters, this assessment does have uncertainties and limitations in the
extent to which the potential effects of implementing any of the scenarios can be estimated. Major
uncertainties are summarized below.
• The estimated annual probability of a tailings dam failure is uncertain and based on design goals
rather than historical experience. Actual failure rates could be higher or lower than the estimated
range of probabilities.
• The proportion of the tailings that would spill in the event of a dam failure could be larger than the
largest value modeled (20%). However, even this conservative assumption results in an initial
outflow beyond the 3 0-km limit of the model.
• The ultimate fate of spilled and deposited tailings in the event of a dam failure could not be
quantified. From principles of geohydrology and review of analogous cases, we know that slurry
would erode from areas of initial deposition and move downstream over a period of more than a
decade. However, the data needed to model that process and the resources to develop the model
were not available.
• It is uncertain whether a tailings spill into a remote roadless area would be remediated, how it could
be remediated, how long it would take to remediate, and to what extent remediation could reduce
effects downstream of the initial runout of the slurry.
• The effects of mining on fish populations could not be quantified because of the lack of quantitative
information concerning Pacific salmon, Dolly Varden, and rainbow trout populations and their
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14^13 April 2013
-------
Chapter 14 Integrated Risk Characterization
responses. The occurrence of salmonid species in rivers and major streams is generally known, but
not their abundances, productivities, or limiting factors. Estimating changes in populations would
require population modeling, which requires knowledge of life-stage-specific survival and
production as well as knowledge of limiting factors and processes that were not available for this
case. Further, it requires knowledge of how temperature, habitat structure, prey availability, density
dependence, and sublethal toxicity influence life-stage-specific survival and production, which is not
available. Obtaining that information would require more detailed monitoring and experimentation.
Further, salmon populations naturally vary in size because of a great many factors that vary among
locations and years. Collecting sufficient data to establish reliable salmon population estimates takes
many years. Thus, we used estimated effects of mining on habitat as a reasonable surrogate for
estimated effects on fish populations.
• Standard leaching test data are available for test tailings and waste rocks from the Pebble deposit,
but these results are uncertain predictors of the actual leachate composition from a tailings
impoundment, tailings deposited in streams and their floodplains, and waste rocks in piles. Test
conditions are artificial, and the materials tested may not be representative. In particular, the pyritic
tailings were not tested. Additionally, data and resources were insufficient to allow geochemical
modeling of water quality expected in the TSF or downstream of the mine site under varied
chemical and hydrological conditions, or to model expected pit water chemistry at closure.
• The effects of tailings and product concentrate deposited in spawning and rearing habitat are
uncertain. It is clear that they would be harmful to salmonid eggs, alevins, or sheltering fry due to
both physical and toxicological effects, but the concentration in spawning gravels required to reduce
reproductive success of salmonids is unknown.
• The actual response of Alaska Native cultures to any of these scenarios is uncertain. Interviews with
tribal Elders and culture bearers and other evidence suggest that responses would involve loss of
food resources and cultural disruption, but it is not possible to predict specific changes in
demographics, cultural practices, or physical and mental health.
• Although some tailings would eventually reach the estuarine portions of the Nushagak River and
even Bristol Bay, exposures at that distance could not be estimated. Therefore, risks to salmonids
resulting from marine and estuarine contamination could not be addressed.
• The assessment is limited by its focus on the effects of mining on salmonids and the indirect effects
of diminished fish resources on wildlife and people. Neither direct effects on humans, wildlife, and
terrestrial ecosystems nor secondary development associated with mine development are included
in this assessment.
• Some sources, such as air pollution from a power plant, were not addressed because they are less
related to the Clean Water Act or because they were judged to pose less risk to salmonids.
• Climate change will affect both the probability and magnitude of failures in a mining operation, as
well as change the habitat quality and biology of salmonids. These climate effects are highly
uncertain, but their likely qualitative influences are described in Box 14-2.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14-14 April 2013
-------
Chapter 14 Integrated Risk Characterization
BOX 14 2. CLIMATE CHANGE AND POTENTIAL RISKS OF LARGE SCALE MINING
Climate change in the Bristol Bay region (Section 3.8) will likely result in changes in snowpack and the timing of
snowmelt, a greater chance for rain-on-snow events, and an increase in flooding. These changes are likely to affect
multiple aspects of any large-scale mining in the area, including mine infrastructure, the transportation corridor,
water treatment and discharge, and post-closure management (Pearce et al. 2011).
Mine infrastructure (e.g., buildings, waste rock piles, tailings retention structures, and water retention facilities), and
the transportation corridor (i.e., roads, associated bridges and culverts, and pipelines), would likely be affected by
extreme weather events resulting in increased flooding (Instanes et al. 2005, Pearce et al. 2011). Mine
infrastructure and the transportation corridor would need to be designed for potential increases in flood frequency
and magnitude as well as potential changes in storm patterns, because these changes could weaken structural
integrity, increase embankment instability, and accelerate erosion (Instanes et al. 2005, Pearce et al. 2011).
Water management would be a major challenge at the mine site, and changes result!ngfrom climate change could
exacerbate the challenge. Climate change could contribute to future changes in temperature, precipitation,
evapotranspiration, hydrology, and seasonal flooding patterns and dry periods. Changes in water availability and
groundwater recharge could affect the amount and timing of water available and the hydrologic gradients of
groundwater in and around the mine site, thereby requiring changes to water management in the pit and other areas
of the mine site.
Under future climate conditions, the return intervals of various sized storms could change (e.g., medium-sized
storms could become more frequent or the frequency of current 100-year storms could change). Possible increases
in flood magnitudes would require the need to plan for larger and more frequent flood events at the mine site. This
in turn may affect the likelihood of a tailings dam failure, overtopping of ponds, and/or flooding of water
management facilities. Failure to plan for these conditions could result in unintended environmental releases. For
example, in 2008, Minto Mine, a copper-gold mine in Canada, was forced to release untreated water into the Yukon
River system due to torrential rains and the mine's inability to manage this increased water (Pearce et al. 2011).
Mining infrastructure would need to be designed to account for projected climatic changes, such that its structural
integrity can be maintained in perpetuity, even under potentially more extreme climatic conditions. In addition, any
mine reclamation plan would need to consider changing climate conditions and how those changes will directly
affect fish and wildlife populations. The following list includes infrastructure and operations design, maintenance,
and management that would need to consider climate change.
• Mine footprint. Climate change may affect water availability both within and across seasons (e.g., via changes in
snowmelt patterns; amount, type, and timing of precipitation; frequency of large storms; and groundwater
inputs). Water processing associated with these changes could alter flow and temperature in downstream water
bodies.
• Water treatment and discharge. Climate change might result in greater volumes of water requiring treatment,
changes in the dilution provided by receiving streams, changes in temperature that would affect management of
discharged water, and potential overload due to lack of storage and treatment capacity.
• Tailings storage facility (TSF) failure. TSFs may be exposed to greater volumes of water, and the probability of
dam failure due to overtopping may increase with changes in precipitation patterns and/or rapid snowmelt.
• Transportation corridor. Greater flood frequencies and an increase in erosion and sedimentation are likely to
affect streams and wetlands along the transportation corridor.
• Culvert, pipeline, and bridge failures. These failures may be more likely due to changes in precipitation patterns,
rapid snowmelt, larger water volumes, debris issues, and sedimentation.
• Cumulative effects. Climate change issues are likely to affect any mining operation in the area. As the number of
mines increases, the likelihood of having mine and climate change interactions might increase.
Climate change could obscure or complicate efforts to monitor habitat and fish population responses to mine-related
activities. Survey and monitoring designs would need to take potential climate change effects into account through
strategic measurement of stream and lake temperatures, precipitation, water flow, and fish populations throughout
the Bristol Bay watershed. Monitoring design could be aided by models able to downscale climate effects and
project changes at watershed scales. Population monitoring should take salmon adaptation and metapopulation
dynamics into account, and be cognizant of the many interacting processes influencing populations. Protecting
salmon sustainability in an uncertain future will require adaptability of both management and monitoring strategies
(Schindler etal. 2008).
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14-15April 2013
-------
Chapter 14 Integrated Risk Characterization
14.6 Summary of Uncertainties in Mine Design and Operation
In addition to uncertainties in assessment, uncertainties are inherent in planning, designing,
constructing, operating, and closing a mine. Such uncertainties are inherent in any complex enterprise,
particularly when it involves an incompletely characterized natural system. However, the large scales
and long durations of any effort to mine the Pebble deposit make these inherent uncertainties more
prominent.
• Mines are complex systems requiring skilled engineered design and operation. The uncertainties
facing mining and geotechnical engineers include unknown geologic features, uncertain values in
geological properties, limited knowledge of mechanisms and processes at the site, and human error
in design, construction, and operation. Vick (2002) notes that models used to predict the behavior of
an engineered system are "idealizations of the processes they are taken to represent, and it is well
recognized that the necessary simplifications and approximations can introduce error in the model."
Engineers use professional judgment in addressing uncertainty (Vick 2002).
• Accidents are inherently unpredictable. Though systems can be put into place to reduce system
failures, seemingly logical decisions about how to respond to a given situation can have unexpected
consequences resulting from human error—for example, the January 2012 overtopping of the
tailings dam at the Nixon Fork Mine near McGrath, Alaska (Box 8-1). Further, unforeseen events or
events that are more extreme than anticipated can negate the apparent wisdom of prior decisions
(Caldwell and Charlebois 2010).
• The ore deposit would be mined for decades, and the waste would require management for
centuries or even in perpetuity. Engineered waste storage systems of mines have only been in
existence for about 50 years, so their long-term behavior is not known. The performance of modern
technology in the construction of tailings dams is untested and unknown in the face of centuries of
extreme events such as earthquakes and major storms.
• Human institutions change. Over the long time span of mining and post-mining care, generations of
mine operators must exercise due diligence. Priorities are likely to change in the face of financial
crises, changing markets for metals, new information about the resource, political priorities, or any
number of currently unforeseeable changes in circumstance. The promises of today's mine
developers may not be carried through by future generations of operators whose sole obligation is
to the shareholders of their time (Blight 2010). Similarly, governments that are expected to assume
responsibility when mining companies fail may not appropriately manage mine sites or the funds in
performance bonds.
14.7 Summary of Risks under the Mine Scenarios
Even if the mining and mitigation practices described in the mine scenarios were performed perfectly,
an operation of this size would inevitably destroy or degrade habitat of salmonids. The mine scenario
footprints would eliminate, block, or dewater streams known to support spawning and rearing habitat
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14-16April 2013
-------
Chapter 14
Integrated Risk Characterization
for coho, Chinook, and sockeye salmon and Dolly Varden (Table 14-2). Wetlands would be filled or
excavated in 5,12.4, and 19.4 km2 of the mine footprints under the Pebble 0.25, 2.0, and 6.5 scenarios,
respectively. Reduced flow from water use would significantly degrade additional stream reaches (Table
14-2) and an unquantifiable area of wetland habitat. Leachates and other wastewater would be collected
and treated to meet standards, but leakage would be sufficient to cause direct toxic effects to fish in up
to 83 km of stream and indirect effects due to loss of invertebrate food species in up to 130 km of stream
(Table 14-2). In addition, the temperature and distribution of effluents could further degrade habitat.
Streams between the transportation corridor and Iliamna Lake would receive silt and deicing chemicals,
which would reduce habitat quality.
Table 14 2. Summary of estimated effects on streams under the three mine scenarios, assuming
routine operations.
Effect
Eliminated, blocked, or dewatered
Eliminated, blocked, or dewatered— anadromous
>20%flow reduction
Direct toxicity to fish
Direct toxicity to invertebrates
Downstream of transportation corridor
Stream Length Affected (km)
Pebble 0.25
38
8
15
0
15
290
Pebble 2.0
90
24
26
29
62
290
Pebble 6.5
145
35
54
57
83
290
This assessment considered failures of a tailings dam; product concentrate, return water, or diesel
pipelines; roads and culverts; and water collection and treatment. Tailings dam failures are improbable
in the sense of having a very low rate of occurrence, but some sort of failure becomes likely in the
extremely long-term. A tailings dam failure could destroy salmonid habitat in more than 30 km of the
North Fork Koktuli River and associated wetlands for years to decades. Product concentrate and diesel
pipeline failures near streams would be expected to occur during the life of a mine. Both would cause
acute lethal effects on invertebrates and fish, and the concentrate could create highly toxic sediment. A
truck wreck near a stream could introduce highly toxic chemicals causing acute lethality to fish and
invertebrates. Culvert failures would be common, unless a higher than usual maintenance program were
maintained, and could block fish passage and degrade downstream habitat. Failures to collect and treat
leachates and other wastewaters could cause releases ranging from short-term and innocuous to long-
term and highly toxic to fish and invertebrates.
14.8 Summary of Cumulative and Watershed-Scale Effects of
Multiple Mines
To provide reasonable realism and detail, this assessment largely addresses the potential effects of a
single mine, at three different sizes, on the Pebble deposit. However, the development of multiple mines,
of various sizes, in the Nushagak and Kvichak River watersheds is plausible. Several known mineral
deposits with potentially significant resources are located in the two watersheds, and active exploration
is underway at a number of claims blocks. The construction of roads, pipelines, and other infrastructure
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
14-17
April 2013
-------
Chapter 14 Integrated Risk Characterization
for one mine would likely facilitate the development of additional mines. Thus, the development of
multiple mines and their associated infrastructure may affect fish populations, wildlife, and Alaska
Native villages distributed across these watersheds.
Outside of Bristol Bay, most ecosystems that support Pacific salmon have been modified by the
cumulative effects of multiple land and water uses. Anadromous fish are particularly susceptible to
regional-scale effects, because they require suitable habitat in spawning areas, rearing areas, and along
migration corridors. Because Pacific salmon, Dolly Varden, and rainbow trout migrate among freshwater
habitats seasonally or between life stages, loss or degradation of habitat in one location can diminish the
ability of other locations to support these species. As a result of their particular susceptibility,
anadromous salmonid fisheries have declined in most of their range due to the combined effects of
habitat loss and degradation, pollution, and harvesting.
The Nushagak and Kvichak River watersheds have not yet experienced these cumulative stresses
associated with human activity, and their ecosystems are relatively undisturbed. Bristol Bay salmon
runs are resilient, because the abundance, diversity, and quality of Bristol Bay habitats result in large
and diverse salmon populations. Fluctuations in habitat availability or quality across the watersheds
caused by natural processes typically result in temporary loss or reduction in a discrete portion of
habitat, but are easily absorbed by Bristol Bay's diverse salmon populations. In contrast, the effects of
mining may be long-lasting and extensive, eliminating habitat for extended periods, and potentially
killing or otherwise eliminating populations offish. Such effects may remove component populations
permanently or for long periods of time, weakening the overall population's ability to absorb and
rebound from disturbance.
To examine the potential cumulative impacts from multiple mines, we consider development of
additional mines at the Pebble South/PEB, Big Chunk North, Big Chunk South, Groundhog, Humble, and
AUDN/Iliamna prospects. The AUDN/Iliamna and Humble prospects are located approximately 90 km
and 135 km, respectively, southwest of the Pebble deposit. All of the other prospects are within 25 km of
the Pebble deposit and may be of the same geologic origin. Construction of mining infrastructure at the
Pebble deposit would substantially reduce development costs for surrounding prospects and could
facilitate creation of a mining district that could include these sites.
The impacts from mine footprints and from accidents and failures associated with mine components
would be similar to those projected under the mine scenarios. The footprints would eliminate
substantial amounts of stream and wetland habitat, both directly and through dewatering. Total stream
length eliminated by these footprints would range from 41 to 64 km, and wetland area lost would range
from 7.4 to 2 5 km2. Further habitat loss and degradation would result from flow alteration. Each
additional mine will increase flow alteration from water removal and retention, increased impervious
surface and road crossings.
The consequences of leachate collection or treatment failure would depend on the chemical nature of
the rock or tailings over which it flows. Because porphyry copper deposits tend to straddle the threshold
between acid-generating and non-acid-generating, some of the waste rock and a portion of the tailings at
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14-18April 2013
-------
Chapter 14 Integrated Risk Characterization
any of these additional mines would be reasonably likely to be acid-generating. Each additional facility
would increase the likelihood of collection and treatment failures, which would increase the frequency
of discharge of untreated leachate or other wastewater in the Nushagak and Kvichak River watersheds,
with each event resulting in an increment of impact. Longer roads and pipelines associated with
additional mines, coupled with a greater number of stream crossings, would increase the frequency of
events such as culvert failures, pipeline breaks, and truck accidents that would damage aquatic systems,
incrementally decreasing habitat value over an extensive area. In the long-term, cessation of
maintenance and treatment would likely result in the degradation of fisheries in waters downstream of
each mine in the Nushagak and Kvichak River watersheds. Extreme natural events such as earthquakes
and floods could cause failures of dams, roads, pipelines, or WWTPs at multiple mines.
Induced development is that which results from the introduction of industry, roads and infrastructure. It
is reasonably foreseeable that infrastructure from large-scale mining in the Nushagak and Kvichak River
watersheds, particularly the transportation corridors, could induce further development in the region.
Existing communities, the tourism industry, and the recreational housing market could benefit if large-
scale mining expanded through the watersheds. Unregulated access to currently roadless wilderness
areas also could expand. The improved access would increase hunting and fishing pressure, as well as
competition with existing subsistence users; increase damage from off-road vehicle, boat, and foot traffic
to currently inaccessible areas; facilitate poaching, dumping, trespassing, and other activities; and lead
to scattered development in the watersheds.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 14-19 April 2013
-------
15.1 References by Chapter
15.1.1 Chapter 1—Introduction
PLP (Pebble Limited Partnership). 2013. About Pebble. Available:
http://corporate.pebblepartnership.com/about. Accessed: April 1, 2013.
USEPA (U.S. Environmental Protection Agency). 2002a. Clinch and Powell Valley Watershed Ecological
Risk Assessment. Washington, DC: Office of Research and Development, National Center for
Environmental Assessment, Washington Office. EPA/600/R-01/050. Available:
.
USEPA (U.S. Environmental Protection Agency). 2002b. Ecological Risk Assessment for the Middle Snake
River, Idaho. Washington, DC: Office of Research and Development, National Center for
Environmental Assessment, Washington Office. EPA/600/R-01/017. Available:
.
USEPA (U.S. Environmental Protection Agency). 2002c. WaquoitBay Watershed Ecological Risk
Assessment: the Effect of Land-Derived Nitrogen Loads on Estuarine Eutrophication. EPA/600/R-
02/079. Available: .
USEPA (U.S. Environmental Protection Agency). 2008. Predicting Future Introductions of Non-indigenous
Species to the Great Lakes (Final Report). Washington, DC. EPA/600/R-08/066F. Available:
.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment
15-1
April 2013
-------
Chapter 15 References
USEPA (U.S. Environmental Protection Agency). 2010. The Effects of Mountaintop Mines and Valley Fills
on Aquatic Ecosystems of the Central Appalachian Coalfields (External Review Draft). Washington, DC.
EPA/600/R-09/138A. Available: .
15.1.2 Chapter 2—Overview of Assessment
Forman, R. T. T., and L. E. Alexander. 1998. Roads and their major ecological effects. Annual Review of
Ecology andSystematics 29:207-231.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska, February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document2004 through 2008,
Anchorage, AK. Available: .
USEPA (U.S. Environmental Protection Agency). 1998. Guidelines for Ecological Risk Assessment
EPA/63 0/R-95/002F. April. Washington, DC: Environmental Protection Agency, Office of Research
and Development
USFWS (U.S. Fish and Wildlife Service). 2012. National Wetland Inventory. Available:
. Accessed: December 11, 2012.
USGS (U.S. Geological Survey). 2012. National Hydrography Dataset, High Resolution, Alaska. Available:
ftp://nhdftp.usgs.gov/DataSets/Staged/States/FileGDB/HighResolution . Accessed: October 16,
2012.
15.1.3 Chapter 3—Region
ACIA (Arctic Climate Impact Assessment). 2004. Impacts of a Warming Arctic: Arctic Climate Impact
Assessment. Cambridge, UK: Cambridge University Press.
Armstrong, J. B., D. E. Schindler, K. L. Omori, C. P. Ruff, and T. P. Quinn. 2010. Thermal heterogeneity
mediates the effects of pulsed subsidies across a landscape. Ecology 91:1445-1454.
Barnett, T. P., J. C. Adam, and D. P. Lettenmaier. 2005. Potential impacts of a warming climate on water
availability in snow-dominated regions. Nature 438:303-309.
Battin, J., M. W. Wiley, M. H. Ruckelshaus, R. N. Palmer, E. Korb, K. K. Bartz, and H. Imaki. 2007. Projected
impacts of climate change on salmon habitat restoration. Proceedings of the National Academy of
Sciences of the United States of America 104(16):6720-6725.
Beer, W. N. and J. J. Anderson. 2001. Effect of spawning day and temperature on salmon emergence:
interpretations of a growth model for Methow River chinook. Canadian Journal of Fisheries and
Aquatic Sciences 58:943-949.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment .^2 April 2013
-------
Chapter 15 References
Berggren, T. J. and M. J. Filardo. 1993. An analysis of variables influencing the migration of juvenile
salmonids in the Columbia River basin. North American Journal of Fisheries Management 13:48-63.
Bjornn, T. C. and D. W. Reiser. 1991. Habitat Requirements of Salmonids in Streams. Pages 83-138 in W.
R. Meehan (ed.), Influences of Forest and Rangeland Management on Salmonid Fishes and their
Habitats. Bethesda, MD: American Fisheries Society.
Bryant, M. 2009. Global climate change and potential effects on Pacific salmonids in freshwater
ecosystems of southeast Alaska. Climatic Change 95:169-193.
Burgner, R. L. 1991. Life history of sockeye salmon (Oncorhynchus nerka]. Pages 1-118 in C. Grootand L.
Margolis, (eds.), Pacific Salmon Life Histories. University of Washington Press, Seattle, WA.
Burnett, K. M., G. H. Reeves, D. J. Miller, S. Clarke, K. Vance-Borland, and K. Christiansen. 2007.
Distribution of salmon-habitat potential relative to landscape characteristics and implications for
conservation. Ecological Applications 17(1):66-80.
Christensen, J. H., B. Hewitson, A. Busuioc, A. Chen, X. Gao, I. Held, R. Jones, R. K. Kolli, W.-T. Kwon, R.
Laprise, V. Magaa Rueda, L. Mearns, C. G. Menndez, J. Risnen, A. Rinke, A. Sarr and P. Whetton, 2007.
Regional Climate Projections. In Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.
Tignor and H.L. Miller (eds.), Climate Change 2007: The Physical Science Basis. Contribution of
Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
Cambridge, UK and New York, NY, USA: Cambridge University Press.
CRU (Climatic Research Unit). 2012. Data. University of East Anglia, UK. Available:
.
Demory, R. L., R. L. Orrell, and D. R. Heinle. 1964. Spawning Ground Catalog of the Kvichak River System,
Bristol Bay, Alaska. Issue 488 of Special Scientific Report—Fisheries. Washington, DC: U.S.
Department of Interior, Bureau of Commercial Fisheries.
Detterman, R. L. 1986. Glaciation of the Alaska Peninsula. Pages 151-170 in Hamilton, T. D., K. M. Reed,
and R. M. Thorson (eds.), Glaciation in Alaska. Alaska Geological Society.
Detterman, R. L. and B. L. Reed. 1973. Surficial Deposits of the Iliamna Quadrangle, Alaska. Geological
Survey Bulletin 1368-A. U.S. Geological Survey. Available:
. Accessed: December 17, 2012.
Eaton, J. G. and R. M. Scheller. 1996. Effects of climate warming on fish thermal habitatin streams of the
United States. Limnology and Oceanography 41:1109-1115.
Eiler, J. H., B. D. Nelson, R. F. 1992. Riverine spawning by sockeye salmon in the Taku River, Alaska and
British Columbia. Transactions of the American Fisheries Society 121(6):701-708.
Feddema, J. J. 2005. A revised Thornthwaite-type global climate classification. Physical Geography
26:442-466.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T April 2013
-------
Chapter 15 References
Gallant, A. L, E. F. Binnian, J. M. Omernik, M. B. Shasby. 1995. Ecoregions of Alaska, U.S. Geological Survey
Professional Paper 1567. Washington DC.
Gesch, D., M. Oimoen, S. Greenlee, C. Nelson, M. Steuck, D. Tyler. 2002. The National Elevation Dataset.
Photogrammatic Engineering and Remote Sensing 68(1):5-11.
Gesch, D. B. 2007. Chapter 4-The National Elevation Dataset. Pages 99-118 in Maune, D. (ed.), Digital
Elevation Model Technologies and Applications: The DEM Users Manual. 2nd edition. Bethesda, MD:
American Society for Photogrammetry and Remote Sensing.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska, February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
Haeussler, P. J. and R. W. Saltus. 2004.26 km Offset on the Lake Clark Fault Since Late Eocene Time.
Studies by the U.S. Geological Survey in Alaska, 2004. U.S. Geological Survey Professional Paper
1709-A. 4 pages.
Haeussler, P. J. and C. F. Waythomas. 2011. Review of the Origin of the Braid Scarp Near the Pebble
Prospect, Southwestern Alaska. U.S. Geological Survey Open File Report2011-1028.14 pages.
Hamon, W. R. 1961. Estimating potential evapotranspiration./ourno/ of the Hydraulics Division,
Proceedings of the American Society of Civil Engineers 87: 107-120.
Hauer, F. R., J. S. Baron, D. H. Campbell, K. D. Fausch, S. W. Hostetler, G. H. Leavesley, P. R. Leavitt, D. M.
McKnight, and J. A. Stanford. 1997. Assessment of climate change and freshwater ecosystems of the
Rocky Mountains, USA and Canada. Hydrological Processes 11:903-924.
Healey, M. 2011. The cumulative impacts of climate change on Fraser River sockeye salmon
(Oncorhynchus nerka) and implications for management. Canadian Journal of Fisheries and Aquatic
Sciences 68:953-953.
Heming, T. A., J. E. Mclnerney, and D. F. Alderdice. 1982. Effect of temperature on initial feeding in
alevins of Chinook salmon (Oncorhynchus tshawytscha]. Canadian Journal of Fisheries and Aquatic
Sciences 39:1554-1562.
Hilborn, R., T. Quinn, D. Schindler, and D. Rogers. 2003. Biocomplexity and fisheries sustainability.
Proceedings of the National Academy of Sciences of the United States of America 100:6564-6568.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment . c_4 APril 2013
-------
Chapter 15 References
Hinzman, L. D., N. D. Bettez, W. R. Bolton, F. S. Chapin, M. B. Dyurgerov, C. L. Fastie, B. Griffith, R. D.
Hollister, A. Hope, H. P. Huntington, A. M. Jensen, G. J. Jia, T. Jorgenson, D. L. Kane, D. R. Klein, G.
Kofinas, A. H. Lynch, A. H. Lloyd, A. D. McGuire, F. E. Nelson, W. C. Oechel, T. E. Osterkamp, C. H.
Racine, V. E. Romanovsky, R. S. Stone, D. A. Stow, M. Sturm, C. E. Tweedie, G. L. Vourlitis, M. D.
Walker, D. A. Walker, P. J. Webber, J. M. Welker, K. S. Winker, and K. Yoshikawa. 2005. Evidence and
implications of recent climate change in Northern Alaska and other Arctic regions. Climatic Change
72:251-298.
Hughes, N. F. and L. M. Dill. 1990. Position choice by drift-feeding salmonids: model and test for Arctic
grayling (Thymallus arcticus) in subarctic mountain streams, Interior Alaska. Canadian Journal of
Fisheries and Aquatic Sciences 47: 2039-2048.
Irvine, J. R., M. Fukuwaka, T. Kaga, J. H. Park, K. B. Seong, S. Kang, V. Karpenko, N. Klovach, H. Bartlett, and
E. Volk. 2009. Pacific Salmon Status and Abundance Trends. North Pacific Anadromous Fish
Commission, Document 1199, Rev. 1.153 pages.
Johnson, J. and P. Blanche. 2012. Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes—Southwestern Region, Effective June 1,2012. Special Publication No. 12-08.
Anchorage, AK: Alaska Department of Fish and Game.
Kisslinger, C. 1976. A review of theories of mechanisms of induced seismicity. Engineering Geology
10:85-98.
Koehler, R. D. and R. D. Reger. 2011. Reconnaissance Evaluation of the Lake Clark Fault, TyonekArea,
Alaska. Alaska Division of Geological & Geophysical Surveys Preliminary Interpretative Report 2011-
1. 8 pages.
Kovach, R. P., A. J. Gharrett, and D. A. Tallmon. 2012. Genetic change for earlier migration timing in a pink
salmon population. Proceedings of the Royal Society B-Biological Sciences 279:3870-3878.
Lawson, P. W., E. A. Logerwell, N. J. Mantua, R. C. Francis, and V. N. Agostini. 2004. Environmental factors
influencing freshwater survival and smolt production in Pacific Northwest coho salmon
(Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Sciences 61:360-373.
Lea, P. D. and C. F. Waythomas. 1990. Late-Pleistocene eolian sand sheets in Alaska. Quaternary Research
34:269-281.
Lea, P.O., S. A. Elias, and S. K. Short. 1991. Stratigraphy and paleoenvironments of Pleistocene nonglacial
deposits in the southern Nushagak lowland, southwestern Alaska, USA. Arctic and Alpine Research
23:375-391.
Lisi, P. J., D. E. Schindler, K. T. Bentley, and G. R. Pess. 2012. Association between geomorphic attributes
of watersheds, water temperature, and salmon spawn timing in Alaskan streams. Geomorphology
185:78-86.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment .^5 April 2013
-------
Chapter 15 References
Lisle, T. 1989. Sediment transport and resulting deposition in spawning gravels, North Coastal
California. Water Resources Research 25:1303-1319.
Listen, G. E., and C. A. Hiemstra. 2011. The changing cryosphere: pan-arctic snow trends (1979-2009).
Journal of Climate 24(2iy.5691-5712.
Luck, M., N. Maumenee, D. Whited, J. Lucotch, S. Chilcote, M. Lorang, D. Goodman, K. McDonald, J. Kimball,
and J. Stanford. 2010. Remote sensing analysis of physical complexity of North Pacific Rim rivers to
assist wild salmon conservation. Earth Surface Processes and Landforms 35:1330-1343.
Mantua, N., I. Tohver, and A. Hamlet. 2010. Climate change impacts on streamflow extremes and
summertime stream temperature and their possible consequences for freshwater salmon habitat in
Washington State. Climatic Change 102:187-223.
Markon, C. J., S. F. Trainor, and F. S. Chapin III, eds. 2012. The United States National Climate Assessment
- Alaska Technical Regional Report. U.S. Geological Survey Circular 1379. Reston, VA. 148 pages.
Martins, E. G., S. G. Hinch, S. J. Cooke, and D. A. Patterson. 2012. Climate effects on growth, phenology,
and survival of sockeye salmon (Oncorhynchus nerka]: a synthesis of the current state of knowledge
and future research directions. Reviews in Fish Biology and Fisheries 22:887-914.
Mellina, E., R. D. Moore, S. G. Hinch, J. S. Macdonald, and G. Pearson. 2002. Stream temperature responses
to clearcut logging in British Columbia: The moderating influences of groundwater and headwater
lakes. Canadian Journal of Fisheries and Aquatic Sciences 59:1886-1900.
Montgomery, D. R., and J. M. Buffington. 1997. Channel-reach morphology in mountain drainage basins.
Bulletin of the Geological Society of America 109(5):596-611.
Montgomery, D. R., E. M. Beamer, G. R. Pess, and T. P. Quinn. 1999. Channel type and salmonid spawning
distribution and abundance. Canadian Journal of Fisheries and Aquatic Sciences 56(3):377-387.
Montgomery, D. R., J. M. Buffington, N. P. Peterson, D. Schuett-Hames, and T. P. Quinn. 1996. Stream-bed
scour, egg burial depths, and the influence of salmonid spawning on bed surface mobility and
embryo survival. Canadian Journal of Fisheries and Aquatic Sciences 53(5): 1061-1070.
Mote, P. W., E. Parson, A. F. Hamlet, W. S. Keeton, D. Lettenmaier, N. Mantua, E. L. Miles, D. Peterson, D. L.
Peterson, R. Slaughter, and A. K. Snover. 2003. Preparing for climatic change: The water, salmon, and
forests of the Pacific Northwest. Climatic Change 61:45-88.
Nehlsen, W., J. Williams, and J. Lichatowich. 1991. Pacific salmon at the crossroads—stocks at risk from
California, Oregon, Idaho, and Washington. Fisheries 16:4-21.
Parks, B. and R. J. Madison. 1985. Estimation of selected flow and water-quality characteristics of Alaskan
streams. U.S. Department of Interior, U.S. Geological Survey.
Peacock, S. 2012. Projected twenty-first century change in temperature, precipitation, and snow cover
over North America in CCSM4Journo/ of Climate 25(13):4405-4429.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T April 2013
-------
Chapter 15 References
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008.
Anchorage, AK. Available: .
Power, G., R. S. Brown, and J. G. Imhof. 1999. Groundwater and fish—Insights from northern North
America. HydrologicalProcesses 13:401-422.
PRISM Climate Group. 2012. Oregon State University. Available: .
Quinn, T. P. 2005. The Behavior and Ecology of Pacific Salmon and Trout. Seattle, WA: University of
Washington Press.
Rains, M. C. 2011. Water sources and hydrodynamics of closed-basin depressions, Cook Inlet Region,
Alaska. Wetlands 31:377-387.
Rich, H. B., Jr., T. P. Quinn, M. D. Scheuerell, D. E. Schindler. 2009. Climate and intraspecific competition
control the growth and life history of juvenile sockeye salmon (Oncorhynchus nerka] in Iliamna Lake,
Alaska. Canadian Journal of Fisheries and Aquatic Sciences 66(2):238-246.
Rogers, L. A. and D. E. Schindler. 2011. Scale and the detection of climatic influences on the productivity
of salmon populations. Global Change Biology 17:2546-2558.
Ruff, C., D. Schindler, J. Armstrong, K. Bentley, G. Brooks, G. Holtgrieve, M. McGlauflin, J. Seeb, and C.
Torgersen. 2011. Temperature-associated population diversity in salmon confers benefits to mobile
consumers. Ecology 92:2073-2084.
Salathe, E. P. 2006. Influences of a shift in North Pacific storm tracks on western North America
precipitation under global warming. Geophysical Research Letters 33(19):L19820.
Schindler, D. E., R. Hilborn, B. Chasco, C. P. Boatright, T. P. Quinn, L. A. Rogers, and M. S. Webster. 2010.
Population diversity and the portfolio effect in an exploited species. Nature 465:609-612.
Selkregg, L. L. 1974. Southwest Alaska, Alaska Regional Profiles. Anchorage, AK: University of Alaska,
Arctic Environmental Information and Data Center.
Serreze, M. C., J. E. Walsh, F. S. Chapin III, T. Osterkamp, M. Dyurgerov, V. Romanovsky, W. C. Oechel, J.
Morison, T. Zhang, and G. Barry. 2000. Observational evidence of recent change in the northern high-
latitude environment. Climatic Change 46(1-2):159-207.
Shallin Busch, D., M. Sheer, K. Burnett, P. McElhany, and T. Cooney. 2011. Landscale-level model to
predict spawning habitat for lower Columbia River fall chinook salmon (Oncorhynchus tshawytscha}.
River Research and Applications 29(3):297-312.
SNAP (Scenarios Network for Alaska and Arctic Planning). 2012. Historical and Projected Decadal
Averages of Annual Total and Annual Mean Temperature, Precipitation, and Evapotranspiration.
Available: . Accessed: February 7, 2013.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T April 2013
-------
Chapter 15 References
Stafford, J. M., G. Wendler, and J. Curtis. 2000. Temperature and precipitation of Alaska: 50 year trend
analysis. Theoretical and Applied Climatology 67(1-2) :3 3-44.
Stanford, J. A., M. S. Lorang, and F. R. Hauer. 2005. The shifting habitat mosaic of river ecosystems.
International Association of Theoretical and Applied Limnology—Proceedings 29:123-136.
Steen, R. P., and T. P. Quinn. 1999. Egg burial depth by sockeye salmon (Oncorhynchus nerka):
implications for survival of embryos and natural selection on female body size. Canadian Journal of
Zoo/ofly77(5):836-841.
Stewart, I. T., D. R. Cayan, and M. D. Dettinger. 2004. Changes in snowmelt runoff timing in western
North America under a 'business as usual' climate change scenario. Climatic Change 62:217-232.
Stilwell, K. B. and D. S. Kaufman. 1996. Late Wisconsin glacial history of the northern Alaska Peninsula.
Arctic and Alpine Research 28: 475-487.
Tang, J., M. D. Bryant, and E. L. Brannon. 1987. Effect of temperature extremes on the mortality and
development rates of coho salmon embryos and alevins. Progressive Fish-Culturist 49:167-174.
Taylor, E. B. 1990. Environmental correlates of life-history variation in juvenile Chinook salmon,
Oncorhynchus tshawytscha (WalbaumJ./ourno/ of Fish Biology 37:1-17.
USGS (U.S. Geological Survey). 2010. Significant United States Earthquakes, 1568-2009. June. Available:
. Accessed: October 16,
2012.
Wahrhaftig, C. 1965. Physiographic Divisions of Alaska. U.S. Geological Survey, Geological Survey
Professional Paper 482.
Walsh, J. E., W. L. Chapman, V. Romanovsky, J. H. Christensen, and M. Stendel. 2008. Global climate model
performance over Alaska and Greenland. Journal of Climate 21:6156-6174.
Wendler, G., L. Chen, and B. Moore. 2012. The first decade of the new century: A cooling trend for most of
Alaska. The Open Atmospheric Science Journal 6.
Winter, T. C. 2001. The concept of hydrologic landscapes. Journal of the American Water Resources
Association 37:335-349.
Wissmar R. C., R. K. Timm, M. D. Bryant. 2010. Radar-derived digital elevation models and field-surveyed
variables to predict distributions of juvenile coho salmon and Dolly Varden in remote streams of
Alaska. Transactions of the American Fisheries Society 139:288-302.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7TT April 2013
-------
Chapter 15 References
Zamzow, K. 2011. Investigations of Surface Water Quality in the Nushagak, Kvichak, and Chulitna
Watersheds, Southwest Alaska 2009-2010. Sutton, AK: Center for Science in Public Participation.
Personal Communications
Wiedmer, Michael. Malma Consulting. Conversation with Joe Ebersole in Sept 2011, about distribution of
salmon across the Nushagak and Kvichak River watersheds and factors controlling that distribution.
15.1.4 Chapter 4—Type of Development
ADNR (Alaska Department of Natural Resources). 2012. Alaska Mapper, Mineral Estate Map. Available:
. Accessed: April 17, 2012.
Anderson, E. D., R. G. Eppinger, K. D. Kelley. 2009. Using regional geochemistry, geology, aeromagnetics,
Landsat, and digital elevation models (DEM) to define favourable areas for porphyry-style
mineralization in southwestern Alaska. Pages 345-348 in D. R. Lentz, K. G. Thorne, K. L. Beal (eds.),
24th International Applied Geochemistry Symposium, Fredericton, New Brunswick, Canada.
Association of Applied Geochemists.
Blight, G. E. 2010. Geotechnical Engineering for Mine Waste Storage Facilities. Boca Raton, FL: CRC Press.
Borden, R. K. 2003. Environmental geochemistry of the Bingham Canyon porphyry copper deposit, Utah.
Environmental Geology 43(7):752-758.
Brodie, M.}., L. M. Broughton, A. M. Robertson. 1991. A Conceptual Rock Classification System for Waste
Management and a Laboratory Method for ARD Prediction from Rock Piles. Preprint for the 2nd
International Conference on the Abatement of Acid Drainage, Montreal, Quebec. September 16-18,
1991.17 pages.
Commonwealth of Australia. 2007. Tailings Management - Leading Practice Sustainable Development
Program for the Mining Industry. Australian Government, Department of Industry, Tourism and
Resources. 88 pages.
Davies, M. P. 2002. Tailings impoundment failures: Are geotechnical engineers listening? Geotechnical
News. September:31-36.
Davies, M. P. 2011. Filtered dry stacked tailings - the fundamentals. Pages 189-197 in D. van Syle (ed.),
Tailings and Mine Waste '11: Proceedings of the 15th International Conference on Tailings and Mine
Waste, Vancouver, British Columbia, Canada, November 6-9, 2011.
Fuerstenau, M. C., G. Jameson, R. H. Yoon (eds.). 2007. Froth Flotation: A Century of Innovation. Littleton,
CO: Society of Mining, Metallurgy, and Exploration. 891 pages
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska. February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 7T7 April 2013
-------
Chapter 15 References
ICOLD (International Commission on Large Dams). 2001. Tailings Dams, Risk of Dangerous Occurrences,
Lessons Learnt from Practical Experiences. United Nations Environmental Programme, Bulletin 121.
John, D. A., R. A. Ayuso, M. D. Barton, R. J. Blakely, R. J. Bodnar, J. H. Dilles, F. Gray, F. T. Graybeal, J. C.
Mars, D. K. McPhee, R. R. Seal, R. D. Taylor, P. G. Vikre. 2010. Porphyry Copper Deposit Model. Chapter
B in Mineral Deposit Models for Resource Assessments. U. S. Geological Survey Investigations Report
2010-5070-B. Reston, VA: U.S. Geological Survey.
Kesler, S. E., S. L. Chryssoulis, G. Simon. 2002. Gold in porphyry copper deposits: Its abundance and fate.
Ore Geology Reviews 21(1-2)\W3-124.
Koerner, R. M., Y. G. Hsuan, and G. R. Koerner. 2011. GRI White Paper #6 on Geomembrane Lifetime
Prediction: Unexposed and Exposed Conditions, Updated 2011. Available: . Accessed: April 20, 2012.
Levit, S. and D. M. Chambers. 2012. Comparison of the Pebble Mine with Other Alaska Large Hard Rock
Mines. Sutton, AK: Center for Science in Public Participation.
Lipman, P. W. and D. A. Sawyer. 1985. Mesozoic ash-flow caldera fragments in southeastern Arizona and
their relation to porphyry copper deposits. Geology 13:652-656.
Logsdon, M. J., K. Hagelstein, T. I. Mudder. 1999. The Management of Cyanide in Gold Extraction. Ottawa,
Ontario, Canada: International Council on Metals and the Environment. 44 pages.
Lusty, P. A. J. and S. D. Hannis. 2009. Commodity Profile: Copper. British Geological Survey Internal
Report OR/09/041. Keyworth, Nottingham.
Marsden, J. 0. and C. I. House. 2006. The Chemistry of Gold Extraction. 2nd Edition. Littleton, CO: Society
of Mining, Metallurgy, and Exploration. 651 pages.
Martin, T. E., M. P. Davies, S. Rice, T. Higgs, P. C. Lighthall. 2002. Stewardship of Tailings Facilities.
Commissioned by the MMSD project of the International Institute for the Environment and
Development (IIEA) and the World Business Council for Sustainable Development (WBCSD),
London, England. 37 pages.
Novagold. 2012. Donlin Creek Gold Project Technical Report. January 2 0.
O'Kane, M. and C. Wels. 2003. Mine Waste Cover System Design - Linking Predicted Performance to
Groundwater and Surface Water Impacts. Pages 341-349 in Proceedings of the 6th Annual
International Conference on Acid Rock Drainage, July 12-18, Cairns, Queensland, Australia.
Porter, K. E. and D. I. Bleiwas. 2003. Physical Aspects of Waste Storage from a Hypothetical Open Pit
Porphyry Copper Operation. Department of the Interior, U.S. Geological Survey Open File Report 03-
143, Version 1.0.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-10 April 20:L3
-------
Chapter 15 References
Price, W. A. 2009. Prediction Manual for Drainage Chemistry from Sulphidic Geologic Materials, 579.
Report prepared by CANMET - Mining and Mineral Sciences Laboratories for the Mine Environment
Neutral Drainage (MEND) Program, Natural Resources Canada.
Price, W. A. and J. C. Errington. 1998. Guidelines for Metal Leaching and Acid Rock Drainage at Mine Sites
in British Columbia. British Columbia, Canada: British Columbia Ministry of Energy and Mines. 92
pages.
Rowe, R. K. 2005. Long-term performance of contaminant barrier systems. Geotechnique 55(9):631-678.
Available: . Accessed:
September 18, 2011.
USAGE (U.S. Army Corps of Engineers). 2011. Permit # POA-1992-574-M19, Fish Creek. Issued March 4,
2011.
USEPA (U.S. Environmental Protection Agency). 1994. Technical Report: Design and Evaluation of
Tailings Dams. Washington, DC: Office of Solid Waste. EPA 530-R-94-038.
Vick, S. G. 1983. Planning, Design, and Analysis of Tailings Dams. New York, NY: John Wiley & Sons.
15.1.5 Chapter 5—Endpoints
ADF&G (Alaska Department of Fish and Game). 2012. Alaska Freshwater Fish Inventory Database.
Available: .
Alaska Department of Health and Social Services. 2005. The Prevalence of Diabetes Among Adult
Alaskans, 2002-2004. Epidemiology Bulletins 1-8.
Alaska Federation of Natives. 2010. The Right to Subsist—Federal Protection of Subsistence in Alaska.
Available: .
Armstrong, R. H. and J. E. Morrow. 1980. The Dolly Varden Charr, Salvelinus malma. Pages 99-140 in E.
K. Balon (ed.), Charrs: Salmonid fishes of the genus Salvelinus. Springer.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-11April 2013
-------
Chapter 15 References
Augerot,X. 2005. Atlas ofPacific Salmon: The First Map-Based Status Assessment of Salmon in the North
Pacific. Portland, OR.
Baker, T. T., L. F. Fair, R. A. Clark, and J. J. Hasbrouck. 2009. Review of Salmon Escapement Goals in Bristol
Bay, Alaska, 2009. Alaska Department of Fish and Game, Division of Sport Fish, Research and
Technical Services.
Ballew, C., A. Ross, R. S. Wells, V. Hiratsuka, K. J. Hamrick, E. D. Nobmann, and S. Bartell. 2004. Final
Report on the Alaska Traditional Diet Survey. Anchorage, AK.
Bartz, K. K. and R. J. Naiman. 2005. Effects of salmon-borne nutrients on riparian soils and vegetation in
Southwest Alaska. Ecosystems 8:529-545.
Bersamin, A., S. Zidenberg-Cherr, J. Stern, and B. Luick. 2007. Nutrient intakes are associated with
adherence to a traditional diet among Yup'ik Eskimos living in remote Alaska native communities:
The CANHR Study. International Journal of Circumpolar Health 66:62-70.
Blair, G. B., D. E. Rogers, and T. P. Quinn. 1993. Variation in life history characteristics and morphology of
sockeye salmon in the Kvichak River system, Bristol Bay, Alaska. Transactions of the American
Fisheries Society 122:550-559.
Brna, P. J. and L. A. Verbrugge (eds.). 2013. Wildlife Resources of the Nushagak and Kvichak River
Watersheds, Alaska. Final Report. Anchorage, AK: Anchorage Fish and Wildlife Field Office, U.S. Fish
and Wildlife Service. 177 pages.
Burger, C. V. and L. A. Gwartney. 1986. A Radio Tagging Study ofNaknek Drainage Rainbow Trout.
Anchorage, AK: U.S. National Park Service, Alaska Regional Office.
Butler, L. G. 2004. Unit 9 moose management report. Pages 113-120 in C. Brown, (ed.), Moose
Management Report of Survey and Inventory Activities: 1 July 2001-30 June 2003. Juneau, AK: Alaska
Department of Fish and Game.
Butler, L.G. 2009. Unit 9 and 10 Wolf Management Report. Pages 66-70 in P. Harper (ed.), Wolf
Management Report of Survey and Inventory Activities 1 July 2 005-30 June 2008. Juneau, AK: Alaska
Department of Fish and Game.
Cederholm C. J., M. D. Kunze, T. Murota, A. Sibatani 1999. Pacfic salmon carcasses: essential contributors
of nutrients and energy for aquatic and terrestrial ecosystems. Fisheries 24:6-15.
Chan, H., K. Fediuk, S. Hamilton, L. Rostas, H. Caughey, G. Egeland, and E. Loring. 2006. Food security in
Nunavut, Canada: Barriers and recommendations. International Journal of Circumpolar Health
65:416-431.
Christie, K. S. and T. E. Reimchen. 2008. Presence of salmon increases passerine density on Pacific
Northwest streams. Auk 125:51-59.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-12April 2013
-------
Chapter 15 References
Colombi, B. J. and J. F. Brooks. 2012. Keystone Nations: Indigenous Peoples and Salmon Across the North
Pacific. Santa Fe, NM: School for Advanced Research.
Crane, P., M. Lisac, B. Spearman, E. Kretschmer, C. Lewis, S. Miller, and J. Wenburg. 2003. Development
and Application of Microsatellites to Population Structure and Mixed-Stock Analyses of Dolly Varden
from the Togiak River Drainage, Final Report (Study No. 00-011). Anchorage AK: U.S. Fish and
Wildlife Service, Conservation Genetics Laboratory.
Cunningham, C. J., R. Hilborn, J. Seeb, and T. Branch. 2012. Reconstruction of Bristol Bay Sockeye Salmon
Returns Using Age and Genetic Composition of Catch. University of Washington, School of Aquatic and
Fishery Sciences: AFS-UW-1202.
Darimont, C. T., T. E. Reimchen, and P. C. Paquet. 2003. Foraging behavior by gray wolves on salmon
streams in coastal British Columbia. Canadian Journal of Zoology 81:349-353.
DeCicco, A. L. 1992. Long-distance movements of anadromous Dolly Varden between Alaska and the
USSRMrct/c45(2):120-123.
DeCicco, A. L. 1997. Movements of postsmolt anadromous Dolly Varden in northwestern Alaska.
American Fisheries Symposium 19:175-183.
Denton, K. P., H. B. Rich Jr., and T. P. Quinn. 2009. Diet, movement, and growth of Dolly Varden in
response to sockeye salmon subsidies. Transactions of the American Fisheries Society 138(6):1207-
1219.
Dewailly, E., C. Blanchet, S. Lemieux, L. Sauve, S. Gingras, P. Ayotte, and B. J. Holub. 2001. N-3 Fatty acids
and cardiovascular disease risk factors among the Inuit of Nunavik. The American Journal of Clinical
Nutrition 74:464-473.
Dewailly, E., C. Blanchet, S. Gingras, S. Lemieux, and B. J. Halub. 2002. Cardiovascular disease risk factors
and n-3 fatty acid status in the adult population of James Bay Cree. The American Journal of Clinical
Nutrition 76:85-92.
Dittman, A. H. and T. P. Quinn. 1996. Homing in Pacific salmon: Mechanisms and ecological basis. Journal
of Experimental Biology 199:83-91.
Duhaime, G. and N. Bernard (eds.). 2008. Arctic Food Security. Canadian Circumpolar Institute (CCI)
Press, University of Alberta and Centre inter-universitaire d'etudes etde recherche autochtones
(CIERA), Universite Laval.
Dye, J. E. and C. J. Schwanke. 2009. Report to the Alaska Board of Fisheries for the Recreational Fisheries of
Bristol Bay, 2007, 2008, and 2009. Alaska Department of Fish and Game, Division of Sport Fish,
Research and Technical Services.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-13 April 2013
-------
Chapter 15 References
Dye, J. E., C. J. Schwanke, and T. A. Jaecks. 2006. Report to the Alaska Board of Fisheries for the
Recreational Fisheries of Bristol Bay, 2004, 2005, and 2006. Special Publication No. 06-29. Alaska
Department of Fish and Game, Divisions of Sport Fish and Commercial Fisheries.
Ebbesson, S. 0. E. and M. T. Tejero. 2007. Fatty acid consumption and metabolic syndrome components:
The GOCADAN study. Journal of the Cardiometabolic Syndrome 2(4):244-2 49.
Eliason, E., T. Clark, M. Hague, L. Hanson, Z. Gallagher, K. Jeffries, M. Gale, D. Patterson, S. Hinch, and A.
Farrell. 2011. Differences in thermal tolerance among sockeye salmon populations. Science
332:109-112.
Fair, L. E. 2003. Critical elements of Kvichak River sockeye salmon management. Alaska Fishery Research
Bulletin 10:95-103.
Fair, L. F., C. E. Barzil, X. Zyhang, R. A. Clark, and J. W. Erickson. 2012. Review of Salmon Escapement Goals
in Bristol Bay, Alaska, 2012. Anchorage, AK: Alaska Department of Fish and Game, Fishery
Manuscript Series No. 12-0.
Fall, J. A., D. L. Holen, B. Davis, T. Krieg, and D. Roster. 2006. Subsistence Harvests and Uses of Wild
Resources in Iliamna, Newhalen, Nondalton, Pedro Bay, and Port Alsworth, Alaska 2004. Technical
Paper No. 302. Juneau, AK: Alaska Department of Fish and Game, Division of Subsistence.
Fall, J. A., T. M. Krieg, and D. Holen. 2009. Overview of the Subsistence Fishery of the Bristol Bay
Management Area. Special Publication No. EOF 2009-07. Anchorage AK: Alaska Department of Fish
and Game, Division of Subsistence.
Fancy, S. G. 1983. Movements and activity budgets of caribou near oil drilling sites in the Sagavanirktok
River Floodplain, Alaska. Arctic 36:193-197.
Gende, S. M., M. F. Willson. 2001. Passerine densities in riparian forests of Southeast Alaska: potential
effects of anadromous spawning salmon. Condor 103:624-629.
Gende, S. M., R. T. Edwards, M. F. Willson, M. S. Wipfli. 2002. Pacific salmon in aquatic and terrestrial
ecosystems. BioScience 52(10):917-928.
Gerrard J. M., Gerrard P., Maher W. J., Whitfield D. W. A. 1975. Factors influencing nest site selection of
Bald Eagles in northern Saskatchewan and Manitoba. Blue Jay 33:169-176.
Goldsmith, D., A. Hill, T. Hull, M. Markowski, R. Unswort. 1998. Economic Assessment of Bristol Bay Area
National Wildlife Refuges: Alaska Peninsula, Becherof Izembek, Togiak. U.S. Department of the
Interior, Fish and Wildlife Service.
Gresh, T., J. Lichatowich, P. Schoonmaker. 2000. An estimation of historic and current levels of salmon
production in the northwest Pacific ecosystem: evidence of a nutrient deficit in the freshwater
systems of the Pacific Northwest. Fisheries 25(1):15-21.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-14April 2013
-------
Chapter 15 References
Gustafson, R. G., R. S. Waples, J. M. Myers, L. A. Weitkamp, G. J. Bryant, 0. W. Johnson, and J. J. Hard. 2007.
Pacific salmon extinctions: quantifying lost and remaining diversity. Conservation Biology 21:1009-
1020.
Gwartney, L. A. 1985. Naknek Drainage Rainbow Trout Study in the Katmai National Park and Preserve.
King Salmon, AK: Alaska Department of Fish and Game, Division of Sport Fish and the U.S.
Department of Interior, National Park Service.
Hansen A. J. 1987. Regulation of Bald Eagle reproductive rates in Southeast Alaska. Ecology 68:1387-
1392.
Heard W. R. 1991. Life history of pink salmon [Oncorhynchus gorbuschd). Pages 119-230 in C. Grootand
L. Margolis (eds.), Pacific Salmon Life Histories. Vancouver, BC: UBC Press
Helfield, J. M. and R. J. Naiman. 2006. Keystone interactions: salmon and bear in riparian forests of
Alaska. Ecosystems 9:167-180.
Hilborn, R. 2006. Fisheries success and failure: The case of the Bristol Bay salmon fishery. Bulletin of
Marine Science 78(3):487-498.
Hilborn, R., T. Quinn, D. Schindler, and D. Rogers. 2003. Biocomplexity and fisheries sustainability.
Proceedings of the National Academy of Sciences of the United States of America 100:6564-6568.
Holen, D., andT. Lemons. 2010. Subsistence Harvests and Uses of Wild Resources in Lime Village, Alaska,
2007. Technical Paper No. 355. Anchorage, AK: Alaska Department of Fish and Game, Division of
Subsistence.
Holen, D., T. M. Krieg, andT. Lemons. 2011. Subsistence Harvests and Uses of Wild Resources in King
Salmon, Naknek, and South Naknek, Alaska, 2007. Technical Paper No. 360. Anchorage, AK: Alaska
Department of Fish and Game, Division of Subsistence.
Holen, D., J. Stariwat, T. M. Krieg, and T. Lemons. 2012. Subsistence Harvest and Uses of Wild Resources in
Aleknagik, Clark's Point, and Manokotak, Alaska, 2008. Technical Paper No. 368. Anchorage, AK:
Alaska Department of Fish and Game, Division of Subsistence.
Johnson, J. and P. Blanche. 2012. Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes—Southwestern Region, Effective June 1, 2012. Special Publication No. 12-08.
Anchorage, AK: Alaska Department of Fish and Game.
Johnson, S. W., J. F. Thedinga, and A. S. Feldhausen. 1994. Juvenile salmonid densities and habitat use in
the main-stem Situk River, Alaska, and potential effects of glacial flooding. Northwest Science
68(4):319-325.
Johnson, Jennifer S., E. Nobmann, E. Asay, A. Lanier. 2009. Dietary intake of Alaska Native people in two
regions and implications for health: The Alaska Native Dietary and Subsistence Food Assessment
Project International journal of Circumpolar Health 68:109-122.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-15April 2013
-------
Chapter 15 References
King, R. S., C. M. Walker, D. F. Whigham, S. Baird, and J. A. Back. 2012. Catchment topography and
wetland geomorphology drive macroinvertebrate community structure and juvenile salmonid
distributions in southcentral Alaska headwater streams. Freshwater Science 31(2):341-364.
Krieg, T., M. Chythlook, P. Coiley-Kenner, D. Holen, K. Kamletz, and H. Nicholson. 2005. Freshwater Fish
Harvest and Use in Communities of the Kvichak watershed, 2003. Technical Data Report No. 297.
Juneau AK: Alaska Department of Fish and Game, Division of Subsistence.
Krieg, T. M., D. L. Holen, and D. Koster. 2009. Subsistence Harvests and Uses of Wild Resources in Igiugig,
Kokhanok, Koliganek, Levelock, and New Stuyahok, Alaska 2005. Technical Paper No. 322. Dillingham,
AK: Alaska Department of Fish and Game, Division of Subsistence.
Krueger, C. C., M. J. Lisac, S. J. Miller, and W. H. Spearman. 1999. Genetic differentiation of rainbow trout
(Oncorhynchus mykiss] in the Togiak National Wildlife Refuge, Alaska. Alaska Fisheries Technical
Report No. 55. Anchorage, AK: U.S. Fish and Wildlife Service, Fish Genetics Laboratory.
Larkin, G. A. and P. A. Slaney. 1997. Implications of trends in marine-derived nutrient influx to South
Coastal British Columbia salmonid production. Fisheries 22:16-24.
Lindley, S. T., C. B. Grimes, M. S. Mohr, W. Peterson. J. Stein, J. T. Anderson, L. W. Botsford, D. L. Bottom, C.
A. Busack, T. K. Collier, J. Ferguson, J. C. Garza, A. M. Grover, D. G. Hankin, R. G. Kope, P. W. Lawson, A.
Low, R. B. MacFarlane, K. Moore, M. Palmer-Zwahlen, F. B. Schwing, J. Smith, C. Tracy, R. Webb, B. K.
Wells, T. H. Williams. 2009. What Caused the Sacramento River Fall Chinook Stock Collapse? National
Oceanic and Atmospheric Administration, National Marine Fisheries Service, Southwest Fisheries
Science Center. NOAA-TM-NMFS-SWFSC-447. 61 pages.
Lisac, M. J. 2009. Seasonal Distribution and Biological Characteristics of Dolly Varden in the Goodnews
River, Togiak National Wildlife Refuge, Alaska, 2005-2006. Alaska Fisheries Technical Report No. 103.
Dillingham, AK: U.S. Fish and Wildlife Service, Togiak National Wildlife Refuge.
Lisac, M. J. and R. D. Nelle. 2000. Migratory Behavior and Seasonal Distribution of Dolly Varden Salvelinus
malma in the Togiak River Watershed, Togiak National Wildlife Refuge. Dillingham, AK: U.S. Fish and
Wildlife Service.
McGlauflin, M. T., D. E. Schindler, L. W. Seeb, C. T. Smith, C. Habicht, and J. E. Seeb. 2011. Spawning habitat
and geography influence population structure and juvenile migration timing of sockeye salmon in
the Wood River lakes, Alaska. Transactions of the American Fisheries Society 140:763-782.
Meka, J. M., E. E. Knudsen, D. C. Douglas, and R. B. Benter. 2003. Variable migratory patterns of different
adult rainbow trout life history types in a southwest Alaska watershed. Transactions of the American
Fisheries Society 132:717-732.
Miller, S. D. and D. W. McCollum. 1994. Nonresident Hunters: Their Characteristics and Attitudes Toward
Wildlife. Anchorage, AK: Alaska Department of Fish and Game. 468 pages.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-16April 2013
-------
Chapter 15 References
Miller, S. D., G. C. White, R. A. Sellers, H. V. Reynolds, J. W. Schoen, K. Titus, and V. G. Barnes . 1997. Brown
and black bear density estimation in Alaska using radiotelemetry and replicated mark resight
techniques. Wildlife Monographs 133:1-55.
Milner, A. M. and R. G. Bailey. 1989. Salmonid colonization of new streams in Glacier Bay National Park,
Alaska, USA. Aquaculture and Fisheries Management 20:179-192.
Minard, R. E., M. Alexandersdottir, and S. Sonnichsen. 1992. Estimation of Abundance, Seasonal
Distribution, and Size and Age Composition of Rainbow Trout in the Kvichak River, Alaska, 1986-1991.
Fishery Data Series No. 92-51. Anchorage, AK: Alaska Department of Fish and Game, Division of
Sport Fish.
Moore, J. and D. Schindler. 2004. Nutrient export from freshwater ecosystems by anadromous sockeye
salmon (Oncorhynchus nerka]. Canadian Journal of Fisheries and Aquatic Sciences 61:1582-1589.
Morstad S. and C.E. Brazil. 2012. Kvichak River Sockeye Salmon StockStatus and Action Plan, 2012: A
Report to the Alaska Board of Fisheries. Anchorage, AK: Alaska Department of Fish and Game,
Division of Sport Fish, Research and Technical Services.
Murphy, N. J., C. D. Schraer, E. Boyko, L. Bulkow, B. Doty, and A. Lanier. 1995. Dietary change and obesity
associated with glucose intolerance in Alaska Natives. Journal of the American Dietetic Association
95:676-682.
Nehlsen, W., J. Williams, and J. Lichatowich. 1991. Pacific salmon at the crossroads—Stocks at risk from
California, Oregon, Idaho, and Washington. Fisheries 16:4-21.
Nickelson, T. E., J. D. Rodgers, S. L. Johnson, and M. F. Solazzi. 1992. Seasonal changes in habitat use by
juvenile coho salmon [Oncorhynchus kisutch) in Oregon coastal streams. Canadian Journal of
Fisheries and Aquatic Sciences 49:783-789.
Nobmann, E. D., R. Ponce, C. Mattil, R. Devereux, B. Dyke, S. 0. E. Ebbesson, S. Laston, J. MacCluer, D.
Robbins, T. Romenesko, G. Ruotolo, C. R. Wenger, and B. V. Howard. 2005. Dietary intakes vary with
age among Eskimo adults of northwest Alaska in the GOCADAN study, 2000-2003. The Journal of
Nutrition 135:856-862.
NRC (National Research Council). 1996. Upstream: Salmon and Society in the Pacific Northwest.
Washington, DC.
Ostberg, C. 0., S. D. Pavlov, and L. Hauser. 2009. Evolutionary relationships among sympatric life history
forms of Dolly Varden inhabiting the landlocked Kronotsky Lake, Kamchatka, and a neighboring
anadromous population. Transactions of the American Fisheries Society 138(1): 1-14.
Pinsky, M. L., D. B. Springmeyer, M. N. Goslin, and X. Augerot. 2009. Range-wide selection of catchments
for Pacific salmon conservation. Conservation Biology 23:680-691.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-17April 2013
-------
Chapter 15 References
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008.
Anchorage, AK. Available: .
Quinn, T. P. 2005. The Behavior and Ecology of Pacific Salmon and Trout. Seattle, WA: University of
Washington Press.
Rahr, G., J. Lichatowich, R. Hubley, and S. Whidden. 1998. Sanctuaries for native salmon: a conservation
strategy for the 21st century. Fisheries 23:6-36.
Reynolds, J.B. 2000. Life History Analysis ofTogiak River Char through Otolith Microchemistry. Fairbanks,
AK: Alaska Cooperative Fish and Wildlife Research Unit, University of Alaska Fairbanks.
Ruggerone, G. T., R. M. Peterman, and B. Dorner. 2010. Magnitude and trends in abundance of hatchery
and wild pink salmon, chum salmon, and sockeye salmon in the North Pacific Ocean. Marine and
Coastal Fisheries: Dynamics, Management, and Ecosystem Science 2:306-328.
Russell, R. 1977. Rainbow trout life history studies in Lower Talarik Creek - Kvichak drainage.
Anchorage AK: Alaska Department of Fish and Game.
Salo, E. 0.1991. Life history of chum salmon (Oncorhynchus ketd). Pages 231-310 in C. Groot and L.
Margolis (eds.), Pacific Salmon Life Histories. Vancouver BC: UBC Press.
Salomone, P., S. Morstad, T. Sands, M. Jones, T. Baker, G. Buck, F. West, and T. Kreig. 2011. Bristol Bay
Area Management Report. Anchorage, AK.
Scanlon, B. 2000. The Ecology of Arctic Char and the Dolly Varden in the BecharofLake Drainage, Alaska.
Fairbanks, AK: University of Alaska.
Scheuerell, M., J. Moore, D. Schindler, and C. Harvey. 2007. Varying effects of anadromous sockeye
salmon on the trophic ecology of two species of resident salmonids in southwest Alaska. Freshwater
Biology 52:1944-1956.
Schindler, D. E., P. R. Leavitt, C. S. Brock, S. P. Johnson, P. D. Quay. 2005. Marine-derived nutrients,
commercial fisheries, and production of salmon and lake algae in Alaska. Ecology 86:3225-3231.
Schindler, D. E., R. Hilborn, B. Chasco, C. P. Boatright, T. P. Quinn, L. A. Rogers, and M. S. Webster. 2010.
Population diversity and the portfolio effect in an exploited species. Nature 465:609-612.
Smith, R. J., K. A. Hobson, H. N. Koopman, and D. M. Lavigne. 1996. Distinguishing between populations of
freshwater and saltwater harbor seals (Phoca vitulina] using stable isotope ratios and fatty acid
profiles. Canadian Journal of Fisheries and Aquatic Sciences 53:272-279.
Solazzi, M., T. Nickelson, S. Johnson, and J. Rodgers. 2000. Effects of increasing winter rearing habitat on
abundance of salmonids in two coastal Oregon streams. Canadian Journal of Fisheries and Aquatic
Sciences 57:906-914.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-18April 2013
-------
Chapter 15 References
U.S. Census Bureau. 2010. Alaska QuickFacts from the United States Census Bureau. Available:
.
USEPA (U.S. Environmental Protection Agency). 1998. Guidelines for Ecological Risk Assessment
EPA/630/R-95/002F. April. Washington, DC: Environmental Protection Agency, Office of Research
and Development
Valkenburg, P., R. A. Sellers, R. C. Squibb, J. D. Woolington, A. R. Aderman, and B. W. Dale. 2003.
Population dynamics of caribou herds in southwestern Alaska. Rangifer Special Issue 14:131-142.
Wentworth, C. 2007. Subsistence Migratory Bird Harvest Survey: Bristol Bay 2001-2005. Unpublished,
Anchorage, AK: U.S. Fish and Wildlife Service.
Willson, M. F., Gende S. M., Marston B. H. 1998. Fishes and the forest: expanding perspectives on fish-
wildlife interactions. Bioscience 48:455-462.
Wong D., and C. Wentworth. 1999. Subsistence Waterfowl Harvest Survey, Bristol Bay 1995-1998.
Unpublished. Anchorage, AK: U.S. Fish and Wildlife Service.
Woody, C. A. and S. L. O'Neal. 2010. Fish Surveys in Headwater Streams of the Nushagak and Kvichak River
Drainages, Bristol Bay, Alaska, 2008-2010. Anchorage, AK: Fisheries Research and Consulting.
Woolington, J. D. 2004. Unit 17 moose management report. Pages 246-266 in C. Brown (ed.), Moose
Management Report of Survey and Inventory Activities: 1 July 2 001-30 June 2 003. Juneau, AK: Alaska
Department of Fish and Game.
Woolington, J. D. 2009. Mulchatna caribou management report, Units 9B, 17,18 south, 19A & 19B. Pages
11 31 in P. Harper (ed.), Caribou Management Report Of Survey And Inventory Activities, 1 July 2006-
1 July 2008. Juneau, AK: Alaska Department of Fish and Game.
15.1.6 Chapter 6—Mine Scenarios
ADEC (Alaska Department of Environmental Conservation). 2012. Situation Report: Fairbanks Gold
Mining, Inc. Cyanide Water Spill. Anchorage AK: Alaska Department of Environment and
Conservation, Division of Spill Prevention and Response, Prevention and Emergency Response
Program.
ADF&G (Alaska Department of Fish and Game). 2013. Invasive Species. Overview.
. Accessed: March 4, 2013.
BC (Province of British Columbia, Canada). 2001. Ambient Water Quality Guidelines for Manganese.
Ministry of Environment.
BC (Province of British Columbia, Canada). 2008. Ambient Water Quality Guidelines for Iron. Ministry of
Environment.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-19April 2013
-------
Chapter 15 References
Biesinger, K. E. and G. M. Christensen. 1972. Effects of various metals on survival, growth, reproduction,
and metabolism ofDaphnia magna. Journal of the Fisheries Board of Canada 39:1691-1700.
Blight, G. E. 2010. Geotechnical Engineering for Mine Waste Storage Facilities, Boca Raton, FL: CRC Press.
Blight, G. E. and A. B. Fourie. 2003. A Review of Catastrophic Flow Failures of Deposit of Mine Waste and
Municipal Refuse. In L. Picarelli (ed.), Proceedings of the International Workshop on Occurrence and
Mechanisms of Flow-like Landslides in Natural Slopes and Earthfllls, Sorrento, Italy. May 14-16, 2003.
Blowes, D. W., C. J. Ptacek, J. Jurjovec. 2003. Mill Tailings: Hydrogeology and Geochemistry. Pages 95-116
in J. L. Jambar, D. W. Blowes, A. I. M. Ritchie (eds.), Environmental Aspects of Mine Wastes:
Mineralogical Association of Canada Short Course Series. Volume 31.
Boutet, C. and C. Chaisemartin. 1973. Specific toxic properties of metallic salts. Comptes Rendus Des
Seances De La Societe De Biologie Et De Ses Filiales 167(12) :193 3-1938.
CCME (Canadian Council of Ministers of the Environment). 1999. Canadian Water Quality Guidelines for
the Protection of Aquatic Life: Molybdenum. Winnipeg.
CCME (Canadian Council of Ministers of the Environment). 2009. Canadian Water Quality Guidelines for
the Protection of Aquatic Life: Boron. Winnipeg.
CEA (Chugach Electric Association). 2011. Facilities. Available:
. Accessed: January 3,
2013.
Chapman, P.M., H. Bailey, and E. Canaria. 2000. Toxicity of total dissolved solids associated with two
mine effluents to chironomid larvae and early life stages of rainbow trout. Environmental Toxicology
and Chemistry 19:210-214.
Eddmaps. 2013. Species Information. Alaska Exotic Plant Information Clearinghouse. Available:
. Accessed: March 4, 2013.
Environment Canada and Health Canada. 2010. Screening Assessment for the Challenge: Antimony
trioxide (antimony oxide). Available: . Ottawa, Ontario.
Environment Canada and Health Canada. 2011. Screening Assessment for the Challenge: Cobalt, cobalt
chloride, sulfuric acid, cobalt (2+) salt (1:1), andsulfuric acid, cobalt salt. Available:
. Ottawa, Ontario.
Etcheverry, D. 2012. Aquatic Invasive Plants: A New Focus for Alaskan Invasive Species Management.
Presentation at the 13th Annual Invasive Species Conference, University of Alaska Fairbanks.
. Accessed: March 4, 2013.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-20 April 2013
-------
Chapter 15 References
Gammons, C. H., L. N. Harris, J. M. Castro, P. A. Cott and B. W. Hanna. 2009. Creating lakes from open pit
mines: processes and considerations, with emphasis on northern environments. Canadian Technical
Report of Fisheries and Aquatic Sciences 2826.106 pages.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska, February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
Goldman, C. 1960. Primary productivity and limiting factors in three lakes of the Alaska Peninsula.
Ecological Monographs 30:207-230.
Hidalgo, P. and I.G.R. Gutz. 2001. Determination of low concentrations of the flotation reagent ethyl
xanthate by sampled DC polarography and flow injection with the amperometric detection. Talanta
54:403-409.
HSWCD (Homer Soil and Water Conservation District). 2007. Managing invasive plants in wetlands of the
Kenai Peninsula: Developing a management program of reed canarygrass infestations, FY2007
Progress Report Summary. Homer, AK.
Laurence, D. 2011. Establishing a sustainable mining operation: an overview. Journal of Cleaner
Production 19(2-3):278-284.
Marsik, T. 2009. Basic Study of Renewable Energy Alternatives for Electricity Generation in
Dillingham/Aleknagik Region. Prepared for Nushagak Electric and Telephone Cooperative, Inc.
Bristol Bay Environmental Science Lab, University of Alaska Fairbanks, Bristol Bay Campus.
Merz, J. E., J. R. Smith, M. L. Workman, J. D. Setka, and B. Mulchaey. 2008. Aquatic macrophyte
encroachment in Chinook salmon spawning beds: Lessons learned from gravel enhancement
monitoring in the Lower Mokelumne River, California. North American Journal of Fisheries
Management 28:1568-1577.
ML&P (Municipal Light and Power). 2012. About ML&P. Available:
. Accessed: January 3, 2013.
Moore, J. and D. Schindler. 2004. Nutrient export from freshwater ecosystems by anadromous sockeye
salmon (Oncorhynchus nerka}. Canadian Journal of Fisheries and Aquatic Sciences 61:1582-1589.
NDM (Northern Dynasty Minerals). 2006. Water Right Applications Submitted to the Alaska Department
of Natural Resources. Available: . Accessed: October 12, 2012.
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008.
Anchorage, AK. Available: .
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-21April 2013
-------
Chapter 15 References
Reeve, A. S. and M. B. Gracz. 2008. Simulating the hydrogeologic setting of peatlands in the Kenai
Peninsula Lowlands, Alaska. Wetlands 28:92-106.
Singer, D. A., V. I. Berger, B. C. Moring. 2008. Porphyry Copper Deposits of the World: Database and
Grade and Tonnage Models, 2008. U.S. Geological Survey Open File Report 2008-1155. 45 pages.
Swedish Chemicals Inspectorate. 2008. European Union Risk Assessment Report: Diantimony trioxide Risk
Assessment. Luxembourg: Office for Official Publications of the European Communities.
USEPA (U.S. Environmental Protection Agency). 1976. Quality criteria for Water. Washington, DC: U.S.
Environmental Protection Agency. 534 pages.
USEPA (U.S. Environmental Protection Agency). 1980. Ambient Water Quality Criteria for Antimony.
EPA/440/5-80-020. Washington, DC: U.S. Environmental Protection Agency, Office of Water.
USEPA (U.S. Environmental Protection Agency). 1997. EPA Can Do More to Minimize Hardrock Mine
Liabilities. Washington, DC: U.S. Environmental Protection Agency, Office of the Inspector General.
EiDMF6-08-0016-7100223. June 11.
USEPA (U.S. Environmental Protection Agency). 1998. Guidelines for Ecological Risk Assessment
EPA/630/R-95/002F. April. Washington, DC: Environmental Protection Agency, Office of Research
and Development.
USEPA (U.S. Environmental Protection Agency). 2011.^4 Field-Based Aquatic Life Benchmark for
Conductivity in Central Appalachian Streams (Final Report). Washington, DC: U.S. Environmental
Protection Agency. EPA/600/R-10/023F.
USFWS (U.S. Fish and Wildlife Service). 2012. National Wetland Inventory. Available:
. Accessed: December 11, 2012.
USGS (U.S. Geological Survey). 2012a. National Hydrography Dataset, High Resolution, Alaska. Available:
ftp://nhdftp.usgs.gov/DataSets/Staged/States/FileGDB/HighResolution. Accessed: October 16,
2012.
USGS (U.S. Geological Survey). 2012b. Water Resources of Alaska. USGS Alaska Science Center Water
Resources Office. Last updated: December 12, 2012. Available:
. Accessed: April 17, 2013.
Wells, P. S. 2011. Long Term In Situ Behaviour of Oil Sands Fine Tailings in Suncor's Pond 1A. In
Proceedings from the Tailings and Mine Waste 2011 Conference, Vancouver, British Columbia,
November 6-9, 2011.
15.1.7 Chapter 7—Mine Footprint
ADF&G (Alaska Department of Fish and Game). 2012. Alaska Freshwater Fish Inventory Database.
Available: .
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-22 April 2013
-------
Chapter 15 References
ADF&G (Alaska Department of Fish and Game). 2Ql3.Anadramous Waters Catalog, Overview. Available:
. Accessed: April 9, 2013.
Alexander, R., E. Boyer, R. Smith, G. Schwarz, and R. Moore. 2007. The role of headwater streams in
downstream water quality. Journal of the American Water Resources Association 43:41-59.
Anderson, K. E., A. J. Paul, E. McCauley, L. J. Jackson, J. R. Post, and R. M. Nisbet. 2006. Instream flow
needs in streams and rivers: the importance of understanding ecological dynamics. Frontiers in
Ecology and the Environment 4:309-318.
Armstrong, J. B., D. E. Schindler, K. L. Omori, C. P. Ruff, and T. P. Quinn. 2010. Thermal heterogeneity
mediates the effects of pulsed subsidies across a landscape. Ecology 91:1445-1454.
Arthington, A. H., S. E. Bunn, N. L. Poff, and R. J. Naiman. 2006. The challenge of providing environmental
flowrules to sustain river ecosystems. Ecological Applications 16:1311-1318.
Bartholow, J. M. 2010. Constructing an Interdisciplinary Flow Regime Recommendation. Journal of the
American Water Resources Association 46:892-906.
Beechie, T. J., E. Beamer, and L. Wasserman. 1994. Estimating coho salmon rearing habitat and smolt
production losses in a large river basin, and implications for habitat restoration. North American
Journal of Fisheries Management 14:797-811.
Bovee, K. D. 1982. A Guide to Stream Habitat Analysis Using the Instream Flow Incremental Methodology.
U.S. Fish and Wildlife Service Instream Flow Group Information Paper No. 12, FWS/OBS-2/26, Fort
Collins, CO.
Bovee, K.D., B. L. Lamb, J. M. Bartholow, C. D. Stalnaker, J. Taylor, and J. Henrikson. 1998. Stream Habitat
Analysis Using the Instream Flow Incremental Methodology. U.S. Geological Survey, Biological
Resources Division, Information and Technical Report USGS/BRD-1998-0004.131 pages.
Bradford, M. J., G. C. Taylor, J. A. Allan, and P. S. Higgins. 1995. An experimental study of the stranding of
juvenile coho salmon and rainbow trout during rapid flow decreases under winter conditions. North
American Journal of Fisheries Management 15:47-479.
Brannon, E. 1987. Mechanisms stabilizing salmonid fry emergence timing. Canadian Special Publication
of Fisheries and Aquatic Sciences 96:120-124.
Brown, T. G. and G. F. Hartman. 1988. Contribution of seasonally flooded lands and minor tributaries to
the production of coho salmon in Carnation Creek, British Columbia. Transactions of the American
Fisheries Society 117:546-551.
Brown, R. S., W. A. Hubert, and S. F. Daly. 2011. A primer on winter, ice, and fish: what fisheries biologists
should know about winter ice processes and stream-dwelling fish. Fisheries 36:8-26.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-23 April 2013
-------
Chapter 15 References
Bryant, M. D., N. D. Zymonas, and B. E. Wright. 2004. Salmonids on the fringe: Abundance, species
composition, and habitat use of salmonids in high-gradient headwater streams, southeast Alaska.
Transactions of the American Fisheries Society 133:1529-1538.
Bue, B. G., S. M. Fried, S. Sharr, D. G. Sharp, J. A. Wilcock, and H. J. Geiger. 1988. Estimating salmon
escapement using area-under-the-curve, aerial observer efficiency, and stream-life estimates: the
Prince William Sound pink salmon example. North Pacific Anadromous Fish Commission Bulletin
1:240-250.
Chadwick, M. A. and A. D. Huryn. 2007. Role of habitat in determining macroinvertebrate production in
an intermittent-stream system. Freshwater Biology 52:240-251.
Collen, P. and R. J. Gibson. 2001. The general ecology of beavers (Castor spp.), as related to their
influence on stream ecosystems and riparian habitats, and the subsequent effects on fish—A review.
Reviews in Fish Biology and Fisheries 10:439-461.
Compton, J. E., M. R. Church, S. T. Larned, and W. E. Hogsett. 2003. Nitrogen export from forested
watersheds in the Oregon Coast Range: The role of N2-fixing red alder. Ecosystems 6:773-785.
Cunjak, R. A. 1996. Winter habitat of selected stream fishes and potential impacts from land-use activity.
Canadian Journal of Fisheries and Aquatic Sciences 53 (Supplement l):267-282.
Curry, R. A., J. Gehrels, D. L. G. Noakes, and R. Swainson. 1994. Effects of river flow fluctuations on
groundwater discharge through brook trout, Salvelinusfontinalis, spawning and incubation habitats.
Hydrobiologia 277:121-134.
Dekar, M. P., R. S. King, J. A. Back, D. F. Whigham, and C. M. Walker. 2012. Allochthonous inputs from
grass-dominated wetlands support juvenile salmonids in headwater streams: evidence from stable
isotopes of carbon, hydrogen, and nitrogen. Freshwater Science 31:121-132.
Doucett, R. R., G. Power, D. R. Barton, R. J. Drimmie, and R. A. Cunjak. 1996. Stable isotope analysis of
nutrient pathways leading to Atlantic salmon. Canadian journal of Fisheries and Aquatic Sciences
53:2058-2066.
Dye, J. E. and C. J. Schwanke. 2009. Report to the Alaska Board of Fisheries for the Recreational Fisheries of
Bristol Bay, 2007,2008, and 2009. Alaska Deptartment of Fish and Game, Division of Sport Fish,
Research and Technical Services.
Elliott, G. V. and J. E. Finn. 1983. Fish Utilization of Several Kenai River Tributaries: 1982 Field Report.
Anchorage AK: U.S. Fish and Wildlife Service, Special Studies.
Estes, C. C. 1998. Annual Summary oflnstream Flow Reservations and Protection in Alaska. Alaska
Department of Fish and Game, Division of Sport Fish, Research and Technical Services.
Ferreira, V., V. Gulis, and M. A. S. Graca. 2006. Whole-stream nitrate addition affects litter decomposition
and associated fungi but not invertebrates. Oecologia 149:718-729.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-24 April 2013
-------
Chapter 15 References
Gesch, D., M. Oimoen, S. Greenlee, C. Nelson, M. Steuck, D. Tyler. 2002. The National Elevation Dataset.
Photogrammatic Engineering and Remote Sensing 68(1):5-11.
Gesch, D. B. 2007. Chapter 4-The National Elevation Dataset. Pages 99-118 in Maune, D. (ed.), Digital
Elevation Model Technologies and Applications: The DEM Users Manual. 2nd edition. Bethesda, MD:
American Society for Photogrammetry and Remote Sensing.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska, February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
Hancock, P. J. 2002. Human impacts on the stream-groundwater exchange zone. Environmental
Management 29:763-781.
Hendry, A. P., J. E. Hensleigh, and R. R. Reisenbichler. 1998. Incubation temperature, developmental
biology, and the divergence of sockeye salmon (Oncorhynchus nerka] within Lake Washington.
Canadian Journal of Fisheries and Aquatic Sciences 55:1387-1394.
Henning, J. A., R. E. Gresswell, and I. A. Fleming. 2006. Juvenile salmonid use of freshwater emergent
wetlands in the floodplain and its implications for conservation management. North American
Journal of Fisheries Management 26:367-376.
Hillman, T. W., J. W. Mullan, and J. S. Griffith. 1992. Accuracy of underwater counts of juvenile Chinook
salmon, coho salmon, and steelhead. North American Journal of Fisheries Management 12:598-603.
Hodgson, S. andT. P. Quinn. 2002. The timing of adult sockeye salmon migration into fresh water:
adaptations by populations to prevailing thermal regimes. Canadian Journal of Zoology 80:542-555.
Huusko, A., L. Greenberg, M. Stickler, T. Linnansaari, M. Nykanen, T. Vehanen, S. Koljonen, P. Louhi, and
K. Alfredsen. 2007. Life in the ice lane: the winter ecology of stream salmonids. River Research and
Applications 23:469-491.
Irons J. G., Ill, S. R. Ray, L. K. Miller, M. W. Oswood. 1989. Spatial and seasonal patterns of streambed
water temperatures in an Alaskan subarctic stream. Proceedings of the Symposium on Headwaters
Hydrology June: 381-390. American Water Resources Association.
Johnson, J. and P. Blanche. 2012. Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes - Southwestern Region, Effective June 1, 2012. Special Publication No. 12-08.
Anchorage, AK: Alaska Department of Fish and Game.
Jones, E. L., Ill, S. Heinl, and K. Pahlke. 2007. Aerial counts. Pages 399-410 in D. H. Johnson, B. M. Shrier,
J. S. O'Neal, J. A. Knutzen, X. Augerot, T. A. O'Neil, and T. N. Pearsons, (eds.), Salmonid field protocols
handbook: techniques for assessing status and trends in salmon and trout populations. Bethesda, MD:
American Fisheries Society.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-25 April 2013
-------
Chapter 15 References
King, R. S., C. M. Walker, D. F. Whigham, S. Baird, and J. A. Back. 2012. Catchment topography and
wetland geomorphology drive macroinvertebrate community structure and juvenile salmonid
distributions in southcentral Alaska headwater streams. Freshwater Science 31(2):341-364.
Lang, D. W., G. H. Reeves, J. D. Hall, and M. S. Wipfli. 2006. The influence of fall-spawning coho salmon
(Oncorhynchus kisutch) on growth and production of juvenile coho salmon rearing in beaver ponds
on the Copper River Delta, Alaska. Canadian journal of Fisheries and Aquatic Sciences 63:917-930.
Lytle, D. A. and N. L. R. Poff. 2004. Adaptation to natural flow regimes. Trends in Ecology & Evolution
19:94-100.
Matthews, W. J. and E. Marsh-Matthews. 2003. Effects of drought on fish across axes of space, time and
ecological complexity. Freshwater Biology 48:1232-1253.
Morstad, S. 2003. Kvichak River Sockeye Salmon Spawning Ground Surveys, 1955-2002. Regional
Information Number 2A02-32. Anchorage, AK: Alaska Department of Fish and Game.
Nickelson, T. E., J. D. Rodgers, S. L. Johnson, and M. F. Solazzi. 1992. Seasonal changes in habitat use by
juvenile coho salmon (Oncorhynchus kisutch] in Oregon coastal streams. Canadian Journal of
Fisheries and Aquatic Sciences 49:783-789.
Pess, G. R., S. A. Morley, J. L. Hall, and R. K. Timm. 2005. Monitoring Floodplain Restoration. Pages 127-
166 in P. Roni, (ed.), Monitoring Stream and Watershed Restoration. Bethesda, MD: American
Fisheries Society.
Pess, G. R., D. R. Montgomery, E. A. Steel, R. E. Bilby, B. E. Feist, and H. M. Greenberg. 2002. Landscape
characteristics, land use, and coho salmon (Oncorhynchus kisutch} abundance, Snohomish River,
Wash., U.S.A. Canadian Journal of Fisheries and Aquatic Sciences 59:613-623.
Piccolo, J. J. and M. S. Wipfli. 2002. Does red alder (Alnus rubra] in upland riparian forests elevate
macroinvertebrate and detritus export from headwater streams to downstream habitats in
southeastern Alaska? Canadian Journal of Fisheries and Aquatic Sciences 59:503-513.
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008.
Anchorage, AK. Available: .
Poff, N. L., J. D. Allan, M. B. Bain, J. R. Karr, K. L. Prestegaard, B. D. Richter, R. E. Sparks, and J. C.
Stromberg. 1997. The natural flow regime: A paradigm for river conservation and restoration.
Bioscience 47:769-784.
Poff, N., J. Olden, D. Pepin, and B. Bledsoe. 2006. Placing global stream flow variability in geographic and
geomorphic contexts. Regulated Rivers: Research & Management 22:149-166.
Poff, N., B. Richter, A. Arthington, S. Bunn, R. Naiman, C. Apse, E. Kendy, A. Warner, R. Jacobson, and K.
Rogers. 2009. The ecological limits of hydrologic alteration (ELOHA): A new framework for
developing regional environmental flow standards. Freshwater Biology 55:147-170.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-26 April 2013
-------
Chapter 15 References
Poff, N. L. and J. K. H. Zimmerman. 2010. Ecological responses to altered flow regimes: a literature
review to inform the science and management of environmental flows. Freshwater Biology 55:194-
205.
Pollock, M. M., M. Heim, and D. Werner. 2003. Hydrologic and geomorphic effects of beaver dams and
their influence on fishes. The Ecology and Management of Wood in World Rivers 37:213-233.
Pollock, M. M., G. R. Pess, and T. J. Beechie. 2004. The importance of beaver ponds to coho salmon
production in the Stillaguamish River basin, Washington, USA. North American Journal of Fisheries
Management 24:7 49-7 60.
Postel, S. and B. Richter. 2003. Rivers for Life: Managing Water for People and Nature, Washington, DC:
Island Press.
Power, G., R. S. Brown, and J. G. Imhof. 1999. Groundwater and fish—Insights from northern North
America. Hydrological Processes 13:401-422.
Quinn, T. P. 2005. The Behavior and Ecology of Pacific Salmon and Trout Seattle, WA: University of
Washington Press.
Rains, M. C. 2011. Water sources and hydrodynamics of closed-basin depressions, Cook Inlet Region,
Alaska. Wetlands 31:377-387.
Richter, B. D. 2010. Re-thinking environmental flows: from allocations and reserves to sustainability
boundaries. River Research and Applications 26:1052-1063.
Richter, B., M. Davis, C. Apse, and C. Konrad. 2012. A presumptive standard for environmental flow
protection. River Research and Applications 28:1312-1321.
Rogers, L. A. and D. E. Schindler. 2008. Asynchrony in population dynamics of sockeye salmon in
southwest Alaska. Oikos 117:1578-1586.
Ruff, C., D. Schindler, J. Armstrong, K. Bentley, G. Brooks, G. Holtgrieve, M. McGlauflin, J. Seeb, and C.
Torgersen. 2011. Temperature-associated population diversity in salmon confers benefits to mobile
consumers. Ecology 92:2073-2084.
Schindler, D. E., R. Hilborn, B. Chasco, C. P. Boatright, T. P. Quinn, L. A. Rogers, and M. S. Webster. 2010.
Population diversity and the portfolio effect in an exploited species. Nature 465:609-612.
Shaftel, R., R. King, and J. Back. 2011. Breakdown rates, nutrient concentrations, and macroinvertebrate
colonization of bluejoint grass litter in headwater streams of the Kenai Peninsula, Alaska. Journal of
the North American Benthological Society 30:386-398.
Shaftel, R., R. King, and J. Back. 2012. Alder cover drives nitrogen availability in Kenai lowland
headwater streams, Alaska. Biogeochemistry 107:135-148.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-27 April 2013
-------
Chapter 15 References
Slaughter, C. W. 1990. Aufeis formation and prevention. Pages 433-458 in W. L. Ryan and R. D. Crissman
(eds.), Cold Regions Hydrology and Hydraulics. New York, NY: American Society of Civil Engineers.
Solazzi, M., T. Nickelson, S. Johnson, and J. Rodgers. 2000. Effects of increasing winter rearing habitat on
abundance of salmonids in two coastal Oregon streams. Canadian Journal of Fisheries and Aquatic
Sciences 57:906-914.
Sommer, T. R., M. L. Nobriga, W. C. Harrell, W. Batham, and W. J. Kimmerer. 2001. Floodplain rearing of
juvenile chinook salmon: evidence of enhanced growth and survival. Canadian Journal of Fisheries
and Aquatic Sciences 58:325-333.
Szarzi, N. J, S. J. Fleischman, R. A. Clark, and C. M. Kerkvliet. 2007. Stock status and recommended
escapement goal for Anchor River Chinook salmon. Fisheries Manuscript No. 07-05, Alaska
Department of Fish and Game, Divisions of Sport Fish and Commercial Fisheries.
Tank, J. L., E. J. Rosi-Marshall, N. A. Griffiths, S. A. Entrekin, and M. L. Stephen. 2010. A review of
allochthonous organic matter dynamics and metabolism in streams, journal of the North American
Benthological Society 29:118-146.
USFWS (U.S. Fish and Wildlife Service). 2012. National Wetland Inventory. Available:
. Accessed: December 11, 2012.
USGS (U.S. Geological Survey). 2012a. National Hydrography Dataset. High Resolution, Alaska. Available:
. Accessed: October 16,
2012.
USGS (U.S. Geological Survey). 2012b. Water Resources of Alaska. USGS Alaska Science Center Water
Resources Office. Last updated: December 12, 2012. Available:
. Accessed: April 17, 2013.
Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Gushing. 1980. The river continuum
concept. Canadian Journal of Fisheries and Aquatic Sciences 37:130-137.
Webster, J. R. and E. F. Benfield. 1986. Vascular plant breakdown in fresh-water ecosystems. Annual
Review of Ecology and Systematics 17:567-594.
West, R. L., M. W. Smith, W. E. Barber, J. B. Reynolds, and H. Hop. 1992. Autumn migration and
overwintering of arctic grayling in coastal streams of the Arctic National Wildlife Refuge, Alaska.
Transactions of the American Fisheries Society 121:709-715.
Wipfli, M. S. and D. P. Gregovich. 2002. Export of invertebrates and detritus from fishless headwater
streams in southeastern Alaska: Implications for downstream salmonid production. Freshwater
Biology 47:957-969.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-28 April 2013
-------
Chapter 15 References
Wipfli, M. S. and J. Musslewhite. 2004. Density of red alder (Alnus rubrd) in headwaters influences
invertebrate and detritus subsidies to downstream fish habitats in Alaska. Hydrobiologia 520:153-
163.
Wipfli, M. S., J. S. Richardson, and R. J. Naiman. 2007. Ecological linkages between headwaters and
downstream ecosystems: Transport of organic matter, invertebrates, and wood down headwater
channels. Journal of the American Water Resources Association 43:72-85.
Wobus, C., A. Maest, B. Prucha, and D. Albert. 2012. Potential Hydrologic and Water Quality Alteration
from Large-Scale Mining of the Pebble Deposit in Bristol Bay, Alaska: Results from an Integrated
Hydrologic Model of a Preliminary Mine Design (review draft). Boulder CO: Stratus Consulting.
Woody, C. A. andB. Higman. 2011. Groundwater as Essential Salmon Habitat in Nushagak and Kvichak
River Headwaters. Unpublished report. Available:
.
Personal Communications
Morstad, Steve. Fishery Biologist III, ADF&G. September 2011—Email of unpublished data to Rebecca
Shaftel.
15.1.8 Chapter 8—Water Collection, Treatment, and Discharge
ADEC (Alaska Department of Environmental Conservation). 2011. Water Quality Standards: Alaska
Department of Environmental Conservation. Report 18-AAC-70.
ADEC (Alaska Department of Environmental Conservation). 2012. Water Quality Standards Amended as
of April 8,2012. Juneau, AK: Alaska Department of Environmental Conservation.
ADED (Alaska Division of Economic Development). 2012. Nixon ForkMine. Available:
.
ADNR (Alaska Department of Natural Resources). 2012. Nixon ForkMine. Available:
Accessed: April 11, 2012.
Angilletta, M. J., E. Ashley Steel, K. K. Bartz, J. G. Kingsolver, M. D. Scheuerell, B. R. Beckman, and L. G.
Crozier. 2008. Big dams and salmon evolution: changes in thermal regimes and their potential
evolutionary consequences. Evolutionary Applications 1:286-299.
ANZECC (Australia and New Zealand Environment and Conservation Council). 2000. Australian and New
Zealand guidelines for fresh and marine water quality. Volume 2. Aquatic Ecosystems: Rationale and
Background Information. Canberra, Australia: Australia and New Zealand Environment and
Conservation Council and Agriculture and Resource Management Council of Australia and New
Zealand.
Armstrong, J. B., D. E. Schindler, K. L. Omori, C. P. Ruff, and T. P. Quinn. 2010. Thermal heterogeneity
mediates the effects of pulsed subsidies across a landscape. Ecology 91:1445-1454.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-29 April 2013
-------
Chapter 15 References
Barry, K. L, J. A. Grout, C. D. Levings, B. H. Nidle, and G. E. Piercey. 2000. Impacts of acid mine drainage
on juvenile salmonids in an estuary near Britannia Beach in Howe Sound, British Columbia.
Canadian Journal of Fisheries and Aquatic Sciences 57:2032-2043.
Beacham, T. D. and C. B. Murray. 1990. Temperature, egg size, and development of embryos and alevins
of five species of Pacific salmon: A comparative analysis. Transactions of the American Fisheries
Society 119:927-945.
Borgmann, U., C. Janssen, R. Blust, K. Brix, R. Dwyer, R. Erickson, L. Hare, S. Luoma, P. Paquin, C. Roberts,
and W.-X. Wang. 2005. Incorporation of dietborne metals exposure into regulatory frameworks.
Pages 153-190 in J. Meyer, W. Adams, K. Brix, S. Luoma, D. Mount, W. Stubblefield, and C. Wood,
editors. Toxicity of Dietborne Metals to Aquatic Organisms. Pensacola, FL: SETAC Press.
Brannon, E. 1987. Mechanisms stabilizing salmonid fry emergence timing. Canadian Special Publication
of Fisheries and Aquatic Sciences 96:120-124.
Brix, K. V., D. K. DeForest, and W. J. Adams. 2011. The sensitivity of aquatic insects to divalent metals: A
comparative analysis of laboratory and field data. Science of the Total Environment 409:4187'-4197.
Bryant, M. 2009. Global climate change and potential effects on Pacific salmonids in freshwater
ecosystems of southeast Alaska. Climatic Change 95:169-193.
Buchwalter, D. B., D. J. Cain, W. H. Clements, and S. N. Luoma. 2008. Using biodynamic models to
reconcile differences between laboratory toxicity tests and field biomonitoring with aquatic insects
(vol. 41, pg 4821, 2007). Environmental Science & Technology 42:3117-3117.
CH2M Hill and The Clark Fork and Blackfoot, LLC. 2004. Biological Assessment Bull Trout: Biological
Assessment of the Milltown Reservoir Sediments Operable Unit Revised Proposed Plan and the
Surrender Application for the Milltown Hydroelectric Project (FERCNo. 2543). Helena, MT: U.S.
Environmental Protection Agency and Federal Energy Regulatory Commission.
Chapman, P. M., H. Bailey, and E. Canaria. 2000. Toxicity of total dissolved solids associated with two
mine effluents to chironomid larvae and early life stages of rainbow trout. Environmental Toxicology
and Chemistry 19:210-214.
Clements, W. H., D. M. Carlisle, J. M. Lazorchak, and P. C. Johnson. 2000. Heavy metals structure benthic
communities in Colorado mountain streams. Ecological Applications 10:626-638.
Cohen, B. I. 2010. Fraser River Sockeye Salmon: Past Declines. Future Sustainability? Interim Report.
Vancouver, BC: Commission of Inquiry into the Decline of Sockeye Salmon in the Fraser River.
Cohen, B. I. 2012. The Uncertain Future of Fraser River Sockeye. Volume 2: Causes of the Decline. Final
Report CP32-93/2012E-2. Ottawa, Ontario, Canada: Commission of Inquiry into the Decline of
Sockeye Salmon in the Fraser River.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-30 April 2013
-------
Chapter 15 References
Courtney, L. A. and W. H. Clements. 2002. Assessing the influence of water and substratum quality on
benthic macroinvertebrate communities in a metal-polluted stream: an experimental approach.
Freshwater Biology 47:1766-1778.
DeForest, D. K. and E. J. Van Genderen. 2012. Application of U.S. EPA guidelines in a bioavailability-based
assessment of ambient water quality criteria for zinc in freshwater. Environmental Toxicology and
Chemistry 31:1264-1272.
DeForest, D., R. Gensemer, E. Van Genderen, and J. Gorsuch. 2011a. Protectiveness of water quality
criteria for copper in western United States waters relative to predicted olfactory responses in
juvenile pacific salmon. Integrated Environmental Assessment and Management 1:336-347.
DeForest, D., J. Meyer, R. Genseer, B. Shephard, W. Adams, R. Dwyer, J. Gorsuch, and E. VanGenderen.
2011b. Are ambient water quality criteria for copper protective of olfactory impairment in fish?
Integrated Environmental Assessment and Management 7:145-146.
Earthworks. 2012. U.S. Copper Porphyry Mines Report: The Track Record of'Water Quality Impacts
Resulting from Pipeline Spills, Tailings Failures and Water Collection and Treatment Failure.
Washington, DC.
Eisler, R. 2000. Handbook of Chemical Risk Assessment: Health Hazards to Humans, Plants, and Animals,
Volume 1: Metals. Boca Raton, FL: Lewis Publisher. 738 pages.
Environment Canada and Health Canada. 2011. Screening Assessment for the Challenge: Cobalt; cobalt
chloride; sulfuric acid, cobalt (2+) salt (1:1); andsulfuric acid, cobalt salt. Available:
. Ottawa, Ontario.
Farag, A. M., C. J. Boese, D. F. Woodward, and H. L. Bergman. 1994. Physiological-changes and tissue
metal accumulation in rainbow-trout exposed to foodborne and waterborne metals. Environmental
Toxicology and Chemistry 13:2021-2029.
Farag, A. M., D. E. Woodward, W. Brumbaugh, J. N. Goldstein, E. MacConnell, C. Hogstrand, and F. T.
Barrows. 1999. Dietary effects of metals-contaminated invertebrates from the Coeur d'Alene River,
Idaho, on cutthroat trout. Transactions of the American Fisheries Society 128:578-592.
Fuerstenau, M. C., B. M. Wakawa, R. K. Price, and R. D. Wellik. 1974. Toxicity of selected sulfhydryl
collectors to rainbow trout. Transactions of the Society of Mining Engineers 256:337-341.
Gensemer, R. W. and R. C. Playle. 1999. The bioavailability and toxicity of aluminum in aquatic
environments. Critical Reviews in Environmental Science and Technology 29:315-450.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska. February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-31April 2013
-------
Chapter 15 References
Griffith, M. B., J. M. Lazorchak, and A. T. Herlihy. 2004. Relationships among exceedences of metals
criteria, the results of ambient bioassays, and community metrics in mining-impacted streams.
Environmental Toxicology and Chemistry 23:1786-1795.
Gustafson, 0. 2012. Active Metal Mines of'the Fraser River Basin and Fish - CaseStudies. Pebble Limited
Partnership, Technical Paper Series No. 1, White Paper No. 3.
Hansen, J. A., J. C. A. Marr, J. Lipton, D. Cacela, and H. L. Bergman. 1999. Differences in neurobehavioral
response of Chinook salmon (Oncorhynchus tshawytscha] and rainbow trout (Oncorhynchus mykiss]
exposed to copper and cobalt: Behavioral avoidance. Environmental Toxicology and Chemistry 18:
1972-1978.
Hecht, S., D. Baldwin, C. Mebane, T. Hawkes, S. Gross, and N. Scholz. 2007. An Overview of Sensory Effects
on Juvenile Salmonids Exposed to Dissolved Copper: Applying a Benchmark Concentration Approach to
Evaluate Sublethal Neurobehavioral Toxicity. Seattle, WA: National Oceanic and Atmospheric
Administration, Northwest Fisheries Science Center.
Hodgson, S. andT. P. Quinn. 2002. The timing of adult sockeye salmon migration into fresh water:
adaptations by populations to prevailing thermal regimes. Canadian Journal of Zoology 80:542-555.
IPCS (International Programme on Chemical Safety). 2004. Concise International Chemical Assessment
Document 63, Manganese and its Compounds: Environmental Aspects. Geneva, Switzerland: World
Health Organization.
ITRC (Interstate Technology & Regulatory Council). 2010. Bingham Canyon Water Treatment Plant,
Kennecott South Zone. Mining Waste Treatment Technology Selection Web. Washington DC:
Interstate Technology & Regulatory Council, Mining Waste Team, .
Jennings, S., D. Neuman, and P. Blicker. 2008. Acid Mine Drainage and Effects on Fish Health and Ecology:
A Review. Bozeman, Montana: Reclamation Research Group, LLC.
Jensen, A. and B. Johnsen. 1999. The functional relationship between peak spring floods and survival and
growth of juvenile Atlantic salmon (Salmo salar) and brown trout (Salmo truttd). Functional Ecology
13:778-785.
Joling, D. 2011. Mine Executive Says Pebble, Fishing Can Co-Exist Anchorage Daily News, Anchorage, AK.
March 4, 2011.
Kiffney, P. M. and W. H. Clements. 1996. Size-dependent response of macroinvertebrates to metals in
experimental streams. Environmental Toxicology and Chemistry 15:1352-1356.
Kuipers, J. R., A. S. Maest, K. A. MacHardy, and G. Lawson. 2006. Comparison of Predicted and Actual
Water Quality at Hardrock Mines: The Reliability of Predictions in Environmental Impact Statements.
Luoma, S. N. and P. S. Rainbow. 2005. Why is metal bioaccumulation so variable? Biodynamics as a
unifying concept. Environmental Science & Technology 39:1921-1931.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-32 April 2013
-------
Chapter 15 References
MacDonald, D., J. Sinclair, M. Crawford, H. Principe, M. Meneghetti. 2011. Potential Effects of
Contaminants on Fraser River Sockeye Salmon. Cohen Commission Technical Report 2. Vancouver,
BC: MacDonald Environmental Sciences, Ltd.
Marchand, E. A. 2002. Minerals and mine drainage. Water Environment Research 74:746-772.
McCullough, D. A., J. M. Bartholow, H. I. Jager, R. L. Beschta, E. F. Cheslak, M. L. Deas, J. L. Ebersole, J. S.
Foott, S. L. Johnson, K. R. Marine, M. G. Mesa, J. H. Petersen, Y. Souchon, K. F. Tiffan, and W. A.
Wurtsbaugh. 2009. Research in thermal biology: Burning questions for coldwater stream fishes.
Reviews in Fisheries Science 17:90-115.
Mclntyre, J. K., D. H. Baldwin, J. P. Meador, and N. L. Scholz. 2008a. Chemosensory deprivation in juvenile
coho salmon exposed to dissolved copper under varying water chemistry conditions. Environmental
Science & Technology 42:1352-1358.
Mclntyre, J. K., D. H. Baldwin, J. P. Meador, and N. L. Scholz. 2008b. Chemosensory deprivation in juvenile
coho salmon exposed to dissolved copper under varying water chemistry conditions (42:1352-1358,
2008). Environmental Science & Technology 42:6774-6775 (Addition or correction).
Mebane, C. A. 2010. Cadmium Risks to Freshwater Life: Derivation and Validation of Low-Effect Criteria
Values Using Laboratory and Field Studies. Investigations Report 2 006-5245, Version 1.2. Reston VA:
U.S. Geological Survey.
Mellina, E., R. D. Moore, S. G. Hinch, J. S. Macdonald, and G. Pearson. 2002. Stream temperature responses
to clearcut logging in British Columbia: The moderating influences of groundwater and headwater
lakes. Canadian Journal of Fisheries and Aquatic Sciences 59:1886-1900.
Meyer, J. S. and W. J. Adams. 2010. Relationship between biotic ligand model-based water quality criteria
and avoidance and olfactory responses to copper by fish. Environmental Toxicology and Chemistry
29:2096-2103.
Meyer, J., W. Adams, K. Brix, D. Mount, W. Stubblefield, and C. Wood. 2005. Toxicity ofDietborne Metals to
Aquatic Organisms. Pensacola, FL: SETAC Press.
Mortazavi, S. 2008. Application of membrane separation technology to mitigation of mine effluent and
acidic mine drainage. Mine Environment Neutral Drainage Program (MEND) Report 3.15.1.
Nelitz, M., M. Porterm, K. Wieckowski, D. Marmorek, K. Bryan, A. Hall, and D. Abraham. 2011. Evaluating
the Status of Fraser River Sockeye Salmon and Role Of Freshwater Ecology in Their Decline. Cohen
Commission Technical Report 3. Vancouver, BC: ESSA Technologies, Ltd.
NICNAS (National Industrial Chemicals Notification and Assessment Scheme). 1995. Sodium Ethyl
Xanthate, Priority Existing Chemical No. 5. Full Public Report. Canberra, Australia: Australian
Government Publishing Service.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-33April 2013
-------
Chapter 15 References
NICNAS (National Industrial Chemicals Notification and Assessment Scheme). 2000. Sodium Ethyl
Xanthate, Priority Existing Chemical Secondary Notification Assessment. Report No. 5S. Canberra,
Australia: Australian Government Publishing Service.
O'Neal, S. and C. A. Woody. 2011. Canada's Fraser River: Reasons for Sockeye Salmon Declines with a
Comparison to Bristol Bay. Anchorage, AK: Fisheries Research and Consulting.
Ott, A. G. and P. Weber Scannell. 1994. Fish Monitoring Study, Red Dog Mine in the Wulik River Drainage,
Emphasis on Dolly Varden fSalvelinus malmaj. Summary Report 1990-1993. Anchorage, AK: Alaska
Department of Fish and Game.
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008,
Anchorage, AK. Available: .
Poole, G. C., J. B. Dunham, D. M. Keenan, S. T. Saulter, D. A. McCoullough, C. Mebane, J. C. Lockwood, D. A.
Essig, M. P. Hicks, D. J. Sturdevant, E. J. Materna, S. A. Spalding, J. Risely, and M. Deppman. 2004. The
case for regime-based water quality standards. BioScience 54:155-161.
Power, G., R. S. Brown, and J. G. Imhof. 1999. Groundwater and fish - Insights from northern North
America. Hydrological Processes 13:401-422.
Quinn, T. P. 2005. The Behavior and Ecology of Pacific Salmon and Trout Seattle, WA: University of
Washington Press.
Quinn, T. P. and D. J. Adams. 1996. Environmental changes affecting the migratory timing of American
shad and sockeye salmon. Ecology 77:1151-1162.
Reimer, P. S. 1999. Environmental effects of manganese and proposed freshwater guidelines to protect
aquatic life in British Columbia. M.S. Thesis, University of British Columbia, Department of Chemical
and Bio-Resource Engineering.
Rudolph, B. L, I. Andreller, and C. J. Kennedy. 2008. Reproductive success, early life stage development,
and survival of westslope cutthroat trout (Oncorhynchus clarki lewisi) exposed to elevated selenium
in an area of active coal mining. Environmental Science and Technology 42:3109-3114.
Schindler, D. E., R. Hilborn, B. Chasco, C. P. Boatright, T. P. Quinn, L. A. Rogers, and M. S. Webster. 2010.
Population diversity and the portfolio effect in an exploited species. Nature 465:60-612.
Schmidt, T. S., W. H. Clements, K. A. Mitchell, S. E. Church, R. B. Wanty, D. L. Fey, P. L. Verplanck, and C. A.
San Juan. 2010. Development of a new toxic-unit model for the bioassessment of metals in streams.
Environmental Toxicology and Chemistry 29:2432-2442.
Shao, E., J. Wei, A. Yo, and R. Levy. 2009. Application ofultraflltration and reverse osmosis for mine waste
water reuse. Water in Mining Conference, September 9, 2009, Perth, WA.
USEPA (U.S. Environmental Protection Agency). 1985a. Ambient Water Quality Criteria for Copper -1984.
EPA 440/5-84-031. Washington, DC.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-34April 2013
-------
Chapter 15 References
USEPA (U.S. Environmental Protection Agency). 1985b. Guidelines for Deriving Numerical National Water
Quality Criteria for the Protection of Aquatic Organisms and Their Uses. Washington, DC.
USEPA (U.S. Environmental Protection Agency). 1986. Ambient Aquatic Life Water Quality Criteria for
Lead. Washington, DC: Office of Water, Regulations and Standards, Criteria and Standards Division.
USEPA (U.S. Environmental Protection Agency). 1987. Ambient Water Quality Criteria for Zinc -1987.
EPA-440/6-87-003. Washington, DC: Office of Water, Regulations and Standards, Criteria and
Standards Division.
USEPA (U.S. Environmental Protection Agency). 1998. Alaska's Red Dog Mine Required to Reduce
Pollutant Discharges. Available: .
USEPA (U.S. Environmental Protection Agency). 2004. Draft Aquatic Life Water Quality Criteria for
Selenium - 2004. Washington DC: U.S. Environmental Protection Agency, Office of Water. EPA-822-
D-04-001.
USEPA (U.S. Environmental Protection Agency). 2007. Aquatic Life Ambient Freshwater Quality Criteria—
Copper. 2007 Revision. Washington, DC.
USEPA (U.S. Environmental Protection Agency). 2008. The U.S. Environmental Protection Agency (EPA)
Plans to Re-Issue a Wastewater Discharge Permit to: Tech Cominco Alaska, Inc., Red Dog Mine Near
Kotze, Alaska, and the State of Alaska Proposes to Certify the Permit. Anchorage, AK.
USEPA (U.S. Environmental Protection Agency). 2011. A Field-Based Aquatic Life Benchmark for
Conductivity in Central Appalachian Streams (Final Report). Washington, DC: U.S. Environmental
Protection Agency. EPA/600/R-10/023F.
Webb, M., H. Ruber, and G. Leduc. 1976. The toxicity of various mining flotation reagents to rainbow
trout (Salmo gairdneri). Water Research 10:303-306.
Weber Scannell, P. K. 1991. Influence of Temperature on Freshwater Fishes: A Literature Review with
Emphasis on Species in Alaska. Juneau, AK: Alaska Department of Fish and Game.
WE (Windward Environmental LLC). 2002. Development of Site-Specific Water Quality Criteria for the
South Fork Coeur D'alene River, Idaho: Derivation of Acute and Chronic Criteria for Lead and Zinc.
Boise, ID: Idaho Department of Environmental Quality.
Wobus, C., A. Maest, B. Prucha, and D. Albert. 2012. Potential Hydrologic and Water Quality Alteration
from Large-Scale Mining of the Pebble Deposit in Bristol Bay, Alaska: Results from an Integrated
Hydrologic Model of a Preliminary Mine Design (review draft). Boulder CO: Stratus Consulting.
Woodward, D. F., W. G. Brumbaugh, A. J. Delonay, E. E. Little, and C. E. Smith. 1994. Effects on rainbow-
trout fry of a metals-contaminated diet of benthic invertebrates from the Clark Fork River, Montana.
Transactions of the American Fisheries Society 123:51-62.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-35April 2013
-------
Chapter 15 References
Woodward, D. F., A. M. Farag, H. L. Bergman, A. J. Delonay, E. E. Little, C. E. Smith, and F. T. Barrows. 1995.
Metals-contaminated benthic invertebrates in the Clark-Fork River, Montana—Effects on age-0
brown trout and rainbow-trout. Canadian Journal of Fisheries and Aquatic Sciences 52:1994-2004.
Woody, C. A. and B. Higman. 2011. Groundwater as Essential Salmon Habitat in Nushagak and Kvichak
River Headwaters. Unpublished report available:
.
Xu, Y., J. P. Lay, and F. Korte. 1988. Fate and effects of xanthates in laboratory freshwater systems.
Bulletin of Environmental Contamination and Toxicology 41:683-689.
Zamzow, K. 2011. Investigations of Surface Water Quality in the Nushagak, Kvichak, and Chulitna
Watersheds, Southwest Alaska 2009-2010. Sutton, AK: Center for Science in Public Participation.
15.1.9 Chapter 9—Tailings Dam Failure
ADEC (Alaska Department of Environmental Conservation). 2011. Water Quality Standards: Alaska
Department of Environmental Conservation. Report 18-AAC-70.
ADF&G (Alaska Department of Fish and Game). 2011. Moose Species Profile. Juneau, AK.
ADF&G (Alaska Department of Fish and Game). 2012. Alaska Freshwater Fish Inventory Database.
Available: .
ADNR (Alaska Department of Natural Resources). 2005. Guidelines for Cooperation with the Alaska Dam
Safety Program. Dam Safety and Construction Unit, Water Resources Section, Division of Mining,
Land, and Waters. 230 pages.
AECOM. 2009. Root Cause Analysis ofTVA Kingston Dredge Pond Failure on December 22, 2008.
Ankley, G. T., D. M. DiToro, D. J. Hansen, and W. J. Berry. 1996. Technical basis and proposal for deriving
sediment quality criteria for metals. Environmental Toxicology and Chemistry 15:2056-2066.
ARCO (Atlantic Richfield Company). 1998. Milltown Reservoir Sediments NPL Site, Clark Fork River
Operable Unit, Final Draft Remedial Investigation Report. Anaconda, MT.
Azam, S. and Q. Li. 2010. Tailings dam failures: a review of the last one hundred years. Geotechnical
News. December:50-53.
Bisson, P. and R. Bilby. 1982. Avoidance of suspended sediment by juvenile coho salmon. North American
Journal of Fisheries Management 2:371-374.
Blight, G. E. 2010. Geotechnical Engineering for Mine Waste Storage Facilities. Boca Raton, FL: CRC Press.
Bogan, D., D. Rinella, and R. Shaftel. 2012. Baseline Macroinvertebrate and Diatom Surveys in Wadeable
Streams of the Kvichak and Nushagak Watersheds, Bristol Bay, Alaska. Report prepared for The
Nature Conservancy, Anchorage, AK. Alaska Natural Heritage Program, University of Alaska,
Anchorage.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1536 April 2013
-------
Chapter 15 References
Borgmann, U., C. Janssen, R. Blust, K. Brix, R. Dwyer, R. Erickson, L. Hare, S. Luoma, P. Paquin, C. Roberts,
and W.-X. Wang. 2005. Incorporation of dietborne metals exposure into regulatory frameworks.
Pages 153-190 in J. Meyer, W. Adams, K. Brix, S. Luoma, D. Mount, W. Stubblefield, and C. Wood,
editors. Toxicity of Dietborne Metals to Aquatic Organisms. Pensacola, FL: SETAC Press.
Brabets, T. P. 1992. Hydrologic Data for the Lower Copper River, Alaska, May to September 1991. Open-
File Report 92-89. Anchorage, AK: U.S. Geological Survey.
Breitenbach, A. 2007. Rockflll for Mines. Infomine E-Book. Available:
.
Brown, R. S., W. A. Hubert, and S. F. Daly. 2011. A primer on winter, ice, and fish: what fisheries biologists
should know about winter ice processes and stream-dwelling fish. Fisheries 36:8-26.
Browne, R. 2 011. Tailings Dam Break Model in Nevada - A Practitioner's Perspective. PowerPoint
presentation, August 24, 2011, Colder Associates. Available:
.
Brumbaugh, W. G., C. G. Ingersoll, N. E. Kemble, T. W. May, and J. L. Zajicek. 1994. Chemical
characterization of sediments and pore-water from the upper Clark-Fork River and Milltown
Reservoir, Montana. Environmental Toxicology and Chemistry 13:1971-1983.
Bryce, S. A., G. A. Lomnicky, and P. R. Kaufmann. 2010. Protecting sediment-sensitive aquatic species in
mountain streams through the application of biologically based streambed sediment criteria. Journal
of the North American Benthological Society 29:657-672.
Buck, G.B., C.E. Brazil, F. West, L.F. Fair, X. Zhang, and S.L. Maxwell. 2012. Stock Assessment of Chinook,
Sockeye, and Chum Salmon in the Nushagak River. Anchorage, AK: Alaska Department of Fish and
Game, Fishery Manuscript Series No. 12-05.
Bustard, D. R. and D. W. Narver. 1975. Aspects of the winter ecology of juvenile coho salmon
(Oncorhynchus kisutch) and steelhead trout (Salmo g air dneri). journal of the Fisheries Research
Board of Canada 32:667-680.
Canfield, T. J., N. E. Kemble, W. G. Brumbaugh, F. J. Dwyer, C. G. Ingersoll, and J. F. Fairchild. 1994. Use of
benthic invertebrate community structure and the sediment quality triad to evaluate metal-
contaminated sediment in the upper Clark-Fork River, Montana. Environmental Toxicology and
Chemistry 13:1999-2012.
Chambers, D. M., B. Higman. 2011. Long-Term Risks of Tailings Dam Failure. Report. Bozeman MT: Center
for Science in Public Participation. Available: .
Chapman, D. W. and K. P. McLeod. 1987. Development of Criteria for Fine Sediment in Northern Rockies
Ecoregion. Page 279. Final report prepared for EPA Region 10. Battelle Columbus Laboratories. EPA
910/9-87-162.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-37April 2013
-------
Chapter 15 References
Gulp,}., F. Wrona, and R. Davies. 1986. Response of stream benthos and drift to fine sediment deposition
versus transport. Canadian journal of Zoology 64:1345-1351.
Cunjak, R. A. 1996. Winter habitat of selected stream fishes and potential impacts from land-use activity.
Canadian Journal of Fisheries and Aquatic Sciences 53 (Supplement l):267-282.
Dalpatram, A. 2011. Estimation of Tailings Dam Break Discharges. USSD Workshop on Dam Break
Analysis Applied to Tailings Dam, August 24-26, 2011, Denver CO.
Dann, T. H., C. Habicht, H. A. Hoyt, T. T. Baker, and F. W. West. 2011. Genetic Stock Composition of the
Commercial Harvest of Sockeye Salmon in Bristol Bay, Alaska, 2009. Fishery Data Series 11-21. Alaska
Department of Fish and Game, Division of Sport Fish, Research and Technical Services.
Davies, M. P. 2002. Tailings impoundment failures: Are geotechnical engineers listening? Geotechnical
News. September:31-36.
Davies, M. and T. Martin. 2009. Mining Market Cycles and Tailings Dam Incidents. Pages 3-14 in Tailings
and Mine Waste '09, Proceedings of the 13th International Conference on Tailings and Mine Waste.
Edmonton, Canada: University of Alberta Geotechnical Center.
Davies, M. P., T. E. Martin, P. Lighthall. 2000. Mine Tailings Dams: When Things Go Wrong. Pages 261-
273 in Tailings Dams 2000. Association of State Dam Safety Officials, U.S. Committee on Large Dams,
Las Vegas, Nevada.
Dye, J. E. and C. J. Schwanke. 2009. Report to the Alaska Board of Fisheries for the Recreational Fisheries of
Bristol Bay, 2007,2008, and 2009. Alaska Deptartment of Fish and Game, Division of Sport Fish,
Research and Technical Services.
El Mejjati, S., J. Bell, J. Botz, A. Faulkner, and S. Maxwell. 2010. Using hydroacoustic methods to
enumerate migrating salmon in the Copper River, Miles Lake sonar project 2007. Fishery Data Series
No. 10-98. Anchorage, AK: Alaska Department of Fish and Game, Division of Commercial Fisheries.
Farag, A. M., D. F. Woodward, J. N. Goldstein, W. Brumbaugh, and J. S. Meyer. 1998. Concentrations of
metals associated with mining waste in sediments, biofilm, benthic macroinvertebrates, and fish
from the Coeur d'Alene River Basin, Idaho. Archives of Environmental Contamination and Toxicology
34:119-127.
Farag, A. M., D. E. Woodward, W. Brumbaugh, J. N. Goldstein, E. MacConnell, C. Hogstrand, and F. T.
Barrows. 1999. Dietary effects of metals-contaminated invertebrates from the Coeur d'Alene River,
Idaho, on cutthroat trout. Transactions of the American Fisheries Society 128:578-592.
FERC (Federal Energy Regulatory Commission). 1991. Embankment Dams. Chapter 4 in Engineering
Guidelines for the Evaluation ofHydropower Projects. Available:
.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-38 April 2013
-------
Chapter 15 References
Fitzpatrick, A. 2012. Update: Tailings dam collapse atformer Gullbridge mine site. The Telegraph, St.
John's, Newfoundland, December 17, 2012.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska, February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
Goldstein, J. N., D. F. Woodward, and A. M. Farag. 1999. Movements of adult Chinook salmon during
spawning migration in a metals-contaminated system, Coeur d'Alene River, Idaho. Transactions of
the American Fisheries Society 128:121-129.
Goodyear, K. L. and S. McNeill. 1999. Bioaccumulation of heavy metals by aquatic macro-invertebrates of
different feeding guilds: a review. Science of the Total Environment 229:1-19.
Griffith, M. B., J. M. Lazorchak, and A. T. Herlihy. 2004. Relationships among exceedences of metals
criteria, the results of ambient bioassays, and community metrics in mining-impacted streams.
Environmental Toxicology and Chemistry 23:1786-1795.
Hancock, P. J. 2002. Human impacts on the stream-groundwater exchange zone. Environmental
Management 29:763-781.
Holland, W. K., F. W. Rabe, and R. C. Biggam. 1994. Recovery of macroinvertebrate communities from
metal pollution in the south-fork and mainstem of the Coeur d'Alene River, Idaho. Water
Environment Research 66:84-88.
Huusko, A., L. Greenberg, M. Stickler, T. Linnansaari, M. Nykanen, T. Vehanen, S. Koljonen, P. Louhi, and
K. Alfredsen. 2007. Life in the ice lane: The winter ecology of stream salmonids. River Research and
Applications 23:469-491.
ICOLD (International Commission on Large Dams). 2001. Tailings Dams, Risk of Dangerous Occurrences,
Lessons Learnt from Practical Experiences. United Nations Environmental Programme, Bulletin 121.
Ingersoll, C. G., W. G. Brumbaugh, F. J. Dwyer, and N. E. Kemble. 1994. Bioaccumulation of metals by
Hyalella azteca exposed to contaminated sediments from the upper Clark-Fork River, Montana.
Environmental Toxicology and Chemistry 13:2013-2020.
Ingersoll, C. G., P. S. Haverland, E. L. Brunson, T. J. Canfield, F. J. Dwyer, C. E. Henke, N. E. Kemble, D. R.
Mount, and R. G. Fox. 1996. Calculation and evaluation of sediment effect concentrations for the
amphipod Hyalella azteca and the midge Chironomus riparius. Journal of Great Lakes Research
22:602-623.
Johnson, J. and P. Blanche. 2012. Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes—Southwestern Region, Effective June 1,2012. Special Publication No. 12-08.
Anchorage, AK: Alaska Department of Fish and Game.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1539 April 2013
-------
Chapter 15 References
Jones, E. L, T. J. Quinn, and B. W. Van Alen. 1998. Observer accuracy and precision in aerial and foot
survey counts of pink salmon in a southeast Alaska stream. North American Journal of Fisheries
Management 18:832-846.
Jones, J. I., J. F. Murphy, A. L. Collins, D. A. Sear, P. S. Naden, and P. D. Armitage. 2011. The impact of fine
sediment on macro-invertebrates. River Research and Applications 28:1055-1071.
Jones, M., T. Sands, S. Morstad, P. Salomone, G. Buck, F. West, T. Baker, and T. Krieg. 2012. 2011 Bristol
Bay area annual management report. Anchorage, AK: Alaska Department of Fish and Game, Fishery
Management Report No. 12-21.
Kemble, N. E., W. G. Brumbaugh, E. L. Brunson, F. J. Dwyer, C. G. Ingersoll, D. P. Monda, and D. F.
Woodward. 1994. Toxicity of metal-contaminated sediments from the Upper Clark-Fork River,
Montana, to aquatic invertebrates and fish in laboratory exposures. Environmental Toxicology and
Chemistry 13:1985-1997.
Kondolf, G. 2000. Assessing salmonid spawning gravel quality. Transactions of the American Fisheries
Society 129:262-281.
MacDonald, D. D., C. G. Ingersoll, and T. A. Berger. 2000. Development and evaluation of consensus-based
sediment quality guidelines for freshwater ecosystems. Archives of Environmental Contamination
and Toxicology 39:20-31.
Major, J., T. Pierson, R. L. Dinehart, and J. E. Costa. 2000. Sediment yield following severe volcanic
disturbance—A two-decade perspective from Mount St. Helens. Geology 28:819-822.
Marcus, W. A., G. A. Meyer, and D. R. Nimmo. 2001. Geomorphic control of persistent mine impacts in a
Yellowstone Park stream and implications for the recovery of fluvial systems. Geology 29:355-358.
Maret, T. R. and D. E. MacCoy. 2002. Fish assemblages and environmental variables associated with
hard-rock mining in the Coeur d'Alene River Basin, Idaho. Transactions of the American Fisheries
Society 131:865-884.
Maret, T. R., D. J. Cain, D. E. MacCoy, and T. M. Short. 2003. Response of benthic invertebrate assemblages
to metal exposure and bioaccumulation associated with hard-rock mining in northwestern streams,
USA. Journal of the North American Benthological Society 22:598-620.
Miller, John F. 1963. Probable Maximum Precipitation and Rainfall Frequency Data for Alaska for Areas to
400 Square Miles, Durations to 24 Hours, and Return Periods from 1 -100 Years. U.S. Weather Bureau,
Hydrologic Services Division, Cooperative Studies Section. Technical Paper 47. Prepared for U.S.
Department of Agriculture, Soil Conservation Service, Engineering Division.
Miller, J. R. 1997. The role of fluvial geomorphic processes in the dispersal of heavy metals from mine
sites. Journal ofGeochemicalExploration 58:101-118.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1540 April 2013
-------
Chapter 15 References
Milner, A. and R. Piorkowski. 2004. Macroinvertebrate assemblages in streams of interior Alaska
following alluvial gold mining. River Research and Applications 20:719-731.
Moore, J. and H. Langner. 2012. Can a river heal itself? Natural attenuation of metal contamination in
river sediment. Environmental Science and Technology 46:2616-2623.
Morgenstern, N. R. 2011. Improving the safety of mine waste impoundments. Pages 3-10 in Tailings and
Mine Waste '10: Proceedings of the 14th International Conference on Tailings and Mine Waste, Vail, CO.
October 17-20, 2010. Boca Raton FL: CRC Press.
Nagorski, S. A., J. N. Moore, and T. E. McKinnon. 2003. Geochemical response to variable streamflow
conditions in contaminated and uncontaminated streams. Water Resources Research 39:1044-1058.
National Inventory of Dams. 2005.2005 United States National Inventory of Dams database. Summary
information available at: .
NDM (Northern Dynasty Minerals). 2006. Pebble Project Tailings Impoundment G - Initial Application
Report. Prepared by Knight Piesold Ltd.
Newcombe, C. and J. Jensen. 1996. Channel suspended sediment and fisheries: A synthesis for
quantitative assessment of risk and impact. North American Journal of Fisheries Management
Nielsen, J. 1992. Microhabitat-specific foraging behavior, diet, and growth of juvenile coho salmon.
Transactions of the American Fisheries Society 121:617-634.
Nimik, D. and J. Moore. 1991. Prediction of water soluble metal concentrations in fluvially deposited
tailings sediments, Upper Clark Fork Valley, Montana, U.S.A. Applied Geochemistry 6:635-646.
Nimmo, D. R., M. J. Willox, T. D. Lafrancois, P. L. Chapman, S. F. Brinkman, and J. C. Greene. 1998. Effects
of metal mining and milling on boundary waters of Yellowstone National Park, USA. Environmental
Management 22:913-926.
NRC (National Research Council). 2005. Superfund and Mining Megasites. Washington, DC.
Oswood, M. 1989. Community structure of benthic invertebrates in interior Alaskan (USA) streams and
rivers. Hydrobiologia 172:97-110.
Pascoe, G. A., R. J. Blanchet, G. Under, D. Palawski, W. G. Brumbaugh, T. J. Canfield, N. E. Kemble, C. G.
Ingersoll, A. Farag, and J. A. Dalsoglio. 1994. Characterization of ecological risks at the Milltown
Reservoir Clark-Fork River sediments Superfund site, Montana. Environmental Toxicology and
Chemistry 13:2043-2058.
Peck, P. 2007. Avoiding Tailings Dam Failures: Good Practice in Prevention. Workshop on the Safety of
Tailings Management Facilities, November 12,2007. Yerevan, Armenia. Communication on Progress
to the United Nations Convention of the Transboundary Effects of Industrial Accidents.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1541April 2013
-------
Chapter 15 References
Peck, D. V., A. T. Herlihy, B. H. Hill, R. M. Hughes, P. R. Kaufmann, D. J. Klemm, J. M. Lazorchak, F. H.
McCormick, S. A. Peterson, P. L. Ringold, T. Magee, and M. R. Cappaert. 2006. Environmental
Monitoring and Assessment Program - Surface Waters Western Pilot Study: Field Operations Manual
for Wadeable Streams. EPA 600/R-06/003. Washington, DC: U.S. Environmental Protection Agency,
Office of Research and Development.
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008.
Anchorage, AK. Available: .
Quinn, J., R. Daviescolley, C. Hickey, M. Vickers, and P. Ryan. 1992. Effects of clay discharges on streams.
2. Benthic invertebrates. Hydrobiologia 248:235-247.
Rico, M. G., G. Benito, A. Diez-Herrero. 2008. Floods from tailings dam failures. Journal of Hazardous
Materials 154(1-3) :79-87.
Rowe, M., D. Essig, and B. Jessup. 2003. Guide to Selection of Sediment Targets for Use in Idaho TMDLs.
Boise, ID: Idaho Department of Environmental Quality.
Ruggerone, G. T., B. A. Agler, S. E. Gilk, D. B. Molyneaux, D. J. Costello, D. E. Orabutt Jr., and D. B. Young.
2011. Habitat and growth of river-type sockeye salmon in the Kuskokwim watershed, Alaska. Pages
49-89 in S. E. Gilk, D. B. Molyneaux, and Z. W. Liller (eds.), Kuskokwim River Sockeye Salmon
Investigations, 2006 and 2007. Alaska Department of Fish and Game, Fishery Manuscript Series No.
11-04, Anchorage, AK.
Scheuerell, M., J. Moore, D. Schindler, and C. Harvey. 2007. Varying effects of anadromous sockeye
salmon on the trophic ecology of two species of resident salmonids in southwest Alaska. Freshwater
Biology 52:1944-1956.
Schmetterling, D. A., C. G. Clancy, and T. M. Brandt. 2001. Effects of riprap bank reinforcement on stream
salmonids in the western United States. Fisheries 26:6-23.
Semko, R.S. 1954. The stocks of west Kamchatka and their commercial utilization. Izvestiia
Tikhookeanskovo Nauchno-Issledovatelskovo Instituta Rybnovo Khozyaistva i Okeanografli 41:3-109
(translated in 1960, Fisheries Research Board of Canada Translation Series 288).
Shirazi, M. A., W. K. Seim, and D. H. Lewis. 1981. Characterization of Spawning Gravel and Stream System
Evaluation. Proceedings of the Conference on Salmon-Spawning Gravel. Pullman, WA: Washington
Water Resources Research Center.
Silva, F. T., T. W. Lambe, and W. A. Marr. 2008. Probability and Risk of Slope Failure. Journal of
Geotechnical and Geoenvironmental Engineering 134:1691-1699.
Stantec Consulting. 2011. Dam safety review, various former mine sites, former Gullbridge Copper Mine,
NL. Final Inspection Report. Prepared for NL Department of Natural Resources, File No. 121612646,
St. John's, NL.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1542 April 2013
-------
Chapter 15 References
Suter, G. W., II and S. M. Cormier. 2011. Why and how to combine evidence in environmental
assessments: Weighing evidence and building cases. Science of the Total Environment 409:1406-
1417.
USAGE (U.S. Army Corps of Engineers). 2003. Slope Stability Engineer Manual EM 1110-2-1902.
Available: .
USEPA (U.S. Environmental Protection Agency). 2001. Final Coeur d'Alene Basin Ecological Risk
Assessment. Region 10, Seattle, WA.
USEPA (U.S. Environmental Protection Agency). 2007. Aquatic Life Ambient Freshwater Quality Criteria -
Copper, 2007Revision. Washington, DC.
USEPA (U.S. Environmental Protection Agency). 2012a. Clark Fork River Operable Unit. Available:
. Accessed: May 2, 2012.
USEPA (U.S. Environmental Protection Agency). 2012b. Bunker Hill Mining and Metallurgical Superfund
Site (Coeur d'Alene Basin). Available: .
Accessed: May 2, 2012.
USGS (U.S. Geological Survey). 2004. Integrated Investigations of Environmental Effects of Historical
Mining in the Basin and Boulder Mining Districts, Boulder River Watershed, Jefferson County, Montana.
Reston, VA: U.S. Geological Survey.
USGS (U.S. Geological Survey). 2005. Stream-Sediment Geochemistry in Mining-Impacted Streams:
Sediment Mobilized by Floods in the Coeur d'Alene-Spokane River System, Idaho and Washington.
Washington, DC: U.S. Geological Survey.
USGS (U.S. Geological Survey). 2007. Integrated Investigations of Environmental Effects of Historical
Mining in theAnimas River Watershed, San Juan County, Colorado. Reston, VA: U.S. Geological Survey.
Wagener, S. M. and J. D. LaPerriere. 1985. Effects of placer mining on the invertebrate communities of
interior Alaska streams. Freshwater Invertebrate Biology 4:2 08-214.
Wenning, R.J., M. Sorensen, and V.S. Magar. 2006. Importance of implementation and residual risk
analyses in sediment remediation. Integrated Environmental Assessment and Management 2:59-65.
West, R. L, M. W. Smith, W. E. Barber, J. B. Reynolds, and H. Hop. 1992. Autumn migration and
overwintering of Arctic grayling in coastal streams of the Arctic National Wildlife Refuge, Alaska.
Transactions of the American Fisheries Society 121:709-715.
Westley, P. A. H., D. E. Schindler, T. P. Quinn, G. T. Ruggerone, and R. Hilborn. 2010. Natural habitat
change, commercial fishing, climate, and dispersal interact to restructure an Alaskan fish
metacommunity. Oecologia 163:471-484.
Wood, P. and P. Armitage. 1997. Biological effects of fine sediment in the lotic environment.
Environmental Management 21:203-217.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-43April 2013
-------
Chapter 15 References
World Information on Service Energy. 2012. Uranium Project website. Last updated: June 12, 2012.
Available: .
Personal Communications
Rinella, D. J. University of Alaska, Anchorage. Unpublished data.
15.1.10 Chapter 10—Transportation Corridor
ADEC (Alaska Department of Environmental Conservation). 2003. Alaska Water Quality Criteria Manual
for Toxic and Other Deleterious Organic and Inorganic Substances, May.
ADEC (Alaska Department of Environmental Conservation). 2011a. Alaska Pollutant Discharge
Elimination System General Permit for Discharges from Large and Small Construction Activities,
Permit Number: AKR100000. Anchorage, AK.
ADEC (Alaska Department of Environmental Conservation). 201 lb. Alaska Storm Water Guide.
December. Anchorage, AK: Division of Water.
ADF&G (Alaska Department of Fish and Game). 2012. Alaska Freshwater Fish Inventory Database.
Available: .
ADF&G and ADOT (Alaska Department of Fish and Game, Alaska Department of Transportation and
Public Facilities). 2001. Memorandum of Agreement Between Alaska Department of Fish and Game
and Alaska Department of Transportation and Public Facilities for the Design, Permitting, and
Construction of Culverts for Fish Passage, Anchorage, AK.
ADNR (Alaska Department of Natural Resources). 2013. Invasive Weed and Agricultural Pest
Management, Weed Free Gravel, Available: . Palmer, AK: Division of Agriculture, Plant Materials Center. Accessed: March 22, 2013.
ADOT (Alaska Department of Transportation and Public Facilities). 1995. Alaska Highway Drainage
Manual, Available: .
ADOT (Alaska Department of Transportation and Public Facilities). 2001. Iliamna-Nondalton Road
Improvements, Revised Environmental Assessment and Finding of No Significant Impact, Project No.
STP-0214(3)/51951. Prepared for Federal Highway Administration.
Angermeier, P. L, A. P. Wheeler, and A. E. Rosenberger. 2004. A conceptual framework for assessing
impacts of roads on aquatic biota. Fisheries 29:19-29.
Arsan, E. L. and J. L. Bartholomew. 2008. Potential for dissemination of the nonnative salmonid parasite
Myxobolus cerebralis in Alaska. Journal of Aquatic Animal Health 20:136-149.
Auerbach, N. A., M. D. Walker, and D. A. Walker. 1997. Effects of roadside disturbance on substrate and
vegetation properties in Arctic tundra. Ecological Applications 7:218-235.
Bader, C. D. 1997. Controlling Dust. Erosion Control 4:]u\y/August.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1544 April 2013
-------
Chapter 15 References
Behlke, C. E., D. L. Kane, R. F. McLean, and M. D. Travis. 1991. Fundamentals of Culvert Design for Passage
of Weak-Swimming Fish, Final Report No. FHWA-AK-RD-90-10. Prepared for Alaska Department of
Transportation and Public Facilities, in cooperation with U.S. Department of Transportation, Federal
Highway Administration.
Bella, E. M. 2011. Invasion prediction on Alaska trails: distribution, habitat, and trail use. Invasive Plant
Science and Management 4:296-3 05.
Berejikian, B. A. 1992. Feeding Ecology of Rainbow Trout with Comparisons to Arctic Char in Iliamna Lake,
Alaska, M.S. Thesis, University of Washington. 72 pages.
Blossey, B., L. C. Skinner, and J. Taylor. 2001. Impact and management of purple loosestrife (Lythrum
salicarid) in North America. Biodiversity and Conservation 10:1787-1807.
Bolander, P., and A. Yamada. 1999. Dust Palliative Selection and Application Guide, U.S. Departmentof
Agriculture Forest Service. Available:
.
Brown, T. G. and G. F. Hartman. 1988. Contribution of seasonally flooded lands and minor tributaries to
the production of coho salmon in Carnation Creek, British Columbia. Transactions of the American
Fisheries Society 117:546-551.
Bryant, M. D., N. D. Zymonas, and B. E. Wright. 2004. Salmonids on the fringe: Abundance, species
composition, and habitat use of salmonids in high-gradient headwater streams, southeast Alaska.
Transactions of the American Fisheries Society 133:1529-1538.
Christen, D. C. and G. R. Matlack. 2009. The habitat and conduit functions of roads in the spread of three
invasive plant species. Biological Invasions 11:453-465.
Collen, P., and R. J. Gibson. 2001. The general ecology of beavers (Castor spp.), as related to their
influence on stream ecosystems and riparian habitats, and the subsequent effects on fish—A review.
Reviews in Fish Biology and Fisheries 10:439-461.
Copp, G. H. 1989. The habitat diversity and fish reproductive function of floodplain ecosystems.
Environmental Biology of Fishes 26:1-27.
Cunjak, R. A. 1996. Winter habitat of selected stream fishes and potential impacts from land-use activity.
Canadian Journal of Fisheries and Aquatic Sciences 53 (Supplement l):267-282.
Darnell, R. M., W. E. Pequegnat, F. J. Benson, and R. A. Defenbaugh. 1976. Impacts of Construction
Activities in Wetlands of the United States. Washington, DC: U.S. Environmental Protection Agency.
Dekar, M. P., R. S. King, J. A. Back, D. F. Whigham, and C. M. Walker. 2012. Allochthonous inputs from
grass-dominated wetlands support juvenile salmonids in headwater streams: evidence from stable
isotopes of carbon, hydrogen, and nitrogen. Freshwater Science 31:121-132.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1545 April 2013
-------
Chapter 15 References
Demory, R. L, R. L. Orrell, and D. R. Heinle. 1964. Spawning Ground Catalog of the Kvichak River System,
Bristol Bay, Alaska. Issue 488 of Special Scientific Report—Fisheries. Washington, DC: U.S.
Department of Interior, Bureau of Commercial Fisheries.
Denton, K. P., H. B. Rich, Jr., and T. P. Quinn. 2009. Diet, movement, and growth of Dolly Varden in
response to sockeye salmon subsidies. Transactions of the American Fisheries Society 138:1207-
1219.
Ehrenfeld, J. G. and J. P. Schneider. 1991. Chamaecyparis thyoides wetlands and suburbanization: effects
on hydrology, water quality and plant community composition. Journal of Applied Ecology 28:467-
490.
Environment Canada. 2005. Profile of Chloride-based Dust Suppressants Used in Canada, Gatineau,
Quebec: Government of Canada, Road Salts Working Group. Available:
.
ESRI (Environmental Systems Research Institute). 2013. World Imagery, ArcMap 10.0.
Forman, R. T. 2000. Estimate of the area affected ecologically by the road system in the United States.
Conservation Biology 14:31-35.
Forman, R. T. T. and L. E. Alexander. 1998. Roads and their major ecological effects. Annual Review of
Ecology andSystematics 29:207-231.
FSSSWG (Forest Service Stream-Simulation Working Group). 2008. Stream Simulation: An Ecological
Approach to Providing Passage for Aquatic Organisms at Road-Stream Crossings, U.S. Department of
Agriculture, U.S. Forest Service.
Furniss, M. J., T. D. Roelofs, and C. S. Yee. 1991. Road construction and maintenance. Pages 29-323 in W.
R. Meehan (ed.), Influences of Forest and Rangeland Management on Salmonid Fishes and Their
Habitats. American Fisheries Society Special Publication 19.
Gesch, D. B. 2007. Chapter 4-The National Elevation Dataset. Pages 99-118 in Maune, D. (ed.), Digital
Elevation Model Technologies and Applications: The DEM Users Manual. 2nd edition. Bethesda, MD:
American Society for Photogrammetry and Remote Sensing.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska, February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
Gibson, R. J., R. L. Haedrich, and C. M. Wernerheirn. 2005. Loss offish habitat as a consequence of
inappropriately constructed stream crossings. Fisheries 30:10-17.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-46 April 20:L3
-------
Chapter 15 References
Gomez-Uchida, D., J. E. Seeb, M. J. Smith, C. Habicht, T. P. Quinn, and L. W. Seeb. 2011. Single nucleotide
polymorphisms unravel hierarchical divergence and signatures of selection among Alaskan sockeye
salmon (Oncorhynchus nerka] populations. BMC Evolutionary Biology 11:48.
Grout, J. A., C. D. Levings, and J. S. Richardson. 1997. Decomposition rates of purple loosestrife (Lythrum
salicaria) and Lyngbyei's sedge (Carex lyngbyei) in the Fraser River estuary. Estuaries 20:96-102.
Gucinski, H., M. J. Furniss, R. R. Ziemer, and M. H. Brookes. 2001. Forest Roads: A Synthesis of Scientific
Information. General Technical Report. PNW-509. Portland, OR: U.S. Department of Agriculture.
Hamilton, T. D. 2007. Surflcial Geologic Map of the Pebble Project. Report C-l. Anchorage, AK: The Pebble
Partnership.
Hancock, P. J. 2002. Human impacts on the stream-groundwater exchange zone. Environmental
Management 29:763-781.
Harwood, D. W., and E. R. Russell. 1990. Present Practices of Highway Transportation of Hazardous
Materials, August 86 - November 89. May. Kansas City, MO: Midwest Research Institute; Washington,
DC: Federal Highway Administration.
Hastings, K. 2005. Long-Term Persistence of Isolated Fish Populations in the Alexander Archipelago.
Missoula, MT: University of Montana.
Hilderbrand, R. H. and J. L. Kershner. 2000. Conserving inland cutthroat trout in small streams: How
much stream is enough? North American Journal of Fisheries Management 20:513-520.
Hoover, J. 1981. Emission-Oriented Dust Control and Surface Improvement Processes for Unpaved Roads.
Project 1308, Iowa State University, Ames, IA.
Hoover, J. M., K. L. Bergeson, D. E. Fox, C. K. Denny, and R. L. Handy. 1973. Surface Improvement and Dust
Palliation of Unpaved Secondary Roads and Streets. Final report, Iowa Highway Research Board
Project HR-151.
Johnson, J. and P. Blanche. 2012. Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes—Southwestern Region, Effective June 1, 2012. Special Publication No. 12-08.
Anchorage, AK: Alaska Department of Fish and Game.
Johnson, L. E., A. Ricciardi, and J. T. Carlton. 2001. Overland dispersal of aquatic invasive species: A risk
assessment of transient recreational boating. Ecological Applications 11:1789-1799.
Kane, D. L., and P. M. Wellen. 1985. Hydraulic Evaluation of Fish Passage through Roadway Culverts in
Alaska. Report No. FHWA-AK-RD-85-24 & 85-24A. Water Center/Institute of Northern Engineering,
Fairbanks AK, in cooperation with U.S. Department of Transportation, Federal Highway
Administration.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-47 April 2013
-------
Chapter 15 References
King, R. S., C. M. Walker, D. F. Whigham, S. Baird, and J. A. Back. 2012. Catchment topography and
wetland geomorphology drive macroinvertebrate community structure and juvenile salmonid
distributions in southcentral Alaska headwater streams. Freshwater Science 31(2):341-364.
Lake and Peninsula Borough. 2009. Multi-Hazard Mitigation Plan, February.
Lang, D. W., G. H. Reeves, J. D. Hall, and M. S. Wipfli. 2006. The influence of fall-spawning coho salmon
(Oncorhynchus kisutch) on growth and production of juvenile coho salmon rearing in beaver ponds
on the Copper River Delta, Alaska. Canadian journal of Fisheries and Aquatic Sciences 63:917-930.
Langill, D. A. and P. J. Zamora. 2002. An Audit of 'Small Culvert Installations in Nova Scotia: Habitat Loss
and Habitat Fragmentation. 2422, Dartmouth, Nova Scotia: Canadian Department of Fisheries and
Oceans, Habitat Management Division.
Lee, J. E., and S. L. Chown. 2009. Quantifying the propagule load associated with the construction of an
Antarctic research station. Antarctic Science 21:471-475.
Meka, J. M., E. E. Knudsen, D. C. Douglas, and R. B. Benter. 2003. Variable migratory patterns of different
adult rainbow trout life history types in a southwest Alaska watershed. Transactions of the American
Fisheries Society 132:717-732.
Merz, J. E., J. R. Smith, M. L. Workman, J. D. Setka, and B. Mulchaey. 2008. Aquatic macrophyte
encroachment in Chinook salmon spawning beds: Lessons learned from gravel enhancement
monitoring in the Lower Mokelumne River, California. North American Journal of Fisheries
Management 28:1568-1577.
Morgan, E. L., W. F. Porak, and J. A. Arway. 1983. Controlling acidic-toxic metal leachates from southern
Appalachian construction slopes: mitigating stream damage. Transportation Research Record
948:10-16.
Morstad, S. 2003. Kvichak River Sockeye Salmon Spawning Ground Surveys, 1955-2002. Regional
Information Number 2A02-32. Anchorage, AK: Alaska Department of Fish and Game.
Mount, D. R., D. D. Gully, J. R. Hockett, T. D. Garrison, and J. M. Evans. 1997. Statistical models to predict
the toxicity of major ions to Ceriodaphnia dubia, Daphnia magna and Pimephales promelas (fathead
minnow). Environmental Toxicolology & Chemistry 16:2009-2019.
Nehring, R. B. and P. G. Walker. 1996. Whirling disease in the wild: The new reality in the intermountain
west. Fisheries 21:28-30.
Newcombe, C. P. and D. D. MacDonald. 1991. Effects of suspended sediments on aquatic ecosystems.
North American Journal of Fisheries Management 11:72-82.
Nickelson, T. E., J. D. Rodgers, S. L. Johnson, and M. F. Solazzi. 1992. Seasonal changes in habitat use by
juvenile coho salmon (Oncorhynchus kisutch] in Oregon coastal streams. Canadian Journal of
Fisheries and Aquatic Sciences 49:783-789.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1548 April 2013
-------
Chapter 15 References
NICNAS (National Industrial Chemicals Notification and Assessment Scheme). 2000. Sodium ethyl
xanthate, priority existing chemical secondary notification assessment. Canberra, Australia.
Opperman, J.}., R. Luster, B. A. McKenney, M. Roberts, and A. W. Meadows, 2010. Ecologically functional
floodplains: connectivity, flow regime, and scale. Journal of the American Water Resources
Association 46:211-226.
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document2004 through 2008.
Anchorage, AK. Available: .
Price, D. M., T. Quinn, and R. J. Barnard. 2010. Fish passage effectiveness of recently constructed road
crossing culverts in the Puget Sound region of Washington State. North American Journal of Fisheries
Management 30:1110-1125.
Quinn, T. P., H. B. J. Rich, D. Gosse, and N. Schtickzelle. 2012. Population dynamics and asynchrony at fine
spatial scales: a case history of sockeye salmon (Oncorhynchus nerka] population structure in Alaska,
USA. Canadian Journal of Fish and Aquatic Science 69:297-3 06.
Reid, L.M. and T. Dunne. 1984. Sediment production from forest road surfaces. Water Resources Research
20:1753-1761.
Richardson, E. V., B. Simons, S. Karaki, M. Mahmood, and M. A. Stevens. 1975. Highways in the River
Environment: Hydraulic and Environmental Design Considerations Training and Design Manual.
Washington, DC: U.S. Department of Transportation.
Ruff, C.P., D.E. Schindler, J.B. Armstrong, K.T. Bentley, G.T. Brooks, G.W. Holtgrieve, M.T. McGlauflin, C.E.
Torgersen, and J.E. Seeb. 2011. Temperature-associated population diversity in salmon confers
benefits to mobile consumers. Ecology 92:2073-2084.
Russell, R. 1977. Rainbow Trout Life History Studies in Lower Talarik Creek - Kvichak Drainage.
Anchorage, AK: Alaska Department of Fish and Game.
Schultz, R. and E. Dibble. 2012. Effects of invasive macrophytes on freshwater fish and
macroinvertebrate communities: the role of invasive plant traits. Hydrobiologia 684:1-14.
Shaftel, R., R. King, and J. Back. 2011. Breakdown rates, nutrient concentrations, and macroinvertebrate
colonization of bluejoint grass litter in headwater streams of the Kenai Peninsula, Alaska. Journal of
the North American Benthological Society 30:386-398.
Slaughter, C.W. 1982. Occurrence and recurrence of aufeis in an upland taiga catchment. 4th Can.
Permafrost Conference 1982:182-188.
Smokorowski, K. E. and T. C. Pratt. 2007. Effect of a change in physical structure and cover on fish and
fish habitat in freshwater ecosystems: A review and meta-analysis. Environmental Reviews 15:15-
41.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 1549April 2013
-------
Chapter 15 References
Sommer, T. R., M. L. Nobriga, W. C. Harrell, W. Batham, and W. J. Kimmerer. 2001. Floodplain rearing of
juvenile Chinook salmon: evidence of enhanced growth and survival. Canadian Journal of Fisheries
and Aquatic Sciences 58:325-333.
Spellman, B. T. and T. L. Wurtz. 2011. Invasive sweetclover (Melilotus alba) impacts native seedling
recruitment along floodplains of interior Alaska. Biological Invasions 13:1779-1790.
Stanford, J. A. and J. V. Ward. 1993. An ecosystem perspective of alluvial rivers - connectivity and the
hyporheic corridor. Journal of the North American Benthological Society 12:48-60.
Suttle, K., M. Power, J. Levine, and C. McNeely. 2004. How fine sediment in riverbeds impairs growth and
survival of juvenile salmonids. Ecological Applications 14:969-974.
Urgenson, L. S., S. H. Reichard, and C. B. Halpern. 2009. Community and ecosystem consequences of giant
knotweed (Polygonum sachalinense] invasion into riparian forests of western Washington, USA.
Biological Conservation 142:1536-1541.
USEPA (U.S. Environmental Protection Agency). 2003a. Final Environmental Impact Statement, Pogo Gold
Mine Project. Seattle, WA: Region 10.
USEPA (U.S. Environmental Protection Agency). 2003b. EPA and Hardrock Mining: A Source Book for
Industry in the Northwest and Alaska. Appendix H. Erosion and Sedimentation. Seattle, WA: Region
10.
USEPA (U.S. Environmental Protection Agency). 2006. Industrial Stormwater Fact Sheet Series. Sector G:
Metal Mining (Ore Mining and Dressing) Facilities. EPA-833-F-06-022. Office of Water.
USFWS (U.S. Fish and Wildlife Service). 2012. National Wetland Inventory. Available:
. Accessed: December 11, 2012.
USGS (U.S. Geological Survey). 2012. National Hydrography Dataset, High Resolution, Alaska. Available:
ftp://nhdftp.usgs.gov/DataSets/Staged/States/FileGDB/HighResolution . Accessed: October 16,
2012.
Van Bohemen, H. D. and W. H. J. Van de Laak. 2003. The influence of road infrastructure and traffic on
soil, water, and air quality. Environmental Management 31:50-68.
Van Hassel, J. H., J. J. Hey, and D. L. Garling. 1980. Heavy metals in a stream ecosystem at sites near
highways. Transactions of the American Fisheries Society 109:636-643.
Walker, D. A. and K. R. Everett. 1987. Road dust and its environmental impact on Alaskan taiga and
tundra. Arctic and Alpine Research 19:479-489.
Walker, C. M., R. S. King, D. F. Whigham, and S. J. Baird. 2012. Landscape and wetland influences on
headwater stream chemistry in the Kenai Lowlands, Alaska. Wetlands 32:301-310.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-50 April 2013
-------
Chapter 15 References
Ware, C., D. M. Bergstrom, E. Muller, and I. G. Alsos. 2012. Humans introduce viable seeds to the Arctic on
footwear. Biological Invasions 14:567-577.
Weaver, W.E., D.K. Hagans, and M.A. Made). 1987. Managing forest roads to control cumulative erosion
and sedimentation effects. Pages 119-124 in Proceedings of the California Watershed Management
Conference, 18-20 November, 1986, West Sacramento, California. Report 11. Berkeley, CA: Wildland
Resources Center, University of California.
Weber Scannell, P. K. and L. K. Duffy. 2007. Effects of total dissolved solids on aquatic organisms: A
review of literature and recommendation for salmonid species. American Journal of Environmental
Sciences 3:1-6.
Whitton, B. A., N. T. W. Ellwood, and B. Kawecka. 2009. Biology of the freshwater diatom
Didymosphenia: a review. Hydrobiologia 630:1-37.
Wipfli, M. S. and C. V. Baxter. 2010. Linking ecosystems, food webs, and fish production: subsidies in
salmonid watersheds. Fisheries 35:373-387.
Wissmar, R. C., R. K. Timm, and M. D. Bryant. 2010. Radar-derived digital elevation models and field-
surveyed variables to predict distribution of juvenile coho salmon and Dolly Varden in remote
streams of Alaska. Transactions of the American Fisheries Society 139:288-302.
Withycombe, E. and R. Dulla. 2006. Alaska Rural Dust Control Alternatives. Report No. SR2006-03-03.
Sierra Research, Inc. for Alaska Department of Environmental Conservation.
Woody, C. A. and S. L. O'Neal. 2010. Fish Surveys in Headwater Streams of the Nushagak and Kvichak River
Drainages, Bristol Bay, Alaska, 2008-2010. Anchorage, AK: Fisheries Research and Consulting.
Wurtz, T. L., M. J. Macander, and B. Spellman, T. 2010. Spread of invasive plants from roads to river
systems in Alaska: A network model. Pages 699-708 in J. M. Pye, H. M. Rauscher, Y. Sands, D. C. Lee,
and J. S. Beatty (eds.), Advances in Threat Assessment and Their Application to Forest and Rangeland
Management. PNW-GTR-802. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific
Northwest and Southern Research Stations.
Young, M. K., P. M. Geunther-Gloss, and A. D. Ficke. 2005. Predicting cutthroat trout (Oncorhynchus
clarki) abundance in high-elevation streams: revisiting a model of translocation success. Canadian
Journal of Fisheries and Aquatic Sciences 62:2399-2408.
Personal Communications
Morstad, Steve. Fishery Biologist III, ADF&G. September 2011—Email of unpublished data to Rebecca
Shaftel.
15.1.11 Chapter 11—Pipeline Failures
ADEC (Alaska Department of Environmental Conservation). 2011. Water Quality Standards: Alaska
Department of Environmental Conservation Report 18-AAC-70.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-51April 2013
-------
Chapter 15 References
ADEC (Alaska Department of Environmental Conservation). 2012. Situation Report: Fairbanks Gold
Mining, Inc. Cyanide Water Spill. Anchorage AK: Alaska Department of Environment and
Conservation, Division of Spill Prevention and Response, Prevention and Emergency Response
Program.
Alexander, V. and K. VanCleve. 1983. The Alaska pipeline: a success story. Annual Review of Ecology and
Systematics 14:443-463.
API (American Petroleum Institute). 2003. Robust Summary of Information on Distillate Fuels. Report No.
201-14835B2. Washington DC: American Petroleum Institute.
Barrett, D. 2012. Notice of Probable Violation and Proposed Civil Penalty: Letter to Mr. Richard Adams,
Vice President, U.S. Operations, Enbridge Energy, Limited Partnership. Kansas City, MO: U.S.
Department of Transportation, Pipeline and Hazardous Materials Safety Administration.
Bass, M. L., J. R. Voshell, and R. W. Young. 1987. Assessment of damage to aquatic communities resulting
from a refined oil product spill. Pages 353-378 in J. H. Vandermeulen and S. E. Hrudey (eds.), Oil in
Freshwater: Chemistry, Biology, Countermeasure Technology. Pergamon Press, New York.
Briceno, F. and F. Bajak. 2012. Townsfolk sickened after Peru toxic spill. BusinessWeek.
http://www.businessweek.com/ap/2012-08-16/townsfolks-sickened-after-toxic-spill
Bury, R. B. 1972. The effects of diesel fuel on a stream fauna. California Fish and Game 58:291-295.
Caleyo, F. 2007. Statistical and tolerance uncertainties in pipeline failure rate estimates. Noveno Congreso
y Expo Internacional de Ductos, Villahermosa, Mexico.
Caterpillar Global Mining. 2009. Minera Antamina: Setting the Standard, Driving the Future. Viewpoint:
Perspectives on Modern Mining 2009:12-19.
Clap, L. 2004. Xstrata's "Best Practice" in Argentina. MAC: Mines and Communities. Available:
.
Coghlan, S. M., Jr. and J. L. Lund. 2005. Rapid assessment of benthic faunal responses to a small
petroleum spill in a headwaters stream. Journal of Freshwater Ecology 20:777-779.
Das, P. K. M. K. and S. K. Konar. 1988. Acute toxicity of petroleum products, crude oil and oil refinery
effluent on plankton, benthic invertebrates and fish. Environment and Ecology 6:885-891.
DNV Consulting. 2006. Frequency-Volume Study of Keystone Pipeline. Report No. 70015849-2 rev 1.
Report for TransCanada Pipelines, Ltd.
Earthworks. 2012. U.S. Copper Porphyry Mines Report: The Track Record of Water Quality Impacts
Resulting from Pipeline Spills, Tailings Failures and Water Collection and Treatment Failure.
Washington, DC.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-52 April 2013
-------
Chapter 15 References
ERCB (Energy Resources Conservation Board). 2013. ST57-2012: Field Surveillance and Operations
Branch - Field Operations Provincial Summary 2011. Calgary, Alberta, Canada.
EUB (Energy and Utilities Board). 1998. Pipeline Performance in Alberta 1980-1997. Calgary, Alberta,
Canada: Alberta Energy and Utilities Board.
Geosphere and CH2M Hill. 2006. Hydrocarbon Characterization for Use in the Hydrocarbon Risk
Calculator and Example Characterizations of Selected Alaska Fuels, Technical Background Document.
Draft report prepared for the Alaska Statement of Cooperation Working Group.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska. February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
Giddings,}., B. Parkhurst, C. Gehrs, and R. Millemann. 1980. Toxicity of a coal liquefaction product to
aquatic organisms. Bulletin of Environmental Contamination and Toxicology 25:1-6.
Goldstein, J. N., D. F. Woodward, and A. M. Farag. 1999. Movements of adult Chinook salmon during
spawning migration in a metals-contaminated system, Coeur d'Alene River, Idaho. Transactions of
the American Fisheries Society 128:121-129.
Guard, H. E., N. G. James, and R. B. Laughlin, Jr. 1983. Characterization of gasolines, diesel fuels and their
water soluble fractions. Oakland, CA: Naval Biosciences Laboratory, Naval Supply Center.
Hayes, M. 0., R. Hoff, J. Michel, D. Scholz, and G. Shigenaka. 1992. Introduction to Coastal Habitats and
Biological Resources for Oil Spill Response. NMRAD-92-4. Seattle, WA: National Oceanic and
Atmospheric Administration, Coastal Monitoring and Bioeffects Assessment Division.
Hebert, G. W. and R. H. Kussat. 1972. A Laboratory Evaluation of the Toxicity of Certain Oils and Chemical
Oil Dispersants to Juvenile Coho Salmon and Staghorn Sculpins. Canada Department of the
Environment, Fisheries Service, Pacific Region, Northern Operations Branch. Technical Report 1972-
6.
HydroQual. 2007. Biotic Ligand Model Windows Interface, Version 2.2.3, User's Guide and Reference
Manual. Mahwah, NJ: HydroQual, Inc.
Ingersoll, C. G., P. S. Haverland, E. L. Brunson, T. J. Canfield, F. J. Dwyer, C. E. Henke, N. E. Kemble, D. R.
Mount, and R. G. Fox. 1996. Calculation and evaluation of sediment effect concentrations for the
amphipod Hyalella azteca and the midge Chironomus riparius. Journal of Great Lakes Research
22:602-623.
Khan, N., M. A. Warith, and G. Luk. 2007. A comparison of acute toxicity of biodiesel, biodiesel blends,
and diesel on aquatic organisms. Journal of the Air and Waste Management Association 57:286-296.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-53April 2013
-------
Chapter 15 References
Kubach, K. M., M. C. Scott, and J. S. Bulak. 2011. Recovery of a temperate riverine fish assemblage from a
major diesel oil spill. Freshwater Biology 56:503-518.
Lockhart, W. L, R. W. Danell, and D. A. Murray. 1987. Acute toxicity bioassays with petroleum products:
influence of exposure conditions. Pages 335-344 in J. H. Vandermeulen and S. E. Hrudey (eds.), Oil in
Freshwater: Chemistry, Biology, Countermeasure Technology. New York: Pergamon Press.
Lytle, D. A. and B. L. Peckarsky. 2001. Spatial and temporal impacts of a diesel fuel spill on stream
invertebrates. Freshwater Biology 46:693-704.
MacDonald, D. D., C. G. Ingersoll, and T. A. Berger. 2000. Development and evaluation of consensus-based
sediment quality guidelines for freshwater ecosystems. Archives of Environmental Contamination
and Toxicology 39:20-31.
Meyer, J., W. Adams, K. Brix, D. Mount, W. Stubblefield, and C. Wood. 2005. Toxicity ofDietborne Metals to
Aquatic Organisms. Pensacola, FL: SETAC Press.
Minera Alumbrera. 2004. Sustainability Report for 2004. Minera Alumbrera Ltd. andXstrata Copper,
Catamarca, Argentina. Available: .
Minera Alumbrera. 2005. Minera Alumbrera Sustainability Report2005. Minera Alumbrera YMAD-UTE
and Xstrata Copper, Catamarca, Argentina. Available:
.
Minera Alumbrera. 2006. Minera Alumbrera Sustainability Report 2006. Minera Alumbrera and Xstrata
Copper, Catamarca, Argentina. Available:
.
Minera Alumbrera. 2007. Minera Alumbrera Sustainability Report 2 007. Minera Alumbrera andXstrata
Copper, Catamarca, Argentina. Available:
.
Minera Alumbrera. 2008. Minera Alumbrera Sustainability Report2008. Minera Alumbrera YMAD-UTE
and Xstrata Copper, Catamarca, Argentina. Available;
.
Minera Alumbrera. 2009. Minera Alumbrera Sustainability Report2009. Minera Alumbrera YMAD-UTE
and Xstrata Copper, Catamarca, Argentina. Available:
.
Minera Alumbrera. 2010. Minera Alumbrera Sustainability Report2010. Minera Alumbrera YMDA-UTE,
and Xstrata Copper, Catamarca, Argentina. Available:
.
Mining Watch Canada. 2005. Wheaton River Profits from Destruction of Bajo La Alumbrera, Argentina.
Mines Alerte. Available: .
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-54 April 2013
-------
Chapter 15 References
Taj, M. and T. Cespedes. 2012. Peru's Antamina mine faces fallout from spill. Reuters.com. Accessed:
August 2 0,2012.
Mos, L, G. A. Cooper, K. Serben, M. Cameron, and B. F. Koop. 2008. Effects of diesel on survival, growth,
and gene expression in rainbow trout (Oncorhynchus mykiss) fry. Environmental Science and
Technology 42:2656-2662.
Nauman, J. W. and D. R. Kernodle. 1975. The effect of a fuel oil spill on benthic invertebrates and water
quality on the Alaskan Arctic Slope, Happy Valley Creek near Sagwon, Alaska, journal of Research of
the U.S. Geological Survey 3:495-500.
NICNAS (National Industrial Chemicals Notification and Assessment Scheme). 1995. Sodium Ethyl
Xanthate, Priority Existing Chemical No. 5. Full Public Report. Canberra, Australia: Australian
Government Publishing Service.
NICNAS (National Industrial Chemicals Notification and Assessment Scheme). 2000. Sodium Ethyl
Xanthate, Priority Existing Chemical Secondary Notification Assessment. Report No. 5S. Canberra,
Australia: Australian Government Publishing Service.
NOAA. 2006. Small diesel spills (50-5,000 gallons). Seattle, WA: National Oceanic and Atmospheric
Administration, Office of Response and Restoration.
NOAA and API. 1994. Options for Minimizing Environmental Impacts of Freshwater Spill Response.
National Oceanic and Atmospheric Administration, Hazardous Materials Response and Assessment
Division and the American Petroleum Institute.
O'Brien's Response Management. 2009. TransCanada Emergency Response Plan: Worst Case Discharge
Analysis and Scenarios. Appendix B, Version 21.0.1. O'Brien's Response Management Inc.
OGP. 2010. Risk Assessment Data Directory, "Riser & pipeline release frequencies." Report No. 434-4,
International Association of Oil & Gas Producers.
Overstreet, R. and J. A. Gait. 1995. Physical Processes Affecting the Movement and Spreading of Oils in
Inland Environments. HMRAD 95-7. Seattle, WA: National Oceanic and Atmospheric Administration,
Hazardous Materials Response and Assessment Division.
Pereira. S., V. Araiijo, M Reboucas, F. Vieira, M. de Almeida, F. Chinalia, and I. Nascimento. 2012. Toxicity
of biodiesel, diesel, and biodiesel/diesel blends: comparative sub-lethal effects of water-soluble
fractions to microalgae species. Bulletin of Environmental Contamination and Toxicology 88:234-
238.
PHMSA (Pipeline and Hazardous Materials Safety Administration). 2012. Pipeline Basics. Pipeline Safety
Stakeholder Communications, U.S. Department of Transportation. Available:
. Accessed: February 21,
2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-55April 2013
-------
Chapter 15 References
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008.
Anchorage, AK. Available: .
Redman, A. D., T. F. Parkerton, J. A. McGrath, and D. M. Di Toro. 2012. PETROTOX: An aquatic toxicity
model for petroleum substances. Environmental Toxicology and Chemistry 31:2498-2506.
Rice, S., J. Short, M. Carls, A. Moles, and R. Spies. 2007. The Exxon Valdez oil spill. Pages 419-520 in R.
Spies, editor. Long-Term Ecological Change in the Northern Gulf of Alaska. Elsevier B.V., Amsterdam,
The Netherlands.
Schein, A., J. A. Scott, L. Mos, and P. V. Hodson. 2009. Oil dispersion increases the apparent bioavailability
and toxicity of diesel to rainbow trout (Oncorhynchus mykiss]. Environmental Toxicology and
Chemistry 28:595-602.
Smith, P., D. Snook, A. Muscatt, and A. Smith. 2010. Effects of a diesel spill on freshwater
macroinvertebrates in two urban watercourses, Wiltshire, UK. Water and Environment Journal
24:249-260.
Suter, G. W., II and S. M. Cormier. 2011. Why and how to combine evidence in environmental
assessments: Weighing evidence and building cases. Science of the Total Environment 409:1406-
1417.
Taj, M. and T. Cespedes. 2012. Peru's Antamina mine faces fallout from spill. Reuters.com. Accessed:
August 20, 2012.
URS. 2000. Final Longhorn Pipeline Environmental Assessment, Volume 4 Responsiveness Summary,
Appendix D, Leak Frequencies of Other Pipeline Operations.
Van Derveer, W. D., R. J. Nadeau, and G. L. Case. 1995. A screening-level evaluation of impacts to a
Montana lotic macroinvertebrate community from a fuel oil spill. Pages 391-397 in Proceedings of
the 14th Bien, API et al. Oil Spill (Achieving and Maintaining Preparedness) Internal Conference. Long
Beach, CA. February 27-March 2,1995.
Wan, M. T., R. G. Watts, and D. J. Moul. 1990. Acute toxicity to juvenile pacific salmonids and rainbow
trout of butoxyethyl esters of 2,4-D, 2,4-DP and their formulated product: Weedone CB and its
carrier. Bulletin of Environmental Contamination and Toxicology 45:604-611.
Personal Communication
Adams, William. General Manager Legacy, Rio Tinto. January 17, 2013—Email to Glenn Suter.
15.1.12 Chapter 12—Fish-Mediated Effects
ARCO (Atlantic Richfield Company). 1998. Milltown Reservoir Sediments NPL Site, Clark Fork River
Operable Unit, Final Draft Remedial Investigation Report. Anaconda, MT.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-56April 2013
-------
Chapter 15 References
Bersamin, A., B. Luick, E. Ruppert, J. Stern, and S. Zidenberg-Cherr. 2006. Diet quality among Yup'ik
Eskimos living in rural communities is low: The center for Alaska native health research pilot study.
Journal of 'the American Dietetic Association 106:1055-1063.
Braund, S. R. and Associates. 2009. Impacts of Oil and Gas Development to Barrow, Nuiqsut, Wainwright,
andAtqasuk harvesters. Report prepared for North Slope Borough, Department of Wildlife
Management
Braund, S. R. and J. Kruse (eds). 2009. Synthesis: Three Decades of Research on Socioeconomic Effects
Related to Offshore Petroleum Development in Coastal Alaska. MMS OCS Study 2009-006.
Braund, S. R. and Associates. 2011. Chapter 23: Subsistence Uses and Traditional Knowledge, Bristol Bay
Drainages. In Pebble Limited Partnership. 2011. Pebble Project Environmental Baseline Document
2004 through 2008. Available:.
Brna, P .J. and L. A. Verbrugge (eds.). 2013. Wildlife Resources of the Nushagak and Kvichak River
Watersheds, Alaska. Final Report. Anchorage, AK: Anchorage Fish and Wildlife Field Office, U.S. Fish
and Wildlife Service. 177 pages.
CEAA (Canadian Environmental Assessment Agency). 2010. Reportof the Federal Review Panel
Established by the Minister of the Environment: Taseko Mines Limited's Prosperity Gold-Copper Mine
Project. Ottawa, Ontario [cited 2011 May 4]. Available:
.
Chapman, P. M., W. J. Adams, M. L. Brooks, C. G. Delos, S. N. Luoma, W. A. Maher, H. M. Ohlendorf, T. S.
Presser, and D. P. Shaw. 2010. Ecological Assessment of Selenium in the Aquatic Environment.
Pensacola, FL: Society of Environmental Toxicology and Chemistry (SETAC).
Eisler, R. 2000. Handbook of Chemical Risk Assessment: Health Hazards to Humans, Plants, and Animals.
Volume 1: Metals. Boca Raton, FL: Lewis Publishers. 738 pages.
Goldsmith, S. 2007. The Remote Rural Economy of Alaska. Institute of Social and Economic Research,
University of Alaska Anchorage. April 12. Available:
.
Helfield, J. M. and R. J. Naiman. 2006. Keystone interactions: salmon and bear in riparian forests of
Alaska. Ecosystems 9:167-180.
Hilderbrand, G. V., T. A. Hanley, C. T. Robbins, and D. D. Schwartz. 1999. Role of brown bears (Ursus
arctos) in the flow of marine nitrogen into a terrestrial ecosystem. Oecologia 121:546-550.
Holen, D. 2009. The dynamic context of cultural and social sustainability of communities in southwest
Alaska. Journal of Enterprising Communities 3:306-316.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-57 April 2013
-------
Chapter 15 References
Joyce, A. 2008. Risk and Opportunity in British Columbia Shellfisheries: The Role of Limited Property
Rights in Aquaculture Development. University of British Columbia, Vancouver, British Columbia
[cited 2011 May 4]. Available: .
Kruse, J. A., M. Baring-Gould, W. Schneider, J. Gross, G. Knapp and G. Sherrod. 1983. A Description of the
Socioeconomics of the North Slope Borough. Technical Report No. 85. Prepared by the Institute of
Social and Economic Research, University of Alaska for the Minerals Management Service, Alaska
Outer Continental Shelf Region, Anchorage, AK.
Kuhnlein, H., 0. Receveur, and H. Chan. 2001. Traditional food systems research with Canadian
indigenous peoples. International Journal ofdrcumpolar Health 60:112-112.
Loring, P. A., L. K. Duffy, and M. S. Murray. 2010. A risk-benefit analysis of wild fish consumption for
various species in Alaska reveals shortcomings in data and monitoring needs. Science of the Total
Environment 408:4532-4541.
NMFS (National Marine Fisheries Service). 2011. Effects of Oil and Gas Activities in the Arctic Ocean, Draft
Environmental Impact Statement. United States Department of Commerce, National Oceanic and
Atmospheric Administration, National Marine Fisheries Service, Office of Protected Resources.
Available: (update August
2012).
NRC (National Research Council). 2003. Cumulative Environmental Effects of Oil and Gas Activities on
Alaska's North Slope. Washington, DC: The National Academies Press.
Palinkas, L. A., M. A. Downs, J. S. Petterson, and J. Russell. 1993. Social, cultural, and psychological
impacts of the Exxon Valdez oil spill. Human Organization 52:1-12.
PLP (Pebble Limited Partnership). 2011. Environmental Baseline Document 2004 through 2008.
Anchorage, AK. Available: .
Poppel, B., J. Kruse, G. Duhaime, and L. Abryutina. 2007. Survey of Living Conditions in the Arctic (SLiCA)
Results. Anchorage: Institute of Social and Economic Research, University of Alaska Anchorage [cited
2011 May 4]. Available: .
Storey, K. and L. C. Hamilton. 2004. Planning for the impacts of megaprojects: Two North American
examples. Pages 28-302 in R. 0. Rasmussen and N. E. Koroleva (eds.), Social and Economic Impacts
in the North. Dordrecht, Netherlands: Kluwer Academic Publishers.
USEPA (U.S. Environmental Protection Agency). 1985. Ambient Water Quality Criteria for Copper - 1984.
EPA 440/5-84-031. Washington, DC.
USEPA (U.S. Environmental Protection Agency). 2009. Red Dog Mine Extension Aqqaluk Project, Final
Environmental Impact Statement, October.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-58April 2013
-------
Chapter 15 References
USEPA (U.S. Environmental Protection Agency). 2011. A Field-Based Aquatic Life Benchmark for
Conductivity in Central Appalachian Streams. Washington, DC: Office of Research and Development,
National Center for Environmental Assessment. EPA/600/R-10/023F. Available:
15.1.13 Chapter 13—Cumulative Effects of Large-Scale Mining
ADF&G (Alaska Department of Fish and Game). 2012. Alaska Freshwater Fish Inventory Database.
Available: .
ADNR (Alaska Department of Natural Resources). 2005. Bristol Bay Area Plan (BBAP) for State Lands.
Resource Assessment and Development Section, Division of Mining, Land, and Water. Available:
. Accessed: December 27, 2012.
ADNR (Alaska Department of Natural Resources). 2012a. Temporary Water Use Permit F2012-275 for
TNR Gold Corp (Shotgun Prospect). Accessed: November 29, 2012 at
.
ADNR (Alaska Department of Natural Resources). 2012b. Case Abstract, ADL 656311,James W. Caswell.
Available:
.
Accessed: December 10, 2012.
ADNR (Alaska Department of Natural Resources). 2012c. Case Abstracts ADL 701294 etal, Alaska Earth
Resources, Inc., Kody andKoksetna claims. Available:
.
Accessed: December 10, 2012.
ADNR (Alaska Department of Natural Resources). 2012d. Case Abstracts ADL 644576 etal., TNR Gold
Corp., Winchester claims. Available:
.
Accessed: December 10, 2012.
ADNR (Alaska Department of Natural Resources). 2012e. Case Abstracts, ADL 552871 etal., Full Metal
Minerals (USA), Inc., South Pebble claims. Available:
.
Accessed: December 10, 2012.
ADNR (Alaska Department of Natural Resources). 2012f. Case Abstracts, ADL 648509 et al., Alaska Earth
Services, Inc., Groundhog and Nika claims. Available:
.
Accessed: December 10, 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-59 April 2013
-------
Chapter 15 References
ADNR (Alaska Department of Natural Resources). 2012g. Case Abstracts ADL 709821 etal, Millrock
Alaska, LLC,AUDN claims. Available:
.
Accessed: December 10, 2012.
ADNR (Alaska Department of Natural Resources). 2012h. State Mining Claims Dataset, Alaska State Geo-
SpatialData Clearinghouse. Available: . Accessed: November 19,
2012.
ADOT (Alaska Department of Transportation and Public Facilities). 2001. Revised Environmental
Assessment and Finding of No Significant Impact, Iliamna-Nondalton Road Improvements. Project No.
STP-0214(3)/51951. Available: .
Accessed: December 18, 2012
ADOT (Alaska Department of Transportation and Public Facilities). 2004. Southwest Alaska
Transportation Plan. Anchorage, AK. Available:
. Accessed:
December 19, 2012.
ADOT (Alaska Department of Transportation and Public Facilities). 2011a. Ambler Mining District Access.
Available: . Accessed: December 6, 2012.
ADOT (Alaska Department of Transportation and Public Facilities). 2011b. Aleknagik: Wood River Bridge.
Available: . Accessed: December 6, 2012.
AERI (Alaska Earth Resources, Inc.). 2008. Affidavit of'Annual Labor for Mining (Kamishak claims). 2008-
000296-0. Available:
.
Accessed: November 29, 2012.
AREA. 2011. Alaska Hardrock Exploration Application A113099 by Alaska Earth Sciences, Inc., for
Groundhog prospect. January 31.
AREA. 2012. Alaska Hardrock Exploration Application A122793 by Big Chunk Corp., July 11, 2012.
Alix Resources. 2008. News Release. Available:
.
Accessed: January 3, 2013.
Big Chunk Corp. 2012. Affidavit of Annual Labor for Mining, Big Chunk Corporation/Liberty Star. Doc. No.
2012-000351-0. Available:
.
Accessed: November 29, 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-60 April 2013
-------
Chapter 15 References
Bristol Exploration Co., Inc. 2011. Affidavit of'Annual Labour for Mining (HSS Claims). Doc. No. 2011-
000463-0. Available:
.
Accessed: November 29, 2012.
Brna, P .J. and L. A. Verbrugge (eds.). 2013. Wildlife Resources of the Nushagak and Kvichak River
Watersheds, Alaska. Final Report. Anchorage, AK: Anchorage Fish and Wildlife Field Office, U.S. Fish
and Wildlife Service. 177 pages.
Charles, S., R. Bravo De La Parra, J. P. Mallet, H. Persat, and P. Auger. 2000. Annual spawning migration in
modeling brown trout population dynamics inside an arborescent river network. Ecological
Modelling 133:15-31.
Detterman, R. L. and B. L. Reed. 1973. Surficial Deposits of the Iliamna Quadrangle, Alaska. Geological
Survey Bulletin 1368-A. U.S. Geological Survey. Available:
. Accessed: December 17, 2012.
Fall, J. A., D. L. Holen, B. Davis, T. Krieg, D. Roster. 2006. Subsistence Harvests and Uses of Wild Resources
in Iliamna, Newhalen, Nondalton, Pedro Bay, and Port Alsworth, Alaska 2004. Technical Paper No.
302. Juneau, AK: Alaska Department of Fish and Game, Division of Subsistence.
Fall, J. A., J. C. Schichnes, M. Chythlook, and R. J. Walker. 1986. Patterns of Wild Resources Use in
Dillingham: Hunting and Fishing in an Alaskan Regional Center. Technical Paper No. 135. Division of
Subsistence, Alaska Department of Fish and Game.
Full Metal Minerals, Inc. 2008. Affidavit of Annual Labor for Mining (South Pebble claims). Doc. No. 2008-
000298-0. Available:
.
Accessed: November 29, 2012.
Full Metal Minerals. 2012. Pebble South. Available:
. Accessed: November 29, 2012.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska. Prepared for Northern Dynasty
Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia. Document
1056140100-REP-R0001-00. 579 pages.
Golden Lynx, LLC. 2009. Affidavit of'Annual Labor for Mining (Kisa prospect). Doc. No. 2009-000419-0.
Available:
.
Accessed: November 29, 2009.
Hilborn, R. 2006. Fisheries success and failure: The case of the Bristol Bay salmon fishery. Bulletin of
Marine Science 78(3):487-498.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15^61 April 2013
-------
Chapter 15 References
Hilborn, R., T. Quinn, D. Schindler, and D. Rogers. 2003. Biocomplexity and fisheries sustainability.
Proceedings of the National Academy of Sciences of the United States of America 100:6564-6568.
Helen, D., J. Stariwat, T. M. Krieg, andT. Lemons. 2012. Subsistence Harvest and Uses of Wild Resources in
Aleknagik, Clark's Point, and Manokotak, Alaska, 2008. Technical Paper No. 368. Anchorage, AK:
Alaska Department of Fish and Game, Division of Subsistence.
Johnson, J. and P. Blanche. 2012. Catalog of Waters Important for Spawning, Rearing, or Migration of
Anadromous Fishes - Southwestern Region, Effective June 1,2012. Special Publication No. 12-08.
Anchorage, AK: Alaska Department of Fish and Game.
Kennecott Exploration Co. 2011. State of Alaska Affidavit of Annual Labor for Mining Groundhog and Nika
Group of 484 Claims. Doc. No. 2011-000223-0. Available:
.
Accessed: November 29, 2012.
Krieg, T. M., D. L. Holen, D. Koster. 2009. Subsistence Harvests and Uses of Wild Resources in Igiugig,
Kokhanok, Koliganek, Levelock, and New Stuyahok, Alaska 2005. Technical Paper No. 322. Dillingham,
AK: Alaska Department of Fish and Game, Division of Subsistence.
Levelock Village Council. 2005. Levelock Watershed Community Planning Project, Final Summary Report.
Available: . Accessed:
December 14, 2012.
Liberty Star. 2012a. Big Chunk Super Project. Available:
. Accessed: November
29,2012.
Liberty Star. 2012b. Gold Copper Explorer Liberty Star Signs New Agreement with Northern Dynasty
Minerals Ltd, Southwest Alaska. NR 144. Available:
. Accessed:
Decembers, 2012.
Millrock Alaska. 2012a. Affidavit of'Annual Labor for State Mining Claims (AUDN), Millrock Alaska LLC.
Doc. No. 2012-000236-0. Available: <
http://dnr.alaska.gov/ssd/recoff/sag/DocDisplay.cfm?SelectedDoc=20120002360&District=310>.
Accessed: November 29, 2012.
Millrock Alaska. 2 012b. Affidavit of Annual Labor for State Mining Claims (KM), Millrock Alaska LLC. Doc.
No. 2012-000470-0. Available:
.
Accessed: November 29, 2012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-62 April 2013
-------
Chapter 15 References
Millrock Resources. 2011. Millrock Provides Exploration Update, Available:
. Accessed:
November 29, 2012.
Nehlsen, W., J. Williams, and J. Lichatowich. 1991. Pacific salmon at the crossroads—stocks at risk from
California, Oregon, Idaho, and Washington. Fisheries 16:4-21.
Northern Bonanza Trust. 2011. Affidavit of Annual Labor for Mining. Doc. No. 2011-000248-0. Available:
.
Accessed: November 30, 2012.
NRC (National Research Council). 1996. Upstream: Salmon and Society in the Pacific Northwest.
Washington, DC.
PLP (Pebble Limited Partnership). 2011. Affidavit of'Annual Labor for Mining. Doc. No. 2011-000252-0.
Available:
.
Accessed January 3, 2013.
Ruckleshaus, M. H., P. Levin, J. B. Johnson, and P. M. Kareiva. 2002. The Pacific salmon wars: What
science brings to the challenge of recovering species. Annual Review of Ecology Systematics 33:655-
706.
Schichnes, J. and M. Chythlook. 1991. Contemporary Use of Fish and Wildlife in Ekwok, Koliganek, and New
Stuyahok, Alaska. Technical Paper No. 185, Division of Subsistence, Alaska Department of Fish and
Game.
Schindler, D. E., R. Hilborn, B. Chasco, C. P. Boatright, T. P. Quinn, L. A. Rogers, and M. S. Webster. 2010.
Population diversity and the portfolio effect in an exploited species. Nature 465:609-612.
Stickman, K., A. Balluta, M. McBurney, and D. Young. 2003. K'ezghlegh, Nondalton Traditional Ecological
Knowledge of Freshwater Fish. Fisheries Project 01-075, Fisheries Information Services, U.S. Fish and
Wildlife Service. 65 pages. Available: . Accessed: December 11, 2012.
Stuy Mines, LLC. 2010. Affidavit of'Annual Labor for Mining. Doc. No. 2010-000602-0. Available:
.
Accessed: November 29, 2012.
Szumigala, D. J., L. A. Harbo, J. N. Adleman. 2011. Alaska's Mineral Industry 2010. Special Report 65.
Fairbanks, AK: Alaska Department of Natural Resources, Division of Geological and Geophysical
Surveys.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15-63 April 20:L3
-------
Chapter 15 References
Thor Gold Alaska, Inc. 2011. Affidavit of Annual Labor for Mining (Sleitat Mountain prospect). Doc. No.
2011-000443-0. Available:
.
Accessed: November 29, 2012.
TNR Gold Corp. 2011. Affidavit of'Annual Labour for Mining (Shotgun prospect). Doc. No. 2011-000202-0.
Available:
. Accessed: November 29, 2012.
TNR Gold Corp. 2012a. Shotgun/Winchester. Available: .
Accessed: November 29, 2012.
TNR Gold Corp. 2012b. Iliamna. Available: . Accessed:
November 29, 2012.
U5 Resources, Inc. 2010. Affidavit of'Annual Labor for State Mining Claims (Big Chunk claims). Doc. No.
2010-000-591-0. Available:
.
Accessed: January 3, 2013.
USCG (U.S. Coast Guard). 2012. Public Notices of Bridge Projects—Public Notice 17-02-12. Available:
. Accessed: December 10, 2012.
USFWS (U.S. Fish and Wildlife Service). 2012. National Wetland Inventory. Available:
. Accessed: December 11, 2012.
USGS (U. S. Geological Survey). 2008. Alaska Resource Data File, Fog Lake Prospect. Available:
. Accessed: January 3, 2013.
USGS (U.S. Geological Survey). 2012. National Hydrography Dataset. Last Revised: May 3. Available:
. Accessed: May 5, 2012.
Woody, C. A. and D. Young. 2006. Life History and Essential Habitats of Humpback Whiteflsh in Lake Clark
National Park, Kvichak River Watershed, Alaska. Annual Report FIS 05-403. Available:
-------
Chapter 15 References
Caldwell, J. A., and L. Charlebois. 2010. Tailings Impoundment Failures, Black Swans, Incident Avoidance
and Checklists. Pages 3-39 in Tailings and Mine Waste 2010: Proceedings of the 14th International
Conference on Tailings and Mine Waste, Vail, CO. October 17-20, 2010. Boca Raton, FL: CRC Press.
Dye, J. E. and C. J. Schwanke. 2009. Report to the Alaska Board of Fisheries for the Recreational Fisheries of
Bristol Bay, 2007,2008, and 2009. Alaska Department of Fish and Game, Division of Sport Fish,
Research and Technical Services.
Earthworks. 2012. U.S. Copper Porphyry Mines Report: The Track Record of Water Quality Impacts
Resulting from Pipeline Spills, Tailings Failures and Water Collection and Treatment Failure.
Washington, DC.
Ghaffari, H., R. S. Morrison, M. A., deRuijeter, A. Zivkovic, T. Hantelmann, D. Ramsey, and S. Cowie. 2011.
Preliminary Assessment of the Pebble Project, Southwest Alaska. February 15. Prepared for Northern
Dynasty Minerals Ltd., by WARDROP (a Tetra Tech Company). Vancouver, British Columbia.
Document 1056140100-REP-R0001-00. 579 pages.
Gibson, R. J., R. L. Haedrich, and C. M. Wernerheirn. 2005. Loss offish habitat as a consequence of
inappropriately constructed stream crossings. Fisheries 30:10-17.
Instanes, A., 0. Anisimov, L. Brigham, D. Goering, L. N. Khrustalev, B. Ladanyi, and J. 0. Larsen. 2005.
Infrastructure: buildings, support systems, and industrial facilities. Pages 907-944 in the Arctic
Climate Impact Assessment. New York, NY: Cambridge University Press.
Kuipers, J. R., A. S. Maest, K. A. MacHardy, and G. Lawson. 2006. Comparison of Predicted and Actual
Water Quality at Hardrock Mines: The Reliability of Predictions in Environmental Impact Statements.
Buka Environmental and Kuipers & Associates.
Kyhn, C. and B. Elberling. 2001. Frozen cover actions limiting AMD from mine waste deposited on land in
Arctic Canada. Cold Regions Science and Technology 32:133-142.
Langill, D. A. and P. J. Zamora. 2002. An Audit of 'Small Culvert Installations in Nova Scotia: Habitat Loss
and Habitat Fragmentation. 2422, Dartmouth, Nova Scotia: Canadian Department of Fisheries and
Oceans, Habitat Management Division.
Pearce, T. D., J. D. Ford, J. Prno, F. Duerden, J. Pittman, M. Beaumier, L. Berrany-Ford, and B. Smit. 2011.
Climate change and mining in Canada. Mitigation and Adaptation Strategies for Global Change
16:347-368.
Price, D. M., T. Quinn, and R. J. Barnard. 2010. Fish passage effectiveness of recently constructed road
crossing culverts in the Puget Sound region of Washington State. North American Journal of Fisheries
Management 30:1110-1125.
Schindler, D. E., R. Hilborn, B. Chasco, C. P. Boatright, T. P. Quinn, L. A. Rogers, and M. S. Webster. 2010.
Population diversity and the portfolio effect in an exploited species. Nature 465:609-612.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment ±5-65 April 2013
-------
Chapter 15 References
Schindler, D. E., X. Augerot, E. Fleishman, N. J. Mantua, B. Riddell, M. Ruckelshaus, J. Seeb, and M.
Webster. 2008. Climate change, ecosystem impacts, and management for Pacific salmon. Fisheries
33:502-506.
Vick, S. G. 2002. Degrees of Belief: Subjective Probability and Engineering Judgement. Reston, VA:
American Society of Civil Engineers.
Personal Communications
Buck, Gregory. Assistant Area Research Biologist (Bristol Bay). ADF&G, Anchorage, AK. August 30,
2011—Unpublished data provided by email to Rebecca S. Shaftel, University of Alaska, Anchorage.
15.2 GIS Base Map Citations
Alaska Department of Natural Resources. 2006. Alaska Legislatively Designated Areas—State Parks.
Anchorage, AK: Information Resource Management Section. Available:
Accessed: February 7, 2013.
Alaska Department of Natural Resources. 1998. Populated Alaska Places. Anchorage, AK: Information
Resource Management Section. Available: . Accessed: February 7,
2013.
Alaska Department of Transportation. 2012. Department of Transportation Road System
[DOT_RoadSystem_061212.zip]. Available:
. Accessed: October 31, 2012.
Alaska Department of Transportation. 2012. Public Airports in Alaska [Public_Airports_Oct2012.zip].
Available: . Accessed: October
31,2012.
Environmental Systems Research Institute. 2013. World Base—Light Gray Canvas. ArcMap 10.0.
Environmental Systems Research Institute. 2013. World Shaded Relief. ArcMap 10.0.
Environmental Systems Research Institute. 2007. U.S. States (Generalized). ESRI ArcMap 9.3 USA Census
Data.
Environmental Systems Research Institute. 2007. U.S. and Canada Water Polygons. ESRI ArcMap 10.0
Streetmap.
National Atlas of the United States. 2005. Federal Lands of the United States. Reston, VA. Available:
. Accessed: February 7, 2013.
U.S. Geological Survey. National Hydrography Dataset. High Resolution, Alaska [NHDH_AK_931v210.zip].
Available: . Accessed:
October 16, 2 012.
SECOND EXTERNAL REVIEW DRAFT-DO NOT CITE OR QUOTE
This document is a draft for review purposes only and does not constitute Agency policy.
Bristol Bay Assessment 15^66 April 2013
-------
Second External Review Draft - Do Not Cite or Quote
United States
Environmental Protection
Agency
iegion 10, Seattle, WA
Recycled/Recyclable
Printed with vegetable-based ink on
contains a minimum of 50% post-consumer
fiber and is processed chlorine free.
------- |