March 12, 2014 Lisa Fay, EIS Project Manager Minnesota Department of Natural Resources Division of Ecological and

Water Resources Environmental Review Unit 500 Lafayette Road, Box 25 St. Paul, MN 55155-4025 Re: Comments on the NorthMet Project Supplemental Draft Environmental Impact Statement (December 2013) Dear Ms. Fay: On behalf of the board, staff, and members of the Friends of the Boundary Waters Wilderness (Friends), I submit this letter and attached documents as comments on the NorthMet Project Supplemental Draft Environmental Impact Statement (SDEIS). Thank you for this opportunity to provide our input, information, concerns and recommendations for this proposed project of PolyMet Mining, Inc. The mission of the Friends of the Boundary Waters Wilderness is to protect, preserve and restore the wilderness character of the Boundary Waters Canoe Area Wilderness (BWCAW) and the Quetico-Superior Ecosystem. We have nearly 3,000 members in Minnesota and across the United States, and regularly communicate with about 10,000 supporters. Our organization values healthy ecosystems, clean water, wilderness character, and primitive recreation. Our members enjoy the Superior National Forest and the BWCAW for canoeing, camping, fishing, hunting, bird-watching, and many other reasons, as well as the regions natural, largelyundeveloped character. The risks to many of these activities and attributes from nonferrous mining have been a significant concern for our organization for many years. The Friends has conducted an extensive analysis of PolyMets proposed mine near Hoyt Lakes, MN. As part of our review, we worked with six technical experts with knowledge in the areas of geophysics, hydrology, geochemistry, mining engineering, and wetlands ecology. These individuals studied the SDEIS and its associated technical background documents. Their written reports include relevant information, concerns and recommendations. These reports are attached and should be considered part of the official comments by the Friends. Friends of the Boundary Waters Wilderness 401 North Third Street, Suite 290, Minneapolis, MN 55401 612-332-9630 www.friends-bwca.org

Listed here are the individuals and their areas of focus in reviewing the SDEIS: Dr. David Chambers, geophysicist, and Stuart Levit, mining reclamation and bonding; focus: mining engineering and planning, reclamation and financial assurance Dr. Paul Glaser, wetland geohydrologist; focus: hydrology and wetlands Dr. Ann Maest, geochemist; focus: geochemistry and water quality Dr. Glenn Miller, geochemist; focus: water quality and treatment Dr. Tom Myers, hydrologist; focus: hydrologic modeling Dr. Michael A. Malusis, geotechnical and geoenvironmental engineer; focus: barrier and containment strategies

The technical experts have included their credentials with their reports. In our comments below, we refer to these reports and cite these individuals, but their entire reports should be read for their complete analyses and we ask that the agencies respond not only to the issues we identify in our comments but to each issue and concern identified in the expert reports. The Friends appreciates the effort and time involved in developing the SDEIS. It is evident that a great deal of resources were involved in preparing it. However, despite the large amount of information provided in the SDEIS, many significant issues associated with this proposed project remain unaddressed. Based on our review of the SDEIS and guided by input from the technical experts, the Friends has significant concerns about the PolyMet proposed mining operation as designed. During the summer of 2013, the Friends as part of a coalition called Mining Truth, posed four clean water principles that should be adhered to when contemplating sulfide mining projects in Minnesota: 1. Will Minnesotas water stay safe and clean? 2. Are there strong safeguards in place for when things go wrong? 3. Will the company leave the site clean and maintenance free? 4. Will Minnesotas taxpayers be protected? On September 25, 2013, Mining Minnesota, a sulfide mining trade association representing PolyMet, responded with a letter to Governor Dayton. The letter noted: We agree that these are all good questions and that there are fact-based answers that will give you and all the citizens of Minnesota the utmost confidence that new coppernickel mining projects can bring unprecedented economic opportunity to Minnesota while protecting our precious natural environment. In short, can Minnesota enjoy both mining growth and a healthy environment? The answer is unequivocally YES (Mining Minnesota, 2013). The Friends is alarmed that the SDEIS for the PolyMet proposal fails to meet any of the four clean water principles and fails to match the commitment made by Mining Minnesota to adhere to these principles. We also believe the state and federal Co-lead Agencies reviewing this

project can and must make substantial changes to the SDEIS so that they too can comply with these clean water principles. Our comments detail the ways the SDEIS fails to meet these clean water principles. We also identify additional significant concerns with the proposed mine project. The flaws of this project pose unacceptable risks to human health and the environmental well-being of a potentially large area of northeastern Minnesota. Four years ago, the 2009 Draft Environmental Impact Statement for this project was delayed for critical revisions to be made. Unfortunately, this revision, the SDEIS, continues to contain fundamental problems that need to be addressed before this project can proceed. Given this level of risk and the failure of the mine design to meet basic clean water principles, the Friends cannot support the PolyMet project as proposed. We recommend the Co-lead Agencies either select the No Action Alternative for this project, or make substantial revisions to the design of the proposed mine and re-submit for public review. We provide detailed recommendations for critically important revisions to this project. Thank you for your consideration of our input and recommendations. Sincerely,

Betsy Daub Policy Director Friends of the Boundary Waters Wilderness Attachments: 1. Dr. Michael Malusis: technical report on viability of the tailings and waste containment systems. 2. Dr. David Chambers and Stuart Levit: technical report on mining engineering, reclamation and financial assurance 3. Dr. Paul Glaser: technical report on hydrology and wetlands 4. Dr. Anne Maest, technical report on geochemistry and water quality 5. Dr. Glenn Miller: technical report on geochemistry 6. Dr. Tom Myers: technical report on hydrologic modeling (3 part report) 7. Dr. Michael A. Malusis: technical report on the viability of the tailings and waste rock containment systems 8. CVs for the technical experts (6) 9. Edison Investment Research Limited: PolyMet Mining Corp: Low-cost polymetallic development project. November 21, 2013 10. Earthworks: U.S. copper porphyry mines: the track record of water quality impacts resulting form pipeline spills, tailings failures and water collection and treatment failures. July 2012.

11. State of Alaska, Department of Health and Social Services: Technical Guidance for Health Impact Assessment (HIA) in Alaska. July 2011. 12. Human Impact Partners. Frequently Asked Questions About Integrating Health Impact Assessment into Environmental Impact Assessment. 2014. 13. Mining Minnesota: Letter to Governor Dayton re: Minnesota should say yes to coppernickel mineral development. September 25, 2013.

Table of Contents I. Failure to Meet Clean Water Principle #1: Minnesotas Water Will Not Stay Safe and Clean A. 200 to 500 Years of Water Pollution B. Millions of Gallons of Polluted Water Will Escape Capture C. SDEIS Not Complete or Transparent About Water Quality Impacts D. A Critical and Risky Mitigation Plan Not Described in SDEIS E. Major Flaws With the Water Model F. Mercury Contamination G. Waste Rock Production and Management H. Tailings Basin and Hydrometallurgical Residue Facility Pollution I. Transportation Corridor Pollution Concerns J. Cumulative Impacts K. Concerns About Reliability of Reverse Osmosis L. Engineering Design Problems M. Lack of a Reclamation Plan N. Under-estimation of Acid and Leachate Generation II. Failure to Meet Clean Water Principle #2: There Are Not Safeguards in Place for When Things Go Wrong III. Failure to Meet Clean Water Principle #3: The Mine Plan Does Not Leave the Site Clean and Maintenance Free or Comply with Minnesota Laws A. Minnesota Rules 6132.3200 B. Minnesota Rules 6132.4800 C. Minnesota Rules 6132.1300 D. Minnesota Rules 6132.1200 E. Minnesota Rules 6132.1100 IV. Failure to Meet Clean Water Principle #4: Minnesota Taxpayers Will Not Be Protected A. Financial Assurance Information Lacking B. What Financial Assurance Should Include V. Failure to Analyze Alternatives A. Underground Mine Alternative Not Adequately Considered B. West Pit Backfill Alternative Not Adequately Considered

Page Number 1 1 3 4 8 10 16 21 24 28 30 32 33 37 38 39

40 41 41 42 42 42

43 43 45 47 48 49

VI. Wetland Impacts A. Loss of High Quality Wetlands 100 Mile Swamp B. Inadequate Wetlands Mitigation C. Inaccurate Wetlands Characterization and Groundwater Connection D. Under-Estimated Pit Dewatering Impacts E. Failure to Assess Indirect Wetland Impacts F. Fugitive Dust and Wetland Impacts G. Loss of Peatlands and Climate Change Impacts VII. Ecological, Wildlife and Vegetation Impacts A. 100 Mile Swamp Impacts B. Canada Lynx Impacts C. Moose Impacts D. Impacts To Wildlife Corridors E. Wildlife Using The Project Area F. Impacts To Rare Plants G. Impacts To Wild Rice H. Use of Non-native Plants VIII. Climate Change and Air Issues A. Climate Change Impacts B. Air Quality and Visibility Issues IX. Impacts to Human Communities, Cultures and Health A. Impacts to Subsistence Users and the Safety of Fish Consumption B. Potential Colby Lake Impacts C. Exposure to Asbestos-like Fibers D. Human Health Impact Assessment Needed E. Inaccurate And Incomplete Information About Jobs F. Unreliable Information About Taxes X. Land Exchange A. Failure to Meet USFS Exchange Objectives B. Failure to Meet USFS Forest Plan Goals C. Fragmented Lands Do Not Have Equivalent Ecological Value D. The Exchange and Climate Change

52 52 54 55 57 57 58 58 58 59 59 62 62 63 63 64 65 66 66 69 72 72 73 74 75 76 78 79 79 80 82 82

XI. Agency Roles and Responsibilities

83

XII. Significant Omissions and Unclear Information

85

XIII. References

87

I. Failure to Meet Clean Water Principle #1: Minnesotas Waters Will Not Stay Safe and Clean The SDEIS presents a design that contains omissions, faulty assumptions, and risky and unexplained mitigation measures. In addition, the proposed mine would be located in a waterrich environment that has inherent risks for wide-spread water contamination. As noted by Dr. Miller (see attached report): Unlike proposals in the arid regions of the western United States, where evaporative processes will largely keep water quality impacts localized to the area around the mine, this mine is being proposed in a well-watered region of the U.S. and will be a potential source of sulfate and metals release to surface waters that will potentially affect the environment distant from the mine (Miller 2014). Dr. Miller also notes that the duration of expected water pollution provides Minnesota with an important decision: Whether a mine of this type should even be permitted remains an open question, since the impacts and management requirements will exist for thousands of years (Miller 2014). A. 200 to 500 Years of Water Pollution The SDEIS predicts centuries of water pollution that will require active treatment as a result of developing the NorthMet mine proposal. Mechanical water treatment is part of the modeled NorthMet Project Proposed Action for the duration of the simulations (200 years at the Mine Site, and 500 years at the Plant Site)It is uncertain how long the NorthMet Project Proposed Action would require water treatment, but it is expected to be long term. (SDEIS ES-11). Elsewhere, the SDEIS notes, Both mechanical and non-mechanical treatment would require periodic maintenance and monitoring activities. Based on current GoldSim P90 model predictions, treatment activities could be required for a minimum of 200 years at the Mine Site (SDEIS 3-72). Data in SDEIS supporting documents and analyses show contaminant levels well above water quality standards at both the mine site and plant site centuries after mine operations cease. In the Water Modeling Data Package Volume 1 Mine Site (12 March 2013) from PolyMet Mining, Figure H-16-24.2 shows sulfate levels in the Waste Water Treatment Facility Influent at year 200 between four and ten times higher than the 10,000 ug/L standard. Copper concentrations are also high, between 10 and 60 times higher than the legal standard (PolyMet Mining, Water Modeling Data Package Mine Site, March 2013). Data in these documents show the potential for 10 different contaminants exceeding safe and legal water quality levels at year 200 in the mines influent into the treatment plant, including arsenic, cobalt, lead, manganese and nickel. The graphs indicate a need to continue treating the water at least 200 years, and likely longer, given that they show no downward trend in the pollution levels. These same data packages also graph contamination levels for Category 1 Stockpile Drainage showing concentrations of pollutants above water quality standards at year 200.

Similarly, data packages for collected drainage at the plant site show water contamination needing long-term treatment. PolyMet modeled simulations for 500 years, but only graphed the first 80 years for the contamination levels in the influent into the Waste Water Treatment Plant. Fifteen different contaminants, including arsenic, lead, sulfate, cobalt and copper, show levels above safe and legal water quality standards at year 80 for water influent into the Waste Water Treatment Plant. The levels show no indications of downward trends (PolyMet Mining, Water Modeling Data Package, Plant Site, March 2013). Dr. Millers analysis of the project highlights the length of required treatment as a significant concern. The long term treatment of water from the West Pit and the Category 1 Stockpile is discussed in terms of 200-500 years, and in fact, is likely to be required in perpetuity. The Minnesota regulatory agencies need to realize that this length of treatment is almost unheard of for mining projects (Miller 2014). He also notes that in contrast with more arid regions where this type of mining has already occurred, Minnesota is different than these arid lands states in that the rain and snow fall will preclude any walk-away solution. The Category 1 Stockpile will almost certainly generate and release sulfates or sulfuric acid and contaminating metals far into the future, well beyond the time frames where institutional and regulatory control has existed in the past (Miller 2014). The fact that water treatment will be needed for centuries is reason for concern. Not only are there financial and logistical complications associated with active treatment for this period of time (which we will explore in greater detail later in these comments) but the generation of pollution for hundreds of years creates a lasting contamination risk to Minnesotas waters. Over centuries, it is likely polluted waters will enter the surrounding environment, from waters that escape capture and from such events as spills, leaks, accidents and flooding. PolyMets NorthMet proposal asks Minnesota to commit future generations to a significant pollution risk. Alternative mine designs must be explored to reduce the pollution risks. The Underground Mine Alternative and West Pit Backfilling Alternatives have not been fully assessed, but offer the potential to reduce water pollution problems. See our comments on Failure to Analyze Alternatives, Section V. for more reflection on these options. Recommendation: A mine plan that anticipates creating centuries of water pollution does not sufficiently protect Minnesotas waters. The SDEIS must more fully examine mine design alternatives, such as the Underground Mine Alternative and the West Pit Backfilling Alternative, that could reduce pollution risks. Approving a mine that leaves future generations with significant pollution problems for hundreds of years is not responsible stewardship of our natural resources, and the Minnesota Department of Natural Resources (Minnesota DNR) should reject any mine proposal that has this as part of its design. The SDEIS should clearly state that active water treatment will be needed for at least 200 years at the Mine Site and for at least 500 years at the Plant Site, or give a revised measured length of treatment time needed, instead of vague references to it is expected to be long term. And the SDEIS should include graphs and other depictions of water quality data, similar to what is found in the water modeling data packages. These graphs are

ways to help the public understand the information, and they should not exclusively reside in technical data packages that most of the public will never read. B. Millions of Gallons of Polluted Water Will Escape Capture The SDEIS acknowledges that some polluted water will not be captured through the various collection systems, and will seep into the surrounding environment. There are multiple sources of this contamination. For example, At the Mine Site, about 10 gallons per minute of untreated water would be released during closure (all related to groundwater seepage)At the Tailings Basin, about 21 gallons per minute of untreated water would be released during closure (all related to Tailings Basin seepage that bypasses the groundwater containment system) (SDEIS ES-36). This equals 5,184,000 gallons per year at the Mine Site and 10,886,400 gallons per year at the Tailings Basin, a combined amount of over 16 million gallons a year of untreated water entering groundwater and the environment. Some of this seepage will come from the mine pits. Beginning in year 33, the West Pit water level would rise above the top of bedrock and begin to release pit lake water into the West Pit surficial groundwater flowpath. and Beginning in year 21, the water level in the East Pit saturated backfill would rise above the top of bedrock and begin release of pit water into the East Pit Cat 2/3 Surficial Flowpath (SDEIS 5-97). Because the East Pit will be backfilled with Category 2/3 and Category 4 waste rock, the most reactive waste rock anticipated at the NorthMet site, drainage from this pit will contain toxic levels of pollution. Graphs in the Water Modeling Data Package (PolyMet Mining, Water Modeling Data Package Mine Site, March 2013) from PolyMet Mining show East Pit and West Pit concentrations of contaminants that will exceed water quality standards for centuries. It is the content of these pits that is expected to seep into the surrounding groundwater flowpaths. The pits will also be sources of significant sulfate pollution. "It is expected that eventually the sulfate concentration in the pits would stabilize to more or less steady-state values, although the timeframe for this would likely be greater than 200 years as indicted by Figures 5.2.37 to 5.2.239, which show solute concentrations continuing to decrease at year 200, although still above water quality standards" (SDEIS 5-155). Seepage from the pits will carry with it these elevated levels of sulfate. In addition, leakage from Category 2/3 and 4 stockpiles and the Ore Surge Pile will also release contaminated water during the operation of the mine, and the Category 1 stockpile will be an ongoing source of pollution (PolyMet Mining, Water Modeling Data Package Mine Site, March 2013). But the SDEIS accounting for the amount of uncaptured pollution is likely low. For example, according to the SDEIS, A maximum of 7 percent of seepage is estimated to bypass the containment system surrounding the Category 1 Stockpile. The SDEIS predicts that it would all flow as groundwater into the West Pit where it would be collected and pumped to the WWTF for treatment. (SDEIS 5-101). But earlier the SDEIS had noted that beginning in year 33, West Pit water would enter the surficial groundwater flowpath. So, some of the Category 1 polluted

drainage is entering the West Pit, and leaving by way of groundwater seepage, rather than being sent to the Waste Water Treatment Facility. This amount, whether the full seven percent or some other amount, is not accounted for in the estimated amount of water that bypasses treatment. Table 5.2.2-26 lists the contaminant sources to groundwater flowpaths, and Category 1 seepage is incorrectly not among these sources (SDEIS 5-123). The amount of polluted water uncaptured and untreated at the mine and plant sites is also likely to be under-accounted for due to faulty assumptions in the SDEIS water model used to predict impacts. The water model makes assumptions about the baseflow rates, travel times for pollution, effectiveness at capturing polluted water, and lack of bedrock fractures that are not supported by evidence and that could result in understated pollution risks. See our comments in the Water Model Section for a detailed outline of issues with the water modeling. Recommendation: The SDEIS needs to contain a full and accurate accounting of the amount of uncaptured polluted water that is expected to seep from mine and plant site features. The water model needs to be corrected and re-run with accurate information about baseflow and bedrock fractures, and revised results for pollution seepage and travel times clearly presented. C. SDEIS Not Complete or Transparent About Water Quality Impacts The SDEIS does not clearly nor fully present information to the public about the water quality impacts expected with the proposed NorthMet project Despite more than 2,000 pages of information, the document lacks key information, does not present information clearly, presents sometimes inconsistent information, and fails to answer critical water quality questions. In an Environmental Impact Statement of this type, the public wants to know the answers to questions such as: How long water treatment is needed, what is in the polluted water before treatment, what is in the water after treatment, and how effective is the water treatment? However, the SDEIS does not contain this fundamental information. Dr. Millers analysis (see attached report) indicates that the water impacts have been significantly underestimated in the SDEIS: the sulfate and metals loadings to the East Pit (refilled) and the West Pit Lake are, I believe, substantially underestimated in the proposed PolyMet plansulfate and metals loading is very likely to exceed the groundwater standards of the state of Minnesota (Miller 2014). 1. Duration of Active Water Treatment While the water modeling data packages show data that indicate the need for active water treatment for at least 200 years at the Mine Site and at least 500 years at the Plant Site, the SDEIS states that water treatment will be needed long term. In public statements and testimony, Minnesota DNR representatives indicate that they did not seek to answer the question of how long water treatment would be necessary. Its kind of a paradigm shiftWe came in with an assumption that it would be a problem for a long time, so we were interested in answering a different question with the model (MPR News January 27, 2014).

Explanations in the SDEIS and agency statements have often made it confusing to know what the Minnesota DNR intended to model. The SDEIS notes that the 200 and 500 years were the length of time that the model was run based upon capturing the highest predicted concentrations (SDEIS 3-5). Other agency statements suggest the model results depict travel times for uncaptured polluted water, or examines if capturing and treating 90 percent of water at the mine and plant sites ensure that the other 10 percent of polluted water that escapes still meets state water quality standards (MPR News January 27, 2014). According to the Minnesota DNR, the agency did not ask or try to answer a fundamental question that the public wants and needs to know: How long will water treatment be necessary? This is a key question that an environmental impact statement should examine and seek to answer. Understanding environmental impacts and crafting financial assurance decisions depend on analyzing this question. 2. What is in the Water and Wastes, and How Effective is Treatment? The SDEIS does not present information about what is in the polluted water that is sent for treatment at the Mine and Plant sites. It does not reveal the contaminants in the water nor their concentration levels. Neither is post-treatment water quality information presented to enable the public to compare and evaluate the effectiveness of water treatment. Also missing is information about the make-up of Tailings Basin water, Hydrometallurgical Facility water and contents, the content of the waste sludge and solids, and the make-up of Reject Concentrate that is transported from the Plant Site to the Mine Site by rail car. These are all sources of contaminated water and yet the SDEIS is not transparent about what kinds of pollution and what levels of pollution are contained within them. Dr. Miller (see attached report) identifies the lack of information about the make up of the sludge and solid wastes as an SDEIS deficiency. Essentially no mention is madeof where the thousands of tons of precipitated sludge will be deposited, other than it will be moved offsite. This is a serious deficiency of the SDEIS, since the precipitated sludgecontains contaminants that are water soluble and will readily leach in the Minnesota climate (Miller 2014). Miller identifies a series of questions about the sludge that should be answered in the SDEIS. 3. Inconsistent Waste Rock Characterization The SDEIS does not consistently characterize the pollution risks from Category 1 waste rock. Table 3.2-8 on page 3-45, Waste Rock Categorization Properties, identifies Category 1 waste rock as having the low potential to generate acid. But in Table 3.2-2, Key Phases and Activities (Mine Site) on page 3-17, Category 1 waste rock is identified as non-acid-generating waste rock. Page 5-52 states that Category 1 waste rock would not produce acidic leachate. 4. Unsupported Statements on Water Quality and Treatment Time The SDEIS notes that the concentrations of certain contaminants in the pits are expected to exceed water quality standards but, without any supporting data, the SDEIS suggests that the need for water treatment may cease. Regarding sulfate levels in the pits, the SDEIS says, "It is expected that eventually the sulfate concentration in the pits would

stabilize to more or less steady-state values, although the timeframe for this would likely be greater than 200 years as indicted by Figures 5.2.37 to 5.2.2-39, which show solute concentrations continuing to decrease at year 200, although still above water quality standards" (SDEIS 5-155). But without any supporting information, the SDEIS asserts about the Waste Water Treatment Facility (WWTF), These predicted improvements in water quality suggest that the WWTF may not need to operate permanently, but that at some point, non-mechanical treatment systems may be sufficient to meet water quality standards" (SDEIS 5-155). There are no data presented that show that the solute concentrations would continue to decrease or even reach a steady state value below water quality standards. There is no information provided that supports the idea that non-mechanical treatment would work and be sufficient to control remaining solute loads. 5. Missing Information Section 3.2.2.1.3, New Construction and Pre-production Development, states that among the construction activities is constructing water management features includingthe Treated Water Pipeline (SDEIS 3-37) and refers to Section 3.2.2.1.8 for more information. But Section 3.2.2.1.8 does not describe anything about the Treated Water Pipeline. The figures in the SDEIS do not appear to depict the treated water pipelines running into or out of the plant site. 6. Water Data Not Transparent The SDEIS does not present water quality information in a transparent fashion. Many graphs, charts, and tables that could help explain information are either missing completely, or found in technical data packages and supporting documents to the SDEIS, but not in the SDEIS itself. Most members of the public will not know to find this information outside of the SDEIS. For example, there is no graph showing what is in the water before treatment and what the concentration levels look like over time. There is no chart that compares pollution levels in pre and post-treatment water. There is no table that shows what the contents of the Hydrometallurgical Residue Facility are, a basin that is double lined for safety. This is key information that the public needs to evaluate this proposal. 7. Problems with Dewatering Water Dr. Maest (see attached report) notes that the SDEIS does not indicate that dewatering wells will be used around the pits, not just within the pits. She states that dewatering wells around the pits will be essential to address water inflow issues. But, dewatering water from wells near pits could have elevated concentrations of nitrate and ammonia from blasting agents, sulfate from oxidation of sulfides or unexpected infiltration of stockpile seepage, and potentially metals and other contaminants (Maest 2014). Nitrate and ammonia will also be constituents in the water collected within the pits themselves. But Dr. Maest notes that The treatment pilot tests assumed dewatering water would have near-background levels of nitrate and ammonia, so the effectiveness of the methods of removal of this constituents under mining conditions is not known (Maest 2014). If excess mine water is discharged during mine operations and it is not adequately treated,

it could increase concentrations of these constituents above levels that would be harmful to human health, aquatic biota, and wild rice (Maest 2014). In addition, the volume of water collected from dewatering is likely to be more than what the SDEIS has anticipated for recycling through use in mine processes. If dewatering operations produce more water than can be used or stored and this is a reasonable possibility during a large storm or snowmelt event the excess dewatering water will require treatment before being discharged to streams (Maest 2014). PolyMet, however, has assumed that all process water would be used in the mine processes for the first 11 years. The SDEIS has not planned for a contingency that involves the production of more water than they can use. Dr. Miller also concludes that the volume of water could exceed that anticipated in the SDEIS. Pits in Nevada, a much drier place, require extensive wells around the pits in order to keep water out of the pit. As far as I can tell, there are no proposed dewatering wellsIf the pit dewatering calculations are incorrect, the volume of water needed to be removed could substantially exceed the capacity of the needs for mining and milling, and would need to be dischargedThe water treatment, as proposed, will not be sufficient to manage this water, as well as the volumes indicated in the SDEIS (Miller 2014). 8. Passive Water Treatment Unsupported By Evidence The SDEIS describes the potential to transition to non-mechanical water treatment at some undefined time in the future. However, the SDEIS provides no documentation, case studies, or data to support the feasibility of non-mechanical treatment successfully working for the volumes of water anticipated. The SDEIS has a responsibility to present evidence that methods of water treatment can successfully operate to protect human health and the environment. Unsupported concepts should not appear as viable options in the SDEIS. Dr. Millers report (see attached) identifies this lack of supporting information as problematic. This section implies that a biological treatment alternative actually exists and could treat water being discharged from the various contaminated sources from the closed PolyMet mineNone of these systems that I have observed can reduce sulfate to under 30 mg/L. At the least, there are no completely passive biological systems that have been shown to operate continuously. This sectionborders on being disingenuous in that there is an implication that water can be treated in a walk away design (Miller 2014). Recommendation: The SDEIS must analyze and provide information fundamental to this proposed project. The SDEIS should analyze and state how long active water treatment is needed and what the makeup and concentration levels are of all water on the Mine and Plant sites, including water collected for treatment, water after treatment, water that is expected to escape capture, contents of the mine pits, the Tailings Basin and Hydrometallurgical Residue Facility, waste solids and sludges, and the Reject Concentrate.

The acid generating potential of all waste rock should be accurately and consistently characterized throughout the SDEIS and should include its potential to leach metals and sulfates, given that waste rock pollution is not limited to acidic conditions. The SDEIS should remove statements that are not supported by data or other information. If solute levels do not decline to below water quality standards, the SDEIS should not suggest that the need for water treatment may cease. Any discussions of non-mechanical water treatment effectiveness need to be based on information and evidence from case studies, and not on hopeful but unsupported projections. All treated and non-treated water conveyances should be fully described and portrayed in figures in the SDEIS. Pipeline failures and breaks are among the common mishaps at mines of this type, with associated risks of spreading contaminated water. The SDEIS needs to fully describe and depict these features for the public to better evaluate the proposed project. The SDEIS should include graphs, tables, charts and other depictions of water quality data to help the public understand the information. We suggest: contents and concentrations of contaminated water over time, side-by-side comparisons of pre and post treated water, contents of mine pits, Tailings Basin, Hydrometallurgical Residue Facility, and the Reject Concentrate. The SDEIS needs to describe if chemical make-up of pit and nearby well dewatering water. It should describe how treatment methods will address any constituents of concern. And the SDEIS should re-evaluate the volume of dewatering water anticipated, and how it might manage a situation where the volume exceeds the need for mine processes. The SDEIS should provide documentation and supporting data for the use of non-mechanical water treatment or eliminate the idea from consideration. D. A Critical and Risky Mitigation Plan Not Described in SDEIS The SDEIS fails to describe and analyze a significant mitigation measure that is briefly discussed in the Water Modeling documents but referred to only obliquely and with no details in the SDEIS (pg. 5-102). The proposal includes pumping water from the East Pit backfill for treatment in the WWTF. PolyMet proposes pumping water at 1,750 gpm during years 22-31. Not only is 1,750 gpm a large volume of water to be treated, it may imperil the ability of the East Pit to remain saturated. If the East Pit is not saturated, it no longer offers the benefits of subaqueous disposal. Pumping this volume of water from the pit may risk exposing it to air, causing reactivity. This pumping rate is approximately four times the pumping rate used to dewater the pit, therefore pumping alone would obviously dry the pit and desaturate the backfill (Myers 2014). This proposal appears to be critical to the reclamation plan, but it is not explained, analyzed, or presented in the SDEIS. In over 2,000 pages in the SDEIS, the only reference to this plan is this vague sentence about the East Pit and the Waste Water Treatment Plant (WWTP): Additional concentration reduction would occur by cycling the East Pit backfill pore water through the WWTP (SDEIS 5-102). It is unclear if the plan is feasible, if the backfill can remain saturated, 8

if the plan would dramatically increase pollution, and if the WWTP can handle the volume of water this plan would generate. Neither the SDEIS nor its supporting describe anything about the return of water from this treatment process to the East Pit (Myers 2014). The Water Modeling Data Package for the mine site offers very little additional information. In 2,870 pages, a short discussion without details, occurs on page 154 of the document. This brief discussion is buried in a large data document, and still does not provide answers to fundamental questions about this significant component of the project. This is not adequate public disclosure. The Water Modeling Data Package for the mine site does, include, however, conflicting information about the length of time required for pumping and treating the East Pit water. A range of time from Year 25 to Year 60 is given, but then the document notes that the time required for pumping to the WWTF to be complete is uncertain (PolyMet Mining March 2013). Confusing information about the amount of water pumped and treated is also presented in PolyMets documents. The Water Modeling Data Package for the Mine Site indicates that 1,750 gpm would be pumped during Years 22-31, but after Year 31 this rate could go up due to decreasing flows from other sources. However, pumping rates higher than 1,750 gpm are not reflected in the data in the document, nor in the data presented in the Adaptive Water Management Plan (PolyMet Mining March 7, 2013). Dr. Miller (see attached report) is skeptical the East Pit rinsing method can work for such large volumes of water at such high sulfate levels. He also notes such a method has not been successfully demonstrated elsewhere. The overall goal of this process is to reduce sulfate concentrations to less than 250 mg/L, which can then be discharged back into the Eat Pit. However, I was unable to locate any data that indicated that sulfate cold be reduced from 20003000 mg/L to 250 mg/L using this approach. I seriously doubt that sulfate can be reduced by this amount, and in the absence of any scalable data for a continuous process, the MDNR should assume that it cannot (Miller 2014). Indeed, Dr. Miller concludes about this proposed mitigation plan: Rinsing surfaces in a refilled pit to meet a water quality standard has not even been attempted to my knowledge, and is quite frankly, an absurdly optimistic proposal (Miller 2014). Additionally confusing is that the SDEIS indicates this water treatment will take place at the WWTP, while the Water Modeling Data Package for the mine site indicates it will occur at the WWTF. It is unclear if the SDEIS meant to describe pumping East Pit water to the WWTF at the mine site or if it is really intended to be pumped all the way to the WWTP as noted. Recommendation: The failure to describe this mitigation plan in the SDEIS is a significant omission. The SDEIS needs to fully explain this plan, analyze the potential for this to desaturate the backfill, and account for potential impacts. The SDEIS should include an analysis of adding water containment systems to the East Pit, with the necessary adaptation to the wastewater treatment facility to handle this additional water treatment. The SDEIS needs to be clear if it is the WWTP or the WWTF that is intended to treat East Pit water, and if it is the WWTP a full description of the process for piping the water needs to be included. The SDEIS needs to reveal if pumping rates will go higher than 1,750 gpm and provide details of that rate and how the backfill will remain saturated under such pumping conditions. The SDEIS should include examples of mines that have used this rinse method and had success at bringing sulfate levels to 250 mg/L. 9

E. Major Flaws With the Water Model The SDEIS relies on a computer model to assess potential environmental impacts of the proposed NorthMet project. But, the model is built upon a variety of assumptions that are likely incorrect, and as a result, under-estimates the potential for water pollution. 1. Baseflow Assumptions Too Low Tribal cooperating agencies and hydrologist Dr. Tom Myers (see attached report and comments E.5 below) voice concerns that the baseflow numbers used in the model for basic site surface water flow hydrology at the Mine Site are lower than measured data suggest for the site. Indeed, the Minnesota DNRs own hydrologists identified this as a problem in a December 17, 2013 memo (Kruse 2013). The baseflow is a way of assessing the amount of groundwater that flows through the mine site and contributes to the flow in the Upper Partridge River. The amount of baseflow impacts how far and how quickly contamination from the mine and tailings basins might be distributed in nearby waters. The SDEIS used modeled estimates of baseflow rather than measured values from flow gaging stations nearby. The SDEIS itself acknowledges that the modeled values are lower than those measured. For all locations along the Partridge River, the XP-SWMMestimated baseflow is less than the MDNR-measured winter flow (SDEIS 4-67). Table 4.2.2-9 on page 4-67 compares average values at Minnesota DNR gaging stations with modeled estimates. In many cases, the SDEIS model underestimated flows during winter by as much as a factor of five. Other flow data recorded over a two year period at a measurement site on the Dunka Road also show higher baseflow numbers than used by PolyMet in the model. Scientists with the Great Lakes Indian Fish and Wildlife Commission (GLIFWC) in communications with agency staff, conclude that the measured flow numbers suggest that a baseflow estimate for site SW003 of 1-2 cfs would be reasonable (SDEIS, Appendix C, GLIFWC Comments, 2013-07-02). Table 4.2.2-9 reveals an average measured value of 5.0 cfs. But the PolyMet model used a significantly lower rate than either of these numbers with a rate of 0.51 cfs. Low baseflow values in the model likely result in an underestimation of water quality impacts. As GLIFWC scientists note, the faulty baseflow assumptions in the model can influence predictions about pit inflow, dewatering impacts to the Partridge River, water treatment needs, groundwater flow rates, contaminant transport times and concentrations, and contaminant dilution in the Partridge watershed (SDEIS, Appendix C, GLIFWC Comments, 2013-07-02). In addition, the mischaracterization of the baseflow could also mean transport of greater quantities of pollutants to the river and more drawdown of the Partridge River. (GLIFWC Comments, SDEIS Appendix C, 2013-07-02). The NorthMet SDEIS predicts travel times of contaminants to the Partridge River that are between decades and 110 years in duration (Tables 5.2.2-21 and 5.2.2-26). The

10

document also predicts that water quality standards will not be exceeded as a result of the mine project. Yet, without building accurate baseflow assumptions into the model, the predictions in the SDEIS are not reliable. Pollution travel times, water quality standards compliance, and drawdown effects on the Partridge River cannot be confidently evaluated using the flawed model in the SDEIS. The Minnesota DNRs memo on the issue notes, Additional years of data and more detailed dewatering records will significantly improve base flow estimates for this location (Kruse 2013). 2. Lack Of Bedrock Fractures Is A Faulty Assumption In modeling water quality impacts, the SDEIS makes the assumption that the Duluth Complex and the bedrock at the Mine and Plant Sites are not fractured. From this assumption, the SDEIS predicts a nearly perfect and complete ability to re-route and capture water from mine and plant features, and little to no movement of contaminated water into bedrock with further movement. This assumption has been challenged by the U.S. EPA, tribal cooperating scientists, and other technical experts. In an August 7, 2013 letter to the Co-lead Agencies, the U.S. Environmental Protection Agency (U.S. EPA) identified the assumption of no bedrock fractures as incorrect. Fractures are known to commonly occur in the bedrock of the Duluth Complex (U.S. EPA letter, August 7, 2013). The EPA recommended that the SDEIS acknowledge the potential occurrence of fractures. But the SDEIS continues to assert that fractures do not occur. The geologic conditions are favorable for a cutoff wall due to the presence of low permeability bedrock. Performance modeling of the containment systems performed by PolyMet and reviewed by the Co-leads provides strong evidence that the capture efficiency would be greater than 90 percent (SDEIS 3-46). If the U.S. EPA and others are correct about the presence of fractures, the SDEIS has made a critically flawed assumption about the permeability of the site, with potentially disastrous pollution transport implications. For example, the SDEIS notes: The attenuation effect resulting from sorption is significant enough that arsenic, copper, and nickel are not predicted to travel from source areas to any evaluation locations or the Partridge River within the 200-year model simulation period (SDEIS 5-56). Other metals are also not predicted to travel to monitoring sites for centuries (SDEIS 5-56). The presence of fractures stands to significantly change these travel time predictions. The tribal cooperating agency also reference information that suggest potentially faster constituent travel times. The agencies note that PolyMets own well water collection data show this potential: The water samples from wells P-2 and P-4 exceeded the nitrogen (ammonia as N) criteriaThe presence of ammonia nitrogen in the samples likely indicates that there is a hydraulic connection between the bedrock aquifer and the surficial aquifer (PolyMet Mining 2006). In addition, The samples from pumping

11

well P-2 all contained measureable tritium, indicating that at least a portion of the source water is post-1952 (PolyMet Mining March 2007). The tribal agencies note that the Peter Mitchell Pit lies about one mile north of the proposed PolyMet mine pit and that taconite production began there in 1955. Though PolyMet did not determine what the source unionized ammonia or tritium found in the deep boreholes was, it seems likely that because of the Peter Mitchell Pits close proximity to the Proposed PolyMet mine site, the Peter Mitchell Pit is the source of contamination. The approximate fifty-year travel time of the pollutants found in the P-2 bore hole from the Peter Mitchell pit were not used to estimate travel time for pollutants leaving the PolyMet mine pit and reaching the Partridge River, or even to calibrate the model (SDEIS, Appendix C, Tribal Cooperating Agencies Cumulative Effects Analysis). In addition, the SDEIS does not examine the potential for blasting activities at the mine site to create new fractures and pathways for contaminated water to move off site. whatever degree of fractures now existing in the rock, blasting at the levels proposed by PolyMet will create damage to rock masses and rock fractures over an extensive area, including the entire mine site and extensive adjacent wetlands areas (SDEIS, Appendix C, Tribal Cooperating Agencies Cumulative Effects Analysis). 3. Incorrect Assumptions About Naturally Occurring Levels The mine plan assumes naturally occurring elevated levels of beryllium, manganese and thallium, and uses these elevated levels in the water model as what occurs naturally (SDEIS 5-10). However, the U.S. EPA argues that there is not evidence that this assumption is accurate. The computations provided in this table assume that the concentrations of beryllium, manganese, and thallium solutes are naturally occurring. However, the PSDEIS does not support this assumption with evidence that these concentrations are truly natural and not anthropogenic. (US EPA August 7, 2013). This faulty assumption in the mine plan means the SDEIS uses a higher evaluation criteria for these constituents than the U.S. EPAs maximum contaminant levels or the States Health Risk Limit (SDEIS 5-11 to 5-12). The U.S. EPA states that Evaluation criteria in the SDEIS should be set at Minnesotas water quality standards (WQS) unless evidencebased analysis shows that levels are due solely to natural background and not anthropogenic sources. (US EPA August 7, 2013). The SDEIS does not provide this evidence. 4. The SDEIS Does Not Use Water Quality-Based Effluent Limits The SDEIS uses a P90 Criterion to determine if water evaluation criteria are being met. The output is expressed as a cumulative frequency distribution of predicted concentrations (SDEIS 5-77). But the SDEIS notes that this is not equivalent to how water quality-based effluent limits (WQBELs) would be developed for National Pollution Discharge Elimination System (NPDES) permitting (SDEIS 5-77). Appropriate WQBELs would be derived based on water quality standards and implemented in the permit (SDEIS 5-77).

12

The use of the P90 approach makes it difficult for the public to understand if water quality standards will be met as the mine has been designed. It would make more sense and make for better understanding of the projects impacts, if the SDEIS used the same evaluation method as will be required in permitting. The US EPA commented on this problem too. The co-lead agencies are using a 90th percentile (P90) projection to evaluate whether or not evaluation criteria are being met. As we have discussed, this is not equivalent to how water quality based effluent limits (WQBELs) will be developed for NPDES permitting. Rather, appropriate WQBELs must be derived based on water quality standards (US EPA August 7, 2013). Tribal cooperating agencies and GLIFWC also find the evaluation criteria used to be problematic: Grand Portage and GLIFWC note that evaluation criteria are not equivalent to water quality standards (SDEIS 8-19). 5. Expert Evaluation of Model and Assumptions Technical expert, Dr. Tom Myers (see attached report), reviewed the water modeling for the NorthMet project and has provided extensive analysis and recommendations. He assessed the accuracy of the conceptual and numerical modeling presented by PolyMet, and developed an alternative flow model using different recharge rates than PolyMets. He then ran simulations of the mine development including assessments of mine dewatering and changes in seepage rates and contaminant transport (Myers 2014). His work outlines alarming problems with the model used by PolyMet, with the assumptions PolyMet put into the model, and with the results achieved. The flaws are so many, so significant, that it represents an indictment of core findings in the SDEIS. The Co-lead Agencies must make substantial changes in a revised water quality model or risk significant pollution problems from this project. Below the Friends highlights some of the major issues Dr. Myers identifies, but his full report, analysis, and alternative modeling should be read and incorporated into NorthMet project revisions: After about 60 years from the onset of mining, sulfates reach the Partridge River at concentrations that exceed the legal standard. This occurs at year 60 if the West Pits re-filling is assisted by pumping water into it. If the West Pit is allowed to refill on its own, sulfates reach the Partridge River at about year 75, also in concentrations exceeding the standard. The SDEIS predicts the project would cause no significant effects. Copper concentrations reaching the Partridge River would also exceed the water quality standard by large amounts after 60 to 80 years (Myers 2014). The SDEIS predicts the project would have no significant effects. PolyMet has underestimated recharge rates by three to five times what is to be expected. Higher recharge rates lead to higher conductivity (Myers 2014). Dewatering rates in the East and West Pits would be about twice that predicted by PolyMet due to higher recharge and conductivity. Higher dewatering rates have the potential to impact river flows and wetlands far more than have been predicted in the

13

SDEIS. Based on the simulation herein, the Partridge River reach above the confluence with the South Partridge River is most at risk from dewatering and that risk continues for years during reclamation and closure if the pit refills naturally (Myers 2014). During some periods, drainage from the Category 1 Stockpile would flow north to the Partridge River, and not solely toward the West Pit as the SDEIS predicts. This is significant, in that the SDEIS counts on contaminants flowing from the Category 1 Stockpile to the West Pit and through waste water treatment as a way of containing pollution. During reclamation, the East Pit water would be pumped, treated to bring sulfate levels down to 250 mg/L, and then redistributed to the East Pit, all while supposedly keeping the backfill waste rock in the East Pit saturated. Dr. Myers questions the ability to successfully do this. While pumping this water, they claim they will keep the backfill submerged to prevent further oxidation. Considering that the required pumpage vastly exceeds the rate required to dewater the pit, this seems infeasible (Myers 2014). The effort would likely cause large areas of the backfill to become desaturated. Pumping may even facilitate the movement of oxygen and moisture through the backfill, conditions involved in the development of acid mine drainage. PolyMet assumes it can work without presenting a design or model (Myers 2014). Hastening the West Pits refilling through pumping has the effect of changing gradients which cause contaminant loads to reach the river more quickly. Pumping also adds very significant loads to the groundwater system on the west end of the mine for SO4 (Myers 2014). Dr. Myers challenges PolyMets claim that GoldSim is a widely used model in the industry. This reviewer has reviewed many environmental documents for mining projects all over the country and somewhat globally and has never seen the model used (Myers 2014). PolyMet should provide a list of mines for which GoldSim has been used to simulate contaminant transport. The PolyMet GoldSim model contains inaccuracies with the parameters, the assumptions, the conceptual model, and even the equations used in the model to solve the transport. It indicates the overall results are inaccurate and probably give a false sense of security (Myers 2014). The assumption of a lack of bedrock fractures is incorrect, demonstrated even by PolyMets own data. Dr. Myers points out that the low conductivity values arrived at by PolyMet are the result of extremely low values for three wells pulling the mean in a direction that suggests a low conductivity, while 9 of the wells showed higher conductivity values. it would be better to consider that 70% of the site had bedrock K greater than 0.1 ft/d (Myers 2014). A 30-day pump test conducted to assess the connection between the bedrock and surficial aquifers was done on well P-2, the well with the second lowest conductivity value. Had this test been done on any one of the 9 wells showing higher conductivity, results may have been different. This is far too little data with which to assume little connection between the bedrock and surficial aquifers (Myers 2014). PolyMet relies on perfect engineering to limit the seepage rates through three waste stockpiles, an ore surge pile, and the tailings impoundment (Myers 2014).

14

If their many modeling assumptions do not fully manifest or their engineering designs do not work as well as expectedthis load will reach the river and cause the project to exceed many standards (Myers 2014).

In its analyses, PolyMet frequently uses mean values for model inputs, resulting in an underestimation of the potential environmental impacts. Dr. Maest notes, for example, when assessing the relationship of the sulfur content of the rock to the potential to release sulfate, PolyMet used a method that gives an average release rate for Category 1 and 2/3 wastes, and ignores higher sulfate release rates that would produce higher sulfate concentrationsThe lack of inclusion of higher values suggest that the approach used to estimate sulfate release rates would underpredict maximum sulfate concentrations for Category 1 and 2/3 wastes and should not be used (Maest 2014). In addition, PolyMets modeling places limits or concentration caps on the concentrations leached from mined wastes. This approach results in the high end of potential contaminant concentrations being excluded from consideration in the model (Maest 2014). And the model assumes that essentially all contaminants will be adsorbed or otherwise removed as mine-affected waters travel along prescribed flowpaths to receptors (Maest 2014). However, the adsorption of metals depends on numerous factors which are not understood for the mine site. Adsorption of metals and other mine-related constituents is strongly dependent upon a number of factors that can change markedly along groundwater flowpaths, including groundwater flow and water quality characteristics, the characteristics of the geologic materials, and the availability of adsorption sites. Given the alack of site-specific information on these factors that affect contaminant adsorption along potential groundwater flowpaths, a modeling run with no adsorption should be conducted and included in the modeling results presented in the FSEIS (Maest 2014). Dr. Maest also identifies problems with the Monte Carlo simulations that were run in the model: Monte Carlo simulations used in the water quality model do not simulate the timing or characteristics of natural processes or the potential environmental consequences of the interaction of these processes with mined materials (Maest 2014). This is a troubling issue, given how much of the SDEIS evaluated impacts relies on a model that simulates natural processes and interactions. Dr. Millers report (see attached) also highlights issues with the modeling that have likely underestimated water quality impacts. The SDEIS for the PolyMet mine, and accompanying documents have uncertainty estimates for sulfate release, and they utilize the 90-95% confidence estimates for sulfate release. This same type of Monte Carlo simulations have been completed for many mines in Nevada, but are still wrong by sometimes over an order of magnitude, primarily a result of utilizing an incorrect basic conceptual model. I believe that this is the case for the East and West pits of the PolyMet mine (Miller 2014). He also notes regarding the modeling for the West Pit and the cone of depression that will form, I believe that the physical model proposed for the PolyMet West Pit Lake to be fundamentally incorrect, and certainly not consistent with pit lakes formed in gold bearing deposits in Nevada (Miller 2014).

15

Recommendation: The SDEIS must present a substantially revised water model that has corrected the fatal flaws found in the current version. Without doing so, none of the impact assessments within the EIS can be understood with any level of confidence. The many issues identified in the technical experts reports should be addressed and their recommendations fully incorporated. The SDEIS should reconsider if the GoldSim model is the appropriate model for assessing this projects impacts. The water modeling assumptions and flaws need to be fixed and the model re-run for more accurate results in the SDEIS. Water modeling should be based on accurate field observations. The model should be re-calibrated with targets based on observed baseflow numbers and new water level data from newly installed mine site wells. The SDEIS should incorporate the issues identified in the Minnesota DNRs own memo identifying incorrect flow numbers, and collect the additional data needed to calculate a more accurate flow. The degree of bedrock fracturing should be more thoroughly analyzed and this information incorporated into re-running the water model for more accurate results on pollution transport from all mine and plant contaminant sources. Particular attention should be given to reevaluating claims that the discharge of sulfates and metals will not impact wetlands and will not exceed water quality standards. The SDEIS should either provide evidence to support using elevated levels of beryllium, manganese and thallium as naturally occurring, or adjust the water model to use evaluation criteria set at Minnesotas water quality standards. The SDEIS should use the same water quality based effluent limits as will be used in NPDES permitting to enable the public to understand and evaluate impacts. The SDEIS should be revised to incorporate the issues outlined in Dr. Maests report. F. Mercury Contamination The SDEIS is critically inadequate in its analysis of potential mercury impacts from the proposed mine project. 1. Potential Sources Of Mercury Not Evaluated The SDEIS fails to evaluate several potential sources of mercury contamination to the waters surrounding the NorthMet proposed mine and plant sites. Constructed Wetlands. After mine operations cease, the SDEIS calls for wetlands to be constructed at various locations including the East Pit, the Tailings Basin, the Temporary Category 2/3 Stockpile, the Overburden Storage and Laydown Area, some haul roads and adjacent ditches, and the WWTF ponds and process water ponds. But, as GLIFWC notes in its comments in the SDEIS (Appendix C), Wetlands are known to promote enhanced mercury methylation. The extent of methylation depends on the amount of mercury and sulfate in the sediment and water of the constructed

16

wetland. The East Pit and the tailings basin are regions of potentially elevated mercury and sulfate. Therefore, there is a reasonable potential for the constructed wetlands to be significant sources of methylmercury to the aquatic foodweb (SDEIS Appendix C, GLIFWC Mercury Comments). Waterfowl and other water birds may be especially vulnerable to mercury exposure from using these constructed wetlands. West Pit. The SDEIS uses data from 16 mine pit lakes and five natural headwater/seepage lakes in northeastern Minnesota to evaluate the potential for the West Pit to be a source of mercury contamination. The SDEIS notes that only two of these lakes/pits had average total mercury concentrations above the Great Lakes Initiative standard for mercury of 1.3 ug/L (SDEIS 5-202), and uses these data to conclude West Pit mercury concentrations would fall below the Great Lakes standard. But GLIFWC scientists disagree with this conclusion. The two lakes/pits that were examined that exceeded the Great Lakes standard, exceeded the standard by significant amounts, in some cases double the acceptable level. In addition, for the remaining 16 examples, the SDEIS provides only an average concentration, but does not reveal if any, and how many, of the examples had individual samples that exceeded the standard. The potential for West Pit overflow to exceed the Great Lakes standard is not fully examined in the SDEIS. Peat and Unsaturated Overburden In The Overburden Storage and Laydown Area. In the development of the proposed mine, peat and unsaturated overburden will be stripped from the land and placed in an unlined area referred to as the Overburden Storage and Laydown Area. any potential water seepage into the ground below the Overburden Storage and Laydown Area will flow directly into the Partridge River. The result is a potentially unaccounted for and unquantified mercury pulse into the Partridge River (SDEIS, Appendix C, GLIFWC Mercury Comments). The SDEIS predicts the proposed project would result in 4.6 pounds per year of mercury emissions to the air from the Plant Site (SDEIS 5-430). But the analysis does not include evaluating the mercury emissions from electricity generation for the project, or from the burning of fuel by mining vehicles and other equipment. Given that a significant portion of the electricity provided to the site would be generated through coal burning power plants, the most significant source of airborne mercury, this is a critical omission. Indirect mercury emissions, like indirect CO2 emissions, should be a part of the SDEIS analysis.

2. Mercury Not Included in GoldSim Model and Erroneous Assumptions Mercury was excluded from evaluation in the GoldSim water quality modeling due to insufficient data and a general lack of definitive understanding of mercury dynamics (SDEIS 5-201). The SDEIS notes, Current scientific understanding of the factors and mechanisms affecting mercury methylation and bioaccumulation is limited (SDEIS 5201). The failure to model a contaminant of major concern for this proposed project is a significant shortcoming of the SDEIS. Even more troubling is that even without this evaluation and without evidence, the SDEIS claims mercury contamination will not be an issue for the proposed mine.

17

the mass balance modeling and analog data from other natural lakes and mine pit lakes in northeastern Minnesota suggest that the mercury concentration in the West Pit Lake, the only surface water discharge at the Mine Site, would stabilize below the GLI standard at approximately 0.9 ng/L (SDEIS ES-36). But the model did not include a numeric analysis of the mines effects on mercury in fish tissue, and the SDEIS notes that the ability to understand mercurys bioaccumulation is limited. The SDEIS makes the assumption that the project will reduce the amount of sulfates entering the environment, and thus mercury methylation will not increase. The NorthMet Project Proposed Action is predicted to decrease mercury loadings in the Partridge River. The net effect of these changes would be an overall reduction in mercury loadings to the downstream St. Louis River (SDEIS ES-36). But there are no data to support this claim. In addition, the claim depends upon a mine site that has a nearly perfect polluted water capture rate and the uninterrupted treatment of contaminated water for centuries, perhaps in perpetuity. If any of these conditions are not met, the amount of mercury contamination could be significantly higher.

With limited understanding of the mechanisms affecting mercury methylation and bioaccumulation, the SDEIS cannot and should not make claims about the project reducing mercury loadings in the St. Louis River. Even the Minnesota Pollution Control Agency (MPCA) has delayed completing a mercury Total Maximum Daily Load (TMDL) process for the St. Louis River until more data have been collected. The St. Louis River is acknowledged to have a complex and unique mercury dynamic that is the subject of ongoing research. This should be recognized in the SDEIS and assertions about the NorthMet projects mercury impacts should be supported with relevant data. The agencies overseeing the NorthMet proposed mine need to insist on the collection of more data before drawing conclusions about mercurys impact on the watershed. 3. No Evidence That Tailings Basin Would Be Mercury Sink The SDEIS assumes that mercury will be sequestered in the LTVSMC tailings, with the Tailings Basin acting like a sink for mercury (SDEIS 5-205). GLIFWC scientists highlight a number of flaws with this assumption: The claim is based on a small scale study, the details of which are not provided in the SDEIS or supporting documents. The small study did not simulate real conditions at the proposed mine site. the experiment used process water that was 3.3 ng/L to test the adsorption capacity of the tailings. Buta pilot study found the process water fro the project would contain an estimated 11.2 ng/L of mercury (3.4 times higher than the experimental concentration)Process water with a much higher mercury concentration might not experience mercury reductions to the same degree as was seen in the small-scale bench top study (SDEIS, Appendix C, GLIFWC comments). The SDEIS assumes taconite and sulfide tailings adsorption capacities will be similar. But as GLIFWC notes, There are inherent differences in composition between taconite tailings and the tailings that would come from the NorthMet PGM type project. These differences are likely to affect metallic binding potential (SDEIS, Appendix C, GLIFWC Comments).

18

The SDEIS assumes that adsorption can continue indefinitely, but does not consider the possibility that the tailings may reach a saturation point beyond which further adsorption is not possible. At this point, if the Tailings Basin had been functioning as a sink, it could shift to become a mercury source of contamination. The time scale on which the experiments were conducted are not adequate for predicting the longterm behavior of mercury and its interactions with tailings materials (SDEIS, Appendix C, GLIFWC Comments).

4. Flawed Assumptions About Impacts to Partridge and Embarrass Rivers The SDEIS notes, The NorthMet Project Proposed Action is predicted to increase mercury loadings in the Embarrass River Watershed but decrease mercury loadings in the Partridge River. The net effect of these changes would be an overall reduction in mercury loadings to the downstream St. Louis River (SDEIS ES-36). The statement suggests that the impacts from increasing mercury loadings in the Embarrass River can be overlooked because loadings in the Partridge River are expected to decrease. This is not an appropriate response to the pollution issues facing the Embarrass River should this project proceed. In addition, the assumption that the Partridge River and St. Louis River will experience decreased mercury loadings are dependent upon a nearly perfect polluted water capture system, and the uninterrupted treatment of contaminated water for centuries. If any of these conditions are not met, the amount of mercury contamination could be significantly higher. The potential for this should be examined. Conclusions about impacts to these rivers do not take into consideration the potential for additional mercury loading from hydrological changes and water level fluctuations that can release mercury. The SDEIS assumes that the proposed mine would have few effects on the flows in the Partridge and Embarrass Rivers. Our wetlands comments reflect on concerns that the SDEIS under-estimates flow impacts to these rivers. Fluctuations in flow levels carry the risk of stimulating sulfate-reducing bacteria that can result in mercury methylation and enhance its bioaccumulation. Dr. Maests report (see attached) also concludes that it is likely that mercury loads could increase to the Embarrass River and the Partridge watershed (Maest 2014). Her analysis suggests that the more conventional treatment approaches proposed during operations will not remove mercury to meet the standard (Maest 2014). 5. Lack of Analysis of Wildlife Impacts The SDEIS fails to assess or properly evaluate potential mercury issues to wildlife. The Wildlife Section and the Aquatic Species Section of the SDEIS do not review mercury contamination impacts to species that could consume mercury contaminated food sources. In other places, the SDEIS mentions possible risks to wildlife, but does not fully evaluate the impacts nor propose a response: Wildlife, specifically aquatic birds, may utilize open water habitat created by the NorthMet Project Proposed ActionSome wildlife species, specifically those that feed on aquatic prey, may be susceptible to mercury exposure directly from open water sources such as the pit lake and Tailings

19

Basin pond, and indirectly at the Partridge River and Embarrass RiverSpecific species such as loons, osprey, mink, and otter may be affected (SDEIS 5-373-374). GLIFWC scientist have identified the use of mine and plant features by birds to be a significant potential pathway of mercury exposure to these individuals (SDEIS, Appendix C, GLIFWC Mercury Comments). GLIFWC also notes the failure of the SDEIS to evaluate constructed wetlands over the East Pit and elsewhere as potential sources of mercury exposure and bioaccumulation. Stating that risks may exist to wildlife does not fulfill the obligation of the SDEIS to fully evaluate the threats and propose mitigations. 6. Inconsistent Statements on Impacts to Partridge River Mercury impacts to the Partridge River are inconsistently characterized in the SDEIS, creating confusion about what the likely impacts may be. In some sections, the SDEIS notes that, The NorthMet Project Proposed Action is predicted to increase mercury loadings in the Embarrass River Watershed but decrease mercury loadings in the Partridge River (SDEIS ES-36). But after noting that the Partridge River is expected to experience decreased mercury loadings, the SDEIS states, Some wildlife species, specifically those that feed on aquatic prey, may be susceptible to mercury exposure directly from open water sources such as the pit lake and Tailings Basin pond, and indirectly at the Partridge River and Embarrass River (5-373-374). The mercury impacts to the Partridge River and the wildlife that use it should be consistently and accurately reflected in the SDEIS. 7. Incorrect Statements About Ongoing Mercury Issues The SDEIS states that: Deposition of mercury from the NorthMet Project Proposed Action would cease at closure, but mercury bioaccumulation in fish tissue an exiting fish consumption limits could persist beyond the mines operational life (SDEIS 5-510). This statement is incorrect according to PolyMets own data. Mine and Plant Site Water Data Packages show sulfate concentrations in mine and plant water far above water quality standards hundreds of years after the mine closes. High sulfate concentrations can stimulate the methylation of mercury in the environment. Given that the SDEIS acknowledges significant amounts of polluted seepage will escape capture and treatment, the SDEIS cannot assume that mercury deposition ceases at closure. The SDEIS itself notes concentrations of sulfates in the pits, projected to seep contaminated water, will continue to remain high: "It is expected that eventually the sulfate concentration in the pits would stabilize to more or less steady-state values, although the timeframe for this would likely be greater than 200 years as indicted by Figures 5.2.37 to 5.2.2-39, which show solute concentrations continuing to decrease at year 200, although still above water quality standards" (SDEIS 5-155). Recommendation: The evaluation of mercury impacts in the SDEIS is critically inadequate. A more complete and accurate assessment of mercury impacts must be completed. Additional data collection should

20

occur to enable the evaluation of mercury contamination. If insufficient data or understanding still exist to allow for the inclusion of mercury in modeling, the SDEIS should refrain from drawing conclusions that cannot be supported. No claims should be made about mercury reductions, increases, stabilizing levels, travel times, impacts to wildlife without data and information to support those claims. The SDEIS should include analysis of potential mercury sources that have not been addressed, including all constructed wetlands, the West Pit, and the Overburden Storage and Laydown Area. Mercury air emissions must be evaluated from the generation of electricity for the proposed project and from mining vehicles and equipment. If the evaluation continues to use other lakes/pits as references for what may occur in the West Pit or other mine features, all relevant data from those reference lakes/pits should be provided, including individual mercury samples that may have exceeded mercury standards. The SDEIS must include data to support the claim that the Tailings Basin would function as a mercury sink. The evaluation should answer questions that include: Are there conditions under which the tailings would shift from a sink to a source of mercury? Is the mercury permanently and irreversibly adsorbed to the tailings? Are there data to support the claim that 95 percent of the mercury is expected to be adsorbed to the tailings and hydrometallurgical residue? Mercury loadings to the Embarrass River from the proposed NorthMet project should be fully disclosed, the impacts fully evaluated, and mitigation measures proposed. The SDEIS should examine mercury impacts if polluted water collection systems fail to meet capture expectations (perhaps assess various capture rate scenarios) and if water treatment is interrupted or functions at a lower than expected rate. The SDEIS should revisit the analysis of flow level fluctuations in the Partridge River and more thoroughly examine the potential for drying and re-wetting cycles to release mercury. The SDEIS needs to include evaluation of impacts to wildlife and aquatic life from mercury contamination. All mine features should be fully assessed for their potential to contaminate wildlife. Mitigation responses should be provided. Mercury impacts in the Partridge River should be consistently and accurately described in the SDEIS. Descriptions of mercury contamination risks beyond the life of the mine should be fully and accurately described, and supporting data provided. G. Waste Rock Production and Management One of the major sources of potential water contamination at the proposed NorthMet mine is the waste rock piles. The proposed mine would generate 308 million tons of waste rock over the life of the mine (SDEIS 3-39). Each year, it would excavate 15 million tons of reactive waste rock

21

(SDEIS 1-5). The SDEIS describes a plan to manage this waste that needs greater clarification and adjustments: 1. Category 2/3 and 4 Waste Rock and the Ore Surge Pile Liner Systems Unlike the Category 1 Waste Rock Pile, the area around the Category 2/3 and Category 4 Waste Rock Piles and around the Ore Surge Pile do not have a water collection system with a cutoff wall down to bedrock. Instead, the plan calls for using liners underneath the piles with a drainage layer over it. "The liner system would consist of an impermeable barrier layer (to limit the downward infiltration of water through the liner system) and an overlying drainage layer....Foundation underdrains would be used, if necessary, to provide gravity drainage should elevated groundwater be encountered, to prevent or minimize the potential for excess pore pressures as the stockpile is loaded" (SDEIS 3-51). Unclear in the SDEIS is what would trigger the decision to add underdrains and the feasibility of adding them if the need arises. What if circumstances occur that cause the groundwater to be more elevated than when the stockpile liner system was being built? The SDEIS is not clear how this would be managed to ensure ground and surface waters are not impacted. 2. Category 1 Waste Rock Drainage Containment The plan in the SDEIS calls for creating a cutoff wall around the Category 1 Waste Rock Pile by excavating a trench down to bedrock. The cutoff wall will be constructed of either compacted soil or by placing a manufactured geosynthetic clay barrier in the trench (SDEIS 3-46). The SDEIS does not describe how or why this determination would be made. What are the differences in capture efficacy? This information should be provided. The containment system is designed to work by drawing down the water table on the stockpile side of the wall thus creating an inward gradient and minimizing the potential for drainage passing through the cutoff wall (SDEIS 3-46). The SDEIS states that performance modeling was done that showed a greater than 90 percent efficacy in the capture rate with this system (SDEIS 3-46). The SDEIS does not indicate if this was ever field-tested. No supporting data is given for the efficacy of such a system at another mine or similar structure. Nor does the SDEIS describe a contingency plan for if the containment system does not capture 90 percent of polluted drainage. This information is necessary to properly evaluate the potential effectiveness of a major pollution-control engineered design for the mine. In addition, the 90 percent efficacy assumes bedrock beneath the mine site is not fractured, and thus water that drains down to the cutoff wall is captured and sent for treatment. The geologic conditions are favorable for a cutoff wall due to the presence of low permeability bedrock (SDEIS 3-46). But the U.S. EPA and other scientists note that the Duluth Complex is known to be fractured. The failure of the SDEIS to account for these fractures means the bedrock is more permeable than PolyMets model has accounted for, and that the Category 1 collection system will likely not experience a 90

22

percent efficacy for water capture. This would result in a much greater transport of pollution than the SDEIS has anticipated. 3. Waste Rock Separation The SDEIS has little information about the process to separate the different waste rock types into their appropriate stockpiles. Any mistakes made with this process have potential for pollution risks. For example, if Category 3 waste rock is improperly placed in the unlined Category 1 stockpile, pollution issues can arise that have no control systems in place to manage. The SDEIS notes, Proper placement of waste rock and overburden in the appropriate stockpile and for ultimate disposal would be important to achieve the NorthMet Project Proposed Actions predicted water quality. PolyMet has developed a Rock and Overburden Management Plan for monitoring and testing of waste rock during mine operations. The USEPA, MDNR and MPCA have agreed that they will review this Plan and include requirements for waste rock testing and monitoring to ensure it is properly categorized and managed during permitting (SDEIS 5-219). However, the SDEIS does not describe this waste rock management plan that is critical to reducing water pollution risks. The public cannot fully evaluate if the separation plan can be done with consistent accuracy without more information. The public cannot fully evaluate environmental impacts and potential mitigation strategies if the information is not provided. This is a significant omission in the SDEIS. In her report, Dr. Maest identifies waste separation issues as a potential problem for the project. it is highly unlikely that the sulfur content of waste rock that ends up in the Category 1 Stockpile will only contain material with a %S of <0.12 (Maest 2014). Category 1 waste rock is likely to include Category 2/3 wastes, which would lead to leaching higher concentrations of contaminants from the Category 1 Stockpile. 4. Waste Rock Fugitive Dust Fugitive dust pollution concerns regarding loading and transporting ore are described elsewhere in the Friends comments. But Chambers and Levit in their report (see attached) also express concerns that the waste rock piles may also represent a fugitive dust source of contamination to surrounding waters and lands. The plan anticipates that dormant portions of the pile will be covered when not activeBecause the waste rock is highly reactive it stands to reason that its dust, notably fugitive dust, may be highly reactive (Chambers and Levit 2014). The potential for this to be a source of pollution should be assessed in the SDEIS. 5. Problems with Waste Characterization Dr. Maest identifies several concerns with how waste rock was characterized for the project. She notes that too few samples were analyzed in characterizing the waste rock for the volume of waste that is predicted to be generated. only 89 samples of waste rock from actual NorthMet project core have been characterizedThe number amounts to one sample for each 3.5 million tons of waste rock. I believe that the number of

23

samples usedis lower than for any other mine Environmental Impact Statement Ive reviewed (Maest 2014). Proper characterization of waste rock is essential for identifying reactivity potential and for managing the rock to limit its pollution risks. In addition, Maest finds that No traditional acid-base accounting (ABA) analyses were conducted on the samples- only sulfur, carbonate, and paste pH (Maest 2014). She notes that the near total absence of carbonate minerals suggests that ABA testing would show that all samples are potentially acid generating (Maest 2014). The SDEIS characterization of Category 1 Waste Rock as not acid generating could be wrong. MDNR cannot rule out the possibility that all wastes produced from the NorthMet Project will produce acid or leach metals within the 20-year timeframe of proposed mining (Maest 2014). Recommendation: Triggers for potential changes to designs to collect and manage polluted drainage at the mine site should be outlined in the SDEIS. Why or how underdrains would be added to certain stockpiles should be fully described. The Category 1 Stockpile Containment System should be field tested at an appropriate scale to assess the efficacy of this vital pollution control system. Results should be fully described in the SDEIS. Examples of the use of this method at mines or other operations at a similar scale should be found and their success rates described. Contingency plans should be developed for scenarios of water capture that are less than 90 percent, and these plans outlined in the SDEIS. The SDEIS should include a re-worked water model that assumes the presence of bedrock fractures and fully analyzes what this means for the transport of polluted water off-site. The Rock and Overburden Management Plan should be described in the SDEIS. So too should the SDEIS describe the waste rock separation process. Waste rock fugitive dust should be analyzed to ensure it is not a hazard or threat to human health or the environment. The SDEIS should contain a plan for managing waste rock fugitive dust if it is determined that this is a pollution risk. The SDEIS should reanalyze the characterization of waste rock and provide supporting evidence for its findings. H. Tailings Basin and Hydrometallurgical Residue Facility Pollution The Tailings Basin and Hydrometallurgical Residue Facility are two additional sources of potential water pollution. Each year, over 11 million tons of contaminated tailings and hydrometallurgical residue waste are expected to be generated (SDEIS 1-5). The SDEIS describes a plan to manage this waste that needs greater clarification and adjustments:

24

1. Existing Pollution Problems At Tailings Basin The Tailings Basin that PolyMet proposes to use for disposal of its processing waste is an existing basin with tailings from the previous LTVSMC operations. And the basin has on-going polluted seepage problems. The SDEIS states that while groundwater has been impacted by the existing tailings basin, the data suggest that the concentrations of the pollutants are not increasing or decreasing (SDEIS 4-111). But the last time measurements were taken was 2004 (SDEIS 4-109 and 4-110 Table 4.2.2-23). It is unknown if concentrations have increased or decreased over the past nine years. A full and current understanding of existing pollution issues at the Tailings Basin is essential given that PolyMet plans to add additional waste to the basin with additional seepage of contaminants expected. "The NorthMet Project Proposed Action would increase Tailings Basin seepage rates by 67 percent and increase surface seepage by about 75 percent" beyond the existing seepage from the LTVSMC Tailings Basin (SDEIS 5-159). More current information is needed to understand what the existing pollution issues are before adding additional waste as PolyMet proposes. 2. Tailings Basin Containment System Similar to the Category 1 Stockpile, the NorthMet plan calls for developing a pollution containment system that would involve the construction of a cutoff wall down to bedrock to divert and collect drainage from the Tailings Basin. a water containment system would be installed around the northern and western Tailings Basin dams to intercept the seepage that emerges as surface water near the toeand greater than 90 percent of all the seepage that remains in the ground as groundwaterPerformance modeling of the containment systems performed by PolyMet and reviewed by the Co-leads provides strong evidence that the capture efficiency would be greater than 90 percent (SDEIS 3116-117). The SDEIS does not indicate if this was ever field-tested. No supporting data are given for the efficacy of such a system at another mine or similar structure. The SDEIS does not provide PolyMets data that shows the 90 percent capture efficiency rate. Nor does the SDEIS describe a contingency plan for if the containment system does not capture 90 percent of polluted drainage. This information is necessary to properly evaluate the potential effectiveness of a major engineered pollution-control design for the Tailings Basin. In addition, the 90 percent efficacy assumes bedrock beneath the Plant Sited is not fractured and thus water that drains down to the cutoff wall is captured and sent for treatment. The geologic conditions are favorable for such a containment system due to the presence of low permeability bedrock (SDEIS 3-117). But the U.S. EPA and other scientists note that the Duluth Complex is known to be fractured. The failure of the SDEIS to account for these fractures means the bedrock is more permeable than PolyMets model has accounted for, and that the Tailings Basin collection system will likely not experience a 90 percent efficacy for water capture. This would result in a much greater transport of pollution than the SDEIS has anticipated.

25

3. Unsupported Assertions The SDEIS describes a Tailings Basin containment system that is assumed to have 100 percent effectiveness at capturing surface seepage and 90 percent effectiveness at capturing groundwater seepage (SDEIS 5-159). These are modeled assumptions that seem to lack field-testing and that do not take into account bedrock fractures or containment system failures or inadequacies. This is an example of the SDEIS making assertions about low probability of water quality impacts without providing supporting data. The SDEIS should provide supporting data from similar engineering designs at other mines that have proven efficiency rates of 90 to 100 percent. The SDEIS also makes unsupported assertions about the engineered design of the Tailings Basin. Regarding the function of the pond that would remain by design in the Tailings Basin post-closure: "The presence of the pond in closure would provide benefits as it would create a saturated layer that would permanently reduce the oxygen flux and associated solute release in the underlying tailings" (SDEIS 5-161). But the SDEIS does not provide data or evidence to support this assertion. Would the depth of the pond influence the efficiency of this, and what is the expected depth of the Tailings Basin pond? Would wind and turbidity add oxygen to the system? The SDEIS needs to back up its claims with evidence that shows the design will work and answers common sense questions. The SDEIS describes a plan at closure to try to limit downward seepage from the Tailings Basin: "PolyMet would also place a bentonite layer at the bottom of the tailings pond to reduce downward percolation of pond water into the tailings" (SDEIS 5-162). Yet the SDEIS does not describe if this technique has been applied elsewhere and with what success. This is an important plan to limit water pollution, but the SDEIS does not show it can work or has worked elsewhere. 4. Functioning of Emergency Overflow Channel The design for the Tailings Basin calls for the construction of a channel should storm events or other issues risk water escaping the basin. "An emergency overflow channel would be constructed as a backup means of controlling pond elevation...The emergency overflow would be provided for the protection of the dams in the event that freeboard within the Tailings Basin is not sufficient to contain all stormwater" (SDEIS 3-117). Yet details about this channel are not provided in the SDEIS. The public does not know the capacity of the overflow channel or if it is intended to be lined. If it is an unlined channel, what measures will be in place to prevent contaminated seepage from escaping the channel? Has such a method been used at other mines and with what success rates? The SDEIS needs to provide more information so the public can evaluate if this backup plan will work as intended and if water contamination can be prevented. 5. Functioning of the Hydrometallurgical Residue Facility The SDEIS calls for Hydrometallurgical Residue Facility to be double-lined to protect water quality. Between the liners, a collection system would be put in place to capture contaminated drainage. "The amount of water pumped from the leak collection system would be monitored on a long-term basis" (SDEIS 5-157). The SDEIS should clearly

26

state that long-term likely means centuries of management, given the results of the modeling as revealed in the Plant Site Water Data Packages. The need to double line the Hydrometallurgical Residue Facility is because the contents of this basin are expected to contain high concentrations of contaminants. The SDEIS starts to describe a contingency plan for unexpected issues with the facility, but then fails to address the issues in a meaningful way: "If the amount of pumpage were to increase or if there were any other indications of increased leakage, appropriate repairs and mitigation measures would be undertaken. For these reason, it is assumed that the leakage from this facility into underlying groundwater or adjacent surface water would be negligible and this potential effect is not discussed further" (SDEIS 5-157). This is not an adequate contingency plan for unplanned events at this basin. The SDEIS does not describe what kind of measures would be taken or how quickly a response might occur. What if a quick and adequate response does not transpire? Without a thorough contingency plan for the project, it is not possible to know the true scope of impact. The SDEIS cannot consider the impacts to be negligible without a contingency plan in place and without supporting information. It is especially troubling that the authors of the SDEIS found that describing the potential effects did not warrant further discussion. This is an example of the SDEIS assuming all designs work flawlessly forever, resulting in no water quality impacts. Supporting evidence is needed in the SDEIS for such assertions to have credibility. Recommendation: New data measuring existing contamination at the LTV Tailings Basin should be collected and the results fully described in the SDEIS. The Tailings Basin Containment System should be field-tested at an appropriate scale to assess the efficacy of this vital pollution control system. Results should be fully described in the SDEIS. Examples of the use of this method at tailings basins at other operations should be found and their success rates described. Contingency plans should be developed for scenarios of water capture that are less than 90 percent, and these plans should be outlined in the SDEIS. The SDEIS should include a re-worked water model that assumes the presence of bedrock fractures and fully analyzes what this means for the transport of polluted water off-site. The SDEIS should provide supporting evidence that a pond at the Tailings Basin will reduce the oxygen flux and associated solute release in the underlying tailings. It should describe the depth of the pond and how wind action will not introduce oxygen to the system. Supporting evidence needs to be provided in the SDEIS about the functioning of the bentonite layer that will be added to the Tailings Basin at closure. Examples of this application and its success rate at other tailings basins should be provided.

27

The SDEIS needs to contain more information about the emergency channel for the Tailings Basin, how it would function, and how seepage of contaminated waters would be prevented from this feature. The SDEIS needs to clearly state how long water collection systems and monitoring will be required based on modeling results. The SDEIS needs to develop and fully describe contingency plans for all components of the Tailings Basin and Hydrometallurgical Residue Facility. Scenarios should be developed looking at impacts of capture rates less than 90 percent. Capture rate expectations need to be supported with data and examples from other mine operations. I. Transportation Corridor Pollution Concerns 1. Risk of Ore Spillage and Fugitive Dust The SDEIS describes a plan to transport ore from the Mine Site to the Plant Site along an eight-mile railroad transportation corridor. The ore would be sent via open rail cars over a route that crosses wetlands and three creeks. We are concerned that ore spillage and the release of ore fines pose significant water quality threats to water systems along this route. Fallen ore and fugitive dust have the potential to contaminate wetlands and soils along the rail line, leading to acid drainage and metal leaching in nearby water bodies. The SDEIS plans to reduce these risks by using rail cars designed to centralize the ore fines to the central portion of the rail car to minimize the potential for spillage during transport (SDEIS 5-434) and by using side-dump cars (SDEIS 3-83). The SDEIS expects the amount of spillage and fugitive dust that escape would be minimal (SDEIS 5434). But these designs are not adequate to ensure spillage does not occur. GLIFWC shares these concerns: GLIFWC staff disagree that the amount of ore that could escape from rail cars would be small (SDEIS, Appendix C, GLIFWC Comments on Transport of Ore). The rail cars would not be sealed and would contain gaps in the side hinges from which small ore particles could escape. As GLIFWC notes, Even a loss of only one thousandth of one percent (0.001%) of the extracted material to tracking, dusting or spillage would result in 6,220 tons of fine leachable material being released into the environment (SDEIS, Appendix C, GLIFWC Comments on Transport of Ore). GLIFWCs comments point out that existing spillage of taconite pellets along this corridor are visible, clearly showing this kind of contamination is likely. The SDEIS provides no supporting data to show the centerline loading method will adequately contain spillage. The 2009 DEIS for the NorthMet project also failed to design rail cars to adequately protect against spillage. At that time, Dr. Chambers highlighted this as a concern. He noted that the effectiveness of the centerline loading method was quite frankly, not good at containing spillage. Unless rail cars are designed to be completely enclosed, there will be spillage and most probably metals

28

contamination along the rail haulage-line. Given time, spillage from the rail cars could be spread from the rail line across a wide area by wind (Chambers 2010). Chambers and Levit re-emphasize concerns about fugitive dust and ore spillage in their 2014 report (see attached). They note that ore dust contains numerous contaminants that could accumulate over years of use and present a pollution risk. They note that the mine ore transfer and loading process will generate a significant amount of dust in the vicinity, and in the predominant downwind direction of the Rail Transfer Hopper. And, since the rail cars are open-top, there will be an accumulation of dust along the rail corridor between the mine and mill (Chambers and Levit 2014). The SDEIS needs to address the pollution risks from dust at loading and during transport. While the SDEIS calls for monitoring of spillage along the rail line, this too is an inadequate response given that contamination will have already occurred. 2. Transportation of Reject Concentrate and Sludge In addition to issues associated with transporting ore from the Mine Site to the Plant Site, there is also the potential for spills from transporting Reject Concentrate by rail tank cars and from transporting waste sludge and solids by truck along the corridors roadway. The SDEIS plan calls for rail transport of Reject Concentrate from the Plant Sites Waste Water Treatment Plant to the Mine Sites Waste Water Treatment Facility. In addition, waste sludge and solids that are by-products of the treatment process at the WWTF at the Mine Site would be trucked off-site to other disposal facilities. The SDEIS does not discuss the possibility of accidents, derailments, or spills that could result in the contents of these transports being released into the surrounding environment. Nor does the SDEIS describe safeguards to deal with this possibility or any plan for how to manage potential spills. The SDEIS only mentions the possible risk of ore spillage, but does not mention contamination risks from Reject Concentrate or waste sludge and solids: No effects on groundwater quality along the Transportation and Utility Corridor are anticipated during construction or closure as part of the NorthMet Project Proposed Action. There is the potential, however, for ore spillage from rail cars in transport from the Mine Site to the processing plant during operations (SDEIS 5-98). Plant Site Water Data Packages from PolyMet Mining show that the Reject Concentrate is expected to contain contaminants at levels far above safe and legal water quality standards. Arsenic, chloride, cobalt, copper, fluoride, iron, manganese, sulfate are among the pollutants showing dangerously high concentrations levels in the Reject Concentrate. Sulfate levels are modeled at times at more than 1,000 times the legal water quality standard (PolyMet Mining, Water Modeling Data Package, March 2013). The contents of the sludge and solid wastes are not disclosed in the SDEIS or its supporting documents. The SDEIS does not describe how the rail line will be safely shared between the cars transporting ore and the tank cars transporting Reject Concentrate.

29

Recommendation: The risk of widespread and localized contamination from ore spillage and fugitive dust during uncontained rail transport is unacceptable and avoidable. The SDEIS needs to include a transport plan that calls for enclosed rail cars. The Rail Transfer Hopper and rail car loading conveyor and platform should be in an enclosed structure. The SDEIS needs to describe how the transportation corridor will be shared between the cars transporting ore and the tank cars transporting Reject Concentrate. The SDEIS needs to disclose the contents of the Reject Concentrate and the sludge and solid wastes, and give the concentration levels of the contents. It should provide a contingency plan for managing spills and accidents from the transportation of Reject Concentrate and sludge and solid wastes. The SDEIS should include modeled impacts of the distribution of fugitive dust. J. Cumulative Impacts The water pollution issues associated with the proposed NorthMet project would contribute to cumulative impacts within the region. These impacts have not been adequately addressed in the SDEIS. The SDEIS identifies both surface and groundwater concentrations that will exceed existing and base-line conditions if the mine proceeds. For examples Table 5.2.2-22 and Table 5.2.2-30 indicate that cobalt, lead, copper, nickel, arsenic, cadmium are constituents that will increase over what would occur under a No Action decision (SDEIS 5-109 and 5-129). The U.S. EPA identified this as a gap in the SDEIS because no analysis has been done on what this means for cumulative impacts when combined with existing contamination from other sources. The U.S. EPA found that the proposed project does not consider how the increased contaminant concentrations caused by the NorthMet Project, combined with all other past, present, and reasonably foreseeable future actionsmay cumulatively affect aquatic resources (U.S. EPA August 7, 2013). The letter states that, The SDEIS should acknowledge and consider how the modeled impacts of the NorthMet Project, including projected increased contaminant concentrations above baseline or no action levels in combination with other past, present and reasonably foreseeable actions, may cumulatively impact aquatic resources (U.S. EPA August 7, 2013). Elevated levels of sulfate are a current concern for the Partridge River. The SDEIS acknowledges that the project would increase this loading. The NorthMet sulfate load to the Partridge River would total an average of about 5 kg/d, which represents a 0.1 percent increase over existing loads (SDEIS 6-30). While the SDEIS asserts it will not contribute loadings that could promote mercury methylation, our concerns noted elsewhere in these comments indicates the SDEIS lacks data to support this claim. Mercury is also a concern for the Embarrass River, and the projects impacts are likely to add to this problem. The NorthMet Project Proposed Action is predicted to result in a net increase in mercury loadings to the Embarrass River of up to 0.6 grams per yearwhich represents about a 3 percent increase (SDEIS 6-34). Several lakes downstream from the proposed project along

30

the Embarrass River are already listed for mercury in fish tissue impairment, including Sabin, Wynne, Embarrass, and Esquagama lakes (SDEIS 6-33). The SDEIS shows in Table 5.2.2-38 that over a 500 year time, concentrations of constituents from the Tailings Basin will increase existing concentrations as a result of the project (SDEIS 5169). However, the cumulative impact of this increase has not been examined or proposed for mitigation in the SDEIS. Rather than investigating the cumulative impact, the SDEIS offers only a statement conceding the possibility of impact: "These contaminant loads from the NorthMet Project Proposed Action, however, could contribute to cumulative effects in combination with contaminant-loading from other projects" (SDEIS 5-153). The cumulative effects analysis also limits its scope for impacts to water resources to the Partridge and Embarrass River watersheds. Pollution issues from the proposed NorthMet project are likely to impact farther downstream, including waters of the St. Louis River. Given that water quality is already impacted in the St. Louis River from existing projects, the NorthMet proposed project should be considered capable of distributing constituents into the St. Louis River watershed. The appropriate scale for cumulative impacts to water resources should include the St. Louis River watershed. The cumulative effects analysis in the SDEIS also does not include foreseeable mine projects such as the Twin Metals project or potential projects by Teck. For the purposes of this assessment, cumulative actions are those current and permitted mine projects located in the portion of the Mesabi Iron Range within St. Louis, Lake, and Cook counties (SDEIS 6-99) The heading in this section says Past, Present, and Reasonably Foreseeable Future Actions, but the definition of what is included for evaluation does not include reasonably foreseeable actions like Twin Metals. This is an omission that needs to be corrected. In addition, the SDEIS does not evaluate the relationship between the development of the NorthMet proposed project and the prospects for development of other mine projects. If developed, the NorthMet processing facility is likely to be made available to other mining operations, possibly making other mine projects economically viable. The potential for the NorthMet project to enable the development of other mining activities needs to be evaluated. The Cumulative Effects Analysis should also include a geographic area not just south and west of the proposed project area, but also north and east where other mining activities and development are occurring and proposed. Limiting the scope to south and west creates an artificial area of evaluation that potentially under-represents the cumulative effects. Recommendation: The SDEIS needs to include a more thorough evaluation of cumulative impacts of the project. Increases in constituent loadings to the areas water systems should be evaluated for their impacts when combined with existing and foreseeable future contributions from other sources. The Twin Metals project should be included in this evaluation. Water quality models need to be re-run with corrections made to faulty assumptions (see Friends comments about Major Flaws with the Water Model, Section I.E.) and the cumulative effects re-evaluated using the new results. Cumulative effects analysis for water resources should include the St. Louis River and

31

should include areas north and east of the NorthMet proposed project. The relationship between the development of the NorthMet project and the development of other mining projects should be analyzed and discussed in the SDEIS. K. Concerns About Reliability of Reverse Osmosis The NorthMet mine plan calls for employing a reverse osmosis (RO) technology to treat the water and bring constituent levels down to legal water quality standards. In particular, the RO process is intended to bring sulfate levels below the 10 mg/L wild rice standard (SDEIS 3-123). There is little information in the SDEIS about how this technology works. The SDEIS has no information about the successful application of RO elsewhere at the scale proposed by PolyMet. Dr. Miller (see attached report) notes in his analysis: While the treatment technology (e.g. reverse osmosis treatment) has been shown to be effective in much smaller treatment facilities, it has not, to my knowledge, been utilized on such a grand scale and certainly not for centuries, as is projected in the SDEIS (Miller 2014). Additionally, the document provides no contingency plans for if the RO system does not work as intended. A May 31, 2011 memo from Barr Engineering highlights some of the uncertainties with water treatment in general and RO in particular. The memo notes, At the current time there are no technically feasible water treatment technologies that can be used to ensure that all the required discharge limits for alkalinity, TDS, hardness, and specific conductance can be met. There are currently no viable treatment options to treat Pit 1 water to less than 10 mg sulfate per liter (Barr Engineering May 31, 2011). The memo indicates that RO could in theory be used to reduce sulfate to the standard, but that since no proven method has been developed for disposal of the brines from nanofiltration or RO systems, these treatment methods are currently not viable treatment options for Pit 1 discharge water (Barr Engineering May 31, 2011). Presumably, PolyMets implementation of the evaporation/crystallization process is intended to deal with managing the brine problem. This needs to be clarified in the SDEIS. But the Barr memo raises many questions and concerns that are also not addressed in the SDEIS: If the RO system can be applied successfully for sulfate, can it perform as intended for the other constituents mentioned in the Barr memo, such as alkalinity, TDS, hardness, and specific conductance? The SDEIS should describe how water treatment has been designed to address the issues that the 2011 memo mentions, specifically the lack of feasible water treatment technology. What changes in the RO design have been made to address the issue in the 2011 memo that states RO is not viable in treating Pit 1 discharge water?

If the addition of the evaporation/crystallization process is intended to solve the RO issues mentioned in the memo, the SDEIS fails to provide examples of the successful application of this at a similar scale. Indeed, the Barr Engineering memo notes that examples of successful applications do not exist. Such large scale systems have not been built anywhere in Minnesota,

32

or elsewhere to our knowledge. While others may have proposed to construct such systems, none have in fact been built (Barr Engineering May 31, 2011). If examples have emerged since 2011, the SDEIS should describe them. Given that the RO system is critical to the SDEIS assertion of limited water quality impacts, the SDEIS should address Barrs characterization of the unproven application of this technology (Barr Engineering May 31, 2011). In addition to the reliability issues identified in the Barr memo, the Eagle Mine in Michigan highlights other issues with RO application. The Eagle Mine is a copper-nickel mine in Upper Michigan. Like the proposed NorthMet mine, the Eagle Mine has employed state of the art RO technology to treat its mine water. Blasting at the mine began in 2011. Even before the blasting has reached the most reactive ore, independent water quality monitoring has found 47 exceedances of ground water standards. The results can be found at www.cempmonitoring.com. Constituents that have exceeded standards include arsenic, copper, lead, molybdenum, silver, vanadium and pH (Halley January 19, 2014). It appears that the mines RO system is not working as intended. The SDEIS needs to examine this mines RO issues and explore how the NorthMet RO system will not experience similar problems. Recommendation: The SDEIS needs to thoroughly review the RO technology and how it intends to address the projects issue of contaminated water. The SDEIS does not provide the information or level of confidence that the public needs about this critically important technology that is central to managing pollution. The SDEIS needs to give detailed examination that shows how this has been applied successfully elsewhere and how it will avoid the problems encountered elsewhere. Without such detailed examination, the entire ability of the project to manage polluted water is called into question. The SDEIS needs to describe if and how the treatment process and RO can manage water issues in addition to sulfate, such as hardness, TDS, and specific conductance. All the issues identified in the Barr memo must be explained. L. Engineering Design Problems The SDEIS relies heavily on engineered methods to prevent and mitigate environmental impacts that are unproven, unreliable, and assume levels of effectiveness that are not realistic. Dr. Miller in his report (see attached) notes that the mitigation measures outlined in the SDEIS rely on large water collection and treatment processes that are unparalleled in modern mines and that many, such as the reverse osmosis process have not been applied at the scale proposed in the SDEIS (Miller 2014). He identifies the plan to pump and treat high-sulfate waters from the East Pit as unlikely to work: Particularly for the East Pit, the plan to rinse and withdraw the expected highly contaminated water from the pore spaces in the waste rock is untested, and very likely to fail due to the extreme difficulty of efficiently rinsing rock that has s been added to the pit (Miller 2014). Technical expert Dr. Michael Malusis has identified numerous concerns with the technical viability of the tailings and waste rock containment systems as outlined in the SDEIS and its supporting documents. Dr. Malusis full report is attached with the Friends comments and

33

should be reviewed for his detailed remarks and recommendations. The engineering design problems he highlights are significant issues that could result in the failure of systems to function as intended. Failures of these systems can in turn translate into significant water quality issues for the surrounding environment. Some highlights of Dr. Malusis concerns: 1. Category 1 Stockpile and Liner Issues - The SDEIS does not indicate if the containment system wall will include a geosynthetic barrier, and if so, what type will be used. The absence of such commitments and details is problematic, as the options proposed are not commonly employed for long-term containment of contaminated groundwater and do not have a strong track record of success in these applications (Malusis 2014). - For the cutoff wall, the SDEIS implies a geosynthetic clay liner rather than a conventional geomembrane will be used. But Dr. Malusis indicates a lack of evidence in the literature to show that barriers other than geomembranes have been used in vertical barriers. Research indicates walls with geomembrane alone are not likely to be effective for controlling underseepage. - The SDEIS provides for a maximum allowable hydraulic conductivity for the cutoff wall that is 10 to 100 times higher than the typical conductivity for vertical barriers used in long-term hydraulic control. If constructed at 10-5 cm/s, the wall likely would be ineffective as a long-term barrier (Malusis 2014). Chambers and Levit also identify the permeability as problematic, and recommend a lower permeability of 10-6 cm/s. - The SDEIS model simulations do not consider the possibility that the efficiency of the groundwater collection system could diminish over time. - The groundwater modeling assumes that groundwater from outside the containment area does not make its way in. Unless the cutoff wall is keyed into the bedrock, significant underseepage could occur, either into the containment system or out of the containment area (Malusis 2014). Seepage out of the containment system means contaminated water is escaping capture and is entering the environment at levels not anticipated. - The slope that is proposed for the final cover system is too flat and would likely result in areas where water ponding will occur. - Dr. Miller (see attached report) also highlights concerns about the Category 1 stockpile liner system, in particular the issue of the liner degrading over time. The liner on the Category 1 stockpile is made of carbon polymers, and those liners will oxidize, crack and degrade over timewhen the liner ultimately breaks down, as it surely will, the drainage from that waste rock dump will increase both in volume and contaminant load. Those future generations will then need to decide how to handle the waste, and may or may not have the resources to manage the waste (Miller 2014). He notes that the contaminated drainage will continue for thousands of years.

34

Recommendation: The SDEIS needs to address all of the engineering design problems of the Category 1 Stockpile system identified in the experts reports. The design should include the use of a geosynthetic barrier in the wall, and the use of geomembranes with a low-permeability, soil-bentonite backfill. The maximum hydraulic conductivity of the cutoff wall should be adjusted in the SDEIS to a much lower threshold of 10-6 to 10-7 cm/s. Modeling needs to be re-run to include scenarios under which the efficiency of the groundwater collection system diminishes over time due to clogging, and a response plan needs to be identified to address this. The design must include keying the cutoff wall into bedrock to prevent underseepage, and the cover system slope should be adjusted to 2 to 5 percent. The SDEIS should describe the plan for the Category 1 Stockpile liners to degrade and how this will be addressed and financed in the future. The SDEIS should evaluate backfilling the West Pit with the Category 1 waste rock, to avoid pollution issues associated with degrading liners. Overall, the long term stability of the wasterock deposited in the pits and covered with water is a much lower long-term risk than leaving it on the surface with a plastic sheet over it (Miller 2014). 2. Tailings Basin Containment System Issues - The design plan will need to commit to maintaining a certain magnitude of inward gradient or head difference across the cutoff wall (similar to the Category 1 collection system) to ensure the system functions. - Unlike the Category 1 Stockpile containment system, this cutoff wall is identified as being keyed into bedrock, but the minimum depth of the key is not described. - The SDEIS inconsistently describes the thickness of the bentonite-amended layers to be installed on the dam raises. In some places it indicates a thickness of 12 inches, elsewhere 18 inches. - The application of 3 percent granular bentonite over predominantly coarse tailings is not likely to produce a uniform barrier with low hydraulic conductivity. The flow paths within the barrier would need to include bentonite to function properly. - The SDEIS indicates that the betntonite-amended layer would be at or above 90 percent saturation so that it would function as an oxygen barrier. However, no evidence is presentedto show that a saturation of > 90 % would be maintained in this layer over the long term (Malusis 2014). - The SDEIS proposes three possible methods by which it would create a bentonite seal at the bottom of the Tailings Basin pond. Two of these proposals are, at best, experimental (Malusis 2014). Case study evidence is necessary. And PolyMet needs to explain why draining the pond before installing the bentonite seal is not considered. Recommendation: The SDEIS needs to address all of the engineering design problems of the Tailings Basin collection system identified in Dr. Malusis report. The design plan needs to commit to maintaining the inward gradient essential for the functioning of the system. The minimum depth of the keyed cutoff wall should be described. The SDEIS needs to consistently describe the thickness of the bentonite-amended layers on the tailings waste. The SDEIS should include

35

bench-scale tests to determine the percentage of bentonite necessary to form an effective oxygen barrier. The SDEIS needs to present evidence to show that a greater than or equal to 90 percent saturation level would be maintained in the long-term for the waste cover system. And case study evidence of the bentonite seal application method for the Tailings Basin pond must be demonstrated as working for similar projects. The SDEIS should explore the option of draining the pond before applying the bentonite seal, or explain why this option is not considered. 3. Tailings Basin and Hydrometallurgical Residue Facility Geotechnical Stability Chambers and Levit note in their report that the tailings dam construction method chosen for the NorthMet Tailings Basin is the least expensive dam construction approach, but poses the most risk to long term seismic stability (Chambers and Levit 2014). The method, called upstream method, involves placing existing tailings on top of the existing LTVSMC tailings embankment and to perimeter embankments (SDEIS 5-561). Chambers and Levit highlight that most tailings dam failures that have occurred elsewhere have been associated with upstream dam construction. Chambers and Levit also point out several flaws in the calculations in the SDEIS for seismic risks, including underestimating the potential earthquake size and using a method of assessing dam stability in an earthquake that most federal regulatory agencies would not accept. It is not possible to predict failure by pseudostatic analysis, and other types of analysis are generally required to provide a more reliable basis for evaluating field performance (Chambers and Levit 2014). Similar problems occur for the Hydrometallurgical Residue Facility geotechnical stability design, evaluation and modeling. In addition, the SDEIS calls for establishing the Hydrometallurgical Residue Facility on a layer of existing taconite tailings. Chambers and Levit highlight this as potential stability risk and suggest removing the tailings before constructing this facility. Recommendation: The SDEIS should take into account the many geotechnical stability concerns identified in the Chambers and Levit report and implement the recommendations to ensure engineered designs that provide stability and protection. Underlying original ground and taconite waste should be removed from beneath the Hydrometallurgical Residue Facility and an engineered stable base installed. Dynamic modeling should be performed for both the Tailings Basin and the Hydrometallurgical Residue Facility. 4. Hydrometallurgical Residue Facility Engineering Issues - The SDEIS assumes a liner leakage defect frequency of 2.5 defects per acre. However, field evidence indicates that liner leakage predictions assuming such low frequencies of defects are lower bound estimates that may significantly underestimate the actual leakage unless strict construction quality assurance (CQA) procedures are followed during installation (Malusis 2014). - The Hydrometallurgical Residue Facility does not employ an electrical leak detection system, something that is becoming more commonplace as a CQA tool for geomembranes, as this technique provides greater confidence in the quality of the installation (Malusis 2014).

36

Recommendation: The SDEIS needs to address all of the engineering design problems of the Hydrometallurgical Residue Facility collection system identified in Dr. Malusis report. The SDEIS should describe a rigorous construction quality assurance process that includes continuous observation during installation. The SDEIS should include an electrical leak detection system for the liners. 5. Category 2/3 and Category 4 Stockpile Liners - The SDEIS does not describe a rigorous construction quality assurance process that includes continuous observation during installation and other methods. Otherwise, the predicted leakage rates, which are based on a 90th percentile frequency of only 4 defects per acre, may grossly underestimate the true leakage through the liners (Malusis 2014). - The maximum hydraulic conductivity proposed for the Category 2/3 stockpile liner differs from that proposed for the Category 4 stockpile liner. The SDEIS does not describe why the Category 2/3 allowance is not as rigorous at the Category 4. Chambers and Levit highlight this problem as well in their report. Recommendation: The SDEIS needs to address all of the engineering design problems of the Category 2/3 and 4 Stockpile liners identified in Dr. Malusis report and the Chambers and Levit report. The SDEIS should include a rigorous construction quality assurance process that incorporates the elements in both the Malusis and Chambers/Levit reports. The maximum allowable hydraulic conductivity for the Category 2/3 stockpile liner should match that for Category 4. M. Lack of a Reclamation Plan Chambers and Levit in their report (see attached) identify the lack of a mine Reclamation Plan to be a critical omission in the SDEIS. Every DEIS should contain a Draft Reclamation Plan which includes a detailed analysis of the financial surety, so that it is demonstrated before the project is allowed to proceed that project closure can be accomplished using demonstrated reclamation techniques, and at a cost that is affordable (Chambers and Levit 2014). The information about reclamation that has been provided in various background documents to the SDEIS, such as the Wetlands Management Plan and the Water Management Plan, lack detail and specific goals and methods that would allow for evaluation of the likelihood of success. As Chambers and Levit note, these plans are more accurately plans to plan reclamation than actual reclamation plans (Chambers and Levit 2014). A reclamation plan for the NorthMet project must also include a plan to address the structural issues and seepage associated with existing 2W tailings cell. The SDEIS notes that the rapid construction of this Cell resulted in oversteepened dams which has led to seepage and erosion (SDEIS 4-371). But the SDEIS does not describe a plan to address this existing problem with Cell 2W. This should be addressed.

37

Dr. Myers notes the failure of the SDEIS to describe a reclamation plan that includes the timing of activities Myers 2014). Recommendation: The SDEIS should include a full Reclamation Plan that provides details about goals, methods, financial surety, reclamation techniques that have been demonstrated, and timing of activities. Chambers and Levit outline numerous reclamation activity recommendations that are extremely important to address in the SDEIS. They also provide an example of a reclamation plan with an appropriate level of detail and cost estimate for use as a model (the Pogo Project Reclamation and Closure Plan, December 2002). Their report provides information about how to access this plan. The reclamation plan must include a plan to address the oversteepening of the Cell 2W embankments and the associated seepage. Chambers and Levit make other reclamation activity recommendations that should be included in the SDEIS including characterization of waste disposed of in the East Pit, evaluation of abandoned pipelines, and a tracking system for hazardous materials to ensure appropriate disposal and compliance with laws. In addition, the Reclamation Plan should describe the water balance desired for the pits post-closure. The SDEIS should contain details about wetlands construction. N. Under-Estimation of Acid and Leachate Generation Dr. Maests analysis of the proposed project leads her to the conclusion that the SDEIS underestimates the potential for the mine to generate acidic conditions and polluted drainage. it is reasonable to assume that leachate concentrations from the Category 1 and other stockpiles could be higher than predicted (Maest 2014). She describes her reviews of other mines and the waste rock leachate issues at those operations, and compares their pollution experiences with the NorthMet plan. Case study mines in this category were more likely to have adverse effects to groundwater and surface water resources (Maest 2014). If the SDEIS had under-estimated the potential for acid and metal leachate generation, the well-being of the groundwater and surface water in the area depends more than ever on a perfect functioning of the containment and capture systems. Dr. Maests review of similar mines, suggests the SDEIS places too much reliance on engineered mitigation measures. In addition, Dr. Maest finds that No traditional acid-base accounting (ABA) analyses were conducted on the samples- only sulfur, carbonate, and paste pH (Maest 2014). She notes that the near total absence of carbonate minerals suggests that ABA testing would show that all samples are potentially acid generating (Maest 2014). The SDEIS characterization of Category 1 Waste Rock as not acid generating could be wrong. MDNR cannot rule out the possibility that all wastes produced from the NorthMet Project will produce acid or leach metals within the 20year timeframe of proposed mining (Maest 2014). Dr. Maest disagrees with the assumption in the SDEIS that Category 1 wastes will never generate acid: Given operational realities and existing characterization results, this seems like a poor assumption and does not consider the potential to leach other contaminants, including those added by mining (Maest 2014).

38

Dr. Maests report describes in more detail her assessment of why the acid generating potential for the project has been under-reported. And she also notes that the SDEIS should not overly focus on acid drainage, given that lower sulfur waste has been shown to leach other contaminants at concentrations dangerous to aquatic life and human health. Recommendation: The SDEIS must re-examine the potential for all waste rock to generate acid and other leachate. It should incorporate the recommendation by Dr. Maest to backfill all the waste rock, including Category 1 wastes. The SDEIS should accurately portray data that indicate the acid generating potential of waste rock.

II. Failure to Meet Clean Water Principle #2: There Are Not Safeguards in Place for When Things Go Wrong A major area of concern for the Friends is that the SDEIS for the NorthMet proposed mine does not describe contingency plans for accidents, mishaps, spills, mechanical breakdowns and other unplanned events. While such events cannot be foreseen, they are not uncommon among mines of this type. Emergency plans should be in place to provide crisis management guidelines for the most anticipated types of events. A crisis is not the best time to make detailed emergency decisions. But the SDEIS provides none of this information. An Earthworks report examining 14 modern copper mines in five states showed that between 1988 to 2012, 100 percent of these mines experienced pipeline spills or other accidental releases. Of these mines, 92 percent experienced failures with water collection and treatment systems that resulted in significant water quality impacts (Earthworks 2012). The SDEIS needs to incorporate plans for addressing the most common mishaps that occur at modern copper mines. For example, during mine operations, over 6.2 million gallons of contaminated water will need to be treated every day between the WWTF and the WWTP. Should these treatment plants need to shut down, temporarily or for longer, what is the plan for dealing with this large amount of polluted water? What is the emergency response plan if a tanker car with Reject Concentrate derails and spills highly toxic material into the water systems along the transportation corridor? What if the amount of contaminated water that is collected from the Hydrometallurgical Residue Facility is higher than anticipated? What is the response plan if a pipeline breaks or if the emergency overflow channel at the Tailings Basin leaks contaminants? The SDEIS describes doing Adaptive Management as a means of addressing unplanned events and circumstances. In the event that the monitoring identifies the potential for any water quality exceedances, PolyMet has proposed an Adaptive Water Management Plan (AWMP) that identifies additional measures the firm could take if necessary to prevent any exceedances of water quality standards (SDEIS 5-8 to 5-9). The SDEIS also states, There would be few cumulative effects from the NorthMet Project Proposed Action after proposed mitigation and adaptive management measures are applied (SDEIS ES-42), but the SDEIS does not describe

39

what the adaptive management measures would be, and the public cannot assess if they adequately mitigate the impacts. The adaptive management concept is described more in a supporting document to the SDEIS called NorthMet Project Adaptive Water Management Plan, March 7, 2013. But even in this document there is little information about effectiveness of possible actions. For example, if after installation the geomembrane cover system on the Category 1 Waste Rock pile is not performing correctly, the document describes certain actions that might happen such as overseeding and/or fertilizer application to improve vegetation density or organic matter addition to rooting zone layer to improve vegetation density (PolyMet Mining March 7, 2013). But the document does not say if this has been done elsewhere and how effective this is at addressing pollution discharge from a waste rock pile. If the WWTP at the Tailings Basin is not performing as needed, the document states steps can be taken to adapt the system, such as: selection of alternative membranes for either the Reverse Osmosis (RO) or the Vibratory Shear Enhanced Processing (VSEP) process units to modify the removal efficiencies of such parameters across these systems or chemical addition to increase metals removal by the WWTP (PolyMet Mining March 7, 2013). Again, it does not say if these approaches have been used at other mines or how effective they are. Missing from the SDEIS analysis is a description of the most likely unplanned events and effective measures and protocols to address them. Examples and evidence of successful response measures need to be provided. Without this information, the public cannot fully assess environmental impacts and reliable mitigation responses. Recommendation: The SDEIS must contain descriptions of the most likely unplanned events, such as pipeline breaks, spills, failure of treatment or collection systems, mechanical problems, liner and cover failures. The SDEIS also needs to outline responses to these types of events. Crisis management works best when a response framework has been prepared in advance. The public needs to see this response plan in order to assess the range of environmental impacts that may occur. An Environmental Impact Statement should have this information in it. The SDEIS needs to include supporting evidence for the success of responses at managing unplanned events, including examples from other mines. Dr. Myers recommends that the probabilistic modeling should be expanded to include assumptions that liners and containment systems work poorly. Instead of 99% capture, they should test what occurs with as little as 10% captured. Also they should consider what occurs if the capture fails for a given period of time (Myers 2014). He suggest modeling explore what happens if the cutoff wall and pump systems do not perform as designed.

III. Failure to Meet Clean Water Principle #3: The Mine Plan Does Not Leave the Site Clean and Maintenance Free or Comply with Minnesota Laws The SDEIS outlines a mine plan that the Friends believes would violate several of Minnesota State Rules.

40

A. Minnesota Rules 6132.3200 Minnesota Rules 6132.3200 states: The mining area shall be closed so that it is stable, free of hazards, minimizes hydrologic impacts, minimizes the release of substances that adversely impact other natural resources, and is maintenance free. Yet the SDEIS describes a mine and connected Tailings Basin that would require centuries of maintenance. Some examples of this on-going maintenance include: The operation of two wastewater treatment plants for at least 200 years at the Mine Site and at least 500 years at the Plant Site Seepage collection system management at the Tailings Basin and Category 1 Stockpile Erosion repair Removal of woody species from Category 1 Stockpile cover system Maintenance of pipelines Monitoring of various types including the Hydrometallurgical Residue Facility for the presence of water and of restored wetlands Passive water treatment wetlands, if created Pumping water to maintain levels at the West Pit Pumping water to maintain levels at the Tailings Basin

Active water treatment for centuries defies the rule calling for the site to be maintenance free at closure. B. Minnesota Rules 6132.4800 Minnesota Rules 6132.4800 states: No release from the permit to mine under part 6132.4800 shall be granted for those portions of the mining area that require postclosure maintenance until the necessity for maintenance ceases. If this provision were followed, PolyMet or whatever mining company appears on the permit, would not be released from responsibility for those portions of the mine that still require maintenance. Given that the NorthMet SDEIS calls for at least 500 years of active water treatment and other activities likely to last as long, PolyMet would never be released from its obligations. Common sense, however, indicates that holding a company to a permit obligation for 500 years is neither practical or feasible. If this is the case, then PolyMet and the State of Minnesota would enter into a permit agreement to which both parties do not expect to adhere. From the outset, PolyMet would know it would not be expected to bring the project to a real and meaningful closure and that compliance with Minnesota Rule 6132.4800 will not be asked of them. Minnesota mining permits should only be granted if all legal conditions can be met and if the State of Minnesota plans to hold the mining company responsible to all provisions in the law. This SDEIS describes a mine that cannot comply with the law as it is presently conceived.

41

C. Minnesota Rules 6132.1300 Minnesota Rules 6132.1300 requires the submission of an annual report by the permittee to: describe actual mining and reclamation completed during the past year, the mining and reclamation activities planned for the upcoming year, and a contingency reclamation plan to be implemented if operations cease in the upcoming year. Among other updates to be provided, the report must include, a description of the status of ongoing postclosure maintenance activities, and a discussion of changes in ownership or organizational structure of the permittee, and notification of intent to close a mining area or portion of an area. Common sense suggests that PolyMet or whatever mining company is listed on the permit, will not be in a position 500 years from now to file annual reports with descriptions of postclosure maintenance activities. Entering into a permit agreement with the mine as currently designed would violate this rule, and send a message to the mining company that the State of Minnesota does not intend to hold them accountable to the terms of the permit. D. Minnesota Rules 6132.1200 Minnesota Rules 6132.1200 states: The commissioner shall release the permittee from the responsibility to maintain financial assurance when the commissioner determines, through inspection of the mining area, that: (1) all reclamation activities have been completed according to this part and the permit to mine; (2) conditions necessitating postclosure maintenance no longer exist and are not likely to recur; and (3) corrective actions have been successfully accomplished. The mining company would never be released from financial assurance responsibilities given that the need for postclosure maintenance would not cease. Again, common sense suggests this is not a practical arrangement. But if the company is released from its financial assurance responsibilities before maintenance needs have ceased, this Minnesota Rule will be violated. The risk is that the mining company knows from the outset that the State of Minnesota will not hold them to their permit obligations. E. Minnesota Rules 6132.1100 Minnesota Rules 6132.1100 requires: a certificate issued by an insurance company authorized to do business in the United States under Minnesota Statutes, section 93.481, subdivision 1, clause (2), confirming that the applicant has a public liability insurance policy in force for the mining operation for which the permit is sought or evidence that the applicant has satisfied other state or federal self-insurance requirements, to provide personal injury and property damage protection in an amount adequate to compensate persons who might be damaged as a result of the mining operation or any reclamation or restoration connected with the operation In addition, the Annual Report described in Minnesota Rules 6132.1300 includes a requirement that proof of insurance be submitted annually. Common sense suggests that PolyMet or whatever mining company will be listed on the permit will not be in operation 500 years from now to maintain the required liability insurance. Since 6132.1100 requires this insurance to cover liabilities related to reclamation and restoration, which is expected to be long-term, this suggests that the State of Minnesota does not intend to hold them accountable to the terms of the permit.

42

Recommendation: Minnesota should not approve a mine plan that cannot comply with Minnesotas existing laws. The SDEIS describes a mine that, as presently conceived, could not comply with Minnesotas Rules for nonferrous mining. The SDEIS should include a revised mine plan that can meet these laws.

IV. Failure to Meet Clean Water Principle #4: Minnesota Taxpayers Will Not Be Protected Minnesotas financial assurance provisions for nonferrous mining are designed to protect Minnesota taxpayers from financial liabilities that belong to the mining company. However, the SDEIS fails to provide information that enables the public to determine if it is being protected. Financial assurance is also designed to ensure funds are available to address environmental impacts, but without financial assurance details, the SDEIS does not enable the public to fully analyze the full environmental impacts from the proposed NorthMet project. A. Financial Assurance Information Lacking The SDEIS fails to provide information about financial assurance for the proposed project. This is a fundamental problem with the SDEIS. While the document provides a general list of the kinds of activities that might be factored into a financial assurance calculation, it does not provide details or specifics. A preliminary cost estimate range is provided at years 1, 11, 20 and for post-closure, but the specifics of how these numbers were calculated and what factors were included are not provided. The SDEIS notes that PolyMet has developed preliminary cost estimate ranges, (SDEIS 3-137) but does not indicate that the state and federal co-lead agencies have verified that these numbers are realistic preliminary estimates. Inclusion of financial assurance in the SDEIS is key if the public is to understand and assess the potential effectiveness of reclamation and closure activities. The SDEIS indicates that active water treatment will be needed long-term, likely for centuries, making a timely analysis of financial assurance essential in fully appreciating the potential environmental impacts. While the SDEIS notes that the cost estimates would be finalized by the MDNR during the permitting process (SDEIS 3-138), the details of preliminary cost estimates are available now. PolyMet based their estimates on page 3-138 in the SDEIS on calculations they have run, and these should be made public in the environmental review process. After years of design preparation by PolyMet, it is unlikely that significantly new information regarding design details will transpire between the release of the SDEIS and permitting. The U.S. EPA has urged that financial assurance disclosure occur during the National Environmental Policy Act (NEPA) process. The U.S. EPAs National Hardrock Mining Framework recommends that the: EPA should evaluate the adequacy of EISs for mining operations in predicting the long-term environmental impacts of mining operations. Assessment of financial assurance mechanism that will be utilized to provide funding of required long term environmental management systems is critical to this analysis (U.S. EPA 1997). 43

The U.S. EPA has specifically urged that financial assurance details appear in the PolyMet EIS documents. In an August 2009 letter regarding the NorthMet project, the U.S. EPA stated regarding the lack of financial disclosure for this project: The EPA recommends including financial assurance information because one key component to determining the environmental impacts of a mine is the effectiveness of reclamation and closure activitiesEPA has recognized the importance of disclosing financial assurance in EISs (U.S. EPA August 2009). The lack of financial assurance information in the NorthMet projects first DEIS was part of the U.S. EPAs major criticisms of the project when in February 2010, it gave the NorthMet Projects DEIS a rank of Environmentally Unsatisfactory-Inadequate. In that 2010 letter, the U.S. EPA stated, EPA believes financial assurance information should have been included in the DEISFinancial assurance information includes a description of State and/or federal agency requirements, closure costs, estimated bond amounts needed for each closure and reclamation activity, and how the bonds should be modified should additional temporary, long-term, or perpetual treatment and/or remediation needs be determined during operations (U.S. EPA February 18, 2010). The revised SDEIS still lacks the financial assurance components the U.S. EPA asked for four years ago. Missing is information that illuminates annual projected operating costs, costs associated with operating the reverse osmosis treatment process, capital replacement costs for pollution control infrastructure such as pipelines and pumps, and an analysis of replacement intervals and costs for equipment. The SDEIS does not include information comparing the value of a dollar today with the value of that dollar in the future, the Net Present Value of the financial assurance. Dr. David Chambers has also highlighted the omission of financial assurance calculation as a major flaw in the document. Probably the most glaring omission is that there is only the most scant analysis of the financial surety that will be needed for this projectThe financial surety for this project could be in excess of $400 million. This is very significant potential impact not only to the financial requirement s of the mine owners, but also to the citizens of Minnesota, who are ultimately accountable should the mine operator go bankrupt without an adequate financial surety to close the mine and treat waste water (Chambers and Levit 2014). Dr. Miller also identifies the lack of financial assurance disclosure in the SDEIS to be a significant omission. Bonding and long-term water treatment estimates are major environmental issues. No estimates of bonding for reclamation and long-term water treatment and management were provided, and this is a major deficit of the analysis presented in the SDEIS (Miller 2014). The SDEIS asserts financial assurance information is not yet available for disclosure in the SDEIS: The level of engineering design and planning required to calculate detailed financial assurance amounts is typically made available during the permitting process (SDEIS 2-10). Chambers and Levits report refutes this assertion: This statement is probably not correct. The engineering design and planning to calculate detailed financial assurance is available at the point a project reaches the DEIS stage. For a company to reach this stage without having this

44

information available, and without having performed this calculation internally, would be fiscally irresponsible to the companys shareholders and board of directors (Chambers and Levit 2014). PolyMet Mining has already made initial calculations about financial costs that should be a part of the SDEIS. A Barr Engineering memo from May 2011, provides calculations about the net present value for water treatment systems at an estimated $91.7 million for a twenty year operation life (Barr Engineering May 32, 2011). So too, the SDEIS provides an annual postclosure monitoring and maintenance estimate of $3.5 to $6 million (SDEIS 3-138). The existence of this estimate means preliminary calculations have been made by PolyMet, and these numbers should be shared with the public. While it may be customary for the Minnesota DNR to address financial assurance in the permitting phase, there is no legal reason preventing disclosing the information in the environmental review process, and every cautionary reason to do so. The EIS process is when the public has an opportunity to review the project as a whole and to assess both the environmental impacts and the resources necessary to mitigate those impacts. Because so many various permit applications need to be granted by a variety of state and federal agencies, it becomes much more difficult for the public to assess the project as a whole in the permitting phase. Pairing environmental review with financial assurance information makes more sense and provides for greater public transparency. Recommendation: Including financial assurance information in Environmental Impact Statements has become standard practice, especially for mining projects in which long-term water treatment is expected. The PolyMet project anticipates centuries of water treatment and the SDEIS should include the necessary financial assurance calculations for the public to assess risks and impacts. The financial assurance disclosures should include closure costs, estimated bond amounts needed for each closure and reclamation activity, and how the bonds should be modified should additional temporary, long-term, or perpetual treatment and/or remediation needs be determined during operations. Disclosed information should include capital replacement costs and frequencies and assumptions on interest rates. B. What Financial Assurance Should Include Financial assurance calculations need to include the costs of unexpected events, accidents, developing pollution problems, and the staffing needed to maintain equipment and manage the site. The SDEIS notes that Sulfide-bearing rock from the NorthMet Project Proposed Action may oxidize for several years before producing acidic leachate (SDEIS 5-51) and if the porewater pH were to shift from neutral to acidic, then the rate of sulfide mineral oxidation and associated release of some metal cationswould increase dramatically (SDEIS 5-51). Calculations of financial assurance need to include funding for contingencies such as an increased rate of release of metal contamination. The SDEIS does not assess how many employees will be required to conduct the long-term postclosure work of running the treatment facilities, maintaining pipelines and collection systems, 45

pumps and waste rock piles. Centuries of long-term maintenance will be needed at the mine and plant sites. Regarding the number of workers, the SDEIS notes, Current estimates.include 30 to 50 FTEs for the first 7 yearsand 5 to 10 FTEs for the following 30 years (SDEIS 5-499). It is not clear why the SDEIS stops accounting for the number of workers needed after 37 years when the document indicates a need for maintenance activities for centuries. The SDEIS provides an annual post-closure monitoring and maintenance estimate of $3.5 to $6 million (SDEIS 3-138). However, the details of how this estimate was calculated are not disclosed. This is a glaring omission. The public cannot evaluate the liability risks or the ability of the estimate to adequately address environmental impacts without a full understanding of what was included in calculating the estimate. Also troubling, is that it appears that the Co-lead Agencies have not analyzed the details or adequacy of this estimate. One of PolyMets consulting companies, Foth Infrastructure and Environment, LLC, developed these estimates and presented them in a March 11, 2013 memo to PolyMet Mining. It appears that the Co-lead Agencies adopted these estimates for inclusion in the SDEIS without verification that they represent realistic cost numbers and without asking for details of what was or was not included. This is not the regulatory oversight the public expects and should receive from our state and federal regulatory agencies. The responsibility of the Co-lead Agencies is to verify information provided by the mining company. Chambers and Levit urge in their report that initial financial assurance calculations be based on worst-case water quality predictions, given that there are too many examples of situations where conditions turned out to be worse than the worst-case prediction (Chambers and Levit 2014). Chambers and Levit developed a Net Present Value (NPV) spreadsheet in their report to estimate the NPV of the long-term costs for the project (Chambers and Levit 2014). This is the calculation needed to determine what estimate of financial surety may be required for the NorthMet project. In their calculations, Chambers and Levit show that different interest and return on investment rates, as well as estimates for operating costs, can make hundreds of millions of dollars difference in the financial surety required. They note that the difference between discount rate and inflation is significant for the calculations, not the individual values of the discount rate and inflation themselves (Chambers and Levit 2014). Small changes in the calculations can result in significant differences in how much financial assurance should be available. This is why it is so critical that the public has an opportunity to review these estimates and their details within the EIS process. Dr. Miller urges that financial assurance include items including costs for long-term treatment, the cost of repairing the covering on the Category 1 Stockpile, and the replacement and upgrading the water treatment facilities every 50-75 years (Miller 2014). The SDEIS also does not describe how mitigation of potentially several thousands of acres of indirect wetland impacts would be financed. Past large-scale wetland mitigation projects have required millions of dollars of financial assurance. The SDEIS has not proposed any financial assurance for wetland mitigation. This leaves the public without confidence that the State will have the ability to mitigate the impacts, and will be able to ensure that no net-loss of wetlands occurs as a result of the project.

46

Recommendation: The SDEIS must include in financial assurance calculations funds to cover accidents and unexpected events. Financial assurance costs should include the number of employees necessary to maintain the plant and mine sites for the centuries of management activities that are anticipated. The SDEIS must include detailed information about financial assurance with transparency about what the estimates include. The SDEIS should indicate that the calculations and estimates have all been reviewed and verified by the Co-lead Agencies. Calculations initially should be based on predicted worst-case scenarios until operational data can prove conditions to the contrary. Because the interest and return on investment rates can result in financial assurance variations that are hundreds of millions of dollars different, the SDEIS must make the assumptions and calculations available for comment. Financial assurance calculations should include mitigating indirect wetland impacts and should be fully described in the SDEIS.

V. Failure to Analyze Alternatives Both state and federal regulations require that the Environmental Impact Statement consider alternatives to the proposed project. The National Environmental Policy Act requires that a range of alternatives be examined in the environmental review process. This includes all practicable alternatives, which must be rigorously explored and objectively evaluatedThe emphasis is on what is practicable rather than on whether a proponent or applicant prefers or is itself capable of carrying out a particular alternative (SDEIS 1-13). Despite this, the SDEIS provides no meaningful consideration as required by law of alternative means to achieve the project Purpose and Need. The SDEIS fails to show that the project proposers have given rigorous evaluation to alternatives that may be feasible and mitigate environmental impacts. Serious evaluation is only give to an alternative that would convey fewer acres in the land exchange, and the No Action Alternative. The SDEIS acknowledges this limited scope: As a result of screening and analysis, the NorthMet Project No Action Alternative (i.e., the NorthMet Project Proposed Action would not occur) is the only alternative evaluated in detail in the SDEIS (SDEIS 3-5). The SDEIS fails to show adequate evaluation of two other proposed alternatives that are feasible and offer the potential for reducing environmental impacts. Limiting the scope of alternative evaluation to the No Action Alternative and a minor change in the land exchange does not meet the legal requirement for alternative evaluation. Additionally troubling is that the SDEIS is not clear what the overall project Purpose and Need is. In screening for alternatives, one criteria used is whether an alternative can meet the Purpose and Need for the project (SDEIS 3-140). But the SDEIS is not clear what the project Purpose and Need is. The SDEIS provides an Applicants Purpose and Need, and it provides Co-lead Agencies statements of Purpose and Need. It does not say which of these is used to screen for alternatives. As noted in the Friends comments about Agency Roles and Responsibilities, the Co-lead Agencies statements in the SDEIS are not aligned with the agencies legally mandated

47

missions. The SDEIS needs to be clear about what the project Purpose and Need is for use in screening alternatives. A. Underground Mine Alternative Not Adequately Considered While the NorthMet proposed mine would develop three open pit mines, the option of constructing an underground mine has been proposed as a way to limit the surface impacts and reduce water quality impacts. The SDEIS acknowledges these potentially significant environmental benefits from an underground mine: An underground minewould result in a smaller surface footprintreduced effects on wetlands, vegetation, and wildlife habitat, less fugitive air emissions, and less waste rock and processing wastethus reducing the scale and duration of potential water quality effects (SDEIS 3-150). An underground mine would also eliminate the need for a land exchange with the U.S. Forest Service and allow the landscape to continue to be available for public use. In addition, the Co-lead Agencies concede that the alternative is technically feasible: The underground mining alternative is technically feasible for the NorthMet Deposit (Co-lead Agencies September 27, 2013). Yet the SDEIS dismisses the underground alternative without any further examination: PolyMet concluded underground mining would not be economically feasible and would not generate enough revenue to pay for all costs associated with underground mining (SDEIS 3150). The economic feasibility was evaluated using a model called InfoMine. Neither the SDEIS nor the Underground Mining Alternative Assessment provide details of this model or its assumptions. Nor is it clear that the Co-lead Agencies reviewed the model and its assumptions. It appears the co-lead agencies have accepted PolyMets assertion without verification that an underground mine is not economically feasible. Chambers and Levit note in their report (see attached) that the economic viability of an underground option varies with changes in metals prices. They propose sensitivity analyses that would perform better at assessing the potential for market changes to affect the viability of this option (Chambers and Levit 2014). GLIFWCs comments also point out the failure of the SDEIS to give adequate consideration to an underground mine alternative. GLIFWC notes that the co-leads have not included in their analysis that an underground mine would not require a $4 million land exchange with the Forest Service, and that the alternative would also provide economic benefits from wetlands that would no longer need to be destroyed with the creation of an open pit mine. Additionally, an underground mine could provide economic benefits by potentially not requiring perpetual water treatment and maintenance (SDEIS, Appendix C, GLIFWC Comments, Underground Mine and West Pit Backfill Alternatives). An underground mine is technically feasible and offers the potential for economic benefits and significant environmental benefits. This alternative deserves full analysis in the EIS.

48

B. West Pit Backfill Alternative Not Adequately Considered The NorthMet proposed mine would excavate three open pits. At the end of the mines operations, two would be backfilled with Category 2/3 and Category 4 waste rock. But the West Pit is designed to remain a pit lake, while Category 1 waste rock remains on the surface. While backfilled waste can continue to be a source of pollution, it is considered less of a pollution risk than leaving waste rock on the surface. Surface waste requires cover systems and collection systems, and on-going maintenance to ensure geomembrane liners are not perforated. Backfilling the West Pit with the Category 1 waste rock has been proposed as a method to help limit water pollution risks. And the SDEIS acknowledges some of the potentially significant benefits from backfilling: Removal of the Category 1 Stockpile would allow for reclamation of the affected surface footprint, including potential to recreate wetland areas and restore function (SDEIS 3-151). However, the SDEIS does not give this alternative serious evaluation, claiming the benefits are not substantial enough. it was determined that it would not offer substantial environmental or socioeconomic benefits as compared to the NorthMet Project Proposed Action (SDEIS 3-151). The arguments against greater evaluation of the West Pit Backfill concept are not persuasive and reveal a failure to meet the requirement for rigorous review: 1. Backfilling and Water Quality Benefits The SDEIS provides illogical reasoning to back its claims that backfilling would not offer water quality benefits. The SDEIS even asserts backfilling could exacerbate water quality problems: Backfilling would affect the water quality in the West Pit by increasing constituent loads, so additional mechanical treatment of water in the West Pit may be required for a certain timeframe following backfilling (SDEIS 3-151). This argument makes little sense for several reasons. First, the SDEIS has already identified the need to conduct long-term (centuries) mechanical treatment of West Pit drainage. There is no evidence that adding Category 1 waste rock would extend this period of time. Second, treating drainage from the Category 1 Stockpile has also been identified as a need for the long-term (centuries). Backfilling the West Pit with Category 1 waste rock reduces the need to collect polluted drainage from two sources to one source a source already planned for long-term treatment anyway. This is not an insignificant benefit when one considers that the Category 1 Stockpile is expected to be 526 acres and 240 feet high, and must be covered, monitored, and large quantities of drainage collected. The SDEIS fails to acknowledge this benefit. Third, the SDEIS plan already calls for backfilling the East/Central Pits with waste rock that is more reactive than what would go into the West Pit. The backfilling approach is not eliminated from consideration for this part of the mine. Why is it considered an appropriate method for the East/Central Pits but not for the West Pit as well?

49

In addition, the SDEIS claim that the backfill alternative would not offer substantial environmental or socioeconomic benefits is not based on any information that the Colead Agencies have gathered or been provided by PolyMet. The Co-lead Agencies note in the Co-lead Agencies Consideration of a West Pit Backfill Alternative from April 11, 2013, that they did not assess quantitative surface and groundwater quality projections from the Mine Site GoldSim model under a backfill condition, or detailed cost/saving estimates for long-term water treatment and maintenance under a backfilled condition. The Agencies do not know what the environmental or socioeconomic benefits might be from this alternative because they have not analyzed them. Dr. Tom Myers (see attached report) also concludes that the West Pit Backfill Alternative was rejected without substantial argument or technical analysis (Myers 2014). He notes that the constituent load reaching groundwater from backfilling could be substantially less the perennial load from the Category 1 Stockpile, especially if the Cat 1 stockpile containment system does not work as well as expected (Myers 2014). Dr. Miller (see attached report) also concludes that backfilling mine waste into the West Pit should be considered. The SDEIS should thoroughly evaluate the alternative of putting all of the Category 1 Waste material into the East Pit, or potentially, the West Pit, or a combination of the various pits. Leaving the reactive rock on the surface provides an in perpetuity water treatment requirement, and a very long term management problem when the plastic liners degrade (Miller 2014). GLIFWC notes regarding the backfill option: it is clear that this alternative meets the purpose and need, is available, is technically feasible and is economically feasible (SDEIS, Appendix C, GLIFWC Comments, Underground Mine and West Pit Backfill Alternative). 2. Backfilling and Encumbering Private Minerals The SDEIS also eliminates backfilling the West Pit from consideration because it would encumber private mineral resources that are deeper than the proposed West Pit. Such an encumbrance is in conflict with the terms of PolyMets current private mineral leases (SDEIS 3-152). This makes little sense. The minerals would be no more encumbered than they are now, with 696 feet of soil, rock and ore above them. It is not PolyMets responsibility, nor the Minnesota publics, to provide the holders of mineral rights to deposits deeper than the current NorthMet deposit, with a new, more accessible portal to access their minerals. This rationale does not belong in an Environmental Impact Statements evaluation of viable alternatives and it should certainly not be used as a reason to eliminate a potentially important alternative from further analysis. In addition, if a plan already exists to use the West Pit as a portal for future mining, this concept must be addressed and analyzed in the SDEIS. As the tribal cooperating agencies note: If the West Pit is to be used as a portal for this future mining, then that

50

should be described in the PSDEIS and the environmental consequences assessedIt is unclear why a regulatory agency would prefer this method without first conducting an analysis (SDEIS, Appendix C, GLIFWC Comments, Underground Mine and West Pit Backfill Alternative). Such a plan would constitute a new, connected, and potentially significant aspect of the proposed NorthMet project and must be evaluated in the EIS process. There is information that suggests this portal concept may indeed be a planned but unanalyzed part of the NorthMet project. Edison Investment Research Limited conducted an investment analysis of PolyMets NorthMet proposed mine project and released a report in November 2013. In this report, which used information provided by PolyMet Mining, Edison states: We believe there is a good chance PolyMet will be able to expand the size of its resource by 50-100 % based on what we learned on a site visit. The eastern end of the pit is cut off by the property boundary with the Teck-Mesaba project. However, down dip to the south and west the geology is openAlso, based on drill work to date there is a chance PolyMet will be able to identify economic mineralization on the hanging wall and at depth (Edison Investment Research Limited, November 21, 2013). The area described in this statement is at the base of where the West Pit would be located. The report also notes that expansion of PolyMets project would increase the companys production by 90,000 tons per day. We assume PolyMet would begin working on permitting the expansion to 90,000t/d within six months of receiving its permits for Phase 1 (Edison Investment Research Limited, November 21, 2013). The Edison report is evidence that PolyMet Mining plans an expansion, most likely using the West Pit as a portal. This is undoubtedly the reason the company has resisted a West Pit Backfill Alternative analysis. This planned expansion needs to be evaluated as part of this environmental review process in this SDEIS. Until this planned expansion through the West Pit is included and evaluated as part of this SDEIS, it cannot be used to eliminate from consideration a backfill option that presents the possibility for meaningful environmental benefits. The Co-lead Agencies have a responsibility to ensure that all connected actions to this project are evaluated, presented truthfully to the public, and that reasonable and useful alternatives are given full consideration. Recommendation: The SDEIS must include complete evaluations of the Underground Mine Alternative and the West Pit Backfilling Alternative. Details about the economic modeling and the models assumptions for the Underground Mine Alternative should be given to the Co-lead Agencies, thoroughly reviewed by the agencies, and described in the SDEIS for the public. The evaluation must include reviewing the economic benefits that would transpire from not doing the land exchange, from not destroying the wetlands, and any changes in the need for long-term water treatment. A metals cost sensitivity analysis should be included in the evaluation of the Underground Alternative to verify that the option is not economical with higher metals prices.

51

The SDEIS must fully evaluate the benefits of backfilling the West Pit. The evaluation should not include arguments or assertions that have no evidence or information to support them. Claims about needing longer water treatment with this alternative must be supported with evidence or be dropped. If this assertion is retained, it must be applied in an evaluation of backfilling the East/Central Pits as well. The SDEIS needs to include a complete evaluation of the water quality impacts from the Backfill Alternative in the GoldSim modeling. It should include detailed cost/saving estimates for longterm water treatment and maintenance under a backfilled condition. The SDEIS should cease using the minerals encumbrance issue in its evaluation and rejection of the backfill alternative. It is not the responsibility of the regulatory agencies or the Minnesota public to provide potential future projects with access at the expense of useful alternatives to this current proposed project. The SDEIS must include real and meaningful evaluations of all alternatives and not limit the scope of alternatives to those that are simply what the applicant prefers. Additionally, the Co-lead Agencies need to insist that any planned expansions from the West Pit be incorporated into this environmental review process for evaluation as a connected action.

VI. Wetland Impacts The proposed NorthMet project would result in significant wetlands impacts, both direct destruction from the development of the mine, and indirect impacts from creating hydrological changes in the area. The project would directly affect 912.5 acres of wetlands located within the NorthMet Project area, mostly within the Mine Site, as a result of activities such as filling, excavation, and installation of a containment system within the wetland boundary. These wetlands would be permanently lost. The SDEIS estimates that between 6,498.1 and 7,350.7 acres of indirect impacts to wetlands may result from the project (SDEIS ES-49). The SDEIS does not fully describe the current wetland values, does not provide an adequate wetlands mitigation plan, does not accurately characterize the hydrologic connection between wetlands and underlying groundwater, and does not accurately describe the likely wetland impacts. A. Loss of High Quality Wetlands 100 Mile Swamp The wetlands that would experience direct and indirect impacts are especially high quality ecosystems. Approximately 92 percent of the wetlands in the Mine Site are of high overall wetland quality..." (SDEIS 4-157). "The wetlands along the Transportation and Utility Corridor have all been rated as high-quality" (SDEIS 4-157). As the SDEIS acknowledges, The majority of wetlands that would be affected by the NorthMet Project Proposed Action would be difficult to replace (coniferous bog, open bog, coniferous

52

swamp, and hardwood swamp) (SDEIS 5-313). But the SDEIS does not fully disclose how unique these wetlands are. The proposed mine site was identified in the late 1990s as special habitat worthy of protection in two assessments, one conducted by the U.S. Forest Service and another by the MN DNR. In these evaluations, the scientists concluded that the mine site, part of what is called the 100 Mile Swamp, represents some of the highest quality habitat remaining in the landscape. In January 1997, as part of its preparation for the Forest Plan Revision, the Superior National Forest released a report titled, Identification of Potential Natural Areas, Including Representative Ecosystems, on the Superior National Forest by forest biologist Robin Vora. The report summarized a process to identify natural areas on the forest that might qualify for permanent protection. The focus of the study was to develop a list and general description of the highest quality remaining examples of common ecosystems present in each Landtype Association (LTA or landscape). The report notes, A network of natural areas helps to protect biological diversity at the genetic, species, ecosystem, and landscape scales. Natural areas representative of common ecosystems in natural conditions serve as baseline or reference (Vora 1997). The PolyMet mine site was identified in this report as part of the 100 Mile Swamp. It sits within LTA 8A, and was considered an ecosystems in natural condition that was representative of its landscape, and a good candidate for protected status. The site is especially important given that LTA 8A was noted as lacking ecosystem representation in protected areas. Features that gave this area a high ranking were its watershed integrity, the size of its wetlands, the presence of riverine ecosystems, and the large amount of interior forest present. In December 1997, the Minnesota DNR released a report called, Evaluation of Selected Potential Candidate Research and Natural Areas as Representative Ecological Landtype Associations on the Superior National Forest, Minnesota. This study was conducted by plant ecologist Chel Anderson. The purpose of this assessment was to continue the evaluations begun by Robin Vora. The assessment was to further assist the Superior National Forest in evaluating areas for protection for the Forest Plan Revision process. The study analyzed the 93 sites identified by Vora and developed a shorter list of 45 sites worthy of consideration as protected natural areas. The assessment notes that these sites represent the highest-quality remaining examples of characteristic ecosystems in each ecological Landtype Association on the Superior National Forest. Again, the 100 Mile Swamp appears on this list of worthy candidates. The report notes, Inclusion of the 100 Mile Swamp site would very likely complete representation of the prominent ELTs [ecological landtypes], and provide some additional upland diversity (Anderson 1997). Representative ecological systems within the Laurentian Highlands, the ecological subsection in which the PolyMet mine site is located, include forested communities dominated by aspen-birch, jack pine barrens, red and white pine on uplands, and conifer bogs and swamps in the lowlands. The PolyMet mine site sits on a conifer bog and lowland swamp. The 100 Mile Swamp remains an important example of its ecosystem within its landscape.

53

The industrialization of this area would be an irrevocable loss to the State of Minnesota. Habitat fragmentation, loss of biological diversity, loss of representative habitats types, and climate change are threats to Minnesotas natural resources that only make sites like the 100 Mile Swamp more valuable. Recommendation: The SDEIS should acknowledge the high quality habitat that this proposed mine would destroy and that these wetlands are irreplaceable. Their value and loss should be placed in a statewide and regional context within the EIS. B. Inadequate Wetlands Mitigation The SDEIS does not provide an adequate mitigation and replacement plan for the significant wetlands losses that are expected. To mitigate the more than 900 acres of direct impacts, the SDEIS plans the off-site restoration of 1,631.4 acres. Restoration sites are identified in Hinckley, Aitkin, and Zim, Minnesota. Only the Zim site is located within the NorthMet project watershed. Aitkin and Hinckley are located outside the projects watershed. This means that the majority of wetland restoration, 1,445 acres, would take place in watersheds other than where the wetland destruction occurred. The wetland functions performed by the destroyed wetlands will not be replaced for the project area. Dr. Glasers report also points out the inadequacies of the mitigation plan. The NorthMet team seems to be committed to their previous plans to restore wetlands at the Zim, Hinckley, and Aitken County sitesWe are missing a golden opportunity to preserve and restore one of the most outstanding peatland complexes in northeastern Minnesota (Glaser 2014). Dr. Glasers report describes a wetland complex in the St. Louis River watershed that has been previously disturbed, but contains excellent examples of raised bogs, the most outstanding patterned fen in northeastern Minnesota, and populations of several rare, endangered or threatened plant species... (Glaser 2014). He notes the site is on tax-forfeited land and available for purchase by the State. The SDEIS should acknowledge that its replacement plan cannot compensate for the loss of centuries-old peatlands and their unique functions. The SDEIS notes, it is not practicable to replace all affected wetland types with an equivalent area of in-kind wetlands (SDEIS ES-37). Mitigation of indirect wetland impacts is not even described in the SDEIS. Wetland mitigation for potential indirect wetland effects would be determined by the agencies during permitting (SDEIS ES-38). The EIS process is the stage at which the public evaluates environmental impacts. If wetlands mitigation plans are absent, then a full understanding of the impacts cannot be made. The SDEIS is incomplete without a mitigation plan for indirect wetlands impacts. In addition, it is unclear where the compensatory wetlands might even be found. Given that full replacement wetlands have not yet been identified for those wetlands expected to be directly impacted, it is hard to understand how the 6,000 to 7,000 acres of wetlands potentially damaged indirectly will be adequately mitigated. If monitoring of wetlands for potential indirect effects 54

did determine effects were occurring, additional compensation may be requiredPermit conditions would likely include an adaptive management plan to account for any additional effects that may be identified in the annual monitoring and reporting (SDEIS 5-336). The SDEIS fails to describe what this adaptive management plan might include. The absence of information leads to worries that many impacted wetlands may never be compensated. A robust, full discussion of this is necessary in the SDEIS. In addition, the U.S. Army Corps of Engineers (USACE) has not yet made a determination about the wetlands compensation ratios required for the project. This is also information important to have in the EIS to evaluate the environmental impacts of this proposed project. The SDEIS also does not describe how mitigation of potentially several thousands of acres of indirect wetland impacts would be financed. Past large-scale wetland mitigation projects have required millions of dollars of financial assurance. The SDEIS has not proposed any financial assurance for wetland mitigation. This leaves the public without confidence that the State will have the ability to mitigate the impacts, and will be able to ensure that no net-loss of wetlands occurs as a result of the project. Recommendation: The SDEIS needs to address the inadequacies of the mitigation plan outlined in Dr. Glasers report. The SDEIS should include a mitigation plan for both direct and indirect wetlands. It should include analysis of possible compensatory wetlands for direct and indirect wetlands. A framework should be described for how any additional effects to wetlands may be mitigated. Wetlands compensation ratios need to be included in the SDEIS. The SDEIS should also analyze the wetland identified in Dr. Glasers report as a possible mitigation site and explain any reasons to exclude it as an option. The SDEIS should seek to replace wetlands within the affected watershed. If appropriate acreage cannot be located, the SDEIS should clearly acknowledge that these wetland functions are permanently being lost within the affected watershed. The SDEIS needs to contain a discussion of what cumulative wetlands impacts this has within northeastern Minnesota and at a state level. The SDEIS should include financial assurance for indirect wetlands mitigation and should describe this fully. C. Inaccurate Wetlands Characterization and Groundwater Connection The approach the SDEIS takes to classifying wetlands and their connection to groundwater has potentially under-represented indirect wetlands impacts from the proposed mine. The SDEIS relies on surface observations of plant communities to classify bog wetlands as ombrotrophic (spring fed, but with minimal or no groundwater connection) or minerotrophic (stream fed with connection to groundwater). SDEIS states: The Mine Site is covered by extensive wetlands, many of which have only minimal hydraulic connection to the underlying groundwater (SDEIS 4-46). Over 90 percent of the Mine Site wetlands are characterized in the SDEIS as ombrotrophic. The SDEIS uses no hydrologic data to support that the wetlands are perched and disconnected from interaction with groundwater. This is important because if a connection exists, the potential

55

for mine dewatering to create greater wetland impacts exists. No data were collected at the site to show that all the ombrotrophic wetlands on site are perched and would remain perched under mine induced drawdown conditions. Dr. Glasers technical report (attached) outlines his concerns about the mischaracterization of the wetlands. Ombrotrophic bogs can only be distinguished by the absence of fen-indicator speciesThe appearance of fen indicator species in the coniferous bog type at the NorthMet site clearly indicates that these peatlands are minerotrophic fens and not ombrotrophic bogs. This interpretation is further supported by the pH values recorded for surface water in coniferous bogs at the NorthMet site (Glaser 2014). Dr. Glaser finds the methods used to characterize the wetlands in the SDEIS to be not consistent with the scientific literature (Glaser 2014). In his report, Dr. Glaser also notes that even if ombrotrophic bogs exist at the mine site, they may still be hydraulically connected to groundwater flow and be sensitive to mine developmentinduced changes. No hydraulic testing was conducted at all for the Tailings Basin, but the SDEIS still makes an unsupported assumption that little conductivity exists: Hydraulic testing in the bedrock has not been performed in the Tailings Basin area, but the bedrock is believed to have a significantly lower hydraulic conductivity than the overlying drift...This is supported by analogy to the bedrock of the Mine Site (Duluth Complex), which, based on hydraulic testing, has been shown to have a significantly lower hydraulic conductivity than the overlying till" (SDEIS 4-95). Dr. Myers also finds the SDEIS does not support its claim of a lack of connectivity. The SDEIS assumes there is little connection throughout, but is incorrect (Myers 2014). Like Glaser, he recommends the installation of piezometers to better asses this relationship. GLIFWC highlights the lack of hydrologic data as a problem in their comments. Regarding the use of plant communities to classify bog wetlands, GLIFWC notes that they agree that this is useful information but we maintain that it is not a substitute for detailed understanding of the relationship of the water table and wetlands at the site (SDEIS, Appendix C, GLIFWC Comments, Wetlands Attachment). GLIFWC points out that data at the mine site indicate large daily fluctuations of vertical water movement within the aquifer, supporting the idea that a hydrologic connection exists between the wetlands and the surficial aquifer. The model used to assess wetland impacts assumed that no connection exists, an assumption based solely on plant lists and surface observations. As GLIFWC notes, We believe this assumption is not supportable (SDEIS, Appendix C, GLIFWC Comments, Wetlands Attachment). If the assumption is incorrect, indirect impacts to wetlands could be even greater than the 7,300 acres described in the SDEIS. The SDEIS needs to classify the wetlands based on an understanding of the relationship of the water table and wetlands at the site. This has not been done. Recommendation: The SDEIS should use hydrologic data to characterize the wetlands and identify groundwater connections with the wetlands. The SDEIS should address all the issues identified in Dr. Glasers report and use the scientifically supported methods he describes for wetland

56

characterization. Hydraulic testing needs to be conducted at the Tailings Basin and this information shared in the SDEIS. D. Under-Estimated Pit Dewatering Impacts The creation of mine pits several hundred feet deep can result in water being pulled into the cone of depression from the surrounding landscape, and lead to significant impacts to surrounding wetlands. In the analysis of impacts, the SDEIS relies on an analog method that uses information from other sites to make best estimations of impacts at the NorthMet site. Analog information can be useful, but it should not be the sole source of assessing impacts. GLIFWC notes, We also reiterate that the lead agencies reliance on analogs as the only source of information to gauge impacts from pit dewatering is not a rigorous approach to impact estimation (SDEIS, Appendix C, GLIFWC Comments, Wetlands Attachment). Dr. Glaser notes in his report that the evaluation of this hydraulic connectivity between the wetlands and groundwater should have included water level measurement tests in piezometer nests installed above and below the hydraulic confining layers (Glaser 2014). He also notes It should also be kept in mind that the magnitude of the pumping tests performed for the EIS is probably orders of magnitude lower than that of actual mine dewatering operations (Glaser 2014). Pump tests may have been insufficient to show the impacts that might occur to the wetlands from mine dewatering activities. In addition, the proposed NorthMet pits would be more than twice the depth of a typical pit on the Mesabi Range and twice the depth of the Canisteo Pit (SDEIS, Appendix C, GLIFWC Comments, Wetlands Attachment). Comparisons of impacts with mine pits that are significantly more shallow risks under-estimating potential impacts at the proposed NorthMet mine. Recommendation: The SDEIS should use hydrologic data to characterize the wetlands and identify groundwater connections with the wetlands. The SDEIS should address all of the issues identified in Dr. Glasers report, and use water level measurement tests to assess the connectivity with groundwater. Analog data from other mines should be used as supporting information but not the sole source of information.

E. Failure to Assess Indirect Wetland Impacts The SDEIS does not give a detailed analysis of potential indirect wetland impacts. GLIFWC, however, in their comments, has provided detailed information assessing the likelihood of impacts in zones around the mine pits. Ranging in distance from 0 feet to 10,000 feet from the pit edge, GLIFWC analyzes what expected drawdowns of water levels might be. Some of this potential drawdown could impact river baseflows leading to reductions in baseflow farther downstream and in riparian wetlands. The SDEIS should contain a much fuller evaluation of drawdowns and their impacts.

57

Dr. Glasers report provides many recommendations for how wetland monitoring should be done to assess indirect wetland impacts during operations should the project proceed. These recommendations should be incorporated and described in the EIS. Recommendation: The SDEIS needs to contain a more complete evaluation of indirect wetland impacts, drawdowns, and impacts to river baseflows and riparian wetlands. The recommendations made in Dr. Glasers report for monitoring wetland impacts from the mine operations should be incorporated in the SDEIS. F. Fugitive Dust and Wetland Impacts Dr. Glasers report indicates that the potential for fugitive mineral dust to impact nearby wetlands has not been fully assessed in the SDEIS. Dust deposition could have potential impact on both ombrotrophic bogs and minerotrophic fens depending on the chemical and mineralogical composition of the dustAlso, overland transport of solutes derived from dust is possible during the flush of spring snowmelt (Glaser 2014). Recommendation: The SDEIS should more thoroughly examine the potential impacts from fugitive dust on wetlands. The SDEIS should include an evaluation of the transport of solutes during the spring snowmelt flush. G. Loss of Peatlands and Climate Change Impacts The loss of thousands of acres of peatlands at the mine site from direct and indirect impacts will result in the release of a significant amount of carbon that is currently sequestered. We share more about our concerns in the Climate Change and Air Issues, Section VIII.

VII. Ecological, Wildlife and Vegetation Impacts The NorthMet Mine Site would be located in an area of high biological significance in the state. Its development would critically impact this area. Acknowledging that the Minnesota Biological Survey (MBS) has identified the site as important, the SDEIS notes that, "The entire 3014.5-acre Mine Site has been characterized by the MBS as various Sites of High Biodiversity Significance due to the presence of the One Hundred Mile Swamp site, which covers 15 percent of the Mine Site, and the Upper Partridge River site, which is 85 percent of the Mine Site" (SDEIS 4-173). If the NorthMet mine is developed: 1,741.1 acres of wetlands of High Biodiversity Significance would be directly destroyed (SDEIS ES-38)

58

698.2 acres of imperiled or vulnerable native plant communities would be destroyed (SDEIS ES-38) Over 4,000 acres of wildlife habitat would be decreased (SDEIS ES-50) 11 plant species on Minnesotas list of Endangered, Threatened and Species of Special Concern would be impacted (SDEIS ES-38) Critical habitat for Canada lynx would be reduced and habitat fragmentation increased (SDEIS 5-365) Wolf habitat would be diminished by 1 to 10 percent of a single wolf pack territory (SDEIS 5-367)

A. 100 Mile Swamp Impacts As the Friends noted in our comments about wetland impacts, the mine would severely impact a area known as the 100 Mile Swamp which has been identified as an area of high biological significance. "The portion of the One Hundred Mile Swamp that is on the federal lands, including part of the Mine Site, has been identified as a Site of High Biodiversity Significance...These sites provide a good representation of most of the LTA's biological and physical attributes a the ELT level...Inclusion of the One Hundred Mile Swamp site would likely complete representation of prominent ELTs in LTA 212LE11" (SDEIS 4-174). Sites of high biological diversity contain important landscape functions, rare species, and high quality examples of rare native plant communities. The impact of the mine to this site and the loss of the functions and rare species are no small matter and not a decision the State of Minnesota should take lightly and without explanation. Recommendation: The SDEIS should examine and describe how it justifies the destruction of an area of high biological significance and the loss of rare plant species associated with this project. The State of Minnesota should provide an explanation to the public of why areas of biological significance are identified and species are placed on our threatened and rare species lists, if they are not to be protected even when on public lands. B. Canada Lynx Impacts The proposed mine would impact federally designated critical habitat for the Canada lynx, a federally listed threatened species with fewer than 200 individuals in Minnesota. Mining activities would destroy about 1,454 acres of designated critical lynx habitat. The mine would decrease local lynx populations, reduce available habitat, and increase habitat fragmentation of lynx habitat (SDEIS ES-50 and 5-365). 1. USFWS Biological Assessment and Opinion Not Included The United States Fish and Wildlife Service (USFWS) is preparing a Biological Assessment of the potential impacts to the Canada lynx from the proposed project, but its

59

results are not included in the SDEIS. The United States Fish and Wildlife Service (USFWS) will review the Biological Assessment and provide a Biological Opinion (SDEIS 1-10). A draft Biological Assessment is available for review, but this information is not included in the SDEIS and a final report is not yet available. The evaluation by the USFWS is important for the publics review of this proposed project. For example, the draft Biological Assessment finds that, Based on the analysis in this BA, the Project would affect lynx, and those effects would not be insignificant or discountable (USFWS November 2013). However, this finding does not appear in the SDEIS. In addition the draft Biological Assessment states, Thus, the Project is likely to adversely affect Canada lynxthe Project may affect but is not likely to adversely affect Canada lynx critical habitat (USFWS November 2013). Findings such as this should be in the EIS and should be explained so the public understands what they mean. 2. Restoration of Lynx Habitat Overstated The SDEIS states that while the project would decrease critical lynx habitat, restoration of the mine area after operations might provide habitat once more. But the mine area is expected to remain fenced-off and highly polluted for centuries. It is unlikely that the area would be suitable for lynx for a very long time. The SDEIS needs to provide a restoration plan to support any claims that the area could be made suitable for lynx again. I addition, the SDEIS does not consider alternatives like the West Pit Backfill and Underground Mine Alternative that may help limit impacts to lynx habitat. 3. Loss of Critical Habitat Via Destruction and Land Exchange Portions of the Mine Site lie within designated "critical habitat" for lynx. The U.S. Forest Service has designated Lynx Analysis Units (LAUs) within the Superior National Forest that comprise landscape-scale analysis areas for lynx management. The SDEIS notes that the Mines Site is located within LAU 12, and that about 96 percent of LAU 12 provides suitable lynx habitat. Mining activities would destroy about 1,454 acres of suitable lynx habitat, and the proposed land exchange would result in further loss of lynx habitat within identified LAUs. While the federal lands within LAU 12 comprise 6,500 acres, the non-federal parcels combined only equal 2,149 acres within LAUs. This represents a total loss of 4,752 acres within LAUs. Tract 1 proposed for the exchange is not located within an LAU at all (SDEIS 5-629 and 5-630). 4. Vehicular Collision Risks Not Fully Evaluated The SDEIS notes, Although there is the potential for incidental take as a result of vehicle collisions with lynx, haul traffic at the Mine Site would likely have little direct effect on lynx (SDEIS 5-366). The SDEIS further states that, this species does not rely upon roads for travel (SDEIS 5-366). These statements are in contradiction with findings in the Draft Biological Assessment (BA) by the USFWS. The Draft BA cites research that showed that lynx will take advantage of roads when present and that a road and trail network increases habitat

60

connectivity for lynxLynx may find that it is more energetically efficient to walk on or alongside of a road (USFWS November 2013). The SDEIS contradicts itself, acknowledging that lynx actually do use roads: Lynx utilize snow packed trails and roads as travel corridors (SDEIS 5-628). The SDEIS should have included the findings of the Draft BA, and should consistently acknowledge that the use of roads by lynx increases the risks of collisions with vehicles. The SDEIS should also support claims of little effect on lynx with actual estimations of the amount of expected collision deaths. Without numbers, the assertions are subjective opinions that cannot be evaluated by the public. 5. Incorrect and Conflicting Information The SDEIS states that a 2006 survey of the Mines Site and nearby area showed no lynx were identified (SDEIS 5-365). But the USFWS Draft BA disagrees, and notes, Tracks and scat of three female lynx were identified during the survey within the study area, concentrated in areas approximately 5 miles east and south of the Mine Site (USFWS November 2013). The Draft BA also describes a lynx survey conducted 12 miles northeast of the Mine Site for the Birch Lake Project and Maturi Project for Franconia Minerals that discovered several lynx using the area based on scat and tracks. This survey is not mentioned in the SDEIS. Without the inclusion of all relevant information, the SDEIS risks under-representing the number of lynx in the area and not properly assessing impacts to the species. The SDEIS also incorrectly states that the Canada lynx is not on the states Endangered, Threatened and Species of Special Concern list (ETSC). However the species is not listed as an ETSC species in Minnesota" (SDEIS 4-201). This statement is inaccurate. On August 19, 2013, the species was placed on the list of Minnesotas Species of Special Concern. The SDEIS should accurately portray the state and federal status of this species for the publics full assessment of risks to its well-being from this project. Recommendation: The SDEIS should include the USFWS findings of the Biological Assessment and the Biological Opinion for lynx impacts. The EIS is incomplete without it. The SDEIS should include a full evaluation of alternatives, such as underground mining, that would disturb less of the area and impact lynx and other wildlife habitat less. The SDEIS should include a restoration plan for lynx habitat once operations cease, with realistic evaluations about the restoration capacity for parts fenced-off and contaminated. The SDEIS should insist that a land exchange include acquiring enough acres within lynx LAUs to equal the losses of acres in LAU 12. The SDEIS should acknowledge the use of roads by lynx, use this information to fully evaluate

61

the risk of vehicular collisions, and provide estimates of the amount of collision deaths anticipated. The SDEIS should also consider and describe mitigation measures such as underpasses and fencing to reduce the collision risk to lynx. The SDEIS needs to correctly describe the Canada lynx as a Minnesota Species of Special Concern. The SDEIS should present all relevant lynx surveys and show that the best available information has been used in assessments of impacts rather than relying on incomplete or outdated information. Included in the SDEIS should be a table and map revealing the locations of lynx tracks, sightings, scat, or kills between 2000 and 2014. C. Moose Impacts The SDEIS makes almost no mention of the projects impact on moose populations and moose habitat, despite the states alarming moose population decline. The SDEIS notes that factors in the population decline include habitat fragmentation and loss, climate change, disease, and predation. And the SDEIS states that the NorthMet proposed project would eliminate 2,775.2 acres of key moose habitat (SDEIS 5-377). But the SDEIS then makes this unsupported statement: the NorthMet Project Proposed Action will affect moose individuals in the vicinity through habitat loss and fragmentation, though not likely at a population level (SDEIS 5-377). The SDEIS provides no data or survey information to support the claim that the loss of habitat would not impact the species population numbers. The SDEIS does not appear to include any analysis of the projects impacts on this species. The SDEIS fails to even note that the moose on August 19, 2013, was designated a State Species of Special Concern. The alarming declines in this species population, even before its recent designation, should have been reason for the SDEIS to include a full evaluation of the projects potential impacts on this species. Recommendation: The SDEIS should include a full evaluation of the projects impacts to moose populations and moose habitat. Statements about the likelihood of impacts must be supported by data. The SDEIS should eliminate references to moose as regionally common as the status has changed to a Species of Special Concern. D. Impacts To Wildlife Corridors The SDEIS acknowledges that the NorthMet project will impact identified wildlife travel corridors. For Wildlife Corridor 17, the SDEIS states, Direct loss and fragmentation (SDEIS 658). Corridor 16 is projected to experience restricted use by wildlife due to noise and activities (SDEIS 6-58). The SDEIS cites a study by Emmons and Oliver Resources Inc. (2006) that identifies 13 major wildlife travel corridors connecting large roadless blocks along the Iron Range. In the 2009 DEIS for the NorthMet project, it was acknowledged that the Emmons and Oliver study considered the loss of any one of these wildlife corridors to be significant (DEIS 4.4-30). The loss of Corridor 17 and restricted use of Corridor 16 due to the NorthMet project should therefore be acknowledged as a significant loss in the SDEIS. 62

In addition to the loss of these Wildlife Corridors, the SDEIS should, but does not, evaluate the project in light of impacts that may have occurred to the other 13 corridors. For example, Corridor 18 identified in the SDEIS appears to already have been lost as a suitable corridor for wildlife due to mining activities, ATV use and other landscape changes. The relocation of Highway 53 may also impact wildlife corridors. The impacts to wildlife corridors from the NorthMet project must be analyzed with these other lost corridors in mind. Recommendation: A Biological Assessment and consultation with the USFWS must be conducted to assess impacts to wildlife travel corridors. The SDEIS must analyze the loss of Corridors 16 and 17 with the loss of other regional corridors as part of its assessment. E. Wildlife Using The Project Area Contaminated mine pits and the Tailings Basin present a risk to wildlife, especially waterfowl, that may attempt to use the area. Other risks to wildlife come from exposure to mine contaminated waters off-site. The SDEIS notes, Wildlife, specifically aquatic birds, may utilize open water habitat created by the NorthMet Project Proposed ActionSome wildlife species, specifically those that feed on aquatic prey, may be susceptible to mercury exposure directly from open water sources such as the pit lake and Tailings Basin pond, and indirectly at the Partridge River and Embarrass RiverSpecific species such as loons, osprey, mink, and otter may be affected (SDEIS 5-373-374). But the SDEIS does not describe a plan to discourage wildlife use of contaminated areas. The SDEIS also fails to adequately address the mercury contamination of nearby waters, which the Friends describes more in the Mercury section. But the SDEIS concludes that the proposed project will not adversely impact wildlife. Such conclusions need to be supported with more information than has been provided in the SDEIS. Recommendation: The SDEIS should include a plan to discourage the use of mine pits and the Tailings Basin by wildlife. The SDEIS should more adequately assess and mitigate potential mercury pollution issues (see our Mercury comments and recommendations). F. Impacts To Rare Plants and Communities The proposed NorthMet project would directly impact 9 of the states 11 listed ETSC plant species, and indirectly impact 2 of the 11 state-listed ETSC plant species (SDEIS 5-348). For the 9 directly impacted, Most of the direct effects would involve the complete loss of colonies within a population (SDEIS 5-348). Particularly at risk is the plant species Floating marsh marigold (Caltha natans). Very few populations are known in Minnesota. Habitat loss is largely the reason behind recent local extirpations of this species in Minnesota" (SDEIS 4-181). The NorthMet project would directly

63

impact, destroy, about 8 percent of its Minnesota population (SDEIS 5-350 and 5-346). This constitutes a risk to the entire population of this plant in Minnesota. Neat spike-rush (Eleocharis nitida) is another rare plant species threatened by the project. Despite the long collection record for this species in Minnesota, relatively few populations have been documented and little is known about the overall distribution of the species throughout the state" (SDEIS 4-181). The project puts entire populations of some of these rare plants at risk. In addition, the project would result in the degradation of 698 acres of black spruce-Jack pine woodlands, systems which are considered imperiled-vulnerable by the Minnesota Biological Survey. The impacts would represent a 20 percent loss of this forest type within the Laurentian Uplands. Recommendation: The SDEIS must examine the potential risk to entire populations of state-listed rare plant species, and assess if this complies with Minnesotas endangered species law. The impacts to vulnerable forest communities must be more thoroughly described and their loss evaluated at a state-wide level. The SDEIS should examine how this loss would be mitigated. G. Impacts To Wild Rice The potential for the NorthMet project to release high concentrations of sulfate puts wild rice in jeopardy. Impacts to wild rice constitute not only a threat to this plant, but also a threat to wildlife dependent upon the plant, as well as a threat to cultural values for the plant (more in our comments about Impacts to Human Communities). Table 5.2.2.30 (SDEIS 5-129) shows that mine site surface waters are expected to exceed the wild rice standard of 10 mg/L for sulfate for the 200 year simulation period. Data in the Water Data Packages for both the Mine and Plant Sites, show concentrations in mine and Tailings Basin drainage water that exceed the wild rice standard for centuries. And the SDEIS acknowledges that 16 million gallons of contaminated water a year will escape capture, a rate we have identified elsewhere in these comments that is likely low. See more about our concerns about sulfate that could escape capture in our comments in the Section I. Failure to Meet Clean Water Principle #1: Minnesotas Waters Will Not Stay Safe and Clean. The SDEIS fails to acknowledge the risk to wild rice from sulfate drainage: Effects on wild rice as a result of the NorthMet Project Proposed action are expected to be minimal (SDEIS 5-341). But the wild rice sulfate standard is already being exceeded from the seepage at the current LTV Tailings Basin and other mining operations in the area. Sulfate concentrations in Trimble Creek, the Embarrass River, and the Partridge River currently exceed the wild rice standard of 10 mg/L (SDEIS, Appendix C, Tribal Cooperating Agencies Cumulative Effects Analysis, September 2013). The NorthMet projects sulfate drainage will be adding additional loads to a

64

watershed already impacting wild rice. This must be acknowledged and more thoroughly analyzed in the SDEIS. The SDEIS is inconsistent about the intent to apply the wild rice sulfate standard seasonally. At one point the SDEIS notes, PolyMet is not seeking application of the seasonal component of this standard for the NorthMet Proposed Action as currently proposed and evaluated in this SDEIS (SDEIS 5-25). But elsewhere, the SDEIS seems to say it will apply the standard seasonally: The West Pit overflow non-mechanical treatment system would be designed to discharge only during September and October in order to comply with the seasonal sulfate discharge criterion for wild rice downstream of the Mine Site (SDEIS 5-216). And in another section the SDEIS states: The Partridge River will, at certain times of the year, exceed the 10mg/L sulfate standard, mostly during winter low-flow conditions (SDEIS 5-342). GLIFWC scientists disagree with the concept of a seasonal application of the standard, citing a lack of evidence or scientific basis for doing so. Research has not shown that the wild rice seed avoids impact from high sulfate levels while it is dormant. The SDEIS should be clear and consistent that the project would not apply a sulfate standard seasonally. The Tribal Cooperating Agencies have also identified concerns about which waters have been designated as wild rice waters for application of the wild rice sulfate standard. Historic information about the presence of wild rice must included in all analyses of the project. The SDEIS must include information from elders about where wild rice has grown. In addition, the SDEIS should analyze the impact from current sulfate conditions on wild rice waters, and not eliminate from designation waters where wild rice has been diminished due to human-caused sulfate levels. Recommendation: The SDEIS must contain language that clearly indicates Minnesotas sulfate standard of 10 mg/L is scientifically defensible and will be applied for the NorthMet project. The SDEIS should insist that the standard is applied all year and not seasonally. Designation of wild rice waters applicable to the NorthMet project should include historic information from elders and should not exclude waters where wild rice once grew but due to human-caused sulfate levels no longer exist. Wild rice at low densities also provides important food to wildlife, and so low-density wild rice waters should also be included for application of the standard for this project. H. Use of Non-native Plants The NorthMet project calls for using non-native plant species for reclamation and erosion control. For example, non-native plants are proposed to be used on the Category 1 Stockpile to control dust and erosion, and used to re-vegetate disturbed areas during mine operations (SDEIS 5-343 and 5-344). Some of the species proposed are considered invasive species, which can have negative impacts on native plant communities. The use of non-native invasive species to reclaim the mine site is unnecessary and irresponsible with respect to the wide-scale problem of non-native species in Minnesota, and the efforts to

65

contain their spread by federal and state agencies. Recommendation: The SDEIS should include a requirement that non-native plants will not be used in reclamation activities or for temporarily stabilizing disturbed areas. Only the use of native, non-invasive species should be permitted.

VIII. Climate Change and Air Quality Issues A. Climate Change Impacts 1. Peatland Destruction Peatlands have been identified as crucial ecosystems in storing carbon that would otherwise contribute to climate change. The destruction of peatlands can release large quantities of previously sequestered CO2 into the atmosphere. The NorthMet projects direct peatland impacts would be nearly 1,000 acres and likely much higher due to indirect impacts (see our comments on Wetlands Impacts, Section VI.). Scientists have calculated that the loss of 1,000 acres of Minnesota peatlands translates to a release of approximately 2.7 million metric tons of CO2 to the atmosphere. This is an increase in Minnesotas total annual emissions of CO2 by approximately two percent (above 2005 levels) (Anderson et al., 2008). PolyMets impacts on Minnesotas carbon emissions are likely to be this level or higher. In 2007, the Minnesota State Legislature requested that the University of Minnesota produce an assessment of the potential capacity for carbon sequestration in Minnesotas terrestrial ecosystems. The Minnesota Terrestrial Carbon Sequestration Project, an interdisciplinary research group, was organized to produce that assessment. The team analyzed existing scientific literature, land existing in broad land use categories, and the role of current state policies and programs on carbon sequestration potentials. In February 2008, the Project produced a report titled, The Potential for Terrestrial Carbon Sequestration in Minnesota. Some of the key findings and recommendations of that team of researchers are: - Peatlands in Minnesota contain the largest carbon stocks in the state, in excess of 4 billion metric tons - Release of this carbon to the atmosphere as CO2 can result from peatland drainage and conversion - Release of this carbon to the atmosphere would accelerate global warming and require greater reductions in CO2 emissions elsewhere - Destruction of 1,000 acres of peatland in Minnesota from mining or other activities would increase the states total CO2 emissions by 2% over 2005 levels

66

The top recommendation of this research group: Preserve the existing large carbon stocks in peatlands and forests by identifying and protecting peatlands and forests vulnerable to conversion, fire, and other preventable threats (Anderson et. al 2008). In December 2006, Governor Tim Pawlenty announced the states Next Generation Energy Initiative, including the development of a comprehensive plan to reduce Minnesotas emissions of greenhouse gases. The Minnesota Climate Change Advisory Group, a broad-based group of Minnesota citizens and leaders, was created to develop state-level policy recommendations to the Governor. In April 2008, the Advisory Group released its report titled, Minnesota Climate Change Advisory Group Final Report: A Report to the Minnesota Legislature. Some of its key findings and recommendations include: The policy goals from the Advisory Group included: - Protect and restore northern peatlands. - By 2015, identify peatlands at risk of releasing greenhouse gases because of !lowered water table or industrial uses such as mining. - Design policies to protect peatlands and wetlands from drainage and other carbon- !releasing land uses. ! The destruction of the peatlands at the NorthMet mine site runs counter to the recommendations of both of these government-initiated studies. ! But the SDEIS fails to discuss the carbon emissions from the destruction of hundreds of year old peatlands. It states it will mitigate greenhouse gas emissions associated with a change in land cover by replacing or protecting other wetlands in three locations, but that these would not be an equivalent acreage of in-kind wetlands. it is impracticable to replace all affected wetland types with an equivalent area of in-kind wetlands (SDEIS 5433). Instead of a thorough analysis and discussion about the carbon emissions from the peatland destruction, the SDEIS provides only short statements that fail to connect to peatland destruction: the construction and operation of the NorthMet Proposed Action would emit gases known to contribute to global climate change (SDEIS 6-86). Information about how much carbon is expected to be released from wetlands destruction is placed in tables where it is hard to find and lacks discussion. Table 5.2.7-9 (SDEIS 5406) indicates that 199,963 metric tons of greenhouse gas emissions will result from wetland carbon loss, and 20 years of emissions from stockpiled peat and emission from peat used in reclamation. Minnesotas Climate Change Advisory Group concluded that: Wetlands have among the highest potential carbon-sequestration capacities for any type of land cover in Minnesota. Peatlands are likely Minnesotas largest single carbon sink, containing 37% of all carbon stored in the state.. (Minnesota Climate Change Advisory Group 2008). The groups final recommendation: Protecting these enormous carbon reservoirs (peatlands)...is critical (Minnesota Climate Change Advisory Group 2008).

67

The SDEIS also fails to appropriately plan for mitigating the loss of these wetlands and their carbon sequestering abilities. Wetland mitigation plan would be implemented to offset increased carbon dioxide emissions to extent possible (SDEIS ES 49-ES-50). But wetland mitigation will not offset the emission of carbon from the peat rich wetlands of the 100 Mile Swamp. Recommendation: The recommendations of both the Climate Change Advisory Group and the Terrestrial Carbon Sequestration Initiative should be heeded, and no significant losses to the states peatlands should be permitted. The SDEIS should clearly identify this project as significant contributor to Minnesotas statewide carbon emissions. The SDEIS needs to include a thorough narrative discussion of the amount of carbon expected to be released from the loss of peatlands and the statewide consequences. The SDEIS should also acknowledge that wetland mitigation will not replace the carbon sequestering abilities lost with the destruction of the 100 Mile Swamp peatlands. 2. Generation of Electricity The SDEIS discussion about emissions of greenhouse gases under-represents the emissions expectations by failing to include emissions from the additional electricity generation needed for the project. The amount of electricity the project will demand is not insignificant. Barr Engineering estimates that the NorthMet project will require 59.5 megawatts of power, which will be supplied from Minnesota Power (Barr 2012). The SDEIS in Table 5.2.7-8 provides emission numbers for both direct emissions (those from activities at project site) and indirect emissions (electricity power plant emissions due to the projects needs), but the discussion within the SDEIS focuses on the direct when it concludes that assessing greenhouse gas (GHG) contributions from the project cannot be calculated. As notedthe NorthMet Proposed action would directly produce approximately 196,342 mtpy of GHGGiven the minor GHG contribution of the NorthMet Propsed Action to global GHG emissions, it is impossible to predict how much the NorthMet Proposed Action would factor into climate change (SDEIS 6-87). Not included in the narrative discussion of greenhouse gas emissions is the indirect emission estimation from the generation of electricity. Table 5.2.7-8 indicates this additional emission is expected to be 511,000 metric tons per year (mtpy) (SDEIS 5-405). This brings the total emission expectations from the NorthMet project to 707,342 mtpy (SDEIS 5-405). Table 5.2.7-9 indicates that the total lifetime Greenhouse Gas emissions from the NorthMet project is expected at 15,790,752 metric tons (SDEIS 5-406). But the SDEIS analysis and discussion do not include these extremely large total emission numbers when exploring impacts. Instead, the SDEIS makes statements that seem to minimize the emission threats: total GHG emissions are less than 100,000 tpy of CO2e and would not exceed the USEPA major source thresholds for GHGs (SDEIS 5-404). When examining cumulative impacts, the SDEIS makes this statement, focused on direct impacts, but ignoring the extremely large and alarming indirect impacts from

68

electricity generation: the NorthMet Proposed Actions direct GHG emissions would be several orders of magnitude lower than total global, national, and even statewide GHG emissions (SDEIS 6-87). It is irresponsible for the SDEIS to fail to incorporate an analysis and discussion of the climate change impacts from greenhouse gas emissions that include the contribution from power generation. It is not accurate to say it is impossible to predict how much the NorthMet Proposed Action would factor into climate change when the lifetime greenhouse gas emissions are expected to be over 15 million metric tons. This is a significant shortcoming of the SDEIS analysis and must be corrected. In addition, the lifetime greenhouse gas emissions for the project have been significantly underestimated. Table 5.2.7-9 indicates that the expected reclamation emissions are based on a 20 year closure period and a 30 year long-term closure period for the WWTF and WWTP (SDEIS 5-406). A 30 year long-term closure period is no where near as long as the two treatment facilities will need to operate. Water quality modeling for the project indicates that the need to treat contaminated water will be at least 500 years at the Plant Site and at least 200 years at the Mine Site, not 30 years. Operating these plants for centuries will continue to require large amounts of energy that will contribute to greenhouse gas emissions. Lifetime greenhouse gas emission estimates must include centuries of operating the treatment plants. The amount of greenhouse gas emissions for this project have been significantly underestimated. Recommendation: The SDEIS must include analysis of the projects expected greenhouse gas emissions both on a yearly and life-time basis and this analysis must include indirect contributions from the generation of electricity. It is not adequate to say that 15 metric tons or more of greenhouse gas emissions cannot be analyzed for climate change impacts. A full written narrative of the climate change impacts from emissions that include power generation must be included in the SDEIS. Estimates of emissions must be based on the need to generate power for centuries to operate the two water treatment facilities. The climate change impacts of this proposed project should be described in terms of what it contributes to state, regional and national carbon emissions. B. Air Quality and Visibility Issues 1. Regional Haze The NorthMet proposed project is located within 300 km of four Class I regions. Under the Clean Air Act, Class I airsheds were established as areas where emissions of particulate matter and sulfur dioxide are to be restricted. Class I areas include federal wilderness areas exceeding 500 acres and national parks. These are places that are allowed only the smallest incremental pollution levels above baseline conditions. The four Class I areas within the vicinity of this project are the Boundary Waters Canoe Area Wilderness, Rainbow Lakes Wilderness, Voyageurs National Park, and Isle Royale National Park.

69

But, the SDEIS inaccurately and without supporting data predicts that project-related emissions will not impair visibility within Class I areas such as the Boundary Waters Canoe Area Wilderness (BWCAW) and Voyageurs National Park. Additionally, the NorthMet Project Proposed Action would not adversely affect visibility in nearby Class I areas, such as the BWCAW and Voyageurs National Park (SDEIS ES-41). The SDEIS cites higher expectations of visibility-reducing emissions such as SO2 and NOx than was anticipated four years ago in the NorthMet Draft EIS (SDEIS 6-83; DEIS 4.6-53 and 54). The analysis in the DEIS concluded that these emission levels, lower than predicted in the SDEIS, would indeed contribute to visibility impairment for as much as 23 days a year in the Boundary Waters Canoe Area Wilderness. The DEIS concluded that the NorthMet project may result in a 5% visibility impact in the BWCAW for 23 days a year, and as much as a 10% impact for one day a year (DEIS 4.6-37). The SDEIS does not explain why higher levels are now not expected to contribute to visibility impairment in Class I areas. The visibility impacts acknowledged in the DEIS are not insignificant impairments from a single project. The Minnesota Regional Haze State Implementation Plan (SIP) is a plan developed by the Minnesota Pollution Control Agency as required by the Regional Haze Rule of 1999. The goal of the SIP is to reduce haze in Class I areas affected by Minnesota emissions, and to reach natural visibility conditions by 2064. The SIP was submitted to the U.S. EPA for review on December 30, 2009. The Haze SIP notes that a 5% or more contribution to visibility impairment from an entire state would be considered significant. The NorthMet project, as described in the DEIS, would contribute a 5% visibility impairment for as many as 23 days, and a 10% impairment for one day per year. Emissions contributions of this level are keeping Minnesota from making progress toward the goals outlined in the SIP. Even the SDEIS concedes that the state is not on target to reach the Regional Haze SIP visibility goals (SDEIS 6-83). The SDEIS indicates that data between 2005 and 2010 show BWCAW visibility is trending toward improvement, but it does not say how much improvement this represents. The SDEIS also notes that visibility improvement in the Class I area of Voyageurs National Park has not experienced improvement (SDEIS 6-85). In addition, the SDEIS acknowledges that Minnesota emission sources are the single largest contributor to regional haze in Minnesotas Class I areas, and that these sources are expected to contribute 30 percent of the visibility impairment at Isle Royal National Park (SDEIS 686). This information does not seem to support the SDEIS assertion that the project would not adversely impact visibility in the Class I areas. In contrast with the DEIS, the SDEIS does not describe potential mitigation measures to address visibility impairments to Class I areas. The SDEIS does not mention mitigation options such as the installation of low-NOx burners or the use of waste heat for work space heating requirements. Instead, the SDEIS relies on an unsupported belief that Class I areas are already experiencing visibility improvements, and that voluntary reductions of power generation facilities, and the foreseeable regulatory actionswould

70

likely reduce emissions of SO2 and NOx in Minnesota (SDEIS 6-85 and 86). But in the very next sentence, the SDEIS notes that the reductions would not be sufficient to meet the Regional Haze goals. 2. Cumulative Impacts The SDEIS also fails to adequately evaluate the cumulative effects on air quality of the NorthMet project when combined with foreseeable projects. Not included in the analysis are the impacts from on-going and future hardrock mineral exploration in and near the Superior National Forest. The nonferrous mine project by Twin Metals, one that is in an advanced stage of development, was not evaluated for its impacts to air quality and visibility. A prefeasibility study for this project is expected mid-year 2014. 3. Anti-Idle Program for Vehicles The SDEIS describes a voluntary anti-idle program for the operation mine vehicles, trucks, equipment and rail transportation. The SDEIS should insist on a commitment from the company to an anti-idle approach to maximize the reduction of NOx and SO2 emissions. 4. Mercury In the Chambers and Levit report (see attached), the authors note that neither the SDEIS nor the supporting documents provide full disclosure about the total amount of mercury expected in the process stream (bulk tailings, hydrometallurgical tailings, or autoclave scrubber waste) and where the scrubber waste would be disposed (Chambers and Levit 2014). The SDEIS describes using first stage and second stage scrubbers to remove mercury, but third stage controls were determined to be uneconomical. Annual emissions from the autoclaves after first and second stage controls are anticipated to be 4.1 pounds of mercury. But the SDEIS fails to describe the total amount of mercury expected in the process stream. And, while the scrubber mercury waste will be placed in the hydrometallurgical tailings facility, this is not mentioned in the SDEIS, only in the background documents which most of the public will not review. The potential for mercury in the hydrometallurgical waste to be a source of mercury contamination must be discussed. Autoclaves can be a significant source of airborne mercury. But as Chambers and Levit note, U.S. EPA standards only call for monitoring mercury emissions from autoclaves once a year. Once a year measurements will not provide enough data to ensure statistically reliable measurements of the efficiency of the mercury capture systems (Chambers and Levit 2014). They recommend more frequent monitoring. Recommendation: The SDEIS gives contradictory information, citing high levels of visibility impairing emissions and the expected failure of Minnesota to meet Regional Haze reducing goals, while at the same time asserting the project will have no impacts on Class I area visibility. The SDEIS must be corrected and should provide a full explanation for any assertion that Class I areas will not be impacted. Analysis of visibility impacts must include the emissions from power plants supplying energy to the project. The SDEIS should explain why conclusions differ from those in the 2009

71

DEIS. The SDEIS needs to explain how visibility improvements will occur in light of the state not on target to meet Regional Haze goals. Cumulative effects analysis must include all reasonably foreseeable projects, including the Twin Metals project and on-going and expected exploration. The SDEIS needs to provide mitigation measures and show how they achieve the goal of reducing this projects air quality and visibility impacts. The SDEIS should include a firm commitment from the mining company to and anti-idle program for its own vehicles and equipment as well as for outside delivery vehicles. The SDEIS should describe the total amount of mercury in the process stream and that the autoclave scrubber waste will be disposed in the Hydrometallurgical Residue Facility. The potential for this mercury to be a pollution risk must be examined. Autoclave mercury emissions should be monitored more frequently than once a year. The SDEIS should contain a monitoring plan that allows for statistically reliable data on the autoclave mercury emissions.

IX. Impacts to Human Communities, Cultures and Health The SDEIS analysis of the range of potential impacts to human communities, economies, cultures and health is incomplete. The analysis largely focuses on employment benefits, with no attention to negative socioeconomic impacts that can occur with mining and with changes to communities. When the SDEIS does identify potential negative impacts, it fails to analyze or mitigate the social, economic and cultural impacts. The SDEIS claims that the approach of this SDEIS is to evaluate environmental and social effects directly, in the appropriate resourcespecific section (SDEIS 5-497), but these sections do not contain analysis of the economic impacts. The result is that the economic impacts to these resources are never assessed in the SDEIS. The SDEIS must contain a full evaluation of the socioeconomic impacts from this project. A. Impacts to Subsistence Users and the Safety of Fish Consumption The SDEIS acknowledges that Operation of the NorthMet Project Proposed Action would make the Mine Site unavailable for subsistence use (SDEIS 5-509). But it does describe what economic or social impact this would have on subsistence users. The SDEIS notes that Operations could affect individuals who consume fish harvested from nearby waterbodies. The NorthMet Project Proposed Action would increase mercury concentrations in the Embarrass River Watershed, as well as some nearby lakes... (SDEIS 5509). But the SDEIS fails to analyze what the economic, social or human health impacts of this increase of mercury in the Embarrass River and nearby lakes. This impact needs to be calculated, described and mitigated. The SDEIS also concedes that the Partridge River may also be a source of mercury contamination for wildlife: Some wildlife species, specifically those that feed on aquatic prey, may be susceptible to mercury exposure directly from open water sources such as the pit lake and Tailings Basin pond, and indirectly at the Partridge River and Embarrass River (SDEIS 5-373-374). If the Partridge River is a source of mercury

72

contamination for wildlife, it is also a potential source of contamination for humans. This needs to be analyzed, described and mitigated. The SDEIS concedes that mercury contamination if fish may present an environmental justice issue (EJ) to certain human communities, especially Band members. As a result, increased mercury concentrations, and associated increases in mercury bioaccumulation in fish tissue could therefore constitute an EJ impact for Band members and other subsistence consumers of fish (SDEIS 5-509). But the SDEIS does not assess how big an impact this might be, and provides no plan to address the impact or mitigate it. The SDEIS cannot simply identify a human health risk and then fail to provide a way to mitigate it. Regarding how long mercury bioaccumulation might remain a problem for consumers of fish, the SDEIS says that: Deposition of mercury from the NorthMet Project Proposed Action would cease at closure, but mercury bioaccumulation in fish tissue an exiting fish consumption limits could persist beyond the mines operational life (SDEIS 5-510). This assertion had numerous flaws. First, PolyMet data show mine and plant site sulfate discharge levels far above water quality standards hundreds of years after the mine closes: "It is expected that eventually the sulfate concentration in the pits would stabilize to more or less steady-state values, although the timeframe for this would likely be greater than 200 years as indicted by Figures 5.2.37 to 5.2.239, which show solute concentrations continuing to decrease at year 200, although still above water quality standards" (SDEIS 5-155). Second, the SDEIS notes that about 16 million gallons a year of discharge will escape capture. The high sulfate discharge will continue to seep into the surrounding waters. This seepage has the potential to methylate mercury and continue acting as a source for mercury contamination in fish. So the assertion that deposition of mercury ceases at closure is not supported. Finally, the SDEIS does not examine the economic, social or human health impacts of this continued bioaccumulation in fish tissue long into the future. How long might this persist? What are the impacts to the economies, communities, and health of the people consuming this fish? This is a major omission in the SDEIS. Recommendation: The SDEIS needs to include a more robust analysis of social, economic, cultural and health impacts from the proposed project. This section of the EIS is extremely incomplete. The SDEIS needs to fully analyze the impacts to subsistence users, recreationists, outdoor-related businesses, band members, and children from expected mercury contamination in fish. Impacts should be fully described and mitigation plans provided. The potential for bioaccumulation in fish to continue after the mine closes should be fully described and mitigation plans provided. B. Potential Colby Lake Impacts The NorthMet project could result in various impacts to Colby Lake water quality and water supply. Colby Lake is a drinking water supply source for the City of Hoyt Lakes. "...any shortfall in water requirements would be made up by raw water from Colby Lake using an existing pump station and pipeline...Throughout operations, the average annual makeup water drawn from Colby Lake would vary between 20 and 810 gallons per minute..." (SDEIS 3-102). The SDEIS does not describe what kind of impact this might have on the lakes water supply.

73

The SDEIS reveals that as a result of the NorthMet project, increases in several contaminants would increase for Colby Lake. Arsenic, barium, cadmium, cobalt, nickel increase in levels over current conditions as a result of the project (SDEIS 5-145; Table 5.2.2-34). And aluminum, iron and manganese would exceed evaluation criteria (SDEIS 5-145). Arsenic water quality evaluation criterion for Colby Lake is 2 micrograms/L because the lake functions as a drinking water supply for the City of Hoyt Lakes. The SDEIS identifies the potential for arsenic concentrations at the nearest evaluation location to Colby Lake (SW-006) to exceed this level. "These 'elevated' concentrations at SW-006...could raise concern for the potential of exceedances of the arsenic evaluation criterion in Colby Lake" (SDEIS 5-151). But the SDEIS downplays this possibility, stating that this would likely occur during late winter, when the Partridge River flows are low, and that these elevated levels would get "blended with highflow/lower concentrations of arsenic such that arsenic concentrations in Colby Lake are not predicted to exceed the 2 micrograms/L evaluation criteria at the 90th percentile probability" (SDEIS 5-152). The SDEIS seems to be describing hopeful thinking with these statements, rather than information-guided projections. The SDEIS should provide evidence that arsenic dilution would take place. A mitigation plan should be in place and described should levels exceeded the standard in the lake. The SDEIS describes a plan to discharge treated water from the WWTP into four creeks near the Tailings Basin to compensate for any reduction of flow from the creation of containment systems at the Tailings Basin. The four creeks are Unnamed Creek, Second Creek, Trimble Creek, and Mud Lake Creek. If treated water from the WWTP is still not enough to compensate, the SDEIS calls for water from Colby Lake to be pumped in to add to the flow (SDEIS 3-123). The SDEIS needs to describe if Colby Lake can sustain this level of pumping, especially when combined with any other demands that the mine project may be making of the lake. Given that aluminum, iron and manganese already exceed evaluation criteria for Colby Lake, would discharging lake water into these creeks present a contamination risk? (SDEIS 5-145). The SDEIS should evaluate this and present mitigation plans. Recommendation: The SDEIS needs to fully analyze and describe impacts to water quality and quantity for Colby Lake. Management approaches of arsenic and other contaminant levels should be based on supporting information. The cumulative impact of additional contaminants to Colby Lake need to be analyzed and described in the SDEIS. Potential impacts to the lake and the City of Hoyt Lakes should be fully assessed should water quality or quantity issues occur. Mitigation plans should be described for scenarios in which exceedances or water quantity issues occur. C. Exposure to Asbestos-like Fibers The NorthMet SDEIS does not adequately examine the risks to worker safety and public health from asbestos-like fibers found in the ore at the mine site. Crushing the ore in processing releases these fibers, which are suspected of causing mesothelioma in mine workers. Studies show miners

74

who worked in the Northshore and LTV mines who were exposed to asbestos-like fibers have higher levels of mesothelioma. The NorthMet project would produce mineral dust even smaller than what is regulated and has been studied. The SDEIS acknowledges the potential exists for the release of amphibole mineral fibers from the proposed operations, which could pose a potential public health risk of uncertain magnitude (SDEIS 5-439). The SDEIS identifies that about 9 percent of the fibers studied in the NorthMet ore were considered amphibole fibers. But the SDEIS characterizes this 9 percent as a small amount (SDEIS 5-440). The SDEIS does not provide information to support the assertion that a 9 percent amphibole fiber content should be considered small and of little concern. Additional data or studies are needed to know if this percentage represents a minimal health risk or a significant health risk. The SDEIS is also lacking in analysis of the cumulative effects of additional amphibole fiber exposure in the area. Given that rates of mesothelioma on the Iron Range are already high, the SDEIS should evaluate what the additional release of these fibers could mean for human health. Recommendation: The SDEIS should include a formal health assessment of the risks to workers at the NorthMet mine from amphibole fibers. The assessment should analyze asbestos-like fibers smaller than 5 microns in length. The SDEIS should include details of the air monitoring at the mine and plant site and in nearby communities, and describe contingency plans to address the risk to public health and worker safety if asbestos-like fibers are detected during construction, operation, closure and post-closure. D. Human Health Impact Assessment Needed The SDEIS for the NorthMet project does not include a Health Impact Assessment (HIA). And HIA is a broad assessment of a projects potential to impact the health and well-being of communities. HIA is formally defined as a combination of procedures, methods and tools that systematically judges the potential, and sometimes unintended, effects of a proposed project plan, or policy on the health of a population and the distribution of the effects within the population (Human Impact Partners 2014). The assessment includes components such as housing, transportation, employment and income, noise, air quality, access to goods and services, social networks all of which have connections to human health and well-being. And these are all community components that can experience significant changes when a large-scale project like a mine is developed. There is a trend among federal agencies to incorporate the development of an HIA within the environmental impact review process under NEPA (Alaska Dept. of Health and Social Services 2011). HIAs have been included in several NEPA documents in Alaska, and while not required by law, are considered a best practice for environmental review of mining projects in Alaska.

75

Recommendation: The SDEIS should include an HIA to review the scope of impacts possible to the human communities affected by the proposed mine. E. Inaccurate And Incomplete Information About Jobs Information about the number of jobs that may result from the NorthMet project is not well displayed or described in the SDEIS, and may provide over-estimates. 1. Number of Expected Jobs May Be Inaccurate The SDEIS states that the project would create 360 direct jobs during operations (SDEIS ES-40). But the report by Edison Investment Research Limited, which relied on data and information that PolyMet provided, suggests that a major job-creating component of the project may never happen. The Edison report indicates that Phase II, the development of the autoclave process to produce nickel precipitate, is likely to not take place. Phase II does not enhance our estimated valuation, so we do net expect Phase II to be built based on current economic assumptions (Edison Investment Research Limited, November 2013, PolyMet Mining). Cutting Phase II from the project would reduce direct jobs by 60 and construction jobs by over 400 workers. This is a significant reduction in the number of jobs projected by the mining company. The potential for Phase II to not happen needs to be described in the SDEIS, and predictions of economic impact should include a scenario where Phase II is not built. PolyMets job projections may be inaccurate due to the influence of mechanization. mining productivity in the Arrowhead region has also increased, due to mechanization and technological innovation. As a result, far fewer miners are now required per unit of extracted material than before, which therefore lessens the effects of booms and busts in mining communities. Continued technologically driven productivity increases could lead to lower employment than assumed by IMPLAN or other projections (SDEIS 5-496497). The analysis of jobs over the life of this project needs to incorporate the likelihood of foreseeable technology replacing human labor. 2. Confusing and Misleading Information About Where The Jobs Come From The way the SDEIS displays information about the number, kind and geographic scope of potential employment is not transparent or useful in understanding the employment impacts of the project. The 2009 DEIS stated that during operations, direct employment would involve about 55 percent non-local labor that would have to relocate to the area. Another 20 percent would commute to the mine from areas such as Duluth. That meant that about 75 percent were not expected to be from the immediate local area. The DEIS said 25 percent of the direct employment was expected to come from the local area (DEIS 4.10-15 to 4.10-16).

76

The SDEIS has changed these numbers and presented them differently and in a less transparent way, but without any explanation for the difference and with no obvious changes in the project that could account for the difference. The SDEIS states that 75 percent of the employment would be local residents, but combines both direct and indirect jobs in this estimate. So it is not possible to compare direct local jobs with the earlier DEIS predictions. But the SDEIS also states that 25 percent of the direct operations jobs are expected to relocate to the area (SDEIS 5-498-499). This 25 percent is a very different number than the 55 percent or the 75 percent number of relocation jobs described in the DEIS. No explanations are provided for these very different numbers. 3. No Transparency on the Kinds of Jobs The 2009 DEIS provided information about the kinds of jobs the project was expected to create (e.g. management, mine operations, mine technical, railroad operations, plant operations). But the SDEIS provides no such information. Without this information, it is not possible to evaluate the accuracy of the SDEIS projections on employment and local hiring. 4. No Information About Jobs at Key Phases The SDEIS fails to transparently describe employment expectations over the life of the mine. Needed is a table that takes information displayed in Table 3.2-2 showing key phases of the mine, that also displays the amount of employment expected with each of these phases. 5. Confusing Information about Jobs After Closure The 2009 DEIS noted that: it is assumed that 95 percent of working-age people formerly employed by the Project would need to secure alternative local employment or would leave the area after this time (DEIS 4.10-16). But he SDEIS states: at mine closure, workers who held operations-phase direct, indirect, and induced jobs would be expected to secure alternative local employment, retire, or relocate out of the area (SDEIS 5-502). The SDEIS does not provide a percentage expectation of numbers that would need to leave the area or be seeking alternative employment locally. This is important to know to fully assess the socioeconomic impacts to the community once mining has ceased. In addition, the SDEIS does not assess how many employees will be needed to conduct the long-term post-closure work of running the treatment facilities, maintaining pipelines and collection systems, pumps and waste rock piles, the long-term maintenance that is going to be needed for centuries. Current estimates.include 30 to 50 FTEs for the first 7 yearsand 5 to 10 FTEs for the following 30 years (SDEIS 5-499). We are not clear why evaluation of these jobs numbers would stop after 30 years when the need for maintenance will continue for centuries. The SDEIS needs to include this information and analysis.

77

Recommendation: The SDEIS needs to explore the option of Phase II of the project not taking place and what this would mean to the beneficial impacts expected for the project. The analysis of jobs over the life of this project needs to incorporate the likelihood of foreseeable technology replacing human labor. The SDEIS should present clear information about direct and indirect job expectations and where these jobs would be expected to come from geographically. Direct and indirect jobs should be presented separately, preferably in table format. The SDEIS needs to include a table similar to Table 4.10-13 on page 4.10-18 in the 2009 DEIS that shows the types and numbers of jobs expected to be created from the project. The SDEIS should include a table similar to Table 3.2-2 on page 3-17, Key Phases and Activities (Mine Site), that describes expected employment at both Mine and Plant Sites at key phases of the project. F. Unreliable Information About Taxes The SDEIS presents confusing and potentially inaccurate information about mining tax revenue that can be expected from the NorthMet project. Particularly confusing is the discrepancy between the 2009 DEIS information on estimates of federal and state and local taxes anticipated to be paid by the mining company and the current SDEIS. No explanations are provided for these differences and no changes in the project that could account for the differences are obvious. The 2009 DEIS states: The majority of economic benefits to the local community through taxes would be realized during the operating period. IMPLAN modeling estimates thatduring a typical year of operation the federal government would receive $17.3 million and the state and local governments would receive $14.5 million in taxes from the operation of the Project, excluding net proceeds tax (DEIS 4.10-19). But the SDEIS says: IMPLAN modeling estimates that, during a typical year of operation, the federal government would receive approximately $30 million, and the state and local governments would receive approximately $39 million in taxes from the operation of the NorthMet Project Proposed Action (SDEIS 5-503). Following the SDEIS numbers, Table 5.2.10-3 presents estimated taxes paid for 2011 if the project had already been in operations (SDEIS 5-503). Assuming this could be expected to be a typical year, one would expect to see similar numbers as modeled in the IMPLAN model. But here the SDEIS presents still different numbers. Here, the state taxes are anticipated to be $15.6 million dollars. It is possible that this differs from the IMPLAN model in that the model incorporates taxes paid from indirect sources too. This is not clarified in the SDEIS. But the table lists federal estimated taxes for 2011 as $64 million without any line-item explaining where this comes from. This is a number far larger than the $30 million described from the IMPLAN model which one would expect to be higher. The SDEIS creates a confusing picture for expected state and federal taxes. It needs to present the information more clearly. Recommendation: The SDEIS needs to present a more clear and consistent picture about expected tax revenue. Any differences from the 2009 DEIS should be explained. An explanation of where the $64

78

million comes from is needed. Without a clearer and consistent portrayal of the information, the public cannot have confidence in these revenue expectations.

X. Land Exchange The open-pit mine plan proposed in the SDEIS depends upon the PolyMet Mining Company completing a land exchange with the U.S. Forest Service (USFS) to obtain the surface rights. While PolyMet has leased the privately held mineral rights at the mine site, the surface rights are in public ownership, managed by the USFS. The 2009 DEIS notes: It is the position of the United States that the mineral rights leased by PolyMet do not include the right to open pit mine the National Forest System land (DEIS S-1). The exchange would result in the public losing 6,650.2 acres from the Superior National Forest. The Superior National Forest would acquire approximately 6,722.5 acres. The SDEIS describes an exchange without supporting evidence that it is in the publics best interest: A. Failure to Meet USFS Exchange Objectives The SDEIS indicates that benefit of the land exchange for the public and the Superior National Forest is that: The Land Exchange Proposed Action represents a transfer of surface rightsto eliminate the conflict between federal surface and private mineral estate (SDEIS 5-577) But this land exchange fails to accomplish this goal. When the USFS considers land exchanges, it must determine that such an action would translate into a benefit in serving the public interest. Among those considerations is whether the exchange consolidates split estates. Many other benefits are considered as well: Regulations provide that the Forest Supervisor may complete an exchange only after a determination is made that the public interest will be well served (36 CFR 254.3(b)). Factors that must be considered include: the opportunity to achieve better management of federal lands and resources, to meet the needs of state and local residents and their economies, and to secure important objectives, including but not limited to: protection of fish and wildlife habitats, cultural resources, watersheds, and wilderness and aesthetic values; enhancement of recreation opportunities and public access; consolidation of lands and/or interests in lands, such as mineral and timber interests, for more logical and efficient management and development; consolidation of split estates; expansion of communities; accommodation of existing or planned land use authorizations; promotion of multiple-use values; implementations of applicable Forest Land and Resource Management Plans; and fulfillment of public needs. See 36 CFR 254.3(b) and 254.4(c)4 (SDEIS 1-14). While the NorthMet exchange would eliminate the mine site from federal ownership, thus divesting the public of lands with a split estate, the proposed new lands that would be acquired also have split estates. Of the five tracts of lands considered for acquisition, four have split estates (SDEIS 3-163). Four of the five tracts fail to meet the Forest Service objective of 79

reuniting mineral and surface rights. Future mining on these lands cannot be ruled out as possible, and thus the public could be acquiring lands that carry the risk of future land-use conflicts. For example, Tract 1, the Hay Lake parcel, has moderate risk for future mineral conflict because of the presence of potential surficial aggregate resources in the far northeastern corner of the tract (SDEIS 5-586). In addition, the SDEIS fails to demonstrate a public benefit from the exchange for several of the other considerations the USFS must weigh. The SDEIS provides no evidence that: - The project meets the needs of state and local residents and their economies. Absent from the SDEIS is a meaningful assessment of potentially negative impacts, an analysis of impacts to tourism and recreation, impacts on real estate values and sales, impacts to commercial fisheries if mercury contamination occurs. - The project promotes the protection of fish and wildlife habitats (e.g. lynx critical habitat, moose habitat). See the Friends comments in the Section on Wildlife and Vegetation Impacts. For example, the exchange results in a loss of 4,752 acres of critical lynx habitat within Lynx Analysis Units, a loss of lynx denning habitat, and a decrease to the federal estate within designated LAUs (SDEIS 5-625, 629, 630). The project results in directly impact to 9 of the states 11 listed ETSC plant species, and indirect impact to 2 of the 11 state-listed ETSC plant species. - The project promotes aesthetic values. The proposed acquired parcels already have aesthetic value, while the aesthetic value of the mine site would be destroyed by the project. The project results in a net loss of aesthetic land values. - The project results in more logical and efficient management and development. The exchange would result in the loss of one contiguous parcel of land, and the acquisition of five scattered pieces of land. The SDEIS does not indicate how this aids the USFS in efficiency of land management. The exchange also fails to meet another important condition of USFS policy: USFS policy (Forest Service Handbook 5409.13 33.43c) provides that the following list of three conditions satisfy the requirements of EOs 11990 and 11988: 1. The value of the wetlands or floodplains for properties received and conveyed is equal (balancing test) and the land exchange is in the public interest (SDEIS 1-15). While wetland acreage in the exchange may be equal, the value of the wetlands is not. The public is losing 100 year old peatlands for fragmented wetlands of other types. B. Failure to Meet USFS Forest Plan Goals The Land and Resource Management Plan, 2004 (Forest Plan) for the Superior National Forest outlines forest-wide goals that guide the management of resources on the Forest. The SDEIS fails to demonstrate that the Forest Plan goals can be met with this exchange. Forest Plan goals that need a high level of attention in this proposed project include:

80

Promote ecosystem health and conservation using a collaborative approach to sustain the nations forests and watersheds Protect, and where appropriate, restore soil, air and water resources Provide for a variety of life by managing biologically diverse ecosystems Provide a variety of uses, values, products and services for present and future generations by managing within the capability of sustainable ecosystems Contribute to local, regional, and national economies by providing natural resources in a socially and environmentally acceptable manner Contribute to efforts to sustain the American Indian way of life, cultural integrity, social cohesion, and economic well-being

The SDEIS does not provide evidence that a land exchange will enable the USFS to meet its Forest Plan goals. The USFS wetland losses from the exchange would be significant, in quality and function. Indirect impacts from hydrologic changes from the mine project threaten to expand the impacts, potentially to a much larger geographic scope than has been analyzed. Two of the non-federal candidate tracts for acquisition are located outside the Lake Superior Watershed. At least three of the candidates are outside the St. Louis River Watershed. The exchange would result in an actual loss of positively contributing wetlands to an important watershed. While candidate lands bring wetland values in different watersheds, they are already existing wetlands that are adding value at the present time. They just are not within public ownership. Through this exchange, Minnesota and the nation lose, but do not gain, wetland function values. The federal lands contain nine known state-listed endangered, threatened or species of special concern (ETSC) plant species. Many of these populations are expected to be destroyed by direct or indirect activities of the mine. The SDEIS describes a plan to use non-native plant species to stabilize disturbed areas and to reclaim the project area. Some of these species are considered to be invasive species. The federal lands considered in the exchange include two important wildlife corridors. The SDEIS acknowledges that the NorthMet project will impact these travel corridors. The proposed land exchange would lead to impacts to resources of significance to the Ojibwe people. Significant impacts to water resources, wetlands, wild rice and wildlife are possible from the project, and would affect natural resources available for Ojibwe peoples use. Compensatory wetlands are proposed for outside the 1854 Ceded Territory and would result in a loss of these lands to the Ojibwe Bands. If the land exchange occurs, giving PolyMet ownership of the mine site lands, this area would be removed from public ownership and would diminish both Ojibwe Band members and the general publics access to these lands. The potential impacts to wild rice beds from increased sulfate concentrations, and the potential impacts to fish consumption from methylmercury contamination pose significant cultural impacts to the Ojibwe Bands.

81

C. Fragmented Lands Do Not Have Equivalent Ecological Value The proposed land exchange would result in the public losing a large, contiguous parcel of land that is 6,650 acres in size, in exchange for five smaller and isolated land pieces, one of which is merely 30 acres in size. Ecologists are aware that contiguous and connected habitats offer much higher biological values than smaller, less connected parcels. The proposed land exchange runs counter to ecological understanding of what provides higher quality habitat. D. The Exchange and Climate Change U.S. Forest Service Chief Tom Tidwell describes a key mission of the Forest Service as developing climate change resiliency. The Forest Service has made a commitment to manage its forest and grassland resources in ways that help address the climate change problem and its impacts on our natural resources. But the proposed land exchange has real implications for carbon emissions in the state of Minnesota. The nonfederal candidate lands, if they have carbon sequestration capabilities already function in that capacity and cannot perform as well as peatlands. The exchange will result in operations that will emit carbon in significant amounts. Recommendation: The SDEIS must contain supporting evidence that the land exchange serves the public interest. The SDEIS must document how the proposal meets both USFS land exchange objectives and the Superior National Forests Forest Plan Goals and Objectives. Claims that the project consolidates mineral and surface rights should be dropped, as the majority of the acquired parcels do not meet this criteria. The public should be fully informed that a land exchange will result in an actual loss of wetlands in Minnesota. Indirect impacts need to be more fully characterized. The SDEIS should be clear that the exchange will result in a loss of carbon sequestering environments. The SDEIS should address how this exchange addresses the recommendations of both the Climate Change Advisory Group and the Terrestrial Carbon Sequestration Initiative. The SDEIS needs to explore the significance of not mitigating the wetlands losses within the same geographic area, with the same wetland type, and how the wetland mitigation plan would affect greenhouse gas emissions. The SDEIS needs to examine how the exchange helps it meet the Forest Plan Goal of providing a variety of uses, values, products and services for present and future generations by managing within the capability of sustainable ecosystems, in light of climate change issues associated with this proposal. The SDEIS should fully analyze the potential negative impacts to communities, employment, and the economies of the area from the loss of the federal lands and this project. It must evaluate the regions reliance on environmental amenities such as tourism and recreation. This assessment should include the potential impacts from scenarios that include water contamination that could extend into the watershed.

82

XI. Agency Roles and Responsibilities The NorthMet proposed mine is a project put forward by PolyMet Mining Company and evaluated by three Co-lead Agencies: The U.S. Forest Service, the U.S. Army Corps of Engineers, and the Minnesota Department of Natural Resources. The involvement of these agencies is required by law to review the potential environmental impacts and make public review and input possible. The Co-lead Agencies have jointly prepared this SDEIS under the National Environmental Policy Act (NEPA) for the federal agencies and under the Minnesota Environmental Policy Act (MEPA) for the MDNR. The SDEIS describes the process the Colead Agencies undertook to evaluate the effects of the NorthMet Project Proposed Action, the Land Exchange Proposed Action, and alternatives developed during the process (SDEIS ES-3). The roles and responsibilities should be clear. But the SDEIS includes Purpose and Need statements from the agencies that suggest some of the Co-leads may be confused about their roles and responsibilities regarding this project. The SDEIS provides a Purpose and Need statement for PolyMet Mining for this project: The applicants stated purpose of the NorthMet Project is to exercise PolyMets mineral lease to continuously mine, via open pit methods, the known ore deposits (NorthMet Deposit) containing copper, nickel, cobalt, and PGEs to produce base and precious metal precipitates and flotation concentrates by uninterrupted utilization of the former LTVSMC processing plant (SDEIS 111). This is an understandable statement for the companys purpose and need. The SDEIS then provides Purpose and Need statements for each of the agency Co-leads. For the U.S. Forest Service, the SDEIS notes: The purpose for the USFS is to meet desired conditions in the Superior National Forest Land and Resource Management Plan (Forest Plan) and ensure that exploring, developing, and producing mineral resources are conducted in an environmentally sound manner so that they may contribute to the economic growth and national defense (from Forest Plan) (SDEIS 1-11 to 1-12). This too is an understandable statement by the USFS. It identifies its role to assess the project within the framework of meeting its goals and objectives in the Forest Plan. But the statements by the Minnesota DNR and the U.S. Army Corps of Engineers are troubling. Rather than identifying their roles as assessing the projects potential impacts in light of their own agency directives, they provide Purpose and Need statements that are identical to PolyMet Mining Companys statement. The Purpose and Need for both the Minnesota DNR and the USACE: The Purpose and Need of the Proposed Action is to produce base and precious metals precipitates and flotation concentrates from ore mined at the NorthMet Deposit by uninterrupted operation of the former LTVSMC processing plant. The processed resources would help meet domestic and global demand by sale of these products to domestic and world markets (SDEIS 1-12). These statements suggest the agencies have confused their roles and responsibilities with that of the mining companys. This is extremely troubling. The public relies on these agencies to impartially evaluate the mine project, to fully examine potential environmental impacts without bias. The public does not expect the Co-lead Agencies to review the proposed mine project with

83

the same goals as PolyMet Mining Company. Indeed, the Co-lead Agencies have been tasked with very different responsibilities. It is alarming that these mandates are not readily apparent in the way the Co-lead Agencies have characterized their roles in this SDEIS. The Minnesota DNRs purpose and responsibilities are guided by Minnesota Statute. It is the policy of the state to provide for the diversification of the state's mineral economy through longterm support of mineral exploration, evaluation, environmental research, development, production, and commercialization (Minnesota Statute 93.001). But at the same time, In determining the extent and type of regulation required, the commissioner shall give due consideration to the effects of mining upon the following: (a) environment; (b) the future utilization of the land upon completion of mining; and (c) the wise utilization and protection of the natural resources including but not limited to the control of erosion, the prevention of land or rock slides, and air and water pollution. The commissioner also shall give due consideration to (a) the future and economic effect of such regulations upon the mine operators and landowners, the surrounding communities, and the state of Minnesota; (b) the effect upon employment in the state; (c) the effect upon the future mining and development of metallic minerals owned by the state of Minnesota and others, and the revenues received therefrom; and (d) the practical problems of the mine operators and mineral owners including, but not limited to, slope gradients as achieved by good mining or soil stabilization practices (Minnesota Statute 93.47). The Minnesota DNRs website summarizes this responsibility with these words: The mission of the Minnesota Department of Natural Resources (DNR) is to work with citizens to conserve and manage the state's natural resources, to provide outdoor recreation opportunities, and to provide for commercial uses of natural resources in a way that creates a sustainable quality of life.DNR supports natural resource-based economies, managing state forest lands for multiple forest values (M.S. 89), ensuring the maximum long-term economic return from school trust lands (M.S. 127A), and providing other economic opportunities in a manner consistent with sound natural resource conservation and management principles (Minnesota DNR website). When the Minnesota DNRs purpose and need in the SDEIS is exactly that of the mining companys, it appears that the agency is no longer performing its legally mandated role to evaluate the project for its ability to be consistent with sound natural resource conservation and management principles. The roles and responsibilities of the USACE are guided by federal laws. The Clean Water Act indicates that, In carrying out his functions relating to the discharge of dredged or fill material under this section, the Secretary may, after notice and opportunity for public hearing, issue general permits on a State, regional, or nationwide basis for any category of activities involving discharges of dredged or fill material if the Secretary determines that the activities in such category are similar in nature, will cause only minimal adverse environmental effects when performed separately, and will have only minimal cumulative adverse effect on the environment (Clean Water Act, Section 404). Here the term Secretary means the Secretary of the Army, acting through the Chief of Engineers. (e)(1) (Clean Water Act, Section 404). The Clean Water Act indicates the USACE is to evaluate and permit projects that cause only minimal adverse environmental effects. The USACE website summarizes its job function:

84

The Regulatory Program is committed to protecting the Nation's aquatic resources, while allowing reasonable development through fair, flexible and balanced permit decisions. The Corps evaluates permit applications for essentially all construction activities that occur in the Nation's waters, including wetlands (USACE website). When the USACEs purpose and need in the SDEIS is exactly that of the mining companys it appears that the agency is no longer performing its legally mandated role to evaluate the project for its ability to cause only minimal adverse environmental impacts. Also confusing and troubling is that the SDEIS is not clear which of the Purpose and Need statements represents the overall Purpose and Need for the project. In screening for alternatives, for example, one criteria used is whether an alternative can meet the Purpose and Need for the project (SDEIS 3-140). But the SDEIS is not clear what the project Purpose and Need is. Is it the Applicants Purpose and Need? Is it the Co-lead Agencies Purpose and Need? If analysis of alternatives relies on the Purpose and Need statements of the Co-lead Agencies, it is all the more vital that their statements reflect the legal mission mandates and not the objectives of a mining company. Recommendation: The statements of Purpose and Need by the USACE and the Minnesota DNR in the SDEIS are inappropriate and undermine the publics confidence in the impartiality of the Agencies and the performance of their legal mandates. The SDEIS should include statements that reflect the legal requirements of each agency and reflect an impartial review of the environmental impacts of the proposed project. The SDEIS should be clear what the project Purpose and Need is.

XII. Significant Omissions and Unclear Information The SDEIS is a long document with a great deal of information to review and process. After our review, the Friends has questions about certain aspects of the project and has identified missing information that a SDEIS should clarify and include. 1. The Category 1 Stockpile cutoff wall is not constructed at first around the west end of the stockpile. Why is this? If the plan is to wait until Year 11 of the project when that stockpile footprint is complete, what is the plan for collecting and managing the drainage off the west end of the growing stockpile in Years 1-11? 2. The SDEIS contains no discussion about the network of pipelines conveying untreated and treated water. How many miles of pipelines are there? What are the contingency plans for breaks/damage to the pipelines? The SDEIS does not describe information about pipeline layout, engineering, or the pumps needed to push the water to various mine and plant locations. 3. Water is conveyed between the Mine and Plant Sites via a pipeline. Treated water will be moved from the Mine Site WWTF to the Tailings Basin, and in turn, water from the Plant Site may be moved to the Mine Site to help flood the West Pit. Are there two pipelines

85

4.

5.

6. 7.

8.

along the corridor between the sites, or a single pipeline that must provide multiple functions? The SDEIS notes that, "The design of the WWTF is based on the predicted water loads and constituents modeling (see Section 5.2.2)" (SDEIS 3-52). But a read of Section 5.2.2 provides no information about what that constituent make-up is, nor does any other part of the SDEIS. The SDEIS states, "Once the West Pit is full (approximately year 40), discharge of treated water from the WWTF to the West Pit would be terminated. The WWTF would be upgraded to RO and include evaporator/crystalizers to convert the RO reject concentrate to residual solids, which would be disposed of at appropriate off-site facilities" (SDEIS 3-72). If the reverse osmosis (RO) is not being added until Year 40, how are sulfates being removed before then? Some may be removed through the chemical filtration, but this wont remove down to a level that meets the standard. Is the "treated water" (but not gone through RO) being sent to the Tailings Basin and it is assumed that sulfates will be dealt with by either being deposited in the Tailings Basin or getting treated again in the RO system at the Plant Site? If so, does this mean the "treated water" flowing in the pipelines from the Mine Site to the Plant Site is high in sulfates? And what is the contingency plan if this pipeline breaks?? Why doesn't the WWTP at the Plant site include the evaporator/crystalizers to convert the RO to reject concentrate? Why transport it to the Mine Site? The SDEIS describes contingencies for a power outage at the Plant processing site: "In the event of a power failure, all process fluids would be contained within the concentrate dewatering and storage building and recycled to the process when power is restored" (SDEIS 3-101). But similar statements are lacking for the WWTF and WWTP. What happens in the event of a power outage at these facilities? The SDEIS provides a list of hazardous materials the project would use in construction, operation an closure phases of the project (Table 5.2.13-1; SDEIS 5-528). But the SDEIS does not specify how these various chemicals exit the NorthMet operations, if they leave the system via water, air, solid waste, or remain stored in tailings waste. If they remain on site, is there risk that they could seep into the surrounding waters through uncollected seepage? What is the plan for addressing such an occurrence? More information is necessary about these hazardous materials in the SDEIS.

86

XIII. References Alaska Department of Health and Human Services. July 2011. Technical Guidance for Health Impact Assessment (HIA) in Alaska. Anderson, Chel. December 1997. Evaluation of selected potential candidate research and natural areas as representative ecological landtype associations on the Superior National Forest, Minnesota. Minnesota Department of Natural Resources. Anderson, Jim. et. al. February 2008. The potential for terrestrial carbon sequestration in Minnesota: a report to the Department of Natural Resources from the Minnesota Terrestrial Carbon Sequestration Initiative. University of Minnesota. Barr Engineering. May 31, 2011. Technical Memorandum: economic consequences of meeting 10 mg/L sulfate standard. Barr Engineering. June 2012. Greenhouse Gas and Climate Change Evaluation: NorthMet Project. Pepared for PolyMet Mining Inc. Chambers, David. 2010. Draft NorthMet Project EIS. Letter to Stuart Arkley and Jon Ahlness, January 25, 2010. Chambers, David and Stuart Levit. 2014. Comments on NorthMet Mining Project and Land Exchange Supplemental Draft Environmental Impact Statement, prepared by Minnesota Department of Natural Resources, United States Army Corps of Engineers, United States Forest Service, November, 2013. Co-lead Agencies: Minnesota Department of Natural Resources, United States Army Corps of Engineers, United States Forest Service. September 27, 2013. Underground Mining Alternative Assessment for the NorthMet Mining Project and Land Exchange Environmental Impact Statement. Draft Environmental Impact Statement (DEIS), NorthMet Project, PolyMet Mining, Inc. October 2009. Prepared by the Minnesota Department of Natural Resources and the U.S. Army Corps of Engineers. Earthworks. July 2012. U.S. Copper Porphyry Mines: The track record of water quality impacts resulting from pipelines spills, tailings failures and water collection and treatment failures. Edison Investment Research Limited. November 21, 2013. PolyMet Mining Corp. Low-cost polymetallic development project. Glaser, Paul H. 2014. Supplemental draft environmental impact statement, November 2013: report on wetland components of the EIS.

87

Halley, F. Michelle. January 19, 2014. Letter to Michigan Department of Environmental Quality re: comments on draft permit GW1810162, Eagle Mine. Human Impact Partners. 2014. Frequently Asked Questions About Integrating Health Impact Assessment into Environmental Impact Assessment. Website. http://www.humanimpact.org/component/jdownloads/finish/11/42 Kruse, Greg. December 17, 2013. Office memorandum, re: Partridge River base flow analysis MDNR Gage #H03155002. Department of Natural Resources, Division of Waters. Maest, Ann. 2014. Technical memorandum re: Polymets Supplemental Draft Environmental Impact statement: water quality and geochemical issues. Malusis, Michael. February 2014. Containment system assessment report, Supplemental Draft Environmental Impact Statement (SDEIS), NorthMet Mining Project and Land Exchange. Miller, Glenn. 2014. Comments on the NorthMet Supplemental Draft Environmental Imapct Statement with attachments. Mining Minnesota. September 25, 2013. Letter to Governor Mark Dayton Re: Minnesota should say yes to copper-nickel mineral development. Minnesota Climate Change Advisory Group. 2008. Minnesota Climate Change Advisory Group final report: a report to the Minnesota Legislature. Minnesota Public Radio News. January 27, 2014. PolyMet environmental study inconclusive on water treatment, fueling confusion. http://www.mprnews.org/story/2014/01/27/polymet-watertreatment-confusion?from=hp. Myers, Tom. March 2014. Three-part technical report: groundwater flow and transport modeling, NorthMet mine and plant site; technical memorandum: review of the Supplemental Draft Environmental Impact Statement; review of the water quality modeling. PolyMet Mining. 2006. Hydrogeologic Investigation-Phase II PolyMet NorthMet Mine Site RS10. Barr Engineering. PolyMet Mining. March 2007. RS10A-Hydrogeological-Drill Hole Monitoring and Data Collection Phase 3. PolyMet Mining. March 7, 2013. NorthMet Project Adaptive Water Management Plan. PolyMet Mining. March 2013. Water Modeling Data Package Vol 1, Mine Site v 12 March 2013. PolyMet Mining. March 2013. Water Modeling Data Package Vol 2, Plant Site, v 9 March 2013.

88

Supplemental Draft Environmental Impact Statement, NorthMet Mining Project and Land Exchange, PolyMet Mining Inc. November 2013. Prepared by the Minnesota Department of Natural Resources, United States Army Corps of Engineers, United States Forest Service. U.S. Environmental Protection Agency. September 1997. EPAs National Hardrock Mining Framework. U.S. Environmental Protection Agency. August 2009. Letter from Kenneth Westlake to Tamara Cameron, Chief Regulatory Branch, St. Paul District. U.S. Environmental Protection Agency. February 18, 2010. Letter from Bharat Mathur to Colonel Jon Christensen Re: NorthMet Project-Draft Environmental Impact Statement. U.S. Environmental Protection Agency. August 7, 2013. Letter to Co-leads Re: Preliminary Supplemental Draft Environmental Impact Statement for the NorthMet Mining Project and Land Exchange, Hoyt Lakes, St. Louis County, Minnesota. U.S. Fish and Wildlife Service. November 2013. Draft biological assessment for the proposed NorthMet mining project and land exchange. Vora, Robin. January 1997. Identification of potential natural areas, including representative ecosystems, on the Superior National Forest. Superior National Forest

89