The U.S. Environmental Protection Agency (EPA) has reviewed the Phoenix Mine Project Final Environmental Impact Statement (EIS), Lander County, NV [CEQ #020001]. Our review and comments are provided pursuant to the National Environmental Policy Act (NEPA), the Council on Environmental Quality’s (CEQ) NEPA Implementation Regulations at 40 CFR 1500-1508, and Section 309 of the Clean Air Act. We appreciate the final deadlline extension we received verbally from Jean Rivers-Council, and are providing our comments within the agreed upon timeframe.

The Bureau of Land Management (BLM) proposes to issue a mining plan of operation to Battle Mountain Gold Company (BMG) for its Phoenix Mine near Battle Mountain, Nevada. BMG is a wholly owned subsidiary of Newmont Mining Corporation. The proposal would significantly expand BMG’s current mining and mineral processing operations at the site, over a 28-year period, including development of two new open pits, expansion of two existing pits, processing of previously mined stockpiled gold ore, expansion and construction of heap leach, milling, waste rock, and tailings facilities, and backfilling of three existing open pits. The proposed project would disturb 6,497 acres (greater than 10 square miles) of public and private lands. Of this, approximately 4,295 acres would be new disturbance (2,382 acres of public lands and 1,913 acres of private lands).

The project area includes mining facilities from more than a century of copper and precious metals mining. Groundwater and surface water in the area currently exceed drinking water standards for numerous contaminants, including pH, total dissolved solids, and metals. The acid-generating potential of the waste rock at the Phoenix Mine is high compared to other active or proposed hard rock mines in the western United States. According to the Final EIS, the majority of waste rock at the site is acid-generating; and acidic water moving through the waste rock and tailings from the proposed project will mobilize metals and sulfates, leaching them into groundwater beneath the site after mine closure.

In order to prevent post-closure off-site migration of the contaminant plumes, BLM’s proposed mitigation includes implementation of the Contingent Long-Term Groundwater Management Plan (Brown and Caldwell, 2001), and establishment of a Long-Term Trust Fund by BMG to ensure that funds are available to implement the Contingent Long-Term Ground Water Management Plan in perpetuity. The funding level of the trust fund is based on the Preliminary Cost Estimate for the Phoenix Project Contingent Long-Term Groundwater Management Plan (“Cost Estimate”), prepared by BMG’s contractor, Brown and Caldwell.

This project will be the first one in the country to include post-closure financial assurance because of the anticipated acid mine drainage, under the recently finalized Surface Mining Regulations for Surface Mineral Operations at 43 CFR 3809 (“3809 Regulations”). Therefore, BLM’s approach for applying the new 3809 Regulations is critical to ensuring the mining operator bears the costs of reclaiming mined lands and fully mitigating the project’s impacts after mining operations have ceased. We believe BLM’s approach in this regard remains flawed. We do not agree that the Long-Term Trust Fund and an additional $1 million surety, as currently proposed by BLM, would adequately cover the cost of implementing the Contingent Long-Term Groundwater Management Plan. EPA estimates that the mining company should set aside $33.5 million into the trust fund at project startup rather than the $408,000 proposed by BLM. The environmental acceptability of this project hinges on an adequately funded long-term mitigation program. While we agree with BLM’s approach to monitoring the trust fund on a triennial basis, BLM’s current strategy does not allow for a timely and informed judgment about the project’s cost-effectiveness or whether an alternative approach such as improved reclamation should be reconsidered. We have attached detailed comments in support of these conclusions [Attachment 1].

EPA believes the project will likely create a perpetual and significant acid mine drainage problem requiring mitigation for hundreds of years. Our conclusion is based on the information provided in the BLM’s NEPA documents for the Phoenix Mine project, including information about current mitigation activities for prior and ongoing mining activities at this site, as well as our experience with sites in northern Nevada and elsewhere in the region. While the modeling done by BLM to predict groundwater impacts may assume a considerable amount of uncertainty, we do not believe its conclusions represent a “worst case” scenario, as represented by BLM in recent correspondence and discussions. We recommend BLM take a prudent approach to a project of this magnitude and projected impact, consistent with the objectives of NEPA, BLM’s 3809 Regulations, and discussions in a recent advanced notice of proposed rulemaking regarding acid mine drainage issued by the Department of Interior’s Office of Surface Mining.

The Final EIS is inadequate as it does not yet include an itemized cost estimate for closing and perpetually operating and maintaining the site, or meaningful assurances that a financial instrument will exist to ensure funds are available in perpetuity to prevent degradation of water quality and impacts to biological resources. From our 20 years of experience with the remediation of mine sites throughout our region, we have acquired extensive technical and financial expertise and knowledge that we believe should be applied to the Contingent Long-Term Groundwater Management Plan for this project. We recommend additional analyses to determine the appropriate levels of funding for the Long-Term Trust Fund before project startup. Given the significance of this project, we believe BLM’s additional analyses and conclusions should be circulated in a Supplemental Draft EIS for public comment, in accordance with NEPA and CEQ’s NEPA Implementation Regulations. In addition, EPA strongly recommends the Record of Decision (ROD) include the following commitments and information:

• BLM will engage an independent third party outside of the Department of Interior and EPA to complete a full analysis of the costs and financial predictions associated with BMG’s ability to implement the Contingent Long-Term Ground Water Management Plan before the project’s first triennial review. EPA can assist you in selecting the appropriate third party.

• BLM will apply the results of the third party evaluation and adjust the funding levels to support the Contingent Long-Term Ground Water Management Plan in the first triennial review of the Long-Term Trust Fund.

• BLM will specify the detailed mechanics of the Long-Term Trust Fund that are critical to ensure that sufficient funds would be available to implement the Contingent Long-Term Groundwater Management Plan in perpetuity.

• BLM will identify specific clear triggers and standards that will require the owner/operator to increase the Long-Term Trust Fund, or allow BLM to transfer money from any unused portion of the reclamation bond into the trust fund.

• BLM will require Newmont to act as guarantor of these financial assurance amounts to ensure that taxpayer dollars will not be required in the event that BMG is unable to meet its financial assurance obligations.

• BLM will include the new reclamation measures that have been added to the Phoenix Mine proposal since publication of the Final EIS.

We realize that your current approach for the post-closure phase of this project was formulated prior to development of BLM’s national guidelines for implementing the new 3809 Regulations. Clearly, our dialogue on the Phoenix mine project would have benefitted from a final guidance document. EPA welcomes the opportunity to work with BLM on these guidelines. We strongly encourage BLM to take advantage of our experience and expertise in developing long-term trusts for future treatment costs as you determine how to implement this new rule.

EPA appreciates the opportunity to review this Final EIS and we look forward to working with you as this project continues. Our goal is to work with you to prevent serious long-term contamination of ground and surface waters. If you have any questions, please call me at (415) 947-8702, or refer staff to Lisa Hanf, Federal Activities Office Manager at (415) 972-3854 or Jeanne Geselbracht, our lead NEPA reviewer for this project, at (415) 972-3853.

Attachment 2: May 8, 2002, Letter from Enrique Manzanilla, EPA, to Robert Abbey, BLM

Attachment 3: May 17, 2002, Letter from Robert Abbey, BLM, to Enrique Manzanilla, EPA

The Phoenix Mine Final Environmental Impact Statement (EIS) establishes that this project may create a significant long-term acid mine drainage problem. According to the Final EIS, baseline groundwater quality has been affected by past mining activities in the project area, and groundwater currently exceeds maximum contaminants levels (MCLs) for several parameters, including arsenic, cadmium, copper, nickel, zinc, selenium, mercury, beryllium, lead, iron, manganese, aluminum, chloride, total dissolved solids, and low pH. The Final EIS states that the proposed mine closure and reclamation activities would not provide sufficient neutralizing material to prevent perpetual acid mine drainage from the site. The Proposed Action is predicted to result in further degradation of groundwater, including exceedences of MCLs for numerous parameters, including low pH, sulfate, and at least 19 other constituents. Surface water quality sampling has also demonstrated exceedences of drinking water standards for several parameters in several streams on or just downstream of the existing mine site. The most acidic surface waters occur adjacent to historic mining facilities in Iron Canyon and Butte Canyon. Surface waters, including springs and seeps in these areas, exceed drinking water standards for antimony, arsenic, beryllium, cadmium, copper, chromium, fluoride, iron, magnesium, manganese, mercury, nickel, nitrate, pH, sulfate, total dissolved solids, and zinc (Final EIS, p. 3.2-14 - 3.2-18). Battle Mountain Gold Company (BMG) has been collecting and treating acidic surface water in these drainages since 1998, and expects to continue doing so until mine closure.

The Contingent Long-Term Ground Water Management Plan could provide acceptable mitigation assuming sufficient funds are available for implementation. However, the Preliminary Cost Estimate for the Phoenix Project Contingent Long-Term Groundwater Management Plan (“Cost Estimate”) prepared for Battle Mountain Gold Company (BMG) by Brown and Caldwell does not demonstrate that adequate funding will be provided.

Based on BMG’s Cost Estimate, BLM indicates the net present value of implementing the Contingent Long-Term Ground Water Management Plan is $408,000, and that this amount should be paid into a Long-Term Trust Fund at project startup in order to cover the costs of implementing that Plan. BLM indicates the proposed monitoring program during project operations would provide sufficient data to determine whether adjustments will be necessary to the current cost estimates or Long-Term Trust Fund; and that BLM can require BMG to make adjustments to the Long-Term Trust Fund if needed. BLM will also require that BMG post a $1 million certificate of deposit for 30 years to ensure that money is available to fully fund the trust should BMG default on additional payments into the trust which BLM deems necessary. Based on EPA’s experience with costs and funding issues at mine sites with perpetual acid rock drainage problems, we do not agree that the Long-Term Trust Fund and additional $1 million surety, as currently proposed, would be adequate to cover the cost of implementing the Contingent Long-Term Ground Water Management Plan. Newmont recently provided EPA with the cost estimating program used by Brown and Caldwell to develop its Cost Estimate. Based on our review of the Cost Estimate and our very brief review of the cost estimating program, EPA estimates that, in order to cover the cost of implementing the Contingent Long-Term Ground Water Management Plan, the mining company should pay $33.5 million into the trust fund at project startup.

EPA also believes that it will be very difficult for BLM to require BMG to make additional payments into the Long-Term Trust Fund due to, among other things, the significant increase in funds required and the amount of time that will pass until information sufficient to trigger such an increase becomes available. Furthermore, BLM’s proposal does not assure that the cost of the mitigation will be paid by BMG, as vital details regarding the mechanics of the Long-Term Trust Fund have not been provided.

If off-site groundwater contamination could not be controlled, with a feasible mitigation program, consistent with our comments on the Draft EIS, the project will be environmentally unacceptable. EPA believes the proposed mitigation may be infeasible because $408,000 in the Long-Term Trust Fund at project startup may be inadequate to yield the funds necessary to implement the Contingent Long-Term Ground Water Management Plan at year 60 or earlier. Furthermore, EPA does not believe monitoring would provide sufficient data to determine whether adjustments are necessary, or that BLM will be able to make such adjustments, for the reasons discussed in Sections 2 and 4 below.

Our discussions with BLM over the past several months have related to: (1) treatment and monitoring costs; (2) net present value of treatment and monitoring costs; (3) mechanics of the Long-Term Trust Fund; and (4) adjustments to the Long-Term Trust Fund. Each of these issues is discussed below.

EPA believes that BLM’s long-term costs are significantly underestimated. Based upon our review of the cost estimate and experience with similar projects, EPA has concluded that BLM should set aside more than three times the current estimated amounts for capital construction and operations and maintenance (O&M) for the long-term treatment activities by a third party and for appropriate contingencies. In addition, a fundamental assumption of the BLM financial assurance program is that the mine operator will not be available to fulfill its financial obligations. For that reason, BLM regulations require that reclamation costs be calculated based upon the assumption that the reclamation is performed by a third party paid by the federal government.

EPA is concerned that the high degree of uncertainty in the costs that BMG projects for extraction and treatment of the expected contaminated groundwater has not been addressed in the Final EIS. The groundwater flow modeling conducted by Baker Consultants assumed that the bedrock is a heterogeneous "porous medium," yet the bedrock in the project area is highly fractured and faulted. This simplified approach was used because of the lack of data at this stage of project planning. This approach introduces significant uncertainty and does not provide a conservative assessment. Faults and fractures in the bedrock create both barriers and conduits for groundwater flow and make it very difficult to predict velocity, direction, and pattern of a contaminant plume. Simulating ground water flow through fractured media is difficult, necessarily involves gross simplifications, and is inherently less accurate than modeling of porous media. Experts in the field do not agree on the applicability or accuracy of such simulations.

The modeling approach used, although reasonable at this stage of planning, would not be acceptable for a final design document. The uncertainties related to the modeling must include contingencies, such as those suggested by EPA in this attachment, for estimating the long-term costs of mitigation.

1. The fractures and faults in the project area generally trend parallel to the predicted groundwater flow. The extraction wells may not capture the contaminant plumes if they travel past on a parallel track. The precise locations of wells to assure the capture of all contaminated groundwater in the fractured bedrock system is expected to be difficult to determine. Additional wells may be required in order to capture the contaminated flows. (The first set of extraction wells is proposed to be four wells. A 50 percent contingency would be enough for two more wells. This is not an unreasonable contingency and could be low).

2. The contaminant could reach the groundwater much sooner than 60 years if the fractures and faults act as conduits for flow.

3. The amount of water and contaminants reaching the wells could be greater, or could be more contaminated than predicted, increasing the costs of pumping, treating, and disposing of the water and treatment sludge. (For example, if the plant treats contaminated groundwater with higher concentrations than predicted, the Contingent Long-Term Ground Water Management Plan indicates that BMG would switch from sodium hydroxide neutralization to lime neutralization. This could double the cost of treatment).

BLM has estimated long-term capital and O&M costs in the Preliminary Cost Estimate for the Phoenix Project Contingent Long-Term Ground Water Management Plan (see table below). These costs include future average annual costs of $64,000 per year for monitoring, which would begin approximately at mine closure, and future average annual costs of $483,000 for extraction and treatment of contaminated groundwater, which would begin approximately 30 years after mine closure. BLM has also estimated the capital costs for facilities necessary to extract contaminated groundwater, treat it, and reinject the treated groundwater. Although information is limited in the Preliminary Cost Estimate for the Phoenix Project Contingent Long-Term Ground Water Management Plan, it appears that BLM cost estimates for several major components include: $659,000 for the groundwater monitoring wells to be constructed 20 years after mine closure; $731,000 for the groundwater extraction and injection wells to be constructed 30 years after mine closure; $2,060,000 for the treatment plant to be constructed 30 years after mine closure; and treatment plant O&M costs of $235,000 per year, as described below.

EPA’s review of these estimates found that the future treatment and monitoring programs could cost more than 3 times BLM’s estimate to construct and operate. For example, EPA believes a realistic cost for the treatment plant is $6.7 million, and a realistic cost for treatment plant O&M is $750,000 per year. The adjusted average undiscounted costs per year would be in the range of $1.8 million per year. However, BLM has not provided EPA with full details of the cost elements and assumptions used in the BLM cost estimate, so EPA’s estimates may not be complete.

Monitoring & Treatment Activities	BLM Estimate	EPA Estimate
Treatment plant (capital costs)	$2,060,000	$6,700,000
Extraction & injection wells (capital costs)	$731,000	$2,400,000
Treatment plan O&M	$235,000	$750,000

EPA’s review also concludes that BLM did not include all appropriate markups for third party construction of the necessary capital improvements and performance of monitoring and O&M as specified in BLM’s applicable guidance (Nevada BLM Bonding Process for Plans of Operations Authorized by 43 CFR 3802/3809, September 2000). Costs associated with performance of construction and O&M by a third party should have included additional markups for contract administration consistent with requirements of the Federal Acquisition Regulations (approximately 20 percent), and provision for overhead for the third party (approximately 7.5 percent).

A few examples illustrate the deficiencies with the current cost estimate and indicate that not all costs specified in applicable guidance are included in BLM’s cost estimates. BLM’s estimated cost for purchasing lime is $72 per ton. At an EPA Superfund site, a private contractor purchases 7,000 tons of lime per year at $72 per ton. The contractor must also pay freight for delivery to the plant, taxes, administrative overhead and include a markup for profit that totals $28 per ton. To price this component of the treatment operations under BLM requirements, the price must include all costs (taxes and freight), an additional mark-up of 18 percent for contract administration, and include a mark-up for profit of 10 percent. The BLM cost estimate does not include the full lime price (since it excludes the costs for freight and taxes) or any of the appropriate markups for federal contracting and third party profit. Many of these items are also required by Nevada BLM for cost estimates. Id. Our review indicates that the cost estimate does not provide for these required elements for lime anywhere else in the estimate. This example leads us to believe that in general the cost estimate does not include all appropriate adjustments to base costs.

Similarly, BLM appears to underestimate labor costs. The cost estimate uses a labor markup of 1.2125. Based upon our experience with labor multipliers for long-term treatment plants, a more realistic and appropriate treatment plant multiplier that addresses all third party costs required for federal contracting would be 1.75 to 2.0. A detailed review also shows that BLM has only included required federal markups on items such as social security and medicare, but has not included markups for overhead and administrative costs that a third party would include in its pricing. These additional markups are necessary to provide an accurate and realistic cost estimate.

Our detailed review of the information provided by BLM also indicates that the alternative treatment approaches that may be required by the Contingent Long-Term Ground Water Management Plan are not factored into BMG’s Cost Estimate. For example, the Cost Estimate indicates that continuous treatment with lime could be required if the acidic drainage is more concentrated than anticipated in the model (pp. 10-12). The costs for this type of treatment alternative would more than double the estimated annual O&M costs from $750,000 per year to $1,800,000 per year, for treatment costs alone. The actual cost estimate, however, does not include any funds to account for this risk (Cost Estimate, Appendix H). Rather than exclude this possibility, the estimate should use a weighted average to develop a more realistic estimate of expected costs.

EPA is concerned that BLM provides only a 10 percent contingency for all costs associated with the Contingent Long-Term Ground Water Management Plan. While consistent with guidance for BLM surface reclamation activities, this low rate is not appropriate for addressing the significant uncertainties associated with controlling a long-term groundwater problem of the magnitude expected under the proposed large-scale expansion of mining into acid-generating rock. BLM has extensive experience with mine reclamation activities which generally involve easily quantified costs, such as capping waste piles and moving known quantities of materials. In contrast, the proposed long-term treatment remedy is similar to the process for a long-term clean-up at a Superfund site. The treatment of groundwater in perpetuity is subject to a wide range of uncertainties such as those produced by reliance on groundwater modeling of a future event, groundwater chemistry, flow rates, and treatment effectiveness. The uncertainty in this case is especially high due to the acid-generating nature of the rock and the scope of the mining that will occur. Given the more complicated and speculative nature of the mitigation, the low contingency used for simpler, more quantifiable projects is not appropriate.

The very low contingency used by BLM is a serious concern given the approach used to estimate costs. BLM’s proposed perpetual treatment program has taken the most optimistic approach in each instance, and anticipates that installation of the minimum number of extraction wells and treatment of the minimum flow of contaminated groundwater are all that will be required. Based on years of experience with complex mining sites, EPA believes that, at this stage in the conceptual formulation of the proposed perpetual treatment activities, and in this extreme type of acid producing rock, it would be reasonable for BLM to include a 50 percent contingency to provide for a significant expansion of the extraction and treatment system. EPA and the U.S. Army Corps of Engineers developed “A Guide to Developing and Documenting Cost Estimates During the Feasibility Study” (July 2000), EPA 540-R-00-002, OSWER 9355.0-75. Based on this guidance, a 25 percent contingency should also be included for normal design, and a 20 percent contingency should also be included for construction. Neither of these contingencies are included in the BLM cost estimates.

Finally, EPA is concerned that BLM provides only an 8 percent markup for all costs associated with engineering support. Based on 20 years of experience remediating mine sites, EPA would expect that markups for project management, design and construction management would total 26 percent or more.

Recommendation: BLM should use the EPA values shown in the table above to calculate the net present value of implementing the Contingent Long-Term Ground Water Management Plan.

BLM’s predicted rate of return is overly optimistic. The foundation of BLM’s long-term financial assurance program is the real investment return on the trust fund. For the reasons stated below, EPA believes BLM’s real return rate is overly optimistic.

The financial aspects of the Long-Term Trust Fund for the Phoenix project should be based upon the assumption that the company will not be available to pay taxes. However, the BLM analysis for this project assumes that Newmont is available to pay taxes and fees for the entire life of the project. If the site operator is not available to pay taxes, either the trust itself would have to pay taxes and fees, or the financial assurances could be converted into a federal trust of some kind (with no taxes being due).

The impact of taxes on the trust earnings is substantial. The rate of tax depends on a number of factors, including the legal structure of the trust and the type of investments in the trust. BLM has not provided information about the tax structure of the trust after insolvency of the operator. However, if the trust were to become a stand alone trust under section 468B of the tax code, the earnings of the trust would be subject to federal taxes at a rate in the range of 40 percent. A 40 percent tax would reduce the nominal return to 60 percent of actual earnings, but the real return would be even less than that. For example, a nominal return of 8 percent would yield an after-tax return of 4.8 percent. After inflation at 4 percent, the net real return would be in the range of 0.8 percent (4.8 percent minus 4 percent) without accounting for trust and investment fees. It is also possible for the trust to invest in tax free instruments, but that would affect the rate of return as tax free instruments generally have a lower yield than taxable ones.

If the corpus of the trust becomes federal funds after the operator becomes insolvent (for example on the theory that the federal government should be considered the owner of the funds), no taxes would be due on the growth of the funds after conversion (although back taxes may be due); however, if the trust is considered federal funds, the investments in the trust would be limited by federal rules, which would make the assumptions that underlie the BLM net present value analysis inapplicable.

Recommendation: BLM should assume that Newmont or BMG will not be available to pay taxes on the trust account. This assumption should be factored into BLM’s real return rate.

The net present value analysis performed by BLM also does not consider the impact of management fees and other trust expenses that affect the rate of return. Management fees vary considerably based on a number of factors, but it is important to consider their impact over time. Fees can range from 0.25 to 1 percent or more depending on the amount in the trust, the type of management and other factors. Wells Fargo Bank, which operates the second largest trading desk for fixed income instruments in the country, would charge an annual rate of 0.56 percent per year on a trust that holds $40 million. If the trust held less than $2 million, the fees would be 1.3 percent. This amount leads to a direct reduction in the annual return.

Recommendation: BLM should not assume that the company is available to pay these fees; rather, BLM should assume, as it does with respect to reclamation costs, that the site operator is not able to pay these costs and that government will be responsible to pay the costs from the financial assurance instrument. For this reason, BLM should deduct these costs from the annual earnings of the trust account. An assumption of at least 0.50 percent should be used, particularly if the trust holds less than $40 million.

The approach used by BLM includes zero margin for safety on the economic projections. Most of the variables discussed in the economic analysis are not static but change on a regular basis due to complex and interrelated factors. This variability increases the risk to adequately funding the project. For example, consider the situation where a project has to spend $3 every year in today’s dollars. One might initially consider the trust to be fully funded if one expected a fixed real return of 3 percent and the trust contains $100. In that case, the trust would be expected to generate $3 in inflation adjusted dollars, and it can continue to do so because the principal remains at an inflation-adjusted $100. However, if market conditions cause the trust to generate only $2 per year for 10 years, the trust will consume more than $10 in principal. When market conditions return to “normal,” the trust would start to earn 3 percent again; however, the trust will generate less than $2.7 dollars because the principal will be less than $90. At that point, the trust will continue to consume principal at a faster and faster rate, until the entire amount is gone. It is possible that a series of good years will follow and allow the fund to “catch-up,” but the conditions required for that to be achieved are uncertain.

EPA understands that Newmont relied upon a deterministic approach to calculate the start value for the trust account. A deterministic approach relies on simple averages and does not take variability into account at all. If the correct average assumptions are used, the approach would provide only a 50 percent chance of adequate funding. A deterministic approach is not appropriate or realistic given the importance of using a conservative approach to provide adequate financial assurances and the certainty that the economic factors will be variable from year to year. Moreover, while the long-term nature of the project does have some averaging effect, an analysis of variability is still an important step to consider because the trust will have unavoidable expenses (treatment plant construction and operation costs) that could occur during periods of poor investment returns. We must assume that BMG is unable to meet those obligations or put more money in the trust fund because BLM anticipates that the trust fund will not be needed until long after BMG closes the site.

Recommendation: EPA recommends that BLM perform an appropriate Monte Carlo analysis, using an appropriate average and standard deviation for each variable, particularly to model the amount necessary from years 60 to 180. The model should rely upon the auto-correlated nature of inflation and the complex but clear relationship between inflation and investment returns. An appropriately conservative probability of success (80 or 90 percent) should be selected, with success being defined as meeting and maintaining an amount necessary to support perpetual treatment. This type of an approach would provide a more realistic assessment of whether the trust fund will be adequate to pay for perpetual treatment. We note that the model must be transparent and replicatable to ensure that the conclusions are valid and defensible.

Another key assumption for the net present value calculation is the projected inflation rate for the life of the project. The initial cost estimate assumes that inflation will have a compound rate of 1.5 percent per year over the life of the project. The basis for this figure has not been provided; however, it would appear to be inconsistent with any reasonable estimate of long-term inflation. Newmont has recently proposed orally that the United States rely on an inflation rate of 3.1 percent. We understand that figure is based on 75 years of inflation data.

EPA agrees that there is much uncertainty about what inflation rates will be in the future. We also believe that an assumption of 3.1 percent is more reasonable than an assumption of 1.5 percent. However, we believe that a more conservative rate of 4 percent is appropriate, based upon the inflation experience during the modern economic era – since World War II. This more conservative approach will provide a higher degree of confidence that an appropriate sum is set aside to for perpetual treatment. The period is also more likely reflective of our current monetary policy (which is not based on the gold standard) than the pre- World War II era. Since the end of World War II, inflation has never been at or below 1.5 percent for any 30- year period. Investment Dimensions 1926-2000, Dimensional Fund Advisors (2001) at 44-45. (DFA 2001). Since 1944, the compound inflation rate has never been less than 3.6 percent per year for any 30- year period and the average for the entire period is 4.1 percent. Over the last 30 years, inflation has an average compound rate of 4.9 percent. Id. The graph below illustrates the degree to which the BLM and Newmont inflation assumptions are inconsistent with inflation rates of the past 50 years.

We are currently in a period of relatively low inflation, so it may be appropriate to use a figure lower than the 4.9 percent average of the last 30 years, but the assumption BLM used for the cost estimate is not supported by the Final EIS and is inconsistent with historic patterns and currently available information. We believe that a figure of 4 percent would be acceptable in these circumstances. That rate is approximately equal to the long-term compound rate since 1944 of 4.1 percent.

Recommendation: We believe that the model should rely upon an average inflation rate of 4 percent.

BLM has assumed that the trust fund will earn a nominal rate of return of 8 percent at the same time it assumes inflation will be 1.5 percent. Newmont has told EPA it believes the net present value calculation should be based on a nominal return rate of 9.8 percent. A nominal rate of return of 8 percent appears overly optimistic, particularly in periods of average or low inflation or if low risk investments are used. Since the prime objective in this matter is to determine the real rate of return, one must use consistent assumptions for inflation and the nominal return. The nominal return on a bond held to maturity is a function of the risk free real return rate, the risk premium, and expected inflation. Thus, in periods of low inflation expectations the nominal rates are lower while in periods of high inflation expectations the nominal rates are higher; but in both high and low inflation environments, the real rate of return remains relatively constant within a range. This section discusses currently available rates and historic rates in the context of inflation.

The 8 percent assumption used by BLM is higher than the return rates available today on low risk fixed-income investments. Currently, AAA corporate bonds yield between 3.12 and 6.83 percent. Bloomberg Yield Curve Number 21 (AAA) (as of May 17, 2002). Longer-term bonds have a higher inflation risk because the coupon does not change in response to increasing inflation, so it is prudent to have a range of maturities in a bond portfolio. Since it would be prudent to have a range of maturities for the investments to manage inflation risk, the expected return on a blend of high grade bonds would be in the 5 percent range. The blended yield on a comparable series of government bonds would be lower.

EPA has also contacted several trust departments at major banks that handle these types of trusts. Senior portfolio managers in trust departments at major banks confirm that return rates in the 8 percent range are not currently available under current market conditions for conservative fixed-income instruments. These institutions also indicate that it is not appropriate to rely on a nominal return rate of 8 percent for long-term planning purposes in an average or low inflation environment.

Similarly, the average historic returns on low-risk fixed income instruments have been significantly less than 8 percent during periods of average inflation. See DFA at 50-51. The

nominal return on long-term government bonds has averaged 5.6 percent since 1944. A review of data from the DFA indicates returns above 8 percent do not occur during periods of low inflation (such as the 1.5 percent used by BLM). While low risk bonds may have returns of 8 percent during periods of above-average inflation, the net real return on the bonds remains in the 2 percent range. For example, once inflation is taken into account, long-term government bonds have averaged 1.4 percent since 1944, with the 30-year average ranging between -2.1 and 4.1 percent. The graph below shows the extent to which the real return rate assumed by BLM is higher than the 30-year average since the end of World War II. For example, from 1944 to 1974, the 30-year average compound return on long-term bonds was -1.4 percent. The negative real return is due to the fact that after the bonds were purchased, inflation increased above the rate expected at the time the bonds were purchased. Similar risks exist today because currently inflation expectations are low so nominal bond returns are low. If inflation increases, long-term bond holders face the risk of negative real returns. All of these historic averages and currently available yields are significantly lower than the 6.5 percent real return assumed by BLM.

Actual data from historic returns on mine closure funds also indicates that the 8 percent or higher return rate may not be realistic. Barrick Goldstrike Mines Inc. established two post-closure funds

in 1991 to pay for long-term monitoring and environmental contingencies at the Betze Project near Carlin, Nevada. The period 1991 to the present is considered to be an exceptional period not only for equities, but also for fixed income instruments like government and corporate bonds. In part, the returns have been highly favorable on bonds because the period is generally characterized as one of lowering inflation expectations and lowering interest rates. For example, since 1991, the average return on long-term government bonds over the last 10 years is 76 percent above average per year. DFA at 20-21. During this same period, the long-term contingency fund established by Barrick has earned on average 5.73 percent. The fact that this long-term trust fund earned only 5.73 percent over this highly favorable period indicates that the 8 percent proposed by BLM is not a reasonable or “conservative” assumption for long-term performance. During this period, inflation has averaged 2.7 percent, so that trust fund has earned 3.03 percent after inflation during a period of exceptional economic returns and declining inflation expectations. The 3.03 percent earned by the mining trust fund during this highly favorable period is less than half of the return rate proposed by BLM.

EPA understands that the high nominal return proposed by Newmont is based upon the assumption that the trust will be invested 70 percent in equities and 30 percent in bonds throughout the life of the trust. There are many reasons that EPA believes the net present value figure should be based upon investment in low-risk instruments and not equities. First, the underlying assumption that the company will not be financially viable should compel the conclusion that post-closure treatment, like reclamation, will ultimately be performed by a third party paid by the government. Since the government, at that time, will need to rely on the financial assurances to pay for perpetual treatment, the government would invest the trust fund like it does its own funds. Government rules generally do not permit BLM to invest public funds in the equities markets. Second, there are also public policy considerations for the executive branch of government owning a portion of private companies or otherwise controlling investment decisions. Third, stocks are inherently more risky than government bonds, so it is not appropriate to put money intended for the government to use to pay for long-term environmental needs into highly variable investments. Government bonds guarantee a nominal return over the life of the bond with zero default risk. The nominal return on stocks is uncertain. While historic averages indicate that stocks have generally had higher returns over time, it is widely accepted that historic performance is no guarantee of future performance. It is even possible that the growing perception over the last several decades that stocks are not risky over the long-term has in fact reduced the equity premium to a point that stocks will not yield higher risk adjusted returns than bonds. Moreover, stock selection introduces additional risk, such that reliance on historic averages may not provide an accurate indicator of performance even during that time period. These are just a few of the reasons that it is not appropriate to assume that the financial assurances for long-term treatment of groundwater problems will be invested in equities.

Newmont has stated orally to EPA that it believes the 6.5 percent real return is conservative because it is less than the 7 percent real discount rate discussed in OMB Circular A-94. That document applies to cost-benefit analysis for government planning, but does not answer the question presented here – namely if money is set aside in a trust fund and invested, what is an appropriately conservative approach for conducting a discount analysis. That document does not take into account taxes and investment fees or the risks introduced by mandatory withdrawal from a trust invested in variable instruments or stocks. In addition, the discount rates for government planning activities take into account factors other than investment returns, such as opportunity costs of funds to the federal government and other policy considerations. For these reasons, reliance on that circular is not appropriate for this type of analysis.

Even if it is assumed that the trust has some equity exposure, an assumption of 9.8 percent nominal return and a real return of 6.5 percent (as suggested by Newmont) is not a conservative assumption. A lower nominal return rate is appropriate, particularly given a mix between stocks and bonds and the need to be conservative in establishing the trust fund. For example, nuclear decomissioning trusts have been established pursuant to statute. Those trusts invest in a range of investments. A survey of Nuclear Decommission Trust Sponsors indicates an implied real rate of return after taxes and fees generally in the range of 1 to 2 percent from the period of 1992 to 2000, an exceptional period for stock and bond performance. See Nuclear Decommission Trusts – 2000 Survey of Trust Sponsors at 5, NISA Investment Advisors, L.L.C. 2000 (nisanet.com).

Recommendation: EPA believes that BLM should rely upon the types of investments listed in the BLM 3809 Regulations for reclamation funds to calculate the net present value of long-term water treatment operations. Those low-risk instruments provide a higher degree of assurance that the funds will be available to pay for treatment when necessary. Those instruments would be expected to have a nominal return in the range of 5 to 6 percent.

BLM has both understated the rate of inflation and overstated the expected return on the fund. In addition, the approach does not take into account other factors that cause BLM to understate the amount necessary to fund the treatment activities. More appropriate assumptions based upon historic rates and currently available investment instruments would be 4 percent and 6 percent for inflation and nominal return rate. These factors lead to a more realistic real return rate projection of 2 percent after inflation. Taxes, fees, and an appropriate statistical model should also be considered and addressed by BLM.

f. The Unreasonably Optimistic Real Return Rate Significantly Underestimates Site Costs.

The effect of using more realistic economic assumptions leads to a profound increase in the amount needed to fund the post-closure activities. The funds needed at the start of the project depend on the factors identified above (the projected annual costs, the nominal return rate, the inflation rate, costs and fees and variability of those variables). The economic variables are combined to produce a discount rate to calculate the net present value of the future expenses. If one uses a fixed assumption for the discount rate and future costs, it is possible to predict the amount needed to fund the expected costs. Additional analysis is necessary to account for the variability of these factors, but this analysis provides a reasonable starting point. The table below shows the amounts that would be needed in the Long-Term Trust Fund to fund the post-closure activities at various discount rates. For purposes of showing the growth rates over time, the table shows the amount that would be necessary at these rates at the start of the project (2002 NPV Base Case) and the amount that would be necessary at these rates at year 20 (2022 NPV Base Case) and project year 60 (year 2062).

Year	2002	2022	2062
Real Return (net inflation, taxes and fees)	NPV Base Case	NPV Base Case (nominal dollars with 4% inflation)	NPV Base Case (nominal dollars with 4% inflation)

0.0%	$261,966,240	$565,847,078	$857,775,099
1.0%	$87,683,850	$184,593,755	$498,552,323
2.0%	$33,530,301	$84,285,567	$324,687,170
3.0%	$14,362,568	$43,034,308	$231,971,267
4.0%	$6,741,581	$24,036,682	$177,702,108
6.5%	$1,362,815	$7,451,477	$110,588,096
Proposed Surety Amount	$1,000,000	$1,000,000

• Monitoring starts in year 30 and costs $230,000 per year (including capital and O&M expenses).

• Treatment starts in year 60 and costs $1.9 million per year (including capital and O&M expenses).

• These figures are generally based on the BMG estimate, but include a contingency and markup by a factor of 3.27 based on more standard cost estimating procedures for a project of this type.

• The estimate also assumes that capital costs are amortized over the O&M period. For simplicity, we assumed that capital expenses add 10 percent to monitoring expenses and 20 percent to treatment expenses.

• Does not take into account standard deviation risks and variability, which would tend to increase the funding requirements.

• Costs are projected only through first 180 years. Perpetual model would increase the start amount by 2.5 percent (at a 3 percent discount rate) to 40 percent (at a 1 percent discount rate).

The table shows that, at a real return of 2 percent, the company should place $33,530,301 into the trust fund at the start of the project. This amount should grow to $84,285,567 by year 20 based on those assumptions. If those rates of growth and inflation continue forever and if no treatment is required until year 60, these funds would be just enough to pay for the projected treatment (using the EPA-adjusted figures shown in the table). The table also shows that the additional $1 million surety proposed by BLM would be only a small fraction of the additional amount needed by year 20. At that time, BLM would be required to ask the mining company for more than $80 million dollars, but it will have very little leverage to do so.

Although the figures shown above are significantly higher than the BLM projections, EPA believes that even the estimates outlined in the table may be overly optimistic, and that a risk remains that the project will not be fully funded. For example, a critical assumption of BLM is that the pollution problem created by this mine will not require any treatment for 60 years, by which time the mining will be complete, and the mining company established to perform this mining may no longer exist. The assumption that 60 years will pass before treatment is needed is based on a groundwater model developed by the mining company. We have not had the opportunity to conduct a rigorous review of the groundwater or geochemistry models, but will assume the modeling assumptions are reasonable for this stage of planning. Given large gaps in the data at this pre-mining stage, however, adding contingencies into the cost estimate would be appropriate. As stated previously, the groundwater modeling referenced in the Final EIS assumed a porous medium even though the bedrock in the project area is highly fractured and faulted. This simplified assumption introduces significant uncertainty into the assessment, and cannot be considered to be a conservative assessment. In our experience it is not uncommon for models to be overly optimistic about the timing of such groundwater pollution problems. For example, modeling conducted in 1999 for the McCoy Cove mine, near the Phoenix project, projected a pit lake concentration of 120 milligrams/liter (mg/L) sulfate. The most recent monitoring report indicates 1,250 mg/L sulfate in the pit lake after one year of filling. This was due to an underestimation of the oxidation rate of the pit area.

In the case of the Phoenix Mine, BLM acknowledges that the timing and magnitude of the projected groundwater impacts are uncertain (Final EIS, p. 3.2-55). If the assumption that groundwater impacts will begin in year 60 errs by only 15 years (i.e., impacts begin in year 45), the amount needed in the Long-Term Trust Fund by year 20 increases by 33 percent (at 2 percent return) to more than 150 percent (at 6.5 percent return). This uncertainty should be evaluated and accounted for in the analysis.

These points have been raised with BLM on several occasions. In response, BLM will monitor the trust fund and require increases as appropriate. While we agree that monitoring field conditions and economic performance of the trust fund is important, as discussed in section 4 below, the approach suggested does not address EPA’s concerns, particularly given the amount of time that will pass before such information becomes available and the significant financial shortfall that may exist by that time.

Recommendation: (1) Prior to signing a Record of Decision, (ROD), BLM should perform additional analyses to determine the appropriate levels of funding for the Long-Term Trust Fund. EPA offers to assist BLM in this analysis. This analysis allows for a more meaningful dialogue about the required funding for the trust. We recommend the model examine the effect of differing start dates for water treatment. This approach should lead to a more appropriate figure for the Long-Term Trust Fund. A shift in only 15 years of the start date (from year 60 to year 45) results in BLM holding less than half the financial assurances needed to pay for long-term treatment, even if all of the other BLM assumptions are correct (such as the real return of 6.5 percent).

(2) If BLM does not adjust the initial financial assurances before issuing the ROD for this project, EPA recommends that BLM commit in the ROD to engaging an independent third party to complete a full analysis of the costs and financial predictions associated with BMG’s ability to implement the Contingent Long-Term Ground Water Management Plan before the project’s first triennial review. The reviewer should have appropriate expertise in long-term engineering design, construction, and O&M costs and long-term financial planning. Information provided to the reviewer should include EPA’s cost and financial estimate, included in this attachment, as well as information from BLM. BLM should circulate the reviewer’s report to interested parties, including EPA, for review and comment. The ROD should include a commitment to require a timely increase to the trust fund to the amount determined necessary by the third party.

Critical information regarding the adequacy of the Long-Term Trust Fund to cover perpetual treatment costs has not been provided. Specific details of the Long-Term Trust Fund are critical to determining whether sufficient funds will be available to implement the Contingent Long-Term Groundwater Management Plan in perpetuity. These include: (a) requirements for timing of payments into the trust fund; (b) how BLM ensures that the trust fund is bankruptcy remote; (c) acceptable financial instruments (such as those specified in 43 CFR 3809.555); (d) legal structure of the trust for tax purposes; (e) who will pay the taxes on trust earnings and trust fees and expenses; (f) how taxes and trust fees will be paid on the trust if the mining company goes out of business; (g) who will make investment decisions if the operator is no longer viable; (h) if the federal government controls the investment decisions, what legal and ethical issues arise from BLM controlling investment decisions about investments in private companies, voting stock and similar issues if the trust owns stock; (i) the identity of the trust fund beneficiaries; and (j) the identity and corporate structure of the operator with responsibility/ liability for financial assurance at this site. The level of detail provided by BLM to date does not allow EPA to conclude the financial assurances will be satisfactory. We have requested details on these aspects of the project [Attachment 2], but have not yet received them from BLM [Attachment 3].

Recommendation: Prior to signing a ROD, BLM should address each of the points above to enable appropriate review of these important issues. BLM should also ensure the Long-Term Trust Fund and its earnings are established to ensure that the corpus and earnings are bankruptcy remote. The federal government should have a perfected security interest in the trust and its earnings.

BLM has stated that if monitoring indicates that the modeling underestimates groundwater flow or leachate concentrations beneath the site, they will require BMG to increase the Long-Term Trust Fund by an appropriate amount. We agree that this is an important commitment and should, therefore, be included in the ROD. In addition, BLM should require Newmont to act as guarantor of these financial assurance amounts so that BLM could look to Newmont if BMG is unable to meet its financial assurance obligations. We understand that BMG is a wholly owned subsidiary of Newmont, and that Newmont believes that it would be liable for environmental problems created by its wholly owned subsidiary. This understanding and commitment should be documented in the ROD and other documents.

The ROD should include these critical details of the trust fund once BLM has determined how these factors will affect the ability of the trust fund to cover the costs of perpetual monitoring and treatment at the mine.

BLM’s wait-and-see approach to adjusting the Long-Term Trust Fund does not assure adequate funds will be available to cover the costs of implementing the Contingent Long-Term Groundwater Management Plan in perpetuity. The primary response of BLM to EPA comments has been that BLM can, over the course of the project, adequately adjust the cost estimates and require the mining company to adjust the Long-Term Trust Fund, if need be, based on its proposed annual review of groundwater monitoring data and triennial review of trust fund status. There are several problems with this approach.

First, monitoring technical or economic data will not address the problems raised by EPA above. Second, if the data indicate that the cost estimate is too low, those data may not become available for several decades, at which point BLM may have no effective way of securing those funds from the mining company. Third, all costs of this project, including the cost of long-term treatment, must be evaluated before the project begins so that the mining company and BLM can make an informed decision about whether the project can proceed on a cost-effective basis, or whether an alternative (such as improved reclamation) should be reconsidered. BLM has a responsibility to consider whether the project would be economically viable, based on all potential mining and post-mining costs. These points are further discussed below.

a. Monitoring technical or economic data will not address the problems identified by EPA.

We agree it will be important to monitor the trust fund to ensure that an appropriate amount of funding is provided, but monitoring technical conditions will not solve the problems identified by EPA in a timely manner. Even without adverse monitoring results, we believe the trust is seriously underfunded, based on the treatment costs outlined above. If monitoring revealed acid generation to be more serious than predicted by BLM’s modeling, the cost estimate would be even higher than the current EPA projections.

Similarly, monitoring the performance of the trust fund investments will not remedy the problems identified here because of the long-term nature of this project. For example, EPA has determined that, in order to cover the costs of implementing the Contingent Long-Term Ground Water Management Plan, the Long-Term Trust Fund should grow to approximately $84 million by project year 20. The need for $84 million in the trust fund at year 20 is based on the conditions that will exist from year 20 to 180, not on the conditions that exist in the first 20 years. That is, if the first 20 years are a very good period for fixed income instruments, during that period the fund may earn an above average real return of 5 percent due to unusually favorable economic conditions. In that case, the mining company may conclude that an amount much less than $84 million is needed in the trust account at year 20. However, if economic conditions return to average conditions from year 20 to 180, the trust fund would still require $84 million at year 20 to pay the expected costs starting in year 60, and the failure to have those funds in the trust would result in a significant shortfall.

Another factor that has not been adequately considered, and cannot be resolved through monitoring the fund, is the cyclical nature of economic returns. Favorable conditions are often followed by poor economic conditions due to the cyclical nature of the economy. Over the past ten years, fixed income instruments have exceeded historic real returns by a meaningful amount due to declining interest rates and declining inflation. However, we are now in a low-interest rate environment with below average inflation expectations. When inflation and interest rates rise, long-term bonds purchased today will decrease in value and real return. For example a long-term bond yield of 5 percent would earn zero real return if inflation increased to 5 percent.

For these reasons, among others, monitoring the economic performance of the trust fund as proposed by BLM will not remedy the problems identified here.

b. The wait-and-see approach suggested by BLM will make it very difficult to secure appropriate financial assurances in the future.

The BLM model predicts an extremely long period before environmental problems materialize. As a consequence, it is likely that monitoring data that supports a change in the cost estimate will not be available for at least several decades. In fact, most of the EPA technical comments relate to the cost estimate for long-term treatment. BLM currently predicts that the design for the treatment remedy will not be developed in detail until year 60. By that time, we calculate that the current cost estimate may require an increase of more than $500 million. Rather than face that potential shortfall, we believe it is appropriate and prudent to require the funds be put aside today so that they can grow to an appropriate level when they are needed.

Similarly, monitoring data may indicate that the groundwater model is overly optimistic in terms of timing for treatment or the nature of the contaminants to be treated. In that case, a meaningful increase in the treatment costs could be justified. The BLM model suggests that data of this type would not be available for several decades. By that time it may be financially infeasible for the company to supplement the trust fund.

As discussed above, deferring payments for several decades would require very large sums to be paid by the company. We do not believe BLM would have an effective way to require the company to increase the Long-Term Trust Fund at a distant future date.

Adjustments to the fund or the surety could not be made if the company is not financially viable in the future. It is unclear how BLM could effectively enforce such an increase to the Long-Term Trust Fund in excess of the $1 million surety (a) during project operations, (b) after project closure, and (c) if BLM discovered the need for such increases after the operator went bankrupt. Generally the only asset of a mining company such as BMG is the mine itself, which over time can become a liability, not an asset. The parent corporation will generally attempt to shield itself from liability through corporate walls. Unless the documents require a corporate guarantee in addition to all other financial assurances, there is a significant risk that the parent Newmont would have no long-term obligations at this site. Even a corporate guarantee from Newmont does not ensure Newmont’s continued financial viability. BLM and Newmont have not provided any information regarding this issue.

BLM stated that after BMG has met its reclamation obligations at each phase, BLM can transfer any unused portion of the reclamation bond into the Long-Term Trust Fund if the fund needs to be increased. We have four fundamental issues with BLM’s approach. First, the purpose of the reclamation bond is to cover the cost of reclamation in the event that the operator goes bankrupt and cannot fulfill its reclamation obligations. The reclamation bond must be relied upon for this function and should not be relied upon for any other function. Second, even if the operator reclaims the site, there may not be any unused portion of the bond remaining. Third, it is unclear how BLM would be able to keep any portion of the bond if the area meets water quality standards for one year without needing additional treatment that year, and given that BMG will have established a long-term trust fund. The preamble to the final rule regarding 43 CFR 3809.591 states:

Paragraph (b) states that BLM will release up to 60 percent of a financial guarantee for a portion of a project area when BLM determines the operator has successfully reclaimed that portion of the project area. Paragraph (c) states that BLM will release the remainder of the financial guarantee when we determine the operator has successfully completed reclamation, if the area meets water quality standards for one year without needing additional treatment or if the operator has established a long-term funding mechanism under Sec. 3809.552(c).

Fourth, BLM has stated that it would be unable to refuse release of, and transfer to a long-term fund, a surety (i.e., which is not cash, certificates of deposit, or letter of credit). It is unclear whether BLM could require BMG to post a significant portion of its reclamation bond in cash, certificates of deposit, or letter of credit.

EPA understands from its discussions with Newmont that it believes that BLM cannot transfer the financial assurances for reclamation to the post closure trust fund, since Newmont believes that BLM is obligated to release the financial assurances once the tasks defined as reclamation are complete.

Recommendation: BLM should commit in the ROD to a process to ensure that additional payments will be made into the Long-Term Trust Fund should BLM annual and triennial reviews reveal the need for additional funds. The ROD should identify specific triggers and standards that would require the owner/operator to increase the Long-Term Trust Fund, or allow for BLM to transfer money from any unused portion of the reclamation bond into the trust fund. BLM should clarify in the ROD its interpretation of bond release requirements under 43 CFR 3809.591 and indicate whether BMG would be required to post a portion of its reclamation bond in cash, certificates of deposit, or letter of credit, and what portion of the bond that would be required to be.

c. The wait-and-see approach prevents a full analysis of whether project is cost effective.

EPA also believes that the wait-and-see approach proposed by BLM misses an important opportunity to ensure that all project costs are internalized to the company before a commitment is made to proceed with the project. All costs of mining, including the cost of long-term treatment, should be evaluated before the project begins so that the mining company and BLM can make an informed decision about whether the project can proceed on a cost-effective basis. This approach could lead BLM or the project proponent to evaluate whether an alternative approach such as improved reclamation should be reconsidered. If a thorough analysis of the long-term costs is deferred until additional data become available, BLM misses its opportunity to make a fully informed decision about the project and its economic viability.

Recommendation: For the reasons stated above, it is critical to establish appropriate levels of financial assurances now and not rely on the need to increase the amounts at a later date.

In our comments on the Draft EIS, we raised several concerns regarding proposed reclamation and closure of the Phoenix Mine. These issues should be resolved before BLM issues the Plan of Operation. The ROD should include new reclamation measures that have been added to the Phoenix Mine proposal since publication of the Final EIS.

The Final EIS does not address EPA’s concern that runoff from the cap would exceed drinking water standards. Response 1-19 in the Final EIS discusses how the elevated metals in soil would affect plant growth. EPA remains concerned that the stormwater runoff from the capped waste rock will exceed water quality standards for drinking water or other uses and that there will not be adequate cap rock of suitable quality for the Phoenix project.

The Final EIS (Response 1-17) states that based on the extraction schedule, sufficient non-acid-generating waste rock for capping requirements will be available throughout the life of the project. However, according to the Final EIS (p. 3.2-36), “[t]he results of the kinetic tests indicate that most of the rocks in the project area directly associated with mining operations (pits and waste rock) have the potential to generate acid rock drainage.” We remain concerned that net neutralizing rock will not be available in sufficient quantities throughout the project to provide a five-foot cap on site facilities.

The Final EIS lacks sufficient characterization of the borrow source to ensure that this source will provide an acceptable cap amendment with respect to geochemistry and geotechnical properties.

Recommendation: BLM should calculate the volume of cap material necessary to prevent exceedences of water quality standards in runoff, identify the volume of borrow material that may be needed, and bond for the additional amendment, including additional testing and transportation costs. This information should be included in the ROD.

The borrow source should be thoroughly characterized so that realistic cost adjustments to the reclamation bond, including costs of importing material from other sources, can be made prior to the ROD. This information should be included in the ROD.

The National Environmental Policy Act (NEPA) establishes our national environmental policy and goals for the protection, maintenance, and enhancement of the environment. Based on EPA’s statutory responsibilities under the Clean Water Act and the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA), EPA believes that BLM’s proposal does not meet the intent of NEPA and we continue to believe that the Final EIS is inadequate. If the proposed mitigation is significantly underfunded and, therefore, infeasible, off-site groundwater contamination would not be controlled, and the Federal government would inherit an enormous financial burden to mitigate the problem. BLM’s proposal does not include the use of all practicable financial assurances to promote the general welfare. To fulfill its NEPA obligations, BLM should require BMG to provide adequate financial assurance that the Contingent Long-Term Ground Water Management Plan will be implemented when necessary after mine closure.

NEPA requires that all relevant information concerning environmental impacts be disclosed to the public before decisions are made and before actions are taken. 40 CFR 1500.1 (b); see also 40 CFR 1505.1(d), which requires that relevant documentation accompany a proposal through existing agency review processes so that agency officials use the statement in making decisions, and 40 CFR 1506.6(f), which requires documents underlying EISs to be made public. Agencies shall ensure the professional integrity of the discussions and analyses, and identify any methodologies used and shall make explicit reference by footnote to the scientific and other sources relied upon for conclusions in the EIS. 40 CFR 1502.24. Moreover, no material may be incorporated by reference unless it is reasonably available for inspection by potentially interested persons within the time allowed for comment. 40 CFR 1502.21.

We believe the information we requested in our Draft EIS comment letter is relevant and required for several reasons. First, the viability of the post-closure plan is a critical factor in whether this project may be considered environmentally acceptable. Second, EPA believes this information is essential for an adequate analysis of the proposed project, because it could make the difference between a project sufficiently managed over the long-term by the site operator, or an unfunded/under-funded contaminated site that becomes a liability for the Federal government, e.g, under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). Third, the economic viability of the proposed project, including the cost of long-term treatment, should be evaluated before the project is authorized so that BMG and BLM can make an informed decision about whether the project is cost-effective.

The Final EIS presented a very general summary of the long-term cost estimates and the Long-Term Trust Fund. BLM provided EPA with BMG’s Cost Estimate and a December 14, 2001, letter from Newmont describing in outline form the Long-Term Trust Fund. However, BLM did not release this information to the public until after the close of the comment period. The information supported by these documents forms the basis of BLM’s position that the proposed Long-Term Trust Fund will be adequate to implement the Contingent Long-Term Ground Water Management Plan. This information may be used by BLM to select a project alternative and to determine whether the action will meet environmental policies, regulations, and standards. EPA believes that by not releasing these documents publicly during the public comment period, BLM did not meet NEPA's requirements that all relevant information that forms the basis for an agency's decision be disclosed to the public for review and comment.

Recommendation: BLM should prepare a Supplemental Draft EIS and make all relevant information available to the public during a public comment period, in accordance with NEPA and CEQ’s NEPA Implementation Regulations at 40 CFR 1502.9(c).

D. THE SURFACE MANAGEMENT REGULATIONS FOR SURFACE MINERAL OPERATIONS AT 43 CFR 3809

Historically, mining projects have resulted in the expenditure of billions of dollars by the government for environmental cleanups. There are many examples of large and well capitalized mining companies going bankrupt before their responsibilities for environmental cleanups could be satisfied. In light of this history, BLM recently finalized a new provision in its Surface Management Regulations for Surface Mineral Operations at 43 CFR 3809 authorizing BLM, when it identifies a need for it, to require operators to:

Establish a trust fund or other funding mechanism available to BLM to ensure the continuation of long-term treatment to achieve water quality standards and for other long term, post-mining maintenance requirements. The funding must be adequate to provide for construction, long-term operation, maintenance, or replacement of any treatment facilities and infrastructure, for as long as the treatment and facilities are needed after mine closure. BLM may identify the need for a trust fund or other funding mechanism during plan review or later. [43 CFR 3809.552(c)]

The purpose of this new provision is to protect the public from liability for the cleanup, and BLM has identified the need for such a fund to be established for the Phoenix Mine. As explained above, however, the funding currently proposed for the Long-Term Trust Fund would not be adequate to provide for construction, long-term operation, maintenance, and replacement of treatment facilities and infrastructure, for as long as the treatment and facilities are needed after mine closure. Moreover, the mechanisms BLM proposes to use to adjust the Phoenix Mine Long-Term Trust Fund over the life of the project are inadequate to assure that the Contingent Long-Term Ground Water Management Plan would be implemented.

The Phoenix Mine will be the first hardrock mine that, because of anticipated acid mine drainage, will include post-closure financial assurance under the new rule. BLM’s approach in calculating long-term costs and determining the mechanics of the Phoenix Mine Long-Term Trust Fund will be an important precedent as BLM demonstrates how it will apply the new rule. There are innumerable mines across the country that have created perpetual pollution problems. This issue is likely to arise at many other mines on federally managed lands. Given the magnitude of the Phoenix Mine and its anticipated problems, BLM’s approach to projecting costs and establishing a financial assurance instrument for the Phoenix Mine is not based on conservative estimates and significantly understates the Federal government’s liabilities.

Recommendation: EPA would like to assist BLM in developing a process for developing accurate post closure costs and acceptable trust fund mechanics, based on our extensive work on establishing trust funds at mining Superfund sites.