8. Residential_Commercial Cleanup Action Level Clearance Crieteria - December 2003.pdf

LIBBY ASBESTOS SITE

RESIDENTIAL/COMMERCIAL CLEANUP

ACTION LEVEL AND CLEARANCE CRITERIA

TECHNICAL MEMORANDUM

DRAFT FINAL - December 15, 2003

Prepared by:

U.S. Environmental Protection Agency, Region 8

999 18th Street, Suite 500 Denver, CO 80202

With Technical Assistance From:

Syracuse Research Corporation

999 18th Street, Suite 1975 Denver, CO 80202

This page intentionally left blank.

DRAFT FINAL - December 15, 2003

LIBBY ASBESTOS SITE

RESIDENTIAL/COMMERCIAL CLEANUP

ACTION LEVEL AND CLEARANCE CRITERIA

TECHNICAL MEMORANDUM

I. INTRODUCTION

In 2002, the U.S. Environmental Protection Agency Region 8 (EPA) began systematic investigation and emergency response cleanup of residential and commercial properties in Libby, Montana. The Action Memorandum Amendment dated May 8, 2002 (“the Action Memo”) set forth general requirements and reasons for the emergency response cleanup (EPA 2002).

This document provides specific information about “action levels” that will be used for determining which properties or situations require an emergency response cleanup. It also provides specific information regarding “clearance criteria” that will be used to determine when such a cleanup is complete. The Appendix presents screening-level risk calculations that EPA has developed to help characterize the relationship between asbestos levels in site media and the resultant levels of health risk to residents. Although these relationships are uncertain and difficult to quantify, the calculations provide a frame of reference that helps guide decision-making at the site.

In this memorandum, the term “cleanup” is used generally to imply some type of response action and does not necessarily imply removal of contaminated material. In some instances, EPA’s response action will be isolation or encapsulation of contaminated material. In some cases, where contaminated material is difficult to access or well-contained, and exposure is likely to occur very infrequently or not at all, the material may be managed in place. Details of EPA’s cleanup approach can be found in the Response Action Work Plan (CDM 2003a).

II. REGULATORY PROCESS CONSIDERATIONS

EPA is currently conducting emergency response removal actions in Libby. This document sets forth action levels and clearance criteria that are applicable to these emergency response actions only. While the emergency response continues, EPA is also conducting a Remedial Investigation/Feasibility Study (RI/FS). The RI/FS will conclude with the development of a proposed plan, extensive opportunities for public comment, and publication of Record of Decision (ROD) that will set forth final action levels and cleanup decisions for Remedial Actions at Libby. Until a ROD is published, it is expected that this memorandum will guide decision-making for emergency response cleanup actions at residential and commercial properties in

Libby.

The RI/FS for Libby residential and commercial properties began in 2002. The first phase of the RI was called the Contaminant Screening Study (CSS). The CSS was a screening step intended to collect readily available information (through inspections, verbal interviews, and soil sampling) to be used in decision-making for each property. The objective was to classify properties as either (1) requiring emergency response cleanup, (2) requiring more investigation before a decision can be made, or (3) likely requiring no further action. The criteria in this memorandum, along with other information, are being used to interpret the results of the CSS and in planning for additional sampling and cleanup. Preliminary results of the CSS (soil sample results are not yet available) are found in the CSS Draft Final Technical Memorandum (CDM 2003b). The Technical Memorandum will be updated and expanded when soil sample results are available. As of this writing, more than 1200 properties have been identified that likely require cleanup. The RI/FS is expected to be complete in approximately 2005 at which time the total number of properties requiring cleanup will be known.

Again, final site-specific cleanup standards will be established upon completion of the RI/FS and publication of a ROD. Similarly, the site-specific decision-making approach, action levels, and clearance criteria set forth here may be changed upon receipt of new information. This has two important implications. First, some properties will not meet any of the criteria for emergency response, but may meet lower or different criteria that are established in the future. In such an instance, these properties would be addressed later. Second, it is possible that properties that are earmarked for cleanup based upon the criteria set forth here may not require cleanup if different criteria are established in the future.

III. SAMPLING APPROACH

Throughout this memorandum, there are general references to sampling and analysis methods. A detailed discussion of the many sampling approaches, analysis methods, counting methods, sample preparation methods, and quality assurance steps is beyond the scope of this document.

• Details of EPA’s dust sampling and analysis protocol for various situations can be found in the Indoor Dust Sampling and Analysis Plan (EPA 2003).

• Details of EPA’s air sampling and analysis protocol for various situations can be found in the Response Action Work Plan (CDM 2003a).

• Detail s of EPA’s soil sampling strategy for various situations can be found in the

< CSS Sampling and Analysis Plan (SAP) (CDM 2002) < CSS SAP Revision I (CDM 2003b) < Remedial Investigation SAP (CDM 2003c) < Response Action Work Plan (CDM 2003a)

It is also important to note that EPA’s sampling and cleanup program is based solely upon the presence of “Libby asbestos.” Libby asbestos (LA) is a form of amphibole asbestos unique to the Libby vermiculite deposit and is fundamentally different from more commonly found chrysotile asbestos. Chrysotile asbestos was used in commercial products for decades and is found throughout the environment of the U.S. and the world. EPA will not base cleanup decisions or take action based upon the presence of chrysotile asbestos not associated with the Libby mine, except where necessary to protect worker safety or where it may present an imminent and substantial endangerment.

IV. CLEANUP APPROACH AND DECISIONS REQUIRED FOR EACH PROPERTY

To date, much of EPA’s investigation and cleanup approach has been geared toward finding and addressing sources of LA. The major sources in the area, such as the mine, the screening and export plants, and large vermiculite piles, have already been isolated or cleaned up. Remaining sources at residential/commercial properties generally include (but aren’t necessarily limited to) vermiculite insulation or soils with elevated levels of LA. Sources, through a variety of mechanisms, can serve to contaminate indoor dust and have the potential to release significant amounts of LA when disturbed. Source removal or isolation ensures that loading to indoor dust (one of the most significant exposure pathways over a lifetime) is minimized immediately. With respect to this, each property in Libby generally requires three independent decisions regarding cleanup:

• Cleanup of the attic or interior walls (ATTICS/WALLS)

• Cleanup of indoor living space (INTERIORS)

• Cleanup of outdoor soils (SOILS)

Contamination in one area or bulk source does not automatically imply contamination in another.

For instance, a particular property may require cleanup of an attic but not the interior or outdoor soils. Some properties may require cleanup in the attic, interior, and soils. Any combination is possible depending upon the unique conditions of each property.

EPA’s cleanup approach considers not only the presence of source materials and the concentration of LA within them, but also the likelihood that these source materials may be disturbed. Based upon this approach, some source materials that are less likely to be disturbed may be left in place (such as in walls, below hard surfaces, and at depth). In some situations, EPA may remove or further isolate such materials to prevent even infrequent exposures, depending on the situation.

For those properties requiring emergency response cleanup, EPA is adopting cleanup procedures and criteria that will that help ensure we conduct only one cleanup action at individual properties, even if action levels are lowered or changed in the future. It is obviously inconvenient, impractical, and costly to clean a property twice. This approach is cost effective and protective.

V. ACTION LEVELS

Any one of the following conditions will generally trigger emergency response cleanup for that portion of the property.

ATTICS/WALLS

• Visual confirmation of open, non-contained, or migrating vermiculite insulation

INTERIORS

• Visual confirmation of vermiculite in the indoor living space

• Concentration of LA in an indoor dust sample greater than 5,000 Libby asbestos (LA) structures per square centimeter (s/cm2) using AHERA counting methods. This will be referred to as 5,000 AHERA s/cm2.

Each level, or floor, of a building is evaluated and sampled separately. At least two samples are collected from each floor. Libby sampling data has shown that in many cases, only one floor is highly impacted (e.g. material tracked in from outside on the ground floor). This sampling and cleanup approach allows us to focus cleanup resources on the portion of the interior where the greatest problem exists.

SOILS

• Visual confirmation of vermiculite or other vermiculite mine related materials in “specific use areas.” A specific use area is defined as a garden, former garden, planter, or other defined area of a yard likely to receive significant use and generally not covered with grass.

• Concentration of LA in specific use areas or other yard soils by any analytical method greater than or equal to 1% Libby asbestos.

VI. ACTION LEVEL RATIONALE AND DISCUSSION

ATTICS/WALLS

Based upon available information, EPA has determined that vermiculite insulation found in Libby is a potential source for current and ongoing exposures to LA. Past sampling by EPA in Libby has clearly shown that while LA concentrations in bulk vermiculite insulation may vary considerably (presumably even within the same home), all or most vermiculite insulation has the ability to release LA when disturbed and that disturbance can lead to excessive risk (EPA 2001).

Given the collective magnitude of exposures in Libby, EPA has determined that visual confirmation of open, non-contained, or migrating vermiculite insulation is sufficient justification for emergency response cleanup. Cleanup is not contingent upon other factors or information such as the concentration of LA in bulk insulation, volume of insulation, condition of the building, or LA dust levels in the building, but such factors may be considered when determining how quickly EPA responds and how extensive the cleanup will be.

INTERIORS

Exposure to contaminated indoor dust, even dust with a relatively low level of LA, is an important exposure pathway. This is because people spend most of their lives in their homes and exposure occurs continually. However, indoor dust is a secondary medium - it can only become contaminated through disturbance of some other source of LA. Such sources may include , but are not limited to, vermiculite insulation, on-property soils, off-property soils, or past vermiculite processing operations. Again, the most important step to break this pathway is to address the sources that are contaminating indoor dust or have the potential to contaminate indoor dust in the future. In Libby, EPA is not relying upon measured dust levels to decide if residential/commercial sources must be addressed. Our approach is to find and address sources with the potential to contaminate indoor dust regardless of current indoor dust levels. In this regard, indoor dust action levels should not be considered triggers for overall cleanup, but only a trigger for aggressive interior cleaning by EPA. This approach ensures that situations that may present a short-term exposure hazard are addressed as quickly as possible.

EPA has established two key “action levels” relating to interior cleanup: presence of visible vermiculite in the indoor living space and indoor dust samples that contain greater than 5,000 AHERA s/cm2. During the CSS, EPA is visually inspecting interiors for the presence of visible vermiculite, such as insulation that has migrated into the living space from the attic or walls. If vermiculite is observed on a particular level, even in small amounts, cleanup of that entire level is triggered and no dust samples are collected on that level. If vermiculite is not observed on a particular level of a building, dust samples are collected on that level to determine if cleanup is necessary.

Visible Vermiculite

Using visual observation of uncontrolled vermiculite material as a trigger for an emergency cleanup action is conservative, protective, and cost effective. This approach ensures that primary source materials are removed, as well as any associated contamination. It addresses resident concerns regarding vermiculite in the living space. This approach also reduces the costs associated with routine sampling and analysis of interiors.

Indoor Dust Samples

Dust samples will be collected for those structures or levels of structures not containing visible vermiculite to determine if cleanup of those structures or levels is necessary. Unfortunately, establishing action levels based upon indoor dust levels is not straightforward. There are two primary reasons for this:

• Unlike air samples, there are no established regulatory or health-based standards to guide the determinations of acceptable concentrations of asbestos in indoor dust.

• The relationship between the concentration of asbestos in dust and the resultant concentration in air (the medium that actually determines human exposure and risk) is highly variable. This is because the relationship depends on a long list of different factors, most important of which is the nature and frequency of dust disturbance. This means that it is difficult to calculate a value in dust that corresponds to an acceptable level in air, and it is even harder to try to select a level in dust based on site-specific measurements. This difficulty is discussed more in the Appendix.

Given these difficulties, EPA has developed an interim site-specific action level of 5,000 AHERA s/cm2 for interior cleaning. In choosing this value, EPA considered several factors:

• Screening level risk estimates for exposure to indoor dust presented in the Appendix indicate that if a resident were exposed to 5,000 AHERA s/cm2 in dust throughout the home for 70 years, the risk of cancer would likely be in the range of 1 in 100 to 1 in 1,000, much higher than the level of 1 in 10,000 level that EPA usually considers to be the limit of acceptable risk. Although these calculations are screening-level and uncertain (see the discussion in the Appendix), and even though it is expected that the levels in dust will decline after source removal is complete, it is still clear that dust levels this high constitute a potential risk that warrants aggressive interior cleaning to ensure protectiveness and to stop even short-term exposures.

• Aggressive interior cleaning by EPA is costly and cleaning the interiors of all houses in Libby would be very expensive and significantly extend the cleanup duration. Therefore, a reasonable cutoff must be established below which such time-consuming and costly cleanup can be omitted, so that available resources can be directed to source removal and the situations presenting the most risk. This does not mean that dust cleanup at levels below 5,000 AHERA s/cm2 is not appropriate, but only that less aggressive methods may need to be used. Providing each resident with a HEPA vacuum and encouraging frequent vacuuming and wet-wiping surfaces will be most effective. This will accelerate the process of decreasing dust loading levels and will achieve similar success to that of an EPA cleanup at a fraction of the cost and effort.

• Concentrations of 5,000 AHERA s/cm2 in dust appear to be on the lower end of the background levels of asbestos contaminated dusts observed in residences in other communities (see Appendix). While the types of asbestos fibers found in indoor dust in other communities is generally different from LA, this information provides some context regarding the dust concentrations of asbestos that people are being routinely exposed to in their homes and businesses in other areas. The health risks or consequences stemming solely from these routine dust-related asbestos exposures is unknown.

• Dust levels of 5,000 AHERA structures/cm2 can be readily detected using efficient sampling techniques. The sensitivity of the analytical methods EPA is using to quantify asbestos in dust vary based on several factors, but are typically on the order of 1,000 AHERA s/cm2. Detection of asbestos concentrations below this level in dust require more extensive analysis and are much more costly to achieve.

SOILS

EPA has established two key “action levels” relating to outdoor soil cleanup: presence of visible vermiculite or other vermiculite mine-related material in specific use areas and soil samples containing greater than or equal to 1% LA.

Visible Vermiculite

During the conduct of the CSS in 2002, EPA visually inspected many properties for vermiculite or vermiculite mine related materials (generally referred to just as vermiculite) in soils. If vermiculite was observed in a particular area (e.g. front yard, side yard, garden, etc), no soil sample was collected in that area. If vermiculite was not observed, a soil sample was collected from that area. Past observations showed that when visible vermiculite was noted, samples confirmed the presence of Libby asbestos approximately 70% of the time using polarized light microscopy (PLM) (CDM 2002). If more sensitive methods were used, this number may have been higher. Thus, the presence of visible vermiculite was considered a reasonably good indicator for the presence of LA and material that could serve as a potential source of LA to air or dust. Using visible vermiculite as a trigger for cleanup, rather than sampling every area that contained visible vermiculite, had the benefits of being conservative, protective, and simple.

This same general approach of using visual observations as a trigger for cleanup was also employed during the remediation of other large source areas in Libby, such as the screening plant, export plant, and flyway property. It will also be used for the remediation of the rail yard in Libby by Burlington Northern Santa Fe (BNSF Workplan 2002).

While conducting the CSS in 2002, EPA discovered three key points regarding visible vermiculite in soils:

• The number of properties with visible vermiculite in soils was far greater than originally anticipated.

• While there were exceptions to the trend, the amount of visible vermiculite varied considerably from a few flakes over a generally wide area to very concentrated amounts in small areas. The CSS, as originally planned, had no systematic way to account for this or differentiate it other than sampler observations.

• There were several instances where vermiculite was observed in areas that were difficult to access and where exposure was likely to occur infrequently, if at all. There are likely many more of these situations that were not discovered during the CSS that will become apparent through subsequent, more detailed investigations or during cleanup.

Because of these factors, EPA reevaluated the proposed initial approach to visible vermiculite in soils described in the CSS SAP (CDM 2002). Rather than assuming that all occurrences of visible vermiculite would result in cleanup, EPA decided to limit the emergency response cleanup of visible vermiculite to specific use areas that pose the most substantial opportunities for disturbance. There are several reasons for this:

• Vermiculite was generally used as a soil amendment in specific locations such as gardens.

• Specific use areas are more likely to lack ground cover, such as grass, that would minimize creation of dust.

• Specific use areas are likely to be actively and frequently disturbed through activities such as gardening.

• Specific use areas are generally small and can be cleaned up quickly at low cost. A large scale sampling program may not be justified for these situations, considering that for many of these situations (CSS SAP, CDM 2002) sampling will confirm the presence of LA. Cleaning up entire yards, large portions of yards, or areas that are infrequently accessed or disturbed is a much larger and expensive task and additional sampling is clearly warranted.

Soil Samples

Similar to indoor dust concentrations, establishing action levels based upon soil contamination levels is not straightforward. There are two primary reasons for this:

• Unlike air samples, there are no established regulatory or health-based standards to guide the determination of acceptable concentrations of asbestos in soil.

• It is extremely difficult to predict airborne asbestos exposures (which is the exposure of concern for health and regulatory standards) based upon the asbestos concentrations in outdoor soil samples. A variety of factors can influence the extent of airborne exposures associated with asbestos fibers in soil, the most important of which appears to be disturbance of contaminated soil by human activity. Other factors which may affect the suspension of asbestos fibers into the air, and thus airborne asbestos exposures, include the environmental conditions, moisture content of the soil, concentration of asbestos in the soil, the type of the soil, and the characteristics of the asbestos present.

Thus, development of the site-specific information necessary to accurately predict the risk between concentrations of asbestos in soil and airborne exposures and secondary contamination of indoor dust, if even possible, would require an extensive sampling effort that included numerous outdoor areas under various test conditions and scenarios. Such investigations would be extremely difficult, costly, and lengthy. Further, considering the magnitude of the cleanup in Libby, such an investigation would divert limited financial resources away from the essential work of cleaning up the worst exposures and conditions present today.

At this time, and in the absence of visible vermiculite in a specific use area, EPA has selected a concentration of 1% LA or greater as a site-specific action level for emergency response soil cleanup. When selecting this level, EPA considered several factors.

• Although there are no standards for acceptable concentrations of asbestos in soil, this standard was applied during previous emergency response cleanups in Libby.

• Screening-level risk estimates for exposures at a home with contaminated soil (presented in the Appendix) suggest that risk levels for a resident living for 70 years at a property with a level of 1% asbestos in the soils are on the order of 1 additional cancer expected per 1000 people. This is well above the risk level of 1 in 10,000 that EPA usually considers to be the upper limit of acceptable risk. Although the calculations are uncertain (see the discussion in the Appendix), these results support the conclusion that removal of soils that contain 1% or more asbestos is needed to protect public health.

• Inexpensive analytical methods currently available (e.g. PLM) can detect levels of 1% or greater with some confidence. Site-specific improvements in the use of PLM analysis at

Libby have led to much higher confidence in sampling results and the ability to detect and quantify asbestos levels in soils at 1% and even less than 1%. EPA is currently testing several methods to determine their ability to detect and quantify levels less than 1%.

It is important to note that EPA does not assert that soil concentration of less than 1% LA are necessarily safe or acceptable, and in certain circumstances, soils with less than 1% Libby asbestos may be remediated under the current emergency response program. Depending on the accessibility and frequency of exposure, EPA may elect to remove or isolate soils containing less than 1% LA. Similarly, if a portion of a property meets either emergency response action level for soils (i.e., visible vermiculite in specific use areas or LA greater than 1%), EPA will remediate all soils at the property with any detectable LA. This is primarily so that properties will not have to re-cleaned later if a lower action level is adopted. This is considered protective and cost effective. The approach, however, is to target properties where this is not the case first.

Soils that meet the emergency response action levels take priority whenever possible. This is also consistent with previous cleanups in Libby.

VII. CLEARANCE CRITERIA

Cleanup of a portion of a property is considered complete and the property “clean” when all of the following site-specific criteria are met.

ATTICS/WALLS

• No uncontrolled visible vermiculite remaining in accessible areas

• Any vermiculite remaining is well-contained

• The average of approximately 5 samples of disturbed air collected in the attic indicate less than .01 AHERA structures per cubic centimeter of air (AHERA s/cm3).

INTERIORS

• No visible vermiculite remaining in accessible areas or living space

• No LA structures are detected in any of approximately 5 samples of disturbed air on the level(s) or floor(s), indicating disturbed air concentrations are generally less than .001 AHERA s/cm3.

SOILS

• No substantial visible vermiculite or waste material remains within the area of excavation

• In excavated areas, soil samples collected at the depth of cut are non-detect for LA by

PLM. If maximum depth of cut is reached (12 inches for yards, 18 inches for specific use areas), soil samples collected at the bottom of excavation must be less than 1% LA by PLM. Clean backfill is then placed over the excavation. This approach ensures that no detectable LA (by PLM) remains to a depth of 12-18 inches, but allows small amounts of LA to remain well below ground surface, where soil is unlikely to be disturbed. More information on the clearance sampling approach for soils is found in the Response Action Work Plan (CDM 2003a).

If these criteria are not met, re-cleaning, additional excavation, or other steps may occur, and the process is repeated. If any situations occur where clearance criteria cannot be met, unique approaches may be considered.

VIII. CLEARANCE CRITERIA RATIONALE AND DISCUSSION

ATTICS and INTERIORS

Attics and interior living spaces are both cleared using an approach based generally upon procedures outlined in the Asbestos Hazard Emergency Response Act (AHERA), but with different final numerical standards to account for the different amount of exposure likely for each. Once physical cleanup is complete, and visual inspection shows that all vermiculite is removed or contained, each individual space (e.g. the attic or particular floor of the home) is blown with a 1 horsepower leaf blower for several minutes. The action of walking through the living space and aggressively blowing dust from all surfaces, effectively simulates a high-end exposure. Following this action, fans are set up in the space to keep the air circulating, and air samples are collected.

EPA considered the use of settled dust samples for a clearance criteria, rather than aggressive air sampling. However, because the property was just cleaned, a settled dust sample would likely not be representative and is not as directly correlated with risk estimates as air concentrations.

The use of aggressive air sampling is also feasible in this situation because the resident is already relocated and a controlled environment is present.

When EPA selected these site-specific clearance criteria, we considered several factors:

• Sampling occurs after the source is removed and is conducted after the dust throughout the space is aggressively disturbed. These conditions will not simulate normal living conditions suspected in the future, but rather approach worst-case conditions. The primary intent of the all clearance sampling is to ensure that sources were effectively addressed, not to demonstrate an expected long-term exposure level.

• Requiring a non-detect for each of five samples in the living space, as opposed to calculating the average of the five samples, increases the protectiveness of the cleanup in the interior living space. Under this scenario, the absolute maximum concentration in any one sample that is possible is less than .005 AHERA s/cm2, but average exposure concentrations across the living space are effectively less than 1/5 the detection limit of a single sample, or less than .001 AHERA s/cm3.

• Screening-level risk estimates for exposures to asbestos-contaminated air presented and discussed in the Appendix suggest that risk levels for someone exposed continually to an air concentration of 0.001 AHERA s/cm3 for 70 years are on the order of 1 additional cancer expected per 10,000 people. EPA generally only takes action if risk estimates exceed a level of 1 additional cancer per 10,000 people. Although the risk calculations are uncertain (see the discussion in the Appendix), the results support the conclusion that the clearance level is protective of human health, especially because the long-term average concentration in air is likely to be lower than that measured following active disturbance.

• Because exposure in attics is likely to occur far less frequently than in main living spaces, higher numerical standards are applied and a mean concentration is used. The clearance criteria for interior living space in Libby are currently about 5-10 times more stringent.

• EPA understands that attainment of the chosen indoor clearance criteria will not necessarily indicate complete removal of all interior LA contamination. It is anticipated that if primary sources of contamination are remediated, then settled dust levels will decline over time. To help ensure that these levels do, in fact, decline, EPA will provide a HEPA vacuum to property owners of all properties that EPA has cleaned. EPA will also provide HEPA vacuums to other property owners whose properties are below the action level and have not been cleaned. Details of the scope of the HEPA vacuum program are found in HEPA Vacuum Program Fact Sheet (EPA 2003b). Furthermore, post-cleanup sampling is being planned to ensure that indoor dust levels remain low or decline as expected. This is discussed more in the final section of this memorandum.

SOILS

Soils requiring cleanup are cleared using an step-by-step approach. Limited excavation of the defined area occurs until no visible vermiculite is observed, or until the native soil horizon is reached. At this point, a representative number of soil samples are collected dependent upon the size of the excavation. If these samples are non-detect by PLM, the excavation is considered complete and the area is backfilled with clean soil. If LA is detected by PLM, excavation continues until the maximum depth of excavation is reached (12 inches for general yard areas, 18 inches for specific use areas). At this point, excavation would only proceed further (generally to a maximum of three feet regardless of contaminant levels) if gross contamination is observed or PLM samples indicate higher levels of LA (e.g. greater than 1%).

When establishing this approach, EPA considered the following factors:

• Nearly all exposure comes from near-surface soils. These soils generate dust and are often actively disturbed. In most circumstances, contamination is also limited to near surface soils. To ensure resources are focused on the soils that are most likely to result in human exposures, a maximum depth of excavation was established at 12-18 inches.

These depths are based on based on the depth that typical residential activities may intrude into the soil (such as planting, rototilling, or installation of sprinklers). Below this depth, disturbance and exposure will occur very infrequently if at all.

• It is important to distinguish that at least two variations of PLM analysis are being used in the Libby residential/commercial investigation and cleanup. A site-specific PLM method, which involves off-site preparation of the sample, is currently being used to analyze surface soil samples (where frequent, ongoing exposure and dust generation occurs). This analysis is used to determine which specific areas require cleanup. Other methods of soil sample analysis are also being considered for these samples. All of these methods are intended to provide lower detection limits than “standard” PLM, which has traditionally been used for analysis of asbestos in bulk materials. However, “standard” PLM, by NIOSH Method 9002, was selected as the analytical method for clearance testing of soils at depth because of it’s ability to recognize soils that exceed 1% LA, and because PLM can be performed on-site with a short turn-around time. This allows real time decisions to be made about whether further excavation is needed, and allows the excavation to be closed as soon as possible. Use of alternative analytical methods that might have lower detection limits than PLM are not feasible because they require off-site analysis and results may not be available for days or weeks. Leaving an excavation open for this time is hazardous and very cost inefficient.

IX. FUTURE ACTIONS AND FOLLOW UP

Based on the information available, EPA has developed an emergency response cleanup program that:

• focuses on elimination of exposures that are likely to occur frequently and continually over time;

• removes nearly all identified LA sources, focusing on sources that are most likely to be disturbed;

• isolates sources that are impractical to remove;

• addresses the highest exposures in the quickest manner possible;

• leaves low residual levels of LA and minimizes the likelihood of future re-cleaning;

• considers the many uncertainties regarding asbestos analysis and risk assessment that suggest risks could be higher than anticipated and employs strategies (from sample collection to cleanup) to help compensate for these;

• reduces future management needs; and

• is protective, cost-effective, and implementable.

EPA recognizes the program does not completely eliminate all potential exposure to LA in Libby. In fact, such a program is impossible to fund or implement. Because of this, future management and review will be required to ensure the long-term protectiveness of the remedy.

Nearly all cleanups require some level of long-term management. At Libby, EPA is already taking several steps to address this issue:

• EPA is providing HEPA vacuums and interior cleaning guidance to residents so they can immediately, and continually, address low levels of residual contamination and any particulate contamination that may be reintroduced into their homes. Guidance on additional steps to take to minimize the potential for exposure to residual asbestos and to increase their confidence in the safety of their home or business will be included in a package of information that residents and business owners receive after cleanup of their property.

• EPA, in conjunction with the Montana Department of Environmental Quality and local government, will develop a long-term operations and maintenance (O&M) plan to deal with future management issues. Key points of this plan will likely include ongoing education, guidance for residents encountering or working with vermiculite in the future, and a management system for any necessary removals of vermiculite including cleanup assistance and disposal procedures at the Lincoln County landfill.

• Most importantly, as part of the Remedial Investigation, EPA will institute a monitoring program for properties that were cleaned up. Not all properties will be visited, but a sufficient number to draw statistical conclusions will be sampled. This monitoring will measure actual dust and air levels, allowing EPA to (1) determine the efficacy of the cleanups after some time has passed, (2) test assumptions that affect the cleanup approach (e.g. Have dust levels declined? Have textiles and carpets that were not removed affected ambient conditions? Have heating and ventilation systems reintroduced contamination?), and (3) provide actual exposure data for use in the baseline risk assessment for the site. Based upon the results of this sampling, it is possible that the cleanup approach and/or criteria may be modified. EPA will also continue to implement other sampling programs to help aid in understanding of Site conditions.

• Using the best information available, a baseline risk assessment will be completed for the site. Using this and other information, final cleanup levels will be established. These will be compared to the measured residual levels at properties where cleanup has occurred, and levels at properties that did not meet the criteria for cleanup, to ensure that all necessary cleanup occurs and that final conditions are protective. The final decisions and actions for the Libby cleanup will undergo extensive public review and comment.

X. POTENTIAL QUESTIONS AND ANSWERS

Question: How do Libby indoor air clearance levels compare to existing regulatory standards?

The current standard for worker protection, the Permissable Exposure Limit (PEL), established by the Occupational Safety and Health Administration (OSHA), is no greater than 0.1 Phase Contrast Microscopy Equivalent (PCME) s/cm3 for an eight hour exposure. While few consider this level protective, and OSHA clearly states that it is not intended to be fully protective, it nonetheless is a current standard which governs worker exposure. The clearance standard for Libby interior cleanings, after accounting for differences in counting rules (AHERA vs PCM), and for differences in assumed exposure frequency and duration, is equivalent to approximately 1/50 this amount. This means that Libby clearance levels are much more protective.

Without going into the details of the Asbestos Hazard and Emergency Response Act (AHERA) clearance protocol for schools, it can be generalized to say the clearance standard for asbestos removals is 70 AHERA structures per square millimeter (s/mm2) on the sample filter. This generally corresponds to an airborne concentration of about 0.02 s/cc. This standard is compared against the average of multiple samples, whereas in Libby interiors we compare each of approximately five samples to our clearance standard. Thus, EPA’s effective clearance criteria at Libby (less than 0.001 AHERA s/cm3) is approximately 20-fold lower than the AHERA standard. Again, Libby clearance levels are more protective.

Question: How will indoor air in Libby compare to other locations across the country?

This is hard to say, but it is safe to say that asbestos (mainly chrysotile asbestos) is present in the air and dust nearly all countries, specifically in urban environments. This is due to its widespread use for decades and the fact that it is a naturally occurring mineral. While there is no single widely acknowledged “background” level of asbestos, several studies have shown asbestos levels (mainly chrysotile) in urban or industrial environments on the order of 0.01 s/cm3 or higher (Murchio, 1973; John et al, 1976; Chatfield, 1983). This is approximately ten times the clearance level for Libby. This does not imply these levels are safe or acceptable, but it does illustrate that all urban areas, and many rural homes with asbestos containing construction materials, are impacted by asbestos to some degree due to its widespread use over time.

Data on the levels of amphibole asbestos in other locations around the US are limited, but it is important to recall that products from the Libby vermiculite mine, including vermiculite insulation, were shipped and used in many, many homes across the county. At this point, it is unclear what, if any, cleanup will occur at these homes. At this point, Libby is the only place in country where vermiculite insulation is being systematically removed from attics.

In the recent cleanup of the apartments around the former World Trade Center, EPA used similar techniques for air clearance sampling. While the details of the sampling and analysis are beyond the scope of this memo and the cleanup situation is somewhat different, it is safe to say that the numerical air clearance criteria used at the WTC cleanup is very similar to the criteria we are using at Libby.

Question: A concentration of 0.001 s/cm3 in air seems like a very low number, but 5,000 s/cm2 in dust seems very high. Why is there such a difference?

The main reason is related to the fact that only a small fraction of dust finds its way into air at any given time - perhaps about 0.1%. If 0.1% of dust that contains 5000 s/cm2 becomes distributed in the air of a room, the concentration in air would be about 0.009 s/cc. Thus, even though the numbers are very different, the risk they represent is similar. This is discussed in more detail in the Appendix.

Question: I thought asbestos-containing materials had to contain greater than 1% asbestos to be a problem. Why are you cleaning up soils and insulation with less than 1% asbestos?

A level of 1% asbestos is a regulatory standard that generally applies to asbestos containing commercial products such as brake pads, pipe wrap, and construction materials. The history of this regulatory standard is too long to discuss here, but the standard is not risk-based and does not apply to all situations In Libby, we have based our decisions for soil action levels and clearance criteria on site-specific risk calculations that show soils below 1% may still present risks that are higher than EPA generally considers acceptable.

Question: Why is EPA using AHERA analysis and counting rules for air/dust samples in Libby?

There are many ways to determine what constitutes an asbestos structure or fiber. Different analytical methods have different rules for determining what structures to count and what not to count. In Libby, we have analyzed such a large number of samples and counted so many asbestos fibers that we know what the general distribution of all fibers is. This enables us to convert sample results among different counting methods, no matter what method we used in the first place on a particular sample. We use AHERA for air and dust analysis because: (1) it is a standard method that can be performed by many labs, (2) it is economical, and (3) similar to the ISO 10312 Method, it counts short length fibers (e.g. fibers less than 5 microns in length).

However, we can easily convert our AHERA sample results to PCME or ISO or otherwise. This is discussed in detail in the Appendix.

Question: When you are done, you should be done. Why do you have to go back and revisit cleaned up properties to sample again?

EPA is making every effort to ensure that any cleanup at a property will be sufficient that a return visit to that property will not be needed. However, because we are not sure what the final action levels will be for soil and dust, we cannot guarantee that we may not need to come back in a few cases. In addition, we are not yet certain that re-contamination from residual sources (dust in heating ducts, carpets, upholstery, etc) is not a problem. It is for this reason that we believe the only way to ensure the cleanup worked, and continues to work, is to revisit cleaned-up homes and sample after some time has passed, and compare the measured levels to final action levels.

There is no better measure of potential exposure than testing homes over time while people live in them.

REFERENCES

CDM 2002. Final Sampling and Analysis Plan, Contaminant Screening Study, Libby Asbestos Site. April.

CDM 2003a. Draft Final Response Action Work Plan, Libby Asbestos Site, December.

CDM 2003b. Draft Final CSS Technical Memorandum, March.

CDM 2003c. Draft Final Design Analysis Report, Libby Asbestos Site, December.

CDM 2003d. Final Sampling and Analysis Plan, Contaminant Screening Study, Libby Asbestos Site, Revision 1. May.

CDM 2003e. Final Sampling and Analysis Plan, Remedial Investigation, Libby Asbestos Site.

May.

Chatfield, EJ. 1983. Measurement of Asbestos Fibre Concentrations in Ambient Atmospheres.

Report Prepared for the Royal Commission on Matters of Health and Safety Arising From the Use of Asbestos in Ontario. Toronto, Ontario.

EPA 2001. Memorandum from Christopher Weis to Paul Peronard. December 20.

EPA 2002. Action Memorandum Amendment, Libby Asbestos Site, May 8.

EPA 2003a. Final Sampling and Analysis Plan for Indoor Dust, Libby Asbestos Site, August.

EPA 2003b. Libby Asbestos Site, HEPA Vacuum Program Fact Sheet. December.

John W, Berner A, Smith G, Wesolowki JJ. 1976. Experimental Determination of the Number and Size of Asbestos Fibers in Ambient Air. US National Technical Information Service, Report #PB-254086, Springfield VA.

Murchio, JC. 1973. Asbestos Fibers in Ambient Air of California. California Air Resources Board.

APPENDIX

SCREENING LEVEL ESTIMATES OF EXPOSURE AND RISK

FROM LIBBY AMPHIBOLE IN AIR, DUST, AND SOIL

1.0 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1

2.0 BASIC APPROACH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1

3.0 RISKS FROM ASBESTOS FIBERS IN AIR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2

3.1 Inhalation Risk Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2

3.2 Methods for Estimating PCM and BCPS Concentrations in Air . . . . . . . . . . . . . . . A-4

3.3 Calculation of Risk-Based Concentrations in Air . . . . . . . . . . . . . . . . . . . . . . . . . . A-6

3.4 Risk Estimates for Indoor Air Concentrations Observed in Libby . . . . . . . . . . . . . A-7

3.5 Risk Estimate for EPA's Air Clearance Criterion . . . . . . . . . . . . . . . . . . . . . . . . . A-11

4.0 RISKS FROM ASBESTOS FIBERS IN DUST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-11

4.1 Basic Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-11

4.2 Parameter Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-12

4.3 Calculation of Cancer Risk-Based Loadings for Dust . . . . . . . . . . . . . . . . . . . . . . A-15

4.4 Cancer Risk Estimates for Dust Levels Observed in Libby . . . . . . . . . . . . . . . . . A-15

4.5 Evaluation of Cancer Risk Associated with EPA's Action Level for Dust . . . . . . A-16

5.0 RISKS FROM ASBESTOS FIBERS IN SOIL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-17

5.1 Basic Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-17

5.2 Parameter Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-18

5.3 Calculation of Cancer Risk-Based Concentrations for Soil . . . . . . . . . . . . . . . . . . A-21

5.4 Cancer Risk Estimates for Soil Levels Observed in Libby . . . . . . . . . . . . . . . . . . A-21

5.5 Estimated Cancer Risks at EPA's Action Level for Soil . . . . . . . . . . . . . . . . . . . . A-22

6.0 SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-22

7.0 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-24

This page intentionally left blank.

A-1

APPENDIX

SCREENING LEVEL ESTIMATES OF EXPOSURE AND RISK

FROM LIBBY AMPHIBOLE IN AIR, DUST, AND SOIL

NOTE: All numeric values derived from the Libby database are based on data available at the time of document preparation and are subject to change pending updates or corrections to the database or revision of data selection procedures.

1.0 INTRODUCTION

This appendix is a description of the methods used by EPA to perform a screening-level evaluation of the potential risks of cancer to residents in Libby, MT, from inhalation exposures to amphibole asbestos fibers in air. For convenience, amphibole fibers of this type are referred to as Libby Amphiboles (LA). At present, quantitative methods are not available for estimating the level of non-cancer risks from asbestos exposures.

The methods used in this appendix to evaluate risks from asbestos in air are the same as those that have been used previously (Weis 2000, 2001a, 2001b), but take advantage of new information on concentration and particle size distribution derived from site-specific samples. In addition, the methods used here begin to assess the relationship between asbestos concentrations in various source media (indoor dust, outdoor soil) and potential health risks to residents for asbestos-related cancers.

These assessments are intended to help provide a basis and frame-of-reference for establishing action levels and cleanup criteria for the ongoing emergency response actions in Libby. As additional information becomes available and as the cleanup progresses towards final remedial decisions, these approaches may be modified as appropriate, in consultation with the community, to ensure a safe and healthy environment.

2.0 BASIC APPROACH

Risk from asbestos is associated mainly with inhalation exposure of suspended asbestos fibers.

Because asbestos fibers are heavier than air and settle out onto surfaces, they will not typically be found in air measurements unless they have been released from contaminated surfaces or source materials by a disturbance:

A-2

Disturbance Source -------------> Air -----> Inhalation Exposure -----> Increased Risk of Lung Disease

The concentration of fibers that occur in air following disturbance of a source and the…