ML20247B952

Forwards Info to Support NRC Disagreement W/Proposed Memorandum of Understanding W/Epa to Assist EPA in Developing Stds for Airborne Radionuclide Emissions.Proposed Clean Air Act Regulations Impose Regulatory Burden on NRC

ML20247B952
Person / Time
Issue date:	05/15/1989
From:	Zech L NRC COMMISSION (OCM)
To:	Reilly W ENVIRONMENTAL PROTECTION AGENCY
References
NUDOCS 8905240298
Download: ML20247B952 (57)

Text

_. -

l ' %. h b,

/ o g"h, UNITED STATES l/

l  ! ? . ,. , .

2 NUCLEAR REGULATORY COMMISSION l j .. . s wassiNcTom. o. c. 20sss h.4.!y w

May 15, 1989 CHA4RMAN The Honorable William K. Reilly Administrator

• Environmental Protection Agency 401 M Street, S.W.

Washington, D.C. 20460

Dear Administrator Reilly:

Under the terms of the Memorandum of Understanding signed by the Environmental Protection Agency and the Nuclear Regulatory Comission in 1980, the NRC com-citted itself to assist EPA in developing standards for airborne radionuclides emissions, inter alia by "providing information and recommendations" and

" reviewing and comenting on proposed standards " In accordance with that comitment, we are providing herewith our strong objections on the proposed radionuclides emission standards (40 CFR Part 61) published in the Federal R:gister on March 7, 1989, at 54 Federal Register 9612. We believe the proposed rule is not necessary as a matter of health risk, regulatory policy, cr law. -

The EPA is proposing to regulate under the Clean Air Act airborne radionuclides emissions that currently result in radiation exposures to the public that are orders of magnitude below the exposures due to natural background radiation and routine diagnostic medical procedures. These proposed regulations are unnecessarily inconsistent with the system currently in place under the Atomic Energy Act by which the NRC and 29 Agreement States regulate all uses of source, special nuclear, and byproduct material under a program that applies internationally accepted standards, endorsed by the EPA under its Atomic Energy Act authority, to all pathways to exposure including those from airborne emissions. The EPA proposed Clean Air Act regulations will add a considerable duplicative regulatory burden and additional costs on NRC and Agreement State licensees, most of whom are educational or medical institutions and small cntities, just to demonstrate compliance, and with no attendant significant improvement in public safety.

In the enclosures to this letter, the NRC is providing its basis for disagreement with the proposed action and the staff's detailed technical cvaluation of the proposed rule. While Approach A (the case-by-case approach) described in the March 1989 notice is the one that is likely to result in the 1:ast duplication of the existing regulations, the Comission believes that none of the four options deserves to be adopted.

\

8905240298 890515 PDR COMMS NRCC CORRESPONDENCE FDC

3 '1 e

The Honorable William K. Reilly May 15, 1989 ioreover, a reversal of EPA's long-held position, that such regulation of NRC-licensed facilities under the Clean Air Act is not necessary, is not needed as a matter of law, es we explain in detail in the enclosure.

The March 1989 notice indicates that the reassessment took place in response to the decision of the U.S. Court of Appeals for the District of Columbia Circuit in Natural Resources Defense Council v. EPA, 824 F.2d 2146 (1987). Thac decision held that EPA cannot take costs into account in deciding what constitutes an adequate level of safety to protect public health. At that time, EPA had taken costs into account in its approach to radionuclides emissions only with respect to coal-fired boilers and elemental phosphorus plants; with respect to NRC-licensed facilities, uranium fuel cycle facilities, and other categories of sources, the decision had been based exclusively on the insignificance of the risk to public health. There is no reason why the U.S. Court of Appeals decision should require any revision of the approach taken by EPA in 1984 and 1985 with regard to the regulation of airborne radionuclides emissions.

The Commission believes that the EPA need not and should not issue NESHAPs for radionuclides emissions from NRC.and Agreement State licensed facilities, since such emissions are already regulated by other agencies at levels which provide a very high degree of protection of public health and safety.

Sincerely, b L/.

Lando W. Zech r.

Enclosures:

1. NRC Position on NESHAPs

2. Program Office Comments 4

ENCLOSURE I NRC POSITION ON RADIONUCLIDES NATIONAL EMISSION STANDARDS FOR HAZARDOUS AIR POLLUTANTS (NESHAPs)

In the proposed regulations, EPA has embarked on the very difficult task of attempting to define a " safe" health risk level for airborne radioactive pollutants, consistent with the standard set forth by the United States Court of.

1146: App1987)

(eals ("

forVinyl the District of Columbia Chloride"). That decisionCircuitheld in NRDC v. EPA, that "the 824 F.2d Administrator must determine what inferences should be drawn from available scientific data and decide what risks are acceptable in'the world in which we live."

-824 F.2d 1146, 1165.

The same task has engaged the radiological sc.ientific community for more than half a century. The National Council on Radiation Protection and Measurements (NCRP) and the International Comission on Radiological Protection (ICRP),

both founded in the 1920's, have established standards for radiation control and progressively refined them so as to limit exposure to ionizing radiation to levels that provide adequate protection of public health and safety. In addition, for some four decades the Atomic Energy Comission and (since 1975) the Nuclear Regulatory Commission have been engaged in the effort to limit exposure to radiation from licensed activities. In the 1960's and 1970's the -

Federal Radiation Council and later the EPA was assigned the responsibility to establish radiation protection standards for use by all Federal agencies.

The general consensus of all these efforts has resulted in current recomen-dations and standards limiting public radiological exposures from all sources except natural background and medical exposures to about 100 millirem per year for individuals but maintaining all exposure at levels as low as reasonably achievable (ALARA). We believe that this consensus standard, reflected in the Comission's recently proposed changes to the radiation protection regulations of 10 CFR Part 20, provides a high degree of protection for public health and safety.

In considering what levels of radiation protection are adequate, moreover, there are certain unique characteristics of radiation that must be taken into account. The first is that, unlike any other pollutant, radiation is ubiquitous in nature, forming part of the environment in which all living species have evolved and exist today. There is wide variation as well in the levels of background radiation to which different persons are exposed, depending on geography, housing type, and personal lifestyles. For example, although the average American receives approximately 100 millirem per year from background radiation (excluding radon), a resident of Denver receives some 50 percent more than that. If radon doses are included, background exposure in the United States averages 300 millirem per year with variations greater than 250 percent.

= _ _ . _ _ _ _ _ . _ -___ - - -_- - _ _ - _ _-_

The proposals put forward by EPA thus deal with amounts of radiation -- 10 millirem and less -- that are not only significantly smaller than natural background, but also' smaller than the variation in natural background in different peographic locations in this country. Those amounts are also substantia ly smaller than radiation received as a result of activities which have never been considered to require regulation, and are not proposed to be regulated under the current EPA proposal, such as airplane travel (about 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> of flight results in a dose of 10 millirem) and dwelling in a masonry house (5-15 millirem / year more than a frame house). Indeed, the level of radiation exposure permitted under Approach D. - .03 millirem per year -- is less than the additional radiktion exposure that results from having one's bedroom on the'second floor rather than'the first floor of one's house. -The question may reasonably be asked whether, if EPA were to establish the " safe" threshold for radionuclides emissions at 10 millirem per year or less, why the public should accept any higher level of exposure from other radiation. sources.

Does this standard imply that the background radiation levels in Denver are

' unsafe" as compared to Washington, D.C.?

Another unique aspect of the regulation of radiological health and safety is that the health effects of ionizing radiation have been studied in great detail over a period of more than 40 years, with a wealth of laboratory data on animals and extensive human data as well. The latter includes data on -

radiation in a variety of forms, including: exposure to the atomic bombs dropped on Hiroshima cnd Nagasaki; fallout on Marshallese Islanders; X-ray used medically for ankylosing spondylitis, tinea capitis, enlarged thymus, tonsils and adenoids, and other disorders; and exposure to radium in the workplace, such as clock dial painters. Unlike those carcinogens for which estimates of harmful levels must be extrapolated from limited data, there is in the area of radiological health a vast literature, and it is this which forms the basis for the general scientific consensus that 100 millirem per year represents a conservative figure for protection against radiation. It should be emphasized, however, that even this 100 millirem per year figure represents an extra-polation,.using the linear hypothesis, from health effects observed at high radiation dose and dose rates; there are no scientific data directly demon-strating any health effects from exposure levels in the range of the variation present in natural background.

It is also highly significant that although there is a consensus in the radiological. scientific comunity in favor of the linear hypothesis, that consensus does not necessarily extend to extrapolations at extremely low levels, such as those involved in the four approaches outlined in the current proposed rule. Indeed, the United Nations Scientific Committee on the Effects of Atomic Radiption in their 1988 report to the General Assembly stated, "there was a need for a reduction factor to modify the risks...for low doses and low dose rates...an appropriate j range to be applied to total risk for low dose and low dose rate should be between 2 and 10." The report also stated, "The product of the risk coefficients appropriate for individual risk and the relevant collective dose will give the expected number of cancer deaths in the exposed population, provided that the collective

i dose is at least of the order of 100 man Sv. (10,000 person-rem). If the collective. dose is only a few man Sv, (200-300 person-rem) the most likely outcome is zero deaths."

Under Atomic Energy Act authorities in place for 35 years, the NRC and its predecessor, the Atomic Energy Commission, have been regulating source, special nuclear, and byproduct material under a program that applies these internationally accepted standards and takes into account all pathways to exposure,. including that from emissions to the ' atmosphere. The NRC's radiation protection program is based on principles that will ensure that exposures are, fir-... adequate to protect he&lth and safety, and second, "as low as reasonably achievable," taking into account costs and the state of the art of technology for reducing exposures. For emissions from the uranium fuel-cycle and uranium mills and mill. tailings, our regulations also implement existing EPA standards for these activities.

The current.NESHAP of 25 millirem per year to the whole body and 75 millirem

. per year to any organ represents an apportioning to the air pathway of this 100 millirem per year standard. We consider this standard, which is consistent with NRC's regulatory approach, to be reasonable; and indeed, EPA's Supplementary Information accompanying the proposed rule confirms that existing regulations and standards provide a very high degree of public protection. -

For' example, Table 6 shows that the calculated incidence of fatal cancers from NRC licensees operating under current emission limits is 0.13 instances per year in a national population'of 240 million. It further shows that a majority of the risk comes from the exposure of the entire 240-million U.S. population to extremely sma?1 increments of risk. Table 8 shows very similar results for existing non-radon emission standards from uranium fuel cycle facilties.

EPA's own analyses indicate that the current 25/75 millirem per year standard squates to a maximum lifetime individual risk of fatal- cancer of 7 in 10,000.

Comparing this risk to the 10 millirem per year alternative under Approach A, EPA notes at one point in its proposed rule that "as a practical matter, this alternative is the same as the current NESHAP." (See Table 7, 54 Fed. Reg. 9630.) Thus EPA's own calculations indicate that none Gf the proposed alternatives for Subpart I are likely to produce signif*; ant reductions in the already low levels of risk under applicable current requirements. At a calculated incidence of 0.13 fatal cancers nationwide per year from the entire category of NRC and Agreement State licensees, this current risk level is well below the one-per-year level proposed under Approach B.

Furthennore, as noted above, natural sources of radioactivity cause exposures wel.1 in excess of any of the options proposed under Subparts I, T, or W. NRC staff calculations using EPA's COMPLY computer code show that dust from the uncontaminated soil for a 40-acre site would result in doses about three times greater than the .03 millirem per year limit under Approach D to Subpart I.

The lower range of the radon flux limits EPA is proposing for disposed tailings piles under Subpart T (.02 to 2 pC1/m2-sec) is equal to on less than the radon flux from typical western soils. It would hardly be logical to require mill tailings piles to have emission levels lower than the uncon-taminated soil that lies adjacent to them.

_=_ _ -_ _ _ - _ . ._ _ _- __---_ __ _ __ _ - _ -

9 's A further problem with the proposed rule is that it requires that calculations are to be based upon the maximally exposed individual, who is assumed to remain in the same location throughout a 70-year lifetime, grow the food to be consumed at that location, and be subject to the same level of exposure throughout that time. This combination of assumptions is too unrealistic to be appropriate, as the International Commission on Radiation Protection has made clear. In its Publication No. 26, the ICRP recommended that its dose' limits apply to the dose equivalent "in the case of exposure to the population, to the average of ...

these quantities over a group of individuals" called the " critical group."

This group, as defined in ICRP Publication No. 43, is to be " reasonably homogeneous" but " representative of those individuals in the population expected to receive the highest dose equivalent." The ICRP publication went on to say, "In habit surveys it is not necessary to research for the most exposed individual within the critical group in order to base the controls on that one person. The results of a habit survey ... should be regarded as an indicator of an underlying distribution and the value adopted for the mean should not be unduly influenced by the discovery of one or two individuals with extreme habits." (Emphasis added.) Contrary to this guidance, EPA states in Appendix A to Volume I of the supporting documentation for the rule that "the location of one or more persons on the assessment grid which provides the greatest lifetime risk (all pathways considered) was chosen for the nearby individuals" for whom the dose limits were calculated. -

in our view, EPA may have underestimated the difficulty and expense licensees will face in demonstrating compliance with the proposed rule. Nuclear power plants would have to discard well-documented, site-specific, NRC-approved I off-site dose calculation manuals and use a currently undocumented and generic COMPLY code to show compliance for airborne releases. Though these plants are likely to be in compliance with either Approach A or Approach B, using new calculational techniques to prove compliance may be problematic. The greatest burden of demonstrating compliance is likely to fall on research institutions, both medical dpd academic, since they use a wide variety of isotopes, have many release points, are licensed for relatively high possession limits, and include older institutions located in areas with a mix of laboratories, classrooms, and residential facilities.

Zn sum, we see little to be gained by establishing another system calling for additional calculations when the best information available indicates that the current system provides an ample margin of safety against unsafe radiation exposure. The likely effect of such a system is to require major expenditures not to reduce radiation emissions below current levels, but rather to produce calculations and reports on compliance with emission standards. The effect is thus to multiply paperwork without conferring any health benefit on any individual.

We believe that before EPA embarks on a course of defining " safe" levels of radiation, it should address squarely the relationship between any new air pollution standards and the recommendations of sucn national and interr.ational groups as the NCRP and the ICRP. If EPA proposes to set standards below what the current radiological scientific consensus regards as acceptably safe levels, 1

1 it should present clear and explicit justification for its determination that existing levels provide insufficient protection for the public health and safety.

What is particularly troubling to us about the proposed rule is that we see no q clear explication of why EPA has deemed it necessary, as a matter of health '

risk, regulatory policy, or law. The approach taken in the proposed rule, of promulgating emission standards for NRC-licensed facilities (such as hospitals  !

and radiopharmaceutical facilities) and uranium fuel cycle facilities (such as nuclear power reactors), represents a reversal of EPA's long-held position that no such regulation was necessary. Our review of the March 1989

. notice and of EPA's statements on this issue in previous years leaves us in some doubt as to the basis for this change of position.

t!hile the March 1989 notice indicates that the reassessment took place in response to the decision of the U.S. Court of Appeals for the District of Columbia Circuit in Natural Resources Defense Council v. EPA, 824 F.2d 1146 (1987) (" Vinyl Choride"), we read that dectston as holding that EPA cannot take costs into account in deciding what constitutes an adequate level of safety to protect public health. But EPA had taken costs into account in its approach to radionuclides. emissions only with respect to coal-fired boilers and elemental phosphorus plants; with respect to NRC-licensed facilities, uranium -

fuel cycle facilities, and_ other categories of sources, the decision had been based exclusively on the insignificance of the risk to public health. See 49 Federal Register 43906 (October 31, 1984); 50 Federal Register 5190 (February 6, 1985). Thus we see no reason why the court's Vin 1 Chloride decision should require any revision of the approach taken by n 1984 and 1985 with regard to the regulation of airborne radionuclides emissions.

Of course, EPA is at liberty to reconsider positions it has taken in the past, even in the absence of any legal compulsion to do so. However, we believe that there was and is no infirmity, either in law or policy, in the approach taken by EPA in 1984 and 1985, when these issues were extensively aired.

In its Federal Register notice of October 31 to NRC-licensed facilities (e.g., hospitals),, that 1984,theEPA declared, record with "does not respect support the conclusion that regulation of [these] ... facilities is necessary to

. protect the public health with an ample margin of safety." 49 Federal Register at 43912. EPA estimated the risk from such facilities as "no more than 0.02 fatal cancer per year, or less than one case every fifty years." Id. This risk, the notice said, was comparable to other risks that EPA consT Ered

" insufficient to warrant regulation in similar Section 112 proceedings." EPA est.imated the' risk of developing fatal cancer, based on a lifetime of exposure to the most concentrated emissions from an NRC-licensed facility, at no more than 2 in 10,000. In the expectation that emissions would continue to be kept as low as reasonably achievable, and were unlikely to increase, EPA saw no need for it to issue emission standards. With regard to uranium fuel cycle facilities, EPA noted that it had already promulgated emission standards for these facilities under 40 CFR Part 190, and that to issue a Clean Air Act standard in addition "would be duplicative and ... would not offer any I additional public health protection." 49 Federal Register at 43915.

1

, i t'

After being found in contempt by in Federal District Court in a decision which i EPA believed to be erroneous, Sierra Club v. Ruckcishaus, 602 F.2d 892 (N.D.

Cal.1984),EPAissuedfinalemissionstandardsforDOEfacilities,NRC-licensed

' facilities, and elemental phosphorus plants. 50 Federal Register 5190 (February 6,1985). EPA's notice made clear that it viewed the emission standards as unnecessary:

- EPA continues to believe existing emissions from these sources are already so low that the public health is already protected with an ample margin of safety, even without regulations. ... Since the beginning of r69ulation under Section 112 EPA has interpreted this section as not requiring regulation in cases where the risks from c category of sources do not exceed a certain minimum threshold. Indeed, contrary interpretations lead to results that are hard to defend from any logical or policy perspective.

(Emphasis added.) 50 Federal Register at 5191.

Consistent with its position that EPA regulation of these airborne emissions served no useful purpose, the agency set limits for NRC-licensed facilities of -

25 millirem per year (whole body) or 75 millirem per year (critical organ) for any member of the public. It also established an alternative means of demonstrating compliance, by which the facility operator could seek to show that no member of the public would receive a continuous exposure of more than 100 millirem per year equivalent and a noncontinuous exposure of more than 500 millirem per year effective dose equivalent from all sources, excluding background and medical procedures. The notice explained that these limits

" reflect current emission levels achieved by existing control technology";

since the great majority of NRC-licensed facility operators would be unlikely to violate the standard, they would be exempted from initial reporting requirements. 50 Federal Register at 5192.

When EPA's action was appealed to the U.S. Court of Appeals for the District of Columbia Circuit in Environmental Defense Fund v. Thomas, No. 84-1524, EPA l

made a number of legal and policy arguments against EPA regulation of radionuclides. These arguments included the following:

1. Petitioners' claim that EPA must regulate radionuclides, even if existing regulations (such as those issued by NRC) are already adequate to protect public health with an ample margjn of safety, " lim in the face of the statute." It is immaterial whethe, ifnse existing regulations were based upon health considerations alone or took other factors into account, so long as they keep risks sufficiently low.

Brief for Respondents at 68.

l

• I

2. "The statutory language explicitly evinces Congress' desire that the Administrator not unnecessarily duplicate NRC 3 regulation." Brief at 69, n. 54. j l

3. Contrary to petitioners' claims, NRC regulations assure both i public access to data on emissions (Parts 20, 40, and 60) and a means to institute proceedings to enforce emissions limits (Section2.206). Brief at 69.

4. The Administrator's judgment that categories of sources presenting insignificant risks require no further regulation is entitled to great deference. Brief at 33-35.

5. It was reasonable for the Administrator, when compelled by the district court to promulgate standards, to set them at a level that maintained emissions at existing levels, since emissions were already at a level that protected health with an ample margin of safety. Brief at 308.

The Cocmission believes that these cogent arguments were sound, both as a matter of law and of policy, at the time they were made and that they are still sound and convincing. We would stress in particular the specific -

directive in the Clean 1.ir Act that EPA and NRC shall, "to the maximum extent practicable..., minimize duplication of effort and conserve administrative resources in the establishment, implementation, and enforcement of

[ radiological]emissionlimitations,standardsofperformance,andother requirements and authorities (substantive and procedural)...." 42 U.S.C.

7422(c)(2). We are unaware of any change in circumstances, either in relevant technical information or in the law, that would invalidate the arguments put forth so cogently by EPA in 1984 and 1985, nor does the March 1989 notice explain why those arguments should not be considered persuasive.

We believe that EPA's reasoning at the time it addressed the issue of radionuclides NESHAPs in 1984 and 1985 was sound, and until such time as a flaw in that reasoning can be persuasively articulated, we believe the existing regulatory structure should be retained.

While our comments are particularly addressed to the proposed standards to be applied to NRC and Agreement State licensed activities, the basic scientific principles discussed apply to the standards proposed fur other categories as well.

1 4

, *f..

l l: ENCLOSURE'2 1

l PROGRAM 0FFICE COMMENTS A. Presentation by Dr. Carl Paperiello to EPA Science Advisory Board B. Memorandum from Robert M. Bernero to Joseph F. Scinto C. Memorandum from Frank J. Congel to Martin G. Malsch

. g 0

o. '.

i i

PRESENTATION BY OR. CARL J. PAPERIELLO ACTING DIRECTOR, OFFICE OF NUCLEAR MATERIAL SAFETY AND SAFEGUARDS U.S. NUCLEAR REGULATORY COMMISSION BEFORE U.S. ENVIRONMENTAL PROTECTION AGENCY SCIENCE ADVISORY BOARD -

RADIATION ADVISORY COMMITTEE APRIL 26, 1989 4

' Good morning Mr. Chairman and members of the Radiation Advisory Comittee of the Science Advisory Board. I am Dr. Carl Paperiello. I am the Acting Director of the Office of Nuclear Material Safety and Safequards of the United States Nuclear Regulatory Comission (NRC). I am here today to summarize the major technical concerns of the NRC staff regarding the Environmental Protection Agency's (EPA's) proposed emission standards for radionuclides for consideration by the Science Advisory Board. In particular, I will address proposed Subparts I, T and W, which apply, respectively, to emissions from'NRC licensed reactor and materials facilities, operating uranium mill tailings piles, and such tailings piles after disposal. i Emissions of radionuclides regulated under the Atomic Energy Act differ from many of the other emissions EPA regulates under the Clean Air Act in two "

fundamental ways. First, these materials have been regulated under a separate program based on long-standing, carefully developed, internationally recognized principles of radiation protection. This is reflected in an NRC regulatory structure which incorporates, among other things, Federal guidance  !

and applicable EPA standards. Second, unlike most other pollutants, radioactive materials occur naturally and pervasively in the environment at such relatively high levels that it will be virtually impossible to demonstrate, using EPA provided methods, that all man-made radionuclides emissions meet the standards if required to be controlled to the alternative levels propos,ed. I will discuss in this statement the implications of both these differences in the nature and regulation of radioactive pollutants.

L

~*

2 9

- Under Atomic Energy Act authorities in place for thirty-five years, the NRC and its predecessor, the Atomic Energy Commission, have been regulating source, special nuclear, and byproduct material under a program that applies internationally-accepted standards and takes into account all pathways of exposure, including that from emissions to the atmosphere. The NRC's radiation protection program is based on principles that will ensure that exposures are, first, adequate to protect health and safety, and second, "as low as reasonably achievable," taking into account costs and the state of the art of technology for reducing-exposures. For emissions from the uranium fuel cycle and uranium mills and mill tailings, our regulations also implement existing EPA standards for these activities.

We consider that the present regulatory system mentioned previously is working well to protect public health and safety, and EPA has presented nothing in its proposed rule to demonstrate that this protection is inadequate. Indeed, evidence presented in EPA's Supplementary Information accompanying the proposed

. standards clearly demonstrates that existing regulations and standards continue to provide adequate protection as well as an adequate margin of safety. For example, "able 6. shows that the calculated incidence of fatal cancers from NRC licensees operating at current emission limits is 0.13 instances per year in a national population of 240 million. The most stringent proposed alternative analyzed in Table 6 would only reduce the calculated incidence to 0.12 fatal cancers per year. Table 6 further shows that a majority of the risk comes from the exposure of the entire 240-million U.S. population to extremely small increments of risk. Table 8. shows very similar results for existing non-radon emission standards from uranium fuel cycle facilities.

3 Current EPA National Emission Standards for Hazardous Air Pollutants (NESHAPs) for radionuclides are 25 mrem /yr to the whole body and 75 mrem / year to any organ. By EPA's calculation, this equates to a maximum lifetime individual risk of fatal cancer of 7 in 10,000. Comparing this risk to its 10-mrem /yr alternative under Approach A, EPA notes at one point in its proposed rule that "as a practical matter, this alternative is the same as the current NESHAP."

(SeeTable7,.54FRp.9630). Thus, EPA's own analyses indicate that none of the proposed alternatives for Subpart I are likely to produce significant reductions in the already low levels of risk under applicable current requirements. At a calculated incidence of 0.13 fatal cancers nationwide per year, this current risk level is well below the one-per-year level proposed under Approach B. "

Furthermore, it is important to keep in mind that natural sources of radioactivity cause exposures well in excess of any of the options proposed under Subparts I, T, or W. In fact, the proposed standards are within the range of variation in natural background from place to place within the United States and within the range of variation measured from time to time at the same location. Our calculations using EPA's COMPLY computer code show that dust from the uncontaminated soil for a 40-acre site would result in doses about three times greater than the .03 mrem / year limit under Approach D of Subpart I.

The lower range of the radon flux limits EPA is proposing for disposed tailings under Subpart T (0.02 to 2 pCi/m -sec) is comparable to or less than the radon flux from typical western soils. It is illogical to impose, and impractical to implement, r. requirement to reduce emissions from mill tailings piles to levels less than emissions from adjacent soils.

, , '. 4 The proposed dose limits are also comparable to the exposures received in many human activities, such as transcontinental airline flights, working in the upper levels of a very tall building and choosing to live in a masonry instead of a frame home. As noted in the Supplementary Information for the proposed rule, the calculated lifetime risk of fatal cancer due to radon in homes ranges up to one in ten, and radon in homes is estimated to cause 5,000 to 20,000 fatal cancers per year. These statistics are particularly relevant when considering radon emissions from uranium mills and mill tailings piles, where EPA calculates an incidence rate of 1.6 fatal cancers per year for operating mills and an incidence of 2.7 fatal cancers a year from previously stabilized tailings. While these impacts are higher than EPA would prefer under the case-by-case approach, they are a small fraction of the risks from natural -

background. We question whether they provide an adequate basis to adopt more restrictive emission standards, particularly considering the compounding of multiple conservative assumptions in the dose assessment methodology EPA used to calculate the risks.

One of these compounding assumptions is that the proposed standards, which are already lower than needed to provide adequate protection to public health, must be applied to the maximally exposed individual. This maximally exposed individual is assumed to remain in the same location throughout a 70 year lifetime, grow the food to be consumed at that same location, and be subject to the same level of exposure throughout that time from the same source. The combination of these assumptions is unrealistic and inappropriate. In its

{: ,

5 Publication No. 26, the International Connission on Radiation Protection (ICRP) reconsnended that its dose limits apply to the dose equivalent "in the case of exposure of the population, to the average of ... these quantities over a group of individuals" called the critical group. This group, as defined in ICRP Publication No. 43, is to be " reasonably homogeneous" but " representative of those individuals in the population expected to receive the highest dose equivalent." The ICRP went on to say that "in habit surveys it is not necessary to research for the most exposed individual within the critical group in order to base the controls on that one person. The results of a habit survey ... should be regarded as an indicator of an underlying distribution and the value adopted for the mean should not be unduly influenced by the discovery of one or two individuals with extreme habits." Contrary to this guidance, EPAr said in Appendix A to Volume I of the supporting documentation for the rule that "the location of one or more persons on the assessment grid which provides the greatest lifetite risk (all pathways considered) was chosen for the nearby individuals" for whom the dose limits were calculated.

Thus, because the standards were based on the individuals hypothetically at maximum risk, and because the contribution to risk from meeting the resulting standards cannot easily be distinguished from the contribution to risk from

- natural background radioactivity, we do not believe that a demonstration of additional risk reduction from current requirements to the proposed levels is appropriate or even achievable. If EPA believes it must undertake a rulemaking to establish new NESHAPs for radionuclides, we believe it should first address why public health is not adequately protected by standards recommended by

1 6

national and international groups such as the National Council on Radiation Protection and Measurements (NCRP) and the International Commission on Radiological Protection (ICRP), and EPA standards under its Atomic Energy Act l Authority to limit public exposure to radiation from any source. Specifically, L te note, it has been recommended, and the proposed NRC regulations in 10 CFR Part 20 set 100 mrem per year as the limit for exposures to members of the public. In addition, the regulations require further reduction of exposure by use of the ALAPA principle. This structure provides a well endorsed limit as well as an additional margin of safety. In particular, the implications for other federal regulatory programs could be profound if EPA finds that annual dose equivalents above 10 mrem /yr or some lower value are " unsafe" or an

" unacceptable health risk" from airborne releases. For an example, a

fundamental radiation protection standard of 10 mrem /yr, raises the question of why any practice that results in higher exposures should be acceptable. A review of recent NCRP Reports, Nos. 93, 94, and 95, shows numerous sources of technologically enhanced radiation that are currently unregulated and likely to result in exposures at these levels.

The NRC recommends that before defining any level below current standards as unsafe or unacceptable, an international symposium of recognized health professionals and epidemiologist be arranged to review and comment on the health effect.s of these very low levels of proposed radiation protection standards.

Under the current regulatory scheme, the collective dose to the population from NRC- and Agreement State-regulated activities is not only a small fraction of the total dose to the public, it is even a small fraction of the dose from discretionary activities that are currently uncontrolled. These proposed

7 regulations will add an increased regulatory burden on licensees, but the principal burden will be to demonstrate compliance, not reduce emissions.

Nuclear power plants would probably be in compliance with option A or B with no effluent reduction but would have to significantly alter well-documented, site-specific, and NRC-approved off-site dose calculation manuals and use a currently undocumented and generic COMPLY code to show compliance for airborne releases. The NRC believes that the EPA has underestimated the cost of these proposals to smaller licensees. We also believe that one of the greater burdens will fall on research institutions, both medical and academic, because they use a wide variety of radioisotopes, have many release points, are licensed for relatively high possession limits, and include older institutions located in areas with a mix of laboratories, classrooms, and residential -

facilities.

The NRC anticipates that it will file further comments pursuant to the Federal Register Notice. Such comments may include legal considerations and additional technical connents. This concludes my prepared testimony, and I would be glad to answer any questions you may have as time permits. With me today are Dr. Donald Cool, Peter Crane and Frank Congal from the NRC staff who have been involved in the preparation of NRC comments and whom are available to assist in answers to questions.

4 I

-- - 1

<. . iso e UNITED STATES

,1 NUCLEAR REGULATORY COMMISSION y, -

wAsmworow. o. c. noses O

g %...

)

MEMORANDbMFOR: ' Joseph F. Scinto Acting Deputy General Counsel j for Licensing and Regulation, FROM: Robert M. Bernero, Director Office of Nuclear Material Safety and Safeguards

SUBJECT:

OFFICE COWlENTS ON PROPOSED CLEAN AIR ACT STANDARDS FOR RADIONUCLIDES Enclosed are the cosmients of the Office of Nuclear Material Safety and Safeguards on the Environmental Protection Agency's (EPA's) proposed rule to establish National Emission Standards for Hazardous Air Pollutants-(NESHAPs) applicable to radionuclides used and possessed by materials licensees. Because" EPA has not provided sufficient time to review all of the more than two dozen supporting documents for this proposed rule, the numerous comments provided here should not be taken as exhaustive.

In essence, we believe no new standards are needed; current NRC regulations and EPA standards adequately protect public health, but EPA volunteered to undertake this rulemaking. EPA also chose to impose a regulatory scheme for vinyl chloride, a chemical compound not found in nature, on radioactivity, which is ubiquitous in the environment. Because the contribution to risk from licensed activities cannot easily be distinguished from the contribution to risk from natural background radioactivity at the extremely low dose levels proposed, we do not believe that a demonstration of additional risk reduction from current requirements to these levels is appropriate or even achievable.

If EPA believes it must undertake a rulemaking to establish NESHAPs for

> radionuclides, existing NRC and EPA rules and standards should be incorporated along with our revised 10 CFR 20 radiation protection standards. If EPA determines to proceed with one of the alternatives in this proposed rule, however, our position should be that this proposal is flawed on several grounds, and should not be promulgated until these flaws are addressed.

Under Subpart I of its 40 CFR 61 NESHAPs rule, EPA is proposing four alternative standards to protect public health from radioactive emissions of NRC and Agreement State licensees. Each is based on a different approach, but two result in the same numerical effective dose equivalent. Approach A, the case-by-case approach to maximum individual risk of fatal cancer, would set a dose limit of 10 millirem per year (ares /yr). Approach B, seeking to limit fatal cancer incidence in the population to no more than one a year, would set

Joseph F. Scinto 2 the same 10 mres/yr dose limit on any member of the public. Approach C, seeking to limit maximum individual risk of fatal cancer over a presumed l 70-year lifetime to no more than one chance in 10,000, would set a dose limit 1 of 3 mres/yr. Approach D, seeking a lifetime fatal cancer risk to an individual of one chance in a million, would set a limit of 0.03 mrom/yr.

Current EPA NESHAPs for radionuclides are 25 mres/yr to the whole body and 75 mrom/ year to any organ. By EPA's calculation, this equates to a maximum lifetime individual risk of fatal cancer of 7 in 10,000. Comparing this risk to its 10-ares /yr alternative, EPA notes at one point in its proposed rule that "[a]s a practical matter, this alternative is the same as the current NESHAP." (SeeTable7,54FRp.9630). EPA also calculates that a dose limit of .3 ares /yr, the midpoint between Approaches C and D, would result in a reduction in the incidence of fatal cancers of 0.06 a year below the calculated incidence of .13 fatalities under current regulations. (Ibid.,p.9630).

Thus, EPA's own analyses indicate that none of the alternatives in the proposed,.

rule are likely to produce significant reductions in the alrea# low levels of risk under applicable current requirements. At an incidence of 0.13 fatal cancers nationwide per year, this current risk level is well below the one-per-year level proposed under Approach B.

The available supporting documentation fails to show that significant implementation difficulties will be avoided, however. The proposed rule does not consider reasonable alternatives, including the alternative of no change in current requirements, to Approaches A through D. In addition, the documentation does not show how compliance with the more stringent alternative 1 standards (Approaches C and D) can be demonstrated. Je can only make an educated guess about the likely difficulty of implementing them, based on the deficiencies of compliance methods made available for Approach A, the most i flexible alternative proposed. '

Our belief is that most materials licensees will have some degree of difficulty showing that they meet any of the four proposed standards, particularly those under Approaches C and D. Our runs of EPA's COMPLY computer code show that dust from the uncontaminated soil cover of a 40-acre uranium mill tailings site would result in doses more than three times the .03 mrem / year limit under .

Approach D. E cept for the highly conservative screening methods in Appendix I E COMPLY is t e only method permissible for licensees to demonstrate I compliance under the proposed rule. Some licensees will be unable to show compliance because the code does not list the nuclides they are licensed to I possess. Other deficiencies in available EPA compliance demonstration procedures are elaborated in the enclosed detailed comments, l

l

Joseph F. Scinto 3 Based on our review, we believe that currently operating, licensed uranium l conversion plants and uranium processing and fabrication facilities, other than uranium recovery facilities, can meet the 10 mrem / year committed effective dose equivalent using reasonable assumptions for dose calculations. It is not l realistic, however, to expect that all of these licensees will be able to demonstrate compliance using the COMPLY code.

Uranium mills will probably not be able to operate under any of the dose limits in Subpart I without more realistic compliance methods or new practices or a i combination of both. The proposed rules have provisions that are inconsistent with provisions of law and codified EPA and NRC rules for mill tailings. For example, the flux limits for disposed tailings piles include contributions from the cover material, not just the tailings. This is contrary to EPA and NRC .

rules. Another example is the proposed ban on the production of mill tailings altogether under Approach D. This appears to be contrary to the spirit of Section 170B of the Atomic Energy Act, as amended. Section 1708 requires annual reports to Congress on the viability of the domestic uranium mining and, milling industry, and provides for investigations under the Trade Expansion Act of 1962 to determine the effects of source and special nuclear material imports on national security.

Other provisions of the proposed mill 2tailings requirements pose technical problems. At the proposed 0.02 pC1/m s level for radon flux, for example, synthetic covers would have to be used. Also, HMSS recommends that Subpart T, which is a performance standard for radon releases from mill tailings after closure, be deleted or made a design standard for closure consistent with 40 CFR Part 192, so that continuing compliance demonstration to meet Subparts I and T will not be required for post-closure emissions. Otherwise, EPA will probably have to address the question of financial assurances for the considerable regulatory and potential remedial actions being imposed. Our preference is that EPA defer to the extensive set of EPA and NRC rules for aills.

have significant adverse impacts on the development of The proposed new low-level rule may(LLW) disposal facilities under the Low-Level Radioactive waste Waste Policy Amendments Act of 1985 (LLRWPAA), even using Approach A. All of {

the proposed standards for licensees under Subpart I appear to apply not only j to ' operations it these facilities, but to the post-operational period for the j duration of the license. As custodian of the site and holder of the license 1 I

during this institutional control period, the host state or federal government l wocid be responsible for monitoring and maintaining emissions within the prescribed dose limits. As a precondition for license approval or transfer. {

the developer or license applicant would also have to show, possibly for at )

4 I

_ _ _ - - - -- __ __ __ )

Joseph F. Scinto 4 '

least 106 years following site closure, that the standard would be met. This will be exceedingly difficult, since the COMPLY code was apparently not designed with the flexibility to accommodate emissions from waste packages or non-point sources such as LLW disposal units. To avoid possible disruptions J

from missing LLRWPAA milestones over this uncertainty if EPA proceeds as "

proposed, 7 MSS recommends that these sites be exempted from Subpart I after closure is completed. This is consistent with EPA's approach to high-level taste disposal. If post-closure emissions from LLW disposal sites are to be covered, however, EPA should provide a method of determining compliance.

ConsideringthepotentialimpactsoflicenseecompliancewiththeseproposN standards, either directly or indirectly through curtailment of disposal capacity or higher disposal fees, we are at a 'oss to determine how EPA could have found that this proposed rule is not a " major rule" under Executive Order }"

12291, and that it will have "little or no impact" on small businesses under the Regulatory Flexibility Act. Compliance demonstration methods were made available only for Approach A, and EPA stated that it would be burdensome to' '

implement Approach D and impossible to predict its impact (see Table 7, 54 FR

p. 9630). Yet EPA was able to affirm that " virtually all small businesses covered by this final rule [ sic] already comply." -

The apparent lack of support for these findings signifies a fundamental flaw in EPA's presentation of its proposed alternatives. The only alternative for which supporting documentation is reasonably complete is Approach A, and this appears to be the only proposed standard EPA could promulgate on the basis of informed public coment as to itn, likely impacts. If EPA can defensibly choose q only Approach A, the Agency has in effect asked the public to coment on I alternatives that are not real. If on the other hand EPA chooses Approach B, C, or D, for which there is no readily available means to demonstrate compliance, the public will not have had the same opportunity as for Approach A j to weigh the burden of implementation against the benefit of additional avoided health risks. If EPA determines not to leave the existing standards in place, nor to pursue the option of codifying current requirements and revised Part 20 limits as NESHAPs, we believe it should defer further action until it has J provided a comparable degree of information for all its proposed options.

]

EPA should also be aware that if it pursues a rulemaking based on any of the approaches set forth in this proposed rule, NRC does not believe it will have a sufficient basis to enforce the resulting standard under the current NRC-EPA

. Memorandum of Understanding (MOU). By EPA's own acknowledgement, compliance with even the proposed 10 mrem /yr dose limit cannot be validated by currently available measurement techniques. In its draft Background Information Document, Procedures Approved for Demonstrating Compliance with 40 CFR Part _6h Subpartj,,EPAconcedestfiatatlevelsconsistentwiththelimitofthe

i .

1 l.

Joseph F. Scinto 5 l standard, " concentrations of many radionuclides will be below the minimum detection level of even state-of-the-art measurement technology." Even when the concentrations can be adequately measured, EPA noted, "it may not be possible to distinguish the portion attributable to the emissions from that which is due to background radioactivity." (Seep.1-4). Thus, even the least stringent proposed standard is already so stringent that compliance is not amenable to real-world verification.

In sum, we believe that more stringent new NESHAPs for NRC-licensed users of radionuclides would be both burdensome and unnecessary. EPA's own analyses suggest that real dose reductions from compliance with these proposed standards would be impossible to measure for most licensees, and probably impossible to demonstrate by calculation if these licensees are limited to using the methods.

proposed. If EPA believes that new NESHAPs are required, it should simply incorporate existing EPA standards and NRC rules, including revised 10 CFR 20 radiation protection standards. There are ample grounds in science and cosmon "

sense to establish NESHAPs for radionuclides differently from NESHAPS for hazardous chemical pollutants, and there is no evident legal requirement to use the sam approach for both.

Robert M. Bernero, Director Office of Nuclear Material Safety and Safeguards

Enclosure:

1. Detailed Comments on Approach, Implementation, and Impacts of Proposed NESHAPs 4

NMSS STAFF C069fENTS ON APPROACH, IMPLEMENTATION, AND IMPACTS OF PROPOSED EPA CLEAN AIR ACT STANDARDS FOR RADIONUCLIDES A. PRESENTATION AND COMPARISON OF THE FOUR EPA APPROACHES TO SETTING NATIONALEMISSIONSTANDARDSFORHAZARDOUSAIRPOLLUTANTS.(NESHAPs)

FOR RADIONUCLIDES Following a procedure developed from a court case on an EPA hazardous air pollutant standard for vinyl chloride, the proposed rule for radionuclides l l NESHAPs proceeds in two steps. The first is to establish a dose limit sufficent to protect public health without regard to costs, and the second is to {i establish a limit to provide an ample margin of safety. This second step i permits consideration of costs, and may or may not be more stringent than the dose limit to protect health, but may not be less.

This proposed rule provides four alternative approaches to the setting of health-protective dose limits. Approach A, the case-by-case approach, would permit. EPA to take into account all relevant information in establishing an ,

acceptable level of health risk'for each source category, including population impact, risk distribution, and uncertainties. Taking into account relevant evidence concerning the reliability of risk estimates, Approach A results in 4 ,

dose limit of 10 mrem effective dose equivalent (EDE) per year to main maximum individual risk of on the order of one chance in a 10,000(10~gaina ) of .

contracting fatal cancer from these sources over the individual.'s presumed 70-year lifetime.

The other three approaches do not consider the uncertainties of risk estimation for each source category, but focus instead on a single measure of risk.

Approach B would establish dose limits to assure that the incidence of fatal cancers from radionuclides emissions is no more than one per year per source category. The resulting dose limit is about 10 mram EDE.

A major difficulty associated with the incidence based approach is that the criterion is applied to a source category without regard to the number of facilities, type of facility, level of airborne emissions from facilities within that sousce category, or the size of the population exposed to these emissions.

Individual facilities in source categories containing a large nurber of facilities would have to reduce their airborne emissions more than facilities in categories containing only a few facilities. This results in an uneven and inequitable distribution of risks. Although an " incidence-based" approach may be feasible, using the source category as the basis for risk assignment does not appear to produce uniform or equitable individual risk levels.

Facilities exposing large populations to very small risks may be required to install additional control equipment because of the " collective risk" criterion embodied in Approach 8. However, facilities having emissions that produce i

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - i

2 substantially greater individual risks may not have to employ additional controls if the size of the exposed population is small. Consequently, the level of individual protection can vary considerably among source categories depending upon the size of the exposed population.

Approach C would establish dose limits to maintain a maximum individual lifetime risk of fatal cancer of ene chance in 10,000. Taking this risk level at face

.value, and disregarding such factors as the likely degree of over- or underestimation of risk characterization, or the match between measured and modelled exposures, the calculation results in a dose limit of 3 mres/ year.

A lifetime risk of 10 d translates approximately to an annual risk of one in a nillion (1.4 x 10'6 yr y). This annual risk is comparable to the lower end of therangesuggestedbgtheInternationalCommissiononRadiologicalProtection (ICRP) of 10 to 10' per year as being commensurate with other everyday risks that the general public faces and which is likely to be acceptable to any individuel member of the general public. (See"Recommendationsofthe International Comission on Radiological Protection," ICRP Publication No. 26 (1977), Paragraph 118).

This approach, which employs a rigid fixed criterion of a 10'4 lifetime risk, does not appear to offer any significant advantage over Approach A except simplicity as stated in the EPA Notice (54 FR 9625). As noted by EPA, this- -

approach has the significant disadvantage of not permitting consideration of the risk distribution, incidence estimates and other data that could be considered in Approach A. Approach A would have roughly the same level of protection as Approach C but would have significantly greater flexibility and would permit greater usage of available information.

Approach D wculd establish a maximum individual lifetime risk of a one-in-a-million chance of fatal cancers. This lifetime risk is equivalent to a dose Ifmit gf 0.03 mrom/yr. The choice of a control level equal to a lifetime risk of 10' is not consistent with other EPA or other societal decisions on risk management. Although we recognize that the Clean Air Act has unique statutory requirements, we believe that greater uniformity in EPA decisions for controlling radioactive materials.and other toxic materials would be in the best l interests of the public. J The 11fetime risk criterion of 10 4 re 10'8 Much higher levels (10'6 to 10'gresents an annualbeen

) have generally risk considered of about 1.4trivial x or de minimis. (See International Atomic Energy Agency, " Principles for the Exemption of Radiation Sources and Practices from Re I Safety Series Report No. 89, Vienna (September ). 1988)gulatory Control," IAEAFor exam draft standards for low level radioactive waste disposal (40 CFR Part 193), a {

1evel of four millirem per year is under consideration for deregulation as l "below regulatory concern." This level is over a factor of one-hundred greater than the 0.03 millirem per year value. Thus EPA has proposed a limit under ]

i

3

' Approach D that is equivalent to a level that EPA deems to be "below regulatory concern under its Atomic Energy Act authorities.

Given the four types of. approaches to setting limits for radionuclides NESHAPs, the NRC believes that Approach A is preferable, based upon the greater amount of information and flexibility available in reaching a decision. If NRC-licensed activities are to be further regulated, we would recommend that the EPA administrator adopt the philosophy of Approach A.

B. REALISM OF RISK ASSUMPTIONS CnMON TO ALL PROPOSED APPROACHES To develop the risk levels on which the proposed alternative dose limits are based, EPA used assumptions about the magnitude and duration of radionuclides exposures that are unnecessarily conservative. EPA has assumed that the maximally exposed individuals offsite from a licensed facility live continuously for 70 years at the point of maximum exposure. This would be realistic only under an extremely rare set of circumstances. The probability that an individual will choose to live 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> a day for 70 years in the same location during the period of operation is exceedingly small. We also believe that very for facilities will continue in existence for 70 years, and only a fraction of.,

these few facilities will continue to have the same mix and level of emissions for many decades. The EPA assumption about the maximally exposed individuals is thus contradictory to the claim (see 54 FR 9616) that EPA has tried to give- -

"best estimates" of individual and population risks. Further, such an assumption is contrary to the way in which other radiation. protection standards have been developed on both the national and international. level.

Despite the unrealistically conservative assumptions used in the development of the alternatives, EPA's analysis of these limits suggests that there will be little additional reduction of risk regardless of the alternative chosen. Table 6 in the Supplementary Information for the proposed rule (see Federal Register Vol. 54, No. 43, p. 3629) indicates that under both the most stringent r mit proposed and the no action alternative of leaving the current requirements in place, there will be approximately one fatal cancer about every eight years from the airborne emissions of NRC- and Agreement State- licensed non-fuel cycle facilities, for the population living within 80 kilometers of those facilities.

For uranium fuel cycle facilities, including power reactors, mill tailings, and ,

all other facilities except those operated by DOE for uranium enrichment, the figures for fatal cancer incidence are on the same order of magnitude. Under an alternative of 0.3 mram/yr, approximately 95 additional individuals in the U.S. '

population would have their level of fatal cancer risk reduced to below one in 10,000 (see Table 8, 54 FR 9631). For comparison, these risk levels are an order of magnitude less than the probability of death by tripping and falling over a 70-year, lifetime. We also note that variations in background radiation exposures in the United States are substantially greater than the values proposed by the EPA as necessary to protect public health and safety. Thus the need for further regulation is questionable in light of what the U.S. Court of

_ .-____m___.__.__ ____.___ __m_mm_._ ___._____._________.-__m -

Appeals had in mind when it spoke of " risks (that) are acceptable in the world in which we live" in the Virtyl Chloride case.

~

The NRC notes that for each of the NESHAPs' options in this proposed rule, EPA

.proposes no additional requirements to provide the " ample margin of safety" by reducing doses below the levels believed necessary to protect health. We believe that the ALARA requirements in NRC regulations and the recommendations of national and international organizations such as the National Council on RadiationProtectionandMeasurements(NCRP)andtheICRPprovideanadditional.

degree of protection essentially equivalent to the " ample margin of safety" sought by the proposed rule. Further, we believe that the values proposed by the EPA are, in reality, " ample margin of safety" values and are not the values necessary to protect health.

The absence of any additional margin of safety in the proposed rule, and EPA's view that the health-protective dose limit for any of the proposed approaches would itself provide an adequate margin, is an indication that this rulemaking is more accurately viewed as an exercise in establishing a margin of safety than a dose Ifmit above which public health is no longer protected. If all of the four proposed approaches are well below the limits set by competent national and

international radiation safety bodies, the critical question is not "Which-option is required to protect health?" but "Do any of them provide the optimum balance between averting unwanted health effects and avoiding unneeded costs?"-

In view of the difficulties posed by unrealistic assumptions and the negligible benefits of further regulation, we believe that EPA should consider carefully the need to promulgate further regulations in this area. However, if further regulation is required for various reasons, then we believe that EPA should establish a health and safety limit consistent with the recommendations of the ICRP and NCRP. A reduction in the overall dose limit to a level such as 25 mree/yr effective dose equivalent for air emissions would be consistent with the ICRP and NCRP recommendations for apportionment to sources and pathways of exposure. If EPA believes that additional reductions in the dose to 10 mrem as proposed in Approach A are necessary, then these reductions should be characterized as those to provide the " ample margin of safety", rather than as necessary for health and safety.

C. BURDEN OF DEMONSTRATING COMPLIANCE JUDGED AGAINST LIKELIHOOD OF FURTHER RISK REDUCTIONS

1. In its Regulatory Flexibility Act analysis (see 54 FR p. 9651. EPA finds that " virtually all small businesses covered by this fTnal rule [ sic] already comply [with the proposed standards). ... Therefore this rule will have little or.no impact An small businesses [of 750 employees or fewer]." As no compliance demonstration procedures are provided for any proposed standard but Approach A, the one for which compliance is comparatively easiest to demonstrate, and no procedures are available at all for many nuclides, the staff is unable to find the basis for EPA's assertion. Given the compliance demonstration uncertainties

)

discussed below for even the 10 mres/yr dose' limit alternative, and the possibility that EPA might choose a more stringent emission standard than this alternative, the staff believes that EPA has not provided sufficient information to assess the impact of all its alternative proposed standards, and the rule I should not be promulgated until the missing information and an opportunity to i comment on it has been made available.

l-

2. Unrealistic assumptions about the magnitude and nature of the risk effectively make it impossible for small operations to use proposed EPA screening procedures to avoid more elaborate compliance demonstration efforts.

L Under these procedures, licensees are required to assume that there is no dispersion of emissions from the point of release to the point of intake, and that all food ingested by those receiving doses from the source are grown at the i

place of residence. For more than half the radionuclides listed in Table 2 of AppendixE,theseasgumptionshaveresultedinscreeningconcentrationssolow-(E -13 to E -15 Ci/m ) that it is doubtful that some levels can even be distinguished from background and detected. The stated purpose of these screening procedures -- simp 1tfying compliance demonstration for small entities

-- would be better served by applying conservative but reasonable assumptions.

e.g., allowing.some amount of dispersion and recognizing that not all food is grown at the dose recipient's residence.

3. In additio'n, the first three methods for compliance demonstration provided 4n the COMPLY code list only 54 isotopes. NRC and Agreement States license possession and use of considerably more than 54 isotopes in unsealed form. Any licensee,.regardless of size, who possesses in unsealed form an isotope not listed in these methods would have no choice but to use the fourth and most demanding compliance procedure, which lists 460 isotopes. Even these 460  ;

isotopes do not include the following isotopes licensed by NRC fcr use in '

unsealed form, however:

Ac-227, Am-242, Am-243, Br-76, Cd-115, Cd-115m, Ce-137, Cs-139, Cu-67, Dy-159, Gd-151, Gd-162, Hf-172, Hf-180m, Hg-195, Lu-172, Po-208 Po-209, Sa-145, Sn-121, Sn-121m, Ta-183, Tb-161, Th-225, Th-233, Tm-166, Tm-172, W-181. W-188, Yb-177.

Although EPA has an effort underway to expand the number of isotopes covered in the first three procedures, the final rule should not be promulgated until complete radionuclides tables are developed and issued for comment.

4..By EPA's own acknowledgement, compliance with even the least stringent proposed alternative dose limit 10 arem/yr cannot be validated by currently available measurement techniques. In its draft Background Information Document, Procedures Approved for Demonstrating Compliance with 40 CFR Part 61, Subpart I, EPA concedes that "[WJhile, in principle, the doses resulting from the release 1

I

of radionuclides to the atmosphere can be determined by environmental monitoring, valid measurements are often impossible to obtain. At the levels consistent with the Ifmit of the standard, concentrations of many radionuclides will be below the minimum detection level of even state-of-the-art measurement technology. Even when the concentrations can be adequately measured, it may not be possible to distinguish the portion attributable to the emissions from that which is due to background radioactivity." (Seep.1-4). Thus, even the least stringent proposed standard is already so stringent that compliance can only be shown by indirect methods, such as inside-the-plant measurements and models of exposure pathways. -

l D. EXPECTED IMPACTS OF PROPOSED NESHAPs ON NRC MATERIALS LICENSEES

1. Fuel Cycle Facilities Based on our review, the staff believes that currently operating, ifcensed uranium conversion plants and uranium processing and fabrication facilities (excluding uranium ore mills) can meet the 10 mres/ year committed effective dose equivalent (EDE). However, it is not realistic to expect that these licensees i will be able to demonstrate compliance with the COMPLY code in some cases.

l- Implementation of the proposed 3 mrem / year EDE would be difficult for some of the major fuel cycle facilities, primarily because of the constraints of the -

system imposed for determining compliance. In order to establish compliance with 40 CFR 190, licensees were required to put continuous air monitoring stations at the locations of nearby members of the public, and to do extensive and costly analysis of the samples obtained. This was done in part because of the large uncertainties in dispersion modeling, and in part to enable more (

accurate dosimetry. In order to ensure compliance with the 3 mres/ year EDE, additional equipment besides that mentioned above would need to be installed, and additional monitoring programs established. Demonstrating compliance would be very difficult, and most likely could not be achieved within the constraints proposed in 54 FR 9612.

The proposal of 0.03 mrem / year EDE is an unrealistically restrictive level.

Such a limit, besides requiring a vast investment in conventional air cleaning, would require large efforts to find and control minuscule sources of fugitive emissions.

2. Industrial and Medical Users, Suppliers, and Supporting Operations This proposed standard will present fundamental problems for many licensees such as manufacturers who currently have possession limits of radioisotopes that equal those of the EPA annual possession quantities. The estimated number of facilities subject to this proposed rule approaches 15,000 when Agreement State licensees are included. Although the 10 mres/ year EDE Ifmit may be considared an achievable limit for most types of facilities, many of NRC's materials licensees do not hve the necessary environmental monitoring nor computer

i l_ , '.

, 1 equipment to demonstrate compliance. Many licensees will have to use the most

, complicated level of the COMPLY code (level 4) because the initial levels use L severely overestimated doses based on unrealistically conservative assumptions.

Thus, the EPA assertion that most NRC licensees will be able to demonstrate

, compliance using the simple screening levels appears to be incorrect.

The EPA. limits are established on a yearly basis (C1/yr). In contrast, NRC l licenses stipulate specific possession quantities at Jan one time. Therefore, l' It is possible for a licensee to have a weekly thrTughput on the order of the yearly EPA limit. This is clearly the case for example, in Type A Resea:ch and )

Development licensees that order several cur,ies of one radiopharmaceutical on a weekly or monthly basis equal to that of their possession limit. For I-131, that could easily be 2 curies per month, which would be an average of 24 curies per year which is well in excess of the 6 ::urie per year limit that the EPA  ;

NESHAPs' propose. As shown_in Table 1, eight types of program codes were run and identified as those being most likely to have difficulty establishing compliance (based on possession limits) with the proposed regulations. In particular, radioisotopes such as An-241, Cs-137, Co-60, H-3, P-32, 3-35 Tc-99m and Xe-133 appear to be in excess (across the board) for many different types of  ;

licensees. Although only a small porticn of the approximately 12,000 NRC licensees were considered for this evaluation, it should not be concluded that other categories will not have problems in demonstrating compliance.

The principle shortcoming in regulating the doses listed in 54 FR 9612 comes in the difficulty of measuring concentration levels as listed in Table 2, Appendix E. Forsomeparticularisogopes,sughasPb-210[ Table 2,AppendixE concentration of 3E-15 C1/m (mCf/cm )], the average natural background radiation in various parts of the country exceeds this limit (see NCRP 94). A further re-evaluation of the vclues in these tables is needed to ensure that these types of discrepancies do not occur.

In the case of medical and hos consihrabledifficultiesfor.pitallicensees,theproposedNESHAPscouldcause dospitals that maintain on-going programs to investigate new uses of radiopharmaceuticals for improved redical techniques.

The proposed restrictions on yearly possession limits could eliminate such prograns and result in reduced health care simply to avoid the problems associated with demonstrating compliance.

In addition to the compliance problems that our licensees wculd have with this proposed rule, NRC foresses additional administrative problems that could potentially effect the methodology behind the way licenses are currently written (in terms of possession limits). This may need to be re-evaluated in order to help alleviate the problem of how a licensee determines compliance. We estimate thatiseveral hundred license amendments may be requested by NRC licenseec to lower their possession limits in order to establish compliance with the proposed EPA standards.

3. Uranium Mills and Tailings Piles (SeeAttachmentA)

4. Low-Level Waste Disposal Facilities (See. Attachment A)

E. OTHER CONCERNS AND COM ENTS

1. Considering the noted deficiencies with the COMPLY code, we recommend that NRC licensees be permitted to utilize the additional flexibility and realism afforded by AIRDOS-EPA, RADRISK, and other EPA-approved codes in order to ensure compliance. NRC originally had requested EPA to use a compliance demonstration code that could be run on a personal computer rather than on a large mainframe computer due to the large number of Itcensees that did not have ready access to larger computers. NRC did not intend for EPA to prohibit the use of other codes in ensuring compliance with the regulations. Most of the NRC licensees who have in the past demonstrated compliance have done so by using AIRDOS-EPA, which allows them the flexibility and capabi11ty for adequate evaluation of site-specific factors.

2. The applicability of the proposed standard to temporary work sites and . ,,

outdoor sites must be further avaluated. In tracer studies, liquid radioactive material is released to r well or river to determine the groundwater flow rate.

These types of activities causes airborne emissions as the radioisotope is allowed to evaporate into the atmosphere. Because there is not a specific release point at which to measure the concentration of the radionuclides, it becomes impossible for a licenses to determine if these activities are in compliance. Furthermore, there is no criteria in the COMPLY code which allows for these types of licensees to demonstrate compliance. Another example of an -

air emission which is difficult to measure and determine is that of the waste j compaction services licensees. Portable waste packaging services such as compactors and supercompactors cause air emissions under constantly changing conditions and at various locations (in the case of the mobile supercompactor) which may effect the overall compliance of the facility where operations are being performed. Section 61.100 states that the standard applies only to licensed facilities. In many cases, however, licensed material is authorized for use at temporary job sites. If it is EPA's intent not to regulate these operations under this rule, then further clarification of this issue is needed.

3. The definition of " sealed source" is unclear. The Supplementary Information to the proposed rule refers to the burial of " sealed sources" of high-level waste (HLW), while earlier guidance documents referred to " packaged materials."

In ' Subsection 61.100, EPA states that this proposed rule does not apply to "any licensees that posesses and uses radionuclides only in the form of sealed sources." Does this imply that those licensees that have both sealed sources 9

I J , .!

1 i

and other forms must treat sealed sources as solids? The intent of this provision should be clarified.

4. The final rule should specify the basis for coverage of accidents, the special provisions, if any, concerning the treatment of accidents for compliance i purposes, and types of accidents that were included in EPA's supporting assessments for the proposed rule.

5. Under Subsection 61.I00, low energy accelerators are exempted from this proposed rule in Subpart I, which covers NRC and Agreement State-licensed facilities and non-DOE federal facilities. It is unclear whether the proposed rule is intended to cover releases of NARM from facilities licensed by NRC. (An example would be a hospital licensed to use radioactive materials, but that also operates e materials)quipment that uses The purpose cobalt-57 of this or other exemption andunlicensed accelerator-produced its implications for emissions of NARM should be explained.

6. The proposed rule draws a distinction between the disposal and the management, processing,andstorageofhigh-levelwaste(HLW). It also exempts from regulation " facilities constructed and dedicated to long term disposal" of HLW " pursuant to regulations to be promulgated at 40 CFR 191 Subpart 8'. This, implies that the surface facilities at a repository (which are currently subjeci.

to Subpart A of EPA's HLW standards) would be subject to a Clean Air Act -

standard superseding current standards. Because Subpart 8 applies to

. repositories after disposal has taken place, and the exemption under this proposed rule applies only to facilities " constructed and dedicated" to '

disposal, it appears that only closed and sealed disposal facilities are exempted. If so, surface facilities would be subject to radionuclides NESHAPs during repository operations. In addition, any operations and accidents resulting in releases from the underground facility would also have to meet the dose limits. If this is the intent of the proposed rule, it should clarified ed justified.

7. Subpart I of the proposed rule would apply to nonradon airborne releases of radionuclides from mill tailings and low-level land disposal facilities during the Itcensed long term care periods after closure. The intent of the CAA and Subpart I would appear to focus more on operations and stored wastes than on properly closed disposal sites. NRC recommends that low-level waste and tallings disposal sites be exempted from Subpart I after closure is completed.

ff the long term care period is not exempted and releases must be evaluated, EPA must provide a way to demonstrate compliance. EPA must either provide the flexibility to use appropriate alternative compliance demonstration methods or provide specific methods to determine compliance. Similarly, NRC recossends that Subpart T, which is a performance standard for radon releases from mill impoundments after closure, be deleted or made a design standard for closure consistent with 40 CFR Part 192. As proposed, it is a performance standard requiring active EPA regulation and radon measurements forever by the long term care agency.

8. If post-closure emissions are to be actively regulated under Subparts I and T contrary to NRC recommendations, EPA should address providing financial assurances for evaluations, monitoring, reporting, filing facility modification requests, and taking remedial actions. Financial planning is needed to enable long term governmental caretakers of closed disposal sites to conduct the '

required regulatory actions and to fund potential corrective actions.

9. The' EPA should consider including a variance provision similar to that in 40 CFR Part 190. Such a provision would allow specific recognition that it could be more beneficial to society to allow facility operations to continue, pending correction of an emissions control problem.

4 l

I TABLE 1. NRC LICENSEES SPECIFICALLY EFFECTED BY THE PROPOSED EPA E5EXPTFOR REGULATION UNDER THE CLEAN AIR ACT (54 FR 9612-9668)

PROGRAM TYPE OF NUMBER RADI0 ACTIVE MATERIAL POSSESSION EPA LIMIT CODE LICENSE OF (Liquid or Powder LIMIT (C1/yr)

LICENSEES Unless otherwise stated) (C1) 01100 Academic (57) Cs-137 1.5 2E-02 Broad Scope Am-241 10.0 2E-04 Po-210 100.0 5E-02 Kgg85 (Foil) 1.5 No Limit A 200.0 ---

Co-60 2.0 1E-02 Cs-137 1.0 2E-02 Gd-162 1.0 ---

I-129 1.0 ---

02110 Medical (75) Ir-192 2.0 8E-01 Institution S-35 8.0 3E+00 Broad Scope Co-60 10.0 1E-02 Cs-137 12.0 2E-02 Dy-165 25.0 ---

Sr-90 5.0 8E-01 Cl-36 0.05 ---

I-125 10.0 SE+00 I-131 10.0 6E+00 l 03110 Well Logging (22) Am-24},. 5.0 2E-04 and/or I-131 6.0 6E+00 Tracer Studies Ir-192 1.0 8E-01 l Co-60 0.2 IE-02 l Sc-46 0.5 ---

l Ag-110m 0.4 8E-02 Au-198 1.0 ---

Br-82 1.0 ---

La-140 0.5 ---

Sb-124 0.5 ---

03113 Field Flood (3) Ag-110m 5.0 8E-02 Studies Br-82 5. 0 ---

EPA Limit is for a liquid or powder unless the radioactive material is denoted as a gas.

• • Any isotope with atomic numbers 1 through 83.

• • • Although these isotopes meet the proposed criteria based on possession limits, it does not mean that they will meet the EPA limits on an annual basis.

TABL.E h (Continued)

PROGRAM TYPE OF NUMBER RAOI0 ACTIVE MATERIAL POSSESSION EPA LIMIT CODE LICENSE OF (Liquid or Powder LIMIT (C1/yr)

LICENSEES Unlessotherwisestated) (C1) 03113 Field Fiood (3) Ca-45 0.5 ---

Studies C1-36 0.5 ---

(Continued) H-3 1,000.0 8E+00 Zn-65 5.0 3E-01 Ar-37 250.0 ---

Co-60 5.0 1E-02 C-14 (G) 40.0 3E-01 Sr-90 5.0 8E-01 S-35 5.0 3E+00 Xe-133 (G) 250.0 4E+01  !

03211 Manufacturing (16) Co-60 200.0 IE-02 Ir-192 45,000.0 8E-01 Cs-137 12,000.0 2E-02 H-3 150,000.0 8E+03 -

Kr-85 10,000.0 SE+02 Ni-63., 1,000.0 3E+02 P-32 550.0 SE+00 Sr-90 3,000.0 2E+00 5-35 1,000.0 3E+00  !

Xe-133 1,500.0 4E+01 1-131 500.0 6E+00 Tc-99m 10,000.0 IE+03 Am-241 0.1 2E-04 '

Ir-192 1.0 8E-01 Mo-99 10,000.0 7E+04 Se-75 200.0 IE+00 Ag-110m 1.0 8E-02 03232 Waste (9) C-14 25,000.0 3E+02 Disposal Hg) 25,000.0 8E+03 A 50,000.0 ---

Am-241 80.0 2E-04 Cs-137 20.0 26-02 l

EPA Limit.is for a liquid or powder unless the radioactive material is i denoted as a gas.

• • Any isotope with atomic numbers 1 through 83.

• • • Although these isotopes meet the proposed criteria based on possession "

limits, it does not mean that they will meet the EPA limits on an annual l

basis.

L-__-----------------.---.------------- -

1 - -

13 TABLE L (Continued)

PROGRAM TYPE OF NUMBER RADI0 ACTIVE MATERIAL POSSESSION EPA LIMIT CODE LICENSE OF (Liquid or Powder LIMIT (Ci/yr)

LICENSEES Unless otherwise stated) (C1)

~

03234 WasteBrokers(6) A 50,000.0 ---

Cs-137 100.0 2E-02 Co-60 50.0 1E-02 03610 R & D TYPE A (129) Am-241 5.0 2E-04 (Universities, Co-60 12.0 1E-02 Private Industries H-3 (G) 1,000.0 8E+00 FederalFacilities} H-3 5,800.0 8E+03 Po-210 5.0 9E-02 Sm-151 5.0 ---

Te-123m 3.0 ---

W-181 10.0 ---

Cs-137 8,000.0 2E-02 Lu-177 10.0 ---

Kr-85 (Foil) 10.0 No Limit -

EPA Limit is for a liquid or powder unless the radioactive material is denoted as a gas.

• • Any isotope with atomic numbers 1 through 83.

NOTE: Those isotopes that did not have an annual possession limits listed in Table 1 of Appendix E (54 FR 9668) as denoted with three dashes "-- "

in the far right column.

4 l

, .. j

. i l

Appendix A COMMENTS SPECIFIC TO LOW-LEVEL WASTE DISPOSAL SITES AND -

URANIUM RECOVERY FACILITIES A. Low-level waste disposal sites

.i

1. EPA ~ should clarify the status and assess the impacts for this category of facilities. Low-level waste disposal sites are covered by Subpart I. The sites also appear to be subject to active dual NRC/ Agreement State and EPA regulation and compliance as long as there is an NRC or Agreement State license at the site (f.e., for the 20 or so years of operation and for up to 100 years of long term care). Thus, all releases during the operational, closure, post closure, and long term care periods must comply with the final standard adopted. NRC staff recossend that releases from the sites after closure should be exempted and that EPA should not actively regulate the sites during long term care. The existing institutional care provisions in 10 CFR Part 61 provide ample assurances that major unanticipated events will be addressed.

during the period dictated by the radiological hazard of the low-level waste for new sites and existing sites are all government owned and following 10 CFR Part 61 to the extent practicable. If the sites are covered by the final "

Subpart I to 40 CFR Part 61 as they are in the proposed rule, EPA must provide means to demonstrate compliance during post-closure as well as during operations and closure. EPA should also address financial assurance for responding to EPA regulation during long term care if such EPA regulation will be required by the final rule.

Although time and resources did not permit careful review of all the supporting documents, low-level weste disposal sites seem to only have been addressed in passing. For example, waste receivers / shippers and disposal facilities are described briefly in Section 2.2.6 of EPA /520/1-89-001. This ducriptive summary implies that they are covered and the lack of exclusion in the standard in 40 CFR 61.100 would appear to confirm coverage. However, no risk assessment appears to have been done and no consideration appears to have been given in the mandatory compliance procedures. For example, the sites are not discussed in EPA 520/1-89-006-1, the draft environmental impact statement.

2. EPA has not provided a clear means for low-level waste sites to determine compliance using the options provided in the proposed rule. Problems with nuclide inventories, waste form, site configuration, and emissions were encountered in trying to evaluate the prescribed compliance procedures. The

. sites currently accept all radionuclides, although the concentrations or quantities may be limited. In general, wastes are packaged, but some wastes may be received in bulk form or be treated in some way. Air emissions at existing sites may result from many activities at the sites, including DOT permitted surface contamination, venting from high integrity containers, noble gas release from opening shipping casks to remove drums or liners, and waste decomposition and biodegradation. Many of these potential sources of air emissions are not point source releases in that potential releases such as releases from an array of undamaged and damaged packages in a disposal trench

could occur. COMPLY's most comprehensive list of radionuclides is in compliance Level 4, but all possible nuclides are not covered. COMPLY is based on point source releases. All input parameters are based on information for the point of release. Multiple release points can be acconnodated, but multiple calculations are required.

The combination of the nature of low-level waste site operations and potential for emissions and the requirements and limitations of the code lead to a number of problems. The code is not sufficiently flexible to do what it is supposed to do. Level 4 is required to attempt to cover the possible radionuclides.

All nuclides should be included in the code.or alternate compliance methods should be provided. Alternatively, EPA could provide a means of limiting analyses to the lists provided or provide exemptions such as exempting very short half Ifved isotopes which are unlikely to still be present when wastes are received at the sites. The COMPLY code and instructions do not address how to deal with area or volume sources or packaged wastes. The NCRP document for manual calculation of releases using levels 1-3 suggests in Section 2.2 of the Introduction that * ...an area source, e.g., a contaminated plowed field, rather than a point source, then the user must assume that all the radioactive material is being emitted from a single point at the center of the area." In this one respect, the approach would be underestimating impacts.

The code or other compliance methods would be more useful for making more realistic assumptions for low-level waste sites if additional flexibility was provided. The COMPLY code includes many simplifying assumptions and restrictions. For example, the user may only choose whether the food supply is grown at his residence or at another receptor. No other guidance to address area or nonpoint sources was found. The code has to be tricked to deal with nonpoint sources and the validity of calculations so outside the intent of the programmers is questionable. The COMPLY code, or alternatives EPA may consider for the final rule, should have some capability to deal with nonpoint sources.

(This flexibility is likely to be needed for other categories of licensees as well). The code is not configu ed to use measured concentration data at the receptor or at site locations between the " point" of release and site boundary. ,

Flexibility to use such measurements should be provided in the final compliance procedures, particularly in view of the difficulty and complexity of calculating releases at these sites. The terms " sealed source," " packages,"

and " containers are not consistently used or sufficiently defined to determine when and how low-level wastes must be counted in inventories and used to estimate releases. (Note that special requirements for estimating releases for uranium mills was proposed.) Other flexibility in COMPLY that would be helpful in using the code are to provide the option to turn off or reduce the impact from one or more of the ingestion pathways and the option to input totally by using diskettes. Using data from diskettes would reduce typing and editing efforts for multiple calculations and allow changes to one data set without retyping unchanged data.

3. EPA could rely on codified NRC standards for low-level waste sites. Under 10 CFR Part 61, radionuclides releases from all pathways combined must not exceed annual dose limits of 25 millirem to the whole body, 75 millirem to the thyroid, and 25 millirem to other organs. Releases must also be ALARA below these Ifmits (see 10 CFR 61.41).

2

4. EPA should also consider relegating the detailed, prescriptive compliance matters to the guidance documents rather than incorporating them into the rule.

If not EPA should include options other than COMPY or at least the flexibility for EPA approved alternatives. Doses from all pathways, including air pathways, must be explicitly considered in meeting the performance objective for radionuclides releases. Air releases were addressed as part of other scenarios such as evaporator releases from leachate treatment or incinerator releasesL at regional centers. No centralized or consolidated analysis for air emissions was presented in the supporting environmental impact statements. The Itcensing guidance documents (e.g., NUREG-1199 and NUREG-1200) developed since 10 CFR Part 61 was issued call for the submittal and evaluation of extensive site-specific information on airborne releases and transport mechanisms.

However, NRC has not specified specific computer codes for estimating and evaluating air emissions. Thus, licensees have flexibility in evaluating air releases and NRC is not in a position to recommend specific alternative codes j er guidance documents for low-level waste sites.

5. EPA needs to consider the potential impact of the three proposed dose limits on the development of new low-level waste disposal sites. Having to demonstrate compliance with release limits of 3 millirem / year or 0.03 nillirem/ year could limit the viability of engineered alternative disposal methods now being considered by States. Some of these facilities may have ventilation systems to allow operational access for weste emplacement or inspection or retrieval of wastes. For perspective, the prototype application ,.

developed by DOE for earth mounded concrete bunkers (EMCBs) included estimates of the air emissions during operations. The estimates did not use COMPLY, but did address a full range of potential exposures, as does COMPLY, including direct inhalation, deposition and resulting direct radiation, and ingestion from airborne deposition. The estimated whole body equivalent dose to a menber of the public from routine operations was 3.3 millirem / year. If an incinerator to treat the wastes was added to the facility, it might contribute additional doses of about 3 millirem / year, resulting total doses of more than 6 millirem / year. An incinerator would be used to convert waste to a form more suitable for engineered disposal. EPA should evaluate the impacts and provide reasonable compliance methods for new low-level waste disposal facilities. The 10 millirem limit coupled with reasonable compliance methods may be the only option that would not be disruptive to current siting and development efforts. ,

6. The proposed dose limit of 0.03 millirem / year is essentially zero as demonstrated by the following results using COMPLY. Using emission methods l from Regulatory Guide 3.59 (required by the proposed rule for mills), Th-230 doses from soil blown from 10 acres to a receptor 100 meters away is 0.01 millirem / year. Thorium-230 is only one of many naturally occurring radionuclides in soil. A second example using COMPLY illustrates the problem with the 0.037nillirem/ year limit when all radionuclides in uncontaminated soil l are modeled. The naturally occurring radionuclides U-234 U235, U-238 Th-230, Th-232, Pa-231, Th-228, Ra-226, Ra-228, Rn-222, Pb-210, and Po-210 in soil for a 40 acre source (maximum impoundment size under Subpart W for mills) were evaluated. The methods in Regulatory Guide 3.59 were used to estimate dust emissions. The receptor was almost 800 meters away. The food pathway was l eliminated by distance. The point of release was assumed to be at the center j l r.f the 40 acres. For both examples, typical national values for soil l concentrations such as 0.6 pCi/ gram of Radium-226 were used. Western soils at

{ '

l 1

3 1 - - - - - . - - - - - - - _ l

i mills could easily contain concentrations at least a factor of two higher. The ,

resulting dose was 0.098 millirem / year (0.1 to an appropriate single digit l value). Even allowable surface contamination for packages shipped under DOT l rules could be a factor at this level for any type of facility.  ?

7. EPA should consider the most neutral conversion of the annual dose limits of i 25 millires to the whole body, 75 millires to the thyroid, and 25 millirem to other organs in 10 CFR Part 61 and 40 CFR Part 190 when discussing the matter for reader perspective. These uncertainties should also be factored in EPA's  !

discussion of the uncertainties in estimating doses and risks. Consistent with ICRP, NRC would view a 25 millfrem whole body equivalent dose as equivalent in risk to the 25/75/25 limits and a neutral conversion. The proposed and pending final amendments to 10 CFR Part 20 follow a whole body to whole body equivalent conversion in that doses in the new system are still 100 and 500 millfrem for members of the public. A 25 millirem whole body equivalent limit would be a (

better choice of limit for mills and low-level waste sites based on adopting i existing NRC and EPA rules instead of the 10 millirem proposed.

The conversion can be presented in two less neutral ways. The first way would result in a whole bo# dose equivalent of 51.5 millires. This figure results from the fact that the 25/75/25 limits work independently, i.e., the whole body dose cannot exceed 25 millirem, the thyroid dose cannot exceed 75 millires, and the dose to any other organ cannot exceed 25 millires. If the mix of radionuclides and pathways were assumed to give the whole bo# a dose of 25-millfrem, the thyroid 75 millf rem, and the six specific organs with weighting factors and the next five organs 25 millirem, the result is 51.5 millires whole body equivalent. The conversion is given in the following table:

Conversion of 10 CFR 61.41 Annual Dose Limits to Whole Body Equivalent 10 CFR 61.41 Equivalent Calculation (61.41 dose x weighting factor) 25 millirem whole body 25 x 1.0 = 25.0 and j 75 millirem thyroid 75 x .03 = 2.25 and 25 millirem other organs Gonads 25 x .25 = 6.25 j Red bone marrow 25 x .12 = 3.0 Bone surfaces 25 x .03 = .75 Breast 25 x .15 = 3.75 l Lung 25 x .12 = 3.0 Next 5 organs 25 x .06 x 5 = 7.5

. Total: 51.5 millirem I l

(Note that applicable to internal exposures only)

EPA usually discusses the conversion in a way to maximize the potential organ doses under the whole body equivalent system. EPA's method assumes that the entire whole body dose equivalent is available to the given organ. Using this assumption and the same weighting factors as in the table above, the maximum 4  !

____---- -- i

. 1 organ dose is calculated by dividing 25 millfrem by the weighting factor. The results are: 25 millirem for the whole body, 833 millfrem to the thyroid,100 millires to the gonads, 208 millfrem to the red bone marrow, 833 millirem to bone surfaces, 167 millfrem to the breast, 208 millfrem to the lung, and 417 millires to any of the next five organs. EPA uses these possible organ doses to conclude that a 10 millirem whole body dose equivalent corresponds more closely to 25/75/25 in the new system and presents 10 whole body equivalent as somewhat of a relaxing of the limits. The whole point of the new system is that even though organ doses may be higher, they are normalized or weighted l according to organ sensitivity and that the resultin organs is equivalent to uniform whole body exposure.g whole body risk from the B. Uranium Recovery Licensees BACKGROUND l

Uranium mill licensees are covered by Subparts I, T, and W of the proposed rule. Other uranium recovery licensees such as in situ facilities would be subject only to Subpart I. Subpart I covers emillions of uranium and thorium and daughters, except for radon, from all uranium recovery licensees. Thorium mill or recovery Itcensees would be covered by Subpart I. Subpart T wou1& also apply to all Title I mill sites being reclaimed by DOE. DOE's Title I tailings sites are not licensed by NRC until after the reclamation is complete, but' would be covered once the site is licensed. -

Uranium recovery licensees are already covered by a number of EPA and NRC codified standards applicable during operations and over the long term after sites are closed or reclaimed. For example, they are a art of the uranium fuel cycle and are already covered by 40 CFR Part 190. Note that 40 CFR Part 190 has been incorporated in NRC rules in 10 CFR 20.105( ) and 20.106(g) and applies to operational exposures and releases to the public, except for radon.)

Provisions of 10 CFR Part 20 not superseded by 40 CFR Part 190 are also applicable (e.g., radon releases to unrestricted areas). Management and disposal of uranium and thorium byproduct materials (the wastes from recovery of uranium and thorium) are explicitly covered by 40 CFR Part 192 and Appendix A to 10 CFR Part 40. Operational releases are also covered by 10 CFR Part 40 by provisions such as Criterion 8 dealing with ALARA at mills and yellowcake stack operation. Radon releases from operating mills are covered by work practice standards in the existing Subpart W of 40 CFR Part 61.

SUBPART I

1. The proposed rules have provisions that are inconsistent with many of the provisions of the codified rules identified in the background discussion. EPA has not prpsented cony'ncing arguments for why the combination of these

. existing regulations is not adequately protective. Examples of inconsistencies and deficiencies in rationale are presented in several of the comments which follow.

2. Subpart I would reduce the existing annual limits for nonradon emissions from the existing 25/75/25 millirem of 40 CFR Part 190 and 25 millfree whole body /75 milltrem organ limits of 40 CFR Part 61 to either 10, 3, or 0.03 millirem whole body equivalent per year. EPA hos argued here and elsewhere 5

that the 10 millires limit is actually less stringent than the 25/75/25 provisions, but as noted under comment 3 for low-level waste sites, this is not necessarily true. Compliat.ce with the 25/75/25 11mits in the past has also been demonstrated using ICRP 2 methodologies, not the ICRP 26 and 30 methodologies and using computer codes such as AIRDOS and MILDOS. If EPA used 40 CFR Part 190 data and methods, the results may not be valid using the new methodologies. NRC use of the new methodologies led to significant reductions in allowable concentrations in unrestricted areas for radionuclides of concern in uranium and thorium recovery activities in proposed revisions of 10 CFR Part

20. It is not clear that EPA used the latest methods in its risk assessment supporting the rule or reflected the impacts they might have in the discussion of risk uncertainties. Compliance with and the impacts of the more restrictive 3 and 0.03 millirem options were clearly not addressed.

3. The proposed compliance procedures require the use of NRC's Regulatory Guide 3.59, " Methods for Estimating Radioactive and Toxic Airborne Source Terms for Uranium Milling Operations." NRC staff agree that this document is appropriate for estimating radionuclides emissions for the variety of activities at the mills. It is sufficiently generic to also be useful for estimating nonradiological emissions at uranium mills and both radiological and nonradiological emissions for other types of extraction facilities.

4. EPA must provide more flexible and more realistic methods for demonstrating compliance than are currently provided by the COMPLY computer code. OtherwJse, it is likely that no mill can operate without major process or facility modification. All the code problems discussed in comment 2 for low-level waste sites apply except that all the radionuclides of concern are c~ overed in level 4 and the use of Regulatory Guide 3.59 provides a method for estimating emissions. NRC staff used the COMPLY code, Regulatory Guide 3.59, and model millparametersfromNRC'sgenericenvironmentalimpactstatement(NUREG-0706, GEIS) to calculate the potential doses from one major source of emission at a mill. Emissions and doses from the yellowcake drying stack were calculated using three sets of assumptions about the receptor and stack building. The results were 16.9,17.3, and 20.5 millirem / year. All of these values from this one source at an operating mill exceed the highest ifmit of 10 millirem / year proposed by EPA. The same source term and wind rose was used for each case.

The source term was based on percent of yellowcake produced to eliminate hours of operation as a variable. Midrange treatment methods were assumed. The nearest receptor was located at 1000 meters in each case. For the 16.9 and 17.3 millfres/ year cases, release heights of 20 meters, building height of 0 meters, and stack diameter of 0.5 were used. The 16.9 case tried to determine the dose with no food pathway by using distances of 8 million meters for vegetable, meat, and milk sources. The 17.3 case assumed vegetables grown at home and meat and milk at 2000 meters. Thus the food pathway appears to add little to.the dose based on comparing 16.9 and 17.3. Based on the GEIS and real sites *, vegetables at 1000 meters and meat at 2000 meters is reasonable.

Milk production close by is not. The 20.5 millirem / year case eliminated the food pathway as was done fcr the 16.9 case. Parameters on the building were varied from the other two cases to test the sensitivity, even though the distance to receptor is large. Building dimensions of 10 meters (H) x 15 meters (W) x 20 meters (L) were used. Building assumptions are more critical based on comparing 16.9'and 20.5. Receptors closer than 1000 meters (over a 6

half a mile) would obviously result in higher doses, except perhaps for radon, and greater noncompliance.

5. Code and machine difficulties were encountered in early efforts to perform sample calculations for level 4, which is the only option for low-level waste sites and uranium recovery sites. A dual drive IBM Personal Computer with DOS 2.10 will not execute at level 4. Data are accepted but messages such as Error at line 419 in DOSEIT4, Error number 3012, were repeatedly displayed. The same problem occured with the sample level 4 problem in Appendix 8 of EPA /520/1-89-003. Levels 1 to 3 ran without difficulty on the dual drive.

Level 4 operates without difficulty on hard. disk machines. EPA should caution users or resolve the problem. EPA should also develop guidance documents and training for the designated contacts to deal with such problems and other problems such as emission estimates and area and volume sources.

SUBPART W

1. The two major differences between the current Subpart W and the proposed revision are: 1) the timin new work practices and 2)the g for phasingDout Approach existing on prohibition impoundments and using the tailings production.

The same work practices are continued. .

Approaches A and B call for phase out two years from the effective date.

Licensees must obtain both EPA and NRC approval of the new icroundments. TAe proposed rule would remove the provision that EPA approval be obtained before applying to NRC. NRC supports the flexibility for review in parallel.

However, two years is a very short time for two regulatory actions to be taken and a new impoundment constructed. NRC is not sure that this is a realistic time. In agency coments on the existing Subpart W, NRC stated that major license amendments involving new technology and 40 CFR Part 192 requirements could take longer than a year and that at least two construction seasons are usually needed following approval of the design.

The Approach C requirement for use of the new technology on the effective date would obviously mean that only mills already meeting the new technology could operate. Others would have to wait until the new impoundments were approved and constructed (i.e., two years or more).

Approach D impacts to prohibit production of tailings are obvious. EPA should consider adding the flexibility to approve methods that could meet the levels should such methods be developed rather than an out right ban as proposed for Approach D. For example, some closed system method involving emplacement in underground mines or well injection might be developed that could limit releases. While it is true that alternate uranium recovery methods such as in -

situ mining are currently providing as much as half the domestic uranium production, the industry has been found nonviable for several years by the Secretary of Energy. Conventional mill production is a necessary component for a viable domestic uranium industry, which Congress has supported as a matter of national security in Section 170b of the AEA. If foreign sources were disrupted, conventional mills would be needed.

2. EPA should add the flexibility to consider alternatives to 40 CFR Part 192 requirements when approving new impoundments or disposal facilities. Congress 7

4 provided that NRC and Agreement States could approve licensee proposed alternatives provided they are equivalent to to EPA's standards to the extent practicable (e.g. Section 84c of the Atomic Energy Act.) The Tenth Circuit has upheld NRC's reading of Section 84c in a decision dated January 27, 1989. (

3. EPA claims that action under the CAA is necessary "...because there has been little, if any, action taken to dispose of the piles." EPA's information is not current. Two piles have been reclaimed. Eight are in the process of reclaiming ~the piles in accordance with approved plans.. (Note that total deconnissioning of a site, including disposal of the mill and the tailings, routinely ~ takes 5 to 10 years from start to finish.) One mill which is still in standby is actively reclaiming 80% of the tailings on site. Four mills operate at least intermittently and eight are in standby. Five mills have declared the intent to deconnission and reclaim the piles but are still in the ,

process of obtaining final approval for deconnissioning and reclamation plans. l A stable regulatory framework is an important factor for continued action to reclaim the impoundments. The changes in allowable flux in Subpart T may actually delay initiation or completion of closure while revised plans are developed and approved. Early closure which would be required by Subpart W may be offset by the new requirements in Subpart T.

4. EPA claims that setting an emission or air concentration standard is not feasible for operating mills. Since EPA appears to be able to set flux standards with precision in Subpart T, the rationale for this position is not ,

clear. In fact, the design standard for disposed tailings at Title I sites in 40 CFR Part 192 includes an ambient air concentration of 0.5 pCi/ liter at the site boundary. NRC sets concentration limits'for radon in 10 CFR Part 20 which apply to operating facilities. EPA should reconsider setting an air concentration value. Such an approach might simplify the extent of EPA review required in facility permits and reduce dual regulation. Compliance with the complex ground-water programs in 40 CFR Part 192(a) incorporated by reference in Subpart W, including ground-water monitoring and standard setting for hazardous constituents, must be explicitly considered by EPA based on the language in the existing and proposed Subpart W.

5. "Immediate" disposal of tailings is required for the continuous disposal method. It is not clear that this is always practical with the limited construction seasons where most of the mills are located and the size of the tailings impoundments. Existing uranium mill tailings impoundments are large when compared to other practices. New impoundments limited to 40 acres wou,d still be large when compared to hazardous impoundments of a few acres. Some acconnodations should be provided.

SUBPART T 1.'Subpart T p'roposes flux standards for radon emission from tailings impoundments that are inconsistent with the codified limits of 20 pC1/m2s in 40 CFR Pgrt 192 and 10 CFR Part g0. Approaches A and B specify211mits of 6 pCi/m s;, Approach C, 2 pCi/m s;, and Approach D, 0.02 pC1/m s. The flux limits include contributions from the cover material, not just the flux from the tailings. This is also inconsistent with the existing provisions that exclude contributions from cover materials. EPA should be consistent with the 8

e existing standards unless a clear unevaluated hazard or convincing rationale is developed.

2. EPA should either withdraw Subpart T or make it a design standard. Subpart T makes the flux limits a perform nce standard that must be measured rather than a design standard for the cover. This change is inconsistent with the codified rules and was rejected by both agencies in issuing existing rules.

The existing rules specify a design life of 1000 years for the covers, or at least 200 years if the longer time cannot be met. The proposed Subpart T does not include this ifmit and is thus inconsistent in this regard also. The proposed rule requires the impoundment to meet the flux as long as there is a NRC or Agreement State license and requires the licensee to measure it regularly unless exempted by the Administrator. EPA sPould not actively regulate the impoundments after disposal. As proposed, both NRC and EPA would be dual regulators forever. No standard or a design standard would accomplish this goal.

3. The flux levels themselves pose technical problems as the rule is structured. Since the radon from cover materials will be measured along with radon from the tailings, the background levels in soils likely to be used as radon2 barriers becomes more significant as the limits are lowered. Ag6 .

pCi/m s, typical western soil contributions of one or up to two pCf/m s could betojerated. Overburden used as cover would be severely restricted. At the 2 pC1/m s level, virtually no contribution froa the tailings themselvet could-be tolerated. At the 0.02 level, soil could not be used without a synthetic material over it. Synthetic covers would certainly require maintenance and replacement over periods of decades, much less hundreds or thousands of years.

It is illogical to impose such a burden to reduce emissions from piles to less than emissions from adjacent soils. Such long term repairs and costs are at conflict with UMTRCA and would use untried technology.

4. EPA does not address whether it expects Title II uranium mill licensees to provide long term care funds for DOE or the State, if the State will be providing long term care, for use by the government agency to conduct the radon testing, reporting, and more importantly take mitigative action. The proposed rule does not address financial assurances for any of the activities covered.

5. The rule and preamble indicate that a mill licensee can admit noncompliance instead of testing. The significance and consequences of this choice are not explained. Since EPA has not provided a means of providing alternative limits in the proposed rule, the intent of this provision is not clear. Under the rules, EPA can only grant waivers in time for compliance, not the levels. EPA should add a provision for site-specific alternative limits.

6. The definition of " operational" is inconsistent with the definition of

" operation" in Subpart W and NRC Ifcensing and poses practical implementation problems. In Subpart W " operation" ends when final closure begins. Closure is not included in Subpart T. Section 61.223 requires testing 60 days after the pile ceases to be " operational." This appears to require closure to be completed in 60 days, a totally impractical time period. EPA should provide a reasonable closure period. NRC licenses persons to process source material and dispose of the tailings in a specified manner. Strictly speaking, NRC does not license " piles." The term "nonoperational" is not defined in Subpart T, but is 9

N e

implied from " operational." If EPA adopts a design standard as recommended, these types of problems go away.

7. EPA states that ..."the dose / risk relationship for radon is well established..." so that it is easy to set flux limits under Approaches C and D.

The uncertainties may be an order of magnitude when factors such as actual particle size daughters may be attached to when exposure occurs, percent equilibrium with daughters, and differing human sensitivities and lifestyles are considered. It is not clear how these factors were considered or that the decision is so simple. In view of such uncertaintic, EPA might consider providing realistic error limits for the numerical haf ts.

8. Time and resources did not permit review of the specific measuring techniques for radon from the closed piles included in the proposed rule. The proposed flux limits range over two orders of magnitude. It would be unlikely that the same techniques would be equally applicable at all levels. EPA did not discuss detection limits and uncertainties in the imposed measurement techniques or justify their application. NRC staff is aware that it is difficult to reliably measure radon emissions from uncovered piles which are generally at least two orders of magnitude higher than the proposed closed pile limits. Measurements are complex and uncertain because of the release l mechanisms and large number of variables.

DETAILED COMMENTS ,

Based on the low-level waste and mill tailings examples, EPA is developing performance standards'that could represent significant long-term institutional and financial burdens 'nd a commitments. A quick scan of the provisions in 40 CFR Parts 60-69 indicates no general provisions to address financial assurances for these long-term commitments. This may be due in part to the assumption that the CAA applies to operational emissions that stop when operations stop. Such is not the case for waste storage or disposal activities. Piles, heaps, trenches, etc., represent area sources and emit, or potentially emit, radioactivity or other emissions over long periods of time, and in some cases, virtually forever.

EPA should consider developing new regulations similar to those provided for hazardous wastes to provide financial assurances for compliance and corrective measures. This is a generic issue, not just a NRC licensee issue and should be considered to provide a means to assure compliance for any standards where long-term (decades to thousands of years) emissions of chemical, physical, or l radioactive materials are being regulated. However, if EPA chooses to adopt Approach D, financial assurances would not help since EPA would be effectively '

closing down many industries with no legal way to store or dispose of existing inventories. Financial assurances would only make sense for achievable standards and viable industries.

4 u

4 10

f

'/ UMTED STATES 8- NUCLEAR REGULATORY COMMISSION WASHINGTON, D. c.20585

% APR 141989 MEMORANDUM FOR: Martin G. Malsch, Deputy General Counsel for Licensing and Regulation-Office of the General Counsel FROM: Frank J. Congel, Director Division of Radiation Protection and Emergency, Preparedness l- Office of Nuclear Reactor Regulation

SUBJECT:

COPMENTS ON EPA PROPOSED RULE, 40 CFR 61 l

L In response to your request of March 10, 1989, we have reviewed EPA's proposed rula (40 CFR 61) which would establish environmental standards for radionuclides as hazardous air pollutants under Section 112 of the Clean Air Act. Our review focussed on the reasonableness of the proposed standards, the

' impact of the four proposed options on nuclear power plants and research/ test reactors, and the difficulties in demonstrating compliance with the standard.

Our first comment relates to the applicability of the proposed standards. The rule, as presently drafted, does not clarify whether the standards are applid'a-ble to accident situations. 'In our view, the proposed'40 CFR 61 should not apply to releases of radioactive materials caused by an accident; rather, they j should apply only to releases arising from normal operations. The term " normal  ;

operation",'as we define it, would include affluents associated with antici-1 pated operational occurrences, i.e., those conditions which are reasonably I expected to occur during the operating lifetime of a facility.

Our other comments pertain to various provisions of the proposed standards and I are detailed in the enclosure. In summary, our comments touch on the following points: {

1

1. EPA could have justified a higher standard than 0.03,.3, or 10 mrem /yr based on recent standards promulgated by the ICRP and NCRP and the risk bases used by these agencies. To this end, we have suggested that a standard of 25-50 mres/yr could be justified for the air pathway's share of a larger generally applicab'le standard (100 mrem /yr seems justified).

2. EPA has compounded a conservative proposal by making it apply to a maxi-mally exposed individual, rather than to the average of a critical popula-tion group, as the ICRP has long-ago recommended (and recently reaffWmed).

CONTACT:

Thomas H. Essig, NRR 492-3147

Martin G. Malsch -2 APR 1419H

3. In its assessment of the reasonableness of the proposed standard, the EPA has defined model plants which are representative of nuclear power plants, as well as research and test reactors. Our assessment of the effluents released from a representative sample of nuclear power plants during 1987 indicated that, on the average, some nuclear power plants would not meet the 3 mrem /yr standard; a few might not meet the 10 mrem /yr standard. The 750 m distance used for the distance to the maximally exposed individual in the vicinity of a model research/ test reactor is unreasonable, in our judgement. The actual value could be 100 m or less, considering faculty and students as the population of interest. Thus, the relatively small Ar-41 releases from these facilities could, for some siting situations, have a considerably higher impact than that characterized by EPA.

4. In determining compliance with the standard, NRC licensees would have to

~ continually keep track of the moving target represented by the maximally exposed individual. A critical group approach, as discussed above, would appear to be less subject to fluctuation with time.

We appreciate this opportunity to comment on these important standards for radionuclides releases. If we can provide any additional assistance, please let me know.

Orig!nd dgacd by Rank J. Co%el Frank J. Congel, Director Division of Radiation Protection j and Emergency Preparedness Office of Nuclear Reactor Regulation

Enclosure:

Comments on EPA 40 CFR 61 1

i

i ENCLOSURE Comments on EPA 40 CFR 61 Rulemaking Office of Nuclear Reactor Reaulation

)

(

In reviewing the standards (40 CFR 61) proposed by EPA to implement the Clean Air Act, we focussed on three areas: a) the reasonableness of the proposed l

standards, b) the impact of the four options presented in the proposed stan- 1 dards on licensees under the purview of NRR, and c) difficulties with demon-strating compliance with the standard.

Reasonableness of the Standard

• The standard contains four approaches (10,10, 3, and 0.03 mrem /yr) to estab-lish an effective dose equivalent (EDE) for maximally exposed members of the general public. We have examined these four approaches in light of'recommenda-tions from other recognized standards promulgating organizations, viz., the

~

National Council on Radiation Protection and Measurements (NCRP) and the International Commission on Radiological Protection (ICRP).

i On the subject of risk associated with dose equivalent limits for individual members of the public, the ICRP stated in its Publication 43:

l l

" Members of the public are exposed to a large number of hazards and the i

magnitude of radiation risks must be considered in the context of other risks. ,The Commission states that: 'On this basis, a risk in the range of 10 s to 10 5 per year would likely be acceptable to an individual member of the public.'" l

.. a (This risk' range was used by tha ICRP to continue recommending a "...whole body Mos'e equivalent limit of 0.5 rem /yr, as applied to critical groups.....r-- -

In a similar. manner, the NCRP stated in its Publication No. 91 that:

"This recommendation [0.1 rem /yr] is designed to limit the exposure of members of the public to reasonable levels of risk comparable with risks from'other common sources, i.e., about 10 s annually."

Based on the above discussion and information contained in EPA rulemaking package, similar acceptable risks to the public have been used to reach marked-ly different recommendations for EDE standards:

Organization Annual Risk Recommended EDE Scope ,.

NCRP 10 s 100 mrem /yr all pathways EPA 2-3 x 10 8 10 mrem /yr air pathways only

- ~

ICRP 10 s.10 5 500 mrem /yr all pathways The risks referenced by NCRP and ICRP bracket that used by EPA, yet the recom-mended EDE is substantially lower. Thus, it appears that EPA would have been justified in recommending an EDE greater than 10 mrem /yr. If the EPA proposal had considered radiation exposures from airborne radioactive materials in the l

context of the total exposure from facility operation, i.e., exposures from air, water, and direct radiation pathways, it is our view that a recommended EDE on the or' der of 25-50 mrem /yr from the air pathway would be a justifiable share of an overall EDE of 100 mrem /yr.

2 1

,0ne other area which warrants comment in terms of the reasonableness of the standar'd is the point of application of the recommended EDE. The EPA has i i

recommended that the EDE be determined for the maximally exposed individual in the vicinity of each site.

4 The ICRP recommended in its Publication No. 26 that "The dose equivalent l

limits... apply to the dose equivalent...in the case of exposure of the popula-tion, to the averaos,of...these quantities over a group of individuals" (empha-sis added). The group of individuals to which the ICRP referred is the n-critical group. This group, as defined in ICRP Publication 43, is'to be

... representative of those individuals in the population expected to receive the highest dose equivalent; the group should be small enough to be relatively homogeneous with respect to age, diet, and those aspects of behavior.that affect the doses received." The ICRP further stated that "...the dose equiva lent limits are intended to apply to the mean dose equivalent in a reasonably homogeneous group...."; that "...the critical group would not consist of one individual nor would it be very large for then homogeneity would be lost....";

and.that "...in habit surveys it is not necessary to research for the most exposed individual within the critical group in order to base the controls on that one person. The results of a habit survey at a particular point in time should be regarded as an indicator of an underlying distribution and the value adopted for the mean should not be unduly influenced by the discovery of one or two individuals with extreme habits." Contrary to this guidance, the EPA stated in Appendix A to Volume I of the rulemaking package that "The location of one or more persons on the assessment grid which provides the greatest lifetime . risk (all pathways considered) was chosen for the nearby individuals."

l 3

f *C 4 .

In ,cnnelusion, we feel first that EPA has recommended EDE standards for the

' ~

generalpublicwhicharelowerthanthatneidedtoprovideIdequateprotection of the public from airborne releases of radioactive materials; and second, that EPA has compounded an already conservative standard by imposing it on the most highly exposed individual, considering all pathways in the offsite population.

Impact of the Proposed Standard Volume 2 of the 40 CFR 61 rulemaking package contains radionuclides release rates for a model BWR and model PWR in Tables 4-25 and 4-26, respectively. The values are shown in units of pCi/yr for several radionuclides. In mo-t cases, the values presented appear to be several orders of magnitude too low. The doses based on these releases (Table 4-28) are not, however, also on the low side. This discrepancy should be resolved, since these data may be used to '

support a conclusion that the 3 and 10 mrem /yr approaches to establishing an EDE standard can easily be met.

Because of the time constraints imposed by the comment period for the rulemaking proceeding, we were not able to compute the EDE for the maximally exposed individual at each nuclear power plant and research/ test reactor site based on actual effluents from each facility. Based on the evaluation of the i EDE for representative sample of nuclear power reactor plants using effluent releases during 1987, we conclude that most nuclear power plants could not meet the 0.03 mrem.EDE standard, assuming operation with currently installed equip-ment and use of current procedures. Using the same assumptions, we estimate l that many plants do not meet the 3 mrem /yr EDE standard. The foregoing conclu- J sions were based on a sample size of 15 plants. The EDE calculated for the 4

. , , ~

c:ximally exposed individual ranged from 0.04-10 mrem /yr for our sample; the mean EDE was 1.7 mrom/yr, with a standard deviation of 2.8.

Time constraints also did not permit us to perform an evaluation of the EDE nd received by the maximally exposed individual in the vicinity of research While we agree with EPA's statement that Ar-41 is the principal test reactors.

lly support the

-radionuclides released from theses facilities, we cannot genera statement that the maximally exposed individual typically resides 750 m from tha release point. Siting practices have been such that the faculty and student population are often situated much closer to the affluent release po On-campus dose rate measurements demon-(albeit on a non-continuous basis).

strated, for examp1'e, that certain students and faculty could receive up to Ar-41 released from a ,

.abcut 10 mrem /yr (credit taken for occupancy time) from i

university research reactor (now being decommissioned) which had been ope at only 100 kW; this exposure situation arose primarily due to the relativel dose proximity of classrooms to the affluent release point.

Difficulties in Demonstrating Compliance Because of requirements already imposed by the NRC on nuclear power plant licensees in the area of offsite dose calculations, no major difficulties are E

anticipated in the procedural aspect of demonstrating compliance with t standard; most, if not all, of the data needed as input to the EPA's COM Research code are already being gathered as inputs to licensee's dose models.

and test reactor licensees are not generally required to make offsite dose l

calculations, but should be able to with minimal difficulty using the COMl code.

5

o' What _ zay cause difficulty in some cases, however, is the need to be continually

' aware of locations of the maximally exposed individual from inhalation , immer-sion, and ingestion perspectives. For both power and research/ test reactors, this awareness process constitutes a moving target. Power reactors currently maintain an awareness of receptor locations in each sector in the vicinity of the plant, and are thus somewhat accustomed to living with such moving targets.

Research and test reactors, on the other hand, do not typically maintain an i awareness of distances to receptor locations in each sector; such a requirement would potentially pose a burden on them in maintaining an awareness of tran-

"~

sient, but very nearby, receptor locations. -

An alternative approach to determining compliance with the standard would be to determine the average exposure of a critical group, as discussed earlier. Once formulated, a critical group approach should be easier to implement and be -

relatively consistent as a function of time, yet provide an appropriate assess-ment of the EDE received by the more highly exposed individuals.

1 4

6 a