Press Release

Table of Contents

Acknowledgements

Main findings

Recommendations

I. Introduction

II. National Primary Drinking Water Regulations – Radionuclides
A. Bone dose estimation in ICRP 2
B. Bone dose estimation, present-day dose conversion factors
1. Bone doses according to FGR 11
2. Bone doses according to FGR 13

III. Conclusions

IV. Costs

V. Estimating the impact of residual radioactivity

VI. Other risks and radionuclides

References

III. Conclusions

The analysis in this report shows that the MCL for alpha-emitting, long-lived transuranic radionuclides should be tightened by about a factor of 100 — that is, it should be reduced from 15 picocuries per liter to 0.15 picocuries per liter. A combined standard for all alpha-emitting, long-lived transuranic radionuclides will simplify the rule and reduce the cost of its enforcement. Moreover, since the plutonium isotopes among these dominate the total curie content of DOE waste and since the dose conversion factors for Pu-238, Pu-239, Pu-240, Pu-242, and Am-241 are nearly the same, using Pu-239 as a reference for deriving the combined standard MCL is reasonable from a health standpoint as well as cost-effective.²⁷

In considering what should be the optimal value for a drinking water standard for alpha-emitting, long-lived transuranic radionuclides radionuclides, we have also examined the values for a plutonium-239 limit that exists in other standards. Specifically, the surface water standard of the State of Colorado is the most relevant, since that state has been host to one of the most important plutonium handling and processing facilities in the United States, namely, the Rocky Flats Plant, near Denver. The statewide standard for plutonium-239 for surface water is 0.15 picocuries per liter.²⁸ It is calculated on the basis of a 30-day rolling average — that is, 30 consecutive measurements are averaged; they may or may not be taken on consecutive days. Colorado's standard is based on the risk of one person in one million developing a cancer from consuming 2 liters of water per day for 30 years.²⁹

The Colorado Department of Health, Water Quality Control Commission describes the background and the rationale for changing from 15 picocuries per liter to 0.15 picocuries per liter as follows:

Background: The Commission previously adopted a basic standard for plutonium of 15 pCi/L and had no basic standard for americium. A basic standard was considered in this hearing for americium because it is closely associated with plutonium and these two radionuclides generally occur together. The current basic standard of 15 pCi/L plutonium was calculated using methodologies in the 1976 National Interim Primary Drinking Water Regulations and was consistent with a goal of keeping exposures below 4 millirem per year. The Basis and Purpose indicated that it was necessary and important to restrict levels because of the difficulty of removing this radionuclide by conventional treatment procedures and because the potential adverse effect on human health suggests that extreme caution be exercised in its release to State waters. Since plutonium is predominantly an alpha emitter, the basic standard was made consistent with the 15 pCi/L alpha standard....

Basis for Commission Decision: Since the previous basic standard was set, several changes have occurred: 1) a new methodology for assessing carcinogens has become the standard practice, 2) new data have resulted in periodic updates to the slope factors used in this methodology, and 3) a more refined Commission policy on appropriate levels of protection for carcinogens has been developed. This latter risk-based policy also parallels a national trend towards risk-based approach to environmental cleanup standards.

The 15 pCi/L dose-based approach was calculated using a "reference-man" and considered exposure during his working life. It was an approach designed to address questions related to occupational exposure. It did not consider sex, age and organ-specific factors over a lifetime. In contrast, the new slope factor methodology, used in EPA's 1989 Risk Assessment Guidance for Superfund Sites, is more complete, more applicable to a general population and has become the standard practice for calculating risk.

The Commission adopted a basic standard of 0.15 pCi/L for plutonium and americium, calculated using a 1 × 10^-6 risk level, based on residential use. This risk level is consistent with the Commission's policy for human health protection.³⁰

This reasoning is based on CERCLA, the Superfund law, but is qualitatively in accord with the reasoning in this analysis. Specifically, the central scientific point of the Colorado rule is that the science has changed, indicating greater risk than previously assumed from exposure to plutonium and americium; therefore the maximum contaminant limits should be adjusted accordingly. Further, the specific value for plutonium and americium recommended in the Colorado rule is just a factor of two lower than the geometric mean of the two values in the last two rows of Table 2 above.

In view of the complexities created by the change from NBS 69 to Federal Guidance Report 13, an MCL for alpha-emitting, long-lived transuranic radionuclides of 0.15 picocuries per liter is reasonable and justifiable. The action we are recommending is consistent with the intent of the National Primary Drinking Water Regulations as originally promulgated and is directly within the framework of the regulation as promulgated then and as it stands at present.

The primacy of the health goal (rather than numerical limits) is clear from the EPA's own description of the Safe Drinking Water Act, pursuant to which the radionuclide maximum contaminant limits are set. Its fact sheet on the Act states:

US EPA sets national standards for tap water which help ensure consistent quality in our nation's water supply. US EPA prioritizes contaminants for potential regulation based on risk and how often they occur in water supplies. (To aid in this effort, certain water systems monitor for the presence of contaminants for which no national standards currently exist and collect information on their occurrence). US EPA sets a health goal based on risk (including risks to the most sensitive people, e.g., infants, children, pregnant women, the elderly, and the immuno-compromised). US EPA then sets a legal limit for the contaminant in drinking water or a required treatment technique.³¹

By this standard, the 15 picocuries per liter limit for transuranic radionuclides is obsolete, not protective of public health, against the spirit of the Safe Drinking Water Act, and, as shown above, not in accord with the intent of the initial regulation. Because of this, the EPA should take up consideration of a tightened standard in its upcoming 2006 drinking water radionuclide review.

Corresponding to the change in the MCL for alpha-emitting, long-lived transuranic radionuclides, there is also a need for a change in the detection limit. Table B in 40 CFR 141.25 should be modified to include a separate detection limit of 0.01 picocuries per liter for each alpha-emitting, long-lived transuranic radionuclide. This detection limit is well within the capabilities of present-day techniques. The current detection limit for these radionuclides is 0.001 picocuries per liter, according to Argonne National Laboratory. The errors at such low levels can be large however. The error at 0.01 picocuries per liter, the recommended detection limit, is estimated by Argonne National Laboratory to be 10 percent.³²

We recognize that alpha-emitting, long-lived transuranic radionuclides are not ubiquitous in significant concentrations, unlike naturally occurring radionuclides like radium-226, thorium-232, and thorium-230. The vast majority of public water systems can therefore be exempted from routine monitoring requirements relating to alpha-emitting, long-lived transuranic radionuclides. The monitoring requirements for these radionuclides should be applied to public water systems that draw water from aquifers or surface water that have potential hydrologic or hydrogeologic connections to areas or facilities with waste tanks, waste burial pits, and other potential sources of alpha-emitting, long-lived transuranic radionuclides in combined totals in excess of 100 curies (see below).³³ Wastes disposed of at shallow and intermediate depths are included in this definition. Alpha-emitting, long-lived transuranic radionuclides that are contained in secure buildings with institutional controls would be exempt from this limit and the associated monitoring requirements.

We recognize that the main recommendation of this report, to set a separate standard for alpha-emitting, long-lived transuranic radionuclides, requires that the present gross alpha limit be split up into two parts — one for alpha-emitting, long-lived transuranic radionuclides and the other for naturally occurring alpha-emitting radionuclides. However, this is not a departure from the content or intent of the present rule, for several reasons.

First, the present rule itself does not have a single standard for alpha-emitting radionuclides. There is a sub-limit for radium-226 and radium-228 of 5 picocuries per liter. Since radium-226 is an alpha emitter, there is in effect a separate sub-limit for an alpha emitter up to maximum of 5 picocuries per liter (depending on how much radium-228, a beta-emitter, is also present). Second, the gross alpha limit excludes uranium and radon. The limit of 30 micrograms per liter of uranium is set on the basis of heavy metal toxicity. However, this amount of uranium causes some amount of harm as a result of its radioactivity. Recent science indicates that the harm from the heavy metal aspects of uranium may be reinforced by its radioactivity. (See Section VI. Other risks and radionuclides, below). Hence, reconsideration of a variety of issues is warranted. In such reconsideration, it would be practical and less costly to separate out alpha-emitting, long-lived transuranic radionuclides. This is because the vast majority of water systems will not require any testing for alpha-emitting, long-lived transuranic radionuclides since they are not at risk.

Footnotes

²⁸ Colorado Reg. 31, 2005. The State also sets standards for other radionuclides and considers different limits for different watersheds. We have not considered these issues, some of which result in more stringent and others of which result in more lax rules. We have simply used the State of Colorado's statewide surface water limit for Pu-239 as a guide for reference.

²⁹ CDPHE 2002.

³⁰ Colorado Reg. 31, 2005, pages 138-139.

³¹ EPA 2004.

³² ANL 1995, Chapter 7, Table 7.1.

³³ For instance, the 100 curie limit is equivalent to 1,000 metric tons of transuranic waste containing alpha-emitting, long-lived transuranic radionuclides at the lower limit of 100 nanocuries per gram. It would be equivalent to a larger mass of low-level waste, since the concentration in such waste (by definition) is less than 100 nanocuries per gram.