PUBLIC HEALTH ASSESSMENT

Plutonium 239 in Sewage Sludge
Used as a Soil or Soil Amendment in the Livermore Community

LAWRENCE LIVERMORE NATIONAL LABORATORY, MAIN SITE (U.S. DOE)
LIVERMORE, ALAMEDA COUNTY, CALIFORNIA

APPENDIX 3: PEER REVIEW AND OTHER COMMENTS AND RESPONSES TO PUBLIC COMMENT RELEASE

ATSDR has received nine sets of comments from various reviewers or sets of reviewers, including three independent peer reviewers. This appendix includes all of the comments that are specific to this public health assessment document along with the ATSDR responses to those comments. The comments have resulted in a number of minor revisions to the public health assessment and have improved the technical accuracy and readability of this document. The ATSDR responses specify how the document was revised relative to each comment or indicate why no change was made.

Peer Reviewer 1

1. An overview of the potential pathways of human exposure from the Pu-contaminated sludge is clearly presented (page 8).

The specific pathways for which doses were calculated from the Pu-contaminated sludge are stated (page 12) to include "external gamma exposure, inhalation of dust/soil, food, milk, and meat ingestion, drinking water, and soil ingestion". In Section 4, Public Health Implications, it is stated (page 21) that "The estimated doses from these soil Pu 239 concentrations include ingestion of soil, inhalation of dust, ingestion of food grown in the soil, and direct external radiation". It would be helpful to provide an explanation for the omission of the drinking water pathway. Data in the appendix appear to indicate that the dose from drinking water is insignificant, which would explain its omission.

ATSDR Response: The description of pathways on page 21 has been revised to indicate that drinking water ingestion was also included in the dose estimates. As you suggest, the drinking water dose is insignificant.

2. The RESRAD model is appropriate to assess doses resulting from Pu-contaminated sludge in soil. It is a well established, reputable model for performing assessments of radionuclides in soil. As noted in the assessment, RESRAD has been used by the U.S. Environmental Protection Agency, the U.S. Army Corps of Engineers, the U.S. Department of Energy, and the U.S. Nuclear Regulatory Commission.

ATSDR Response: Comment noted; no response necessary.

3. Data on gross alpha and Pu concentrations, which form the source term for this assessment, are adequate to show that doses from Pu contaminated sludge in soil would be significantly lower than the MRL See the response to Question 5.

The RESRAD model requires input of many parameters or the use of default values. It is stated that, with the exception of area of contamination and average wind speed, default values were used. This is a reasonable approach.

Dose conversion factors in RESRAD, which are based on Federal Guidance Report 13, were updated to values promulgated by the International Commission on Radiological Protection in its Report Number 71. This provides a strong basis for the dose conversion factors.

ATSDR Response: Comments noted; no response necessary.

4. Together, the main text and Appendix 5 of the assessment adequately describe the relationship between exposure to Pu and the associated health effects. There are uncertainties and differences in approach in evaluating the health effects from exposure to any source of ionizing radiation, and these are presented in Appendix 5.

For completeness, it is suggested that a review of the literature be made for studies of Russian Pu workers that have been published since the ATSDR's "Toxicological Profile for Plutonium" was completed in 1990. However, these studies would not effect the conclusions of this assessment since the Russian workers were exposed to high levels of Pu.

ATSDR Response: Note that the previous Appendix 5 has been deleted from this final version and the relevant information moved into the main document. Comment noted; no response necessary.

5. Using results obtained with the RESRAD model, the assessment determined that an average soil Pu concentration of 816 pCi/g would be required to produce a dose of 100 mrem per year, which is ATSDR's minimal risk level (MRL). (I did not independently verify this calculation.) The assessment was performed using health protective values for parameters in the model including intake rates and exposure durations. Thus, an actual dose would be expected to be lower than the calculated dose.

The maximum average monthly gross alpha concentration in the digester sludge was less than 816 pCi/g. Thus, if the gross alpha concentration in the digester sludge is used as an upper limit for the Pu concentration in soil the dose still would have been less than the MRL As noted in the assessment, the actual soil concentration of Pu would be much lower than the gross alpha concentration in the digester sludge. The gross alpha concentration overestimates the Pu concentration. Also, the concentration of Pu in the processed sludge would be significantly lower than in the digester sludge, and there would be further reduction in the concentration of Pu when mixed in soil.

It is clear from this assessment that any actual doses from Pu-contaminated soil were significantly lower than the MRL

Maximum Pu 239 concentrations measured in soil samples collected from areas where contaminated sludge was placed were less than 2 pCi/g. Even though these measurement results were not the initial Pu soil concentrations, the results provide additional support for the assessment.

ATSDR Response: Comments noted; no response necessary.

6. The conclusions and recommendations are appropriate in view of the potential doses as described in the public health assessment. Based on the assessed doses there would be "no apparent public health hazard".

ATSDR Response: Comment noted; no response necessary.

7. On page iv, second paragraph, second sentence: Should this be "The maximum average monthly --- "?

ATSDR Response: This was indeed the maximum monthly average concentration from digesters 1 and 2. However, monthly average values from digester 1 were higher than those of digester 2, so the reported value is not the only, or highest, maximum monthly average. The reported average from both digesters is the most representative value of overall digester sludge concentrations.

Suggest checking whether the dose presented on page 15, including Figure 2, is "Effective Dose Equivalent" as stated or "Committed Effective Dose Equivalent".

On page 25, fourth paragraph, first sentence: "medical accidents" is repeated.

ATSDR Response: The initial reference to "medical accidents" in this sentence has been changed to "medical exposures".

8. I am impressed with ATSDR's peer review process. The questions asked are the critical ones, and opportunity is provided for any other comments. The logistics of the review, including communications and timeliness, are outstanding.

ATSDR Response: Comment noted; no response necessary.

Peer Reviewer 2

9. The pathways described appear complete and well documented.

10. The RESRAD model as described in the public health assessment is appropriate for the evaluation of the doses. Although a printout of the RESRAD results is presented in Appendix 3, this reviewer could not verify the details of the calculations without an independent parametric study of the RESRAD code. However, the results presented appear reasonable and consistent with the input assumptions.

ATSDR Response: Comments noted; no response necessary.

11. This public health assessment document overstates the degree of certainty about the health protective nature of the use of gross alpha (particularly in sludge samples with high and variable solids) as an over estimate of the content of ^239-240 Pu. This is done on p. v , parag. 1 and p. 17, parag. 5. The latter statement states: " If the analysis process is accurately measuring gross alpha concentrations, the resulting gross alpha measurements are health-protective." This is a BIG IF! Measurements for 1973 (ref. Summary p. v) showing gross alpha is higher than the ^239-240 Pu is not sufficient to prove this occurs consistently for the earlier time periods. One would need information showing routine calibration of the gross alpha for energy level, showing the sample preparation techniques (control of solids) and routine q/a for gross alpha, none of which was presented and probably not available. One example of the difficulty was shown on p. 17. parag. 4, where the LLNL data is 3 times higher than the CDPH data.

Furthermore, the consistency of results does not imply accuracy as implied on p. 3 , parag. 2. Clearly, the historical gross alpha are the only data available and therefore must be used. However, we must be honest to state that there could have been a consistent bias that may never be discovered. The best we can say is that the approach used is the best available, given the data available.

ATSDR Response: The above comment is correct in noting that the quality assurance documentation for the historic analyses is sparse. However, the LRL Environmental Report for 1967 indicates that gross alpha counting efficiency is 34% for the 5.1 MeV decay energy of Pu 239. Further, the Shute letter (appendix 2) indicates that isotopic analyses of the 1967 release were conducted and confirmed the contamination as Pu 239 and Am 241. Although these statements do not provide proof that the gross alpha measurements over-estimate the Pu 239 concentrations, they do suggest that the LLNL measurement techniques adequately measured or corrected for the Pu 239 component of the gross alpha decays.

With regard to differences in the CDPH and LLNL gross alpha measurements, there are several potential explanations. First, when LLNL personnel realized that a release event occurred, they began extensive sampling of various locations within the LWRP. These included discrete daily samples from both digesters. The reported values are averages of all of the discrete samples. The available records from CDPH do not indicate any change in their sampling protocol which does not indicate the specific digester location and were taken as monthly composites (there were significant differences in digester 1 and 2 values). We have also requested any QA/QC data for the historic CDPH data, but as surmised, it is not available. Second, the counting efficiency of the CHPH samples is not specified and it is possible that a broader decay energy range was utilized that did not completely capture or correct for the Pu 239 component.

Finally, it should be noted that a difference in sample data of a factor of 2 may not be a significant difference. All sample data are subject to many sources of error and sampling error is typically many times larger than analytical error. We agree that uncertainty about the accuracy of the measured gross alpha values remains, which is why the above referenced sentences are qualified. However, noting that qualification, we believe the available data indicates that the gross alpha measurements are health-protective estimates of the Pu 239 concentrations.

The above referenced sentence(s) correctly states: "If the analysis process is accurately measuring gross alpha concentrations, the resulting gross alpha measurements are a health-protective estimate of Pu 239 concentrations."

12. This public health assessment adequately describes the relationship between doses and expected health effects. In particular, the justification for no anticipated health effects (even with the weaknesses in the data history) is reasonable. I agree that it is very unlikely that the sludge concentrations could have led to health consequences.

13. The supplemental use of educational materials will be an important addition.

14. I think that soil sampling or other regime would not yield information of high value.

ATSDR Response: Comments noted; no response necessary.

15. The peer review process would be more thorough if the reviewers could start earlier in the development cycle, perhaps by following the development of an assessment technique through its steps. In such a way, reviewers could interact with those developing the assessment technology and review decisions on the choice of models used, the choice of parameters, and feasibility of different approaches. This would lead to a more comprehensive review of the entire assessment process. The limited time available to this reviewer made such an approach impossible.

ATSDR Response: The PHA document is intended to be a complete assessment of the specific public health issues at a site. Consequently, the peer review is focused on a specific review of the resulting PHA document. While ATSDR regularly solicits external informal advice on technical aspects of its public health assessments, such a process is outside of the scope of the peer review of an individual document. Also, it should be noted that if any of the reviewers had identified significant problems in technical aspects of the assessment, the approach, methodology, or parameters would be accordingly revised.

16. While the document does talk briefly about the pathways (in only one sentence on page 8; third paragraph of the "Exposure Pathways" section), it might be beneficial to include a figure detailing the exposure pathways with a bit more text to be certain the reader understands specific pathways used for the assessment. The drinking water pathway is not considered?

ATSDR Response: The pathway components from the RESRAD calculations are listed in Section 3 (Exposure Assessment Method) and shown graphically in Figure A-2. As the descriptions are quite straightforward, an illustrative figure is deemed unnecessary. The drinking water pathway is explicitly included in the dose calculations (see the response to comment 1).

17. In my view, it is appropriate. However, I would think that the use of site-specific values (if available) would be more defensible than using the RESRAD defaults. I don't see it changing the final outcome, but it would just make the analysis more complete.

The use of 30 yrs and 70 yrs into the future isn't explained. I wonder about its validity, as well. Is this meant to be an exposure period? RESRAD calculates transport (and dose change) to those times, but what about further into the future in terms of exposure to the Pu in soils?

ATSDR Response: The RESRAD defaults are intentionally conservative and therefore health-protective. We concur that use of site-specific model parameters beyond those cited, will not significantly affect the resulting dose estimates, and as the defaults are health-protective, are unnecessary.

The 30 year exposure time period was used for several reasons; first it has been approximately 30 years since sludge was distributed to the community, second, 30 years is cited in the EPA Exposure Factors Handbook as an estimate of the time of residential occupation of a house. Also, it should be noted that the specific duration of exposure makes little difference in the annual doses (Figures A-1 and A-2). Seventy years is considered to account for a lifetime dose (EPA 1999). These explanations have been added to the appropriate sections of the document.

18. The document makes a strong case for the appropriateness of the data and the evaluation process. I think the conclusions are justified and the logic is sound.

ATSDR Response: Comment noted; no response necessary.

Peer Reviewer 3

19. It seems to do the job. However, the public health assessment (PHA) mentions "cancer, other illnesses, or death" in the summary and then again in the Introduction (page 2; end of first full paragraph). Yet, also in the summary (and later in the document), the conclusions of the PHA are that "the resulting doses will not cause sickness or death." I understand why cancer risk is not mentioned, but it might be best not to talk about cancer in the beginning unless the conclusions are going to address it in some way. I'd rather see the conclusions addressing cancer risk, but I understand that ATSDR wants to stand clear of the uncertainties of estimating risk.

ATSDR Response: The Public Health Implications section discusses the relative merits of evaluating dose vs. risk. The referenced sentences in the summary and conclusions ("the resulting doses will not cause sickness or death.") has been changed to "the resulting doses are unlikely to cause cancer, other illnesses, or death."

20. The conclusions and recommendations seem entirely appropriate. The PHA does a good job getting the idea across that while it's possible to reach the dose limit, it is overwhelmingly unlikely.

21. Page 2, 2^nd full paragraph, 1^st sentence: The way the sentence is worded makes it sound like the Pu-contaminated sludge is desired for some reason.

ATSDR Response: Comments noted; no response necessary.

22. Page 4, last sentence: It is not clear as to why it "follows" that the Pu remains in solid form since the liquid effluent is a non-detect. "Non-detect" might need to be defined in the text and a few more words should be added to bring the reader along.

ATSDR Response: The parenthetical statement " Pu 239 is below the detection limits of the analytical procedure" has been added to the sentence.

23. Page 7, 2^nd and 3^rd paragraphs: An annual dose is mentioned late in the 2^nd paragraph and the "levels of health concern" are mentioned in the 3^rd paragraph, yet there has been no mention to this point (except in the summary) as to what doses are pertinent to the limits of the PHA, or compared to background, etc. The 2^nd paragraph of the summary should be placed somewhere early in this section to prepare the reader in terms of magnitude of dose that is relevant.

ATSDR Response: The referenced discussion on page 7 presents the results of other studies as background information. As indicated, there is a prior presentation of relevant doses in the summary, as well as a more detailed discussion of doses of public health concern in the following section on "Public Health Implications". No changes to the document have been made in regard to this comment.

24. Page 11: "exposure dose" means what? In several places in the document, the word "exposure" is used in places where the word "dose" should be used.

ATSDR Response: The referenced sentence has been changed to "compare the estimated doses with those that have caused illnesses or death,".

25. Page 13, 3^rd paragraph: I was a bit disappointed when I got to this point and saw that, except for two parameters, the default values had been used.

ATSDR Response: See above response to comment 17.

26. Page 17, middle paragraph: It is a bit disturbing that the State and LLNL measured concentrations of sludge are so different (229 vs 606). Any thoughts on which one is incorrect, and why? Consistent differences are talked about, and trends are similar, but which one is thought to be correct? This area may need more discussion.

ATSDR Response: Please see above response to comment 11.

27. Page 19, Figure 3: Should the word "average" be included in the title? This figure only shows values as high as 300, yet there is talk of 606 and 674 pCi/g. It's a bit confusing as to where the concentrations are being measured, which concentrations are of importance (at the end-use site?), which are averaged, which are point estimates, etc.

ATSDR Response: All of the values used in this figure are explicitly referenced and described in Appendix 5 and represent ongoing and long term monitoring efforts of composited samples. The average gross alpha values of 606 and 674 pCi/g are averages of discrete samples that were collected in response to the release event. As such, the sampling methodologies are not comparable and there are no baseline values to indicate pre-release concentrations. Also please refer to the above response to comment 11.

28. Page 23, 3^rd paragraph: The statement that "the experimental evidence is inconclusive" in reference to plutonium causing an increase in cancer risk seems inappropriate. We have conclusive evidence of an increase in cancer risk from radiation exposure, albeit not from specific nuclides.

ATSDR Response: This sentence refers specifically to cancer induction from low dose exposures and has been so clarified. As stated, there is conclusive evidence of cancer induction at high radiation doses, however, extrapolation of such health effects at those high doses to low doses via risk assessment is based on the LNT which has not been conclusively documented.

29. Page 24, 1^st full paragraph, 1^st two sentences: as with the last comment, it seems inappropriate to be stated as nuclide dependent.

ATSDR Response: The radiological doses estimated and evaluated in this public are necessarily nuclide specific. This section reviews the relevant toxicology of plutonium 239 and plutonium 240.

30. Page 25, bottom: The first sentence of the "Health Effects from background radiation" seems a bit bold; also, the paragraph goes on to say that chromosomal aberrations were found in the Chinese population exposed to high background levels.

ATSDR Response: This statement from the ATSDR Toxicological Profile on Ionizing Radiation, "Since this annual dose of 3.6 mSv/year has not been associated with adverse health effects or increases in the incidences of any types of cancers in humans or other animals, the 3.6 mSv/year is considered a NOAEL for purposes of MRL derivation." Regarding the study of the Guangdong Province in China, a research study by Wei (1986) is summarized by the NRC (1990) and Eisenbud and Gesell (1997) who conclude that "no effects were found in a study of about 80,000 people who have lived for generations in an area in which the soil is enriched in thorium" The NRC summary also states that "Chromosome aberrations and a higher reactivity of T lymphocytes were found in individuals in the high natural background area", but that there was no increased rate of cancer and that there were no differences for a large number of hereditary diseases or congenital defects. Apparently, the chromosomal aberrations are either not significant or do not result in health effects.

31. Page 27, 1^st full paragraph, 5^th sentence: the way the sentence is written, it sounds like "time periods" (?) and intake rates are included in the dose factors. This isn't true. I'm guessing that the sentence refers to those things being included in the overall assessment.

ATSDR Response: The referenced sentence has been changed to "Other exposure factors include exposure durations, body and organ weights,".

32. Page 27, 2^nd full paragraph, 3^rd sentence: "perceived" risk? Risks estimated based on radiation dose are certainly not "perceived". They may be uncertain estimates but it's not a perception.

ATSDR Response: "Perceived" has been replaced with the term, "numerical estimation of risk."

33. Page 28, first full sentence: The sentence is not precise; it says essentially, "studies have demonstrated that nothing has been documented"how can that be?

ATSDR Response: The sentence specifically and correctly states that "no adverse health effects have been documented for exposures of less than 360 mrem/year". This sentence has been changed to "no adverse health effects have been documented from doses less than 360 mrem/year"

34. Page 28, end of same paragraph: Seems like a bit of hand-waving is going on when justifying 100 mrem/yr.

ATSDR Response: The rationale for use of the 100 mrem/year MRL is summarized in the preceding section on page 23. A complete derivation of the MRL is presented in the referenced Toxicological Profile on Ionizing Radiation (ATSDR 1999).

35. Page 29, 2^nd full paragraph: It could be important (for comparison sake) to state the average water content in the wet-weighed sludge samples.

ATSDR Response: The available information does not provide any indication of the water content of those samples reported as "wet weight" analyses.

36. Page 32, 2^nd sentence: how can the "determination of sample locations [be] highly uncertain"? If you have a sample location, you know where it is. The sentence is just not worded precisely.

ATSDR Response: This sentence has been changed to "determination of appropriate locations to sample"

37. Page 32, 2^nd full paragraph: This states that the depth of the contaminated area is 1 meter (but doesn't is say 6 feet in the summary?).

ATSDR Response: The RESRAD calculations assumed a contaminated soil depth of 2 meters (~6 feet). This sentence has been so revised.

38. Pages 83 and 84: The contrast on those Figures needs to be changed so they're easier to read.

ATSDR Response: These figures have been reproduced in color to correct the contrast.

39. Pages 85-91: The contrast in the table headings needs to be changed. Also, in that table, it seems odd to list the values in the "LLNL Dried Sludge" column (for example) multiple times when they are simply 6-month averages. Can the value be entered once with a line through the 6-month period to indicate that it's an average?

ATSDR Response: The table headings have been changed to enhance readability. The repeated values in the spreadsheet are necessary for the graphical presentation in Figure 3, which is produced from the spreadsheet values.

40. Page 94, last paragraph: The issue with the LNT model isn't individual risks, race, sex, radiation type, etc. The issue is whether or not the LNT model is the correct dose-response function at low dose.

ATSDR Response: Comment noted. ATSDR agrees that the primary issue is how the data are evaluated at the range of low dose rate/low dose and then how these data are presented. Furthermore, we believe that the human variability also impacts the dose response curve within a population. This paragraph has been revised accordingly.

41. Page 96: Figure A-3: better contrast is needed.

ATSDR Response: The contrast has been adjusted to improve legibility of this figure.

42. Page 99: It is unclear as to why one calculation results in 2.5 and other results in 2.9. Which is correct? Which was used for this assessment (2.5)?

ATSDR Response: The difference in these soil screening levels of Pu 239, as addressed in the footnote 18 on page 99, is due to changes in the dose conversion factors at the time that each calculation was undertaken. Neither value, except by reference, is explicitly used in estimating doses or assessing potential health effects in this assessment.

Reviewer 4

43. CDHS, Radiologic Health Branch does not use a dose value of 25 mrem/year (above background) as being health protective. California has a dose limit of 100 mrem/year for the members of the general public.

In November 2001, CDHS/RHB adopted the dose limit of 25 mrem/year, with the ALARA approach, for release of unrestricted use of the decommissioned facilitlies. However, such regulation was struck down by the Court in June 2002. At the present time, California does not have a dose based decommissioning regulation; but has a dose limit of 100 mrem/year for the members of the general public.

The 25 mrem/year reference in paragraph 1 and 2, page 21 of the Assessment should be corrected/removed.

ATSDR Response: References to the California dose limit of 100 mrem/year have been changed as suggested.

44. Figure 3 incorrectly states the CDHS "digester sludge values from June-64 through March-65were analyzed as wet weight samples" These samples were dry weight samples as correctly discussed in the "Adequacy of Available Data for Public Health Determination" section, page 29. The statement in Figure 3 should be corrected to reflect dried sample results after May 1963, by removing the last sentence. Appendix 4 correctly reflects the CDHS digester sludge sample results.

ATSDR Response: The last sentence of the Figure 3 caption has been deleted.

Reviewer 5

45. ATSDR has concluded that the different data sets produce similar results and gaps in any particular data set does not "present a critical lack of information". We do not agree with this conclusion for the following reasons:

• There are no data available until January 1960--Sludge was given to the public as early as 1958, which represents a considerable amount of time when levels of gross alpha or plutonium in sludge are not known;

ATSDR Response: The annual gross alpha concentration in sewage from the LLNL site for 1959 is included in a letter from L. Beaufait (US AEC) to Dr. John Heslip (CDPH, Bureau of Radiological Health). An annual value of less than 38 pCi/L is documented for calendar year 1959. While there is no documentation regarding sampling or analytical procedures for this annual value, it does indicate that the sewage was monitored for gross alpha concentrations and there were no apparent releases of the magnitude of the 1967 event. Also, Figure 3 (and Appendix 5) does show measured concentrations of dried sludge for the 1961 calendar year. Considering that any elevated gross alpha concentrations in dried sludge would have to have entered the treatment system at least one year prior to the time of sampling also suggests that there were no significant plutonium releases in the 1958-59 timeframe. We agree that there is some uncertainty regarding the documentation and adequacy of monitoring during the early years of LLNL operation, however, the available data do not indicate any plutonium releases approaching the magnitude of the 1967 event during this time period.

• A discussion as to what constitutes similar results between data sets would be helpful. The data sets used in the evaluation consist of: monthly gross alpha measurements collected by the CDHS/RHB; summary data provided in Lawrence Livermore National Laboratory (LLNL) (formerly Livermore Radiation Laboratory - LRL) semi-annual and annual reports and; some data that was recently (February 2003) released by LLNL in support of LRL environmental reports. The majority of the recently released data provide information relative to 1967, 1969 and 1970. In 1967, these data show gross alpha measurements 2-7 fold higher in digester sludge between June and November, than reported by the state for the same time period. The data provided for 1969 appear to be consistent with the RHB data, for the months reported. However, there are extended time periods (ranging from 4 months to over 2 years) when monthly samples were not collected by the state as a means of comparison and only LLNL summary data were used, and;

ATSDR Response: A discussion of the similarity and utility of the various data sets in included in the section of "Adequacy of Available Data for Public Health Determination". See the above response to comment 11 for comparison of the data values from different sampling protocols.

• There is a great deal of inconsistency in the LRL environmental reports for the years 1961-1968. In many cases only summary statements are made with no supporting data. Without the actual measurements as provided for 1967 and 1969, consistency between the data sets cannot be adequately evaluated or assumed.

ATSDR Response: We agree that, due to changes in sampling or reporting procedures, there is significant uncertainty in any of the individual data sets. However, the temporal consistency of the collective data available provides an adequate basis for the public health assessment. The 1967 release created a gross alpha signature that lasted for about 8 months in the digesters and several years in the dried, processed sludge. Considering that the most significant Pu 239 release did not present a public health hazard, it is very unlikely that the data gaps are obscuring any larger release events.

46. The following comments relate to the toxicology of Pu 239:

• In the Toxicology of Pu 239 section, the discussion of the carcinogenicity of Pu 239 is vague and leads the reader to believe that plutonium has only been shown to cause cancer in laboratory animals. The International Agency for Research on Cancer (IARC) states, "Dose-response relationships have been demonstrated for cancers of the lung, liver, and bone in both men and women exposed to a broad range of doses", and thus IARC has classified plutonium as carcinogenic to humans (Group 1). CDHS suggests that clear language and presentation of the available information be used in the PHA;

ATSDR Response: The suggested sentence has been added to the section on "Toxicology of Pu 239/240".

47. On page 24, ATSDR makes "observations" based on a review of the toxicological profile on plutonium and states, "Neither deaths due to respiratory disease nor reduced respiratory function have been reported among the occupationally exposed cohorts". This statement does not appear to be supported by the literature and is in direct contradiction to IARC (see comment above);

ATSDR Response: This quoted statement from the Toxicological Profile is specific to non-cancer respiratory diseases.

48. In another observation, ATSDR states, "In workers with long-term exposure to plutonium (including workers at Los Alamos National Laboratory, Rocky Flats nuclear Plant) no studies have demonstrated an unequivocal association between exposure to plutonium and cancer mortality 30 or more years after the exposures occurred". It is unclear as to what constitute an "unequivocal" association. Studies of Rocky Flats plutonium workers have shown an increase in cancer, though not statistically significant (ATSDR 1990). It is important to describe the limitations of epidemiological studies of small populations, with regard to showing significance and factors such as the "healthy worker effect" that can bias results. For example, in a recent study of Rocky Flats workers, conducted by Dr. James Ruttenber at the University of Colorado Health Sciences Center, it is suggested that the association between lung cancer and plutonium occurs at levels currently considered safe. The study points out that in general, Rocky Flats workers had lower rates of cancer than the general population, but suggests this may be attributed to the healthy worker effect. The study showed that lung cancer victims had higher levels of plutonium inhalation than the control group.

These comments are pertinent to other "observations" listed in this section (Toxicology of Pu 239, page 24). It would be helpful to the reader for ATSDR to provide additional information from the available literature and be careful not to make over generalizations about the toxicology of Pu 239.

ATSDR Response: ATSDR has received the University of Colorado study and we note that the report reviewed over 16,000 workers. The authors state that of 180 lung cancer deaths matched to 720 controls, there were 93 cases with no lung dose and 87 workers with a lung dose. Furthermore, of the 87 workers with a measurable lung dose, 33 cases of lung cancer were observed at a lung dose of 10,000 millirem or less. This is an organ specific dose - we estimated a dose to the whole body. To estimate the dose to the lung from our estimated doses, one can use the lung weighting factor of 0.12 (the lung dose is about 1/10 of the whole body dose). The researchers identified significantly increased risks when the internal cumulative lung dose was in excess of 400 millisieverts (40,000 millirem). As with other worker studies, these doses are many orders of magnitude greater than the doses estimated from potential exposures to the Pu-contaminated sludge. Additional information on this report is at http://www.cdphe.state.co.us/rf/rfpworkerstudy/index.html.

49. The following comments relate to the discussion and use of a 100 mrem/year dose:

• ATSDR states that use of a 100 mrem/year dose is public health protective. According to the United States Environmental Protection Agency (USEPA), the effective dose limit of 100 mrem/year is not considered health protective because it equates to an "unacceptably high" increased cancer risk of approximately 2 in 1000 (USEPA 1997). The USEPA states that levels of 15mrem/year effective dose equivalent (which equates to approximately a 3x10^-4 cancer risk) or less are protective and achievable (USEPA 1997). CDHS agrees with the USEPAthat exposures resulting in a 2 in 1000 cancer risk are unacceptably high and thus we support the most public health protective approach;

ATSDR Response: The derivation of ATSDR's 100 mrem/year MRL, as described in The Toxicological Profile on Ionizing Radiation (ATSDR 1999), is based on a weight of evidence methodology. Problems associated with use of risk assessments for public health determinations are presented in the Public Health Implications section. It should also be noted that the above cancer risk estimates are based on the Linear No-Threshold dose response model and the revised EPA document on Cancer Risk Assessment promulgates a weight of evidence approach in lieu of the rote application of the LNT model. Also, note that comment 43 indicates that the State of California uses100 mrem/year as the currently accepted dose limit for members of the general public.

50. ATSDR has expanded the definition of the minimal risk level (MRL) (a non-cancer health comparison value) for ionizing radiation to be inclusive of cancer health effects. In doing so, ATSDR is asserting that there is a threshold level for cancer (100mrem/year). This is in direct contradiction to the predominate view within the scientific community that there is no threshold level for carcinogens;

ATSDR Response: As discussed in the ATSDR Toxicological Profile for Ionizing Radiation, the scientific literature is very clear that a radiogenic threshold exists for the induction of cancer. This is clearly seen in the atomic bomb survivors, radium dial painters, uranium miners, and medical patients. The linear non-threshold (LNT) THEORY is appropriate for use in setting regulatory standards with necessary safety factors. However, the LNT theory is not a valid scientific model for evaluating health effects for environmental exposures. This is also discussed in the ATSDR Cancer Framework Policy. There is not a consensus for LNT either for chemical or radiological carcinogenicity. The ATSDR Cancer Framework and the EPA Cancer Risk Assessment (EPA 2003) policies recommend use of a weight of evidence approach rather than rote application of the LNT theory for cancer evaluation. Both of these policy documents have been peer reviewed and may be considered to represent a consensus approach to cancer risk evaluation.

An MRL is an estimate of the daily human exposure to a hazardous substance that is likely to be without appreciable risk of adverse non-cancer health effects over a specified duration of exposure. The MRL is derived from exposure levels observed to produce adverse effects, or from exposure levels that have not produced adverse effects, with uncertainties (or safety factors) incorporated into the value. Thus, MRLs are intended only to serve as a screening tool to help public health professionals decide which release situations require more extensive evaluation. While estimated exposure dose levels below an MRL are not likely to produce non-cancer adverse effects, exposure estimates above an MRL do not mean that adverse effects will occur. ATSDR then evaluates the potential for adverse health effects in an exposed community by comparing levels known to produce adverse effects to the estimated site-related doses. This margin of exposure (MOE) approach, along with an evaluation of available epidemiologic, toxicologic, and medical data, is used by health assessors as part of the public health determination to reach qualitative (rather than quantitative), risk-based decisions about hazards posed by site-specific conditions of exposure.

On the basis of an extensive review of the health studies and documented health effects from radiological exposures, ATSDR established an MRL of 400 mrem for acute duration (14 days or less) of external exposure (ATSDR 1999) to ionizing radiation. The acute MRL is based on external dose levels that did not produce behavioral and/or neurological effects on the developing human embryo and fetus. Similarly, a chronic duration (a year or more) external exposure MRL of 100 mrem/year (above background) has been established based on radiation doses that have not produced observable detrimental health effects in humans. Thus, the ATSDR acute and chronic MRLs for ionizing radiation (including plutonium) are based on doses with No Observed Adverse Effect Levels (NOAELs). While ATSDR MRLs typically include only non-cancer health effects, all the studies on which the chronic MRL for ionizing radiation are based included cancer as the specific end-point. Consequently, the chronic MRL for ionizing radiation is considered protective for both cancer and non-cancer health effects.

51. The following comments relate to the conclusion and recommendation section:

• We agree with ATSDR's conclusion that exposure doses probably do not exceed 100 mrem/year, for the time periods when data are available. However, there are periods of time exceeding two years, when there is no data. Therefore, potential exposures are unknown, prohibiting any health conclusions from being drawn for these time periods;

ATSDR Response: Please see the response to comment 45.

52. ATSDR concludes, "any additional sampling data will be subject to the same types of uncertainties as existing historical data". The uncertainties in the historical data relate mostly to sample collection methodology, radionuclide-specific analysis, quality assurance and control. Any additional sampling in the future would be conducted using approved methodology and would not be subject to these same uncertainties;

• Given the uncertainties with existing data and the data gaps, CDHS recommends a process be implemented that identifies residents who received sludge and offers sampling.

ATSDR Response: All sample data are subject to uncertainty due to sampling and analytical error. We agree that it will be much easier to control aspects of analytical error in relation to future sampling. However, analytical error is typically very small in relation to errors associated with identifying and sampling the appropriate locations (eg., collecting spatially representative samples). Uncertainty associated with identifying specific areas of sludge distribution and modification of those areas by mixing, tilling, etc., are greatly increased relative to historical sampling of documented areas of sludge distribution. So while the minor analytical error component may be controlled and decreased, the much larger locational error component will be greatly increased. Considering that maximum historical sludge concentrations do not result in doses of radiological health hazard, there is no public health basis for conducting additional, necessarily uncertain, soil analyses.

Reviewer 6

53. ATSDR concludes "Pu 239 [plutonium 239] from LLNL was released to the Livermore sewer system and resulted in the contamination of LWRP [Livermore Water Reclamation Plant] sludge which may have been distributed to the Livermore community resulting in areas with Pu 239 concentrations above background." We agree with this conclusion and with certain facts that ATSDR presents in its report. These are:

• LLNL routinely released plutonium into the Livermore sanitary sewage system;
• LLNL also released non-routine quantities of plutonium into the Livermore sanitary sewage system on at least two occasions (1964 and 1967);
• A result of LLNL's releases was plutonium contamination of sludge at the Livermore Water Reclamation Plant (LWRP);
• Plutonium-contaminated sludge was distributed to community residents;
• Many Livermore residents may have been and continue to be exposed to Pu-239 contaminated soil.

ATSDR Response: Comments noted; no response is necessary.

54. The methodology used by ATSDR to assess the potential public health impacts of plutonium-contaminated sludge is biased in the direction of finding that no hazard exists. There are at least four major sources of bias in ATSDR's approach. Together and separately, each of these biases results in the minimization of the potential hazard.

ATSDR assumes that there is a "threshold" dose of radiation below which there are no adverse health effects.

ATSDR states, "doses of public health concern" are doses "that are capable of causing adverse health effects, such as cancer, other illnesses or death" (page iii). A Minimal Risk Level (MRL) of 100 mrem/year (above background) is the criteria for radiologic "doses of public health concern" because according to ATSDR, "no adverse health effects have ever been documented from radiologic doses of 100mrem/year or less (above background)" (page iii). ATSDR's definition of an MRL is the dose of radiation "likely to be without a measurable risk of adverse, noncancerous effects" (page 45), although ATSDR extends this definition to include cancerous effects in the case of ionizing radiation (page 21). ATSDR concludes that "while exposures [to plutonium-sludge] may have occurred, or may still be occurring, the resulting doses will not cause sickness or death" (emphasis added) (page 34). Therefore, ATSDR's conclusion is based on its determination that exposure to radiation at levels less than 100 mrem/year above background will not cause sickness or death. This is the equivalent of a "threshold" for exposure to ionizing radiation, below which level there will not be health effects.

ATSDR's interpretation of health risks from radiation exposure contradicts the standard practice as set forth by the National Academy of Science BEIR V Report (NAS 1995), international commissions (International Commission on Radiation Protection 1991 and UNSCEAR 2000), and ATSDR's 1999 Toxicological Profile for Ionizing Radiation, which is to assume that there is no threshold dose below which health effects of ionizing radiation do not occur. In other words, the amount of evidence concerning radiation health effects is large and it substantially supports the idea any amount of radiation can be harmful.

Therefore, ATSDR's evaluation of the ionizing radiation dose-response curve is not health protective and is inconsistent with current scientific and public policy assumptions regarding radiation and health risk. ATSDR's a priori assumption that there is a "threshold" dose of 100mrem/year, below which "the resulting doses will not cause sickness or death" (page v) means that all potential health impacts of radiologic doses less than 100 mrem/year are ignored by ATSDR's design. This is a significant source of bias in the direction of falsely minimizing the potential health impacts of exposing large populations to a known carcinogen for generations to come.

ATSDR Response: ATSDR, as outlined in its Cancer Framework Policy published in 1993 clearly states that ATSDR does not perform risk assessments. Therefore, the public health assessment was not designed to evaluate the risks associated with radiological releases. As the Cancer Framework Policy states "algorithmically derived numerical risk estimates tend to be conveyed in an artificially precise manner and sometimes used inappropriately in decision-making." Conversely, this public health assessment does address the radiological doses created by exposures to the Pu-contaminated sludge and whether those doses are at levels likely to create any adverse health effects.

As discussed in the ATSDR Toxicological Profile for Ionizing Radiation, the scientific literature is very clear that a radiogenic threshold exists for the induction of cancer. This is clearly seen in the atomic bomb survivors, radium dial painters, uranium miners, and medical patients. The linear non-threshold (LNT) THEORY is appropriate for use in setting regulatory standards with necessary safety factors. However, the LNT theory is not a valid scientific model for evaluating health effects for environmental exposures. This is also discussed in the ATSDR Cancer Framework Policy.

55. ATSDR's implicitly assumes "acceptable risk" to be in the range of 1 additional case of cancer among every 300 persons exposed.

ATSDR acknowledges that exposure from plutonium in sludge occurred and sets 100mrem/year above background as the criterion for doses of public health concern. ATSDR reports (in Appendix 5) that the Nuclear Regulatory Commission (NRC) has established a 100mrem/year dose limit for the general public, and that the NRC's dose limit is associated with a risk of 1 cancer for every 300 people exposed. ATSDR also reports that in1990, the National Academy of Science BEIR V committee estimated that out of 100,000 adults exposed to 100 mrem/year of radiation over a lifetime, anywhere from 410 to 980 men and 500 to 930 women might die of cancer caused by this exposure. We would add that John Goffman, a former Director of the Biomedical Research Division of LLNL contends that BEIR V underestimates the damage caused by low-dose radiation.

ATSDR Response: As discussed in the ATSDR Toxicological Profile for Ionizing Radiation, the scientific literature is very clear that a radiogenic threshold exists for the induction of cancer. This is clearly seen in the atomic bomb survivors, radium dial painters, uranium miners, and medical patients. The above estimates of cancer risk are based on application of the linear non-threshold theory (LNT). The linear non-threshold theory is appropriate for use in setting regulatory standards with necessary safety factors. However, the LNT theory is not a valid scientific model for evaluating health effects for environmental exposures. This is also discussed in the ATSDR Cancer Framework Policy.

The currently accepted risk estimate for radiation induced cancer is 0.0005 per rem per year. However, the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) suggests that uncertainties in cancer risk estimates may be about twofold higher or lower for acute doses where cancer risk can be directly assessed and a further factor of two (higher or lower) for the projection of these risks to very low doses and low dose rates. Therefore the uncertainty can range from about 0.000125 to 0.002 per rem per year.

An assumed dose of 100 mrem/year and using the 0.0005 per rem estimate results in an annual estimated cancer risk of 5 in 100,000, which is well within the EPA acceptable risk range of 1 in 10,000 to 1 in 1,000,000. Further, because risk is a policy statement based on the LNT, rather than an assessment of health effects, and no adverse health effects have ever been associated with normal background radiation doses (~360 mrem/year), these cancer risks are unlikely to translate into any real adverse health effects.

ATSDR correctly states "federal radiation doses and risks are widely distributed even among the same agency, dependent on mission" (page 94). We would addthat evenwithin ATSDR's own report, there appears to be a discrepancy between the methodology used by ATSDR which is explicit in using the MRL as a predictorof noadverse health effects, and ATSDR's definition of an MRL which states, "An MRL should not be used as a predictor of adverse health effects" (page 45).

ATSDR Response: Please see response to above comment.

ATSDR contends that it is using an estimation of dose, not risk as the basis of its assessment. ATSDR contrasts "dose" which is a measure of exposure, with "risk" which is a statistical concept related to the chance of an adverse health impact occurring. ATSDR's stated rationale for not translating estimated doses to risk is its concerns about the inherent limitations of quantitative risk assessment. We share many of the concerns expressed by ATSDR (page 98) regarding the limitations and uncertainties associated with quantitative risk assessment, including: the influence of the model and exposure duration selected on the risk projections, uncertainties associated with extrapolation from high doses to low doses, difficulty in including differences among populations in risk projections (i.e., risk projections based on worker studies and Japanese bomb survivors may differ), and inability to include non-cancer health outcomes in risk projections (i.e., genetic effects, teratogenic effects, and length of loss of life, etc).

To these concerns we would add that quantitative risk assessments look at only one chemical or radioactive exposure at a time. In contrast, exposure to plutonium-contaminated sludge is occurring along with exposure to many other radiologic and chemical exposures, some from LLNL and others non-site related. Together, these exposures may have greater adverse health impacts that predicted by models that look only at a single agent.

However, ATSDR's public health assessment does not live up to its self-described "dose-only" approach, but has embedded a risk assessment in its dose assessment. ATSDR's a priori decision that doses less than 100mrem/year "are not of public health concern" and "will not cause sickness or death" implicitly establishes a negligible risk level for exposure to less than 100mrem/year. The dose associated with ATSDR's implicit determination of "negligible risk" is the same dose estimated by the Nuclear Regulatory Commission to be associated with a risk of 1 additional cancer among every 300 people exposed. This is not a health protective methodology. IF ATSDR prefers to limit its analysis to dose, it should cease making risk-based conclusions.

ATSDR Response: An assumed dose of 100 mrem/year and the 0.0005 per rem risk factor results in an annual estimated cancer risk of 5 in 100,000, well within the EPA acceptable risk range of 1 in 10,000 to 1 in 1,000,000. Further, because risk is a policy statement based on the LNT, rather than an assessment of health effects, and no adverse health effects have ever been associated with normal background radiation doses (~360 mrem/year), these cancer risks are unlikely to translate into any real adverse health effects.

56. ATSDR assumes exposure to low-doses of radiation is harmless in the absence of studies showing such doses result in increased incidence of adverse health outcomes.

In addressing the uncertainties of health outcomes at low doses of exposure, ATSDR takes the approach that "the basis for health outcomes is direct observations using known parameters" (page 27). According to ATSDR, "measurability" is the criteria for adverse health impacts, and doses below levels associated with measurable outcomes have no risk. This criteria also biases ATSDR's findings in the direction of finding no health hazard, because the exposure of men, women and children to plutonium-laden sludge is considered "harmless" until proven guilty in animal or human studies showing increased incidence of adverse health outcomes.

Based on a statement by ATSDR Senior Health Physicist Paul Charp at the February 19, 2003 public meeting, ATSDR staff is well aware of the significant methodological barriers to demonstrating adverse health outcomes in populations exposed to "low-levels" of radiation. These include the need for a large study population, well-documented exposure data, a thorough understanding of the range of potential health outcomes associated with exposure, and well-documented health outcome data.

ATSDR's requirement of "direct observations" is essentially saying "what we don't know can't hurt you." This is not a health protective approach. As our understanding of the risks of exposure to ionizing radiation has improved over time, there has been a 50-fold increase in estimated risk. A most recent example of this trend is from a study of workers at Colorado's Rocky Flats nuclear facility that found people who inhale plutonium have a higher risk of lung cancer than previously believed. Scientists at the University of Colorado who conducted the study found the link between plutonium and lung cancer occurs even at levels currently considered safe.

In considering the potential health impacts of exposure to low doses of radiation, ATSDR's evaluation would be enhanced by the application of a more "precaution-based" scientific approach. This approach is based on the principle that "when an activity raises threats of harm to human health or the environment, precautionary measures should be taken even if some cause-and-effect relationships are not fully established scientifically." Key Elements of the Precautionary Principle include: taking precaution in the face of scientific uncertainty, placing the burden of proof on proponents of an activity rather than on potential victims of the activity, using democratic processes to carry out and enforce the principle - i.e., informed consent, and exploring alternatives to possibly harmful actions.

ATSDR Response: All of the uncertainties underlying the association of radiation doses and adverse health effects have been fully incorporated in the derivation of the ATSDR MRL for ionizing radiation. The MRL is based on a comprehensive review of radiation studies and further acknowledges uncertainty by reducing the No Observed Adverse Effect Levels (NOAELs) by a factor of 3 to accommodate potential uncertainty for human variability. This public health assessment directly addresses the uncertainties underlying the monitoring data and dose assessment process related to Pu-contamination at the Livermore Water Treatment Plant and determines that there is an adequate basis for the resulting public health determination. Consequently, the question to be addressed by the "Precautionary Principle" is not whether there is any uncertainty because there is always some, but rather whether there is adequate certainty.

57. ATSDR assumes that the dose to an individual under a "worst-case exposure scenario" is indicative of the public health impacts of distribution of plutonium-laden sludge.

It is a basic epidemiologic principle of preventive medicine that large numbers of people exposed to "small" risks can lead to a large public health impact. Yet ATSDR presents no population-based dose or risk estimates. The capacity of low level radiation doses to produce health effects is dependent on the population dose and the susceptibility of the individuals exposed, which may be related to their developmental stage, the effectiveness of genetic repair mechanisms, other carcinogenic exposures, reproductive behavior, and other factors. The absence of population-based estimates biases ATSDR's assessment in the direction of underestimating the public health impacts of distributing plutonium-lade sludge.

ATSDR Response: This is a basic fallacy of the risk assessment process. People do not get sick from risk calculations, they get sick because they have been exposed to, and taken in, a specific dose. While risks can be summed across individuals, doses are specific to the conditions of exposure. If the highest potential dose to the most susceptible member of the exposed population is below a level of health concern, then all lower doses are necessarily also below levels of health concern.

58. ATSDR assumes the data are adequate to project the maximum levels of plutonium in the sludge. We disagree that the available data are adequate to determine the highest concentration of plutonium that may have been in the sludge.

• The most relevant data from the standpoint of exposure would be measurements of the sludge that was distributed and measurements at households that received the sludge. ATSDR acknowledges that " there has not been regular monitoring of the processed sewage sludge" (emphasis added)(page 1). Existing data for households that reportedly received sludge are limited to three homes. However, ATSDR presents no data on the sampling or analytical protocols, the time of acquiring sludge and the quantity of sludge used in these residential yards. Therefore there is no basis for assuming that these residential samples represent the maximum levels at these or other households that took the sludge.

ATSDR Response: We agree that such sampling, conducted at the time of placement and distribution would be the best basis for dose assessment. However, as such data was not collected in the 1960's, we have used the available monitoring data to determine whether potential doses could have been at levels of health concern, and subsequently, if additional data are needed to more fully evaluate any potential exposures. Using health protective exposure assumptions and acknowledging the uncertainty in the available data, we have determined that Pu concentrations in any distributed sludge were much lower than those necessary to produce adverse health effects. Consequently, there is no public health basis for recommending additional sampling of areas of potential sludge distribution.

• ATSDR's analysis assumes that the "maximum measured values" are the same as "maximum values" (i.e., the highest concentration measured is the true value of the highest levels present). This is not a supportable assumption for several reasons:



• Contrary to ATSDR assertion that there is "A nearly complete historical record of LWRP gross alpha concentrations for the period 1960-1973" (page iv), data are missing for four and one-half years of time during the period 1958 to 1969 (see Figure in Attachment 1of our comments. Also note that sludge was distributed beginning in 1958, so 1958, not 1960, is the relevant start date for gross alpha monitoring at LWRP). Moreover, interpretation of these data is limited because the sampling and analytical methods that produced these results are not known. ATSDR states "although we have no information on the historic CDHS [California Department of Health Services] Bureau of Radiological Health gross alpha analytical procedures, there is no a priori reason to doubt the validity or utility of their data" (page 6). In contrast to ATSDR, we believe that given the gaps in data, and the lack of information about data collection, analytical methods, data management, and quality assurance for existing data, it cannot be assumed that the existing data are representative of maximum releases for the time periods documented.

ATSDR Response: We agree that there are gaps in the individual data sets. However, for the reasons stated in the document, we believe that the collective data base provides an adequate basis for the public health determination. Additionally, because of the volume of effluent treated and the timeframe of the treatment process, the data gaps are of insufficient duration to obscure releases larger than the 1967 event. As Pu 239 concentrations from sludge from the 1967 release are below levels of public health concern, any smaller releases will also produce concentrations below levels of health concern.

• ATSDR states " measured digester sludge data and the LLNL analyzed effluent data indicate the 1964 and 1967 release episodes represent worst-case sludge concentrations" (page iv), and "time specific gross alpha data which can provide limits on the potential maximum Pu 239 concentrations" (page 2). The problems of missing and un-validated data as cited above are further compounded by several other factors: (1) The source(s) of the 1964 and 1967 releases have not been documented or disclosed; (2) A LLNL report documents that the samplers were inoperative from May 29 to June 2, 1967 and that releases were inferred, not measured for this period; (3) This same LLNL report states that when sampling from the sludge lagoons it was "very difficult to obtain [a] representative sample"; and (4) The California Department of Health Services has determined that Pu 239 concentrations in the sludge at LWRP that resulted from the 1967 release were not adequately characterized because there was no sampling of any waste solids from any part of the LWRP process. Since Pu 239 associates with solids, the amount of Pu 239 may have been underestimated because the activity in solids was never measured.

ATSDR Response: The gross alpha concentrations in digester solids were measured. Because gross alpha concentrations are a health protective estimate of Pu 239 concentrations (all alpha-decaying nuclides in the Pu 239 energy range are included in the gross alpha measurements) and because the digester concentrations over-estimate the resulting drying bed sludge concentrations, the measured digester gross alpha concentrations provide an adequate basis for estimating the Pu 239 concentrations in distributed sludge.

• The maximum values measured in Big Trees Park cannot be assumed to be maximum values related to sludge distribution. ATSDR's assertion that plutonium-laden sludge is the known pathway of plutonium contamination in Big Trees Park is not borne out by soil sampling which had equivocal results, nor is this conclusion supported by statements by the City of Livermore representatives indicating that sludge was not used in Big Trees Park. ATSDR/LLNL have promoted the hypothesis that members of the public may have brought their sludge to the park. These allegations should not be confused with established facts.

ATSDR Response: The conclusions of the Big Trees Park sampling effort are based on distributions and concentrations of heavy metals which are uniformly consistent with heavy metal concentrations in processed sewage sludge. Please refer to the ATSDR health consultation for more information (ATSDR 2000) on this topic.

• The maximum values measured in LLNL's "plutonium-test garden" cannot be assumed to be maximum values related to sludge distribution. The test garden sludge was obtained in the early 1970s. There is no basis to assume that the concentration of plutonium in this sludge was representative of all the sludge distributed in the past, or even at that time.

ATSDR Response: We did not assume that the test garden received the maximum Pu 239 concentrations. We evaluated potential doses for Pu 239 concentrations several orders of magnitude higher than the measured levels in the test garden and compared those levels with potential maximum LWRP digester concentrations.

59. ATSDR should present a comparison of the results of its computer model (RESRAD) used by ATSDR to assess radiologic exposure from plutonium in soil with the results obtained by other agencies cited as using this model.

As presented in the Figure on page 22, ATSDR's results of the computer model (RESRAD) used to evaluate the relationship between radiologic dose and plutonium levels in soil show that for every increasing level of plutonium in soil there is an incremental increase in radiation dose. For example, ATSDR's results show that the 15mrem/year dose set by the US Environmental Protection Agency as being "protective of human health" would be associated with average Pu 239 soil concentrations in a residential yard of 122 pCi/g, and the comparable 25 mrem/year dose set by the California Department of Health Services Radiologic Health Branch would be associated with a soil level of 204 pCi/g.

ATSDR states that "These dose estimates and associated soil concentrations are derived using health protective assumptions for all exposure factors and durations"(page 21). ATSDR notes that RESRAD has been used for deriving limits for radionuclides in soil by US EPA, the US Army Corps of Engineers, US Department of Energy and the US Nuclear Regulatory Commission (page 12). In order to validate that ATSDR's presentation of the relationship between dose and plutonium-soil levels is based on conservative exposure assumptions, we request that ATSDR report on if these other users of RESRAD derived the same slope as ATSDR. Specifically, whether or not any of these other end users derive a steeper slope than the one derived by ATSDR for this assessment, such that for every incremental change in plutonium in soil a proportionately larger increase in dose is expected?

ATSDR Response: RESRAD is a numerical model and we have provided all of the general and site-specific assumptions that we have used in deriving the dose estimates and resulting soil concentrations. Any other person or entity conducting this evaluation using the same model, parameters, and assumptions will necessarily have the same results. Reviewer 8, comment 60 provided an alternate dose assessment model. Their results are essentially identical to those of the RESRAD evaluation in this PHA.

56. ATSDR should recommend LLNL/DOE provide funding to Alameda County to implement a process to address the historic distribution of plutonium-contaminated sludge. Funding should include adequate resources to support soil testing on demand to all households that may have received sludge.

Clearly, the most relevant information to individuals and families living in homes that may have received sludge would be "What is the level of plutonium in the soil in my home and what is the potential health risk of that level? A report by the California Department of Health Services (CDHS) prepared in collaboration with our three community-based organizations, Alameda County and the City of Livermore recommends that "LLNL/DOE provide funding to Alameda County Department of Health Services to implement a process to address the historic distribution of sludge from LWRP).

However, ATSDR does not support any additional sampling, including at households that may have received sludge. ATSDR asserts "any additional sampling data will be subject to the same types of uncertainties as existing historical data" (page 34). These historical limitations are specified by ATSDR as " permit requirements, discharge limits, monitoring and reporting procedures have changed over time which creates difficulties in interpreting Pu 239 concentrations in sludge during this [historical] time period" (pages 1-2) and " currently required practices of data collection, analytical methods, quality assurance and data management cannot be assumed for samples collected and analyzed 30 to 40 years ago" (page 3).

Although we agree with ATSDR that all environmental sampling data are subject to uncertainties, we see a fundamental contradiction in ATSDR conclusions regarding sampling. One the one hand, ATSDR is saying the existing historical data provide sufficient basis for concluding that the concentration of plutonium in any distributed sludge is below a level of health concern. On the other hand, ATSDR is saying that any new sampling would not be worthwhile because it will be no better than historical data with respect to sampling and analysis procedures. We note that not only have environmental sampling analytical procedures improved over the last 30 years, sampling households today that may have received sludge in the past period would provide individuals with a direct measure on which to base their own conclusions regarding health risk.

ATSDR Response: We agree that analytical procedures related to sample analysis have improved greatly over those of 30 years ago. However, uncertainty related to analytical error is usually a very minor component of total sample error (or uncertainty). Knowing where to sample is a much more significant component of sample variability and the passage of time has greatly compounded this source of error. The conclusions of this assessment clearly indicate that there is no public health basis for additional sampling. Promulgation of such a recommendation, in opposition to our public health conclusions would be irresponsible.

57. Finally, we appreciate the opportunity to provide comments on ATSDR's "Public Health Assessment Plutonium 239 Sewage Sludge Used as a Soil or Soil Amendment in the Livermore Community Lawrence Livermore National Laboratory". We note that ATSDR states on the second page of this document " Subsequent to the public comment period, ATSDR will address all public comments and revise or append the document as appropriate" and states in the Forward, "All comments received from the public are responded to in the final version of the report" (emphasis added).

However, we are concerned that ATSDR's request for comments has not been made in good faith. Our concern is based on an email received from ATSDR's Burt Cooper regarding its public health assessment of tritium releases from LLNL (see Attachment 2 of our comments), stating " What we [ATSDR] will do is excerpt the comments, issues, and criticisms related to the PHA and provide responses to these." We note that ATSDR's tritium report contains the same assurances about incorporating comments as this plutonium-sludge report. We believe that all of these comments are pertinent to the public health assessment and anticipate that they will be included in total. ATSDR's refusal to abide by its own written process subverts public participation in decision-making about the health impacts of LLNL activities, and substantially undermines the scientific and ethical credibility of ATSDRs' conclusions and recommendations.

ATSDR Response: ATSDR has always included and responded to public comments related to specific documents. However, because these comments have typically included voluminous contributions and references to multiple documents, ATSDR has never included or listed these submitted documents verbatim. Our consistent and longstanding policy has always been to summarize, and if necessary interpret, lengthy comments in order to include and directly respond to them.

Reviewer 7

58. Based on all of the available information, it does not appear that the City of Livermore, including the City of Livermore Water Reclamation Plant ("LWRP") personnel, were aware of the 1967 plutonium releases from LLNL or of any other releases during the 1960's or early 1970's. Rather, it appears the City did not learn of the Pu release(s) until the LLNL 1973-74 Annual Report was issued.

ATSDR Response: Comment noted; no response is necessary.

Reviewer 8

59. The overall conclusion of this report - that there is no apparent public health hazard associated with the distribution of Pu-239 contaminated sludges in the Livermore, CA community - is probably correct. Although the available data presented in this report contain gaps and inconsistencies, as described below, it is highly unlikely that general public exposure to Pu-239 in sludge has ever resulted in doses approaching more than a small fraction of the 100 mrem/year protective level. This observation is generally consistent with the results of the Interagency Steering Committee on Radiation Standards (ISCORS) Sewage Sludge Subcommittee's project on estimating the possible public health concerns related to typical levels of radionuclides in municipal sewage sludge.

ATSDR Response: Comment noted; no response is necessary.

60. Comparing the reported measurements of gross alpha (which is assumed to be Pu-239) in dried sludge and in soils in Big Trees Park with the dose modeling results obtained by the ISCORS Sewage Sludge project (see www.iscors.org/sewage.htm <http://www.iscors.org/sewage.htm> for the draft Radionuclides in Sludge Dose Assessment) provides an independent perspective on the issue of the possible public health concerns addressed in the ATSDR Public Health Assessment. To make a reasonable comparison of the Livermore sludge data to the ISCORS project results, the following factors and assumptions should be noted.

From the ATSDR report:

• The maximum measured level of Pu-239 in dried sewage sludge (which would have been distributed for unrestricted use) is 60 pCi/gram;



• The maximum soil concentration of Pu-239 measured in 1998 in Big Trees Park is less than 1.0 pCi/gram (see UCRL-ID-134581). The maximum reported level of Pu-239 in soils in a residential area is 1.8 pCi/g (page 7);

From the ISCORS SSS project:

• The ISCORS Sewage Sludge Subcommittee (SSS) dose modeling project included an "Onsite Resident " scenario, where a home was built on land that had previously received sewage sludge applied at agronomic rates (10 MT/ha);



• The ISCORS SSS also assessed a "Recreational" scenario where a public park was constructed on lands that had previously been reclaimed by applying sewage sludge once, at higher than agronomic rates (100 MT/ha);



• These scenarios, while not completely consistent with the Livermore community situation in terms of potential pathways of exposure, are reasonably similar and provide reasonably conservative (i.e., health protective) dose estimates;



• The ISCORS SSS used RESRAD 6.0 to perform calculations that estimate possible doses to the public from these scenarios. The results are stated in Dose-to-Source Ratios (DSR), in mrem/year/pCi/gram of soil, for each radionuclide examined. The calculations assume an initial sludge concentration of 1.0 pCi/gram for each radionuclide, and that the sludge is incorporated into the upper 15 cm of the soil. To estimate dose, it is possible to multiply the actual radionuclide concentration in the sludge by the appropriate DSR. It is also possible to estimate dose by multiplying the actual soil concentration by the appropriate DSR, after correcting for a 227-to-1 dilution factor resulting from incorporation of the sludge into the soil.



• Since the RESRAD dose modeling used a probabilistic method for parameter selection, it is possible to select a very conservative (i.e., health protective) DSR for this analysis. The DSRs selected below for each scenario represent the 95^th percentile peak DSR.

Based on these assumptions, the dose estimation is as follows:

Onsite Resident Scenario

Multiply sludge concentration by Pu-239 DSR, and assume 3 annual applications:

60 pCi Pu-239/g sludge X 3 X (1.35 E-3 mrem/yr/pCi/g) = 0.2 mrem/yr

Multiply soil concentration by Pu-239 DSR, and by 227x dilution factor

1.8 pCi Pu-239/g soil X 227 X (1.35 E-3 mrem/yr/pCi/g) = 0.6 mrem/yr

Recreational Scenario

Multiply sludge concentration by Pu-239 DSR

60 pCi Pu-239/g sludge X (3.97 E-5 mrem/yr/pCi/g) = 0.002 mrem/yr

Multiply soil concentration by Pu-239 DSR

1.8 pCi Pu-239/g soil X 227 X (3.97 E-5 mrem/yr/pCi/g) = 0.02 mrem/yr

Based on this comparison, it is likely that no exposure to Pu-239 in the Livermore community ever exceeded a fraction of 1 mrem/year as a result of distribution of sewage sludge containing Pu-239. This conclusion is consistent with dose estimates discussed in the Public Health Assessment report.

ATSDR Response: Comment noted; no response is necessary.

61. Data gaps and inconsistencies noted in review of the Public Health Assessment document include the following:

• There are no gross alpha analyses for dried sludge after October 1966;



• There are no radionuclide-specific analyses until 1973;



• There are no data on moisture content of digester sludge (June 1964 through March 1965); Gross alpha values are presented in Figure 3 as wet weight analyses for this time period;



• There is no explanation of the relationship between sewer effluent data (in pCi/liter) and annual releases (in Ci/yr). Some estimate of annual volume of wastewater discharges from LLNL would be needed to compare these levels.

ATSDR Response: Comments noted, please see responses to comments 45, 44, and 11 (respectively).

62. On page 16, second and third paragraphs, there is discussion of the data presented in Figure 3. The statement is made that the effluent and release data in Figure 3 track the monthly digester sludge values. This is not an accurate statement. The digester sludge data are consistently greater than sewer effluent and annual release data for 1965 and 1966, and generally below sewer effluent and annual release data from 1967 to 1973. Additionally, the 1964 episodic release, as evidenced by the observed values of 297 pCi/g and 173 pCi/g (wet weight), are not reflected in elevated gross alpha levels in the sewer effluent or annual release values immediately preceding the time of the digester sludge measurements.

ATSDR Response: With respect to absolute numerical values, the digester sludge values are indeed greater than those for the sewer effluent and annual releases. This reflects differences in the respective concentration units and the dynamics of the treatment process. The relatively dilute concentrations of gross alpha in the liquid effluent are concentrated in the digester sludge solids. Similarly, the annual release data are in completely different concentration units as specified in the right hand scale. Temporally, there is very good agreement between releases and measured gross alpha concentrations. With regard to the 1964 release event, both the sewer effluent and annual release data show a spike that corresponds in time with the digester concentrations. As noted the absolute values of these peaks are not the same as the 1967 release event, but this difference may be an artifact of the sampling and reporting process (monthly vs. 6 month and yearly reporting periods).

63. Although the report acknowledges that the digester sludge data from June 1964 through March 1965 is stated on a wet weight basis, and is, therefore, not comparable to all of the other data in Figure 3, nevertheless, the non-comparable data is presented in Figure 3 and discussed throughout the report as though it were comparable (see discussion on page 29). The point noted in #4 above (that the statement that the effluent and release data in Figure 3 track the monthly digester sludge values and the 1967 episodic release is not accurate) becomes more significant if the wet weight values for the digester sludge data are converted to a dry weight basis. An estimated solids content of 2% for the digester sludge could be used to make a crude approximation of the comparable dry weight values for gross alpha for this time period. The resulting estimated concentrations would be 14,850 pCi/g (297 pCi/g x 50) and 8,650 pCi/g (173 pCi/g x 50), respectively. Presenting these data on a dry weight basis would further exacerbate the lack of a consistent pattern in a comparison of the digester sludge data with the sewer effluent and annual release data for 1964. Furthermore, the observation (page 16) that the sewer effluent data and the annual release data "capture and record a short term event such as the May-June1967 Pu-239 release" is not true with regard to the 1964 release, especially when considering the converted dry weight levels above. Sewer effluent and annual release data for the period preceding the 1964 release would have had to be several orders of magnitude greater than the levels in Figure 3 for this statement to be accurate.

ATSDR Response: Please refer to footnote 12 and comment 44.

64. In addressing the issue of the adequacy of the available data for the public health determination (page 28), the report concludes that the Pu-239 levels in the processed (i.e., dried) sludge are lower than the gross alpha levels in the digester sludge. The explanation is that the Pu-239 contaminated sludges are blended with uncontaminated sludges in the drying beds, where residence times may be one to five years. This blending would substantially reduce the Pu-239 levels, thereby allowing the assumption that the gross alpha levels in the digester sludge are always a conservative (i.e., health protective) surrogate for Pu-239 levels in the dried sludge that was actually used as a soil amendment. The ISCORS SSS project concluded that levels of non-degrading contaminants (such as Pu-239) are likely to increase in the sludge, as a result of the normal operations of the treatment plant (in this case, the drying beds), rather than being reduced. The draft ATSDR report concludes that this lack of Pu-239 data in the dried sludge following the 1967 release is not significant if dilution can be assumed. It is possible that such assumed dilution has, in fact, occurred. Without any measure of gross alpha or Pu-239 in the dried sludge following the 1967 release, however, this dilution explanation is merely hypothetical. Another plausible explanation is that the Pu-239 levels remained elevated in the sludge drying beds, as a result of both episodic releases (1964 and 1967). Although the dried sludge levels declined from the 60 pCi/g peak throughout late 1965 and 1966, it is possible that the dried sludge levels increased from 1967 on, reflecting the 1964 release of possibly 15,000 pCi/g, dry weight, as well as the 1967 release. Since residence time in the drying beds could be as long as five years, it is possible that the sludges removed after October 1966 contained elevated levels as a result of the 1964 release, and were not blended and diluted in the drying beds. The data do not support this hypothesis any more than they support the conclusions presented in the report. Absent actual data on Pu-239 (or gross alpha) levels in the dried sludges after October 1966, it is not possible to confirm which of these hypotheses, or any other hypothesis, is correct.

This limitation on the adequacy of the existing dried sludge data can be mitigated (and the dilution assumption can be tested) by estimating the maximum soil concentration of Pu-239 resulting from the highest level measured in the digester sludge converted to a dry weight basis (14,850 pCi Pu-239/g), and assuming the following:

- Levels of Pu-239 in the digester sludge remained at 14, 850 pCi/g for 2 months;

- Pu-239 contaminated sludges were blended in the drying beds for 5 years (60 months) before being distributed for public use;

- The sludge-to-soil dilution factor (227 to 1) used in the ISCORS SSS dose modeling project accounts for eventual incorporation of sludges into the upper portion of the soil.

Based on these assumptions, the following calculation can be made to estimate likely Pu-239 soil concentrations:

14, 850 pCi Pu-239/g (dry) digester sludge X 2 months/60 months = 495 pCi/g dried sludge

495 pCi/g dried sludge / 227 = 2.2 pCi Pu-239/g in soil

This estimate (2.2 pCi Pu-239/g in soil) approximates the maximum levels measured in residential soils (1.8 pCi Pu-239/g). As shown in #2 above, potential doses resulting from soil concentrations in this range would be below 1 mrem/year.

ATSDR Response: Dilution of the digester gross alpha concentrations relative to the dried processed sludge is much more than a hypothesis. It is the necessary conclusion of a simple mass balance consideration of an episodic release into the sewer system and a 5 year period of sludge accumulation in the drying beds before distribution of the resulting sludge mixed with 4+ years of non-contaminated sludge.

65. Definitions: the definition for effective dose equivalent seems to be incorrect. Suggest the definition for total effective dose equivalent be added and throughout the document the terms replaced where needed. Also the definition for dose should conclude with "total effective dose equivalent" not "total effective dose".

Suggested definitions (from 10 CFR 835):

Committed dose equivalent (HT,50) means the dose equivalent calculated to be received by a tissue or organ over a 50-year period after the intake of a radionuclide into the body. It does not include contributions from radiation sources external to the body. Committed dose equivalent is expressed in units of rem (or sievert) (1 rem = 0.01 sievert).

Committed effective dose equivalent (HE,50) means the sum of the committed dose equivalents to various tissues in the body (HT,50), each multiplied by the appropriate weighting factor (wT)--that is, HE,50 = ?wTHT,50. Committed effective dose equivalent is expressed in units of rem (or sievert).

Dose is a general term for absorbed dose, dose equivalent, effective dose equivalent, committed dose equivalent, committed effective dose equivalent, or total effective dose equivalent as defined in this part.

Dose equivalent (H) means the product of absorbed dose (D) in rad (or gray) in tissue, a quality factor (Q), and other modifying factors (N). Dose equivalent is expressed in units of rem (or sievert) (1 rem = 0.01 sievert).

Effective dose equivalent (HE) means the summation of the products of the dose equivalent received by specified tissues of the body (HT) and the appropriate weighting factor (wT)--that is, HE = ?wTHT. It includes the dose from radiation sources internal and/or external to the body. For purposes of compliance with this part, deep dose equivalent to the whole body may be used as effective dose equivalent for external exposures. The effective dose equivalent is expressed in units of rem (or sievert).

Total effective dose equivalent (TEDE) means the sum of the effective dose equivalent (for external exposures) and the committed effective dose equivalent (for internal exposures).

ATSDR Response: The agency appreciates your suggestions for the terms listed in your comment. However, the agency currently has accepted definitions as published in its Toxicological Profile for Ionizing Radiation. The glossary of the public health assessment has been updated with these definitions as follow.

Committed Dose Equivalent (HT₅₀)--The dose equivalent to organs or tissues of reference (T) that will be received from an intake of radioactive material by an individual during the 50-year period following the intake.

Committed Effective Dose--The International Commission on Radiological Protection (ICRP) term for committed effective dose equivalent. (See Committed Effective Dose Equivalent.)

Committed Effective Dose Equivalent (HE₅₀)--The sum of the products of the weighting factors applicable to each of the body organs or tissues that are irradiated and the committed dose equivalent to the organs or tissues (HE₅₀ = W_THT₅₀). The committed effective dose equivalent is used in radiation safety because it implicitly includes the relative carcinogenic sensitivity of the various tissues.

Dose (or Radiation Dose)--A general term denoting the amount of energy from radiation that is absorbed per unit mass of absorber. A generic term meaning absorbed dose, dose equivalent, deep dose equivalent, effective dose, effective dose equivalent, committed dose equivalent, committed effective dose equivalent, equivalent dose, or total effective dose equivalent. For special purposes it must be appropriately qualified. If unqualified, it refers to the absorbed dose.

Dose Equivalent (DE)--A quantity used in radiation protection. It expresses all radiations on a common scale for calculating the dose for purposes of radiation safety. It is the product of the absorbed dose in rad or gray and a quality factor, whose value depends on the radiation. (The unit of dose equivalent is the rem. In SI units, the dose equivalent is the sievert, which equals 100 rem.)

Total Effective Dose Equivalent (TEDE)--The sum of the effective deep dose equivalent from external exposures and the committed effective dose equivalent from internal exposures.

66. Appendix 2 is missing.

ATSDR Response: This reviewer received the document as an attachment to an e-mail. The e-mail indicated that appendix 2 (which is a copy of a 1967 letter) was not in the electronic copy and was sent separately via facsimile.

Reviewer 9

67. You may wish to put a hyphen between Pu and 239 (i.e., Pu-239) throughout the report.

ATSDR Response: The ATSDR editors have indicated that "no hyphen" is our preferred format.

68. Page iv, 2^nd paragraph, 2^nd sentence

Suggest that the word "highest" be inserted before "average monthly gross alpha"

ATSDR Response: Please see the response to comment 7.

69. Page iv, 3^rd paragraph, 1^st sentence

Suggest the word "Because" be used instead of "As" since the later may cause the reader to be searching for a time rather than a reason. Suggest the word "collected" be inserted ahead of "following the 1964 episode."

ATSDR Response: Comments noted; no changes are necessary.

70. Page 2, 1^st partial paragraph, 1^st partial sentence

Suggest inserting a comma after the word "time."

71. Page 2, 1^st full paragraph, 1^st sentence

Suggest inserting a comma between "Livermore community" and "three specific questions."

ATSDR Response: Commas inserted.

72. Page 2, 3^rd paragraph, 2^nd sentence

As with Pu and 239, suggest inserting a hyphen between Am and 241, and Americium and 241. Also, in the same sentence, suggest substituting "were not conducted." For "are not currently available."

ATSDR Response: Please see the response to comment 67. A memorandum from W.P. Bennett and B.L. Rich to D.C. Sewell (appendix 2) indicates that isotopic measurements were conducted in 1967. If so, those results are not currently available.

73. Page3, 3^rd paragraph

Suggest "300 site" be rewritten as Site 300 where ever it occurs in this document.

ATSDR Response: This has been changed as suggested.

74. Page 4, 2^nd full paragraph, 5^th sentence

Needs an ending parenthesis.

ATSDR Response: This has been changed as suggested.

75. Page 4, 3^rd full paragraph, 2^nd sentence

Suggest inserting "(hereafter referred to as digester sludge)" after the words "the digester process" and checking throughout the document for consistency in the use of the terms.

ATSDR Response: The term "solid materials" has been replaced with "digester sludge".

76. Page 6, 2^nd paragraph

Discussion of data before the 1967 release is provided. A brief discussion of the development of after-the-release data would make this paragraph more complete.

ATSDR Response: A reference to the Sewell letter and attached data are referenced in the preceding paragraph. Other monitoring data, which are presented and discussed in later sections were not previously published and would be inappropriate to present as background information.

77. Page 17, 2^nd paragraph

Suggest more clearly labeling the data referred to in the sentence, "These values are included in Appendix 4" within the table or in the footnotes.

ATSDR Response: Each of these referenced data sets have a number of attributes which are included in the appendix. Redundant description of these attributes is unnecessary.

78. Page 18, last paragraph and page 20, 1^st paragraph

Suggest the areas in the 2 paragraphs be made consistent or the reasons for using the 2 different areas be made more clear. We understand that 1/2 acre is the RESRAD default minimum area for modeling and making the exposure and dose calculations. Is 1/4 acre selected to approximate the portion of the average Livermore and vicinity residential lot area that might be vegetated for the purpose of calculating a bounding case dose?

ATSDR Response: The parenthetical statement "or 108 truck loads for a acre lot" has been added to this sentence. The lot sizes are for illustration purposes only. One half acre is used for dose estimation purposes in order to be consistent with the assumptions underlying the derivation of the EPA soil screening levels (appendix 5).

79. Page 20, caption for Figure 4, 6^th sentence

Suggest adding the words "remains the same" to the end of the sentence after "while the baseline contributions of uranium and other radionuclides."

ATSDR Response: This has been changed as suggested.

80. Page 22, 1^stparagraph, 1^st sentence

While exposure to radiation at the level of ATSDR's MRL would result "without appreciable risk of adverse non-cancerous effects", it would be informative to the readers to state whether there may be cancerous effects at that level and to state the levels at which cancerous effects have been observed.

ATSDR Response: Both cancer and non-cancer health effects are mentioned in the referenced sentence and more fully described in the following sections.

81. Page 22, 1^st partial paragraph

Suggest breaking this into 2 paragraphs by ending the 1^st paragraph after the 1^st full sentence and beginning the next paragraph with "The following sections will present".

ATSDR Response: Comment noted.

82. Page 25, 4^th paragraph, 1^st sentence

The phrase "medical accidents and other medical accidents" does not seem to make sense.

ATSDR Response: This has been changed to read "medical exposures and other medical accidents."

83. Page 29, 3^rd full paragraph, 5^th sentence

Suggest substituting "The fact that" for the words "Considering that".

ATSDR Response: This has been changed as suggested.

84. Page 29, footnote 13, last sentence

The work described as being underway "currently" has been completed and provided to the ATSDR, as pointed out on the next page. Suggest deleting sentence that begins "LLNL is currently.

ATSDR Response: This has been changed as suggested.

85. Page 30, 1^st paragraph, 2^nd sentence

Suggest more clearly labeling the data referred to in the sentence, "These data have recently been" within the Appendix 4 table or its footnotes.

ATSDR Response: "These data" has been replaced with "Daily measurements".

86. Page 31, caption for Figure 6

Suggest inserting "(later weekly)" between the words "Daily measurements" to more accurately convey the frequency of the sampling that was performed.

ATSDR Response: This has been changed as suggested.

87. Page 33, 4^th paragraph, last sentence

Suggest that the word "currently" be deleted.

ATSDR Response: Comment noted; no change has been made to document.

88. Page 35, 2^nd paragraph, 2^nd sentence

Suggest that the word "considerable" be replace by "little."

ATSDR Response: Comment noted; no change has been made to document.

89. Page 35, 1^st paragraph, 1^st sentence

Suggest providing an explanation for why there are no recommendations for further sampling. Perhaps rewrite the latter half of the sentence to read, "ATSDR believes at this time that additional soil sampling in areas of known or unknown sludge distribution would most likely be of no or little value in determining health impacts due to the technical difficulties limiting sampling and analysis of such low levels of radioactivity. ATSDR would expect findings to be within the range of normal fallout background and exceptions to this within the range of historic sampling and analysis, including that found in this ATSDR report.

ATSDR Response: Comment noted; no change has been made to document. The referenced sentence does indicate why we are not making any further recommendations, which is discussed in the preceding section.

91. Page 45, definition of MRL

It might be informative for the readers if the MRL definition which refers to exposure "without a measurable risk of adverse, noncancerous effects" were expanded to include a brief discussion of the cancer implications at that level (i.e., 100 mrem per year). Have there been cancerous effects observed at that level or at what levels have cancerous effects have been observed?

ATSDR Response: The glossary contains ATSDR's approved definition of an MRL. Although it would be inappropriate to change this approved definition, the Toxicological Profile for Ionizing Radiation contains the following sentence- "No individual studies were identified that could be used to base a chronic-duration external exposure MRL that did not result in a cancer-producing endpoint." Another sentence indicates- "an MRL for external radiations should be appropriate for internal radiation." Although we cannot change the specific definition of an ATSDR MRL, we have interpreted these sentences to indicate that the MRL for ionizing radiation does address cancer as a potential health effect for both internal and external radiation sources. This is addressed in footnote 8.

92. The use of the concepts of "annual dose" and "committed dose" is inconsistent in the PHA. Specifically, although ATSDR's minimum risk level (MRL) is defined in terms of an annual dose, the results of the dose calculations reported in the PHA are committed doses. As a point of clarification, in this memo, the term" committed dose" refers to the committed effective dose equivalent (CEDE) and the "annual dose" refers to the annual effective dose equivalent (AEDE).

Thus, Figure 2 shows the dose an individual is committed to receive (over the next 50 years) from intakes that occurred each year, not the actual annual EDE received during each year. Since the annual uptake rate is constant, the 50-year CEDE accrued each year is also necessarily constant. Accordingly, Figure 2 is mis-labeled. It simply shows that the CEDE for each year of exposure is the same and that the exposure conditions are not changing. It does not show annual doses that would relate to the MRL.

ATSDR Response: Please see the response to comment 65 concerning definitions. Relative to Figure 2, the doses calculated by RESRAD are derived from the ICRP dose coefficients which are the equivalent doses. Therefore, the graph indicates the radiological dose one would receive following a single annual intake of the radionuclide. That is, the annual radiation dose. For a committed dose, one would multiply the annual dose by the number of years, for example, 70 years. The dose is correctly defined.

93. Page 1, footnote 1

The contributions of other transuranic radionuclides, notably Pu-238 and Am-241, are dismissed here, but the analysis of the sludge taken from the park in the 1990's indicated that these two isotopes can contribute up to 40% of the contaminant alpha activity in the soil. This information was conveyed in a memorandum from Dave Myers to Bill Mcconachie on 1/15/03.¹ This, however, becomes a moot point because the pathway analysis assumes a concentration of 2.5 pCi/g of plutonium rather than the actual (lower) concentrations of ^239/240Pu found in the park. Since the dose conversion factors for all of these radionuclides are very similar, the specific isotopic mix of these radionuclides doesn't significantly affect the conclusions.

94. Page 2, footnote 2

See comment for Page 1, footnote 1

ATSDR Response: Comment noted; similar explanations are presented in footnotes 1 and 2.

95. Page 13, paragraph 2, line 2

It should read "that may have occurred"

ATSDR Response: Changed as suggested.

96. Page 16, paragraph 3, line 6

It should read "release value" rather than "release values" since there is only one value for the year.

ATSDR Response: Changed as suggested.

97. Page 23, paragraph 4, line 7

It should read "99.5% to 99.9%" since the values are percentages.

ATSDR Response: Changed as suggested.

98. Page 68

The annual inhalation rate is listed as 8,400 cubic meters per year. This rate is equivalent to about 24 cubic meters per day (for an exposure frequency of 350 days per year) or 23 cubic meters per day (for an exposure frequency of 365 days per year).

However the table on page 14 states that the inhalation rate was assumed to be 20 cubic meters per day (presumably applied to the listed exposure frequency of 350 days per year). This gives an annual inhalation rate of 7,000 cubic meters per year. This discrepancy in the inhalation rate needs to be resolved.

ATSDR Response: Table 1 has been revised to reflect the default RESRAD inhalation rate, which is ~24 m³/day.

99. Page 68

The fraction of the time spent indoors is specified as 0.683, and the "fraction of the time spent outdoors on site" is specified as 0.317, which results in 100% of the inhalation time on site.

However, in the table on page 14 it states that 68% of the time is spent indoors, 7.3% of time is spent outdoors (on site), and 24.7% of time is spent off site. This discrepancy needs to be resolved.

ATSDR Response: Table 1 has been revised to reflect the RESRAD default values (~68% of time indoors, ~32% of time outdoors, on-site). Time spent off-site is ~15 days per year (365-350).

100. Page 69

The consumption of fruit and vegetables uses the RESRAD default value of 160 kg/y. However, in the table on page 14, it states that the consumption of vegetables is 100 kg/y. This discrepancy needs to be resolved.

ATSDR Response: Table 1 has been revised to reflect the RESRAD default ingestion rate of 160 kg/year (for fruit, grain, and vegetables).

101. Pages 74-82

The RESRAD output tables are presented in portrait rather than landscape and this makes them very difficult to read. The information would be much easier to understand if the output data were put into landscape view.

ATSDR Response: The page formatting is embedded in the RESRAD output and cannot be changed in the WORD document formatting.

102. General

One of the more important findings of the PHA is that over 65% of the hypothetical dose comes from the vegetable and fruit ingestion pathway scenario, in which it is assumed that fully half of the fruits and vegetables consumed are grown on site, in the contaminated soil. Most of the remaining dose ("26%") comes from eating dirt. These conclusions are somewhat "buried" as a footnote to Figure A-2 in the back of the report.

Public understanding of the likelihood of significant doses from the plutonium in the soil could be enhanced by making the contribution from each exposure pathway very clear - and early in the document. We recommend that the discussion of the contribution to total dose from each pathway be included in the general discussion of hypothetical doses on page 15 of the document.

ATSDR Response: We agree that these are significant findings and have included this information in the section on "Estimated Doses from Pu 239/240 Contaminated Soil" and also in the section outlining health protective aspects of the dose estimation approach ("Exposure and Health Evaluation").

Next Section     Table of Contents