PUBLIC HEALTH ASSESSMENT
PANTEX PLANT
AMARILLO, CARSON COUNTY, TEXAS
Public Comments on the Pantex Plant Public Health Assessment and ATSDR's Responses
ATSDR received comments on the Pantex Plant Public Health Assessment, Public Comment Release (December 11, 1997), from individuals, organizations, and agencies. We thank all of those who took time to comment. The comments received reflect a strong and sincere interest from people in the communities surrounding the Pantex Plant. This appendix includes a listing of the public comments received and ATSDR's responses.
The comments received covered many topics. Recommendations included asking us to delete words, add supporting documentation, and reconsider conclusions: some people wanted us to soften them and some wanted us to strengthen them. One commenter recommended that the public reject the assessment out of hand.
A major concern, expressed repeatedly throughout the comments and questions received by ATSDR, was how much contamination in the environment, or exposure to that contamination, is enough to regard that contamination as a health hazard.
A partial answer to that question--how much of a substance is a health hazard?--is that it depends on the situation. ATSDR does not maintain a set of values for substances that define what is and what is not a health hazard. We have health-based comparison values for some chemicals that we use to initially screen environmental data. However, exceeding screening values does not mean a substance is a health hazard. For each situation, we consider what exposure assumptions are appropriate and what is known about the toxicity of the substance. We explore the answer to this question in greater depth in our responses to specific comments.
Related to the question of hazard, several commenters specifically noted that with radioactive materials, there can be no safe level of exposure and any contamination of radioactive materials in the environment poses a public health hazard. In our opinion, based on the overwhelming body of scientific knowledge, this may not be true. We recognize that people who assign risks to radioactive materials usually assume there exists some incremental risk from exposure to radioactive materials no matter how small the quantity of radioactive material in the environment (i.e., an assumption of the no-threshold models), and they calculate incremental risks by linearly extrapolating risks from higher radiation doses. This approach is a conservative one that is used by regulatory agencies to provide adequate protection for workers and the general public from potential radiation hazards. But this assumption is not necessarily an accurate description of the true risks from exposures to low levels of radiation. Unfortunately, we cannot measure those risks directly using currently available scientific methods (e.g., epidemiology and biological testing). Furthermore, considerable mounting evidence from the fields of genetics and cancer research, as well as numerous epidemiological studies refute the linear, no-threshold risk models for exposures to low levels of radiation. Rather than assuming that any quantity of radionuclides in the environment will cause adverse health effects, we consider both existing radionuclide-specific toxicological data and the ongoing recommendations of radiation advisory bodies (e.g., the International Commission on Radiological Protection [ICRP] and the National Council on Radiation Protection and Measurements [NCRP]). ATSDR believes this to be the most sound scientific approach.
Finally; at ATSDR, we bring to bear the collective education and experience of public health scientists, health physicists, toxicologists, epidemiologists, environmental health scientists, physicians, and others, to the assessment process. In this environment, we work to keep current on scientific and social issues, and discuss, both internally and externally, different approaches to the assessment process. Nevertheless, we recognize that not all people will interpret the same information the same way we did, and we respect others who may have opinions different from ours.
We address the comments received in the following section. Page number references for the comments received by ATSDR (comments are numbered and in bold print) refer to text in the previous (Initial Release and Public Comment Release) versions of the Pantex Plant public health assessment (PHA). Page number references for the ATSDR response to each comment indicate where text occurs in this document--not to the pages where the text occurred in earlier versions of the Pantex Plant PHA.
ATSDR responses are provided after each comment or group of comments.
1. The health outcomes section of the Pantex PHA ... definitely references recent Texas cancer incidence and mortality statistics. They note an excess of CLL that "needs to be investigated:, as well as an excess of prostate cancer and muscular dystrophy deaths, and some other assorted findings. The previously reported Carson County excess leukemia mortality (SMR=1.5, not signif.) is not referred to, but the combined data in this report show Carson County leukemia deaths to still be elevated for the period 1985-1994 (SMR=1.9, not signif.). I think this deserves mention and the relevant table is on pg. 112.
Response: We are not entirely sure which previous report the commentor is referring to; however, in both instances the SMRs were not significant. Additionally, with respect to the combined data, the SMR is small (1.9), the confidence interval is relatively narrow (0.98,3.32), and the category "leukemia" includes all types of leukemia, some of which may have differing etiologies. We do not believe that a discussion of these data would add to the report.
2. On pg. 52, there is a sentence in the first full paragraph that the "statistically significant excess could be the result of slight (non-significant) elevations in deaths from many different types of cancer or the result of a large increase deaths in from one or two types of cancer". This is one garbled sentence - it should say "increase in deaths in one or two types of cancer". But more importantly, which is it in Potter County males? They clearly have the data to answer that, but leave the reader guessing.
Response: Apparently the word "in" was misplaced. The sentence should read " a large increase in deaths from one or two types of cancer". The change has been incorporated. The problem with identifying the exact cancer types is that all cancers, with the exception of prostate cancer are reported as a group category referred to as "All Types". This is illustrated in Tables 16 and 17. Therefore, we know prostate cancer is elevated in this particular population; however, the remaining specific type or types cannot be identified with the current database.
3. On pg. 52, at the bottom, there is a statement that "one of three persons alive today will develop some type of cancer in their lifetime". Actually, its gone up and is now more like 4 out of 10 (more in males). The new NCI SEER report has that figure.
Response: We have rewritten the paragraph to read: "Approximately four out of every ten persons alive today will be diagnosed with some type of cancer in their lifetime."
4. On pg. 55, there is the old "smoking and diet" argument posed against the environmental/occupational causes. This is unnecessary in a PHA about a Superfund site.
Response: This statement is written in response to the question; "can the Pantex Plant be the cause of the higher than expected number of people with cancer that the TCR observed". We do not view this to be an argument posing smoking and diet against environmental and occupation. The fact that the majority of cancers can be attributed to individual behaviors is not in dispute and does not exclude the possibility that some cancers may be caused by environmental or occupational exposures. It is important that residents concerned with the occurrence of cancer in their neighborhoods have knowledge of the most common causes of cancer. The presence of cancer in a neighborhood might be alarming; however, people being diagnosed with cancer is not unusual. As the reviewer pointed out in Comment 3 approximately 40 percent of the population will be diagnosed with cancer at some time in their lives (42.52% for males and 38.88% for females). One of the functions of the PHA is to address resident concerns and provide information so that people can have a better knowledge base from which to draw their conclusions and plan future life choices.
5. Use of definitive language that lacks support - The use of definitive language is inappropriate for this Public Health Assessment. ATSDR PHAs are exploratory, preliminary, inventorying types of studies. They are not designed to provide definitive results. Moreover, given the limitations of this review (in large part because of the data available for the review), I do not believe that ATSDR can use definitive language in its conclusions about whether there are, or were, any risks to public health. In my opinion, the limitations of ATSDR's approach warrant more cautious conclusions. In general, I find this report was written with a bias toward the managers of the facility. When in doubt, the report authors have injected conclusions that appear to be intended to reassure the public of there being minimal if any risk.
ATSDR does this even when they have doubts. On page 20 ATSDR writes "contaminant concentrations detected on- and off-site are compared to values that are believed to be without adverse health effects upon exposure." (Emphasis is mine). So, on pg. 20 ATSDR notes some uncertainty about the protection provided by their (unstated) screening standards. But later, in the conclusions and analysis. ATSDR forgets about these qualifiers and makes statements without qualifiers.
Response: A qualifier has been added.
6. On pg. 48, the report states that it is "difficult to provide meaningful interpretation of results" related to fetal death frequencies. In the Discussion section on the same page (48) the report states that "it would be difficult to attach much importance to findings of statistical significance because our confidence in the precision of the data would be limited." Yet, ATSDR seems to have no trouble attaching importance to findings of no statistical significance that are based on inadequate (missing and unreliable) data. The burden of proof is not balanced, and more attention seems to be given to the evidence suggesting negative findings that to the evidence suggesting positive findings. A better approach would be for ATSDR to suggest how to correct the lack of "precision of the data" (if possible) in order to reach more reliable conclusions. This is clearly within the realm of what ATSDR can recommend in a PHA.
Response: With regard to the health outcome data, we reviewed available health outcome data to determine if the observed amount of disease in the area is greater than what might be expected. During this process data are tested for statistical significance; if the data are found to be statistically significant we examine the data to try to make some conclusion as to whether the "statistically significant" finding is significant with respect to the everyday use of the word. In doing so we look at such factors as the magnitude of observed cases, the magnitude of the expected cases, the specificity of the category examined, and the precision of the estimate as measured by the width of the confidence interval. Contrary to the reviewers comment, with respect to health outcome data, we actually pay more attention to positive findings than negative findings. General limitations with respect to health outcome data quality are discussed in the document.
7. On page 25 ATSDR writes that "no pesticides, herbicides,.... have been detected in perched groundwater." This would seem to argue that these sources CANNOT be problems, at least from groundwater exposure; this seems to be the same line of reasoning used to claim that there CANNOT be any harmful effects from Pantex-related contamination. Is ATSDR willing to make this kind of definitive claim?
Response: It is agreed that pesticides and herbicides are not a problem in groundwater because they have not been detected in groundwater. ATSDR's conclusions are based on environmental sampling data. If contaminants, as mentioned in this comment, are not detected during monitoring, then it can be concluded by weight-of-evdence that no exposure is occurring. One must have a contaminant present to have an exposure.
8. A large body of case studies and research on risk communication suggests that the style and tone of the PHA are the very kinds that should be avoided. For example, definitive language that seems to dismiss community concerns (for example, on page. 55) and unclear and technical writing are all factors that have been observed to contribute to lack of trust and credibility in scientific studies--especially in situations where there are already public concerns about the accountability or credibility of an agency/organization and the legitimacy of a process. Moreover, the writing of this PHA is filled with redundant statements that repeatedly suggest that the reader "not worry." Yet, the claims are based on inadequate evidence and an over-reliance on negative findings (which suggest no harm) at the expense of inconclusive or positive findings (which suggest some incremental risk). A document, such as a PHA that is intended for a public audience has a special responsibility to provide information in a way that is clear, complete, and responsive to multiple audiences. This document clearly fails to achieve these responsibilities in regard to interested and affected community members who do not have specialized scientific/technical training.
Response: ATSDR attempted to find information that would lead us to believe people were exposed to releases from the Pantex Plant in quantities that could cause adverse health effects, and was unsuccessful. We have had health physicists, public health experts, and occupational medical professionals help write and review this public health assessment. None of them has made this observation.
9. Inadequate discussion of comparison values and safety thresholds - As I noted in my presentation/paper to the PPCAB, regulatory standards for radionuclides are risk-based--that is they do not provide levels below which safety is ensured--only that risks are below a certain level. Any additional exposure to radionuclides can have adverse health effects, although the probability of harm decreases as exposure decreases. Yet, on pg. 36 the report states "The elevated levels of tritium, however, have NOT resulted in radiation exposures or doses above levels that would cause adverse health effects." The definitive tone of this statement suggests that there are levels of exposure to tritium below which no adverse health effects can occur. In other words, any level of exposure has some incremental non-zero risk of causing harm. This risk may be small, but this is a different claim than that there is NO risk.
Furthermore, claims, in the report, of minimal risk below some threshold of acceptability are not based on the factual evidence in a clear way. ATSDR does not clearly state the levels at which exposures to contaminants (such as tritium) would be expected to cause harm. What level of exposure is assumed to cause a high enough probability of harm that it would be called "significant"? What are the "comparison values" discussed on pg. 20 (and implied on pg. 36 as well as other places in the text)? How close are estimated/measured values to the comparison values? How were the comparison values derived so as to protect the "most sensitive segment of the population"? How do uncertainties and interindividual variabilities affect the relationships to the comparison/safety levels? Are safety factors used? Do they have any basis in biological understanding (e.g., pharmacokinetic modeling) or are they derived by dividing by 10 (a traditional, yet arbitrary number)? Without this type of information in the report readers are unable to evaluate the reports claims completely. A reader cannot evaluate the validity of conclusions and the adequacy of assumptions if they are hidden from view.
On page 33 ATSDR writes "in 1975 elevated levels of uranium were found in soils north of Zone 4 but the values were still less than 0.11 Bq/g, about 3 times above background levels." Those concentrations are "similar" to background levels. What does "similar" mean?
On pg. 35-36 they discuss tritium releases and exposures. In the second to last paragraph they write that in 1982 tritium levels in the air ranged from 0.059 - 0.24 Bq/m3, or 1.6-6.4pCi/m3. According to the EPA, National background levels are about 300 pCi/Liter.
1000 Liters = 1m3 according to my calculations, which means the 1982 levels ranged from 1600 - 6400 pCi/m3. This is 5 to 21 + times more than average national background levels. At what point are elevated levels of tritium expected to cause harm?
On page 33, ATSDR wrote that the radiological dose expected from groundwater contamination is less than 4 mrem per year. The issue is cumulative dose, and 4 mrem adds to the cumulative dose, and since everyone agrees that all additional exposures have some risk, 4 mrem implies more risk. How much more risk is the question. This same issue comes up on pg. 34 when they write "there were no levels of uranium or plutonium found that would be considered a health hazard." ANY additional exposure presents additional risk, which is the same as saying additional hazard. So, how does ATSDR define "a health hazard"? What level of risk must be exceeded?
Response: We believe the commentor is referring to the Linear-No-Threshold Hypothesis (LNT) which was established as a regulatory framework. An hypothesis is a theory and LNT has never been shown to be fact. LNT was based on one set of experiments in the 1920s using fruit flies. Data in the scientific literature does not support the concept of LNT. In fact, much of the data, both laboratory and human, shows that a threshold does exist in a response to radiation exposure and dose. This has led to members of the scientific community challenging the use of LNT. Based on the available radiation protection information, radiation health information, and other scientific peer reviewed data the levels of dose associated with releases from the Pantex Plant are not expected to cause adverse health effects.
Radiation evaluations are not performed in a manner such as those prepared for chemical exposures. Chemical toxicological information is based on experiments with laboratory animals and limited experimental data, with insufficient information on pharmacokinetics associated with chemical exposure. Because of these limited data, toxicologists use factors to ensure safety in the case of exposures. In some cases, these factors may result in reductions 10, 100, 1000 or more. ATSDR as well as other federal agencies have developed comparison values for the evaluation of chemical exposures; ATSDR has not developed comparison values for radionuclides in the environment. Therefore, ATSDR uses a different protocol to evaluation the radiation levels in the environment. In the case of radiation dose estimates, the data are more precise as the location and metabolism can easily be tracked. Furthermore, the energy deposited within the specified tissues have been calculated by international organizations. These energy calculations have resulted in the tabulation of dose conversion factors (DCF). These DCFs were used by ATSDR to develop the radiological dose assessment.
In discussing background levels and other levels "similar," or comparable to background, one must realize that variations exist in nature. In the evaluation of background levels, statistics such as the 2 sigma counting error are very important. This error represents the 95% confidence levels of the determinations. In the evaluation of uranium levels associated with the Amarillo area, the variations in background and the associated 2 sigma error suggest that the levels are "similar" throughout the area.
The commentor is misquoting the scientific literature. The value he is quoting is the tritium levels found throughout the country in surface waters. We recommend that he review any surface water data prepared by the US Environmental Protection Agency for their Environmental Radiation Ambient Monitoring System. To further show that the commentor is incorrect, one can calculate the actual tritium concentration. As an example, according to the March 1979 NCRP Report Number 62 (Tritium in the Environment), during the height of atmospheric nuclear testing, the concentration of tritium was 0.48 atoms per cubic centimeter of air at sea level. Using standard values and physical constants, it can be easily shown that this concentration of tritium was 0.3 picocuries per cubic meter, a value 10,000 lower than that quoted by the commentor. Other estimates can be made using the total amount of tritium in the atmosphere and taking the total volume of the atmosphere and these estimates are even lower than the value of 0.3 picocuries per cubic meter.
See our response above regarding thresholds, and risks. Based on information being developed, current data suggests that a radiological dose of 2 rem to the entire body may exhibit adverse health effects. In the case of external exposures, ATSDR has issued health advisories when the external whole body dose, over a period of time, has exceeded 500 millirem per year to a member of the public. Without ATSDR involvement, exposures above this level may well have continued far into the future. The commentor can request copies of these health advisories for his information. The sites at which these health advisories were issued include: the Radium Chemical Company, Queens, New York; the Brown Vandever Navajo Uranium Mines in Bluewater, New Mexico; Austin Avenue, Lansdowne, Pennsylvania; and the H&K Aircraft Components and D&L Sales in Benton Harbor and Belding, Michigan.
10. For example, ATSDR writes in regard to groundwater that "concentrations were below health screening values" (e.g. pg. 25). What are these values? How close are the actual values to the screening levels? Are concentrations likely to be equal to or exceed these screening levels sometime in the future because of additional releases and movement of contaminants?
Response: The current ATSDR health comparison values selected for those contaminants related to the comment are 400 ppb for HMX explosive; 30 ppb for RDX explosive; 5 ppb for the 1,3,5-TNB and 2,4,6-TNT explosive; 20 ppb for 2,4-DNT explosive; 5 ppb for benzene, 1,2-dichloroethane, methylene chloride and trichloroethane; 300 ppb for tetrachloroethane; 100 ppb for total chromium; and 50 ppb for hexavalent chromium. It is critical to note that, even if the contaminant concentrations should exceed those values in the future, a resident would have to be drawing water from that particular perched aquifer, discussed in this public health assessment, for a completed exposure pathway to occur. To date, ATSDR has not been able to identify any residents utilizing that particular perched aquifer downgradient of the contaminant plume.
ATSDR's comparison values are included in the ATSDR database, HAZDAT, which is available to the public on ATSDR's web page (http://www.atsdr.cdc.gov/). It is critical to emphasize that the choice of which comparison value to use for a particular contaminant is determined by the exposure scenario being evaluated and the health assessor's professional judgement. There is no set rule for comparison value selection. Data that exceed our screening values do not automatically represent a public health hazard. We also consider the relative accessibility of the contamination, which populations may be exposed, the magnitude of the exposure, and the toxicity of the contaminants. Our public health calls are weight-of-evidence decisions.
11. ATSDR does not discuss issues of cumulative and multiple exposures. An uniformed reader might be led to believe these are irrelevant. Yet, there is no evidence that even if exposures to single contaminants are below regulatory standards, that there can be no adverse effect from the cumulative/multiple exposures. This issue needs to be addressed in a clear and thorough manner.
Response: A discussion of additive and synergistic effects has been added to the Toxicological Evaluation Section.
12. Validity of databases and monitoring systems - There is no evaluation of the adequacy, completeness, or validity of the databases that ATSDR uses. They do make comments that information is missing (e.g., monitoring for particular contaminants began as late as the 1980s), not reliable (e.g., disease incidence, infant death certificates), etc. However, they do tell us how important these missing, inaccurate, incomplete, etc. data are. How much uncertainty do they introduce? Is there any way to improve the databases? ATSDR can make these kinds of recommendations as part of PHAs. There should be less passive acceptance of the data as provided by Pantex Plant and more evaluation as to data gaps, uncertainties, inconsistencies, and reliability. Again, this kind of discussion would provide readers with better understandings and abilities to evaluate the reports claims completely.
Response: ATSDR is obliged to make a public health call, which is based on an interpretation of available databases. In a perfect world, historical data would be of the quality we collect today; however, we do not believe the historical databases are invalid because they do not meet today's standards. Even if there are data gaps, the data we have are numerous and consistent enough that they would require large deviations to move the average values significantly. None of our ways of looking at the data (e.g., air concentrations, release quantities, CAP88 modeling, soil concentrations) indicate that there are large inconsistencies in the data. We cannot turn back the clock and sample people's blood or urine, which would provide the best measure of exposures to environmental contamination. We have to use the information that we have available--usually, environmental sampling data (such as concentrations of contaminants in the air and water)--to estimate people's exposures, and available health outcome data to determine if adverse outcomes are elevated for an area relative to other areas. Although not perfect, ATSDR concluded that databases reviewed for the Pantex Plant public health assessment were sufficient and appropriate.
To improve the quality of its health outcome database, TDH is currently in the process of expanding an active birth defect surveillance system into the Pantex area beginning with 1998 childbirth deliveries. Data from this active surveillance system will be used to evaluate the accuracy and completeness of vital records in the Pantex area.
13. Specifically in regard to tritium, there are questions about the available data. It is unclear whether the measurement techniques used account for the amount of organically bound tritium in plants.
There is evidence that there can be more organically bound tritium in plants than the amount that is present in the free water in the plants. Do they measure this? Why not, since analytic techniques are available?
In addition, the Pantex 1995 Environmental Report claims that higher observed levels in plants may have been due to a change in laboratories--not in actual levels. Did ATSDR evaluate this possibility? This kind of problem points to potential problems in extant databases, which can affect analyses. Moreover, why do they use background levels (in the 1995 report) as the 1989-1991 observed values? There are, apparently, tritium monitoring data from as early as 1974 (see page 32).
What can be told about the adequacy of the monitoring systems? Were they evaluated and does ATSDR have any recommendations about how to improve them? For example, potential problems are caused by conducting discrete monitoring for contaminants on pre-known schedules (pg. 28). What about the distribution of monitoring wells? There seem to be none in the southeast buffer area--so how can anyone know with a high degree of certainty where the groundwater plumes are going? In addition, on pg. 34 ATSDR writes that no uranium releases were reported in 1987, yet samples show high uranium in soils (3rd full paragraph from top). Are these samples from 1987? If so, what does this say about the quality of the monitoring system and the reporting system? In the 5th full paragraph ATSDR writes that "actual concentrations could be higher or lower than reported because uranium was present in quality control samples." Again, what does this say about the reliability of the quality control? Or of the sampling system? Or of the reporting system? They can't all be working right.
Response: The tritium measured in vegetation was used to determine the tritium in biota. Organically bound tritium (OBT), however, relates to tritium bound to organic molecules. Depending on the type of organic compound to which the tritium is bound, the biological half-life will vary. Pantex did not determine the OBT in biota; however, the dose assessment prepared by ATSDR took into account the total tritium in the vegetation.
ATSDR reviewed the monitor locations, background locations, and windrose charts. ATSDR believes the monitoring system as it exists at the Pantex Plant is adequate for the purposes of environmental assessment.
DOE and the Pantex Plant are aware of the quality control issues and are taking appropriate measures.
14. The contamination of the perched aquifer is not a problem, in the view of ATSDR, because it is not used for drinking water. They then recommend that institutional controls be implemented (pg.65) to ensure that exposures not occur in the future. Are such institutional controls likely to succeed? What evidence/evaluation of institutional controls in such situations exists? Should they be attempting to clean-up? Given the lack of information about the hydrology of the aquifers (pg.7), why doesn't ATSDR recommend additional research on characterizing the hydrology?
Again in regard to groundwater contamination, ATSDR claims that there are no completed pathways of exposure through this media because domestic wells in one location were capped/modified. Is there continuous monitoring of this "potential pathway"?
Response: According to TNRCC, the current limits of the contaminated perched aquifer associated with the Pantex Plant are delineated; therefore, ATSDR does not recommend additional hydrogeological studies of that particular perched aquifer at this time. A single private drinking water well, drawing water from the Ogallala and not the contaminated perched aquifer, is located downgradient of the contaminant plume. That well is monitored by both Pantex Plant and TNRCC staff to ensure that contamination from the perched aquifer does not migrate to the Ogallala without detection. Given TNRCC's reputation for proactive oversight and the fact that the office responsible for Pantex Plant environmental activity regulation is located in Amarillo, we feel confident that the recommended institutional controls will be successful.
15. Inaccurate assessment of pathways - The definition of "potential pathway" and the way it is used in the Pathway Analysis section is odd and unclear. On pg. 38 a "pathway" is defined. A "potential pathway" is also defined--it is one in which one of the five components is missing. The report then goes on to claim that "a potential pathway was eliminated because the old well was closed and sealed, and the new well was constructed in such a manner as to prevent intrusion of the contaminated perched aquifer" (pg.38). But this is a potential pathway according to the definition given one paragraph earlier, "an exposure pathway is defined as potential if one of more of the necessary components is missing or if information for evaluating the pathway is not available." By closing and sealing a well one of the components of a pathway is eliminated, but it is still a potential pathway. This strange writing suggests sloppiness in the analysis and/or a rush to minimize understandings of risk.
More confusion arises when a pathway is defined as potential when "information for
evaluating the pathway is not available." There is an important distinction that should be
made when claiming a pathway is not complete because:
| a) | we have enough information to evaluate it and we can conclude within reasonable levels of certainty that a contaminant will not lead to exposure of
a population, and |
| b) | we do NOT have enough information to know whether a contaminant can lead to exposure of a population. |
ATSDR conflates the two. In the second case (where information is lacking) definitive language about no risks is highly inappropriate. The community would be much better served with a clear analysis of what information is lacking, whether it can be obtained (and with how much difficulty), how reliable it might be, etc. Yet, ATSDR makes definitive claims and it provides no help in evaluating the quality of available data or the importance of missing data.
Finally, ATSDR does not clearly distinguish between past, present, and future potential or completed pathways. Clearly, there were completed pathways with regard to contaminants in a domestic drinking well. There were emissions of uranium and other contaminants from high explosives testing and burning of materials. There were tritium releases that led to elevated levels near the plant. These were completed pathways. At the same time, there are still off-site emissions from high explosive testing and burning of materials. Thus, the statement "No on-site potential or completed exposure pathways are evident from Pantex Plant activities" (pg. 38) is factually incorrect. One might be able to argue that the pathways did not lead to high risk exposures, but this is a different issue. ATSDR seems to equate the existence of a (potential or completed) pathway with the existence of high-risk exposures in a population. These are analytically different; ATSDR's logic is confusing and lacks credibility in the pathways analysis section.
Response: The pathways discussion has been revised.
16. The legal definition of a Health Assessment has not been met - The legal definition of a health assessment, under CERCLA (Comprehensive Environmental Response, Compensation, and Liability Act of 1980), is on the inside cover of this assessment, and is referred to on page i. Compliance with the following items within this legal definition were not met:
"The health assessment shall include...the comparison of expected human exposure levels to the short-term and long-term health effects associated with identified hazardous substances, and any available recommended exposure or tolerance limits for such hazardous substances, and the comparison of existing morbidity and mortality data on diseases that may be associated with the observed levels of exposure."
Nowhere in this document can this information be located. Although there are frequent references to regulatory jargon such as maximum contaminant levels (MCLs), existing regulatory limits are not identified in a comprehensive and clear manner.
One result of this oversight is that the assessment is cluttered with a plethora of regulatory language that has no meaning to the average reader. For example, in Section IV.C.3.a alone (groundwater contamination), there are references to MCL's, risk reduction standards (RRS), health screening values, background levels, and derived concentration guides. Several of these, but not all, are defined on pages 21-22, but none of these are quantified for any contaminant.
A second result is that readers have no indication of the proximity of contaminant levels to regulatory guidelines. All that is known is whether they are above, or beyond, the limits. Considering the frequent use of the word "significant" and other qualifiers in the report, this is a disturbing omission.
Response: Public health implications, including a comparison of health outcome data (such as existing morbidity and mortality data) are discussed in Section VI. The current ATSDR health comparison values selected for those contaminants related to the comment are 400 ppb for HMX explosive; 30 ppb for RDX explosive; 5 ppb for the 1,3,5-TNB and 2,4,6-TNT explosive; 20 ppb for 2,4-DNT explosive; 5 ppb for benzene, 1,2-dichloroethane, methylene chloride and trichloroethane; 300 ppb for tetrachloroethane; 100 ppb for total chromium; and 50 ppb for hexavalent chromium.
17. The Assessment fails to follow its own stated procedures. The Forward contains a discussion of the procedures ATSDR will follow in a health assessment -
a. It states that "when there is not enough environmental information available, the report will indicate what further sampling data is needed." In this report there is not discussion or reference as to the sufficiency of the utilized data.
b. If "people have or could come into contact with hazardous substances," ATSDR is responsible for evaluating "whether or not there will be any harmful effects from these exposures." While ATSDR acknowledges the chemical contamination of the perched aquifer, it evaded its responsibility to determine if people have come into contact with this water source in the past, or what the harmful effects could be if people did come into contact with it in the future.
Since the perched aquifer is contaminated offsite and there are wells that are known to access it, ATSDR should not dismiss this issue by stating the water is not presently used. In reality, the threat to people does exist, and ATSDR should assess this threat fully and clearly. The threat does not vanish because people are prevented from using this water any more than banning fishing in a polluted stream removes the threat of ingesting contaminated fish.
c. ATSDR admits that "the science of environmental health is still developing, and sometimes scientific information on the health effects of certain substances is not available. When this is so, the report will suggest what further research studies are needed." There are no suggestions for further research needs in this assessment. This is particularly disturbing given the definitive nature of the preliminary assessment's conclusions. The reader is given the clear impression that no further research needs are necessary and that perhaps the science is actually fully developed.
d. ATSDR "can issue a public health advisory warning people of the danger" posed by an urgent health threat." Since ATSDR states that the perched aquifer would require treatment prior to it being safe, the action plan should consider a health advisory for the perched aquifer. The importance of this measure is heightened by the fact that DOE is unaware of the exact boundaries of the perched layer, and only has an estimation of the boundary at the present time.
Response:
(a) Although not perfect, ATSDR concluded that environmental sampling databases, reviewed for the Pantex Plant public health assessment, were sufficient and appropriate.
(b) ATSDR, TNRCC, and Pantex Plant staff have been unable to identify any residents utilizing the particular perched aquifer discussed in the Pantex Plant public health assessment as a drinking water source. Local residents and the commentor have been requested to provide the actual locations of those private wells, but have apparently been unable to do so.
(c) To improve the quality of its health outcome database, TDH is currently in the process of expanding an active birth defect surveillance system into the Pantex area beginning with 1998 childbirth deliveries. Data from this active surveillance system will be used to enhance the accuracy and completeness of vital records in the Pantex area.
(d) The exposures from Pantex operations do not meet the requirements for a health advisory. For additional information, see 57 Federal Register 58504 (December 10, 1992) and 58 Federal Register 42732 (August 11, 1993).
18. The Assessment takes a limited approach to contaminant exposure - ATSDR provides no mention of the possible health effects of exposure to multiple contaminants. This omission creates the impression, without any scientific backing, that people can be exposed to multiple contaminants as long as exposure to no individual contaminant exceeds regulatory limits.
ATSDR provides no discussion of the cumulative effects of multiple exposures to multiple contaminants. At a minimum, ATSDR should briefly review the status of knowledge in this field.
Response: A discussion of additive and synergistic effects has been added to the Toxicological Evaluation Section.
19. The Assessment fails to analyze the validity of data sources - There is not a single discussion pertaining to the limitations of the environmental data being used to make the assessment. This is particularly notable since ATSDR states that some systematic studies never occurred before 1995.
Response: See Section IV.D, Quality Assurance and Quality Control.
20. The data base is clearly insufficient to make definitive conclusions - While the assessment makes definitive conclusions about in incidence of cancer, data prior to 1973 is absent, and the early data is admittedly weak. There is also a lack of discussion on the limitations of existing data. Also, since increased regulatory oversight began in the late 1980's, ATSDR should evaluate whether data from this latter era is reflective of historic uses.
Response: Limitations associated with the health outcome data are discussed throughout the public health implications section. We reviewed cancer incidence data (1985 to 1993) and mortality data (1985- 1994) to answer the question posed by local residents " is there too much cancer in the area. According to the Texas Cancer Registry, Texas has cancer incidence data going back to 1976. Incidence data prior to 1976 is very limited and the quality is poor.
21. The Assessment contains numerous errors and inaccuracies -
a. The population estimates for the 1 mile radius are inaccurate. For such a small population, it is curious to see that statistically rigorous methods were used to estimate population densities and demographics. According to a Pantex neighbor, there are at least four children under six years old within one mile of the plant. ATSDR claims there is only a single child under six years old--a 400% error.
b. On Page 25, ATSDR reported, "The risk reduction levels are generally as stringent as the Safe Drinking Water Act MCL's. "Risk reduction level three, which the Pantex Plant Draft Baseline Risk Assessment Work Plan proposes for the perched aquifer, calls for only partial remediation with long-term site controls in some cases.
c. Page 25. "Two new monitoring wells were installed in the perched aquifer on the off-site property in May, 1996. Sampling of those off-site wells in May 1996 also detected explosive constituents; however, concentrations were below health screening values." It is unclear which wells ATSDR is referring to. However, wells PX 06-1030 and PX 06-1031, which are offsite perched monitoring wells, have had RDX levels ranging from 1.54 to 354 parts per billion between June, 1996 and January, 1997 (TNRCC data). All but one reading exceeded the EPA Health Advisory Level for Lifetime Exposure of 2 parts per billion.
Response:
(a) The ATSDR demographic data source was the U.S. Bureau of the Census 1990 Census. Based on residents' input, the number of children under 6 years old residing within one mile of the plant has been corrected.
(b) The operative phrase in the statement is "generally as stringent as the Safe Drinking Water Act MCL's". TNRCC has approved the workplan for the perched aquifer, which is not used as a drinking water source, as appropriate under current conditions.
(c) This paragraph has been revised.
22. The assessment uses language that understates or masks hazardous activities - For example, on page 4, ATSDR reported, "a small fraction of the recovered explosives is thermally treated." ATSDR should report the quantities of recovered explosives that are thermally treated," and identify thermal treatment as open-air burning. Failure to do so conveys technical legitimacy to an archaic method of waste disposal.
Response: The Pantex Plant public health assessment focuses on the air monitoring program and human exposure related to those activities.
Open burning of high explosives as a means of disposal is currently permitted at five other sites in Texas. Lone Star Army Ammunition Plant and Red River Army Depot in Bowie County, Fort Hood in Coryell County, and Camp Bullis in Bexar County are Department of Defense facilities. The GOEX Plant, a private operation in Johnson County, is permitted to dispose of high explosives used in oil drilling applications. Open burning is currently considered a safe and effective means of high expolosive disposal, and is a common method employed throughout the U.S. by both government and private organizations.
23. The assessment is burdened by unprofessional writing and premature conclusions - The narrative contains several examples of overemphasis placed on words. For example, on page 38 it states that "the elevated air levels of tritium, however, have NOT resulted in radiation exposures or doses above levels that would cause adverse health effects." If ATSDR wishes to be a credible, it should make every effort to eliminate a defensive and strident tone in its writing.
In light of the identified shortcomings, with this draft report, the definitive conclusions within it seem unjustifiable at this time. ATSDR appears to be more concerned with pacifying public opinion than fulfilling its role of neutral investigator.
Response: Based on the available peer reviewed scientific data, radiation doses at the levels experienced through off-site exposures have never been shown to cause any adverse health effects. Therefore, we believe no change in wording is necessary.
24. The important role of ATSDR is to:
Response: Comment noted.
25. I 'm not sure this assessment was really needed.
Response: Under the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA), commonly known as the "Superfund" Act, ATSDR is mandated to prepare public health assessments for all sites on EPA's National Priorities List.
The Pantex Plant Department of Energy (DOE) site was proposed for the Environmental Protection Agency's (EPA) National Priorities List (NPL) in July 1991. In August 1991, representatives of Panhandle Area Neighbors and Landowners (PANAL), Peace Farm, Serious Texans Against Nuclear Dumping (STAND), and the Texas Nuclear Waste Task Force petitioned the Agency for Toxic Substances and Disease Registry (ATSDR) to conduct a public health assessment of the Pantex Plant; however, because of the proposed listing, ATSDR was already mandated to conduct a public health assessment of the site. The site was subsequently officially added to the NPL in May 1994.
26. What was the total cost of this report?
Response: The average cost of a completed public health assessment is approximately $200,000. However, cost can vary greatly depending on the commitment of resources to the site.
27. Cancer deaths did not reflect instances of cancer/cancer death of individuals once having lived in the area but are now at other locations.
Response: This limitation has been added to the discussion of cancer mortality data.
28. I found the report (somewhat) ambiguous but it was very informative and well prepared and comprehensive.
Response: Thank you for your comment.
29. I thought the content was not completely factual and its basis was unsubstantial.
There must be some sort of risk working at a plant where nuclear weapons are disarmed.
Tell us the truth, bunnies are running around with seven ears!
The people need to know of every possible concern of plant problems.
I thought it was a lie-don't waste my time.
It is a lie! I envy the genius stature of the person(s) who make the people believe such malarchy. Send out a poll in which you really do want to hear from the public. Thanks.
Response: Thank you for taking the time to comment.
30. Overall, the PHA remains characterized by a lack of documentation on methodology and data sources, and confusion over the scope of the assessment. The document also offers weak arguments that cannot be substantiated under close scrutiny.
It is recommended that ATSDR submit this document to a rigorous, independent professional review by highly qualified experts outside of the internal Pantex and DOE regulatory and oversight structure.
Response: ATSDR submitted the December 11, 1997 Pantex Plant Public Health Assessment, public comment release, to a panel of external scientists and public health professionals for peer review of the technical data presentation format and determination of compliance with ATSDR's public health assessment mandate. The panel consisted entirely of non-ATSDR employees and panel members were chosen for their expertise in epidemiology, toxicology, groundwater, and radiation. The independent peer review process was conducted during January 1998 and the peer reviewers' comments are incorporated in this public health assessment.
31. Define the meaning of the PHA - ATSDR has focused its message on what it can state about Pantex, much to the exclusion of what it does not know about Pantex. ATSDR should make perfectly clear what it cannot determine about the effects of past operations, and identify specifically where in the PHA professional judgement was used when hard quantitative evidence was unavailable.
It is improbable that there is enough information to make any conclusive assessment about the effects of past Pantex operations on public health and safety. ATSDR must make the distinction about what period of the past falls outside of the scope of the PHA due to a lack of information.
ATSDR should make this effort because the primary--and most publicized--conclusions from the PHA include assertions that "the Pantex plant does not pose a threat to public health from any site release at this time" and that Pantex "is probably not the cause of the higher than expected levels of cancer found in the counties surrounding Pantex." (page 1).
This latter statement could easily be interpreted and rewritten, without any loss of accuracy, in the following manner: "There is still a possibility that Pantex could be the cause of the higher than expected levels of cancer found in the counties surrounding Pantex, but there is insufficient data to determine any relationship."
It is only evident that the incidence of cancer and other adverse health problems in the area cannot be traced by evaluating existing operations at Pantex, yet ATSDR implicitly exonerates Pantex of all responsibility. A more generalist, objective approach is needed, and ATSDR should be as definitive about what it cannot conclude and what it can conclude. It appears highly probable that the primary conclusions should include the following:
There is insufficient information to determine whether the incidence of cancer or other adverse health effects in the area are related to past operations at the Pantex plant.
While the higher than expected rate of cancer and birth defects cannot be linked to Pantex operations, Pantex cannot be ruled out either.
There is not enough information to determine whether there were exposure pathways prior to 1970, and the information between 1970 and 1990 may be insufficient to determine whether exposure pathways existed during this time. Any health effects from exposures that may have occurred prior to 1990 cannot be addressed in this document.
There are enough uncertainties about health issues in the area to warrant further investigation.
Response: Response noted.
32. Petition for declassification of classified documents - Throughout the cold war, secrecy compromised the public' s right to know about the impact of DOE facilities on their communities. The cold war is over, and the Department of Energy has made improvements in its openness policies and remains committed--at least on paper--to substantial public access to previously classified documents. In December 1997 Secretary of Energy Pena signed a rule that prohibits DOE from classifying information about nuclear activities that is solely related to public and worker health and safety or environmental quality.''
As a public health agency, ATSDR has an obligation to inform people about past--and present-activities that released contaminants to the environment and raised the risk to the public. It is not the job of ATSDR or any other government agency to decide what the public should know about environmental, safety, and health issues.
ATSDR would fulfill a public service by petitioning the Department of Energy to release all classified documents that were reviewed during the preparation of the PHA pertaining to Pantex operations and that are related to public and worker health and safety or environmental quality.
Response: A review of Pantex Plant classified documents was conducted by the ATSDR Senior Health Physicist during the September 1995 site visit. It was determined that none of the classified documents related to the scope of the Pantex Plant public health assessment. Documents reviewed in preparation of the Pantex Plant public health assessment pertain to waste management, environmental sampling, and environmental restoration activities. Those documents are not classified.
33. Recognize data limitations - During preparation of the PHA, ATSDR relied heavily upon the use of data provided by the Pantex plant, but provided no indication of the validity of the various data sets. This is a serious omission in the PHA, because there is general acknowledgement that pre-1970 environmental monitoring was minimal to nonexistent at Pantex, and that the early environmental monitoring program was insufficient.
It was during this era that environmental abuses during routine operations were more prevalent. During this time, Pantex had substantial releases of depleted uranium to the atmostphere from the firing site, and released wastewater on-site and off-site that was so heavily contaminated that eventually the perched layer of water below Pantex was polluted to levels well above health standards. ATSDR has not offered any evidence in the PHA suggesting that these past activities did not result in exposure pathways that affected people' s health then and now. The Department of Energy and its predecessors are renowned for shortcomings in data collection and analyses, and examples of information gaps at Pantex include:
Prior to 1989, DOE admits that worker exposure records were seriously flawed because internal doses were estimated or not recorded, so the records of DOE nuclear workers who were at risk of internal exposures are "incomplete, misleading, and inaccurate."
Prior to 1990, Pantex admits that the groundwater monitoring program was characterized by a "lack of quality assurance project plans, inadequate sampling protocols, and inadequate sampling techniques." As a result, groundwater data prior to this date is useful for "general trending and rudimentary screening purposes as an indicator of the historical presence of contaminants of concern."
Response: This public health assessment did not consider exposures to workers. Worker health studies are the purview of the National Institutes of Occupational Safety and Health.
ATSDR does not believe the historical databases are completely invalid because they don't meet today's standards. Even if there are data gaps, the data we have are numerous and consistent enough that they would require large deviations to move the average values significantly. None of our ways of looking at the data (e.g., air concentrations, release quantities, CAP88 modeling, soil concentrations) indicate that there are large inconsistencies in the data. Although not perfect, ATSDR concluded that databases reviewed for the Pantex Plant public health assessment were sufficient and appropriate.
34. According to the PHA (page 19), ATSDR, "selects and discusses contaminants based on several factors: sample design, field and laboratory data quality, and companson of chemical concentrations to levels that could cause cancer or other health effects." ATSDR states this occurred, but does not provide any insights on the results beyond generic assurances. The final PHA should contain an appendix that cites each data source used in preparation of the PHA, and that addresses the following factors for each data set:
Statistical validity in terms of sample size and variance
Quality of data in terms of validation of tests, quality assurance programs, sampling protocols, and sampling techniques
Limitations of data in relation to the nature of operations
Additional data needs
The nature of releases and the constraints of monitoring technology in relation to these releases
Response: Approval of Pantex Plant environmental monitoring protocols and quality control programs is the responsibility of the regulatory agencies such as TNRCC and the EPA. ATSDR reviewed the Pantex Plant environmental monitoring program, as approved by TNRCC and EPA and concluded that program is adequate for the purposes of environmental assessment. Appendix B of this document lists the documents we reviewed in reaching that conclusion. Those documents are also available in the Pantex Plant reading rooms.
35. Define limitations of models - ATSDR also relied upon data derived from computational models in preparing the PHA. ATSDR cites the use of the Toxic Chemical Release Inventory that requires manufacturing facilities that use or manufacture certain chemicals in amounts "greater than its specified threshold for manufacture, import, processing, or other use...to estimate their annual releases of such toxic chemicals into the air, water, and land."
Pantex estimates potential air emissions from open burning of high explosives at the Burning Grounds using an air dispersion model. Pantex also made estimated dose assessments from airborne releases of tritium in the 1980's, including the 1989 release of 40,000 curies of tritium. The PHA authors draw conclusive results from these estimates Without any visible analysis of the models used during estimations.
Most of the releases at Pantex are difficult to model because they were, and are, discontinuous, sporadic, and often accidental. If the releases occur in pulses, are the models equipped to assess this variability? ATSDR should provide an appendix in the final Health Assessment that clearly defines the advantages and disadvantages of the models and methods used in estimating air emissions from the Burning Grounds, Firing Sites, and tritium releases. Also, since ATSDR did address accidental tritium releases, it should identify all accidents that have occurred, including uncontrolled wildfires in the 1990's at Pantex.
Response: All models have limitations such as the default values used to define the releases, the release rate variability, the terrain surrounding the release point, the use of standard meteorological data, standardized risk values, the dispersion (plume) characteristics, to name a few. In many cases errors from the use of the models are not major sources of inaccurate model predictions. The documentation included with the CAP88-PC (the users guide) states "Dose and risk estimates from CAP88-PC are applicable only to low-level chronic exposures, since the health effects and dosimetric data are based on low-level chronic intakes. CAP88-PC cannot be used for either short-term or high-level radionuclide intakes." In the case of the Pantex Plant, several limitations previously mentioned are not applicable. For example, the terrain around the facility is flat thus buildings higher than the release point do not affect the dispersion pattern, the terrain is flat with no hills or valleys affecting the plume, the meteorological information was based on historical values, and the low level release continuously occurs from a distinct point; thus the dispersion appears to follow Gaussian characteristics.
36. Correct and clarify - The PHA continues to be troubled by inaccurate, misleading, and/or unsupported statements. ATSDR should make an effort to provide support for these statements or otherwise delete them from the PHA. The list includes the following:
Response:
(a) The Pantex Plant public health assessment reads, "samples from the production wells are used as a secondary check for possible chemical migration from the perched aquifer into the Ogallala aquifer. If contamination of the Ogallala aquifer were to occur as a result of Pantex Plant operations, it would be discovered first in the production wells. In this manner, the Pantex Plant production wells serve as interceptor wells for the downgradient City of Amarillo drinking water production wells." Because the Pantex Plant production wells are located on-site and are draw from the Ogallala, any contaminant intrusion from the perched aquifer to the Ogallala would be detected first in those wells due to their cone of influence.
(b) The monitoring period discussed is 1973 to 1998
(c) A discussion of the EPA Administrative Order has been added to the Surface Waters Section of this document.
(d) The Fourier Transform Infrared air monitor has been in use since 1995 and was an addition to an existing air monitoring network. In September, 1997, TNRCC contracted with Radian Corporation to operate and maintain that monitor. Since that time, the system has functioned reliably.
(e) Exceeding screening values does not mean that adverse health effects will occur. ATSDR considers what exposure assumptions are appropriate and what is known about the toxicity of the substance. In this particular exposure scenario, the air concentration did not pose a public health threat.
(f) ATSDR considered the impact on public health for the release of DU during burning ground detonations. The following parameters were used in the evaluation. The burnings occurred during specific meteorological conditions, that is, low winds and a clear sky. This results in the peak downwind concentration near the release point. DU is very dense and its downwind migration would be limited. We, therefore, used a modifying factor of 1% reaching a distance of 3,281 feet (1 kilometer) from the burning grounds. ATSDR also performed its calculations as an instantaneous release at a height of about 100 feet (30 meters). Using graphs of normalized dispersion factors as published in the Health Physics and Radiological Health Handbook (revised edition), we estimated the downwind concentration of uranium-238 (the major constituent of DU). We then used the inhalation dose conversion factors as published by the ICRP to estimate the whole body dose to an adult. This dose was less than 6 millirem -- a dose not expected to cause adverse health outcomes.
37. (p.i para 4) "... ATSDR scientists then evaluate whether or not there will be any harmful effects from these exposures."
I am amazed that anyone, short of God, would KNOW this. Don't you REALLY mean that you will evaluate whether or not YOU THINK or YOU EXPECT there to be any harmful effects from these exposures? This statement is inaccurate and overstates ATSDR ability.
Response: No change. The sentence states that the evaluation is made by ATSDR scientists.
38. (p. 1 para 1) "No site-related contaminants are currently accessible on- or off-site at levels that would cause adverse health effects."
Please identify WHAT the levels for various contaminants to cause adverse health effects happen to be. Also, please identify which contaminants over what time period and by whom were "monitoring and surveillance off-site" conducted, prior to your drawing this conclusion.
Response: ATSDR comparison values are included in the ATSDR database, HAZDAT, which is available to the public on ATSDR's webpage at http://www.atsdr.cdc.gov/.
39. (p. 1 para 2) "...it is unlikely that area residents come into contact with significant enough amounts of chemicals or radioactive substances... "
What about in the past? The levels of contaminants that were encountered by area residents in the past do not appear to have been evaluated. For example, the release of strontium (used as a tracer during testing) may have resulted in health effects that have not been evaluated. There are likely other examples. Exposures due to past activities do not appear to have been evaluated as reported in this document.
Response: The Pantex Plant public health assessment time period is 1973 to present.
40. (p. 6 para 5) "Rainfall averages between 18 and 20 inches/year and pan evaporation about 95 inches/year, so surface evaporation is up to five times greater than aquifer recharge."
This statement seems to assume that all of the rainfall (and I suspect some of it is actually snowfall) ends up in the aquifer. Technically, this statement appears to be inaccurate.
Response: Sentence revised to " Rainfall averages between 18 and 20 inches/year; however, surface evaporation is up to five times greater than precipitation."
41. (p. 7 para 2) "Overall, pumping of the Ogallala aquifer has lowered the static level of the aquifer by about 2 feet/year, and pumpage exceeds recharge by a factor of 6."
Even this assumption of 4 inches/year recharge to the Ogallala aquifer is not consistent with the DOE statement that the Ogallala aquifer is not in danger of receiving any contaminants that are known to be in the soil and perched aquifers.
Response: Based on updated information, the document has been changed to reflect an aquifer recharge rate of less than 1 inch per year and that pumpage exceeds recharge by a factor of 24.
42. (p. 18 para 1) "effects of gaseous tritium releases (the last one recorded in 1989)"
Was the venting of the Cell One (planned release) that took place after 1989 looked at?
Response: Yes
43. (p. 1 8 para 1) "potential contamination of private and municipal wells;"
There was no evidence provided in this document to suggest that you looked at the "potential contamination of private and municipal wells". What wells did you look at? Were synergistic and antagonistic interactions and effects considered?
Response: ATSDR reviewed sampling data for the City of Amarillo municipal wells and Pantex Plant neighbor wells. In reference to ATSDR's consideration of synergistic and antagonistic interactions and effects, ATSDR uses health comparison values such as Minimal Risk Levels (MRLs) developed by ATSDR when they are available or Reference Doses (RfDs) developed by the U.S. Environmental Protection Agency to evaluate the potential for hazardous substance exposures to cause noncancerous health effects. ATSDR uses various cancer classification schemes developed by organizations such as the National Toxicology Program of the Department of Health and Human Services, the International Agency for Research on Cancer, and the U.S. Environmental Protection Agency to qualitatively identify cancer threats posed by cancer-causing hazardous substances. To quantitate the cancer risk posed by these substances, ATSDR uses cancer potency factors developed by the U.S. Environmental Protection Agency to develop theoretical estimates of how many additional cancer cases might be expected to occur in a given population under specific exposure conditions.
The health comparison values described above and the quantitative estimation of cancer risk do in fact consider cumulative exposures. The MRL and RfD values constitute levels that are believed to be without appreciable risk over a lifetime of exposure. The safety factors used in the derivation of these values also takes into account the possibility of sensitive populations such as infants, the elderly, and immunocompromised individuals being present.
44. (p. 19 para 5) "ATSDR reviewed... non-radiological and radiological data...1973 through l995."
No information as to the quality of the data was provided. Some of the analyses were not conducted by the contract laboratories within the established holding times, for instance. No mention was made that ATSDR eliminated such data in determining whether the data on hand were adequate.
Response: The data we have are numerous and consistent enough that they would require large deviations to move the average values significantly. None of our ways of looking at the data (e.g., air concentrations, release quantities, CAP88 modeling, soil concentrations) indicate that there are large inconsistencies in the data. Although not perfect, ATSDR concluded that databases reviewed for the Pantex Plant public health assessment were sufficient and appropriate.
45. (p. 23 para 2) "The Pantex Plant Groundwater 78 wells [64 , 5 , 7 , and 1]"
The number of wells (64 + 5 + 7 + 1) adds up to 77 wells. Where is the other well?
Response: 77 is the correct number and the change has been incorporated.
46. (p. 23 para 3) "If contamination of the Ogallala aquifer were to occur..., it would be discovered first in the production wells."
This statement does not consider the opportunity for the contaminated water in the perched aquifers (some of which is off-site) to find its way to the Ogallala off-site. Also, it would be more convincing if the lateral distances from various locations of the Ogallala that would be most likely to become contaminated (beneath perched zones, for example) and simple calculations as to various scenarios (if contamination occurred beneath Playa One, for example, what length of time from the moment the Ogallala aquifer became contaminated would it take for the contamination to reach the nearest or the most likely production well?). This information would be useful to understanding the impact of time on discovering any problems that may occur.
Response: To model a worst-case scenario where on-site contamination traveled to drinking water wells, the Draft Final 3-Dimensional Site-Wide Groundwater Flow and Contaminant Transport Model (March 1997) was used as the primary reference. Travel times for groundwater flow within the Ogallala aquifer to selected wells were estimated. The locations of the wells discussed here are illustrated in Figure 12 located at the end of this appendix. Representative values for hydraulic gradient, hydraulic conductivity, and effective porosity were obtained from the primary reference.
First, flow from beneath Playa 1 to Well 15-20 and the Vance Well were considered. Both wells are located approximately 5,000 feet from Playa 1. Using this distance, a gradient of 0.0055, a hydraulic conductivity of 22.5 feet per day, and an effective porosity of 20%, the estimated travel time for Ogallala groundwater to travel from Playa 1 to either receptor well is 8,065 days or 22 years.
A second calculation estimated travel time of potential contamination from PTX06-1038, near the Pantex east gate, to the private wells which are the nearest potential downgradient receptors. Using the same values for hydraulic gradient, hydraulic conductivity, and effective porosity, and an estimated distance of 10,750 feet, the estimated travel time from well PTX06-1038 to the private wells is roughly 19,196 days or 52.6 years.
It is important to note that these numbers are estimates of total travel time based upon average linear advective groundwater flow velocities, and assume uniform hydraulic gradients and properties along the flow path. The assumption is that the contaminants move at the same rate as the groundwater. This does not always occur because the contaminant may chemically interact with the soil, causing retardation of the contaminant movement. No dispersion, dilution, or retardation is accounted for in these calculations; therefore, these are worst-case estimates based on maximum rate of travel. These numbers do not represent any known contaminant loading dates or events within the Ogallala aquifer.
47. (p. 23 para 5) "Constituents analyzed... are listed in Appendix A."
This document suggests that these are the constituents that have been analyzed throughout the period of time being considered (1975 on?). Such is not the case, and thus, by omitting this vital information, this community is misled.
Response: Appendix A is a combined list of environmental monitoring parameters for 1992 through 1997. In preparing the Pantex Plant public health assessment, ATSDR reviewed the non-radiological (chemical) and radiological data collected in support of environmental monitoring at the Pantex Plant for the period 1973 through 1997. Those documents are available in the Pantex Plant reading rooms.
48. (p. 24 para 3) "however, concentrations were below health screening values."
Please explain why ATSDR is unwilling to state what the "health screening values" are in this instance (and in others throughout this report).
Response: The current ATSDR health comparison values selected for those contaminants related to the comment are 400 ppb for HMX explosive; 30 ppb for RDX explosive; 5 ppb for the 1,3,5-TNB and 2,4,6-TNT explosive; 20 ppb for 2,4-DNT explosive; 5 ppb for benzene, 1,2-dichloroethane, methylene chloride and trichloroethane; 300 ppb for tetrachloroethane; 100 ppb for total chromium; and 50 ppb for hexavalent chromium. It is critical to note that, even if the contaminants concentrations should exceed those values in the future, a resident would have to be drawing water from that particular perched aquifer for a completed exposure pathway to occur. To date, ATSDR has not been able to identify any residents utilizing that particular perched aquifer downgradient of the contaminant plume.
49. (p.25 para 3) "Chemical contamination above screening values has not been detected... "
Again, the screening values have not been identified. How can we evaluate future data as they are released if you have not taken the opportunity to identify what the levels of concern for each constituent happens to be?
Response: The current ATSDR health comparison values selected for those contaminants related to the comment are 400 ppb for HMX explosive; 30 ppb for RDX explosive; 5 ppb for the 1,3,5-TNB and 2,4,6-TNT explosive; 20 ppb for 2,4-DNT explosive; 5 ppb for benzene, 1,2-dichloroethane, methylene chloride and trichloroethane; 300 ppb for tetrachloroethane; 100 ppb for total chromium; and 50 ppb for hexavalent chromium. It is critical to note that, even if the contaminants concentrations should exceed those values in the future, a resident would have to be drawing water from that particular perched aquifer for a completed exposure pathway to occur. To date, ATSDR has not been able to identify any residents utilizing that particular perched aquifer downgradient of the contaminant plume.
50. (p. 27 para 3) "Soil samples are collected from two general landscape positions: playa bottoms..."
The recent nature of these soil samples seems to have been overlooked. Please provide the dates that sampling was begun in order to give a more accurate picture of what the current knowledge happens to be.
Response: Routine Sampling of the playa bottoms began 17 March, 1994. The document has been amended to add this statement.
51. (p. 29 para 2) "The only ... inorganic fluoride concentrations in vegetation...."
It is my understanding that the State of Texas no longer has established limits. If this is correct, do these analyses continue to be conducted?
Response: Neither Texas nor the EPA have current limits for inorganic fluoride concentrations in vegetation in effect; however, Pantex continues to monitor this parameter.
52. (p. 37 para 1) Last sentence has a problem, a missing word?
Response: The sentence has been completed as "levels that adversely effect health".
53. (p. 37 para 2) "Quality Assurance and Quality Control summaries from EPA and DOE ... No analytical problems were noted ..."
Please identify the years to which this statement applies. The way it is stated, it applies to the ENTIRE period of data covered by this document. If this is not so, it should be stated. If it is so, it should be clarified.
Response: The summaries covered the environmental monitoring data for the entire period of this report (1973-1998).
54. (p. 61 Question 10) Again, a calculation as to the length of time from when a contaminant would enter the Ogallala aquifer to when it would be detected in the production wells (based on the known rates of movement of water within the aquifer) would be informative.
Response: To model a worst-case scenario where on-site contamination traveled to drinking water wells, the Draft Final 3-Dimensional Site-Wide Groundwater Flow and Contaminant Transport Model (March 1997) was used as the primary reference. Travel times for groundwater flow within the Ogallala aquifer to selected wells were estimated. The locations of the wells discussed here are illustrated in Figure 12 located at the end of this appendix. Representative values for hydraulic gradient, hydraulic conductivity, and effective porosity were obtained from the primary reference.
First, flow from beneath Playa 1 to Well 15-20 and the Vance Well were considered. Both wells are located approximately 5,000 feet from Playa 1. Using this distance, a gradient of 0.0055, a hydraulic conductivity of 22.5 feet per day, and an effective porosity of 20%, the estimated travel time for Ogallala groundwater to travel from Playa 1 to either receptor well is 8,065 days or 22 years.
A second calculation estimated travel time of potential contamination from PTX06-1038, near the Pantex east gate, to the private wells which are the nearest potential downgradient receptors. Using the same values for hydraulic gradient, hydraulic conductivity, and effective porosity, and an estimated distance of 10,750 feet, the estimated travel time from well PTX06-1038 to the private wells is roughly 19,196 days or 52.6 years.
It is important to note that these numbers are estimates of total travel time based upon average linear advective groundwater flow velocities, and assume uniform hydraulic gradients and properties along the flow path. The assumption is that the contaminants move at the same rate as the groundwater. This does not always occur because the contaminant may chemically interact with the soil, causing retardation of the contaminant movement. No dispersion, dilution, or retardation is accounted for in these calculations; therefore, these are worst-case estimates based on maximum rate of travel. These numbers do not represent any known contaminant loading dates or events within the Ogallala aquifer.
55. (p. 71 Appendix A) This list of monitored parameters does NOT apply to the entire period included in this report. Providing this information in this manner is misleading to the community. A more useful tool would be to provide the periodicity of monitoring for EACH of the parameters listed. Then, we might feel somewhat reassured that ATSDR looked in detail at the data for the entire period of time included in this report.
Response: Appendix A is a combined list of environmental monitoring parameters for 1992 through 1997. In preparing the Pantex Plant public health assessment, ATSDR reviewed the non-radiological (chemical) and radiological data collected in support of environmental monitoring at the Pantex Plant for the period 1973 through 1998. Those documents are available in the Pantex Plant reading rooms.
56. (p. 82 Appendix B) Several documents were listed more than once in the review listing: 19 and 33; 23 and 50; and 8, 9, 10 and 57, 58, 59 are some examples of the duplication.
Response: That duplication has been corrected in this document and the references reformatted.
57. (p. 87 Appendix C para 6) "... more than 11,000,000 gallons ... treated."
Please provide a comparative volume that will REQUIRE treating. When will the feat be accomplished? What IS the area of influence of this treatment?
Response: The calculations presented are for the perched aquifer described in this document and located in the southeast corner of the Pantex Plant. The volume of RDX-contaminated water at a concentration above 1 ppb is roughly 1.20 billion gallons. The volume of chromium-contaminated water at a concentration above 100 ppb is roughly 1.15 billion gallons. There is some overlap between the two contaminant plumes, and the total volume of contaminated groundwater would be much less than the 2.35 billion gallon total that you would get by adding the two volumes together. An estimate of about 1.5 billion gallons would be reasonable for the total volume of contaminated groundwater. The process is designed as an interception technique to prevent further plume migration and to ultimately reduce contaminant concentrations; there is no set schedule for project completion. The area of influence is the contaminated perched aquifer.
58. (p. 87 Appendix C para 6) "The treatability system is 100% effective..."
There is absolutely NO scientific basis for this statement! The most that you can possibly state is that the amount that we currently are technically able to MEASURE has been removed (and only if proper analytical methods have confirmed even this statement). And, the detection levels (again) are not identified in this paragraph.
Response: The statement has been amended to read "The treatability system is effective in that treated water contaminant concentrations (explosives and total hexavalent chromium) are below detection limits. The detection limit for explosives is 1 ppb, total chromium is 5 ppb, and hexavalent chromium is 10 ppb."
59. (p. 90 Item 6 para 2) "... which reduces the explosives... to below drinking water standards."
Again, no attempt was made to identify WHAT those STANDARDS happen to be.
Response: In this particular environmental scenario, the appropriate ATSDR drinking water standards for the contaminants of concern are 400 ppb for HMX explosive, 30 ppb for RDX explosive, 5 ppb for TNT explosive, 100 ppb for total chromium, and 50 ppb for hexavalent chromium.
60. (p. 91 Item 8 para 1) " Plant site rainfall and runoff ... flow to four on-site playas."'
This contradicts statements made earlier in the document. See p. 6 para 5.
Response: Page 15, first full paragraph, has been changed to " All surface drainage flows into these playas, of which 4 are located within the DOE-owned Pantex Plant boundaries".
61. What about the claim that each of the persons that irrigated from the pipeline to Pantex Lake have gotten cancer and died? Is this an accurate statement? And if so, what is ATSDR's logical explanation for this seeming anomaly?
Response: ATSDR was unable to locate that population through local inquiries, contact with the Pantex Plant Citizens' Advisory Board, and public meetings. Therefore ATSDR was unable substantiate the claim.
62. ATSDR fails to properly define dose levels considered a public health hazard and misrepresents the meaning of ICRP dose limits - The ATSDR makes several assertions to exposure levels considered to be a public health hazard:
In the introduction of Section IV (Environmental Contamination and other Hazards), the ATSDR report states on page 19 that contaminant concentrations detected on- and off-site are compared to values that are believed to be without adverse health effects upon exposure. The following set of reference values were used in Section IV.C.4 (page 29):
Derived Concentration Guides (DCG): continuous exposure over one year would result in an effective dose equivalent of 100 mrem
Maximum Contaminant Levels (MCL): Regulatory limit for contaminants in drinking water supplies (equivalent to doses of 4 mrem/yr)
Radiation Concentration Guides (RCG): Doses from concentrations in environmental media below RCG values would not exceed 170 mrem.
In the review of environmental data, all of above levels were used even though they clearly represent different dose levels.
The basis for the Derived Concentration Guides are recommendations by the International Commission on Radiological Protection (ICRP). The ICRP recommends a limit of 100 mrem/yr from sources other than medical and those occurring naturally in the environment (up to 500 mrem/yr if 5 year average does not exceed 100 mrem/yr). It should be noted that the ICRP reports do not contain a claim that doses below the limit for the general public would not result in adverse health effects. The 1990 Recommendations of the ICRP (ICRP Publication 60) state that "it must be presumed that even small radiation doses may produce some deleterious health effects" (p.35) and "stochastic effects cannot be completely avoided because no threshold can be invoked for them" (p.25). ICRP further defines its philosophy of setting individual dose limits as follows: "It is the Commission's intention to choose values of dose limits so that any continued exposure just above the dose limits would result in additional risks from the defined practices that could be reasonably be described as "unacceptable" in normal circumstances. Thus the definition and choice of dose limits involve social judgements. These judgements are difficult, partly because the dose limit has to be set at a defined value and there is no discontinuity in the scale of acceptability. For agents like ionizing radiation, for which no threshold can be assumed in the dose-response relationship for some of the consequences of exposure, this difficulty is inescapable and the choice of limits cannot be based on health considerations alone. " (p. 30f.)
ATSDR does not define "levels above which observed adverse health effect levels" in the discussion of tritium releases on p.36. It is stated that an individual would have to consume over 5,000 tons of affected vegetation per year to approach a dose with potential adverse health effects. The reference to a calculation assuming an annual ingestion of 5,000 tons of vegetation is referred to again in the section entitled "Pathway analysis" (p.38) and the "Conclusion" (p.63). No details are provided as to which level of tritium in the vegetation was assumed and which dose was actually calculated.
One way to infer this information is to use the largest dose to a member of the public from tritium releases at Pantex which according to ATSDR was 1.43 mrem in 1989 (Table 2, p.36). ATSDR attributes the doses due to releases of tritium mainly to ingestion of fruits and nonleafy vegetables. The typical consumption of fruits and non-leafy vegetables by an individual is in the order of 100 kg per year. Thus, the same individual receiving a dose of 1.43 mrem per year from the average consumption of fruits and vegetables would receive a dose of ~70 rem per year if he or she ingests the (hypothetical amount of) 5,000 tons per year of vegetation. Such a dose far exceeds the ICRP limit of 0.1 rem per year which was used by ATSDR as one of several reference values "that are believed to be without adverse health effects upon exposure".
In summary, the ATSDR report fails to properly define "values that are believed to be without adverse health effects upon exposure." ATSDR furthermore misrepresents the meaning of implied dose limits by asserting that doses below such limits are associated with no adverse health effects even though the ICRP finds that no threshold can be invoked for stochastic effects such as cancer. ATSDR' s example that the consumption of 5,000 tons per year of tritium contaminated vegetation from the Pantex vicinity would not cause adverse health effects is unsubstantiated and likely to be in error.
Response: Although there have been 100's and perhaps 1000's of studies on low radiation doses from environmental radiation sources or exposures from other types of radiation sources, no adverse health effects have been conclusively shown to exist. In the case of occupational exposure to radiation over a long term, evidence is beginning to surface that a dose of 2 to 5 rem may result in leukemia or other blood related cancers. Exposure to high levels of radiation not found in the environment can lead to deleterious health effects. However, these exposures occur at doses in excess of 10 rem delivered over a short period of time, such as days. Symptoms of these levels of exposure include the classic responses such as hair loss and decreased red blood cell numbers at the doses of 25 to 50 rem.
Regarding the intake of potentially contaminated food stuff, ATSDR has recalculated the values. Using the dose limit of 1 mSv (100 millirem), the amount of fruits and vegetables that would need to be consumed would be 7,500 kilograms or 8.25 tons of food per year. The value in the main text is the amount for documented adverse health effects resulting from exposure to radiation.
63. ATSDR drew far-reaching conclusions from an incomplete set of data - On page 63, the ATSDR report contains the claim that "[o]ff-site radiation exposures to the public living next to the Pantex Plant are well below levels that would cause adverse health effects" (note the present tense). The major source for this far-reaching conclusion is the data contained in the environmental monitoring reports. The monitoring system in place around Pantex was initiated in 1972 and, in ATSDR's own words "[a]t that time (...) contained a limited amount of data" (p.32).
The Atomic Energy Commission (AEC) chose the Pantex site for the expansion of its nuclear weapons assembly facilities in 1951. No information was reviewed by ATSDR for the 22 years from 1951 up and including 1971 before environmental monitoring was initiated in 1972. ATSDR did not comment on potential exposures prior to 1972 and their impact on the health of local residents. This omission is especially serious since it is now reasonably well established that radioactivity releases from nuclear weapons plants were typically far higher in the 1950s and 1960s than in subsequent years.
Response: Although the Pantex plant is considered a nuclear weapons plant, its operations are not comparable to weapons plants such as Hanford, Oak Ridge, or Savannah River. These are materials production centers, not final weapons fabrication (assembly, disassembly, and quality assurance testing) facilities. The nature of the operations at Pantex are much different than the production centers. The Pantex releases to the environment, in ATSDR's opinion, have not impacted the environment to the same extent as the materials production centers. We believe this comment is without basis.
64. ATSDR analyzed available environmental data in an inappropriate fashion - ATSDR summarizes doses to the public resulting from tritium releases on p.36. The largest release of 40,000 Ci occurred in 1989 and is identified as an accidental release. The "estimated dose to the public" is calculated to be 1.43 mrem. The calculation assumed a uniform release during the year. If the same amount of radioactive material is emitted over a short time period rather than evenly distributed over an entire year, the doses to the maximally exposed individual would be expected to be significantly larger. Of course, a smaller number of people would be affected by the release. It is clearly inappropriate to ignore the possibility that some individuals received a much larger dose than the one calculated assuming the uniform release.
In summary, the ATSDR report fails to provide a clear definition as to the level of radiation exposure associated with adverse health effects even though the claim is made that no adverse health effects are to be expected. ATSDR's claim that the consumption of 5,000 tons per year of tritium contaminated vegetation from the Pantex vicinity would not cause adverse health effects is unsubstantiated and likely to be in error. The report fails to address the lack of environmental data for 22 years of Pantex Plant operation. Data used in the assessment of tritium exposures was improperly interpreted. We recommend rejecting the ATSDR report because its major conclusions regarding the public health situation around the Pantex Plant are based on inappropriate methods and the selective use of data.
Response: The comment is comparing two different aspects of the dosimetry question. Bioassay data of the exposed workers indicated that the maximally exposed individual received 1.3 rem, and there was a total of 1.5 man-rem for all monitored individuals. Man-rem is the sum of exposures to all the monitored workers; therefore with a total exposure of 1.5 man-rem with one individual receiving 1.3 of that total exposure, the remainder of the monitored workers collectively received less than 200 millirem. To put the exposures in perspective, the maximum allowable annual exposure dose for monitored workers is 5 rem. The dose estimate at the Pantex Plant boundary for the incident was less than 1 millirem. The tritium released in the gravel covered containment facility has been slowly diffusing from the facility since the day of the accident. It is this slow leak rate on which ATSDR based its dose calculations - not the immediate release. The slow leakage has reached the off-site public areas, not the 40,000 Ci.
65. Is the Pantex Plant the only place in the state where the outdated practice of open burning is permitted?
Response: Open burning is currently considered a safe and effective means of high expolosive disposal, and is a common method employed throughout the U.S. by both government and private organizations. Five other sites in Texas are permitted for open burning of high explosives as a means of disposal. Lone Star Army Ammunition Plant and Red River Army Depot in Bowie County, Fort Hood in Coryell County, and Camp Bullis in Bexar County are Department of Defense facilities. The GOEX Plant, a private operation in Johnson County, is permitted to dispose of high explosives used in oil drilling applications.
Figure 12: Well Locations (15-20, Vance, and PT06-1038)
