UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
RADIATION AND INDOOR ENVIRONMENTS NATIONAL LA80RATORY
P.O. BOX 98517
LAS VEGAS NEVADA 89 193-8517

September 25, 1998

Mr. Kenneth A. Hoar, Director
Environment, Safety & Health Division
U.S. Department of Energy
P.O. Box 98518
Las Vegas, NV 89193-8518

Dear Mr. Hoar:

Enclosed are the EPA, Radiation and Indoor Environments National Laboratory's (R&IE) comments to the Routine Radiological Environmental Monitoring Plan (RREMP). These comments, as well as R&IE's participation in the drafting of the Technical Design Basis Document (TDBD) which preceded the RREMP, represent the best opinion of the laboratory staff, but do not represent the position of EPA Region 9, which has regulatory authority for NESHAPs compliance. Through a conversation with George McNeill of DOE on 9/22/98, we are aware that DOE intends to implement the RREMP on 10/l/98. We strongly encourage that DOE secure approval for any compliance based monitoring plan from EPA Region 9 before implementing any plan.

In addition to technical issues, we are especially concerned about how the RREMP has been described as a consensus plan, and how-EPA has been portrayed as a part of that consensus. In contrast to the TDBD, the RREMP is not a consensus document. Through our laboratory's involvement in the TDBD Working Group, we have provided DOE with what we believe are proposals for revising on and offsite monitoring that meet applicable requirements in an effective, technically defensible, and cost effective manner. We stand behind those proposals, which appear in the consensus based TDBD. Our comments on the RREMP reflect our input to the TDBD, and are detailed in the attached "COMMENTS ON ROUTINE RADIOLOGICAL ENVIRONMENTAL MONITORING PLAN

Recycled/Recyclable - Printed with Vegetable Oil Based Inks on 100% Recycled Paper (40% Postconsumer)

Sincerely;
Jed Harrison
Director


COMMENTS ON ROUTINE RADIOLOGICAL ENVIRONMENTAL MONITORING PLAN

USEPA, ORIA, R&IE

GENERAL COMMENTS

First and foremost, we must emphasize that the document does not accurately portray the EPA's participation in the development of the Routine Radiological Environmental Monitoring Plan (RREMP). The RREMP is not a consensus document, and does not represent the consensus position reached by the Technical Design Basis Document (TDBD) Working Group, which included EPA. The description of the process appearing on page 1-2 indicates that EPA participated on a team which used a consensus based approach to developing details of data quality objectives for each media, and then BN prepared the sampling and analysis approach for each media. EPA did participate on the TDBD Working Group, submitted sampling and analysis approaches for all offsite media, and provided comments on the draft TDBD (which provided a basis for the RREMP). However, EPA has not been included in any of the final decisions on design of the offsite program. In addition, it appears that very little of EPA's proposed sampling and analysis approach, or comments have been incorporated in the final draft.

It is unfortunate, and inappropriate that the EPA seal still appears on maps which have been heavily modified from the TDBD consensus versions. We endorsed the appearance of the EPA seal on the TDBD maps as a symbol of the cooperative effort which took place through the development of the TDBD document (much as described on pg 1-2 of the RREMP). However, significant revisions in the scope of the proposed offsite monitoring program were made after the TDBD process concluded, and the RREMP no longer represents EPA's position.

A notable change from the TDBD that EPA is aware of is that this document has been touted as a consensus document, and the introduction (1 .0) of the RREMP describes a consensus approach, which includes EPA. In fact, DOE's contractor prepared the RREMP without input from the TDBD Working Group, which deviated significantly from the TDBD recommendations. For this reason, the description of the consensus process, as it applies to the RREMP is incorrect, and the use of the EPA seal on the RREMP is inappropriate, and very misleading.

With the evolution of the DOE's mission at the Nevada Test Site (NTS), and associated budgetary impacts, EPA has been aware that reductions in the scope of the environmental monitoring program were technically advisable and fiscally appropriate.

Through our participation on the TDBD Working Group, EPA proposed to DOE cost effective strategies for reducing NTS related environmental monitoring programs that addressed technical and stakeholder driven issues. EPA remains supportive of the consensus plan which was included in the TDBD.

EPA is very supportive of addressing stakeholder concerns in the design plan, and this is consistent with the DQO process. We are pleased that the RREMP attempts to take into account stakeholder concerns, however a clear process for identifying stakeholder interests does not appear in the RREMP. We believe that stakeholder interests should be an input into the design of the monitoring plan, and furthermore, we believe stakeholders should be made aware of planned changes prior to their implementation.

R&IE attended a presentation by Dr. K.C. Chun, where an independent review of the RREMP by Argonne National Laboratory was presented. Argonne's findings were well accepted by participants from DOE, EPA, DRI and BN. In this presentation, one of EPA's key concerns, the need for additional receptor monitoring, was identified. This confirms to EPA that modifications to the plan are warranted prior to implementation.

EPA is disappointed in the reduction in scope portrayed in the RREMP, particularly in the offsite monitoring portions, which depart significantly from what we had recommended in the TDBD. In general, we believe the reductions in offsite monitoring are excessive, drastically reducing segments of the program which have been driven by stakeholder concerns in the past, such as up gradient well, milk, and other human food sampling. In addition, the offsite sampling frequency for down gradient wells near the NTS boundary has been reduced below the level deemed prudent by EPA. Predictive modeling has been relied upon heavily in the RREMP decision making process, but accompanying uncertainty analysis of these predictions is minimal. As both DOE and EPA place protection of human health and the environment as a priority, we believe the loss of confidence that would occur if the public is exposed without adequate measurements is not worth the incremental cost savings afforded by the severe reductions

SPECIFIC COMMENTS

2.4.6, page 2-15: The discussion of plants does not identify ingestion of contaminated garden crops by offsite residents. This is a pathway of concern to stakeholders, and should be considered.

EPA recommends annual samples of crops from offsite resident gardens. 2.4.7, page 2-15: The discussion of animals correctly identifies the milk/meat pathway to man from livestock ingesting contaminated vegetation. No discussion appears here or elsewhere describing why it is not deemed significant enough to monitor.

3.0, page 3-1: The diagram on this page and associated discussion on this and the next page seem to suggest that DOE proposes to monitor for compliance with only those applicable requirements which are both relevant and pertinent. However, all applicable requirements must be met. It might be less confusing if "pertinent" and "relevant" were not depicted as subgroups of applicable. The discussion seems to imply that DOE will not necessarily comply with all applicable requirements.

EPA recommends that this section be clarified.

4.1.4.1, page 4-4: The discussion of shortcomings of receptor monitoring for air obscures the basic issue of uncertainty in resuspension of plutonium for air transport to the offsite. Given the complexity of establishing a credible plutonium source term for the large and widely distributed quantity of plutonium on the NTS, receptor monitoring may be the only way to assure that the offsite public is not being unduly exposed.

Such uncertainty in source term does not exist for tritium, the evaporation and transpiration of which can be convincingly estimated. Moreover, fission products, primarily cesium 137, can easily be detected through high volume air sampling at levels contributing 10% of the dose ranges predicted. Tritium and fission products can be easily addressed by other means; it is plutonium which is problematic.

EPA primary concern in this area has always been to assure that the public near the NTS is adequately protected. The changes in the RREMP to high volume air sampling from what was proposed by EPA appear to undercut this assurance. Reducing this number to 5-6 is not substantiated, since it is clearly an important measurement.

EPA stands by our original proposal for a network of 12 high volume air samplers at near site CTLP and other locations, for the very reason which has been cited as a shortcoming of receptor monitoring - that the location of the Maximally Exposed Individual might vary from year to year.

4.2.2.3, page 4-19: The statement is made that current site knowledge eliminates the possibility of transport of radionuclides from source areas to wells up gradient of the NTS, and this is used as a basis for discontinuing the sampling of these wells.

This statement needs to be substantiated more clearly, since it may be viewed with contention by stakeholders: what site knowledge, what mechanisms of transport are being considered, what persuasive evidence supports discontinuance, etc.?

4.2.2.3, page 4-19: The discussion of water supply monitoring wells proposes to consult with CTLP/stakeholder programs to assure that the concerns of local communities are met. This suggests that stakeholder concerns have not yet been considered for well sampling as is implied by the discussion on page 3-2.

The process of consideration of stakeholder concerns needs to be described in more detail, as observed in the general comments.

4.2.2.4, page 4-20: The use of standard and enriched methods of tritium analysis is discussed, and it is stated that in practice the RREMP may choose the enriched tritium method.

The criteria for use of the enriched method needs to be presented, since it is not needed for determining whether the stated action level has been met, and is considerably more expensive than the standard method.

4.3.1.2, page 4-23: The use of the boundary definition value of 200 pCi/g of plutonium or 25 microR/hr exposure rate from gamma emitters is not substantiated. The U.S. Nuclear Regulatory Commission expects its licensees to clean up plutonium contamination to less than 25 pCi/g in soil, and the exposure value of 47 millirem/yr computed by BN to a resident rancher is several times higher than the maximum target value used in Superfund cleanups (1 E-4 lifetime risk). The criteria used here appears arbitrary. Moreover, it is possible that the use of a lower boundary definition of plutonium and fission product surface contamination might result in a significantly larger estimate of potential air transport source term, since a much larger area may be involved

The use of the boundary definition value of 200 pCi/g of plutonium or 25 microR/hr exposure rate from gamma emitters needs to be substantiated.

4.4.4.2, page 4-32: The selection of one upwind and one downwind milk sampling location appears to have been made to meet the minimum recommendation of the DOE Regulatory Guide DOE/EH-0173T. It is questionable whether a site such as NTS which has a large inventory of uncontrolled surface and subsurface contamination and a history of far more releases of radioactivity than most DOE sites, should consider this minimum criteria to be sufficient. In addition, the sampling frequency for milk is not consistent with the Regulatory Guide, which recommends quarterly composites at minimum when strontium is the contaminant of concern.

The sampling scheme proposed by EPA and conducted in the recent past (8 sampling locations including sampling for goat milk) is considered minimally appropriate.

Table 4.4, page 4-41: The sampling frequencies of offsite springs have been reduced significantly below those proposed by EPA. Specifically, sampling frequency of the Revert Spring (which flows at up to 700gpm) and the Spicer Ranch spring, which lie close to the NTS boundary in the main direction of the modeled groundwater plume have been reduced from quarterly to annually, and the Ash Meadows springs (Crystal Pool and Fairbanks) are to be sampled every 3 years. Implicit in this decision is an apparent confidence in the predictions of the groundwater transport model and the adequacy of the onsite monitoring network and TTR wells to detect a groundwater plume before it shows up offsite. All groundwater models are known to predict imprecisely, and the confidence in the adequacy of the onsite network to detect a plume if present has not been calculated for the model used at NTS. Moreover, in past discussions related to the TDBD Working Group, EPA was informed that worst case predictions of the model indicate that a plume of tritium may already be past the site boundary.

EPA believes that prudence is warranted in this situation. With this in mind EPA is strongly urging that the original proposed sampling frequency be reconsidered and implemented for all down gradient springs and wells.

Table 4.6, page 44-49: The proposed sampling frequency of many of the more distant down gradient wells have been reduced from yearly to every three years. This also suggests a heavy reliance on model predictions and the onsite network and the same comments as above apply. In addition, sampling of water supply and private wells on a three year basis fails to adequately address a significant stakeholder issue. Since the standards for drinking water are expressed in terms of annual dose, a three year sampling cycle allows for the possibility that the standard could be exceeded for up to three years before detection. This is not likely to be very comforting to a member of the public.

EPA urges DOE to reconsider and implement the sampling frequencies which EPA proposed for TDBD.