1. Page A.3: Item 18 seems to be of central importance. Unless the calculated values are supported by both recent and earlier monitoring data, the models will lack credibility. For that reason, early assessment of modeling results and comparison with measured data are crucial and would avoid premature publication of dose values that cannot be supported by the evidence.

PNWD 2023: Parameters Used in the Environmental Pathways (DESCARTES) And Radiological Dose (CIDER) Modules of the Hanford Environmental Dose Reconstruction Integrated Codes (HEDRIC) for the Air Pathway.

This is a useful summary of transfer coefficients and other environmental parameters. Each parameter is assigned a value, or the mode of calculation is indicated, and the technical basis is discussed. References are provided in each case. However, the committee notes the following:

  1. Section 4, Page 1: The use of precision and accuracy continues to be fuzzy. Precision appears to be equated with choosing a distribution function, and accuracy is defined as the range of values that is judged to be consistent with expert opinion or data.

  2. Section 1 Page 2: It seems worthwhile to make and maintain a clear distinction between uncertainty and variability here and throughout the text.

  3. Section 1, Page 3: It is noted that “some additional data sources certainly exist.” That is an important point and one where there is some concern on the part of the community. The amount and nature of data that have not yet been uncovered and the efforts being made to find them must be addressed to allay concerns that important information is being ignored.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Page A.3: Item 18 seems to be of central importance. Unless the calculated values are supported by both recent and earlier monitoring data, the models will lack credibility. For that reason, early assessment of modeling results and comparison with measured data are crucial and would avoid premature publication of dose values that cannot be supported by the evidence. PNWD 2023: Parameters Used in the Environmental Pathways (DESCARTES) And Radiological Dose (CIDER) Modules of the Hanford Environmental Dose Reconstruction Integrated Codes (HEDRIC) for the Air Pathway. This is a useful summary of transfer coefficients and other environmental parameters. Each parameter is assigned a value, or the mode of calculation is indicated, and the technical basis is discussed. References are provided in each case. However, the committee notes the following: Section 4, Page 1: The use of precision and accuracy continues to be fuzzy. Precision appears to be equated with choosing a distribution function, and accuracy is defined as the range of values that is judged to be consistent with expert opinion or data. Section 1 Page 2: It seems worthwhile to make and maintain a clear distinction between uncertainty and variability here and throughout the text. Section 1, Page 3: It is noted that “some additional data sources certainly exist.” That is an important point and one where there is some concern on the part of the community. The amount and nature of data that have not yet been uncovered and the efforts being made to find them must be addressed to allay concerns that important information is being ignored.

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section 2, Page 2: Under “Air Transport Data,” it is stated that “The parameters used by the RATCHET code to estimate the air concentrations and deposition rates are not included in this report.” In fact, much of the detail in this document refers to input from another source (from same page, at bottom: “These diets are stored in a database that will be accessed by DESCARTES”). Much of the information used at Hanford is considered proprietary, and it is difficult to obtain all the information desired on the parameters that are used in the codes. Section 2, Page 3: The word will is used 10 times in referring to what is going to be done. Section 2, Page 10: Variables, such as λperc, that are functions of other (often random) variables might better be termed coefficients than constants. This comment would apply throughout the text. Section 2, Page 11: In the definition of “ρ usl,” “from” should be deleted. Section 2, Page 11: In “(t,l). at location 1, Ci.s/m3,“ per d” should be deleted. Section 2, Page 13: There is only one shielding factor. Was “Sh1” intended to be “Sh1”? This comment would apply throughout the text. Section 3: The methods of “experts” are used to obtain some idea on the uncertainties involved in the analysis and self-assessment process. That is acceptable, but it behooves the authors to list the “experts” used for quantification of the assessment. It is extremely important that these experts not come from Battelle, Pacific Northwest Laboratories if public credence is to be achieved.

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section 3, Pages 2-4: The definitions of random variable, probability, probability density function, and the specifically considered probability functions need to be revisited. The reader should not be asked to consult an external reference to obtain a mathematical definition of a term being used. The definitions should make conceptual interrelationships clear. There should be no confusion between concepts and computer-programming techniques. A probability density function is not a computer-language subprogram. Are the definitions of random variable and of range of a model parameter correct? Section 3, Page 7: “After the literature review was completed, the maximum and minimum parameter values found in the literature (relevant to the release scenario at Hanford) were determined.” The rest of the paragraph states that “either a self assessment or an informal solicitation of expert opinion was used to evaluate whether the maximum and minimum values as obtained from the literature should be changed to reflect physical limitations on parameter values and conditions present in the HEDR study region.” Those procedures are appropriate, but some justification should be provided for the choices of values to reflect local conditions. Section 3, Page 8: “But it is also the most expensive.” What is expensive? What costs are referred to? Do they refer to computer time? If so, considering the cost of this entire project, it is difficult to understand what is meant by “expensive.” Section 3, Page 8: At the bottom is the quotation, “Although this quote may cause one to pause and wonder about the validity of using expert opinion to select pdfs, it is important to remember that all science is subjective to some degree.” Statements like that are matters of philosophy, which are inappropriate in a quantitative analysis, as is to be performed in this case. Section 3, Page 9: In the first paragraph of Section 3.3.3, the second sentence should end with “. and defined in Section 3.1.”

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section 3, Page 10: The distributions in Figure 3.1, with the exception of the top two, are not carefully drawn. The loguniform distribution does not show a clear 1/x behavior between a lower and an upper bound. The lower bound of the triangular distribution is not shown. The normal distribution appears to be asymmetric and to have upper and lower bounds. The lognormal distribution also appears to have an upper bound. Section 3, Page 11: The most likely value for a parameter with a triangular pdf might be at or near the maximum or minimum. Section 3, Page 12: In Section 3.4, it could be made clear that the lowest selection frequency is by realization, in which case the same parameter value is used for all locations and times in a realization. Section 4, Page 1: There is reference to “peer review” of the data quality objectives (DQOs). The nature of the peer reviews and the persons involved are not clear. This might be discussed elsewhere; if so, it should be referred to here for easy access by a reader. If not, such information should be included. Section 6, Page 1: The parameter-specific descriptions after this section could well be moved to Appendix A and Table A.1 to this section. In any case, it seems awkward to have each parameter considered in a separate main section. Section 6, Page 1: It should be noted that the original-source criterion described eliminates most of the “data” in the scientific literature. In any case, the criterion substantially strengthens the study. Section 6, Page 3: The parameter-distribution sections would be strengthened by including the essential parameters of the chosen distribution, e.g., the mean and standard deviation for a parameter with a normal distribution.

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section 6, Page 6: Again, many of the “constants” could be designated as coefficients. Also, a parameter like iodine-131 λleach might better be described as the “leaching rate coefficient from root zone to deep soil.” Section α, Page 1: From the Utah-study dosimetry work, the foliage interception constant is an important parameter that has been studied by a number of researchers, but only three references are cited. How thorough has the review of the literature been? Section α, Pages 1-2: With regard to Miller (1979), it would be clearer to state that the median values for pasture and silage were 1.84 and 1.15 m2/kg(dry), respectively. The corresponding GSD values were 1.55 and 1.31. Section α, Page 1: The committee believes that the rainfall intensity and the chemical form of iodine-131, at least in rainfall, have a strong influence on the value of α. Also, as demonstrated in fallout near the Nevada test site, iodine-131 attached to large aerosols has a smaller value of α than molecular iodine. Section α, Page 1: The paragraph starting at the bottom reads “Miller concluded that a lognormal distribution described this parameter. ” It is important to state how the conclusion was reached, e.g., by eye after plotting or by mathematical analysis. Section α, Page 2: At the top of the page, the same paragraph mentioned in the previous comment states that the geometric standard deviation is given as 0.44 around a geometric mean of 0.61. But it appears that the GSD is 0.61/0.44 = 1.39 and the values for + or - one GSD were 1.05 and 0.17. Is this correct? Section 131I λleach: Several problems are common to this section and to Section 131I λperc. First, the model of Baes and Sharp (1981) is proportional to the net infiltration rate, P + I - E. Note that the sign for the evapotranspiration term, E, is incorrect in both

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS sections. The net infiltration rate through this layer on a day-to-day basis is certainly not constant over an entire year but depends on whether irrigation is being used and whether there is precipitation. Second, the value for d is the depth of the layer being leached. In this case, the nominal depth of the layer is 15.0 - 0.1 = 14.9 cm. The corresponding depth for the upper soil layer is 0.1 cm. The appropriate values for ρ, Kd, and θ depend on soil type and might well depend on location in an area as large as that being assessed for the HEDR project. Unless a sensitivity analysis indicates that expected seasonal (if not daily) and location variations in this quantity are not important for the assessment, the equation should be on the DES equation list and the parameter values selected accordingly. This parameter is of no significance for iodine-131 and does not warrant three pages of discussion. The modified form of the Baes and Sharp equation should be provided, because it is not clear how it was used. A critical assumption that is made is that the farmers overirrigate by 10%; this value is considered to be a constant, so it must be well documented with a reference. Section λweath: All the information indicates that some sort of logarithmic function fits the rate of removal of radionuclides from crop surfaces. Some of the references used (Hoffman and Baes, 1979; Miller and Hoffman, 1983) describe this function as logarithmically distributed. Many things in nature appear to be described by lognormal distributions. That was concluded in this document: “HEDR model developers did not believe the experimental evidence was strong enough to support the assumption of a log-normally distributed weathering half-time. A triangular distribution was assumed.” Selection of the triangular distribution should be justified. Section 131I λperc: As stated above, the depth of this layer is only 0.1 cm, not, 15 cm. As assumed by the authors, evapotranspiration probably does little to reduce the net infiltration from this layer. Other comments for the section above also apply here. The way in which the Baes and Sharp equation was used should be provided; it is not clear what values of P and E were used. Were the results of Barth and Veater (1964), which apply to iodine-131 in an arid environment, ignored in favor of a model that has probably not been validated? If so, why?

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section 131I λsplash: The authors comment in paragraph 2 and the “technical basis” that DESCARTES does not consider the timing of specific weather events. One wonders, then, whether they factored in the additional contribution that rainfall events made to the foliage and ground exposure. No factor seems to have been introduced to account for this, but it could make a substantial contribution—much greater than some of the other factors that are modeled. Section λtrans: The definition of this parameter is not clear. An alternative might be “a rate coefficient used to model the transfer of activity from a nonedible to an edible portion of a crop.” It is likely to depend on the radionuclide considered. Section λweath: As indicated in the text, the parameter value can vary as a result of the weather conditions, the crop under evaluation, and the chemical form of the radioisotope. Here again, the values reported for iodine, this time by Miller and Hoffman (1983), seem to have been ignored, and other, more generic values not specific to iodine, seem to have been chosen. Section ρrz: Equation DES-13 is for the rate of uptake through roots, not the percolation rate constant. The relationship of the identified SCS records to the distribution for this parameter is not clear. Section ρusl: Is there a good reason to believe that the areal bulk density of surface soil (1 mm deep) is the same as that of the upper 5 inches? One would expect the surface soil to consist partly of organic debris. Section BR: The maximal values look high, especially for adults. Consideration could be given to Layton's paper (Health Physics, January 1993) according to which the ICRP estimates are too high. Section 131I CR: This parameter is of little significance for iodine-131 and does not warrant four pages of discussion.

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section 131I DFimm: The wide range of values selected seems to reflect a lack of knowledge by the experts. The calculation of the parameter value is fairly simple, and the results obtained should be similar because they depend essentially on the factor for conversion from exposure to thyroid dose. The committee did not check Kocher's value but could not help noticing that there is an error by a factor of 10 in the conversion from “Sv/yr per Bq/cm3” to “rem/d per Ci/m3.” Section 131I DFing: Another consideration would be the thyroid-function status of the exposed people. This would be affected by the stable-iodine intake of residents and other considerations. Section 131I DFrz: For purposes of calculating external doses to the thyroid, much of the activity calculated for the root zone might actually be in the upper soil layer, inasmuch as nonirrigated land could provide the greatest contribution to the external dose. Section 131I DFrz: The reason why the unit rem is used for external dose calculations (as opposed to rad for internal dose) should be explained. Also, it is not clear whether the iodine-131 vertical profiles used are consistent with the range of values selected for the percolation rates. Section 131I DFusl: The 30% loss due to surface irregularities is probably a low estimate of the losses due to those irregularities and to the substantial shielding that will occur at shallow slant angles because of the finite thickness of the 0.1-cm upper soil layer. Section FSchicken: The assumption that chickens were free-ranging in the mid-1940s does not seem extreme. It would not be surprising to find that free-ranging chickens ingest more soil than those fed a commercial diet.

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section FScow: The reason why the soil ingestion rate by cows is inversely related to the fraction of time on pasture should be made clear. How can a cow ingest 2 kg of soil per day without being on a pasture? Are the other amounts believable? Section 131I ftrans: Why should λtrans be considered to have some dependent relationship on λweath? Why not estimate λtrans on its own and let the value of ftrans = λtrans/(λtrans+λweath) fall where it will? Section 131I ftrans: The committee does not understand the logic of the derivation of the parameter values. The experimentally determined translocation factors for iodine that do exist are primarily from leafy-vegetation experiments, but they are not used, because it is assumed that all surface deposition is translocated to the edible fraction for leafy vegetables and pasture grass. For grain and fruit, experimentally determined values do not seem to exist, but values were selected in this document. Where do they come from? Section kg: Were the data selected to calculate kg (and ks) chosen to represent conditions in the 1940s? Section Iodine Lproc: The experimentally determined values that were published after the Chernobyl accident should have been reviewed. Section ML: Particulate mass loading depends heavily on wind speed. If a sensitivity analysis shows any significance for resuspension of particles, considering it to be independent of meteorologic conditions should be rethought. Section Rio: The values that are selected are not based on iodine measurements. However, iodine-131 was measured in at least one experiment (Megaw, 1962); the average value obtained in that experiment was smaller than the selected minimum.

OCR for page 41
THE HANFORD ENVIRONMENTAL DOSE RECONSTRUCTION PROJECT: A REVIEW OF FOUR DOCUMENTS Section Rv_chicken: The consumption rates for this section would appear to be based on about 15 weeks, rather than 22 weeks, of feeding. Section 131I Sh1: The maximum of 0.93 is highly unlikely. Even if a person remains near an exterior wall with little attenuation, the maximum of the shielding factor is about 0.5 inasmuch as the radiation originates only from one side (the other side of the wall). The terminology used in this section does not always make clear whether the discussion is of a factor Sh1 or its complement 1-Sh1. Section 131I TF milk_herd: The discussion in this section leads one to think that the variance for this factor is simply the variance for an individual cow divided by the size of the herd. It does not account for effects of breed, feed, herd-management practice, or milk-productivity differences, etc. In short, the members of a herd are not a random sample of the cattle considered in Section Iodine-131 TF milk_ind. Section Vd: Vd will be strongly affected by the same sorts of considerations that affect mass loading (ML). Treating the two quantities as though they are independent and not related to other conditions needs to be justified. The minimal and maximal values selected are very high and are not supported by the literature review that is presented. The high values do not seem to be the result of typographic errors (the same values are given in the appendix). An explanation is needed. It is unfortunate that some of the critical values presented in the text are given without units. There is an extensive literature to show that deposition velocity depends also on surface roughness (vegetation, trees, etc.) and humidity. Some description is needed to identify the nature of ground cover in each sector. A single value for Vd is not going to be satisfactory without justification. Particle sizes (AMAD) are presumably in micrometers. Much of the iodine would be deposited by aerosols in the lowest size range considered by Sehmel (1980). The choice in this parameter can easily modify dose values by this pathway by a factor of 10.