ML16314B657

Undated Testimony of Roland Finston Concerning Contentions 4A, 4C, and 4D

ML16314B657
Person / Time
Site:	Diablo Canyon
Issue date:	01/31/2017
From:	Finston R Stanford Univ
To:	Atomic Safety and Licensing Board Panel
References
Download: ML16314B657 (34)

Text

UNXTED STATES OF AMERXCA NUCLEAR REGULATORY COMMXSSION BEFORE THE ATOMIC SAFETY AND LXCENSXNG BOARD.

)

) Docket Nos( 0-275 O.L PACXFIC GAS AND ELECTRIC COMPANY ) 50- O.L.

')

(Diablo Canyon Nuclear Power Plant:, )

Units Nos. 1 and 2) )

TESTXMONY OF ROLAND FINSTON CONCERNING CONTENTIONS 4A, 4C, and 4D

1. I am a healt:h physicist and am employed at: Stanford University, Stanford, California as Acting Director of t:he Healt:h Physics, Safety and Healt:h Office and as a lect:urer in Nuclear Medicine.

2. My professional qualifications include a Bachelor of Science in Physics from the University of Chicago in 1957; a Master of Science in Health Physics from Vanderbilt Universit:y and Oak Ridge National Laboratory in 1959; and a Doctor of Philosophy in Biophysics from Cornell University in 1965. X was an Associate Professor of Radiological Physics at Oregon University in 1965-66, and I have been employed at Stanford 'tate University sinCe 1966 as a health physicist. I have specialized in medical healt:h physics and in this specialty have taught ra-dionuclide dosimetry and have also been responsible for calcu-lat:ing the radiation dose to pat:ients which results from pur-posely administered radiopharmaceuticals. My work in radionuclide

k TESTXiXRK/FINSKN Page 2.

dosimetry has involved analyses of the physical and biological properties of newly recognized radiopharmaceuticals; speci-fically, my research has involved the emissions, biological distribution, retention and resulting radiation dose to humans of such radiopharmaceuticals. For purposes of benefit/risk evaluations, I have also been involved in the evaluation of the somatic and genetic hazards posed by the resulting doses of such radiopharmaceuticals. I have done research on the effects of external radiation on metabolism in bones and used X-ray degradation to study the properties of DNA.

I am a member of the University.'s Human Use Radioisotope Committee which is also approved by the FDA as a Radioactive Drug Research Committee, and have been responsible for dose and risk calculations resulting from administration of radioactive materials in medical research. I have similar responsibilities for external irradiation from X-ray used in medical diagnosis.

3. The purpose of my testimony is to present the Intervenors'osition regarding the inadequacy of the NRC Staff's Final Environmental Statement [ FES ]" as it. has assessed the environmental costs, doses, and effects of low level radiation as to (A) the buildup of concentration of radioisotopes in the food chain; (C) the somatic effects, including incidences of h

human cancers, leukemias, and infant mortalities and genetic

4 t C

TESTlN3iVZ/FIN SIC 0

Page 3.

effects of routine releases on the population within a 50-mile radius of the, plant; and (D) the somatic and genetic effects on plant personnel including inadvertent ingestion of radio-active materials. My evaluation for each contention will be taken in turn.

Contention 4A Buildu of Radioisoto es in the Food Chain.

4. The Appendix I considerations (Letter from Edward Ketchen of the NRC Staff to James Geocaris, dated September 21, 1976, and repeated in the Supplementary Testimony of NRC Staff, Messrs.,Parsont and Boegli; hereafter'"Letter") appear to contain two anomalous values of estimated radioactivity releases which indicate, in one case, that releases may not be as low as readily achievable, and in the other, that calculated doses may be underestimated.

5. Specifically, it is calculated (Table 2 of Letter) that 710 Ci/year/reactor of H are to be released as gases.

This corresponds to 788 Ci/year/reactor if the value of 710 Ci/year/reactor is normalized to a 1000 MW plant operating at 100-. capacity factor. In contrast, data from P61R units of similar size and manufacturer have been performing with much smaller H 3 releases. Reported releases from 6 reactors during the years 1973-74 1/ have been averaging 1/20 as great as that estimated for the DCNGS'units, which estimate was also normalized

'1 TH STIIY/FINSTON Page 4.

for rating and capacity factor (Table 1) . This conclusion was based on an average of the range of H 3 emissions from 6 reac-tors, from 4.35 to 94.2 Ci/year. Only the Zion reactors have approached, but still to less than 50%, the DCNGS estimate (normalized for rating and capacity factor). While this over-estimate of release has resulted in an overestimate of dose, it is of concern to know why this is so. It is unclear, thus, whether the applicants intend to keep the* gaseous releases of 3

H as low as readily achievable as required by 10 CFR Part 50, Section 50.34a. The experience at other similar operating re-actors would seem to suggest, that a lower release of H 3

is pos-sible.

6. The other anomaly concerns the release of fission products, corrosion and activation products to liquid effluents.

Performance data for 8 PNR's operating in the years 1973-74 1/

normalized to rating (1000 MN) and 'capacity factor (100%), show I roleases 5 times greater than the figure estimated for the DCNGS units, also normalized to rating and capacity factor. The five-fold difference (0.38 vs. 1.88 Ci/year) was calculated, (Table 1) by averaging the releases at eight PNR's', a range from 5.61-0.33 Ci/year/reactor, and by comparing it. to the estimate given in Table 2 of the Letter. Xn fact, one PtVR (Indian Point 2) re-leased 30 times more fission products and corrosion and activa-tion prod'ucts to liquid effluents than are estimated for the

TBSTBGNY/EENSION 0

Page 5.

DCNGS. (This unit was deleted from the above analysis on the assumption that thi's represents abnormal operations). Xt would appear, then, that the estimate of 0.34 Ci/year/reactor is un-realistically low in comparison to other reactors whose emissions have been calculated. A realistic release value would in turn result in a-,realistic dose, which is needed to adequately assess the impact of low level emissions that the DCNGS will have on the food chain and consequent population doses.

\

7. Furthermore, my analysis of'performance of PWR's indicates that, with increasing age of PWR plants, environmental releases increase. This conclusion is based on my examination of the emission data for other PWR's reported over several years.

X have requested the emission data from the NRC for 1975 in order to more thoroughly evaluate this trend; conflicting reports as to the availability of that data have been received and have thereby delayed my obtaining it. Another consultant has been

'ontacted by the Xntervenors and will obtain the data prior to the hearing so that an analysis of this observed phenomenon can be presented.'t any rate, neither the FES nor the calculations, pursuant to App.X appear to have taken this phenomenon into account in indicating dose levels over the years, and in this respect, the assumptions for the DCNGS reactors may be unduly optimistic.

TESTING%/FIN STCN Page 6.

Contention 4C Somatic and genetic effects on Population within 50 miles

8. The NRC consultant's analysis of the background radiation and corresponding rate of cancer deaths on the 260,000 person population living within a 50-mile radius of the DCNGS is incorrect. It was assumed 2/ that the background radiation for the 260,000 person population was 30,000 man-rem/year. Data specific to the region (Santa Barbara being the closest area for which figures are available) indicate that the combined in-ternal and external dose of radiation is 82 mrem/year, comprised of 41 cosmic, 2'3 terrestrial, and 18 mrem/year internal radio-activity . The population's corresponding background radiation is 21,000 man-rem/year (82 mrem/year x 260,000 persons = 21,000 man-rem/year). The effect of this lower background radiation figure is to make the relative impact of low level radiation from the DCNGS higher than that originally offered in the Goldman Testimony. Accordingly, the PES should consider data which is more pertinent to the geographical area in which the reactor is located, and take into account the comparative effect of the DCNGS against a lesser background radiation .rate.

9. The National Academy of Sciences (BEIR) report of 1972 actual estimate of the most likely rate of cancer deaths was 152-204 deaths/million man-rem, and not 89, as is stated in the

'L

~

1

Goldman Testimony '.its at p. 4. 4/ The NRC study relied on in the Soldman Testimony only based figure of 89 on the BETR Study; available data should be considered rather than an extrapolation or interpretation of that data. Accordingly, using the figures of 152-204 cancer deaths/million man-rem due to low level emis-sions from the DCNGS, together, with the dose computed for Diablo Canyon effluents (0.3 man-.rem/year, Letter, Enclosure 4) the risk is still'ncreased by a factor of at, least 2 from that reported in the Goldman Testimony. 2/ '0.3 man-rem/year x 152 to 204 can-cer deaths/million man-rem = 5.3 x 10 risk, double the figure o f 2. 7 x 10 given at p. 4 in the Goldman Tes timony 3/ ) . The risk factor of 5.3 x 10 (as properly calculated, using correct BEXR data) is not conservative, as the .Goldman Testimony notes, p."

4, but in fact is recognized as the best estimate currently avaxl-able. Xn fact, the BHXR best. estimate of the number of cancer deaths per million man-rems may be non-conservative by a factor of 2, if the relative risk model applies. That model would pre-dict twice as many cancers to occur from a given dose than does the absolute model /. The efiect of "conservatizing" the BEIR data is to further increase the risk factor of 5.3 x 10

-5 by a

-4 factor of 2, i.e., to 1.1 x 10

10. Moreover, the conclusion that radiation absorbed at.

millirem per day levels. is one-fifth as damaging as when absorbed at. zem per day levels is considered by the EPA to be not a prudent

1 TESmDVZ/rlNSam Page 8.

choice for cent.ral risk estimates 5/ . The basis for the conclusion in the Goldman Testimony that the risk of radiation at low level emission rat(..s is proportionately less than at .higher, i.e. rem per day, levels is the conclusion thai "molecular repair and re-constitution of initial 'lesions'an take place...such that the yield of 'effects'er unit dose is lower than when the dose or rate is high..." (Goldman Testimony at 5). This conclusions is surprising, since the concept of 'reconstitution of radiation lesions in genetic cells at, low doses and dose rates ascribed by Goldman to S. Ab'rahamson is interpreted oppositely by the Environmental Protection Agency 6/ . That report indicates that in mouse oogonia 11 the reduced genetic effects observed are due to cell death rather than to cellular repair mechanisms. In other words, fewer cells indicate genetic mutation at low levels of. radiation, becuase those cells which would contribute to genetic mutation do not survive.

This conclusion in no way indicates that low level emissions are somehow "less damaging" than high levels, an indication which is relied upon in the Goldman Testimony. It is improper, therefore, to reduce a risk factor by five simply because the source is a low-level emitting source. The EPA report would support the con-clusion that the risk factor at low level emission rates is as great as high level (i.e. rem per day) emission rates. Accordingly',

-4 is the factor arrived at using con-the risk factor of 1.1 z 10 servative data. Although the radiation risk may still be negligi-ble (i.e. 444 cancer deaths increased to 444.00011 deaths due to

4 TH STITHY/FINSTOiV Page 9.

operation of the DCNGS), the effect of this increased risk factor, when combined with emissions from other sources (accidents, waste, transportation) should be considered in the FES.

ll. There is, moreover, evidence which indicates that risk t'actor at low levels and low doses may in fact be nicihar than the factor derived linearly from high doses experience 7/' For three different human tumors (thyroid carcinoma, leukemia, and broast cancer), the risk estimates at lo~v'oses or at low dose rateq are either the same as, or in some cases, perhaps higher,

.than risk estimates derived from high doses. This information indicates that the 1.1 x 10 -4 risk factor for DCNGS may be too low and that the cancer deaths may be correspondingly higher than pre-dieted.

12. More troubling is the NRC'onsultant's and Staff's myopic risk evaluation brought about, by standardized calculations of annual doses, without taking regard for the total impact of long-lived emi sions.

Table 2 of the Letter indicates that. the release of Carbon-l,4.for Diablo Canyon will be 8 Ci/year/reactor. The approach taken by 'the NRC Staff to calculate the radiation dose per one year's decay of Carbon-14 is. inadequate, because this figure only calculates the dose from only 0.012% of the radioactive atoms re-leased by the plant. This vastly'underestimates the total radio-logical impact of the long-lived radionuclide., whose average life

I, TEST3 JKSY/I'.INSTON Page 10.

is 8,270 years. The concept of dose commitment has been recog-nized'y an international expert committee as the appropriate means of evaluating the hazards of long-lived nuclides released to the environment 8/ . The NRC also has adopted the concept (10CFR Pt. 50 App. I, Sec. XX-A). For example, the Environmental Pro-tection Agency (EPA) calculates that Carbon-14 is .the major con-tributor to serious health'effects of the nuclear fuel cycle, and may b'e responsible for 88% of the effects attributable to that cycle . Using the EPA estimate, it can be shown that. as the result of nuclear generation of two gigawatt-years of electiic power (one year's operation'f both DCNGS reactors at 90$ capacity) there will be 3 seri'ous health effects perpetrated in the first 100 years of Carbon-14 radioactive decay'. Extrapolation beyond the first century is uncertain; but an upper limit, based on phy-sical 8e'cay alone, would be 250 health effects to the complete 1

decay. Future generations will pay the toll as the result of each year's routine release at DCNGS. The FES has therefore not adequately assessed the health effects attributable to this long-lived radionuclz.de.

Contention 4D -- Somatic and Genetic Effects on Plant Personnel.

13. The FES inadequately evaluates the impact, of low level radiation emissions as to the somatic and genetic effects on personnel working at the plant. Xt is not clear from the Staff's

I TESTIt KNY/FINSTON Page 11.

materials how the figure of 500 man-rem per year per unit for occupational exposures of plant employees was computed. Based upon operating experience during the years 1969-1974, a figure of 1.45 man-rem per Megawatt-year for occupational exposures of PWR plant employees would seem more accurate for the DCNGS re-actors 10/ . (The range is from 0.66 to 2.39 man-rem/megawatt year; the 1975 figure is 0.74 man-rem/Megawatt year) 11/ . The dose to workers at the DCNGS station is therefore computed as follows:

2190 Megawatts X .85 X 1.45 man-rem/

(capacity of site) (DCNGS estimated Megawatt-year capacity factor) 2700 man-rem/year/site.

This is also similar to the annual dose which will be contributed by DCNGS workers to the gene pool dose (adjustment must be made for<

expected child-bearing as a function of age). It is important to compaze the contribution of this dose to the natural background dose and to the other major source of radiation, the genetically significant medical X-ray dose. As stated previously, the back-ground dose for the 260,000 person population living within a 50-mile zadius of the plant is 21,000 man-rem per year. The addition of 2,700 man-rem per year to that, background dose is genetically significant; in fact, it represents more than a twelve percent increase in the dose of radiation to the gene pool. The PES and

) ~

C

TESTIK)NY/FINS TCN Page 12.

e documents reviewed subsequent to the publication of the PES have inadequately assessed the impact of- a twelve percent. increase in the gene pool radiation. Furthermore, the genetically signifi-cant. medical X-ray dose to such a population (260,000 persons) is 5,200 man-rems per year 12/ . An addition of 2,700 man-rem/year on top of this genetically significant. medical X-ray dose to the in-dicated population represents a fifty percent. increase in man-made exposure. The belief that between 33 and 66% of the'edical X-ray-caused gene pool dose may be unnecessary has led to a national effort by FDA and the radiological community to implement gonadal shielding programs l3/ . Hence it follows that an increase equal to 508 of medical X-ray in the DCNGS local population's gene pool dose is a significant environmental effect which the NRC Staff must consider before contending that the PES and amendments are adequate.

14. The dose rate dependence of mutational effects of irradiation is the subject of active scientific debate. The es-timates of rate dependence found. in the BEIR Report. 4/ and relied upon by the Staff in its testimony have been criticized by the Environmental Protection Agency as being perhaps 140 to 220 percent too low /. Using this fact, and the fact that the 500 man-rem/year per unit is low by a factor'of 2.7 (see paragraph 13 above, which computes the occupational exposure at 2,700 man-rem/year per site, which is 2.7 times the 1000 man-rem/year per site claimed), the 20

I ~

TESTZeVZ/FXNSTM Page 13.

genetic "effects" per generation at equilibrium calculated by Goldman {Testimony at p.7) may actually rangee from fro 54 to 173.

That. is, 20 gene zc e ec ts" multiplied by the 2.7-fold increase in occupational radiation results in 54 genetic "effects", which may go as h xg h as 173 genetic "effects" if the EPA analysis of the BEXR data is correct. Properly compared with the s p ontaneous incidence o f 15 000 to 25 000 "effects" in a population of 1/4 million or 260 000 {Goldman Testimony at 7, 60,000 to 100,000'ffects per million x 1/4), the increase in genetic disease is from 0 ~ 2%0 (54 effects/25,000

'I effects) to 1.1-"o- (173 effects/15,000 effects). The.e genetic hffects due to occupational exposure from the plant are certainly signa.ficant enough to merit evaluation in the FES by the Staff.

15. The genetic effect. of low dose radiation on humans 0

li is no longer a matter o f con jecture in ght of rece ntly p ublished observations o f a four-fold increased incidence of severe mental retardation and a statistically signficant increased z.ncidence of

~

Down's Syndrome in populations exposed to natural background ra-diation levels of 1.5 to 3 rems/year 15/ . This exposure rate is similar to that-; experienced by 11% of reactor workers 10/ill/

Therefore, since th 'dence t, e inci en of severe mental retardation may be increased in the children of a population so exposed, the effect, is significant enough to be considere red in the FES.

0 1 TESTED K/FINSTCN Page 14.

16. Finally, the impact of occupational exposure on the health of plant workers should be considered in the FES.

Using BEIR 16/'rom estimates, it. is expected'hat one additional death; delayed effects of radiation cancer will occur for each 6,000 man-rem. Over a twenty-year period of operation, nine workers would be expected to have patent or occult lethal malig-nancies engendered in their bodies as the result of the estimated occupational exposure consequent to the operation of the DCNGS.

Among a work force averaging 1,000 at the site, such a toll should not be overlooked, even if it is not statistically demonstrable above the "background" level of cancer deaths.

A very recent analysis of occupational exposure and ra-diation carcinogenisis makes this a matter of concrete signifi-cance; it may no longer be a matter. of extrapolation 17/ . In a study of Hanford radiation workers, epidemiological methods have revealed an association between radiation exposure and reticulo-endothelial system neoplasms, breast cancer, pancreatic tumors, lI and lung cancers. The dose experience of these workers is no dif-ferent from that experienced by today's nuclear plant employees.

The study further reveals that the amount of radi'ation necessary to double cancer incidence (to i.e., add a number of cancers equal to the spontaneous incidence of cancer) is less, than 10 rads in these four tissues.. This information is significant in light of the fact that the average'nnual exposure to a plant worker is 0.8 rads /. The FES should, therefore, consider the increased danger of cancer death among plant workers.

'0 TABLE 1

~r TABLE OF ANNUAL RADIOACTIVITY RELEASES IN LARGE;"tESTINGHOUSE

'ORMALIZED P')'(R S A I R B 0 R N E Curies/Yr LIQUID, Curies/Yr Date of Cap. Co nver-YEAR Name of Plant Initial HM Factor sion Noble H3 Particu- Halogen tAxed Cri ti cal i ty &AP/ b H

(Ã) Factor Gas lates . F

.197 Di abl o 1060 85 1.11 2953 788 .0023 195 .38 788

~

1.14 1974 Indian Pt.2 9-73 873 43 2.65 14,800 52.7 .757 .763 11.1 127

.0528 1972 H.B.Robinson 2 IC 9-70 700 . 72 1.97 506. 3 4.9 .0048 .048 1.63 '98 c/o 3-71

.5 1973 82 1.74 5390 4.35 .00007 .5 1.05 753

. 097 1974 78 1.83 4220 94.2 .00308 .0942 4. 58 822

.04 5-72 41.3 .00097 .039 .. 0975 437 1973 Surry 1&2 3-73 788 ea 65 .975 844

.24ld/

1974 43 l. 475 10,133 89.1 .0612 .180 5. 61 362

.07

' .07 .033 365 1973 Turkey Pt 3&4 725 ea 62 - 1.11 588 4.6 .0002 -

4.03 1974 62 1.11 5170 10.2 .244 3.83 178 643

.03 6-73 342 .0038 .029 1974 Zion 1&2 1100 ea 24 1.895 5670 .0. 19 12-73 NOTES: TOTAL 47,321 643 6.2 26. 07 4311 a/Data normalized to 1000 HW plant, .69 2.9 479 100% capacity factor. Data incl. AVERAGE 5,258 71. 5 only those years during which otal"Anomalous Figure / 47,321 301.35 2~2 'lb. 97" 4307 plant was in commercial opera- 8 8 8. 8 tion for the full year. vg. ur/o Anomalous Fig. 5,258 37.6 .28 l. GB 538 b/ Fission and Activation Products c/ 1.14 is not the sum of .757+.763; original data in NUREG 0077 does not agree; 1.14 is considered a good estimate.

d/ .241 is probably correct; NUREG 0077 gives .14 for Halogens and Particulates, which is less than the sum of Halogens and Yarticulates reported in another place in 5UREG 0077.

due to abnormal experience.

e/ Unusually large or small values (underlined) are deleted, assuming these are

t TESTXKNY/FINSKM Reference REFERENCES U.S. Nuclear Regulatory Commission, NUREG 0077 and NUREG 75/001.

2/ Testimony of Marvin Goldman for Nuclear Regulatory Commission, undated.

3/ Oakley, D.T., Natural. Radiation Exposure in, the United States.

U.S. Environmental Protection Agency, Office of'adiation Programs (ORP/SID 72-1), 1972, p.36, 37, 50.

National Academy of Sciences/National Research Council.

The Effects on Populations of Exposure to Low Levels of Ionizing Radiation ("BEIR" Report), NAS/NRC, Washington, D.,C.'972, p. 168.

5/ Reactor Safety Study (WAS'400): A Review of the Final Report.

U.S. Environmental Protection Agency, Office of Radiation Programs, June 1976, p.2-5 to 2-8.

Ibid, at p.2-11.

7/ J.M. Brown, "Linearity. vs. Non-Linearity of Dose Response for Radiation Carcinogenesis" Health Physics,'Vol. '31, No. 3, Sept. 1976, p.231.

8/ United Nations Scientific Committee on the Effects of Atomic Radiation, 1962.

9/ U.S.E.P.A., "Draft Environmental Statement for a Proposed Rule-making Action Concerning Environmental Radiation protection Requirements for Normal Operations of Activities in the Uranium Fuel Cycle", p.82 (1975).

10/ NRC,'Occupational Radiation Exposure at Light. Water Cooled Power Reactors, 1969-1974", NURHG-75-032, (June 1975) p.7.

NRC, "Occupational Radiation Exposure at, Light Water Cooled Power Reactors, 1969-1975", NURHG-0109 (August 1976).

12/ U.S. Department of HHW, Food and Drug Administration, Gonad Doses and Genetically Significant Dose from Diagnostic Ra-diology, U.S. 1964 and 1970. Publication FDA 76-8034, April 1976.

13/ Federal Register 41, No. 143 p.30327-9, July 23, 1976.

'4 TESTINX1Y/PIN STCAM%

Reference/Page 2.

14/ Reactor Safety Study (WASH 1400), supra, cote 5 ar 2-11.

15/ Kochupillai, N.; Verma, X.C.; Grewal, M.S.; and Ab-Ramalingsaswami, V., "Down's Syndrome and Related normalities in an Area of High Background Radiation in Coastal Kerala", Nature 262, 60 (1976) 16/ BEIR Report, supra, note 4 at p. 171.

Mancuso, T.F., Stewart, A., Kneale, M.A., "Radiation Exposures of Hanford Workers Dying from Various Causes".

paper presented at The Tenth Midyear Symposium of the Health Physics Society, Saratoga Springs, New York (Oct.

11-13, 1976) .