ML19256F024
| ML19256F024 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 05/06/1979 |
| From: | Battist L, James Buchanan, Congel F Office of Nuclear Reactor Regulation, NRC OFFICE OF STANDARDS DEVELOPMENT |
| To: | FOOD & DRUG ADMINISTRATION |
| Shared Package | |
| ML19210C879 | List: |
| References | |
| NUDOCS 7911200300 | |
| Download: ML19256F024 (86) | |
Text
.
y, a 7m May 6, 1979
., a j, :n 1s' h[.1d MEMORANDUM FOR:
Director, Bureau of Radiological Health, FDA, HEW Acting Deputy Assistant Administrator for Radiation Programs, EPA Director, Office of Standards Development, NRC Director, Office of Inspection and Enforcement, NRC Director, Office of Nuclear Reactor Regulation, NRC FROM:
Members of the Ad Hoc Population Dose Assessment Group Lewis Battist, NRC John Buchanan, NRC Frank Congel, NRC Christopher Nelson, EPA Mark Nelson, CDC, HEW Harold Peterson, NRC Marvin Rosenstein, FDA, HEW SU8 JECT:
TRANSMITTAL OF REPORT The Ad Hoc Population Dose Assessment Group has prepared the attached updated report on " Population Dose and Health Impact of the Accident at the Three Mile Island Nuclear Station."
The report covers the principal population exposure period from March 28 through April 7, 1979.
Additional data that were not available for the April 15, 1979 preliminary report have been included and the analysis of that data has been more systematic.
The Ad Hoc Group has also considered comments on the April 15 report provided by a number of individuals from the three constituent Agencies.
The report is submitted for use as deemed appropriate by each of the three Agencies and supercedes the April 15, 1979 report.
fjm!].C~
D aj e
79j1e00300 1377 060
N-DRAFT POPULATION DOSE and HEALTH IMPACT OF THE ACCIDENT AT THE THREE MILE ISLAND NUCLEAR STATION (for the period March 28 through April 7,1979)
Ad Hoc Population Dose Assessment Grouc Lewis Battist Nuclear Regulatory Commission John Buchanan Nuclear Regulatory Commission Frank Congel Nuclear Regulatory Commission Christopher Nelson Environmental Protection Agency Mark Nelson Center for Disease Control Department of Healtn, Education, and Welfare Harold Peterson Nuclear Regulatory Commission Marvin Rosenstein Food and Drug Administration Department of Health, Education, and Welfare May 6, 1979 1377 061 4
P TABLE OF CONTENTS i.
Preface ii.
Summary and Discussion of Findings 1.
Introduction 2.
Nature of the Radioactivity Released 3.
Dose Assessment from External Exposure A.
Thermoluminescent Dosimeter Data B.
Offsite Population Collective Dose Estimate C.
Offsite Maximum Individual Dose 4.
Potential Health Impact of External Excosure A.
Health Effects from Low-Level Radiation B.
Comparison of Individual Doses with Natural Background C.
Existing Cancer Rates and Risks D.
Summary of Health Effects E.
Dose Rate Effects 5.
Other Sources of Exposures A.
Skin Doses and Health Risks from Beta and Gamma Radiation B.
Inhalation Lung Dose C.
Airborne Radioiodine Concentrations and Doses D.
Ingestion of Iodine-131 in Milk 1377 062
TABLE 0.
CONTENTS (Continued)
Appendix A - Department of Energy Estimate of External Whole Body Gamma Radia-tion Exposure to the Population Around the Three Mile Island Nuclear Station Appendix B - Department of Energy Environmental Deposition Measurements in the Area Surrounding the Three Mile Island Nuclear Power Station Appendix C - Evaluation of Skin Oose Factors Appendix 0 - Estimated Risk of Specific Radiation Induced Cancers Based on the UNSCEAR 1977 Report 1377 063
)
4
4
~
PREFACE This report was prepared by representatives of the Nuclear Regulatory Commission (NRC), the Department of Health, Education and Welfare (HEW) and the Environmental Protection Agency (EPA), who constitute an Ad Hoc Population Dose Assessment Group.
It is an assessment of the health impact on the approxi-mately 2 million offsite re:idents within 50 miles of the Three Mile Island Nuclear Station derived from the dose received by the entire population (collective dose).
The Ad Hoc Group has exacined in detail the available data for the period up to and including April 7, 1979.
Based on a cursory review of data from periods beyond April 7, it appears that the population dose will not be significantly increased due to extention to the period past April 7.
The dose and health effects estimates are based primarily on thermo-luminescent dosimeters placed at specific onsite and offsite locations.
These dosimeters measure the cumulative radiation exposure that occurred at these locations.
They permit the most direct evaluation of dose to the offsite popu-lation from radionuclides released to the environment.
The report also addresses several areas of concern about the types of radionuclides released, about the contribution to population dose from beta radiation emitted from the released radionuclides, about the degree of coverage afforded by available radiation measurements, and about the range of health effects that may result from the estimated population dose.
1377 064
C Based on the current assessment, the Ad Hoc Group concludes that the offsite population dose associated with radioactivity released during the period of March 28 to April 7,1979 represents minimal risks (that is a very small number) of additional health effects to the offsite population.
The numerical statement of this conclusion is developed in the report.
The Ad Hoc Group is not aware of any radiation measurements made during this period that would alter this basic conclusion, although refinement of the numerical estimates can be expected as the data are updated and verified.
The members of the Ad Hoc Group concur that the manner in which the population dose estimates were generated from the exposure measurements was conservative and that the uncertainties in the population dose estimates and health effects are not large enough to alter its basic conclusion.
ACKNOWLEDGMENTS The Ad Hoc Group acknowledges the assistance of Ted Schoenberg of the Department of Energy and Andy Hull of Brookhaven National Laboratory in providing the data and analysis presented in Appendices A and 8.
We also acknowledge the contributions of the following individuals:
Nuclear Regulatory Commission Jeannette Kiminas Walter Pasciak Edward Branagan James Fairobent William Snell Food and Drug Acministration Charles Coyle Dean Elbert Richard Kisielewski Environmental Protection Agency Philip Cuny 1377 065
POPULATION DOSE ANC HEALTH IMPACT OF THE ACCIDENT AT THE THREE MILE ISLAND NUCLEAR STATION (for the period March 28 through April 7,1979)
Summary and Discussion of Findings An interagency team from the Nuclear Regulatory Commission (NRC), the Department of Health, Education and Welfare (HEW) and the Environmental Protec-tion Agency (EPA) has estimated the collective radiation dose received by the approximately 2 million people residing within 50 miles of the Three Mile Island Nuclear Station resulting from the accident of March 28, 1979.
The estimates are for the period from March 28 through April 7, 1979, during which releases occurred that resulted in increased exposure to the offsite population.
The principal dose estimate is based upon ground level radiation measurements from integrating thermoluminescent dosimeters located within 15 miles of the site.
These estimates assume that all the exposure recorded by the dosimeters was from gamma radiation.
The data were obtained from dosimeters placed by Metropolitan Edison Power Company prior to the start of the incident through April 6, 1979, as part of their normal environmental surveillance program, and from dosimeters placed by NRC from noon of March 31 through the afternoon of April 7, 1979.
Both measurement programs are continuing.
The results for the period beyond April 7, 1979 have not been fully examiiec.
An additional 1377 066
4 2
dose estimate developed by the Department of Energy using aerial monitoring that commenced about 4 p.m. on March 28, 1979 is also included.
A variety of other data helpful in assessing relatively minor components of population dose was also reviewed.
The collective dose to the total population within a 50-mile radius of the plant has been estimated to be 3300 person-rem.
This is the mean of four separate estimates that range between 1600 to 5300 person-rem.
The range of the collective dose values is due to different methods of extrapolating from the limited number of dosimeter measurements.
An estimate provided by the Department of Energy (2000 person-rem) also falls within this range.
The projected number of excess fatal cancers due to the accident th.'t could occur over the remaining lifetime of the population within 50 miles is approximately one.
The number of fatal cancers that would be normally expected in a population of this size over its remaining lifetime had the accident not occurred is estimated to be 325,000.
The projected total number of excess health effects, including all cases of cancer (fatal and non-fatal) and genetic ill health to all future generations, is approximately two.
These health effects estimates were derived from intermediate risk estimates within the ranges presented in the 1972 report of the Advisory Committee on the Biological Effects of Ionizing Radiation (BEIR) of the National Academy 1377 067
3 of Sciences.
Preliminary information on the recently updated version of this report indicates that these estimates will not be significantly changed.
It should be noted that there exists a smaller minority of the scientific community that believes the risk factors may be as much as two to ten timec greater than the estimates of the 1972 BEIR report.
There also exists a larger minority of the scientific community that believes that the estimates in the 1972 BEIR report are two to ten times larger than they should be for low doses and low-LET radiation.
The maximum dose to an individual that is based directly on dosimeter measurements is less than 100 mrem.
The estimate represents the cumulative dose recorded by an offsite dosimeter at 0.5 mile east northeast of the site (83 mrem) and is for an individual remaining outcoors at that location for the entire period from March 28 through April 7.
A second estimate (37 mrem) was obtained for an individual who was located on Hill Island 1.1 miles north northwest of the site for a total of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> during the period of March 28 through March 29.
The additional risk (over and above the normal risk) to individuals receiving doses of this magnitude is very small.
The average probability of a fatal cancer from all causes is about 1 in 6; the increase in probab lity of fatal cancer from these doses is about 0.00002 (2 x 10-5), yielding a new probability of 1.00002 in 6.
1377 068
4 4
A number of questions concerning this analysis are posed and briefly answered below.
More detailed discussions are included in the body of the report.
What radionuclides were in the environment?
The predominant radionucliae measured in the environment was xenon-133, together with small amounts of xenon-135, xenon-133m, and iodine-131.
This is based on analyses of airborne radioactivity in the containment structure and waste gas tanks and the analysis of samples collected by aircraft flying in the plumes.
What were the highest radiation exoccures measured outside the plant buildings?
Some of the Metropolitan Edison dosimeters located on or near the Three Mile Island Nuclear Station site during the first days of the accident recorded cumulative doses as high as 1020 mrem.
These recorded exposure readings do not apply directly to individuals located offsite.
However, onsite dosimeter readings were included in the procedure for projecting doses to the offsite population.
This procedure is described in tne report.
What is meant by maximum dose to an individual?
The maximum dose to an individual applies to a real or hypothetical person who remained out-of-doors in the specified offsite area for the specified period of time.
It applies only to that individual or others in the same vicinity.
1377 069
5 What is T.eant by collective dose (person-rem)?
The collective dose is a measure of the total radiation dose which was received by the entire population in the 50-mile radius of the Three liile Island site.
It is obtained by multiplying the number of people in a given sector of the area by the dose estimated for that sector and adding up all of the sector totals.
Were the radiation measurements adecuate to determine pooulation health effects?
The extensive environmental monitoring and food sampling were adequate to characterize the nature of the radionuclides released and the levels of radioactivity in those media.
The measurements performed by Department of Energy (aerial survey) and Metropolitan Edison and Nuclear Regulatory Commis-sion (ground level dosimeters) are sufficient to characterize the size of the collective dose and therefore the long-term health effects, when taken as a whole.
However, a single precise value for the collective dose cannot be assigned because of the limited number of fixed ground level dosimeters deployed during the accident.
How conservative were the collective ccse e:ti. nates?
In projecting the collective dose from the thermoluminescent dosimeter exposures, several simplifying assumptions were made that ignored factors that 1377 070
6 are known to reduce exposure.
In each case, these assumptions introduced signif-icant overestimates of actual doses to the population.
This was done to insure that the estimates erred on the high side.
The three main factors that fall into this category are:
(1) No reduction was made to account for shielding by buildings when people remained indoors.
(2) No reduction was made to account for the population known to have relocated from areas close to the nuclear power plant site as recommended by the Governor of Pennsylvania, or who otherwise lef t the area.
(3) No allowance was made to account for the decrease in the dose to the internal body organs that resulted from reduction in the gamma radiation as it passed through the body.
The exposure reading of the dosimeter was assumed to be the dose to all body organs.
What is the contribution of beta radiation to the total dose?
Beta radiation contributes to radiation dose by two modes, inhalation and skin absorption.
The total beta plus gamma radiation dose to the skin from xenon-133 is estimated to be about 4 times the total body dose from gamma radiation alone.
This additional skin dose could result in small increases 1377 071
7 in the total health effects (about 0.2 health effects) due only to skin cancer.
The increase in total fatal cancers over that estimated for external exposure from gamma radiation alone would be about 0.01 fatal cancers.
This contribu-tion would be considerably decreased by clothing or by being indoors.
The dose to the lungs from inhalation of xenon-133 for both beta and gamma radiation is estimated to be approximately 6 percent of that for total body dose from gamma radiation alone.
In order to inhale radioactive xenon a person would have to be immersed in the plume.
What radionuclides were found in milk and food and what are their significance?
Iodine-131 was detected in milk samples during the period March 31 through April 4.
The maximum level seen in milk (42 pCi/ liter) was 300 times lower than the level at which a recommendation would be made to remove cows from contaminated pasture.
Cesium-137 was also detected in milk, but at concentra-tions expected from residual fallout from previous atmospheric weapons testing.
No increase in radioactivity was found in any other food samples.
Why have the estimates of radiation dose changed?
The original estimate of collective dose (1800 person-rem) presented on April 4 at the hearings before the Senate Subcommittee on Health and Scientific Research covered the period from March 28 through April 2.
The data used for this 2 stimate was obtained from preliminary results for Metropolitan Edision offsite dosimeters for the period March 28 through March 31 and preliminary 1377 072
8 results for NRC dosimeters for April 1 and 2.
On April 10, the estimate of 2500 person-rem presented to the Senate Subcommittee on Nuclear Regulation by NRC Chairman Hendrie included the time period from March 28 through April 7.
The data base for this estimate included additional NRC dosimetry results for April 3 through 7.
The Ad Hoc Committee's preliminary report of April 15 stated a value of 3500 person rem for the time period from March 28 through April 7.
This value resulted from better information on the dosimeter results and an improved procedure for analysis the results.
The current report states an average value of 3300 person-rem (range of 1600 to 5300 person-rem) for the time period from March 28 t'.ough April 7.
Additional dosimeter data were available and better methods were used to deter-mine the collective dose.
Also, the onsite dosimeter measurements are included in a systematic fashion.
The original estimate of maximum dose to an individual presented on April 4 has increased from 80 to 83 mrem as a consequence of adding the contribution from April 2 to April 7.
New information on dosimeter readings on or very near the site was received after the initial analysis.
It was also learned that an individual was present on one of the nearby islands (Hill Island) for a total of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> during the period March 28 to March 29.
The best estimate of the dose which may have been received by the individual is 37 mrem.
The text describe a range of dose estimates for the individual.
1377 OL73
9 9
Will these estimates of dose change again?
The dose and health effects estimates contained in this report are based on the dosimeter results for the period March 28 to April 7, 1979.
There still remain some minor questions concerning interpretation of the dosimeter results.
For example, the best valuas for subtracting natural background contributions from the Nuclear Regulatory Commission dosimeters have not been determined.
Additional dosimeters placed by a second contractor (Radiation Management Corporation) for Metropolitan Edison and Nuclear Regulatory Commission dosimeters that remained at the same 37 sites for the entire period, have not been fully analyzed.
The actual contribution to collective dose from the period after April 7, if any, has not been fully assessed.
Therefore, the numerical dose values may be subject to some modification.
The Ad Hoc Group feels that these factors would represent only minor corrections to the present estimates.
In any case, none of the above refine-ments should cause an increase in any of the current estimates that would alter the basic conclusion regarding the health impact due to the Three Mile Island accident.
1377 074
4 10 1.
INTRODUCTION The Ad Hoc Population Dose Assessment Group was formed from individuals assigned by their respective agencies to the NRC Incident Response Center on Monday, April 2, 1979.
The Ad Hoc Group's objective was to obtain an estimate of the public health consequences of this accident to the offsite population and submit the results to each of the constitutent agencies for their use.
Because of the urgency to prepare estimates of the health impact for presentation at the April 4, 1979 Senate hearings, the group was forced to rely upon very early data that were available at the NRC Incident Response Center or easily obtained through existing communication channels with the Federal coordination center adjacent to the Three Mile Island site.
An interim report was prepared on April 15, 1979, which extended the estimate through April 7, 1979.
The current report is an update of the analysis.
The Ad Hoc Group has also had a chance to review its earlier calculations and analyze the data in a more systematic fashion.
1377 075
11
~
2.
NATURE OF THE RADIOACTIVITY RELEASED The primary radioactive materials released to the environment appear to be xenon-133 (half-life 5.3 days) and xenon-135 (half-life 9.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />) and traces of radioactive iodine, primari'y iodine-131.
This is substantiated by considera-tion of the known course of events, knowledge that the effluents were released through particulate and iodine filters, and from subsequent environmental measure-ments in the diffusing radioactive plume (see Appendix B).
Particulate radio-nuclides such as strontium-90, uranium isotopes, and plutonium would either have been retained in the fuel or if released from the fuel would rem 2in in the coolant water.
These elements have not, to our knowledge, been detected in the environment in the vicinity of Three Mile Island (TMI) nor in the reactor containment or gas decay tanks.
Based on the physical and chemical neture of these radionuclides they would not be released from the plant under the conditions of the TMI accident.
Some of the radioactive krypton isotopes such as krypton-87, krypton-85m and krypton-88 may have been released with the radioactive xenons.
However, these are all relatively short-lived radionuclides and none of the reported gamma-ray spectral analyses detected any measurable quantities of these krypton isotopes.
Appendix 8 describes the environmental surveillance activities of the Department of Energy which measured the radionuclides in the environment from the release.
1377 076
12 3.
DOSE ASSESSMENT FROM EXTERNAL EXP05UPE A.
Thermoluminescent Dosimeter Data The available thermoluminescent dosimeter (TLD) data were used for this evaluation for three reasons:
1.
The TLD's placed by the licensee as part of the environmental radiation surveillance program for routine operation were the only devices for measuring radiation exposure that wera placed at fixed locations throughout the course of the accident, particularly during the first 3 days.
2.
The TLD's are dose-integrating devices and measure total exposure rather than peak erposure rates, which may be transient (of short duration).
3.
The TLD's used can measure exposures of about 1 mR.
(See the folicw-ing description of the TLD's.)
At the time of the accident, Metropolitan Edison had environmental TLD's in place at a total of 20 locations both onsite and offsite.
These locations are described in Table 3-1.
The site locations are given in Figures 3-1 through 3-3.
Except for three locations (1081, 1451, and 16A1), these TLD's had been exposed since December 27, 1978, to measure the environmental radiation exposure 1377 077
13 d Jring the first quarter of 1979.
The three locations, 10B1, 14S1, and 16A1, on islands in the Susquehanna River, had been exposed since September 27, 1978.
All 20 of the Metropolitan Edison locations had environmental TLD's manu-factured and read by Teledyne Isotopes.
These Teledyne Isotopes environmental dosimeters are rectangular Teflon wafers impregnated with 25% CaSO :Dy phosphor 4
contained in black polyethylene pouches in rectangular holders with copper filters to make the energy response more uniform (" flatten" the energy response).
After exposure in the environment, four separate areas of the dosimeter are read in a Teledyne Isotopes Model 8300 TLD reader.
The average of these four readings is used in the calculations.
In the product bulletins, these dosimeters are said to have a " minimum sensitivity" of 0.5 mR and to have a " maximum error (1 standard deviation)" of " 0.2 mR or 23%, whichever is greater" for measurement of exposure from cobalt-60 gamma radiation.
At 10 of their 20 locations, Metropolitan Edison had duplicate dosimeters which were supplied and read by Radiation Management Corporation (RMC) as quality control checks.
These 10 locations are indicated in Table 3-1.
The suffix "Q" added to the station code indicates data from RMC TLD's at the Metropolitan Edison locations.
Two RMC model UD-2005 dosimeters were used at each location.
Each dosimeter contains two CaSO :Tm TLD phosphor elements inside a plastic 4
and metal shield to flatten the energy response.
Thus, four readings (two dosimeters; two readings per dosimeter) of the exposure of the RMC dosimeters 1377 078
9 14 are obtained for each location.
The " sensitivity" of these dosimeters is said to be about O'5 mR.
On March 31, NRC placed TLD's at 37 locations and on April 5, an addi-tional 10 dosimeters were placed at various schools.
The locations of these dosimeters are described in Table 3-2.
The site locations are shown in Figures 3-4 and 3-5.
These dosimeters were also supplied and read by Radiation Management Corporation (RMC) but are different from the TLD's supplied by RMC to Metropolitan Edison.
The RMC dosimeters used by NRC are either the RMC model UD801 dosimeter or the model UD804 environmental dosimeter.
Each of the UD801 dosimeters contains two lib 0 :Cu,Ag phosphor elements and two Co0 :Tm phosphor elements.
One 47 4
Li B 0 element has an open window (to minimize attenuation of beta radiation) 247 2
and the other a 280 mg/cm filter; one of the CaSO elements also has a 280 4
2 mg/cm filter, while the second has a 700 mg/cm filter of lead to flatten the energy response.
The UD801 dosimeters are said to have a " sensitivity, whole body" of "1 mR - 2000 R."
Each of the UD804 environmental dosimeters contains three CaSO :Tm phosphor elements, with a lead filter to flatten the 4
energy response; thus, three readings are obtained from each of these dosimeters.
These UD804 environmental dosimeters are said by RMC to have a " sensitivity" of "1 mR-200 R (30 kev-10 MeV)."
Starting on April 1,1979, at each NRC site, two dosimeters were changed daily; thus, either 6 or 8 dosimeter readings were obtained for each location each day, depending on which type of dosimeter was used.
In addition, beginning on April 1, 1979, two dosimeters were left at each location for longer exposures than the period considered in this report.
1377 079
15 Exposures measured at Metropolitan Edison TLD stations (including both Teledyne Isotopes and RMC data) are listed in Table 3-3.
These Metropolitan Edison data include exposures frcs the time of the accident on March 28, 1979, to April 6, 1979.
Exposures measured at NRC stations are listed in Table 3-4 for the time periods from March 31, 1979 (when the NRC dosimeters were first placed at these locations), until April 7,1979.
Each entry in Tables 3-3 and 3-4 is an average of multiple readings of the exposure at that location for that time period together with the standard deviation of the multiple readings.
Exposures measured at Metropolitan Edison locations during 1978 are listed in Table 3-5 (Teledyne Isotopes data) and Table 3-6 (RMC data).
These data provide an estimate of the background exposures.
All exposure (mR) measurements are based on calibrations with cesium-137 sources.
Samples of each type of TLD placed by the various organizations around the TMI site have been collected and exposed to known sources of xenon-133 at the National Bureau of Standards.
Preliminary results indicate that the energy response of the Metropolitan Edison and NRC TLD's to the gamma radiation from a xenon-133 source varies from about 20% less than the calibration value to about 30% greater than the calibration value.
These experiments were not designed nor should they be interpreted as being a precise calibration of the TLD's under actual field conditions or to 1377 080
16 the exact spectrum of radiation that was emitted during the course of the accident.
For these reasons, these correction factors were not applied to the dosimeter readings.
However, the external exposure calibration does confirm that the dosimeters are sufficiently sensitive to the xenon-133 radiation that their response at low energies would not introduce a major uncertainty in the dose or health impact estimates.
1377 081 A
17 Table 3-1.
METROPOLITAN EDISON TLD STATION LOCATIONS STATION LOCATION DESCRIPTION
- CODE 152**
0.4 miles N of site at N Weather Station 1C1 2.6 miles N of site at Middletown Substation 2S2 0.7 miles NNE of site on light pole in middle of North Bridge 452**
0.3 miles ENE of site on top of dike, East Fence 4A1 0.5 miles ENE of site on Laurel Rd., Met. Ed. pole #668-0L 4Gl**
10 miles ENE of site at Lawn - Met. Ed. Pole #J1813 5S2**
0.2 miles E of site on top of dike, East Fence 5Al**
0.4 miles E of site on north side of Observation Center Building 7Fl**
9 miles SE of site at Drager Farm off Engle's Tollgate Road 7G1 15 miles SE of site at Columbia Water Treatment Plant 6Cl**
2.3 miles SSE of site 9S2 0.4 miles S of site at South Beach of Three Mile Island 9G1 13 miles S of site in Met. Ed. York Load Dispatch Station 1081 1.1 miles SSW of site on south beach of Shelley Island 1151**
0.1 miles SW of site on dike west of Mechanical Draft Towers 1281 1.6 miles WSW of site adjacent to Fishing Creek 14S1 0.4 miles WNW of site at Shelley Island picnic area ISGl**
15 miles NW of site at Wes'. Fairview Substation 16S1**
0.2 miles NNW of site at gate in fence on west side of Three Mile Island 16Al 0.4 miles NNW of site on Kohr Island nAll distances measured from a point midway between the Reactor Buildings of Units One and Two.
All 20 stations had Teledyne-Isotopes Environmental TLD's.
naStations with RMC TLD's.
Data obtained with RMC TLD's at these locations are designated by adding the letter "Q" as a suffix to the station code.
1377 082
wi
'#J 4.
~ ~ _
\\
S wp
\\
, O.,.
>s.-,~
p,y a!., '
- s.
[
- l. :n *..
g
.+
.. ~~-
L*j f
S ' **
, m j ' ' '.
5 y *,
=C2Q h
. ?J;"x','.' ": _
s L ; g
(\\, ' ' '-*
.-_,4 s,
,j
, t.g&
$p
..~
'~s ;:,y 9
2 -. -(
.,c,'~
.i 3 ;
e
)
7, _,
k..: fl '.7-c ~!:%Q~>~.9 +...
e yL*.
' ~-..
.. A. * ;;_ ? M, f 6\\
%,q.
6,.,--- ? ~ 5 =m n d ~ N ss
~:..
.. 2 :-4
- .;$~5. [*. 'j.$i. ?~~
- 6l,-
's 0
~
..--;;. ; ~ -..g ~;.. J.
e b
u
% y. :w.*;;> ~3; q.h.' '
Oi g u {*f.~*;;"]' W
"f
- g E$h? "
?A t,2::'",QL;&
i V
l
\\'.
t
}i
, :,p. ?.;-
t' f.
.i
....s~
e r-Y g
w tr -%~
.6 l-yA
?.
,f ~T5..';5&.
.)
~
n M
~
.' l O' :=
,b, m
s : r
%$i45 i
E ta g.
-gl
/f.7:. 8C!1 f
//;m 1
, :w.% $".. ;~:;- '
7; g llh f,Sh
- U'h-if "ll
,y g
?-
1llll h
9V i+ M : E.:?(y t
~:
54f.1* N
,r,og
,(
')
J.,
f:'
hk..
)
2-S P.
w i
q
~ ~~,
y p
- l' 6'l i *$ $
bo, Wm
- q ~w&
yo f
[
y Mp-6 8
Nl l
W) e nmh&
O I
h
=
- -p ro I ':s f
f m 2w
- /.2' NN l
k k C r==
~
U E.F 3/
"E$g j
d
=
rg c
mae n
%s M
.$Ym5 1. b*
- l1 j
=. e u h.
r qw
~s e
u 4
U1 **
Q
%.m e.
s 74 Q7 g
0m33
- U
[
psn eg u.:.= _=
g E Eh
- ff
[f
/h
,6 C55" y
s-
.,)
'y
. = = '
4>Si
[
/
g e V.
s;u
\\
g $\\ g%n,;
e
,r c
- /
J,v,-
d., 4 5W. '. h.g sn; 4-sl I
a(J{j_.]L. jo g/
}4
~
ss s-w
~
.[ 4
\\u. m/
75Ej l
a a
y
/ag
, c i \\
/
\\ig A=
.m-;
,. 3=-q;n) 2esre s
\\
a ;;
s-
, n7 I $ /.,
\\J 0.),m;a9,5m%,y
. b. i n =:
~
s a-
.I 4y
=6B
.t t i i
- AWitRnM
!7
19 book ekO30kkm
=
m m
- w m
m 2
\\
el m
2 m.
2 g
R$
2 s
=
ea a
l 2
s e
)
81
\\
b2 d
ca =
J
\\
l'
- p. C
%h>
\\
/
(
I 2
- i e
P
/
/
w o
3 5
3 2
5 5-3 s
S
=s5 a
8
- -us
=se8 8.s m.iE e
=~
l 5$.33d!!
\\
- se
=822 -
, / 2eLv*
5 z us=
8E"bh 2
3 3,
awese 3
a Y
m e=
f 1377 084
O' tJORTil HilW title O
'~
f
- 5'e)
, h)
HW p,
.s HE 6
9
+
x W
s c-s
\\
b x
's
\\.
[M
/t
))
\\_
M D
-s WHW f'
N ENE h
\\
' Li. BAH 0tl s
filARRISBURG R[ADIF, 15G i
N
/
6'IGl
/
W EARt.lSL
'{
E g'10
)
S
)
X
-r
- LANCAb b y
Y RK
/
\\
gig 30
\\
s.
s WSW \\
40 ESE
/
Figure 3-3.
location of Metropolitan
/
50 Edison Dosimetry Sites
/
Outside of a Five-Mile Radius for the period
,.e \\
March 28 through April 6, 1979.
SW x
' SE V
m NJ e
C 5 5 *."
SSE CD LT1 5
21 Table 3-2.
NRC TLD LOCATIONS STATION DISTANCE DIRECTION SECTOR DESCRIPTION N-la 2.4 mi 356*
N School (added 4/5/79)
N-1 2.6 mi 358 N
Middletown N-1c 3.0 mi 0*
N School (added 4/5/79)
N-le 3.5 mi 349 N
N-1f
- 4. 0 mi 351 N
N-2 5.1 mi 360*
N Clifton N-3 7.4 mi 6
N Humm11stown N-4 9.3 mi 360*
N Union Deposit N-5 12.6 mi 3*
N NE-1
.8 mi 25*
NNE North Gate NE-2
- 1. 8 mi 19 NNE Geyers Ch NE-3 3.1 mi 17 NNE Township School NE-3a 3.6 mi 44*
NE School (added 4/5/79)
NE-4 6.7 mi 47*
NE E-1
.5 mi 61*
ENE 1200' N of E-la E-5 (E-la) 0.4 mi 90 E
Residence E-3 3.9 mi 94 E
Newville E-4 7.0 mi 94 E
Elizabethtown E-2 2.7 mi 110 ESE Unpopulated area SE-4 4.6 mi 137 SE Highway 441 SE-4a 5.0 mi 146 SE School (added 4/5/79)
SE-5 7.0 mi 135*
SSE Falmouth 5-1 3.2 mi 169*
S York Haven S-la 3.35 mi 173*
S School (added 4/5/79)
S-2 5.3 mi 178*
S Conewago Hts S-3 9.0 mi 181 S
Emigsville S-4 12.0 mi 184*
S Woodland View SW-1 2.2 mi 2c0 SSW Bashore Island SW-2 2.6 mi 203*
SSW Pleasant Grove SW-3 8.3 mi 225 SW Zions View SW-4 10.4 mi 225*
SW Eastmont W-2 1.3 mi 252*
WSW Goldsboro W-3a 4.4 mi 247 WSW School (added 4/5/79)
W-1 1.3 mi 263*
W Goldsboro W-3 2.9 mi 270*
W Unnamed comcanity W-4 S.9 mi 272*
W Lewisberry W-5 7.4 mi 262 W
Lewisberry NW-1 2.6 mi 303 WNW Harrisburg Airport NW-3 7.4 mi 297 WNW New Cumberland NW-2 5.9 mi 310 NW Highspire NW-4 9.6 mi 306 NW Harrisburg NW-5 13.8 mi 312 NW Harrisburg N-lb 2.75 mi 346 NNW School (added 4/5/79)
N-1d 3.5 mi 333*
NNW 13/7 086
y D9 3
n Jo CN 0
b pgb c
E g
M6 i
0 E
B E
S H
E G
E n*N S
i f '
S g
r E
E I
S N
S N
N N
W O
T E
L 2-
- (
N L
D O
T
)
t E5 n
h Y
xs.
y g
6 0N
,a k
0 g
Fi y
V t
r
~
l
- o13, l -N n
y-O i
h 0
yt 1
W
/
rid oWo y
(
H B
t i
asr W
l S
J w
D l ee I
S utP7 S
L gi O
eS rl d
R oi yf r 3
rr p
~#
at sA eeu l mih cid g W
usau w
oRo lf N
D r
f eh g e
onl t oi e
h nim 1 t
w os - 3 i) i se s5 ti vh W
imic a
ll ci Fr oi N
oo a
e or W
W LCaM t
hp cA i
S W
W S
S r
4 D
De L
Lt 3
T Tf A e
C C(
r R
R l
i u
i t
g
)
o i
f
- c. w '- Q N
O' tJ O RTll I
MNE HNW,.. -
C'
')
q, M
/
NW s', NE p
j, -
J'
'y y
\\
a
\\
bd C
s
\\
s N
y N
\\
p n)
'x
\\,
EME WNW
/
/
l}
k s
~
\\ [ADlHG
+5,
'(EBAH0H R
i lllARillSBURG
\\
s;
/
/
]%
[,gg EARLlSL L,3 N g
W 1
u,a m
/
d\\
Jm
'O
/
~~'
' L ANC AS
/
Y
~ i' gg N
Y RK
/
s 30 s
40 ESE WSW \\
\\
/
Figure 3-5.
Location of Nuclear Regulatory 50 Conmission Dosiluetry Sites Out.
. /
/
side of a five-Mile Radius for
'N
__ /
the Period March 31 through u
April 7.
/
\\
J
~
-~
y 3
/
~
ca CD
,/
24 Table 3-3.
METROPOLITAN EDISON TLD DATA - RADIATION EXPOSURES FOR PERIOOS ENDING 04/06/79 Station (1)
Exposure Period 12/27/78 03/29/79 03/31/79 04/03/79
-03/29/79
-03/31/79
-04/03/79
-04/06/79 mR std. deviation per exposure period (includes background) 1C1 20.1 1.3 3.2 0.7 1.4 0.4 0.5 0.1 7F1 24.1 1.8 1.110.1 0.5!0.5 0.9 0.1 7F1Q 23.310.5 0.810.2 1.5 0.2 0.9 0.0 15G1 18.4 2.0 1.910.3
-0.7 0.1 0.5 0.0 15GlQ 17.6 0.6 1.1 0.1 0.8t0.1 0.7 0.2 1281 16.3 0.9 9.411.6 0.2 0.3 1.2 0.2 9G1 21.3 1.4 1.4 0.1(3) 0.10.2 0.6 0.1 SA1 18.6 1.0 8.3 2.8 7.7 2.5 3.0 1.2 5A1Q 16.1 1.3 5.411.0 5.2:0.9 2.0 0.6 4A1 20.2 1.3 34.318.6 41.4 8.5 2.2 3.4 252 43.7 4.4 32.5 5.6 3.4 0.6 0.910.2 152 97.9 1.9 20.013.4
-0.1 0.1 0.6t0.1 152Q 95.7 5.0 15.313.2 1.3 0.1 0.8 0.1 16S1 1044.21128.2 83.7 17.5 7.0t0.7 1.5 0.3 16S1Q 929.4190.5 61.6 12.2 5.6 1.0 1.3 0.5 1151 216.0124.1 107.1 12.7 45.0 15.2 21.8 7.3 1151Q 168.5 15.6 75.7 12.7 35.2 3.3 14.2 1.1 952 25.013.0 25.3 2.6 4.6 1.0 1.8t0.3 452 35.5 4.3 124.3 32.7 28.0 9.1 7.9 2.3 452Q 31.4 1.6 71.4113.0 21.3 6.6 4.7 0.4 5S2 30.5 1.3 49.3 11.2 26.7 5.3 15.5 5.0 SS2Q 27.7 4.0 36.6 0.8 21.2 3.1 11.5 2.4 4G1 17.2 2.1 1.2 0.2 0.6 0.2
- 0. 6t0.1 4GlQ 17.7 0.1
- 0. 6t0.1 1.4 0.1 0.7 0.1 8C1 13.0 0.3 10.7 1.6 1.7 1.1 1.3 0.4 8C1Q 12.6 0.6 8.4 1.0 2.6 0.2 1.1 0.1 7G1 25.810.6 1.0 0.1
-0.5 0.0 0.8 0.0 16Al 907.7 49.4(2) 45.1 2.1 1.7 1.1 0.9 0.1 453.4 12.2(2)
(2) 1451 131.220.p2) 48.8 8.6 9.5 4.3 1.5 0.4 148.3 9.7(2) 1081 40.613.5(2) 14.9 0.9 0.4:0.3 1.10.2 36.611.3 (1) Suffix "Q" indicates RMC data; otherwise data are frem Teledyne-Isotopes.
(2) Results for 6-month exposure period 09/27/78-03/29/79.
(3) Additional values for SA1: 7.8 1.5, 7.4 1.2.
1377 089
Table 3-4.
NRC TLD DATA RADIATION EXPOSURES FOR PERIODS FROM 03/31/79 to 04/06/79 (includes background) 3/31-4/1 4/1-4/2 4/2-4/3 4/3-4/4 4/4-4/5 4/S-4/6 4/6-4/7 mR mR mR mR mR mR mR Station N-1 1.0 1.1
.3
.37 i.08
.32 i.08
.28 i.08
.32 1.04
.43 1.05 N-2 (wet)
.3
.45 i.05
.40 i.06
.33 1.08
.48 i.15
.40 i.05 N-3 1.21.3
.3
.43 1.05
.32 i.08
.34 i.09
.47 i.05
.50 i.11 N-4 1.01.1
.3
.48 i.08
.33 1.05
.37 i.05
.42 i.02
.48 i.10 N-S (wet)
.3
.58 i.08
.37 i.05
.35 1.05
.48 i.10
.52 i.08 NE-1 7.0 1 2.1
.2
.45 i.08
.32 i.04
.45 i.05
.38 i.04
.45 1.08 g;
NE-2 (wet)
.3
.48 i.09
.37 1.10
.33 i.08
.47 i.10
.47 i.12 NE-3
- 1. 6 i. 5
.3
.42 i.09
.38 i.08
.37 i.08
.46 i.05
.45 1.10 NE-4 2.1 1.5
.3
.37 i.05
.38 i.04
.33 1.05
.40 i.09
.43 i.05 E-1 25.0 1 8.1
.4
.53 i.1
.32 1.04 2.6 i.60
.50 i.09
.48 i.08 E-S(E-la) 8.4 1 4.6
.3
.73 i.2
.38 i.08 1.7 i.45 1.2 i.27
.32 i.04 E-2 4.3 i.5
.3
.55 i.7
.SS i.10
.38 i.08
.45 1.10
.35 i.08 E-3 2.1 i.4
.4
.42 i.1
.40 1.06
.50 i.06
.48 i.08
.32 i.08 E-4 2.5 i.4
.3
.4 i.1
.35 i.14
.43 i.19
.42 i.04
.22 1.04 SE-1 10.1 1 2.0
.3 9.1 1 1.6
.43 1.10
.92 i.19
.40 i.00
.55 1.06 u
-J SE-2 3.5 i.5
.3 4.4 i.7
.87 i.16
.38 i.08
.35 1.05
.25 i.05 sa O
4 O
Table 3-4.
(Continued) 3/31-4/1 4/1-4/2 4/2-4/3 4/3-4/4 4/4-4/5 4/5-4/6 4/6-4/7 mR mR mR mR mR mR mR Station SE-3 2.3 i.6
.3 2.8 i.7
.57 i.10
.45 1.05
.40 i.06
.25 i.05 SE-4 3.0 i.4
.3 2.1 1.4
.30 i.06
.53 i.08
.47 i.08
.25 i.05 SE-S 2.5 i.7
.3
.13 i.1
.42 1.04
.37 i.08
.62 i.31
.38 i.13 S-1
- 1. 6 i.1
.4 2.2 i.4 1.1 i.05
.37 i.05
.35 i.05
.40 t.00 S-2 1.0 1.2
.4 1.5 i.2
.52 i.08
.32 1.10
.35 i.05
.43 i.08 92 5-3 1.2 1.3
.4 1.51.3
.47 i.05
.40 1.06
.40 i.06
.55 1.10 S-4 1.2 i.2
.3 1.4 i.2
.33 i.05
.45 i.10
.55 1.18
.42 i.08 SW-1
. 9 i.1
.8 1.2 1.3 1.1 i.18
.37 i.08
.37 i.10
.45 i.05 SW-2
.9 i.2
.5 1.3 1.3
.37 i.12
.30 i.09
.43 i.08
.38 1.08 SW-3 1.1 i.3
.4
.78 i.1
.65 i.10
.45 i.10
.38 i.08
.42 i.02 SW-4
. 9 i.1
.5
.75 i.1
.62
.10
.45 i.14
.50 i.14
.50 i.09 W-1 3.0 1 1.9 1.2 1.4 1.24 1.7 i.35 1.3 i.29
.57 i.10
.48 i.08 W-2
.91.1
.5
- 1. 1.1
.62 i.04
.72 i.04
.37 i.08
.38 i.08 W-3 1.11.1
.5
.78 1.2 1.1 1.15
.42 1.08
.38 i.08
.47 i.08 W-4 1.0 1.2
.4
.67 i.1
.42 i.10
.45 i.14
.45 i.05
.57 i.08 t/4 W-5 1.2 1.2
.6
.4 1.15
.65 1.12
.60 i.13
.40 i.06
.57 i.14 N
.a C
O
Table 3-4.
(Continued) 3/31-4/1 4/1-4/2 4/2-4/3 4/3-4/4 4/4-4/5 4/5-4/6 4/6-4/7 mR mR mR mR mR mR mR Station NW-1
.9 i.2
- 1. 7 1.3 i.25
.30 i.06
.38 i.08
.52 1 12
.53 1.04 NW-2
- 1. 2 i. 5
.4
.62 i.08
.40 i.15
.33 1.05
.35 i.05
.38 i.08 NW-3 1.4 1.7
.8
.63 i.12
.40 1.25
.38 i.04
.40 i.09
.42 1.05 NW-4 S. S i 1. 8
.3
.4 i.06
.30 i.06
.37 i.08
.32 i.04
.45 i.10 NW-5 4.6 1 2.
.4
.42 i.04
.42 i.21
.32 i.04
.48 i.08
.45 i.05 m
S-la Not in Service until 4/5/79
.35 1.05
.43 i
'd SE-4a
.33 i.05
. 25 i. 0!,
W-3a
.65 i.39
.45 i.10 NE-3a
.38 i.08
.57 i.08 N-la
.50 i.19
.47 i.04 N-lb
.40 i.06
.50 i.06 N-1c
.40 i.09
.45 1.08 N-1d
.35 i.05
.50 i.06 N-le
.40 i.06
.44 1.08 N-1f
.47 i.15
.37 i.08 U
-- J
-- -J CD so N
4 28 Table 3-5.
METROPOLITAN EDISON COMPANY:
TELEDYNE-ISOTOPES 00SIMETERS TLD RADIATION EXPOSURE RATES - 1978 Results in Units of mR/ Standard month ( )
12-30-77 03-29-78 06-28-78 09-30-78 to to to to AVERAGE STATION NO.
03-29-78 06-28-78 09-30-78 12-27-78 12a Control Locations TM-ID-7F1 6.5710.17 11.910.3 7.3010.43 7.50 0.20 8.32 4.84 TM-ID-4G1 5.3010.30 8.53 0.40 5.77 0.13 5.90 0.33 6.3812.92 TM-ID-9G1 5.6010.13 9.47 0.50 6.00 0.20 5.97 0.13 6.7613.64 TM-ID-15G1 5.1310.10 8.7310.43 5.57 0.23 5.63 0.50 6.27 3.32 TM-IO-7G1 15.8 0.7 10.4 0.5 7.13 0.63 7.2010.10 10.1 8.2 Indicator Locations TM-ID-152 4.67 0.13 7.37 0.47 5.03 0.13 5.37 0.20 5.61 2.42 TM-ID-252 4.07 0.13 6.0310.17 4.7310.33 4.20 0.20 4.76 1.80 TM-ID-4S2 4.80 0.20 8.07 0.27 5.17 0.13 4.33 0.27 5.59 3.38 TM-IO-552 4.3010.13 8.00 0.27 5.03 0.40 4.2310.10 5.39 3.56 TM-ID-8C1 3.50 0.23 5.5710.30 4.10 0.17 3.50 0.13 4.17 1.96 TM-IO-952 4.67 0.10 8.53 0.33 5.57 0.20 5.67 0.37 6.11 3.34 TM-ID-1151**
5.07 0.20 17.0 0.4 6.50 0.27 5.6010.10 8.54 10.3 TM-ID-14S1**
2.17 0.13 12.2 0.4 5.77 0.73 6.71:10.2 TM-ID-1651 6.40 0.27 19.4 0.7 6.93 0.40 5.60 0.27 9.58:12.2 TM-ID-4A1 4.60 0.20 7.57 0.13 5.03 0.20 5.13 0.30 5.58 2.68 TM-ID-5A1 4.6010.17 7.4710.17 4.57 0.27 4.63 0.23 5.32 2.88 5.00 5.80 TM-ID-16Al 2.03:0.07 7.83 0.37 5.13 0.23 TM-ID-1081 1.9710.20 9.43 0.37 6.57 0.10 5.9917.52 TM-ID-1281 3.57 0.07 6.40 0.30 4.03 0.27 4.10 0.10 4.53 2.54 TM-ID-1C1 4.1010.20 6.43 0.23 4.1310.30 4.3320.27 4.7512.26 Average 4.95 5.70 9.32 7.04 5.50 2.00 5.23 2.18 t2a (1) Standard month = 30.4 days; values originally reported as 1 mrem / standard month assuming 1 mrem = 1 mR.
nTLDs were left in field.
- Originally reported, erroneously, as Stations 1152 and 1452.
1377 093
l Table 3-6.
METROPOLITAN E0IS0N COMPANY:
RADIATION MANAGEMENT CORPORATION DOC' METERS TLD RADIATION EXPOSURE RALES - 1978 Results in Units of mR/ standard month (
12-30-77 3-29-78 6-28-78 9-27-78 STATION to to to to AVERAGE NUMBER 3-29-78 6-28-78 9-28-78 12-27-78 1 20 Control Locations 4
TM-IDH-7FlQ 6.1510.73 7.6010.67 7.7910.29 8.0410.45 7.4011.70 1M-IDH-4GlQ 4.9410.52 S.9510.38 5.6810.46 6.3710.77 S.7411.20 1M-IDH-ISGlQ 4.7010.40 S.6110.38 S.6510.45 6.4710.50 S.6111.45 Indicator Locations TM-IDH-152Q 5.7110.34 S.3210.31 S.3110.42 S.8210.27 S.5410.53 Q;
TM-IDH-4S2Q 4.9110.44 S.6910.24 S.5510.51 S.0510.43 S.3010.76 1M-IDM-SS2Q 4.3210.21 S.1510.56 S.4710.32(4) 5.4410.44 S.1011.07 TM-IDM-11SlQ S.3510.45 9.7210.88 6.7S10.52 6.0910.23 6.9811.92 TM-IDM-16SlQ 3.9310.27 12.0911.31 6.6810.75 6.0210.61 7.1816.95 1M-IDH-SA1Q 4.5710.16 S.1810.38 4.8810.28 5.6010.17 S.0610.88 TM-IDM-8ClQ (1) 4.0710.16 (2) 4.3510.31 4.2110.40 TM-IDM-4A1Q 4.5610.60 (2)
(2)
(2) 4.56 IM-IDM-8SlQ (2)
(2) 4.0410.21 (2) 4.04 Average i 2a 4.9111.33 6.6414.96 S.7812.11 S.9311.96 (1) "TLDs stoien."
(2) "No sample received."
(3) Standard month = 30.4 days; originally reported as " mrem / standard month" assuming 1 mrem = 1 mR.
(4) Originally reported, erroneously, as value for Station "1152".
)
N N
Ow
30 8.
OFFSITE POPULATION COLLECTIVE DOSE ESTIMATE 1.
Introdudtion The coll'ct./e dose for the population within 50 miles of the plant was calculated for the time period of March 28 to April 7, using two independent procedures.
The first procedure utilized the empirical distribution of TLD dose data within each direction sector.
Doses at distances between those locations with measured values were estimated by interpolation.
A power law method was used to extrapolate when necessary.
The second procedure utilized onsite meteorological data with the TLD readings to establish the distribution of dose within a 50-mile radius of the facility.
The distribution of dose was then summed to obtain the collective dose.
The population data used for the dose estimates were the 1980 projected of? site population distribution as presented in the Final Safety Analysis Report (FSAR).* These population distributions are contained in Tables 3-7 and 3-8 covering radii of 0-10 miles and 10-50 miles respectively.
2.
Dosimeter Background Correction The TLD exposure data reported in Tables 3-3 and 3-4 include a back-ground due to terrestrial radiation, cosmic radiation and other sources unrelated to plant releases.
In order to estimate the net exposure due to plant emission this background must be subtracted from the total TLD exposure.
The background Final Safety Analysis Report, Three Mile Island Nuclear Station, Unit 2, Vol-1, Chapter 2, Figures 2.1-5 and 2.1-10.
1377 095
31 Table 3-7.
PROJECTED 1930 POPULATION DISTRIBUTION, 0-10 MILES THREE MILE ISLAND NUCLEAR STATION, UNIT 2 (FROM FIG. 2.1-5 of SAR)
Distance (Miles)
Sector 0-1 1-2 2-3 3-4 4-5 5 - 10 0 - 10 N
19 212 3,970 3,772 415 11,840 20,228 NNE 55 75 169 480 373 11,223 12,375 NE 42 134 271 428 186 2,246 3,307 ENE 58 55 186 461 262 1,567 2,589 E
42 60 39 137 552 10,431 11,261 ESE 6
36 149 214 236 2,809 3,450 SE 6
94 67 203 395 2,095 2,860 SSE 88 197 117 78 43 3,840 4,363 S
0 0
136 817 1,317 12,190 14,460 SSW 84 98 584 217 752 6,883 8,618 SW 84 104 181 562 219 4,297 5,447 WSW 29 273 117 796 237 2,961 4,413 W
36 369 36 331 571 7,155 8,498 WNW 22 106 253 197 235 11,823 12,636 NW 39 106 64 41 1,177 29,482 30,909 NNW 48 98 1,240 942 1,921 16,632 20,881 658
_2,017 7,579 9,676 8,891 137,474 166,295 1377 096
32 Table 3-8.
PROJECTED 1980 POPULATION DISTRIBUTION,10-50 MILES
~
THREE MILE ISLAND NUCLEAR STATION, UNIT 2 (FROM FIG. 2.1-10 of SAR)
Distance (Miles)
Total Sector 30-20 20-30 30-40 40-50 10-50 N
12,663 9,005 8,941 47,588 78,197 NNE 18,240 6,826 14,478 45,115 84,659 NE 39,726 38,979 9,546 62,345 150,596 ENE 10,205 14,757 45,445 177,672 248,079 E
18,853 62,028 42,445 38,754 162,080 ESE 34,339 124,988 27,822 42,737 229,886 SE 20,152 10,000 10,600 26,958 67,710 SSE 44,204 10,774 15,097 66,763 136,838 S
111,002 14,648 13,477 75,781 214,908 SSW 31,917 44,031 18,596 37,729 132,273 SW 11,801 19,931 25,536 18,979 76,247 WSW 5,882 7,996 8,948 23,010 45,836 W
21,769 35,025 10,370 20,602 87,766 WNW 70,460 14,188 5,333 3,681 93,662 NW 99,593 9,308 9,970 12,630 131,501 NNW 26,482 10,517 7,256 12,866 57,121 Total 577,288 433,001 273,860 713,210 1,997,359 0-10 mile popul ation 166,295 0-50 mile population 2,163,654 1377 1397
33 varies from station to station and also depends on the type of TLD being used.
Each set of TLD data requires its own appropriate background estimate.
The background value for each Metropolitan Edison Station with Teledyne Isotope dosimeters in Table 3-3 was estimated on the basis of data collected with similar dosimeters for the period December 30, 1977 - March 29, 1978, as shown in Table 3-5.
Inherent in the use of these data is the assumption that there were no significant plant releases during that time period.
Since the first quarter of 1978 is used as the background for +.he first quarter of 1979, any seasonal eff ct on background should be minimized.
An exception was made for station 7GI, which is inside a brick building at the Columbia water treat-ment piant.
Since the first quarter exposure for 1978 (15.8 mR/Std. mo.) was substantially greater than that for the subsequent quarters, the exposure for the most recent quarter (7.20 mR/std. mo. for the last quarter of 1978) was used in order not to overestimate the background.
As mentioned previously, Metropolitan Edison utilized RMC dosimeters at several sites as a quality control check on the Teledyne Isotope dosimeters.
The RMC results for 1978 (Table 3-6) are in reasonable agreement with those of Teledyne Isotopes for 1978 (Table 3-5) with the possible exception of the second quarter of 1978 which is a period during which fallout from a Chinese nuclear test made a substantial contribution to the measures exposures.
i377 098
34 Also, as mentioned previously, the NRC dosimeters which were also analyzed by RMC are not identical to either the Teledyne Isotopes (TI) or the RMC TLD's used for Metropolitan Edison.
Since these TLD's were not deployed prior to the incident there are no previous data to provide background estimates for these particular dosimeters at the NRC locations.
The assumption was made that the backgrounds for those locations which are located near the Metropolitan Edison stations are the same as for the TI TLD's at those locations.
Pairs of Metropolitan Edison and NRC dosimeters with similar locations are:
N-1 and 1C1, SE-5 and 7F1, NW-5 and 15G1, W-2 and 1281, S-4 and 9G1, E-la (E-5) and 5A1, E-1 and 4A1, NE-1 and 252, SE-3 and 8C1.
The background for the remaining NRC stations was estimated as the mean of the Metropolitan Edison /TI stations for the first quarter of 1978 (except that the value for the last quarter of 1978 was used for station 7G1).
These values probably underestimate the background for the NRC dosimeters at NRC locations and therefore result in an overestimate of the plant's contribution to reported dose readings.
3.
Conversion from TLD Exoosure to Dose
- The dose was estimated by subtracting the appropriate background for each station and time period from the TLD exposure.
This net exposure (mR) was converted to dose equivalent (mrem) assuming a conversion factor of 1 mrem /mR.
AThe term " dose" is used for brevity rather than the more precise term " dose equivalent."
1377 099
35 4.
Standard Grid and Population Data The region surrounding the plant is represented on a circular grid centered at a point midway between the reactor buildings.
This standard grid contains 16 sectors (N clockwise through NNW) centered on the appropriate direc-tion.
Each sector is divided into segments at distances of 2000 ft (.379 mi),
1, 2, 3, 4, 5, 10, 20, 30, 40, and 50 miles.
The 2000-ft distance corresponds to the radius of the exclusion area for the plant.
Tables 3-7 and 3-8 show the estimated 1980 population for each sector segment for distances 0-10 miles and 10-50 miles respectively.
5.
Opse Estimation for Locations Within the Standard Grid The first step in estimating coses based on the TLD measurements for each period is to estimate the doses at each location on the standard grid.
This was accomplished by a method which is equivalent to plotting the measured doses for each sector on log-log graph paper and joining the measured values by straight line segments.
The intersection of each line segment with a standard distance for the grid is taken as the dose at that distance.
Doses between locations where the net measured values are not both greater than 0 are estimated oy simple linear interpolation.
Doses at distances beyond the outermost docimeter or within the inner-most dosimeter were estimated by extrapolation using the assumption that the 1377 100
36 dispersion in a sector is proportional to distance to the (-1.5) power.(1)
The DOE and TLD results indicated that a more rapid dispersal than the distance to the (-1.5) power should be used.
They stated that the data suggest a more rapid decrease of exposure with distance, more consistent with an exponential of (-2) power (Appendix A).
Doses for the standard distances in sectors in which no measurements were made were estimated by interpolating linearly between the dose values of the adjacent sectors for which measured data were available.
The mean dose within each sector segment was estimated by weighting the dose, H(r), t,y the area within the sector
- 2 H(r)rdr ry 5=
rg rdr ry waere 5 is the mean dosa, H(r) is the dose as a function of distance, r, and r and r are the inner and outer radii of the sector segment, respectively.
y 2
(1) M. Smith (Ed.) " Recommended Guide for the Prediction of the Dispersion of Airborne Effluents," American Society of Mechanical Engine (rs, New York (1968), p. 44-46.
This reference shows that the airborne concentration P
varies as x where p can vary from 1.4 (stable conditions) to 1.8 (very unstable conditions).
The value p = 1.5 approximates a daily average value.
1377 101
37 The collective dose for each sector segment is the product of the corresponding mean dose and the population as given in Tables 3-7 and 3-8.
The sum of the collective doses for all sector segments and periods is the total collective dose for the entire assessment area for the total period under consideration.
6.
Collective dose calculations
- Four approaches were used in estimating the total collective dose for the period March 28-April 7.
Each utilize data from the Metropolitan Edison TLD stations for the period March 28 through March 31, since there were no NRC TLD's in place before March 31.
For the first calculational approach, all Metropolitan Edison data for the period March 28-March 31 were used for estimating the collective dose for the periods March 28-29 and March 29-31 (3200 person-rem).
The NRC data, which are all from offsite locations, provided the data for the periods from April 1 through April 7.
The increase in total collective dose with time using this approach is shown in Figure 3-6.
Note that there is a significant contri-bution to the collective dose (1100 person-rem) from the first NRC period (3/31-4/1) and that there is a continuing steady contribution each day for the remaining periods.
A strength of this method is that it utilizes the xA copy of the computer program for generating the collective doses is available from Christopher Nelson, Environmental Protection Agency, Office of Radiation Programs (ANR-461) Washington, D.C. 20460.
1377 102
6,000 5.300 5,000 Met.Ed. + NRC 2
aI e
y 4,000
,t'
'N Het.Ed. only 3,300 O
U 3,000 2,800 i:
o J'".
Met.Ed. + NRC Het.Ed. - x/Q
(<8 miles) w*
"l I
b 2,000
---J Met.Ed. only a
1
(<8 miles) 1,600 28 1,000 1
1 I
i i
1 3
Mar. 28-29 Mar. 29-30 Mar.30-31 Mar. 31-Apr. 1-2 Apr. 2-3 Apr. 3-4 Apr. 4-5 Apr. 5-6 Apr. 6-7 Apr. 1 EXPOSURE PERIOD -- 1919 U
figure 3-6.
Increase in Collective Dose as a Iunction q
of Ilme for Various Calculation Strategies u
39 maximum possible number of individual observations and therefore would be expected to be least dependent on any one of them.
Since the NRC locations are nearly all offsite, they provide better general coverage of the populated areas surround-ing the plant.
However, there are limitations to using this method.
For example a net measurement of a few millirem may easily represent nothing more than a low estimate of the background for that location.
If the location is distant from the facility, and is the only measurement in the sector, it can contribute to a significant overestimate in the collective dose.
Another limitation of this method lies in the uncertainty of the background values for the NRC locations.
As i~ndicated previously, these background values are believed to be low.
The continuing rise in the collective dose in later periods, when there is r.o reason to expect any significant contribution from the facility, confirms this expecta-tion.
The collective dose through April 7 using this methodology is 5300 person-rem and is believed to be a high estimate for the reasons given.
The second approach is based on using only the Metropolitan Edison TLD data.
This approach has the advantage of using a consistent set of data with the same dosimeter type and locations throughout the period.
The back-ground values are reasonably well known by experience for these stations.
A disadvantage to this approach is that there are only 20 dosimeters, so that three sectors (NE, ESE, W) have no measurements at all and seven (NNE, SSE, SSW, SW, WSW, WNW, NW) have only one.
There was concern that the onsite TLD's might be influenced by radiation levels associated with radionuclide contained within the facility and would therefore not be appropriate for estimating offsite 1377 104
40 doses.
This, however, does not appear to be the case.
These dosimeters around the periphery of Three Mile Island show a variation from time period to time period, which would not be expected if they were appreciably affected by unreleased onsite activity.
Onsite radiation monitoring with hand-held radiation monitors also confirms the absence of a significant " direct radiation" component except very close to the containment or auxillary buildings.
The total collective dose through April 6 using this approach is 3300 person-rem.
April 6 becomes the cutoff point in this method because of the 3-day dosimeter cycle under which the Metropolitan Edison TLD's were deployed and read out.
A third approach is based on a subset of the dosimeters used in the first method.
Those locations outside 8 miles were dropped from the analysis.
This has the advantage of minimizing the effect of those locations which are least likely to have measurements that should affect the end result.
While there are many offsite NRC stations, only seven of them are outside 8 miles so that generally good coverage of the offsite exposure is still obtained.
The disadvantage is that a significant dose at a distance greater than 8 miles in a direction where there are no other dosimeters nearer to the facility will be missed completely.
Note that this substantially reduces both the March 28-31 Metropolitan Edison dosimeter contribution to the collective dose and the contri-bution from the first day of NRC observations.
The total collective dose through April 7 using this approach is 2800 person-rem.
1377 105
41
~
The fourth approach is based on using those Metropolitan Edison TLD data from locations that are not more than 8 miles from the facility.
Again the method has the advantage of a consistent base of data for the entire period and the disadvantage of making a small data base even smaller.
The effect of eliminating the distant stations is to reduce the collective dose calculated for the period.
Using approach four, the collective dose through April 6 is 1600 person-rem.
Time did not permit the inclusion in the dose calculations of the Metropolitan Edison - RMC TLD data (which provide independent measurements of the exposure at 10 of the Metropolitan Edison TLD locations).
Inspection of these RMC results, in Tables 3-3 and 3-6, indicates that including them in the calculations would lower the calculated collective dose values.
Given the limited number of observation (especially for the period March 28-31, when it would appear that most of the collective dose was delivered) it is evident that any approach to assessing the collective dose depends entirely on a relatively small number of measurements.
No amount of sophisticated analysis can change this fundamental limitation.
On the other hand, it is also clear that the data de allow reasonable estimates of the collective dose to be made.
7.
Calculations Emoloying Meteorological Dispersion Factors Computed values of the meteorological dispersion factor (x/Q) for the time period of March 28, 4:00 a.m. through March 29, 8:00 a.m. and March 29, 1377 106
42 J
8:00 a.m. through Marc.h 31, 4:00 a.m. were used to estimate population dose.
This method was intended to serve as an independent backup to the methods described earlier.
These values of X/Q calculated hourly were time averaged over these periods for each distance and direction segment.*
The dose H (mrem) for time interval, at, is calculated from the following equation:
H=({}Q(DF)at where H
dose recieved over the time interval, at (mrem)
Q source (Ci/sec) 3 (X/Q) meteorological dispersion factor (sec/m )
3 DF dose factor (mrem m /Ci sec) at length of time interval (::ec).
Assuming that the release rate, Q, is constant over time interval, at, the quotient, H/(X/Q), should be constant for each sector section since the p.; duct, Q(DF)at, is constant.
The TLD readout values of H appearing in Table 3-3 for the first tv time periods were divided by the corresponding X/Q values datermined by interpolation of the meteorological data.
These quotients were then averaged for each time period and the standard root-mean-square (r.m.s. )
These data and calculations are available if requested.
Contact Dr. F. Congel, U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation (P-71'>.), Washington, D.C. 20555.
1377 107
43 error was computed.
The standard r.m.s. error for the first period was 79%,
and for the second period was 73%.
These values provide an estimate of the uncertainty in the subsequent calculations.
Multiplication of these two average H/(X/Q) values by the appropriate X/Q value at different sector segment for the two time intervals yielded an estimate of the dose at those locations for each time interval.
The total population exposcre was estimated by multiplying the sector segment population by the H value of the inner boundary.
Since the inner boundary value is always larger than the outer boundary value for each sector segment, a conservative estimate of population dose is obtained.
The total 0-50 mile population dose for the first and second period was 1900 person-rem and 680 person-rem, respectively, for a total value of about 2600 person-rem.
This value lies in the middle of the range of values estimated using the procedure described in the preceding section (see Figure 3-6).
C.
OFFSITE MAXIMUM DOSE TO AN INDIVIDUAL The estimated maximum dose to an individual depends upon the local meteorological conditions, namely, wind direction, wind speed, and plume disper-sion characteristics.
The known meteorological conditions throughout the accident period indicate that there were three predominant directions in which radioactive material released from the plant would be expected to be found.
These directions 1377 108
44 were characteristic of the near-field 0-5 mile dispersion values (X/Q), as well as the far-field 5-50 mile, dispersion values.
In addition to the meteorological parameters, thermoluminescent dosimeters (TLD's) placed in the field within and beyond the site boundaries, measurements of airborne plume radioactivity by helicopter flights made by the Department of Energy after the onset of the accident and throughout the period, and both onsite and offsite survey meter readings support the meteorological data that suggest three predominant directions of aerial dispersion.
Figure 3-7 depicts these directions.
This figure was prepared from the meteorology data for the period from March 28 through April 3.
The lobes of the uniform exposure lines of Figure 3-7 indicate the predominant wind direction to be towards the NNW, ENE, and SSE sectors.
The maximum exposed individual would be expected to reside in one of these sectors, and also at a location close to the plant within one of these sectors since the airborne concentration of radionuclides in the plume decreases as distance from the source increases.
The lobe extending over a populated area close to the plant, is the one centered around the ENE sector.
The TLD located in this sector at a distance of 1/2 mile registered a net dose for the 5-day period of 83 m-em.
This dose value represents an upper limit dose estimate in that sector for the 5-day period since no individual member of the general public could be closer to the plant in that sector.
The next nearest populated land mass outside the plant boundary in the direction of a lobe is in the SSE sector.
This area is located approximately 0.8 mile from the plant.
The nearest TLD 1377 109
45 X\\,
s
,f Y
l j
\\
Sw
/
i
/
xA
/
?:n:it:a::n::::1.te,
\\.
\\(
w.
s
/
" S"" - sii's#! ?n'H'4*CJ:Ei'l4ce, s
r on Pattern of March 28 through
\\
.5 m.
'N y
,::, :m m,, _
46 located in the south sector is approximately 0.4 mile from the plant.
The accumulated net dose at this location is 41 mrem.
It is expected that the exposure in the populated area in the SSE sector would be less than the 41 mrem as the populated area is twice as far from the plant as the detector.
The nearest populated area in the NNW sector is Hill Island which is 1.1 miles away.
However, the maximum potential dose would be on Kohr Island which was uninhabited during the first three days of the accident.
The next highest dose would be to anyone on Hill Island.
On April 19, 1979, inspectors from the NRC Office of Inspection and Enforcement discovered that three people claimed to be on Hill Island (about 1 mile NNW of the TMI Island plant) during the first three days of the accident.
A subsequent detailed interview was conducted with one of the three individuals.
He stated that he was the only individual present on the island and that he was working on a summer cottage.
He was present only from 11:00 am to 4:30 pm on March 28 and from 11:00 am to 3:00 pm on March 29.
The potential dose to an individual at that particular location on Hill Island had to be estimated from TLD's that were close to the TMI facility and in the same direction sector (NNW) as the island.
The following data were used to estimate the exposure at Hill Island (1.1 miles NNW).
\\bl7
\\\\\\
47 Distance / Direction Dose (mrem) miles NNW 4:00 a.m. 3/28 - 12:10 p.m.
3/29 1020 0.20 900 0.42 440 0.42 12:10 p.m.
3/29 - 10:45 a.m.
3/31 83 0.20 45 0.42 The large discrepancy (900 versus 440 mrem) between the two TLD dosimeter values for the initial time period was investigated, but could not be explained.
However, based on an examination of the meteorological dispersion during that time period, it appears that the 440 mrem value is more plausible.
Using the 440 mrem value, the extrapolated dose at the cottage location on Hill Island would be about 150 mrem for the first time period.
The dose at Hill Island for the second time period dose at Hill Island is about 18 mrem.
Since the person was not present on the island during the entire course of the accident, his exposure has to be reduced accordingly.
This " occupancy factor" is determined in a simple manner as follows:
The individual was present for about 5.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> on March 28 and slightly more than 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> on March 29 until the first TLD's were replaced at 12:10 p.m.
The total individual exposure time was 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> for the first time period.
It was assumed the radioactive releases started at about 7:00 a.m. March 28, and continued uniformly until
\\51i \\\\2
48 the TLD's were replaced at 12:10 p.m. on March 29 for a total TLD exposure time of 29 hours3.356481e-4 days <br />0.00806 hours <br />4.794974e-5 weeks <br />1.10345e-5 months <br /> for the first time period.
The second time period began at 12:10 p.m. March 29, and ended 10:45 a.m. on March 31, for a total TLD exposure time of 47 hours5.439815e-4 days <br />0.0131 hours <br />7.771164e-5 weeks <br />1.78835e-5 months <br />.
The individual on Hill Island was exposed during the second time period for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />.
Using the actual occupancy time on the island, the estimated individual exposure becomes approximately 37 mrem.
If the higher initial period TLD reading (900 mrem) was used, the indi-vidual's dose would be about 180 mrem.
If the two TLD's with the large discrepancy are averaged (670 mrem), the individual's dose would be about 93 mrem.
However, it appears that the nus.
obable estimate of dose to the individual is 37 mrem.
1377 113
~
~.
49 4.
POTENTIAL HEALTH IMPACT OF EXTERNAL EXPOSURE A.
Health Effects from Low-Level Radiation The health risks from low-level radiation are derived by assuming that the effects observed at high doses from high dose rates can be directly and linearly extrapolated to low doses delivered at very much lower dose rctes and also by assuming that there is no absolutely safe dose (or threshold) below which there is no health risk * (linear, non-threshold, dose-rate-independent dose-effect relationship).
These assumptions are generally **
believed to overestimate the health risk from low-level ionizing radiation (1-3) particularly for beta and gamma radiation.
Somatic Effects Somatic radiation effects are those effects that may appear in the irradi-ated individual.
The primary somatic effect observed following high doses of The 1972 BEIR Committee (3) indicates (p. 88) that the possibility of zero as lower limit on the carcinogenic effects cannot be excluded by existing data.
snThere are a few recent studies that suggest that the risks of low-level ionizing radiation might be greater than predicted from linear extrapolation from high doses.
However, the results of these studies have not been generally accepted by the scientific community.
It is important to consider both studies that present higher risk astimates and studies that present lower risk estimates together with the complete body of scientific literature on the effects of ionizing radiation rather than relying on the results of a single, or even a few, studies.
(1) International Commission on Radiological Protection, " Recommendations of the International Commission on Radiological Protection Adopted January 13, 1977" ICRP Publication 26, Pergamon Press, Oxford (1977) Section E pp 6-7.
(2) National Council on Radiation Protection and Measurements, " Review of the Current State of Radiation Protection Philosophy."
NCRP Report No. 43, NCRP, Washington, D.C. (January 15, 1975) p.4.
(3) Advisory Committee on the Biological Effects of Ionizing Radiation (BEIR)
"The Effects on Populations of Exposure to Low Levels of Ionizing Radiation,"
National Academ:r of Sciences - National Research Council, Washington, D.C.
November 1972, Chapter VII,Section IV pp 87-88.
1377 114
50
~
radiation is an increase in cancer deaths (cancer mortality).
The risk of cancer per unit dose of radiation can be expressed in an absolute sense or in relative (comparative) sense.
The absolute risk is the difference in risk between an exposed (irradiated) population and an unirradiated population of similar characteristics.
Under the linear dose-effect relationship, the absolute risk may be expressed as the increased number of radiation-related cases of cancer per year in an exposed population per unit of dose; for example,10 6
deaths per year per million people exposed per rem (10 deaths / year per 10 person-rem).
The relative risk is the ratio between the risk of the irradiated popula-tion and the unirradiated population.
It is usually stated as a fraction or multiple of the natural risk for that particular effect; for example, 0.5%
per year.
In order to convert the relative risk into units comparable to the absolute risk, it is necessary to multiply the relative risk by the natural 5
cancer mortality rate for each type of cancer (cancer deaths per year /10 p,g ),);
for example, for total cancer mortality the death rate is approximately 2000 deaths per year per 1,000,000 people; therefore:
0.005(0.5%) x 2000 cancer deaths = 10 cancer deaths / year rem 10* person years 10* person-rem The risk of cancer does not appear to be increased immediately after irradiation.
It usually requires several years (typically 5-20 years, depending upon the cancer type) before the risk becomes increased.
This time interval between irradiation and the appearance of cancer is called the latent period.
1377 115
\\
51 Following this latent period, there is a period where there is an increased risk of cancer in an irradiated population.
In order to estimate the total risk of cancer from a single dose of radiation, it is necessary to multiply either the absolute risk or the relative risk by the duration (length) of this period.
The exact length of the period of increased risk is not known for most radiation-induced cancers.
Therefore, two assumptions have been made concerning this:
Assumption A is that the risk remains elevated for 30 years following the latent period and then drops to zero.
Assumption B is that the risk remains elevated for the remainder of the individual's lifetime.
The risks of fatal cancer from radiation exposure, estimated from the data in the 1972 BEIR Report (3) for both relative risks and absolute risks and for Assump-tions A and B, are shown in Table 4-1.
Genetic Effects It is firmly established that ionizing radiation can cause genetic mutations and other anomalies.
These effects can be manifested as congenital anomalies (birth defects) or hereditary diseases, such as dwarfism, phenylketonurea, or Down's syndrome, in descendents of an irradiated parent or parents.
- However, the exact numerical value for the risk of genetic injury from low doses is uncertain.
The genetic effects estimated in the 1972 Report of the Advisory Committee on the Biological Effects of Ionizing Radiation (3) are based upon estimates that the radiation dose which would double the natural incidence of genetic anomalies (doubling dose) is between 20 and 200 rem (20,000 and 200,000 mrem).
The lower the doubling dose, the greater the risk from a given radiation 1377 116
l Table 4-1.
RADIATION-INDUCED CANCER MORTALITY ESTIMATED IN Tile 1972 BEIR REPORT (3) 1972 BEIR Report Est.imates Derived Risk Annual number of deaths resulting from Number of Cancer Deaths per person-'em(b) 6 exposure of the U.S. population to a 10 radiation dose rate of 0.1 rem [100 millirem]
per year (d}
Absolute Risk Relative Risk Absolute Risk Relative Risk Model Model Model Model Leukemia S16 738 26 37 Other Fatal Cancers m
AssumptionA:fc N
1210 2436 61 123 d-Assumption 8:
1485 8340 75 421 Total (Range)I*)
1726-2001 3174-9078 87-101 160-458 Nominal Range (I) 1700-2000 3200-9100 90-100 160-460 Geometric mean (90 x 460)1/2 200 (203)
=
6 6
(a) 1967 U.S. population = 197,863,000.
Collective Dose Rate = (198 x 10 eople) x (0.1 rem /yr) = 19.8 x 10 person-rem / year.
From Table 3-3 (Relative Risk and Table 3-4 (Absolute Risk) of the 1972 BEIR Report (3) u pp. 172-173.
6 (b) 1972 BEIR Values (Cancer deaths / year) divided by the collective dose rate of 19.8 [10 person rem]/ year.
N (c) Assumption A:
30 year period of elevated risk following irradiation.
(d) Assumpt. ion B:
Lifetime period of elevated risk following irradiation.
(e)
Low estimate = Leukemia Risk + Assumption A for other fatal cancers.
[
liigh estimate = L;ukemia Risk + Assumption B for other fatal cancers.
(f)
Preceeding values rounded to two significant figures.
53 dose.
Table 4-2 summarizes the calculation of the genetic risk per unit radiation dose from the data given in the 1972 BEIR report.
This calculation was based upon the 1967 birth rate of approximately 18.2 births per year per 1,000 people.
The use of the 1976 birth rate of 14.2 births per year per 1,000 people would give lower risks per person-rem by a factor of 14.2/18.2 or about 0.8.
B.
Comparison of Doses to Individuals from the TMI Accident with Natural Background Radiation and its Variability Man is continually exposed to ionizing radiation which occurs naturally.
There are three primary sources of this natural radiation " background";
(1) solar and galactic cosmic radiation, (2) long-lived radionuclides in the earth's crust (primordial radionuclides) and (3) radionuclides formed in the upper atmosphere from the interactions of the cosmic radiation with gases in the atmosphere (cosmogenic radionuclides).
The magnitude and variation in the radiation dose from these natural radiation sources provides one baseline for comparing the doses and the potential health impact frem the Three Mile Island accident.
Estimates of the dose from background radiation at several locations in the United States are shown in Table 4-3.
None of these values are measured values but they are generally consistent with reported measurements.( ~ )
(4)D.T. Oakley, " Natural Radiation Exposure in the United States," EPA Report ORP/SID 72-1, U.S. Environmental Protection Agency, Washington, D.C. (1972).
(5) National Council on Radiation Protection and Measurements, " Natural Back-ground Radiation in the United States" NCRP Report No. 45, NCRP, Washington, D.C., November 15, 1975.
'\\ 3 I I
)\\0
Table 4-2.
ESTIMATES OF GENETIC EFFECTS OF LOW-LEVEL IONIZING RADIATION 6
Disease Classification Natural Effects per 10 live Estimated Risk Incidence births ( ) of S rem per per 10 person-rem (c) 6 generation ( }
6 (per 10 live births)
First Generation Equilibrium First Generation Equilibrium Dominant diseases 10,000 50 to 500 250 to 2500 6 to 60 30 to 300 Chromosomal and recessive diseases 10,000 relatively very slow relatively very slow slight increase slight increase Congenital anomalies 15,000 Anomalies expressed later 10,000 5 to 500 50 to 5,000 0.6 to 60 6 to 600 Cc itutional and degenerative diseases 15,000 TOTAL 60,000 60 to 1000 300 to 7500 7 to 120 36 to 900 Risk per 10 people 1,200(d)/ ear 6
Geometric Mean 180 (a)from the 1972 BEIR Report (3), Table 4 p. 57 which is believed to be erroneously titled.
This ta ta, like the preceding tables 2-3 pp. 54-55, is believed to be for a population of one million " live births" not u
for a pcuulation of one million.
The range of values corresponds to assumed doubling doses between N
20 rem (high values) and 200 rem (lower values).
(b)A generation is assumed to be 30 years.
6 6
6
[
(') Risk per 10 person rem = (cases /10 live births) x (30 years /S rem) x (4 x 10 live births / year per 0
6 4
2 x 10 people) = 0.12 x cases /10 live births.
'd) Cases /106 6
8 live births x (4 x 10 live births per year / 2 x 10 people).
55 Table 4-3 Estimates of Natural " Background" Radiation Levels in the United States Annual Dose Rate (mrem / year)
Location Cosmic Terrestrial Internal Total Radiation (a)
Radiation (a)
Radiation ( )
Atlanta, Georgia 44.7 57.2 28 130 Denver, Colorado 74.9 89.7 28 193 HARRISBURG, PA.
42.0 45.6 28 116*
Las Vegas, Nev.
49.6 19.E 28 98 New York, NY 41.0(c) 45.6(c) 28 115 PENNSYLVANIA 42.6 36.2 28 107 Washington, DC 41.3 35.4 28 105 UNITED STATES40-160 0-120 28 70-310
(^)Frcm [(4) Table A-1]
(b) Based upon total of internal (gonadal) doses from [(5) Tables 42 and 43, p. 104].
(c)From ((4) Table A-2]
aThe value used elsewhere in this report is 125 mrem / year which is based upon the Final Environmental Statement for the Three Mile Island Facility (AEC, 1972, Section VD 7, p. V-28).
As neither value represents direct measurements and ambient radiation dose rates are expected to vary by at least 25% between locations within a 50-mile radius, these estimates are essentially identical.
1377 120
56 Table 4-4 compares the estimated individual doses from the Three Mile Island accident to some of the variations in annual radiation doses from background radiation.
It should be noted, however, that the " background" doses are delivered continuously (i.e., mrem /yr) whereas the accident doses were delivered over a period of a few days.
The possible significance of this higner dose rate is discussed in a following section on dose-rate effects.
It should also be noted that the " average" doses to individuals within 10 and within 50 miles of the site are numerical averages obtained by dividing the collective population doses by the size of enclosed population.
Clearly, some individuals received more than this dose and others less, depending upon wind direction and distance from the TMI site.
The total collective dose from natural background to the population within 50 miles of the Three Mile Island site is estimated to be 270,000 person-rem per year (0.125 rem per year x 2,164,000 persons).
The potential health consequences of the natural radiation exposure are shown in Table 4-5.
C.
Existing Cancer Rates and Risks Cancer is the second leading cause of death (next to heart disease) in the United States [(6) p. 14)].
The Vital Statistics of the United States, 1976(0) shows that there were 377,312 deaths in the U.S. from cancer, which (6)From American Cancer Society, " Cancer Facts and Figures - 1979," Reproduced by permission of the American Cancer Society who retains copyright.
Sub-sequent quotations should acknowledge the American Cancer Society as the source of these values.
1377 121
57 Table 4-4 Comparison of Individual Doses from the Three Mile Island Accident With Variations in Natural Background Radiation Doses CUMULATIVE THREE MILE ISLAND TOTAL BODY DOSES ACCIDENT DELIVERED THRU 4/7/79 Individual remaining out-of-doors at location of highest estimated offsite dose less than 100 mrem Average dose to a typical individual within:
50 miles of site 1.5 mrem 10 miles of the site 8
mrem (These values correspond to tha 3,300 person-rem collective dose estimate)
ESTIMATED DIFFERENCE IN NATURAL BACKGROUND VARIATION ANNUAL DOSES Living in Denver, Colorado compared to Harrisburg, PA
+ 80 mrem /yr (from Table 4-4)
Living in a brick house instead of a wood frame house [Yeates data in (4)
Table 16, p. 35]
+ 14 mrem /yr Added dose from potassium-40 due to 4.8 mrem /yr being male instead of female
+
(There is 25% less potassium in women than men [(5), p. 106])
1377 122
Table 4-5.
PROJECTED ANNUAL IMPACT OF NATURAL BACKGROUND RADIATION EXPOSURE ON Tile POPULATION RESIDING WITilIN 50 MILES OF Tile TilREE MILE ISLAND SITE Estimated Percentage of Estimated Estimated Impact of Existing Rate Which Hight Existing Rate NaturalBackpggund be Caused by Natural Effect (per year)
Radiation Background Radiation (per year)
Fatal Cancers 3,900 Absolute risk 24-27 0.6-0.7%
Relative risk 43-124(b) 1.1-3.2%
Central Estimate 54 1.4%
Spontaneous Mutations 2,600'C)
(10-245 0.4-9.4%
(Genetic Effects)
Central Estimate 50(b) 9%
(d) Assumed to be 125 millirem (0.125 rem) per year to the 2,163,654 people projected (1980) to live within 50 miles.
This G ves a collective dose rate of 270,500 person-rem per year.
i (b)The central estimate is obtained from the Geometric mean of the risk estimates.
(c)1,200 per year per 10 people (frca 'able 4-2) x 2.163 million people.
6 N
N
~
tra
59 corresponds to a rate of 175.8 cancer deaths per 100,000 people [(6) p. 14].
Cancer deaths accounted for approximately one-fifth (0.198) of ali deaths in the U.S. in 1976.
The existing cancer rate provides an indication of the possibility of detecting any potential increase in cancer incidence due to the Three Mile Island accident.
The cancer death rate for the State of Pennsylvania estimated by The American Cancer Society [(6) p. 12] i. 208 per 100,000 (2.08 x 10-3).
Portions of the State of Maryland are also located within 50 miles of the TMI site.
Maryland has a lower estimated rate (179 per 100,000) which is closer to the estimated U.S. rate af 180 per 100,000 [(6) p. 12].
Applying the U.S. or Pennsylvania values to the 2,164,000 people estimated to reside within 50 miles of the Three Mile Island site gives an approximate estimate of 3,900 (U.S.)
to 4,500 (Pa) deaths per year for the existing cancer death rate for the popula-tion within 50 miles of the TMI site.
Table 4-6 shows the estimated incidence (number of new cases) and death rate for the U.S. population for selected types of cancers.
The American Cancer Society [(6) p.14] estimates that, out of 100,000 people, 25,000 will eventually develop cancer and, of these 25,000, about 15,000 will eventually die of cancer.
This gives an estimate of the risk of cancer death of 0.15.*
Applying this approximate statistic to the population within 50 miles of the Three Mile Island site indicates that approximately 325,000 people in that area would normally die of cancer, nThis has a range between 0.15 and 0.17 depending upon the source of the data and the year to which it applies.
1377 124
60 4
Table 4-6 Estimated New Cancer Cases and Deaths in the United States for 1979 (Existing Rates)
Cancer Site Estimated
- Estimated
- Deaths / Cases (a)
New Cases Deaths Digestive Organs 182,900 105,150 0.57 Lung 112,000 97,500 0.87 Bone 1,900 1,750 0.92 Skin 13,600(D) 5,900(c) 0.43 Breast 106,900 34,500 0.32 Genital Organs 143,500 44,800 0.31 Leukemia 21,500 15,400 0.72 Thyroid 9,000 1,000 0.11 All Sites 765,000 395,000 0.52 (a)If cancer rates and the population (and its age composition) were constant this ratio would be a measure of the probability of dying from having specified types of cancer.
As neither existing cancer rates nor the U.S.
population and its age breakdown are constant, this is only an approximate measure of severity of cancers at a particular site.
(b)This only for melanoma, a rare skin cancer with a high mortality rate.
(c)This includes all fatal skin cancers.
Melanoma accounts for about 4,300 of these cases.
sFrom American Cancer Society, " Cancer Facts and Figures-1979" p. 10.
Repro-duced by permission of the copyright holder, the American Cancer Society.
All subsequent quotations of these values should acknowledge the American Cancer Society as the source of these estimates.
1377 125
61 D.
Summary of the Health Imoact to the Exocsed Population Table 4-7 shows the estimated potential health effects from the Three Mile Island Nuclear Accident.
The central estimate is associated with the mean value of the collective dose 3,300 person-rem delivered to the popula-tion within 50 miles of the reactor.
These estimates consider fatal cancers, non-fatal cancers and genetic ill-health to all future generations.
The projected total number of fatal cancers is approximately 1 (0.7).
The additional number of non-fatal cancers is also approximately 1.
The ranges given represent the extreme values considering bo'h the range of the collective dose estimates and the range of risk estimates given in the 1972 BEIR report.
These values are extremely small compared to either the existing annual incidence of simi-lar effects or the potential effects estimated to result from the natural background radiation.
Comparing the total potential health impact of the acci-dent with the estimated lifetime natural risk indicates that these effects, if they were to occur, would not be discernible.
The uncertainties in the risk from low-level ionizing radiation would not alter this conclusion.
E.
Potential Added Risk to Maximum Individual The added lifetime risk of fatal cancer to the hypothetical maximum exposed individual from the accident is 2.0 x 10-5 (0.00002).
This is based upon a presumed 100 mrem dose rather than the estimated values.
This added risk is extremely small (0.013%) compared to the normal risk (0.15) to an individual 1377 126
Table 4-7.
PROJECTED POTENTIAL llEALTil IMPACT OF Tile TilREE MILE ISLAND ACCIDENT TO Tile OFFSITE POPULATION WITilIN 50 MILES Effect Estimated Potential Impact of Potential Lifetime Impact of Number who Natural Background Population Dose from the THI Accident would normally Radiation from March 28, 1979 through April 7, 1979 develop effect I
ID)
Range ")
Central Estimate Fatal Cancers 325,000(c) 1,700 - 9,000(d) 0.15-2.4(e)
- 0. 7 Non-Fatal Cancers 216,000(I) 1,700 - 9,000(d,g) 0.15 - 2.4(C'#)
0.7 Genetic Effects first generation 78,000(h) 60 - 970(I)
(0.01 - 0.64)(5) all future generations 0.05 - 4.8(k) 0.6 II)
All llealth Effects 0.4 - 10 2.0 Footnotes next page N
~sj N
Footnotes for Table 4-7 (a) This represents the extremes of the range of health effects estimates considering both the range of the collective dose estimates and the range of in the estimates of the risks of low-level ionizing radiation as estimated in the 1972 BEIR Report (3).
(b)
The central estimate is based upon taking the geometric mean (square root of the product) of the upper and lower bounds of the dose-to-health-risk conversion factors and multiplying this by the mean estimate of the population dose (3,300).
(c) Based upon the American Cancer Society projection that the risk of cancer death is 0.15 (0.15 x 2,164,000
= 324,600).
(d) Based upon multiplying the annual rates in Table 4-5 by 70 years, the mean life span.
Baseduponmultiplyingthelowerrangeestimateofthepopulag)iondose(1,600 person-rem)bythelower (e) range of the absolute radiation-induced cancer risk (90 x 10 and the upper range estimate of the population dose (5,300) by upper range of the relative radiation-induced cancer risk (460 x 10 6).
(f) Based upon the difference between the American Cancer Society projection of the risk of getting cancer o,
(0.25) and the risk of dyin0 of cancer (0.15).
The value given is the product of this difference 62 (0.25 - 0.15 = 0.10) and the size of the population (2,164,000).
(g) Based upon the assumption that there are twice as many cancers as there are canceg fatalities.
(h) Based upon the natural annual incidence of genetic effects (1,200 per year per 10 population) from table 4-2 times an assumed reproductive period of 30 years.
(i) Based upon multiplying the risk to the first generation from table 4-2 by an assumed reproductive period of 30 years and by the natural back round dose rate of 270,S00 person-reg)per year.
0 (j)
Based upon multiplying the lower bound of first generation risk (7 x 10 from Table 4-2 by the lower bound of the collective doge estimate (1,600 person-rem) and multiplying the upper bound of the first generation risk (120 x 10
) from Table 4-2 by the upper bound of the collective dose estimate (5,300 person-rem).
The first generation risk is included in the risk to all generations and therefore, should not be separately added into the total.
(k) Based upon the procedure described in (j) but using the equilibrium risk bounds rather than the first generation risk.
(1) This is done for the convenience of providing an estimate of the total potential health impact.
Tech-nically, the effects are not equivalent and cannot be added.
-J
~%
s N
CO
64 of dying from cancer.
It is also small (1.1 percent) compared to the potential lifetime fatal cancer risk that would be associated with natural background radiation using the same dose-to-health effect relationships as used for the accident impact.
F.
Dose Rate Effects The estimated maximum dose to a hypothetical individual (less than 100 mrem) is numerically approximately the same as the annual dose from natural background radiction to residents in the Harrisburg area (115-125 mrem /yr).
There has been some concern that, because this dose was delivered in 1 week instead of 1 year, the biological effects of this accident would be greater than from natural background radiation.
This presumes that radiation delivered at a higher rate is more dangerous than radiation delivered at lower rates (that there is a " dose-rate effect").
If there were such a " dose-rate effect," then the linear extrapolation of the number of effects observed at high doses and dose rates would over-estimate the risk per unit dose at low doses and low dose rates.
This is because the estimates of the health effects of low-level radiation are derived from observations made at much higher doses higher and dose-rates than experienced during the Three Mile Island Accident.
Existing estimates indicate that somatic
.}3ff k2
65 effects (cancer) might be overestimated at low doses by a factor of 2 to 4 (7, 8) and perhaps as much as a factor of 10 (8) and that genetic effects might be overestimated by a factor of 3 (8).
The estimates of the health impact of the Three Mile Island accident have not included any reduction factor to account for reductions due to a dose-rate effect.
(7) United Nations Scientific Committee on The Effects of Atomic Radiation,
" Sources and Effects of Ionizing Radiation - 1977 Report", UNSCEAR, United Nations, N.Y., N.Y. (1977), Annex G, p. 366, paragraph 36.
(8)NCRP Scientific Committee 40, " Influence of Dose and its Distribution in Time on Dose-Effect Relationships for Low-LET Radiation" Draft of February 21, 1979, page 3.
1377 130
66 5.
OTHER SOURCES OF EXPOSURE A.
Skin Doses and Health Risks from Beta and Gamma Radiation The contribution of beta particles from xenon-133 is not addressed by either the dose analysis in Section 3 or the health impact analysis in Section 4.
The principal reasons why this dose was not considered are:
(1) The range of beta particles (electrons) in air is short.
The maximum energy (0.35 MeV; average energy 0.12 MeV) of the beta particle from xenon-133, for example, has a maximum range of only 30 inches; therefore, an individual must be standing in or very near the radio-active plume to be exposed to beta radiation.
The time that any individual would be so exposed is not known.
(2) The beta radiation would be stopped by clothing or by being indoors.
Only small quantities of the gas would diffuse into the clothing or into a house compared to the concentration in the plume.
(3) At the present time, the sensitivity or response of the thermo-luminescent dosimeters to beta irradiation is not known (it is assumed to be zero).*
(4) The composition if the radioactive gases in the cloud is not well known for most of the locations of interest.
n If there were a significant beta dose contribution to the dose recorded by the thermoluminescent dosimeters (TL0s), then the total body dose estimated from the TLD readings would have to be reduced to allow for this non penetrating beta dose contribution.
The beta skin dose is estimated in this section from a theoretical ratio of the beta dose to the gamma dose.
If the " gamma" dose recorded by the TLDs is too high because it includes a beta dose contribution, then the beta skin dose estimated from the beta / gamma ratio would also be overestimated.
\\$11 \\3\\
67 4
(5) The principal health consequences of skin irradiation is skin cancer, which is not a predominant form of radiation-induced fatal cancer.
Although the beta dose cannot be assessed by direct measurement during the accident, it can be estimated from the technical literature.
The depth dose from xenon-133 electrons and beta particles decreases by a factor of 0.39 2
at a skin depth of 0.005 cm or 50 pm (an areal density in tissue of 0.005 g/cm ),
This depth is approximately the thickness of the non-living protective layer of s ki n. (' 1-2) The depth dose to internal organs from these beta particles is essentially zero.
The beta particle skin dose at the 50 pm depth, estimated 0
from the depth-dose calculations of Beijer(3) is 4.7 x 10 mrem /yr per pCi/cm 8
compared to estimates of the xenon-133 gamma-ray total body dose of 1.90 x 10 mrem /yr per pCi/cm,(4) or approximately a factor of 2.5 higher.
The gamma-ray 3
skin dose is 2.55 x 10 mrem /yr per pCi/cm,(4) Therefore, the combined beta 8
3 8
3 and gamma " skin" dose is 7.25 x 10 mrem /yr per pCi/cm, or a factor of 3.8 times the total body gamma-ray dose.
Table 5-1 provides the ratio of the beta (1)" Recommendations of the International Commission on Radiological Protec-tion Adopted January 27, 1977," ICRP Publication 26.
Pergamon Press, Oxford, England, paragraphs (63) and (64), p. 13.
(2) National Council on Radiation Protection and Measurements, " Krypton-85 in the Atmosphere - Accumulation, Biological Significance, and Control Technology," NCRP Report No. 44, National Council on Radiation Protection and Measurements, Washington, D.C., July 1, 1979.
Table 13, p. 30.
(3)M.J. Berger, " Beta-ray dose in tissue-equivalent material immersed in a radioactive cloud," Health Physics, vol. 26 (1): 1-12 (January 1974).
(4)D.C. Kocher, " Dose-Rate Conversion Factors for External Exposure to Photon and Electron Radiation from Radionuclides Occurring in Routine Releases from Nuclear Fuel Cycle Facilities," U.S. Nuclear Regulatory Commission Contract Report NUREG/CR-0494 (Oak Ridge National Laboratory Report ORNL/
NUREG/TM-283), April 1979.
1377 132
~.
68 plus gamma skin dose to the total body gamma dose for the principal radionuclides measured at offsite locations.
For the maximum exposed individual, the beta skin dose from xenon-133 would be about 325 mrem, if the individual were exposed in the plume out-of-doors without benefit of shelter or clothing for the entire period.
The 1972 report of the National Academy of Sciences' Advisory Committee on the Biological Effects of Ionizing Radiation (5) does not provide numerical estimates of the risk at los doses for skin cancers.
Skin cancers from radiation exposure reported in this report are associated with doses above 230,000 mrem in rats and above 450,000 mrem in humans.
This latter dose is sufficient to cause visible effects on the skin and is more than a factor of 1,000 greater than the estimated total (beta and gamma) skin dose to any exposed individual, even neglecting shielding by clothing or by being indoors.
The International Commission on Radiological Protection considers skin to be less likely to develop fatal cancers after irradiation than other tissues (1).
They recommend a lifetime occupational dose limit for skin of 2,000,000 mrem (1) or 5,000 mrem per year for members of the general public [(1) p. 25].
It is also significant that the ICRP has considered the organ at the highest risk (critical organ) for exposure to radioactive noble gases, such as xenon-133, to be the total body and not the skin or lung (6).
(5) Advisory Committee on the Biological Effects of Ionizing Radiation (BEIR),
"The Effects on Populations of Exposure to Lcw-Levels of Ionizing Radiation",
National Academy of Sciences - National Research Council (1972) pp 132-135.
(6)" Recommendations of the International Commission on Radiological Protection.
Report of Committee II on Permissible Dose for Internal Radiation," ICRP Publication 2, Pergamon Press, Oxford, England, 1959.
1377 133
69 Table 5-1.
RATIO OF SKIN DOSE TO TOTAL 800Y GAMMA DOSE Skin Dose (beta + gamma)
Total Body Gamma Dose xenon-133 3.44 xenon-133m 6.45 xenon-135 2.85 iodine-131 1.70 Note:
The skin dose is calculated to a depth of 70 pm, which corresponds to the average skin depth recommended by the International Commission on Radio-logical Protection.
The values for xenon-133 in the text are calculated to the minimum depth of 50 pm.
This difference is responsible for the difference between the factor of 3.8 for xenon-133 in the text and the factor of 3.44 given above.
The technical methodology used for this calculation is presented in Appendix C.
1377 134
70
-6 The ICRP(7) has recommended a fatal skin cancer risk value of 10 er 6
rem (1 per 10 person-rem).
This is in good agreement with risk values of
-6 0.5 x 10 / rem for fatal skin cancer obtained from data in the UNSCEAR Report (8),
as shown in Appendix D.
Because the skin dose from xenon-133 is a factor of 3.44 higher than the total body dose, the ratio of the total fata skin cancer risk from beta and gamma irradiation to the total risk of fatal cancers due to total body gamma irradiation would be:
3.44 skin dose
-6 skin cancer deaths 3
x 1.0 x 10 3.3 x 10 erson-rem total body dose person-rem
= 0.01 fatal skin cancer.
The ratio of fatal skin cancers to all skin cancers is apprcximately 0.06(8)
Assuming that this ratio (morbidity to mortality) is also true for radiation-induced skin cancers, the total number of skin cancers might be 0.01/0.06 =
0.2 (0.17).
Inhalation Lung Dose Radioactive noble gases irradiate the lung in two ways:
(1) from pene-trating gamma radiation from external sources and (2) by beta and gamma radia-tion emitted by radioactive gases inhaled into the lung.
As is the case with (7)Ir.ternational Commission on Radiological Protection, (to be added in final d aft)
(8) United Nations Scientific Committee on the Effects of Atomic Radiation (LNSCEAR), " Sources and Effects of Ionizing Radiation," 1977 Report, United Nations, N.Y. (1977), Annex G, Radiation Carcinogensis in man.
Section H.
pp. 411-412.
1377 135
71 skin irradiation by beta radiation, inhalation of radioactive gases can only occur when the individual is actually standing within the radioactive cloud.
Irradiation of the lung by gamma radiation from radioactive gases outside of the body can occur even though the individual is not actually within the radioactive gas cloud.
This contribution to the radiation dose to the lung is considered in the evaluation of the total body dose.
The potential health effects from this form of lung irradiation are also considered in the evalua-tion of the cancer risk from total body irradiation.
In that evaluation, it was assumed that the lung dose from external gamma radiation was equal to the total body dose.
More refined calculations show that, for xenon-133, the lung dose is about 25% less than the total body dose (4).
The dose to the lung from inhaled radioactive xenons is only a small fraction (2.8 to 7.3 percent) of the dose to the total body from external gamma radiation as shown in Table 5-2.
Because of the difficulty in determining whether or not and how long anyone was actually breathing the radioactive xenon gas, this dose contribution was not considered in the preceding section on health effects.
The risk of a fatal lung cancer per unit dose is about one-fifth (0.22) of the total fatal cancer risk (see Appendix 0).
Even if everyone within 50 miles were continuousi, breathing radioactive xenon-133, the total number of estimated fatal cancers would only be increased slightly.
The magnitude of this increase (for xenon-133 inhalation) wou.o be about a 1 percent increase 1377 136
Table 5-2.
RELATIVE CONTRIBUTIONS OF OTilER NOBLE-GAS DOSES COMPARED 10 Tile GAMMA TOTAL BODY DOSE FROM RADI0 ACTIVE XENON GASES Fraction of Gamma Whole Body Dose Beta Skin Beta Skin Lun0 Dose From Inhalation Internal Whole Body Dose Dose Dose to from Gases Dissolved in (surface) 1000 pm (1 mm)
Body Fluids in Equilibrium Radionuclide Beta Gamma Total Beta Gamma Total Xenon-133 2.71 0.00016 0.039 0.017 0.057 0.0045 0.002 0.0065 Xenon-133m 0.073 0.073 0.009 0.009 Xenon-135 1.83 0.165 0.027 0.0022 0.029 0.003 0.001 0.004 U
Derived from J.L. Russell and F.L. Galpin, " Comparison of Techniques for Calculating Doses to the Whole Body and to the Lungs from Radioactive Noble Gases," in Radiation Protection Standards:
Quo Vadis (W.P. Ilowell) and J.P. Corley, compilers), Proceedings of the Sixth Annual llealth Physics Society Topical Symposium, Richland, Washington, November 1971.
N N
N
\\
73 in the total number of fatal cancers (0.0566 x 0.22 = 0.012).
This contribution is small compared to the other uncertainties in the health impact analysis, especially considering the assumption that all individuals were totally immersed in the cloud.
Small quantities of noble gases may also be dissolved in body fluid (blood) and irradiate the body internally.
However, the dose contribution from this dose contribution is less than 1 percent of the total body dose as shown in Table 5-2.
C.
Airborne Radiciodine Concentrations and Associated Inhalation Doses Metropolitan Edison has, as part of its routine radiological environmental monitoring program, air particulate and radioiodine samplers in the TMI plant vicinity.
The NRC Office of Inspection and Enforcement also monitored the air in the vicinity of TMI for radionuclides including radiciodine.
To obtain an estimate of thyroid dose. it was conservatively assumed thzt a child was present at the Observation Center (" Trailer City") for the time the accident began until April 5, 1979.
The child thyroid dose was calculated using the methods and parameters given in references (_), pg. 29.
At two points during this time period, Metropolitan Edison measured a higher concentration of radioiodine at an offsite location other than the Observation Tower.
It was conservatively assumed that these concentrations existed at the Observation Tower for dose calculational purposes.
Only iodine-131 was used to calculate the dose since iodine-133 data were not 1377 138
\\
74 regularly available at the time of the calculation.
The data and doses are summarized below.
Location Time Iodine-131 Child Thyroid (0.5 mi SSE)
Period Concentration Dose 3
(0.5 mi SSE)
Period (pCi/m )
(mrem)
Observation Tower 3/21-28/79 0.30 0.094 (Met. Ed.)
3/29-31/79 20.
1.8 3/31-4/3/79
- 1. 4 0.19 Trailer City (NRC) 4/1 - 4/79
<0.90 0.12 4/5/79 1.6 0.071 TOTAL 2.3 0.
Thyroid Dose from Ingestion of Iodine-131 in Milk A large number of milk samples were collected during the period March 28 through April 4,1979 from farms and dairies throughout the area surrounding the accident site by the Pennsylvania Department of Environmental Resources, the Food and Drug Administration and Metropolitan Edison.
Aliquots of several of these were also analyzed by the Environmental Protection Agency.
A summary of the results is given below:
Metropolitan Pennsylvania FDA EPA Edison Number of analyses performed 133 106 4
21 Number of positive results 7
42 2
18 Average value of positive results (pCi/ liter) 15 17 17 7
Range of positive results (pCi/ liter) 11-20 13-42 10-24 1-41 Average minimum detectable concentration (pCi/ liter)
<20
<10
<10
<1 1377 139
\\
75 Additional milk samples collected between April 4 and April 17 were below detectable levels.
Five out of 72 samples were positive between April 18-20 with a range of 21 to 36 pCi/ liter.
A summary of all Food and Drug Administra-tion samples collected from March 30 to April 29 is given in Table 5-3.
Using the highest concentration of iodine-131 observed in any single sample of milk as the worst case, 42 pCi/ liter, the dose to the thyroid of an infant drinking 1 liter of milk from that source for the entire duration of the accident would be 5 mrem over the lifetime of the individual.
This is derived from the protective action guide that relates 12,000 pCi/ liter to a 1.5 rem dose to the thyroid.
Under these conditions, an adult drinking the same milk would receive a lifetime thyroid dose of 0.5 mrem, based on a thyroid weight 10 times greater than the infant (20 g versus 2 g).
Cesium-137 was also detected in some of the milk samples at levels generally less than 25 pCf/ liter.
The maximum reported level was 37 pCi/ liter.
The presence of this radionuclide is probably due to fallout produced from previous atmospharic testing.
Review of results from pasteurized milk samples analyzed for the previous year from Pittsburgh and Philadelphia by EPA show the presence of cesium-137 also for several samples during that period.
The levels were less than 12 pCi/ liter.
The Pittsburgh and Philadelphia samples represent milk samples composited from more than one source; the samples collected during the Three Mile Island incident represent specific farms and dairies which exhibit greater variability than composite samples.
1377 140
76 Table 5-3.
10 DINE-131 LEVELS IN MILK (pCi/ liter)
(FC00 AND DRUG ADMINISTRATION MEASUREMENTS)
Number of Positive Number of Positive Sample Date (1979)
Samples Samoles*
Min.
Max.
Range Average 03-31 10 4
13 42 29 25 04-01 19 14 13 36 23 20 04-02 29 13 14 30 16 18 04-03 24 5
16 25 9
20 04-04 24 6
13 22 9
18 04-05 26 04-06 26 04-07 28 04-08 19 04-09 26 04-10 20 04-11 23 04-12 23 04-13 25 04-14 21 04-15 18 04-16 30 04-17 25 04-18 23 2
21 24 3
23 04-19 30 2
19 29 10 24 04-20 22 1
36 36 0
36 04-21 17 04-22 18 04-23 23 04-24 26 04-25 28 04-26 34 04-27 30 04-28 19 04-29 16
- Minimum detectable concentration is 10 pCi/ liter.
1377 14i
APPENDIX A De31rtment of Energy (00E) Estimate of External Whole Body Radiation Exposure to the Population Around the Three Mile Island (TMI) Nuclear Power Station.
A population exposure estimate in the vicinity of the Three Mile Island Nuclear Power Station for the period March 28 through April 10, 1979 has been prepared.
It is based principally upon the average of measurements of the radia-tion levels in the plume made during helicopter flights, supplemented by plant meteorological and projections of the plume location and extent.
Subsequent to the earlier estimate of the population exposure, TLD data obtained by Metropolitan Edison (as supplied by NRC) made it apparent that a substantial portion of the exposure must have occurred during the first day or so after the incident, prior to the time that regular helicopter.neasurements were initiated.
A projection of the probable dose rate in the plume during this interval has been made from the many measurements obtained during the period March 30 through April 9.
Since it sas established that the principal radiogas in the cloud during this period was xenon-133, its concentration was estimated from the measured dose rate and then extrapolated back for the period March 28 through i377 142
s A-2 March 29.
A dose rate for this early period was than projected by considering the dose attributable to the shorter-lived radiogases which should have been present (assuming that there was an equilibrium mixture of fission product gases present at the time of the incident).
The assignement of ground level exposure rate has been made by specific sector and hour after incident during the first 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />, and by average dose rate for each daily interval thereafter.
An exponential relationship of plume dose rate with distance was cbtained using the helicopter measurements.
This 1
2 was compared to curves with 1/R, 1/R.5 and 1/R behavior.
The exponential relationship leads to a smaller dose estimate than do these other curves.
The available TLD data for stations which were at varying distances in a given sector 2
also suggest a rapid decrease of exposure with distance, consistent with 1/R or the observed exponential relationship.
The DOE assessment of the external whole body radiatica exposure to the population around the Three Mile Island (TMI) nuclear power station was based on over 200 aerial radiation measurements taken in the center of the plume of airborne discharges.
These measurements were taken from helicopters, using Geiger-Mueller survey instruments with probes having open, low density windows, to enable measurements of the gamma radiation exposure, plus any contribution from high energy beta radiation.
The radiation survey probe was held external to the helicopter (s) to minimize attenuation of any radiation.
The measurements were made at various distances out to 20 miles from the TMI plant.
At each 1377 143
t A-3 distance, the helicopter (s) were maneuvered to find the maximum radiation exposure rate, and this maximum value was used in the calculation of popula-tion dose within any sector.
The geographical region within a 50-mile radius of the plant was plotted out in concentric and azimuthal sectors, and the population exposure within each annular segment was calculated based on the measured radiation dose rates, records of the helicopter location for each measurement, the path of the plume, the duration of its passage, predictions of its course and speed from current meteorological data, and population figures for each segment projected for the 1980 census.
A factor of 2 reduction was made to account for the aircraft being within the cloud and irradiated both from top and bottom, whereas a person on the ground is only irradiated by material above the ground surface.
The estimate assumes that members of the population were out-of-doors during the entire dura-tion of passage of the plume.
The exposure rates at distances beyond 10 miles from the plant were extrapolated from a curve drawn through the exposure measure-ments measured as a function of distance within 10 miles of the plant.
Exposure rates beyond 10 miles were generally too low to measure.
Figures A-1 and A-2 show exposure profiles for the 0-2 mile and the 0-10 mile radii, respectively, for the average exposure to individuals on the ground.
The collective dose to external radiation within the 50-mile radius using the above data and assumptions was approximately 2000 person-rem (+ 500 or -
1,000 person-rem) through April 3, 1979.
00E estimates that the increase in 1377 144
(
A-4 population dose for the period April 3 through April 10, 1979 to be a total of approximately 50 person-rem.
Table A-1 provides the contributions to the collective dose from each of the population sectors.
The maximum estimated dose was 200 t 50 mrem to an individual located about one mile north-northwest of the station continuously for the entire week following the TMI occurrence.
This location corresponds to Hill Island.
This assessment overestimates the actual exposure because of the following conservative assumptions:
(a) No reduction of the radiation exposure was made for shielding of individuals during periods they would be inside.
(b) Some of the helicopter flights and radiation dose measurements were made in response to known increases in discharges from the plant.
Therefore, they would be higher than average values.
(c) The maximum doses measured in the plume were applied to the entire sector affected.
(d) An expected significant over response of the Geiger-Mueller Survey instrument.
1377 145
i A-5 Table A-1 contains the Department of Energy collective dose results obtained from their surveys.
Table A-2 contains their estimates of individual integrated exposures at specific locations.
1377 146 s
e
A-6 Table A-1.
Collective Dose to Population 0-50 miles from Three Mile Island Nuclear Station March 28 through April 3, 1979 (Department of Energy Aerial Radiation Survey)
Radius Total Average Individual (Mile)
Person-Rem **
Population
- Exposure (mR) 0-1 51.2 658 77.8 1-2 66.7 2,017 33.1 2-3 482.2 7,579 63.3 3-4 352.2 9,676 36.4 4-5 76.4 8,891 8.6 5-10 810.0 137,474 5.9 10-20 137.4 577,288 0.24 20-30 27.3 433,001 0.063 30-40 1.9 273,857 0.0069 40-50 0.3 713,210 0.00048 TOTAL 2,005.7 2,165,651 0.92 (2,000)
(0.9) xEstimated population for 1980, by 22.50 sectors and distance obtained from FSAR for Three Mile Island II.
- Based on projected ground level dose rates under the plume of radioactive gas, which were assumed to have been one-half of those found during the helicopter flights within it.
Table A-2 ESTIMATES FOR INDIVIDUAL INTEGRATED EXPOSURES EVALUATED AT SPECIFIC LOCATIONS Estimated Individual Exposure Location Dist. Dir.
(mR)
Goldsboro 1.5 W 15 3
Falmouth 1.7 SE 10 2
Middletown 3
N 10 2
Harrisburg 10 NW 2
0.5 1377 147
A-7 t
50 10 0 t00 50 g
s
- s
.K s
\\ % $~ O
/
i u
)
%)
\\f' 6
i s
so ;
\\\\\\
==.m
\\
x
/
i;; lid 'in9P "*
i
\\'
5 l.5 M t.
A-8 O.5 f.0 l.0 O'S O.
0 i
e i
\\
I i
1 C.I 4
\\
po 3.,
\\
s 1
l.0 J
0.5 d3
=
9,g
- ";4Rign;"%
"'-^ i;;iiV'ii99P "#
1377 1 0
4 a
APPENDIX B Department of Energy (00E) Environmental Deposition Measurements in the Area Surrounding the Three Mile Island Nuclear Power Station Following the accident at the Three Mile Island Nuc! ear Stations, the 00E established the follcwing environmental monitoring activities starting as of 4:00 p.m. on March 28, at the request of the Commonwealth of Pennsylvania, in accordance with the DOE Radiological Emergency Assistance Program:
(a) Helicopter surveys to locate and measure gamma and beta radiation in the airborne discharges.
(b) Ground vehicle radiation surveys in the path of airborne discharges, including some in-situ radionuclide identification by gamma spectrum analysis.
(:) Collection of environmental soil, grass, surface water, and air samples in the path taken by airborne discharges.
(d) Gamma spectrum analyses of these environmental samples to dete.t, identi fy and quantify any radionuclides present.
1377 150
B-2 (e)
Evaluation and interpretation of survey and analytical data to estimate population exposure.
00E established three field laboratories for analyzing samples of soil, surface water, grass, and air for gamma-emitting radionuclides.
These labora-tories were located at the Capitol City Airport.
Each utilized a sensitive, high efficiency lithium drifted germanium detector and multi-channel gamma spectrum analyzer.
One set of each was brought in and manned by radiochemists from the Brookhaven National Laboratory, Bettis Atomic Power Laboratory, and Knolls Atomic Power Laboratory.
Environmental samples were collected by crews from these laboratories, with specific attention to locations near the plant, and to areas over which the plume of discharges from the plant had persisted, and was known to have touched down.
Attention was also given to assuring that the sampling method would establish if any radioactivity from the plume had been deposited on the ground.
The soil, grass and water specimens were skimmed from the largest surface areas practicable to fill Marinelli geometry containers in order to optimize the sensitivity of the analyses, and thereby increase the likelihood of detection.
The air samples were taken boti' by silver-treated silica gel samplers flown into the plume to ensure capture of any non-ionic radiciodine present.
Charcoal filters were used in ground sampling larger voltmes of air in the plume.
The total number of samples collected and analyzed starting on March 29 has been in excess of 800.
The detection sensitivity achieved (minimun e
1377 151
l B-3 2
detectable activity (MDA)) for iodine-131 was less than one nCi/m for soil
-12 and vegetation, 4 x 10' pCi/ml for water, and 3 x 10 Ci/ml for air.
Even lower MDA's were achieved on many samples by longer counting periods, by further idealizing of geometry, and when background radiation was lower.
These measures 2
enabled sensitivities as low as 0.5 nCi/m for soil, 0.02 nCi/m for grass
~8 and 1 x 10 pCi/ml for water.
The gamma spectrum measured for each sample was examined in its entirety to detect any photopeaks.
The detection sensitivity of this equipment was sufficient to reveal any uranium in the air in the range of allowable occupational concentrations, if any had been present.
The analyses of these environmental samples revealed the presence of iodine-131 in only a few air and grass samples, at barely over the detection limit, when the greater sensitivities were achieved.
In a few soil samples, cesium-137 radioactivity was detected as expected at levels normally fcund due to world-wide fallout from previous atmospheric testing.
Table B-1 summarizes these analytical results.
The silver-treated silica gel air samplers which had been flown through the plume, and the char:oal air sample filters used for the high volume ground level samples in the path of the plume, were returned to Brookhaven National Laboratory for further analysis to detect the presence of beta, or alpha emitters by other techniques.
However, such species are considered entirely unlikely since the properties of the chemical species in which such radionuclides 1377 152
4 B-4 Table B-1
SUMMARY
OF ANALYTICAL RESULTS FOR IODINE-131 IN SAMPLES COLLECTED AND ANALYZED BY DOE No. of No. of No. of Range of Samples Samples Samples Positive Sample Collected less than greater than Values Type MDA*
MDA*
Period Stagnant from Surface Water 122 122 0
3/28-Rain Water 0
0 0
2 4/6 Vegetation 236 234 2
0.1-0.3 nCi/m 2
Soil 235 224 1
0.3 nCi/m
-12 Air 19 11 8
7 x 10 to
~11 3 x 10 pCi/cc Period Stagnant from Surface Watar 60 60 0
4/7 -
Rain Water 17 17 0
2 4/16 Vegetation 78 69 9
0.05 to 0.7 nCi/m Soil 27 27 0
-12 Air 23 11 12 6 x 10 to
~11 9 x 10 pCi/cc
- Minimum detectable activity (concentration) 1377 153
B-5
/
exist are known to promote retention within the reactor fuel and/or coolant.
Containment air samples analyzed on March 30 did not reveal the presence of any such nuclides.
Direct in-situ measurements of radioactivity on the ground were also made by the DOE Environmental Monitoring Laboratory (EML) using two large volume, pressurized ionization chambers, and a very sensitive, high efficiency Lithium drifted Germanium detector gamma spectrorater.
These systems enable detection of variations in radiation levels from natural or man-made radioactivity of a fraction of a microrcentgen per hour.
These vehicle mounted systems were deliberately moved to locations where those few environmental grass samples were taken which, when analyzed in the laboratory indicated iodine-131 at con-centrations just above the MDA.
These EML measurements confirmed both the concentrations measured in the laboratory, and the identification of the specific radionuclide iodine-131.
Other measurements by the EML systems also confirmed the generally negative results found in the laboratory analyses of the environ-mental soil, water and grass samples.
The date, time and specific location of all of the environmental samples, as well as the results of the laboratory analyses are recorded in the Technical Work Record books of the DOE team.
The results of these analyses of the environmental samples, as well as gamma spectrum analyses of the plume made by the EML mobile system, support the conclusion that the predominant radionuclide in the airborne discharges
)377 154
f B-6 was the inert gas xenon-133, with a small amount of iodine-131 also present.
This conclusion is supported by information received from the NRC licensee (Metropolitan Edison) concerning the measured composition of stack discharges, and the analyses of the air radioactivity in the containment.
1377 155 s
T~
APPENDIX C EVALUATION OF SKIN DOSE FACTORS Dose Factor (mrem /yr per pCi/cc )
Radionuclide Total Dose Rate at Body Electron Effective Skin Dose Surface (
)
Depth Dose Factor (C)
(70 pm)
I9)
Id)
Id) Electron (*)
Photon (d'#)
Total Photon Electron 50 pm 70 pm Xenon-133 1.90E08 3.86E08 1.22E09 0.392 0.327 3.99E08 2.SSE08 6.54E08 Xenon-133m 1.69E08 1.66E08 1.28E09 0.753 0.667 8.5,4E08 2.38E08 1.09E09 Xenon-135 1.41E09 2.lSE09 2.83E09 0.869 0.823 2.33E09 1.69E09 4.02E09 Iodine-131 2.12E09 3.31E09 1.71E09 0.750 0.692 1.18E09 2.SSE09 3.73E09 8
Note:
3.86E08 = 3.86 x 10.
Divide by 24 hrs / day x 365.24 days /yr = 8,766 hrs /yr to get hourly dose rates in mrem /hr per pCi/cc.
r3 (a) Values from D.C. Kocher, " Dose-Rate Conversion Factors for External Exposure to Photon and Electron Radiation from Radionuclides Occurring from Routine Releases from Nuclear Fuel Cycle Facilities" Dak Ridge National Laboratory Report (0RNL/NUREG/lH-283) prepared for the Nuclear Regulatory Conunission, NRC Report NUREG/CR-0494 (April 1979)
(b) Kocher, Appendix C, p. 94.
3
-4 2
(c) Ratio of depth dose at z = thickness x 1 g/cm x 10 cm/pm to z = 0.000 g/cm from M.J. Berger,
" Beta-Ray Dose in Tissue Equivalent Material Immersed in a Radioactive Cloud," llealth Physics, 26(1):
pp. 1-12 (January 1974).
Tables 6 and 7 using values with leakage correction.
(d) International Conunission on Radiological Protection (ICRP Publication No. 26, "Reconunendations of the International Conunission on Radiological Protection adopted January 17, 1977."
Pergamon Press, Oxford (1977) paragraph (64) p.
13.
The 50 pm value is reconunended as the average tSickness of the outer " protective" layer of the skin and 70 pm is reconunended as the average depth to be used for evaluating skin dose.
]
(e) Product of body surface electron dose factor and depth dose factor for 70 pm thickness.
(f) From Kocher Appendix C p.
106.
y (g) Sum of two preceeding values.
y
APPENDIX 0 Estimated Risk of Specific Radiation Induced Cancers Based on the UNSCEAR 1977 (
), Annex G Estimated General Population Mortality Estimated Absolute Risk Risk from 1977 UNSCEAR Report (
)
cancer Type Population Cases Per 106 Person-rem Deaths per 106 Person-rem Breast Adolescent Women 440 (36-1500) 30 (20-35) (a)
(pp 385-394)
Women (all ages) 180 (140-230) tung Adult Males 50 (20-150) 50 (20-150) (b)
(pp 394-399)
Skin Adults 5 (2-10) 0.5 (0.2-1.0) (c)
(pp 411-412)
Ihyroid 100 (50-150) 10 (5-15) (d)
(pp 377-385)
Leukemia (pp 370-377)
Adults 25 (15-30) 25 (15-30) (e)
Bone Cancer (pp 399-401)
Adults 3 (2-5) 3 (2-5) (e)
Brain Cancer (p 406)
Fetus 50 (neg-145) child 20 (9-39) 20 (9-39) (e)
Salivary Glands (pp 406-407) child 10 (5-20) 5 (3-10) (f)
Sinus Mucosa 3 (2-5) 3 (2-5) (e)
Digestive Organs 12 (10-15) (e)
Estimated Total Risk 450 (400-500) (f) 230 (200-250) (f)
(pp. 413-414)
(a) Assumes 30% mortality and 50% of the general population is female (b) Assumes 100% mortality and equal risk for women g
y (c) Assumes 10% mortality N
(d) Assumes 10% mortality (e) Assumes 100% mortality (f) Assumes 50% mortality Ln N