ML20056H174

Preliminary Assessment of Potential Human Exposures to Routine Tritium Emissions from Yankee Atomic Electric Co Nuclear Power Facility Located Near Rowe,Ma

ML20056H174
Person / Time
Site:	Yankee Rowe
Issue date:	06/10/1993
From:	Lung S, Macintosh D, Tsai F HARVARD SCHOOL OF PUBLIC HEALTH, BOSTON, MA
To:
Shared Package
ML20056H171	List: ML20056H170 ML20056H174
References
NUDOCS 9309080329
Download: ML20056H174 (29)

Text

O A PRELIMINARYASSESSMENT OF POTENTIAL HUMAN EXPOSURES TO ROUTINE TRITIUM EMISSIONS FROM THE

(

YANKEE ATOMIC ELECTRIC COMPANY NUCLEAR POWER FACILITY' LOCATED NEAR ROWE, MA Prepared by I

D. Macintosh, S. Lung, E Tsai, and J. Spengler Harvard University School of Public Health Department of Environmental Health i

665 Huntington Avenue Boston, Massachusetts 02115 June 10,1993 l

9309080329 930812 0 PDR ADDCK 05000029 P

PDR.

a...- -. -

-. - -. ~ - -

\\

?

i TABLE OF CONTENTS l

1 i

EXECUTIVE

SUMMARY

i

1. INTRO D UCTIO N...............................................

1 i

2. METHO DO LOGY............................................... 4

~

2.1 Estimating Environmental Concentrations...........

4 2.2 Human Exposure Factors and Dose Calculations................... 9 2.3 Source Term Data.........................................

11

3. R ES ULTS.....................................................

13 3.1 Environmental Concentratiors................................

13 3.2 Human Exposures and Doses................................

15 i

4. D I S CU S SI O N..................................................

17_

5. CONCLUSIONS AND RECOMMENDATIONS........... -.............

23 l

t l

6. REFERENCES.................................................

25 l

[

I l

l L

e i

- :r

r EXECUTIVE

SUMMARY

A preliminary assessment was conducted of potential exposures and doses to Deerfield River Valley (DRV) residents of tritium,3H, released from the Yankee Rowe nuclear power facility located near Rowe, Massachusetts. The objective was to explore the upper bound of potential human health impacts by modeling exposures and doses using a variety of simplifying and conservative assumptions. The spatial range of this assessment was limited to the DRV and extended from the Sherman Dam to the town of Greenfield, approximately 60 km downriver from the Yankee Rowe. The temporal scope extended over a 30 year period, approximately equivalent to the operational life of the facility.

3 The assessment assumed all H to be present as a component of water molecules, identified as HTO. The exposure pathways considered were direct ingestion of tritiated water, inhalation of HTO released to the air from the power plant and evaporated from the river, and dermal absorption of HTO from the air. Enhanced evaporation of HTO from the river due to aerosolization was not considered, becaase of our current lack of 3

information regarding this phenomenon. Bioconcentration of H in food webs has not been observed in the field (NCRP 1979) and therefore was not considered.

The exposure scenarios used in the assessment were designed to provide a conservative exposure estimate for the pathways considered. Hence, the following primary assumptions were made; an atmospheric inversion that caps the DRV and limits vertical dispersion of HTO was present at all times, exposed indiviuals were present in the valley at all times, indoor air concentrations were identical to outdoor air i

concentrations, and all exposed individuals met 100% of their daily water needs from the Deerfield River.

Several preliminary conclusions were reached the analysis. First,3H concentrations in DRV environmental media during the operational phase of the Yankee Rowe facility may have been several orders of magnitude greater than those in typical

{

unimpacted areas. Second, given that few DRV residents obtain their water from the 3

Deerfield River, the dominant source of H exposure is inhalation and dermal absorption of airborne HTO as a result of its evaporation from the river. Third,3H exposures from intermittent gaseous releases from Yankee Rowe were probably insignificant compared to river related exposures. And finally,3H doses resulting from Yankee Rowe operations i

4 and delivered as HTO were probably several orders of magnitude lower than those from all natural sources of radiation.

This analysis contains a number of simplifying assumptions which should be examined for their relevance and applicability. Specifically, an investigation should be 3

undertaken of the potential for some of the H to be present in an organic molecule rather than solely as HTO. In addition, the effect of river rapids and falls on HTO evaporation, and possibly organically bound H aerosolization, should also be researched.

3 Finally, because adverse health outcomes for some specific individuals have been i

i 3

anecdotally associated with the Yankee Rowe facility, the potential H doses to these 4

persons should be reconstructed using detailed historics' release data and personal histories.

l I

ii

i i

1. INTRODUCTION The original objective of this study was to review the meteorology of the Deerfield River Valley (DRV) and use the information to estimate potential exposures of the local residents to tritium (3H) released by th'e Yankee Atomic Electric Company nuclear power plant located near Rowe, Massachusetts (hereafter referred to as the Yankee Rowe). Summary data on meteorological conditions collected during 1959 and 1971 were made available to us by the Citizens Awareness Network (CAN) and reviewed. In general, the data indicated wind directions and speeds in the valley are relatively constant and thermal inversions are rather common, occurring approximately 20 to 30% of the time. Thermal inversions are of interest for exposure assessments because they tend to trap pollutants and lead to elevated concentrations in the lower atmosphere. In addition, anecdotal reports from DRV residents indicate that daily episodes of morning fog occur (that may or may not be related to diurnal inversions),

which present another potential exposure route.

The spatial area of interest as presented to us by CAN representatives included I

the DRV extending from the Yankee Rowe to Greenfield, as well as communities located outside the confines of the valley. As discussed in greater detail later, the broad i

i n

spatial area of interest and the extremely complex terrain in the DRV precluded the use of standard Gaussian dispersion models which are commonly used to estimate pollutant concentrations and receptor exposures downwind from point sources. Consequently, we 3

decided to model H dispersion, concentrations, exposures, and doses, based upon a more simplistic set of modeling assumptions.

1 I

r

t 4

A preliminary assessment was conducted of potential exposures and doses to 3

DRV residents of H released from the Yankee Rowe facility. Our objective was to explore the upper bound of potential human health impacts by modeling exposures and doses using a variety of simplifying and conservative assumptions. Therefore, the results of this analysis should be considered only as an indicator of past conditions in the DRV rather than a complete model of the complex environmental situation.

The Yankee Rowe facility is a pressurized water reactor located on the Deerfield River just below the Sherman Dam. The facility was in operation from 1961 through 3H was released 1992. The U.S. Environmental Protection Agency (1971) reported that from the facility in batches ofliquid and gaseous effluent. Batch releases were reported as occurring approximately once per year for gases and 5 times per month for liquids.

All tritium was assumed to be present as tritiated water (HTO).

The spatial range of this assessment was limited to the actual river valley and extended from the Sherman Dam to the town of Greenfield, which is approximately 60 km downriver from the Yankee Rowe. The temporal range of the assessment extended over the life of the plant. The source term data used were provided by the CAN. The exposure estimates assume that all individuals reside within the valley 100% of the time.

The exposure pathways considered are shown in Table L In accordance with the scoping goals of this study, worst case exposure scenarios -

were constructed for each of the above pathways (for example, the valley was assumed to be continually capped by a thermal inversion). However, some exposure conditions and pathways were not considered due to information gaps and time constraints present at 2

Table 1. Exposure pathways considered in this analysis.

Inhalation direct inhalation of gaseous HTO releases

. inhalation of HTO evaporated from the Deerfield River Ingestion direct ingestion of HTO from the Deerfield River

~

Dermal Absorption

. absorption of HTO resulting from gaseous releases absorption of HTO evaporated from the Deerfield River l

this time. Specifically, enhanced evaporation of HTO resulting from aerosolization 3

and the potential for H to be organically bound (e.g., a component of a protein) were I

not considered.

The following section describes the methods, assumptions, and data used to model i

HTO exposures and doses to DRV residents. The third section presents the results of our analysis, followed by a discussion of their interpretation and implications with respect 1

to other sources of enviromnental radiation exposure and current uncertainties and information gaps. The final section contains our conclusions and recommendations for additional research.

)

3 j

2. METHODOLOGY 2.1 Estimating Environmental Concentrations Gaseous Releases. The transport, dispersion, and receptor-specific concentrations i

resulting from pollutant releases ~to the atmosphere are typically modeled with Gaussian t

air dispersion models. These models exist in many forms to suit applications in different L

environmental settings (e.g., homogeneous valleys) and release scenarios (e.g., continuous or short-term). The accuracy of Gaussian dispersion models is related to the degree to 4

which actual environmental conditions correspond to the many simplifying and necessary model assumptions. For example, the standard Gaussian model assumes relatively flat or homogeneous terrain, constant meteorological conditions (such as wind speed, wind direction, and atmospheric stability), and the receptors are not too close (e.g., less than l

100 m) or far (e.g., greater than 40 km) from the source (Turner 1970).

t The actual conditions in the DRV are very different from those required to apply j

a Gaussian dispersion model with confidence. For this reason, direct releases to the atmosphere were modeled in a plug-flow manner. Plug-flow refers to a system of fixed volume and flow rate into which a pollutant is introduced, uniformly distributed in a pre-determined mixing or dilution volume, and transported without loss or dispersion from the beginning to the end of the system. A schematic of the plug-flow system used as a j

model of the DRV is shown in Figure 1.

For our purposes, the plug-flow system is represented by the DRV, described as a l

three-dimensional trapezoid of the following dimensions: length,60 km; width at mid-t height,1.45 km; height,0.26.km. The length was determined from United States d

P 4

?

pigure 1 Schematic 0i dry p\\ug'tioW model (not to scale)-

Yankee AIO 60 km

.,2 a.

i i

i i f

i Greentie\\d 260in

' c :'. _.,-_.. c_.

Dilugon Voiume

%50 m Deettieid giver Total s/oiume " 60 km x 145 km A' 0.26 km = 23 Km^3 4

t i

Geological Survey (U.S.G.S.) topographic maps of the area and represents the distance along the river from the Sherman Dam to the city of Greenfield. The width was based upon the average distance between the valley walls near the Yankee Rowe. The height l

was based upon the assumption that the valley is capped by an atmospheric inversion at all times, and represents the valley depth near the power plant. The dilution volume was

+

based upon an arbitrarily assumed release time of 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> and a typical wind speed (2 m/sec) measured by Austin (1959) in a meteorological survey of the area. The HTO concentration in the plug, [HTO] air,was calculated as:

i Q,)t

[HTO]* =

(1) l uWH i

where, j

3

[HTO] air = HTO concentration in plug of air (Ci/m )

Qair = annual tritium release (Ci) i t = release duration (sec) u = wind speed (m/sec) j

+

W = width of valley at mid-height (m)

H = height of valley (m)

Although the release time and wind speed assumed here were derived arbitrarily, they

~!

have no bearing on estimated annual exposures from gaseous releases because concentration in the plug and release time and wind speed are perfectly inversely related.

{

l Aqueous Releases. Liquid eff.uent from the Yankee Rowe were modeled as occurring continuously. This simplification is felt to be justified because the~ effluent first I

become mixed in the lower Reservoir before continuously flowing over the lower 6

1

[

y i

- ~.

e Reservoir Dam. Liquid HTO releases were assumed to be immediately mixed in the i

Deerfield River using the following equation,

[HTO]mo "

G)

F

where, 3

[HTO]H20 = steady-state tritium concentration in river water (Ci/m )

OH20 = annual tritium release to the river (Ci/yr) 3 F = Deerfield River annual flow rate (m /yr) 8 3 The flow rate of the Deerfield River was assumed to be 1.2 x 10 m /yrbased upon data collected by the U.S.G.S. from 1940 through 1985 at West Deerfield, MA and provided by the CAN.

Evaporation. Evaporation of HTO from H O was studied under laboratory and

]

2 controlled field conditions by Horton et al (1971). They found that the rate of HTO i

evaporation was independent of the moisture content (i.e., relative humidity) of the receiving air. HTO evaporation was found to be well described by the classic two-film model. In this model, evaporation occurs at a rate proportional to the HTO i

concentration gradient between the water body and the atmosphere. The evaporation rate constant (B) for HTO was experimentally determined to be approximately 1.8 x 10-5 cm/sec. The evaporative model can be modi 5ed to represent a number of physical situations including standing bodies of water, batch discharges, and continuous flow through a basin. The latter case best describes the situation in the DRV in which we are modeling ITFO evaporation from the Deerfield River as it flows through the 60 km basin 7

i I

f from the Sherman Dam to Greenfield. The equation that describes HTO. evaporation during continuous flow through a basin is,

= c&

C (3)

C,

where, 3

C = [HTO] in the water at the mouth of the flow-through basin (Ci/m )

3 C = [HTO]H20 n the water at the head of the flow-thmugh basin (from Eq. 2) (Ci/m )

i o

B = proportional evaporation rate constant (6.4 x 10-6cm/sec) 2 A = surface area of the basin (cm )

3 F = flow rate of the river through the basin (cm /sec) 6 2 The surface area of the basin, A, was estimated to be 6 x 10 m from U.S.G.S.

topographic maps of the DRV and the flow rate, F, was set equal to the same value as 8 3 6

3 before,1.2 x 10 m /yr (3.8 x 10 cm /sec). The left-hand side of Eq. 3 represents the i

fraction of HTO remaining in a parcel of water following its travel from one end of the basin to the other. The complement of the left-hand side,1 - C/C, represents the o

fraction of HTO in the water that evaporated to the atmosphere. The amount of HTO that evaporated (HTOeyp) can then be approximated as the product of 1 - C/C and the o

amount of HTO originally released to the water.

The HTO concentration in air resulting from evaporation, [HTO]eyp,was calculated assuming that the amount evaporated, HTO,yp,was uniformly mixed in the i

3 total volume of the DRV system shown in Figure 1.(23 km ). The evaporated HTO concentration was assumed to be constant at all times during the life of the power _ plant.

8 f-

.l 2.2 Human Exposure Factors and Dose Calculations j

l Inhalation. Inhalation exposures and doses were modeled separately for annual

\\

gaseous releases of HTO from the plant and constant concentrations as a result of

, evaporation from the Deerfield River. However, in both cases indoor and outdoor HTO concentrations were assumed to be equal. Inhalation rates for individuals (Ih) were 3

assumed to be 20 m / day (EPA 1989).

Acute exposures frcm the annual gaseous releases were modeled as occurring l

instantaneously, which is a reasonable simplifying assumption given the short exposure period (assumed to be 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />) relative to the 123 day biological half-life of HTO in the human body (Shapiro 1990). Acute inhalation exposures to HTO can be calculated by determining the cumulative radioactivity in the body over the period du-ing which the HTO is present, according to the following equation, HTOEXP, = l [HTO }e ~" dt (4) o 0

where, HTOEXP, = cumulative HTO exposure resulting from a short exposure period (Ci-sec)

HTOo = the amount of HTO instantaneously inhaled (Ci), calculated as [HTO]airx Ih 3

(20 m / day) x Exposure Time (0.125 days or 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />) k = biological elimination rate of HTO from the body (1/ day) (0.693/12.3 days) t = the variable of integration 3

Eq. 4 ignores the radioactive decay of H because the biological half-life is much shorter than the 12 year physical half-life.

9 1

l t

Chronic inhalation exposures as a result of HTO evaporation from the Deerfield River were assumed to occur over a continuous 30-year period (the approximate lifetime of the plant). Chronic inhalation exposures to HTO can be calculated as, HMEXP, = [HTO}, x Ih x t x Exposure Period (5) um

where, HTOEXP = HTO exposure resulting from a chronic exposure (Ci-sec) c 3

[HTO] air = constant HTO concentration in air (Ci/m )

3 Ih = inhalation rate (m / day)

HTo = residence time of HTO in the body (1/k) (days) t 8

Exposure Period = length of chronic exposure (30 years = 9.5 x 10 sec)

Organ-specific HTO doses (rem) from exposures (Ci-sec) can be calculated as, t

i 1 rad

,," *, lam (gy,

• IO E O'0IO

6. 4x10 MeV,

1 Dose = H70EXP x x

x Cisec H70 dis g tissue 43,000 g H2O 1 rad

where, Dose = organ specific radiation dose from HTO (rem)

HTOEXP = the HTO exposure (Ci-sec) 10 3.7 x 10 dis /sec = the definition of a Ci of radioactivity 3

0.018 MeV = the average energy of H beta disintegrations (Shapiro 1990) 7 6.24 x 10 MeV/g tissue = the definition of 1 rad of absorbed dose 43,000 g = the mass of water in the standard human (Shapiro 1990) fH20 = the mass fraction of water in the organ of interest I rem /1 rad = the quality factor conversion for doses from beta emitters H20 n human organs ranges from 60 to 80% (Cember 1983). For this analysis, i

f we assumed a constant water mass fraction of 70% for all organs. Eq. 6 reflects the standard assumption made when calculating organ-specific HTO doses that the IHO is-10

.mmem y

• m

m uniformly mixed in the body water, of which there is about 43 liters in the, typical human (Shapiro 1990).

Ingestion. All DRV residents were assumed to meet 100% of their daily water needs using water from the Deerfield River. The water ingestion rate (Ig) was assumed to be 2 liters per day (EPA 1989). HTO exposures from drinking contaminated water were therefore assumed to be chronic in nature. Hence, HTO exposures and doses were calculated in the same manner as for chronic inhalation exposures (Eq. 5 and Eq. 6) after substituting Ig for Ih and [HTO]H20 or (HTO]eyp.

f Dermal Absorption. HTO is readily absorbed across the skin at a rate that yields exposures and doses approximately 80% of those resulting from inhalation (Shapiro 1990). Thus, HTO exposures and doses were simply assumed to be 80% of the calculated inhalation doses.

2.3 Source Term Data As mentioned previously, source term data collected from annual Yankee Rowe reports were made available to us by the CAN. To simplify the analysis, we assumed l

3 that annual H releases to the atmosphere and Deerfield River were equal to the 3

average of the available annual data from 1961 through 1992. Average annual H releases were calculated as 516 Ci to the Deerfield River and 6.6 Ci to the atmosphere.

It was noted that annual H releases were greater than average prior to 1972 and less 3

than average following 1972, due to a shift to the use of more effective zircalloy cladding i

around the fuel rods at that time. The large discrepancy in releases before and after the j

i 11 i

'[

Figure 2. Estimatec HTO Concentrations in t1e DRV Compared to Natural HTO levels.

1E+01 1E+00 -

m l

l Estimated for DRV

[

V IE-01 1

I Typical Background 7

(Eisenbud 1977)

@ 1E-02 2

!~

Direct Release IE-04 1E-05 Evaporation 2 1E-06

'EW 1E-07 IE-08 Water Air Air

i i

3.2 Human Exposures and Doses Estimated annual HTO exposures and doses to DRV residents by exposure route are shown in Table 2. Estimated 30 year cumulative doses by exposure route are shown j

in Figure 3. Examination of Table 2 and Figure 3 shows that for the conservative exposure scenarios used in this assessment, ingestion of contaminated river water is the dominant exposure route, resulting in doses nearly 100 times greater than all the other i

routes combined. Exposures and doses resulting from inhalation of HTO evaporated l

from the Deerfield River are the next greatest in magnitude, while those resulting from j

annual atmospheric releases are another couple of orders of magnitude lower.

i Table 2. Estimated annual HTO emosures and doses to DRV residents.

e Annual Annual l

Exposure Dose l

Exoosure Route (uCi-sec.) _

_Lmrem)

Ingestion - River Water 5 x 10 0.32 l

6 Inhalation

- Air Release 3003 2 x 10-4

- Evaporation 6 x 10 4 x 10-3 4

Dermal Absorption

- Air Release 2402 2 x 10-4

- Evaporation 4.9 x 10 3 x 10-3 4

s A

'l I

t 15

Figure 3. Estimatec. 30 Year Cumulative HTO Dose to DRV Residents.

100 Total 5

10 River Water 8

E i

1 a

\\.ter 3

Evaporation Evaporation E

0.1 U

v>

0.01 Air Release Air Release 0~001 Ingestion Inhalation Absorption

?

E s

4. DISCUSSION 1

The estimated HTO concentrations in the air and water of the DRV are of interest in that they greatly exceed those present from natural sources. However,it must 1

be noted that H is present naturally in the environment only at very low levels relative 3

to other sources of ionizing and non-ionizing radiation. To illustrate this point, Figure 4 contains the estimated 30 year cumulative HTO dose information shown in Figure 3 along with the typical 30 year cumulative dose received by individuals from all sources of natural radiation (e.g., cosmic rays, radon, radioactive elements in the ground).

Examination of Figure 4 shows that the dose from natural radiation sources is approximately 300 times greater than the total estimated HTO dose from Yankee Rowe I

releases. It should be noted that the y-axis in Figure 4 is on a base 10 logarithmic scale which can distort the information for those unaccustomed to interpreting this type of numerical presentation. Therefore, these data are plotted on a standard y-axis scale in l

l Figure 5. Here, one can clearly see that the estimated HTO doses are minimal 1

compared to those received from natural background radiation.

1 j

Returning to the results displayed in Figure 3, it is interesting to observe that the I

+

dominant HTO exposure route is consumption of contaminated water from the Deerfield River. According to the information provided to us, this is the most implausible i

exposure route because very few individuals obtain their drinking, cooking, or bathing water from the river, either directly or indirectly via a public water treatment facility. If one rejects the ingestion route as implausible, then the dominant exposure route becomes inhalation and dermal absorption of HTO evaporated from the river. This 17 i

=

. ~..

Figure 4. Estimated 30 Year Cumulative HTO Dose to DRV Residents Compared to Raciation Dose from Natural Sources.

10000 Natural Sources 1000 -

{

100 a

Total River Water 10

.i e

.5 y

1 0.1

=

0.01 Air Release Air Release 0.001 Ingestion.

Inhalation Absorption

/

Figure 5. Estimated 30 Year Cumulative HTO Dose to DRV Residents Compared to Radiation Dose from Natural Sources.

3500 Natural Sources 3000 k

8 E 2500 ao O

2000 2

.e l1500 8

w

$ 1000

)

500 Water Water River Water Air Release Evaporation Air Release Evaporation' Total Ingestion Inhalation Absorption f

result suggests that any future efforts should more closely examine the cycling of HTO evaporation and condensation. In this respect, the experimental results of Horton et al (1971) that HTO evaporation is independent of relative humidity are of importance in I

that exposures can be expected to be continuous in nature rather than closely associated i

with the diurnal morning fog reported as common to the valley, i

Health effects associated with exposure to ionizing radiation can be estimated using various dose-response models available in the literature (NAS/BEIR 1990, NRC i

1989, Wing et al 1991). The BEIR V and NRC models are similar because they are based upon health effects observed in survivors from the atomic bombing of Hiroshima and Nagasaki during World War II and uranium miners. A more recent study by Wing et al found statistically significant health impacts in former employees of the Oak Ridge National Laboratory at doses below those commonly thought to produce observable health effects. However, the results of the Wing study are based upon exposures to external penetrating radiation, primarily gamma rays, and therefore are not applicable to 3H.

In accordance with the scoping nature of this analysis, the dose-response models developed by the NRC (1989) were used to project health effects associated with the estimated 30 year cumulative HTO doses. Four health endpoints were selected based upon the availability of appropiiate dose-response models and epidemiological data compiled by Cobb for the CAN; brain cancer, breast cancer, leukemia, and Down's syndmme. The following NRC dose-response models were used, 20

Brain Cancer:

Deaths /1000 = 28 x D Breast Cancer:

Deaths /1000 = 8.4 x D Leukemia:

Deaths /1000 = 9.6 x D 2

Downi:

Cases /1000 = 4.8 x D + 4.8 x D D=3H dose in Gray (1 Gray = 100 rem)

Health impacts estimated from the total 30 year cumulative HTO dose are shown in Table 3 under two cases for what Cobb defined as " river towns" and the city of Greenfield. Case I represents doses calculated for the full exposure scenario described in Section 2 in which all individuals are assumed to obtain their water from the.

7 Deerfield River. Case Il represents what may be a more realistic situation in which only 1% of the DRV population uses the Deerfield River as a sole source of water.

l Table 3. Estimated number of organ-specific cancer deaths and Down's syndrome cases resulting from estimated 30 year cumulative HTO doses in the DRV.

River Towns Greenfield Population 3487 4517 Case I Brain Cancer 0.01 0.01 Breast Cancer 0.003 0.0M Leukemia 0.003 0.0M i

Down's 0.0016 0.0021 l

Case II Brain Cancer 0.0003 0.0004 Breast Cancer 0.0001 0.0001 Leukemia 0.0001 0.0001 Downt 0.00005 0.00007 i

21 r

o The results shown in Table 3 suggest that it is unlikely that HTO exposures to DRV valley residents to routine releases from the Yankee Rowe would result in increased incidence of brain cancer, breast cancer, leukemia, and Downt syndrome.

However, although this analysis contains many conservative and compounding assumptions, it is not complete for several factors remain to be addressed. Principal 3

among these are the potential for some of the H to be organically bound rather than 3

present as HTO. Organically bound H is of special interest because it can become a component of cell nuclei, (e.g., chromosomes) and impart its decay energy directly to genetic material, thus enhancing its mutagenic / carcinogenic potency. Another f

consideration not addressed is the potential for enhanced HTO evaporation as a result of aerosolization associated with rapids and falls in the Deerfield River. Aerosolization would effectively increase the surface area (A) term in Eq. 3, thereby allowing for 3

greater evaporation from the river. In the event that some of the H is a component of 3

organic molecules, aerosolization may also enhance the transfer of H laden micron-sized

~

particles from water to air as described by Belot et al (1982) and Walker et al (1986).

I Finally, nuclear power plants routinely emit a large number of radionuclides in addition to H, and the transport, fate, and human exposure to the predominant ones should be

]

3 1

included in a complete exposure assessment.

i

'l i

l l

22 i

1

~

v

s 4

t

5. CONCLUSIONS AND RECOMMENDATIONS Several preliminary conclusions can be reached from the preceding analysis. First, 3H concentrations in DRV environmental media during the operational phase of the Yankee Rowe facility may have been several orders of magnitude greater than those in f

typical unimpacted areas. Second, assuming that few DRV residents obtain their water j

3 from the Deerfield River, the dominant source of H exposure is inhalation and dermal absorption of airborne HTO as a result ofits evaporation from the river. Third,3H exposures from intermittent gaseous releases from Yankee Rowe were probably insignificant compared to river related exposures. And finally,3H doses resulting from

[

Yankee Rowe operations and delivered as HTO were probably several orders of magnitude lower than those from all natural sources of radiation.

)

However, as noted in Section 4 this analysis contains a number of simplifying, although generally conservative, assumptions which should be examined for their relevance and applicability. Specifically, an im'estigation should be undertaken of the 3

l potential for some of the H to be present in an organic molecule rather than solely as HTO. In addition, the effect of river rapids and falls on HTO evaporation, and possibly 3

J organically bound H aerosolization, should also be researched. Potential exposures to radionuclides other than H should also be addressed. Finally, because the populations i

3 i

of the DRV communities are relatively small and health mtcomes for some specific individuals have been anecdotally associated with the Yank &, Rowe facility, it may be j

possible to reconstruct potential radiation doses to these persons using detailed historical 23 i

f k

I

r j

release data and personal histories. The merit of this type of extensive undertaking i

should be evaluated by the appropriate individuals and organizations.

t

+

i i

k i

F h

i,

?

4 4

.1 i

f s

e 24 a

k

[

i E

L

4

6. REFERENCES i

Belot, Y., C. Caput, and D. Gauthier.1982. Trarsfer of americium from sea water to atmosphere by bubble bursting. ' Atmospheric Environment. 16(1):1463-1466.

Cember, H.1983. Introduction to Health Physics. Pergamon Press, Oxford.

Eisenbud, M. 1987. Environmental radioactivity from natural, industrial, and military l

sources. Academic Press, Inc. Orlando, FL Environmental Pmtection Agency (EPA).1989. Exposure factors handbook.

EPA /600/8-89-043.

Horton, J.H., J.C. Corey, and R.M. Wallace. 1971. Tritium loss from water exposed to the atmosphere. Environ. Sci. Tech. 5(4):338-343.

National Academy of Science (NAS).1990. Health effects of exposure to low levels of.

ionizing radiation, BEIR V. National Academy Press. Washington, D.C.

National Council on Radiation Protection and Measurements (NCRP).1979. Tritium in the environment. NCRP Report No. 62. Washington, D.C..

+

Nuclear Regulatory Commission (NRC).1989. NUREG/CR-4214.

Shapiro, J.1990. Radiation protection. Harvard University Press. Cambridge, MA.

Walker, M.I., W.A. McKmj, N J. P2t'anden, and P.S. Liss. 1986. Actinide enrichment in marine aerosols. Nature. 323:141-143.

Wing. S., C.M. Shy, J.L Wood, S.W. Wolf, D.L f. g and S.L Frome.1991.

. ary. J. American Medical Mortality among workers at Oak Ridge National

+

Association. 265(11):1397-1402.

l

.