ML20197H927
| ML20197H927 | |
| Person / Time | |
|---|---|
| Site: | Vogtle  |
| Issue date: | 08/01/1984 |
| From: | Ballard R Office of Nuclear Reactor Regulation |
| To: | Hulman L Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20197G423 | List: |
| References | |
| FOIA-84-663 NUDOCS 8408070194 | |
| Download: ML20197H927 (15) | |
Text
,.
s a
DISTRIBUTION:
LB6ckets EHEB Rdg i
AUG 1 %4 9 'J Lc2 J
/
Docket Nos. 50-5?4/ErE5-HEMORANDtit FOR:
L. G. Hulman, Chief keident Evaluation Branch, DSI THRU:
William V. Johnston, Assistant Director P,aterials, Chemical & Environmental Technology, DE FROH:
Ronald L. Ballard, Chief Environmental & Hydrologic Engineering Branch, DE
SUBJECT:
LIQUID PATHWAY ASSESSMENT OF CORE fiELT RELEASE -
YOGTLE ELECTRIC CENERATIfiG PLAf T DES Plant Nane: Vogtle Electric Generating Plant Docket fMs.:
50-524/525 Licensing Stage: OL LPri: fi. I:llier Enclosed is the Hydrologic Engineering Section's assessment of the linuid pathway consequences from a postulated core nelt release for nur use in preparing the DES.
The conclusions of the staff are that the liquid pathway consequences at the Vogtle Plant into the Savannah River will be less than those for the river-sited plant in the " Liquid Pathway Ceneric Study". This analysis was perforned by Gary B. Staley (X28003).
- 3gg.,
Ponald L. Ballard, Chief Environnental & Pydroloqic Engineering Eiranch j
L'ivision of Engineering
Enclosure:
As stated cc: w/o enci R. Vollner
- 0. tbiler
,3,
hobo 194e40001 v/ encl 9-
~.)
E. Adensan f ADocK 05000424
- 11. Miller F. Akstulewic(t fi. Fliegel
/ } &(-
\\
rl.
L C. Pa11ard
--R.-Sam orth -
'cyt + t-
.0..Staley.
' DE:EllEB DE:OlEB DE;EHE
/
DEfADMCE.T...l.
- ~ ' "
- f j.
GStaley:ws pHFliegel RBailatW WVJohnston..
.23
"" ' *j 7/%/84 7/..7 /84 7/t//84 7/?/84 s @ T. N,' T 7.f,.
~ - ~ ~
~
orgic;e,ggcopo coev
i
(
i Hydrologic Engineering Section Input to DES Vogtle Electric Generating Plant Docket Number 50-524/525 (Class 9 Liquid Pathway Assessment - Insert for Chapter 5) 5.3 RELEASES TO GROUNDWATER 5.
3.1 INTRODUCTION
This section presents a comparative evaluation of the radiological-consequences which might result following a large accidental release of radionuclides from the VEGP Units 1 and 2 reactors to the local groundwater system.
Such releases could occur following a postulated core meltdown with eventual penetration of the containment basemat.
Core debris which exits the melt hole at elevation 134 ft msl would be released into the ground below the water table, which extends from elevation 134 ft msl to 160 ft as1.
The radionuclides in the debris would then be leached into the groundwater system.
It is also possible for containment sump water, which would be rich in dissolved fission products, to be released via the basemat melt hole into the groundwater system.
An analysis of the potential consequences of such an event is presented by the NRC staff in NUREG-0440, " Liquid Pathway Generic Study" (LPGS).II) It is this generic report which provides the basis for the comparative evaluation of the VEGP units.
The LPGS presents analyses for a Westinghouse PWR at five land-based sites.
Two of the land-based sites analyzed in the LPGS were river sites.
Vogtle is a river site located on the Savannah River 151 miles from the Atlantic Ocean and is most comparable to the LPGS river site.
In the LPGS, parameters for each generic site were chosen to be representa-tive of the full spectrum of similar sites.
Parameters used for analysis in the LPGS, although typical, do not represent any actual plant site.
q
_3 Individual and population doses are reported in the LPGS for the principal liquid pathways; drinking water, aquatic food, and direct exposure from swimming and shoreline usage.
Exposure resulting from crop irrigation was also considered but was found to contribute insignificantly to dose.(2)
Doses to individuals and populations were calculated in the LPGS without taking credit for possible interdiction methods such as isolation of contaminated groundwater, the temporary restriction of fishing or providing alternative sources of drinking water (or additional purification equipment).
Such interdiction methods would be highly successful in preventing exposure to radioactivity and the liquid pathway consequences would therefore be economic and social rather than radiological.
The study concluded that the individual and population doses for the liquid pathway would be a small fraction of the airborne pathway dose which could result from a core meltdown accident.
5.3.2 METHOD OF COMPARIS0N The estimate of the liquid pathway consequences resulting from a radionuclide release at VEGP is developed by comparing, in a series of ratios, the principal parameters applicable to the VEGP site to the parameter values used for the generic river site calculations in the LPGS.
The parameters for which ratio comparisons are developed are the following:
A.
The radionuclide source released to the river.
B.
The population along the river system which obtains drinking water from the river.
. C.
The annual fish harvest on the river system.
D.
The annual recreational usage of the river system.
In a very general way the consequences of a major radionuclide release to the groundwater system at VEGP can be expressed as follows:
EGP source LPGS dose for usage ratio for VEGP dose =
X X
LPGS source the ith pathway the ith pathway Pathway " usage" ratios are the following:
A.
Drinking water population for VEGP river system Drinking Water population for LPGS river system B.
Annual fish harvest for VEGP river system Annual fish harvest for LPGS river system C.
Manhours direct exposure for VEGP river system Manhours direct exposure for LPGS river system To be exact, this summation should be carried out for each radionuclide.
However, it has been found that the liquid pathway doses tend to be dominated by a very few radionuclides.
As will be shown in a subsequent section, the characteristics of the VEGP site are such that most of the important radio-nuclides will undergo substantial decay during the process of groundwater transport to the Savannah River.
Therefore, the general equation above provides an adequate approach to developing a comparative liquid pathways dose evaluation.
5.3.3 SITE CHARACTERISTICS The VEGP is located on the southwest bank of the Savannah River at approximately river mile 151.
This location is about 26 air miles south-southeast of Augusta, Geogia. Vogtle is located on the eastern margin of i
~
+
i
, the Tifton Upland topographic belt, on an elevated area of the Coastal Plain geographic region with plant grade at elevation 220 ft above mean sea level (msl). The Savannah River cuts a deep, transverse valley through the Coastal Plain along the eastern border of the plant site.
The river valley is a mature topographic feature with a broad flood plain at approximately elevation 85 ft asl (135 ft below plant grade). The plant is located about 1097 meters (3600 ft) from the Savannah River at its closest approach to the site.
The principle load bearing structure for the Vogtle plant is the Blue Bluff Marl which is a member of the Lisbon formation.
The Blue Bluff Marl is a clayey marl approximately 21 meters (70 ft) thick; the top of the load bearing horizon is located about 26 meters (85 ft) below grade at elevation 134 ft msl.
The containment building and most other plant structures are built upon this soil structure.
The Blue Bluff Mari consists of a semi-consolidated glauconitic marl with subordinate lenses of dense, well-indurated, and well-cemented limestone.
The marl layer overlies the unnamed sands member of the Lisbon formation.
The hydraulic conductivity of the marl layer is very low, essentially zero, and it is classified as an aquaclude.
The marl effectively confines the groundwater withf" the unnamed sands layer to produce artesian conditions under the site. The. artesian water region is referred to as the Tertiary Groundwater System and is the source of the plant's potable water supply.
Due to the impermeable nature of the marl, recharge to this aquifer is not a direct result of rainwater infiltration at the site.
The influx of meteoric water at the olant site and surrounding area, after percolating through the overlying soil, accumulates above the Blue Bluff Marl to produce water table conditions.
This water table aquifer extends from elevation 160 ft msl down to the top of the Blue Bluff Marl at elevation 134 ft asl.
Hydraulic connection with the Savannah River is precluded by the stratigraphy of the site.
s
. The Blue Bluff Marl formation slopes in a general easterly trend toward the Savannah River.
However, this trend is insufficient for the marl to pass beneath the river. As the Savannah River cut its channel the marl was exposed at elevation 130 ft asl on the southwest bank of the river approximately 14 meters (45 ft) above the flood plain.
The water table aquifer discharges to the surface by seepage through the flanks of adjacent stream beds as they flow toward the Savannah River.
The water table also discharges to surface waters in several free-flowing springs located near the plant site. These springs feed small streams which eventually flow into the Savannah River.
The local groundwater system is shown in FSAR Figure 2.4.12-7 and is described in FSAR Section 2.4.12.
5.3.4 GROUNDWATER TRAVEL TIMF 4
Radionuclides entering the groundwater system above the Blue Bluff Marl would be entrained in the natural groundwater flow to streams feeding into the Savannah River. The Blue Bluff Marl aquaclude would preclude the migrr'. ion of radionuclides from a postulated core melt accident into the underlying confined aquifer.
The VEGP is situated on the northwest side of a relatively flat groundwater plateau.
Radionuclides released in the vicinity of the plant would probably migrate in a northwesterly direction to a spring about 975 meters (3200 ft) from the Unit 2 Containment Building.
However, there is another spring located about 853 meters (2800 f t) southeast of the Unit 1 Containment that would be somewhat closer and would have a steeper average gradient.
Thus, although it is likely that the contaminant pathway would be in a northwesterly direction, the staff conservatively assumed the groundwater pathway for a core melt release will be to the spring
~
853 meters (2800 ft) southeast of the plant (see FSAR Figure 2.4.12-7).
I
. The seepage velocity may be determined with Darcy's Law as follows:
v=ki "e
where; v = seepage velocity k = hydraulic conductivity i = hydraulic gradient n,= effective porosity or specific yield The groundwater drops 6.1 meters (20 ft) over the 853 meters (2800 ft) distance from Unit 1 Containment to the spring giving a hydraulic gradient
-3 of 7.1X10 The applicant provided hydraulic conductivity estimates of 61 to 107 meters / year (200 to 350 ft/ year) from field measurements and 3 to 6096 meters / year (10 to 20,000 ft/ year) from laboratory measurements.
Field measurements are much more reliable and representative of aquifer characteristics, whereas laboratory values represent only a small disturbed sample of the aquifer.
Upper and lower limits are of little value.
An average of many samples representative of the areal and vertical extent of the aquifer would provide a fair approximation of aquifer characteristics.
For this analysis the staff selected the upper limit of field measured hydraulic conductivity of 350 ft/ year as the value that is conservatively representative of aquifer conditions.
The applicant provided only one undocumented estimate of porosity at 0.45.
This is greater than the upper limit quoted in textbooks for medium to fine densely packed sand.
The staff assumed a conservative value of 0.39 for total porosity and 0.30 for the effective porosity.
Using the above parameters the "best estimate" groundwater velocity (v B.E.) is given by:
v B.E. = (107 mete year) (7.1X10-3}
= 2.5 meters / year (8.28 ft/ year)
The "best estimate" of travel time (t B.E.) is thus given by the following:
t B.E. =
= 341 years
=
2.5 et ar
i
(
3 \\
The range of laboratory hydraulic conductivity values (3 - 6,096 meters / year) provided by the applicant are of little value since there is no record of the quantity of samples, the areal location or the vertical location.
- However, since the range was provided in the FSAR, the staff has also provided a
" conservative" (cons.) estimate of groundwater travel time (t cons.) using a hydraulic conductivity of 2438 meters / year (8000 ft/ year).
The resultant conservative groundwater velocity would be 57.7 meters / year (189 ft/ year) and the conservative travel time (t cons.) would be 14.8 years.
N 5.3.5 SOURCE COMPARISON The radionuclide source which is ultimately transmitted through a groundwater system to an adjacent surface water is determined by the following three factors:
The core radionuclide inventory.
The fraction of the core radionuclide inventory released to groundwater via such mechanisms as sump water release and leaching from the core debris.
The attenuation which takes place during transport through the groundwater system, principally from radioactive decay and adsorption.
The LPGS analyses are based on the core inventory for a four-loop Westinghouse PWR similar to the VEGP units.
The fracti.an of the core inventory which could be released to the groundwater depends on numerous factors, such as the specific accident sequence and containment failure mode, containment Y
(
. sump structure, and the nature of the soils which separate the containment basemat from the underlying groundwater system.
For convenience, it is ass'med that the LPGS assumptions apply to the VEGP units.
A number of release cases are considered in the LPGS; however, the worst cases con-sidered (instantaneous release of all sumpwater and all activity available for leaching)(3) are clearly bounding for any plant-site combination.
It was demonstrated in the LPGS(1) that for travel times on the order of years virtually all of the population dose from the liquid pathway in an assumed core melt accident would result from Sr-90 and Cs-137.
These chemically active nuclides would, however, travel through the goundwater pathway at a much slower rate because of the process of sorption onto the soil and rock media.
The degree of retardation is governed by the various physical properties such as bulk density, aquifer porosity, and species equilibrium distribution coefficient.
The relationship between groundwater velocity (or groundwater transport time), radionuclide adsorption, and the (deccyed) radionuclide fraction, which is ultimately transmitted, is given by the following expression:(4)(5) in (T.F.) = - 693 (t cons.) (a), where k
T.F. = transmitted fraction.
t cons. = groundwater transport time.
tg = radionuclide half-life.
a = adsorption retention factor.
Theadsorptionretentionfactorisequalto(1+{K)where:
d l
p = bulk density of the aquifer media.
n = porosity of the aquifer.
f K = distribution coefficient which is defined as the mass d
i I
of radionuclide adsorbed per gram of soil divided by the mass of radionuclide dissolved per milliliter of groundwater.
j
s i
. A typical value of the ratio p/n is 5; however, for consistency the value of 4.1 used in the LPGS is adopted here as well.(6)(7) The retardation factors 3
were calculated,using equilibrium distribution' coefficients of 5 cm /gm for Sr-90, 49 cm /gm for Cs-137, and zero for H-3.
These equilibrium distribu-tion coefficients were derived from an" extensive literature search and are at the low end of the range of values given by Isherwood(8)
The calculated 7
retardation factors for Sr-90, Cs-137 and H-3 are 21.5,165 and 1, respectively.
[
n Table 6.2.1 of the LPGS lists the transmitted fraction for a number of radionuclides, the more important of which are reprodyced as follows:
t Nuclide
% (yrs)
T. F.
H-3 12.1 0.97 SR-90 28 0.87 Cs-137 30 0.31 The values are based on the following data assumed in the LPGS for the generic river site:I )
GWTT = g
= 224 days = 0.61 year a (H-3)
=1 (equivalent values of K = 0) d a (Sr-90) = 9.2 (equivalent values of K = 2) d a (Cs-137) = 83 (equivalent values of K
)
d The conservative groun(twater transport time at the VEGP site is estimated to be 14.8 years.
On the basis of this and the calculated retardation factors, the transmitted / fractions for the principle radionuclides are as follows:
f.
/
. Nuclide W r)
T.F.II)
T.F. (VEGP)/T.F. (LPGS t
H 12.1 0.43 0.44 3
Sr-90 28 0.0005 0.0006 Cs-137 30 0
0 (1) The transmitted fractions using our "best estimate" of travel
~9 time would be 4.2 x 10 or less which would probably meet the 10 CFR Part 20 requirements for potable water in an unrestricted area if the effects of dilution in the ground and surface water regimes are taken into consideration.
The effect of the much longer conservative estimate of groundwater travel time at the VEGP site (14.8 years compared to 0.61 years in the LPGS), even with the relatively small assumed values of K, is very significant.
d Virtually no Cs-137 would be expected to reach the Savannah River.
Only 5/10,000 of the released Sr-90 would reach the river (compared to a trans-mitted fraction of 0.87 in the LPGS).
Tritium (H ) is closer to the LPGS 3
results with a transmitted fraction of 0.43 for VEGP compared to 0.97 for LPGS.
The source effect on liquid pathway consequences can be summarized as follows:
A.
Pathway doses which are dominated by Cs-137 will be nil in comparison to doses calculated in the LPGS.
B.
Pathways doses which are dominated by Sr-90 will be about 4 orders of magnitude lower than those calculated in the LPGS, assuming equal pathways exposure.
C.
Pathways doses from H will be lower but within the same 3
order of magnitude, assuming equal pathway exposure.
At the levels of population dose calculated in both NUREG-0440 and in the Sandia study (10), tritium (11 ) is not a 3
g'
n t
. significant contributor.
This is in part due to the smaller core inventory of tritium (two to three orders of magnitude less the curie content than Sr-90, or Cs-137)(11) and also in 2
part to the relatively low total body dose factor (1 x 10 6
man-rem / curie compared to 1.9 x 10 man / rem / curie for Sr-90 4
and 8 x 10 man-rem / curie for Cs-137).(12) 5.3.6 DRINKING WATER PATHWAY COMPARISON The LPGS river site was assumed to supply drinking water to 620,000 people.(13) As shown in ER Table 2.1-44 the current number of people that get their drinking water supply from the Savannah River is 70,000.
This is only 11.3 percent of the number used in the LPGS.
In addition the drinking water pathway dose is dominated by Sr-90 and Cs-137.II4) Since the transmitted fractions of these radionuclides are much smaller than the LPGS, the drinking water pathway dose for VEGP is about 5 orders of magnitude less than the LPGS dose.
5.3.7 FISH INGESTION PATHWAY COMPARISON In the LPGS river site, it was estimated that the annual fish harvest was 1.2 x 10 kg (1.7 x 10 kg recreational and 3.9 x 10 kg commercial).(15) 6 5
5 The annual recreational fish harvest on the Savannah River from 0 to 187.2 5
miles for 1980 is shown in ER Table 2.1-42 as 1.04 x 10 kg.
The commercial fish harvest is not complete but the average commercial shad harvest is shown 4
in ER Table 2.1-43 as 3.7 x 10 kg.
The amount of fish harvested from the Savannah River is approximately equivalent to the values used in the LPGS river site.
Like the drinking water pathway, the fish ingestion pathway is dominated by Sr-90 and Cs-137.(14) Since the transmitted fraction for Sr-90 is four orders of magnitude lower, it is concluded that the fish ingestion dose is
t I
. about four orders of magnitude lower.
In addition, the economic and social impacts of a severe accident on ocean fish catch should be roughly four orders of magnitude less than that assessed for the LPGS ocean fish catch.
5.3.8 SHORELINE AND IMMERSION PATHWAY COMPARISON The shoreline and immersion pathway includes such activities as swimming, wading, sunbathing, etc.
These are external exposure pathways and dosage is dominated by Cs-137.(14) Since the transmitted fraction for Cs-137 is essentially zero, it is concluded that the direct exposure dose would be nil in comparison to those calculated in the LPGS.
5.
3.9 CONCLUSION
S On the basis of VEGP site features and the specific comparisons of radionuclide source and pathway populations, it is apparent that the spectrum of liquid pathway doses following a Class 9 accident would be much lower for VEGP than the doses calculated in the LPGS for a river-sited plant.
This is mainly due to the much smaller amount of radionuclide reaching the Savannah River, which in turn results mainly from a much longer groundwater transport time.
If shorter times are postulated, the adverse effect would be small and would probably be offset through the assumption of more realistic distribution coefficient (K ) nd hydraulic conductivity coefficient (k) values.
d In the extreme, if one were to postulate the same radionuclide source as in the LPGS, the pathways doses would still be slightly lower, since the pathways population ratios are about the same or lower.
Finally, there are measures which could be taken to further minimize the impact of the liquid pathway.
The staff has conservatively estimated that the minimum ground water travel time from the containment building to the nearest spring would be about 14.8 years.
This would allow ample time for engineering measures such as slurry trenches or well point dewatering to isolate the radioactive contamination near the source and to establish
i q
. a groundwater monitoring program that would insure early detection if any, contaminants should escape the immediate plant area.
A comprehensive discussion of these and other mitigation methods potentially applicable to VEGP is contained in Harris et al (10) and (1 ).
v l
h 4
4
. =,
n
--,,--,-,..n
m t.
I References 1.
U.S. Nuclear Regulatory Commission, " Liquid Pathway Generic Study" NUREG-0440, February 1978.
2.
NUREG-0440; pp 4-26, 4-27.
3.
NUREG-0440; Tables 6.2.16 and 6.2.17; p 6-22.
4.
"The Consequences from Liquid Pathways After a Reactor Meltdown Accident," NUREG/CR-1596, Sandia National Laboratories, Appendix B, Section B.4.3.1, August 1981.
5.
NUREG-0440, p B-23.
6.
NUREG/CR-1596, p 120.
7.
The value of 4.1 is inferred from the transmitted fractions listed in Table 6.2.1 of NUREG-0440 and from the adsorption retention factors listed in NUREG-0440, Table 4.2.4.
8.
Isherwood, D.
Geoscience Data Base Handbook for Modeling a Nuclear Waste Repository, NUREG/CR-0912, Volume 1, U.S. Nuclear Regulatory Commission and Lawrence Livermore Laboratory, January 1981.
t 9.
NUREG-0440, pp 4-18, 4-19.
10.
NUREG/CR-1596, Table 6.4, p 75.
11.
NUREG-0440, Appendix A, Table A07, p A-30.
12.
NUREG-0440, Appendix B, Table B-5, p B-38.
13.
NUREG-0440, Table 4.3.1, p 4-31.
14.
NUREG/CR-1596, Table 6.4, p 74.
15.
NUREG-0440, p 4-31.
16.
Harris, V.
A., et al, " Accident Mitigation:
Slurry Wall Barriers",
Argonne National Laboratory, May 1982.
17.
Harris, V.
A., et al, " Accident Mitigation: Alternative Methods for Isolating Contaminated Groundwater", Argonne National Laboratory, September 1982.
. _ _ _ _ _. _.. - - - - -, - - - - - - -