ML20044E235
| ML20044E235 | |
| Person / Time | |
|---|---|
| Issue date: | 05/19/1993 |
| From: | Taylor J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | |
| References | |
| SECY-93-138, NUDOCS 9305240090 | |
| Download: ML20044E235 (14) | |
Text
\\
~
L RELEASED TO THE PDR
?:'
f * "*%
b S,/79/95 L
'date ' ~
initts g
.g
\\...../
POLICY ISSUE "3
- ' 38 (Notation Vote)
For:
The Commissioners From:
James M. Taylor Executive Director for Operations Sub.iect:
RECOMMENDATION ON LARGE RELEASE DEFINITION
Purpose:
To provide the Commission the staff's evaluation on the development and usefulness of a large release definition as a plant performance objective and to recommend that work on the development of a large release definition should be terminated.
Backaround:
On August 4, 1986, the Commission issued its Safety Goal Policy Statement and approved the use of qualitative and quantitative safety goals in the regulatory process (51 FR 28044). The Commission stated that guidance on the use of the safet/ goals may also include a general performance guideline that was proposed by the Commission for further staff examination.
This guideline was stated as follows:
" Consistent with the traditional defense-in-depth approach and the acc.ident mitigation philosophy requiring reliable performance of containment systems, the overall mean frequency of a large release of radioactive materials to the environment from a reactor accident should be less than 1 in 1,000,000 per year of reactor operation."
On March 30, 1989, the staff proposed a general framework and Safety Goal Pality Implementation in SECY-89-102 (" Implementation of Safety Goal NOTE:
TO BE MADE PUBLICLY AVAILABLE WHEN THE FINAL SRM IS MADE AVAILABLE
Contact:
C. Ader (492-3975)
J. Ridgely (492-39 8) 2 h^ 0SM ooc [yoloo3s
~~~~~J i
gp,
b The Commissioners Policy").
As discussed in SECY-89-102, several different approaches were considered to give more explicit meaning to the term "large release," both quantitatively and qualitatively. At that time, the staff recommended that a large release be defined as follows:
"A large release is a release that has a potential for causing an offsite early fatality."
In a Staff Requirements Memorandum (SRM) dated June 15, 1990, the staff was requested to re-examine and advise the Commission whether a plant performance objective that focuses on accidental releases from the plant (i.e. large release) and eliminates site characteristics could be developed and be useful.
In response to this direction in SECY-90-405, the staff discussed two possible definitions for a large release, proposed one for further evaluation and provided a plan to evaluate its magnitude utilizing representative site characteristics. The definition proposed was the following:
"A large release is a release of radioactivity from the containment to the environment of a magnitude equal to or greater than: (An amount, to be determined by the staff, expressed in curies or fraction of the core inventory, which has the potential, based on representative site characteristics, for causing one or more offsite prompt fatalities.)"
The proposed staff evaluation of large release magnitude was to be consistent with the ACRS proposed guidelines linking the hierarchical levels of the safety goal objectives, where the large release guideline was considered the third level objective (the qualitative and quantitative health objectives are the level one and two objectives, respectively). According to these guidelines, each subordinate level of the safety goal objectives should:
1.
Be consistent with the level above; 2.
Not be so conservative as to create a de facto new policy; 3.
Represent a simplification of the previous level; 4.
Provide a basis for assuring that the Safety Goal Policy objectives are being met; 5.
Be defined to have broad generic applicability; 6.
Be stated in terms that are understandable to the public; and 7.
Generally comport with current PRA usage and practice.
The Commission, in an SRM dated March 21, 1991, approved the staff proposed definition and evaluation plan, provided guidance for use in selecting the representative site characteristics and requested that the representative site parameters be provided for Commission approval before completing final calculations of a large release. A draft of a paper on the representative site definition was provided to the Commission for information in a memorandum dated October 24, 1991.
Several meetings with ACRS have also taken place on this subject over the past two years.
The Commissioners In addition, in a recent memorandum to the staff dated March 2, 1993, Commissioner Remick recognized the difficulty in developing a large release definition that is consistent with the quantitative health objectives and raised the question of whether a large release definition is still needed.
Discussion:
Over the past year and a half the staff has spent considerable time attempting to define a large release magnitude within the framework of the safety goal hierarchy proposed by ACRS and the guidance provided in the Commission's June 15, 1990 and March 21, 1991 SRMs.
A discussion of the analyses performed, the methods and assumptions used and the results are contained in the Enclosurc 1 to this paper.
The overall conclusion reached by the staff is that development of a large release definition and magnitude, beyond a simple qualitative statement related to the 10" per year large release frequency (such as is currently contained in the Commission's Safety Goal Policy Statement), is not practical or required for design or regulatory purposes.
The factors leading to this conclusion are discussed below.
In addition, based upon the work done evaluating the large release, NUREG-1150 and other related activities, the staff notes that the general performance guideline (i.e., large release frequency of 10" per year) proposed in the Safety Goal Policy Statement and the core damage frequency subsidiary objective (i.e., core damage frequency of 10" per year) being used by the staff are not consistent with the quantitative health objectives (QH0s) stated in the Safety Goal Policy l
Statement. This is also discussed below.
Conservatis_m of the Larce Release Guideline The level two hierarchical safety goal objectives are-the quantitative health objectives contained in the Commission's Safety Goal Policy Statement.
Stated numerically the 0.1% individual risk values included in the QH0s correspond to:
l o
prompt fatality goal = 5 x 10 per year risk of a prompt fatality to an average individual within one mile of the site boundary o
latent fatality goal = 2 x 10" per year risk of a latent fatality to an average individual within ten miles of the site boundary, j
Of the two QH0s, the prompt fatality QH0 has been found to be the more l
restrictive objective.
It was recognized in SECY-89-102 that, at an overall mean frequency of less than 1 in 1,000,000 per year, any large release definition would represent a QHO inherently smaller than the prompt fatality
-QHO. (the prompt fatality QHO repesents a 5 x 10" per year risk of a prompt fatality to an average individual within 1 mile of the site boundary).
Consideration of wind direction alone (16 possible wind directions in the MACCS calculations) results in about an order of magnitude conservatism.
This follows from the definition of an average individual given in the Safety Goal Policy Statement. Specifically, if one uses the mean core damage frequency subsidiary objective of 10" per reactor year, a conditional containment I
The Commissioners i 4
failure probability of 0.1 (frequency of a release of about 10 per reactor year), and a probability of a release in a given wind direction of about 1/16, then the maximum risk to an individual can be estimated at 6x10" per reactor year which is approximately the prompt fatality QH0. Similarly if the frequency of a release is taken to be 10" per reactor year, then the maximum risk to an individual is approximately an order of magnitude less than the prompt fatality QHO. An order of magnitude conservatism was accepted by the Commission in U.S June 15, 1990 SRM.
i However, when the individual risk from a *large release" is evaluated using realistic meteorology, realistic release characteristics and realistic protective actions, several more orders of magnitude conservatism can be and are introduced reaardless of how the laroe release is defined. This can be seen from the results presented in NUREG-ll50.
For the five plants studied, all plants had a probability of an early containment failure or bypass between 10" and 10" per year, yet the prompt fatality risk to the average individual for all piants was over an order of magnitude or more below the prompt fatality QHO.
Conversely, the five plants studied in NUREG-Il50 could meet the prompt fatality QH0 even if the frequency of an early containment failure or bypass was higher by an order of magnitude or more. Also, much higher frequency of core damage can be tolerated without exceeding the quantitative safety goals if one were to base regulatory decisions on the QH0s alone.
Larae Release Definition 4
Given a large release frequency of 10 per year, any large release definition will result in a degree of conservatism several orders of magnitude more conservative than the QH0s.
Furthermore, as discussed in Enclosure 1, the specification of the magnitude of any large release definition is very sensitive to the assumptions used for certain parameters in the calculation.
The results will be site, weather, and accident sequence dependent, such as.
the following parameters:
o the energy of release (ground level or elevated) o the timing and duration of the release o
the protective action assumptions used o
the population density of the area immediately surrounding the plant.
Variations in these parameters (given a fixed consequence in terms of risk) can cause the large release magnitude to vary widely and, in some cases, exceed the release estimated to have occurred from the Chernobyl accident.
Other parameters can also affect the magnitude to a lesser extent.
Expressing the large release magnitude in terms of ' equivalent curies of I "" versus 2
fraction of core inventory can eliminate the effect of the timing of release parameter but, the magnitude or quantity released war still subject to wide variation from the assumptions used for the other paremeters. Therefore, to implement a large release guideline expressed in terms of a magnitude of radioactive material released to the environment, a prescriptive analytical methodology would be required. Such a prescriptive methodology would tend to be conservative so as to envelop the variations in the above parameters and its implementation would potentially introduce additional conservatism below i
n
The Commissioners the QH0's. This would lead to a prescription that clearly does not represent a useful simplification relative to the implementation of the quantitative health objectives themselves.
The staff also evaluated a qualitative alternative large release definition as discussed in SECY-90-405 to see if this was practical. This definition was:
"A large release is any release from an event involving severe core damage, reactor coolant system pressure boundary failure, and early failure or significant bypass of the containment."
This evaluation focused on defining reasonable values for "early" and
" failure" and whether these were more useful and subject to large variation from the calculational assumptions. Using the same representative site characteristics and based on one or more prompt fatalities, the analyses indicated that "early" would be defined as occurrin first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> following the onset of core damage.'g within approximately the However, the value of early was also subject to large variations depending upon the other assumptions used. The staff also found it difficult to define " failure."
The staff's efforts were directed toward defining " failure" in terms of containment leak rate, but the staff also found that the value was subject to large variations depending on the assumptions used and would not represent a performance objective that would be simple to understand and readily useable in the regulatory process.
Overall, it was concluded that implementation of this alternative qualitative definition might represent a degree of simplification compared to a quantitative definition, however, it would still result in a large release guideline several orders of magnitude more conservative than the safety goal prompt fatality quantitative health objective.
==
Conclusion:==
In parallel with work evaluating a large release described above, the staff developed interim guidance regarding implementation of the Commission's Safety Goal Policy (SECY-91-270 and SECY-93-043) and criteria for use in the certification review of advanced LWR designs (SECY-90-016 and SECY-93-087).
The interim guidance proposed in the revised Regulatory Analysis Guidelines (SECY-93-043) provides a framework for regulatory decision making utilizing the core damage frequency (CDF) and conditional containment failure probability (CCFP) as the subsidiary safety goal objectives approved for use in the Commission's June 15, 1990 SRM. The staff has recommended that this guidance be issued for public comment. This framework, if adopted, represents
'In NUREG-1150, early containment failure is defined as: "Those containment failures occurring before or within a few minutes of reactor vessel breach for PWRs and those failures occurring before or within 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> of vessel breach for BWRs. Containment bypass failures (e.g., interfacing
. system loss-of-coolant accidents) are categorized separately from early failures."
The Commissioners a relatively easy to use tool to assess the need for proposed changes in the' existing rules and generic requirements considering both core melt prevention and containment performance and is consistent sith the 10" per reactor year CDF contained.in Commission guidance and the 10 per rtactor year large release frequency proposed in the Commission's Safety. Goal Policy Statement.
Further, this approach represents use of the CDF and CCFP values as up-front screening' criteria to determine whether to proceed to an in-depth value-impact analysis on proposed rules and generic requirements. We believe this approach would be consistent with the Commission's intent to maintain the defense-in-depth principle. That is, keep the frequency of a core damage event low in the U.S. plants and to assure that the containment provides appropriate mitigation capability to limit the releases.
The criteria developed for the advanced LWR designs provides for containment performance objectives that could be used to assess overall plant performance in regard to the' containment of radioactive material from severe. accidents, even though the staff has acknowledged that such criterion is likely to result in several orders of magnitude conservatism relative to the constraining QHO.
In addition to capturing the containment performance criteria through design certification rulemakings, the criteria associated with containment performance may become codified via rulemaking as discussed in SECY-92-292,
" Advance Notice of Proposed Rulemaking on Severe Accident Plant Performance Criteria for Future LWRs."
Given the above activities, the staff believes that the need for a precise large release definition has diminished in importance. This, coupled with the difficulty in developing a useful and coherent large. release definition and magnitude, has led us to conclude that it is not necessary tu further pursue a large release definition or magnitude.
Instead, the staff would propose to use the guidance developed for regulatory decisionmaking for assessing reactor designs and safety issues, i.e., Regulatory Analysis Guidelines, for existing
+
plants and Commission approved criteria for reviewing the acceptability of advanced reactor designs.
Coordination:
On April 15, 1993, the staff briefed the Advisory Committee on Reactor Safeguards on the staff's recommendation regarding development of a definition of a large release. The Committee provided its views to the Commission in a
. letter dated April 22, 1993 (Enclosure 2) and agreed with the staff's recommendation to terminate efforts to develop a definition of a large release.
The Office of General Counsel has reviewed this paper and has no legal objection.
I
i 4
The Commissioners \\
Recommendation.
That the Commission direct the staff to terminate further work on the development of a large release definition and magnitude.
f mM4h Atu1"-
^
J mes M. Taylor xecutive Director for Operations
Enclosure:
As stated t
Comnissioners' comments or consent should'be provided directly to the Office of the Secretarf by COB Wednesday, June 2, 1993.
Commission Staff Office comments, if any, should be submitted to the Commissioners NLT Wednesday, May 26, 1993, with an infor-mation copy to the Office of the Secretary.. If the paper is.of such a nature that it requires additional review and comment, the Commissioners and the Secretariat should be apprised of r
when comments may be expected.
t DISTRIBUTION:
Commissioners OGC OCAA
' OIG 1
i l
I
4 Calculations for a tarae Release Definition In examining the definition of a large release, the staff used the Melcor Accident Consequence Code System (MACCS), which was also used in NUREG-1150 to calculate offsite consequences. To calculate consequences, MACCS requires input related to site characteristics (wind speed and direction probabilities; rainfall quantity and duration; population distribution; and exclusion area distance),
protective action assumptions, and source term characteristics (thermal energy, timing, duration, and composition of release). These parameters can affect the characteristics of a releasi that meets the proposed large release definition in SECY-90-405, i.e., having the potential of causing one or more prompt fatalities.
In its March 21, 1991 SRM, the Commission provided guidance in determining the characteristics of a
representative site.
Specifically, real site characteristics that are representative of sites that would fit within the envelope of the proposed revision of 10 CFR Part 100 were to be used.
In addition, the selection of the site characteristics was to reflect a conservative approach such that the resulting site model would encompass the calculated consequences of any actual site.
In developing representative site characteristics and determining potential large release magnitudes in accordance with the Commission's guidance, the staff performed sensitivity studies on selected MACCS input parameters. Key parameters evaluated are discussed below.
Exclusion Area Distance:
The staff is currently proposed to codify in Part 100 an exclusion area distance of 0.4 miles.
The staff performed sensitivity studies of the effect of varying the exclusion area distance between 0.17 and 1.33 miles in ensuring that the prompt fatality QH0 is 2
not exceeded.
Results are shown in Table 1 in terms of fraction of I
released and in equivalent curies.
Because the effect of varying the exclusion area distance was relatively small and to maintain consistency with the revision of Part 100, as discussed above, a value of 0.4 miles was used in characterizing the representative site.
Table 1 - Effect of Exclusion Area Distance Release sheltering Time
% I'" of Core ;
' q[h$
Radius' (Miles) -
r on re 1
12 0.40*
60 5.8E47 1
12 0.17' 42 4.0E+7 1
12 0.55*
73 7.0E+7 1
12 1.33' 93 8.9E+7 Notes:
- Base '(Proposed) Case
- smallest distance of existing plants (SECY-90-341).
- Average distance of. existing plants-(SECY-90-341)
- 1.argest distance of existing plants (sECY 90-341)
Pooulation Density:
The staff is currently proposing to codify the population density guidelines of Regulatory Guide 4.7,
" General Site Suitability Criteria for Nuclear Power Stations," in the proposed revision
2 to Part 100.
For evaluation purposes, this includes an initial value of 500 people per square mile within 30 miles of the reactor and a 40 year projected population density below 1000 people per square mile.
In an effort to ensure the representative site model would encompass the consequences of any actual site, while staying within the envelope of the proposed Part 100, a uniform population density of 1000 people per square mile was used for the representative site.
For comparison, the average population density for current plants within two miles of the plant is approximately 125 people per square mile.
If the population density for the representative site is reduced from 1000 to 100 people per square mile (an order of magnitude), a release of approximately twice the equivalent curies of I"' (1.2x10') would be required to result in one prompt fatality.
A release of this magnitude assumed to occur at 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after the reactor shutdown is equivalent to approximately 100% noble gases, 9% iodine, 9%
Meteoroloav: The meteorological parameters used in the MACCS consequence
- alculations are the atmospheric stability or dispersion characteristics, the amount and duration of rainfall, and the wind direction frequency (percentage of the time that the wind blows in a given sector / direction).
During the staff's efforts to define a large release as a fraction of core inventory, two cases were examined: a mean v&iue case and an 80" percentile case.
The effect of varying the meteorology on the number of prompt fatalities was found to be small. Therefore, reflecting a conservative approach, the 80" percentile meteorology was used for the majority of the sensitivity calculations.
Protective Action Assumptions:
During the staff's efforts to define a large release as a fraction of core inventory, the staff considered four cases. The first case used the mean of the assumptions made in NUREG-IISO (99.5% of the people evacuate with an evacuation speed of 5.9 mph and after a delay time of 1.9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />).
In addition, the staff evaluated several protective action assumptions that were more conservative.
The first conservative case assumed that only 95% of the people evacuate at a speed of 2.5 mph after a delay time of six hours.
A second conservative case, which assumed no protective actions, i.e., the public continues their normal activities, was evaluated in an attempt to decouple the "large release" definition from protective action assump-tions.
This case, assuming a ground level release, results in a relatively small magnitude of a "large release". This is a result of the long term groundshine dose. (For this case, the MACCS code assumes that people will centinue normal activities for seven days after a release.)
Because the no protective action assumption was considered overly conservative, an additional conservative case was evaluated in which it was assumed that the population is sheltered for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, after which 100% of the people were relocated. These analyses resulted in a number of I
l 2
i
different potential "large release" magnitudes ranging from releases of 100% of the noble gases and a few percent of the iodine to release magnitudes of 100% of the noble gases and 20-30% of iodine and cesium along with significant amount of other isotopes.
Much of this wide variation resulted from assumptions of the timing of the release in relationship to the warning time for evacuation.
Because a "large release" performance guideline is for use in assessing plant performance, staff believes it should not be dependent on protective action assumptions. Therefore, to reasonably decouple the large release definition from protective action assumptions, while avoiding overly conservative assumptions, the assumption of 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> sheltering, followed by 100% relocation, is a reasonable one for use in determining a large release magnitude.
However, during its efforts to define a large release in terms of equivalent curies of I, staff did evaluate an eight hour sheltering case and a case with no protective actions. The results of these analyses are shown in Table 2.
Table 2 - Effect of Protective Actions Release
' Release sheltering :
% 1* of '
- Time Duration.
Time '
. Core Equivalent!..
Fraction of Core Inventory (Hours)
(Hours)
- (Hours)
Inventory Curies Nobles Iodine
' Cesium ~
Te 1 2*
1 12 60 5.8E+7 100 2
2
.1 2
1 8
67 6.4E+7 100 2.5 2.5
.1 2
1 None' 26 2.5E+7 93 0
0 0
Notes:
- Base'(Proposed) Case Assumes normal activity for seven days -
Enerav of Release:
In addition to the _ magnitude or the quantity of radioactivity released, the likelihood of prompt fatality also depends on the thermal energy and the timing of the release.
Releases having high thermal energy will ascend to elevated levels in the atmosphere and are more dispersed compared with low energy releases close to the ground. For this reason, high thermal energy releases require a greater magnitude of radioactive material to be released to result in a prompt fatality compared to low energy releases.
If a high energy release is assumed, instead of a low energy release, the magnitude of a "large release" would increase from 6x10' to 2x10' equivalent curies of I.
A release of this magnitude at two hours after reactor shutdown would be equivalent to approximately 100% noble gases,14% of iodine and cesium and 4% tellurium.
Since both high energy and low energy releases are possible, the staff used a low energy release in defining a large release.
Timino & Duration of Releases: The timing of the release is important in two ways.
First, if assuming evacuation, the time of release becomes important in relationship to the time and speed of evacuation. Secondly, f.
3
because of the decay of short lived isotopes, a late release will require larger fractions of the core inventories to be released to result in one or more prompt fatalities. When expressed in terms of aquivalent curies I"*, the release magnitude is independent of the. time of the release.
Table 3 shows the effect of different release times on the radioactive material needed to be released to result in one prompt fatality.
Table 3 - Effect of Release Time Release Release Fraction'of Core Inventory' Time Duration' (Hours)
(Hours) '
' Noble's
' lodine.
' Cesium
' Te 2
1 100 2.0 2.0
<1 4
1 100 4
4
<1 12 1
100 8.0 8.0 2
24 1
100 11 11 3
Note:
releases equivalent to'a release of 6E7 equivalent curies During the staff's efforts to define a large release in terms of equivalent curies I"*, the staff assumed a one hour release duration to ensure that all of the releases occurred early in the sheltering scenario.
However, the staff did evaluate the effect of a longer duration release.
The results of these analyses are shown in Table 4.
Table 4 Effect of Release Duration Release-
% I " of. Core Fraction of Core Inventory Duration -
Inventory:
(nxirs) -
Ecuivalent Curies
. Nobles.
Iodine -
Cesium:
Te~
1 60 5.8E+7 100 2
2
.1 2
91 8.7E+7 100 5
5 1
4 99 9.5E+7 100 6.8 6.8 1.7 Other Modelina Parameters:
In addition to the site parameters discussed above, other modeling parameters also were found to affect the results of the calculations.
These parameters include shielding factors assumed during sheltering, binning of the weather data, and grid size used to represent population distribution.
Therefore, if a quantitative definition of a large release is used, further consideration would need to be given to the selection of these parameters.
In an effort to develop a large release magnitude that would be readily useable for assessing plant performance, the staff initially focused on establishing a j
4 J
4.
large release magnitude in terms of fractions of core inventory released to the environment.
It was believed that this would allow relatively easy comparison with the results of Level II PRAs and could later be converted to equivalent curies for use with reactors of significantly different power levels. Using the range of releases developed for the five plants' examined in NUREG-1150 and the LaSalle plant, a set of simplified source terms were developed to reflect a range of release characteristics, including timing, duration, energy, and composition (fractions of core inventory).
These simplified source terms, adjusted to an assumed, enveloping, reactor power level of 3800 megawatts (thermal), and the range of site characteristics discussed above were used in the MACCS code to identify those releases that would most nearly lead to one prompt fatality.
Discussion A. Fraction of Core Inventory As noted above, the staff performed a number of sensitivity studies to determine the effect of varying selected MACCS code inputs. The important parameters that affect the magnitude of a "large release" are the release characteristics (including the energy of the release and the timing of the release with respect to reactor shutdown) and the protective actions assumed. The effects of varying exclusion area size and meteorology were found to be relatively small.
Furthermore, the impact of varying the assumptions of time and duration of a re' ease is dependent on the protective action assumptions and converse ly, the impact of varying protective action assumptions depencs on the time of a release.
When expressed in terms of equivalent curies of I*, the time of release is important, primarily in conjunction with the protective action assumptions.
The sensitivity calculations resulted in a wide range of possible large release magnitudes, when expressed in terms of fraction of core inventory.
Assuming a low energy release occurring at approximately 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> after reactor shutdown and a bounding assumption of no protective actions, a release of 100% of the noble gases, 3% of the iodine and less than 1% of other isotopes would meet the definition of a "large release," i.e, result in one prompt fatality. Such a release is small enough that many plants likely would not be able to show that its likelihood was less than 10-* per year over the spectrum of potential severe accidents.
Assuming an energetic release, instead of a ground level release, the resulting magnitude of a large release would be increased to 100% noble gases,10%
iodine, cesium and tellurium and less than 5% of other isotopes to result in one prompt fatality.
The effects of assuming realistic protective action assumptions are even more pronounced.
If an energetic release occurring at approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after reactor shutdown is assumed and mean NUREG-1150 protective action assumptions are used, a release of 100%
of the noble gases, 25-30% of the iodine and cesium, 10-20% of the tellurium and strontium, and less than 10% of other isotopes would be required to meet the definition of a large release.
A release of this magnitude is estimated to be larger than that which occurred at Chernobyl.
i This wide variation in a potential definition of a 'large release" suggests that a quantitative definition may not be practical.
5
2 Uhile performing this work, it became apparent that a key factor affecting the magnitude of a large release, when expressed in terms of fraction of core inventory, is the timing of the release.
Because of the wide variations in possible "large release" magnitudes and the difficulties envisioned in justifying any single "large release" expressed in terms of fraction of core inventory, the staff subsequently focused on establishing a large release magnitude in terms of equivalent curies of I"' released to the environment. This approach eliminates much of the variation in magnitude that results from timing of the release.
However, the other key factors that can affect the magnitude (energy of release and protective action assumed) remain.
2 B. Equivalent Curies of 1 "
As discussed above, the staff found that the magnitude of a large release, when expressed in terms of fraction of core inventory, varied widely and was very sensitive to assumptions on timing and energy of the release and protective actions taken. In order to reduce the sensitivity of the large release magnitude to the assumptions of timing of release, the staff has also evaluated the magnitude of a large release expressed in terms of curies released.
Accordingly, the staff has evaluated a larp,e release magnitude in terms of equivalent curies released, using I
- as the representative isotope.
The variation of the magnitude, based on-variationsinenergyofreleaseandprotectiveactionsassumed,wasfrom approximately 2x10 curies to 4x10' curies (over an order of magnitude difference). For a candidate "large release" magnitude, the staff assumed a low energy release of I hour duration with the following " representative site" characteristics:
,[
o Exclusion area boundary - 0.4 miles o
Population density - 1000 people per mi' o
Protective actions - 12 hr. sheltering followed by 100%
relocation o
Meteorology - 80" percentile The resulting magnitude of the "large release" is 6x10' curies of I **,
2 which is equivalent to approximately 60% of the core inventory of 1 '2 2
This is equivalent to a release of approximately 100% of the noble gases and 2% of the iodine and cesium at 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after reactor shutdown (less than 1% of any other isotopes released).
At 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after reactor shutdown, this is equivalent to a release of approximately 100% noble-gases, 8% iodine, 8% cesium, and 2% tellurium.
6
)
/psnecy 'o UNITED STATES
~,'n NUCLEAR REGULATORY COMMISSION
^
.E ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
{
g W ASHINGTON, D. C. 20555 April 22, 1993 The Honorable Ivan Selin Chairman U.S. Nuclear Regulatory Commission Washington, D.C.
20555
Dear Chairman Selin:
SUBJECT:
DEFINITION OF A LARGE RELEASE FOR USE WITH SAFETY GOAL POLICY During the 396th meeting of the Advisory Committee on Reactor Safeguards, April 15-17,
- 1993, we discussed the staff's recommendations in regard to the definition of a large release related to the implementation of the Commission's Safety Goal Policy.
During this meeting, we had the benefit of discussions with members of the NRC staff and of the document referenced.
In the draft Commission paper and in the presentation to the Committee, the staff expressed its belief that the development of the definition of a large release is no longer practical or useful and, therefore, it is requesting Commission approval to terminate efforts in this area.
We believe the staff has made a
conscientious effort with this activity and we agree with its basic conclusions.
Our views are as follows:
1.
A large release definition would either represent a
replacement for the existing safety goals or, if made consistent with the quantitative health objectives (QHOs),
would be redundant and unnecessary.
i l
2.
New guidelines being developed for implementing the Safety -
Goal Policy within regulatory analysis and issue prioritization processes adequately meet the originally perceived need for a large release component of the safety goals.
These utilize a core damage frequency (CDF) and a conditional containment failure probability (CCFP).
3.
Plant performance objectives, viz CDP S10~' and CCFP $0.1, provide an easily understandable and adequate surrogate for the QHos and provide quantitative prioritization for two basic aspects of defense in depth (prevention and mitigation).
These could help ensure that a plant does not end up with great core protection but marginal containment performance.
l
E-w m-2L-A-
The Honorable Ivan Selin 2
April 22, 1993 1
i We support the recommendation that the commission approve the staff's proposal to terminate its effort to develop a definition of a large release.
i Sincerely, h
< w,
-(.
Pau.\\ Shewmon-Chai. man
Reference:
Memorandum dated March 11, 1993, from Warren Minners, Director, RES/DSIR, for John T. Larkins, Acting Executive Director, ACRS,
Subject:
ACRS Review of Draft Commission Paper on Large Release Determination, w/ Enclosure i
t I
~
.i o
f x
I 1
1 I
l s
- 1 se