ML17310A665
| ML17310A665 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 09/27/1993 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML17310A664 | List: |
| References | |
| NUDOCS 9310010182 | |
| Download: ML17310A665 (9) | |
Text
~gS REOIj (4
0 UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION AND THE OFFICE OF NUCLEAR REACTOR RESEARCH RELATING TO THE REDUCTION OF REVIEW LEVEL EARTH UAKE FOR PALO VERDE FROM 0.5G TO 0.3G RIZONA PUBLIC SERVICE COMPANY PALO VERDE NUCLEAR GENERATING STATION UNITS 1
2 AND 3 DOCKET NOS.
50-528 50-529 AND 50-530
1.0 BACKGROUND
In response to the Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities 10 CFR 50.54(f)
(Generic Letter 88-20, Supplement 4), Arizona Public Service Company (APS),
on December 26,
- 1991, submitted a probabilistic seismic hazard (PSH) study (Reference
- 1) which was performed for the Palo Verde Nuclear Generating Station (Palo Verde) site.
The methodology used to obtain the PSH data is essentially the same as that used by the Electric Power Research Institute (EPRI) to estimate the PSH for the nuclear power plants in the eastern U.S.
(Reference 2).
The results of this study were intended to justify APS's request for a reduction of the review level earthquake (RLE) for the Palo Verde site from 0.5g [specified in NUREG-1407, Procedural and Submittal Guidance for the Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities, Table 3.1] to 0.3g.
The NRC determined that the study and its supporting data were not adequate to justify this reduction in the RLE for Palo Verde (References 3 and 5).
On August 4,
- 1992, and Hay 6,
- 1993, APS submitted additional material (References 4 and 6) to improve the technical level and quality of the study.
The staff reviewed the additional material submitted and found that the study in its present form (Revision 2) and its conclusions are adequate to justify the reduction in RLE for Palo Verde from 0.5g to 0.3g.
Additionally, the staff found that it is reasonable for APS to perform a full scope 0.3g IPEEE of one reactor unit and perform focused scope reviews of the other two units.
REVIEW OF SEISMIC HAZARD ANALYSES The staff has reviewed all the material submitted by the licensee (References 1, 4, and 6) in support of its request to reduce the IPEEE RLE for Palo Verde from 0.5g to 0.3g.
The staff held a meeting on February 16, 1993, for APS to present the results of these
- analyses, which included sensitivity studies to assess the ramifications of these studies (Ref.
1 and 4).
The staff also held a conference call on July 2, 1993, to discuss additional staff questions on 9310010182 930927 PDR ADOCK 05000528 P
'PDR
C'f C
h
'I 4 l'
Reference 6.
The seismic hazard analyses at the Palo Verde site consisted of the following:
(a)
Five earth science teams (EST), which included nationally recognized seismicity experts, were used for assessing seismicity, and sensitivity studies of source zone distributions were made in accordance with NRC requests.
(b)
(c)
Ground motion models, applicable to the Palo Verde site, were developed by experts with extensive experience in empirical ground motion modelling.
Soil damping and modulus reduction factors suitable for the Palo Verde site were used for the soil amplificatidn studies.
Historical Seismicit A unique earthquake catalog was developed from four published catalogs which were reviewed by Professor David Brumbaugh of Northern Arizona University to ascertain the completeness.
Some earthquakes were relocated using recently developed crustal velocity models.
This procedure resulted in the relocation of several earthquake epicenters from within the Arizona border to locations outside Arizona.
In particular, the Fort Yuma earthquake of November 29, 1852 (H-6.6),
was relocated to 32.5 degree N., 115.0 degree W.
This relocation agrees with that reported in the earthquake catalog obtained by the NRC staff from the U.S. Geological Survey (USGS) on March 12, 1992.
A USGS earthquake listing of an earthquake in the same area and on a slightly different date, November 9, 1852 (H=7.0), 33.0 degree N., 114.5 degree W., was omitted from the APS catalog.
Because of the date (1852), it is quite possible that these reports of a large earthquake (6.5<
H >7.0) spaced in time by 20 days, obtained from two different catalog sources according to the USGS listing, were in fact reports on one occurrence.
Seismic Zonation The seismic hazard analyses (Ref.
1 and 4) were originally performed by two earth science
- teams, Geomatrix Consultants, Inc.
and J.
M. Montgomery Consulting Engineers.
The additional seismic hazard analyses submitted by the licensee on May 6, 1993 (Reference 6), incorporated three additional earth science teams (Bechtel Corporation, Dames
& Moore, and Woodward-Clyde Consultants),
and consisted of experts in the fields of seismology and geology.
It also evaluated the seismic zonation and seismicity parameters suggested by the NRC staff (Enclosure 3).
The revision increased the number of seismic zonation models from the 2 in the initial study to 9, each with its estimated seismicity parameters.
The methodology used to obtain probabilistic seismic hazard data was essentially that developed by the Electric Power Research Institute (Reference 2).
However, in the case of this study the earthquake activity rates (a-values) and the slopes of the recurrence rates (b-values) were established specifically for the seismic zones (no smoothing of b-values and no prior b-values were used).
For the active faults identified, slip rates and activity rates (probabilities of activity) were
specified.
The slopes (b-values) were tested against the APS earthquake catalog using magnitude 5 earthquakes and greater.
For those seismic zones where ranges of activity rates and b-values were specified, a weighting scheme was utilized to encompass the range of uncertainty (i.e. multiple hazard curves were generated each with a predetermined weighing factor).
Ground Notion and Soil Am lification The ground motion models used in the initial study (Ref. I) were mostly developed based on attenuation relationships from California data.
Additional
- studies, presented in Reference 6, incorporated locally applicable ground motion models recommended by the staff (modified Campbell models applicable to southern Arizona) and models recommended by Abrahamson.
In addition, the study project modified the assessment of the ground motion amplification factors applicable to the Palo Verde site.
These additional analyses increased the number of attenuation models used from 4 to 9.
Four of these models were developed specifically for the southern Arizona area soil sites.
The other five models were developed for southern Arizona rock sites and were augmented with soil amplification factors to incorporate the Palo Verde soil site conditions.
To make the ground motion models developed for rock sites (Joyner-Boore,
- Campbell, and Abrahamson) applicable to the Palo Verde site, the study utilized the method used by EPRI to obtain (generic) amplification factors (Reference
- 2) with the exception that soil parameters specific to the PVNGS site were applied.
Four of the ground motion models (Abrahamson and Campbell) were specifically developed for the southern Arizona geological profile.
A weighing scheme of 0.67 for the five rock/site-specific-soil-amplification models and 0.33 for the four regional soil models was employed.
2.0 DISCUSSION EVALUATION The rationale for the assignment of operating nuclear power plants, in the contiguous U.S., to the different review level earthquakes for the
- IPEEE, was based upon a relative ranking procedure.
This procedure utilized a seismic hazard measure consisting of the combined (weighted) exceedance probabilities for three spectral frequencies using the NUREG/CR-0098 median response spectrum (2.5Hz, 5.0Hz, 10.0Hz) and the (weighted)
PGA (Reference 8, Appendix A).
The results were then tested for agreement between the three eastern U.S.
seismic hazard studies (Lawrence Livermore National Laboratory (LLNL) hazard studies with 5-G experts and 4-G experts and EPRI hazard study).
The APS study for Palo Verde based its conclusion on comparing the Palo Verde seismic hazard results to those obtained by EPRI for the 57 eastern U.S. sites.
Each EST submitted one or more seismic zonation maps showing source
- zones, within which the probabilities of earthquake occurrences were considered to be uniform, together with fault systems where the activities were restricted to the faults indicated (with, in certain instances, a measure of background activity of lesser magnitude earthquakes).
P P'
The results of the study indicate that the host zone (i.e. the zone in which the Palo Verde site is located; namely, the Southern Great Basin zone (Geomatrix),
Sonoran Basin and Range, zone (JHH), Southern Basin and Range zone (Bechtel and Woodward
& Clyde),
Sonoran zone (Dames
& Hoore)) is the major contributor to the seismic hazard at the Palo Verde site for all ESTs.
The significance of this finding is that earthquake occurrences at distances less than 100 km have a dominating influence on the seismic hazard at the PVNGS site.
This is noteworthy in light of the maximum magnitudes (Hmax) estimated for nearby earthquakes which vary from team to team from a low of 4.5 to a high of 6.75. Other significant contributors to the hazard at the Palo Verde site appear to be the San Andreas fault system in southern California, and the Cerro Prieto, and Laguna Salada faults extending from southern California into northern Baja California.
The nine ground motion models used by the study vary significantly in their predictions of attenuation over distance.
For instance, the difference in exceedance probabilities at the I Hz spectral velocities for the Campbell soil model and that of the Joyner-Boore soil model vary by as much as'ne order of magnitude.
However, the manner in which the ESTs chose to use (weight) the attenuation functions reduced these differences considerably.
The diversity and range of the scientific opinions exhibited by the earth science teams and the ground motion experts used for this study project is very similar to that experienced in the LLNL and EPRI seismic hazard characterization studies.
This fact has improved the applicability of these considerations to the Palo Verde site.
As a result, the comparisons of the Palo Verde site with eastern U.S. plant sites, have become more realistic.
In comparisons of hazard measures between Palo Verde and eastern
- sites, the Palo Verde plant is now located in the upper range of eastern plants in the 0.3g bin.
3.0 CONCLUSION
In summary, the staff has concluded that APS has performed a suitable analysis of seismic hazards at the Palo Verde site in comparison to seismic hazards at plants in the eastern U.S.
The analysis shows that Palo Verde is comparable to eastern U.S. plants in the 0.3g bin, and that a lowering of the Review Level Earthquake from 0.5g to 0.3g is justified.
Additionally, in accordance with APS's recommendation, it would be reasonable to perform a full scope review of one unit and perform focused scope reviews of the other two units, provided that any differences between the units are considered.
Principal Contributors:
G. V. Giese-Koch, NRR E. Zurflueh, RES Date:
September 27, 1993
P V
REFERENCES:
Letter from W. F.
Conway of Arizona Public Service Co. to NRC,
Subject:
Response
to Generic Letter 88-20, Suppl. 4, "Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities," dated December 26, 1991.
2.
EPRI NP-4726-A, "Seismic Hazard Hethodology for the Central and Eastern United States."
3.
5.
6.
7.
Letter from C. H. Thompson of NRC to W. F.
Conway of Arizona Public Service Co.,
Subject:
Review of Response to Generic Letter 88-20, Suppl.
4, "Individual Plant Examination of External Events Palo Verde, Nuclear Generating Station",
dated June 17, 1992.
Letter from W. F.
Conway of Arizona Public Service Co. to NRC,
Subject:
Supplemental
Response
to Generic Letter 88-20, Suppl.
4, "Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabili,ties, Pursuant to 10 CFR 50.54(f)", dated August 4, 1992.
Letter from C.
H. Trammell of NRC to W. F.
Conway of Arizona Public Service Co.,
Subject:
Review of Response to Generic Letter 88-20, Suppl.
4, "Individual Plant Examination of External Events Palo Verde Nuclear Generating Station", dated February 9, 1993.
Letter from W. F.
Conway of Arizona Public Service Co. to NRC,
Subject:
Review Level Earthquake for Individual Plant Examination of External Events (IPEEE; Generic Letter 88-20, Suppl. 4), dated Hay 6, 1993.
Hemorandum for G. Bagchi from C. Ader of June 7,
- 1993, "Review request of Palo Verde's Supplemental
Response
to Generic Letter 88-20, Suppl.
4 II 8.
NUREG-1407 "Procedural and Submittal Guidance for the Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities."
p
~
~
t F,
2 tl t