ML16342A537
| ML16342A537 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 10/15/1998 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML16342A536 | List: |
| References | |
| NUDOCS 9810190226 | |
| Download: ML16342A537 (14) | |
Text
8 REOI
~
+
0 Cy ClO I
C Ih0 gO
++*++
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON> D.C. 2055&4001 AEEVLATEFIE E
LAT EL D 0 LO E
B EA BL NY N S
NDEL N
BLO A
CLEA NT N
D
- 2 1.0 D
On January 3, 1997, Pacific Gas and Electric Company (PG&E) declared an Unusual Event at the Diablo Canyon Power Plant due to a mud slide on the county road leading to the plant entrance.
This led to some concern on the part of the NRC staff as to the potential for land slides that could affect Plant Category I structures or equipment, or could potentially isolate the site. Of particular concern were the effects of earthquake vibratory ground motion on soil slope stability following a period of prolonged precipitation. A review of the Diablo Canyon Updated Final Safety Analysis Report (UFSAR) and the NRC Operating License Safety Evaluation Report indicated that the stability of cut slopes adjacent to the plant were analyzed and reviewed as part of the licensing process, but there was no discussion of the stability of natural slopes in either document.
Plant site geologic maps in the UFSAR showed numerous areas marked as landslide deposits.
In addition, an aerial photograph which was taken during the construction of the plant appeared to show a large "landslide scar" on the hillside east of the plant buildings and south of the raw water stoiage reservoirs, with transmission line towers mounted on it. Therefore, the staff requested PG&E to provide an analysis of the landslide hazard in the site vicinityand any potential impact on Category I structures and equipment and other important facilities such as access roads (Ref. 1).
In response to the NRC request, PG8E performed an extensive geologic and geotechnical assessment of slope stability in the plant site area.
The assessment consisted of a review of existing data and documents, aerial and field reconnaissance, interpretation of aerial photography, detailed geologic mapping of selected slopes, kinematic analyses of rock slope stability, and static and dynamic stability analyses of selected soil and rock slopes.
PG&E's assessment covered the stability of the following slopes in the plant site vicinitythat have the potential to adversely affect Category I structures and equipment and other important facilities (Ref. 2):
Cut slope east of Unit 2 and natural slope east of Unit 1.
Slopes in the vicinityof the circulating water intake tunnels and auxiliary saltwater pipes.
Slopes in vicinityof the discharge structure.
Slopes in vicinityof the raw water reservoirs.
98iOf90226 98i015 PDR ADOCK 05000275 P
~
Slopes in vicinityof the 230-kV switchyard and 500-kV switchyard.
Slopes along Diablo Ocean Drive (access road).
For each slope, PG&E evaluated the past slope performance, surface and subsurface drainage control measures, geologic and geotechnical slope conditions (for example, bedrock bedding, fracture and joint orientation and spacing, bedrock and soil strength parameters, and thickness and nature of unconsolidated deposits) and hydrologic conditions (for example, evidence for shallow or perched groundwater, and evidence for past slope saturation).
PG&E's-geologic assessment of the stability of the slopes assumed that strong earthquake vibratory ground motions followed a period of prolonged precipitation.
It also evaluated whether prolonged rainfall would lead to saturation of each slope.
For the slopes east of Units I and 2, PG&E performed new dynamic slope stability analyses, adopting material strength parameters based on a review of previous studies performed for these slopes.
The staff's evaluation of PG&E's stability assessment of the above slopes is given below.
. ~ ~U PG&E performed a helicopter reconnaissance on January 16, 1997, which provided important information on the geologic factors influencing the location and occurrence of landslides.
Most landslides occurred in areas. associated with hillslopes underlain by the Pliocene Squire Member of the Pismo Formation, melange of the Franciscan assemblage, and Miocene diabase rocks (Ref. 2). The January 3, 1997, landslides along the Avila Beach Road occurred as rockfalls and debris flows within sandstone of the Squire Member of the Pismo Formation in wave-undercut sea cliffs, in colluvial-filledswaies on hillslopes, and in areas of preexisting landslides.
Most of the plant site area is underlain primarily by relatively stable bedrock of the Miocene Obispo Formation, a bedrock type that generally is not susceptible to slope instability.
3.0 L
P I
N Io The plant is sited on a large cut pad excavated into bedrock.
Cuts of up to 60 feet were made for the foundation of Unit 2. During construction of the power generating units and auxiliary structures, the lower portion of,the slope along the ma'rgin of the power plant pad east of Unit 2 was excavated to an inclination of about one and a half horizontal to one vertical (1.5: 1), locally requiring cuts as deep as 30 feet. Category I structures and equipment at the toe of the slope include the Unit 2 containment, the fuel handling and auxiliary building, the condensate storage tank, and the refueling water storage tank. The closest approach of these structures to the toe of the slope is greater than 60 feet. The radwaste storage and laundry building is constructed at the base of the slope.
Allof the plant structures are founded on bedrock.
Past Slope Performance PG&E has reported that the interpretation of preconstruction aerial photography (1968) shows that the cut slope is located along the western flank of a geologically stable ridge. The ridge has no evidence of preexisting deep-seated slope instability in the cut slope area.
In addition,
performance of the cut slope has been excellent.
Since construction of the slope in the late 1960s and early 1970s, there has been little erosion or instability of the slope (Ref. 2).
Previous Slope Stability Analyses PG&E evaluated the stability of the cut slope east of Unit 2 previously for static and seismic conditions using two-dimensional, dynamic finite-element analyses.
The dynamic analyses were performed using, as input, ground motions from the double design earthquake, having a peak horizontal bedrock acceleration of 0.4g and a peak vertical acceleration of 0.2g.
Earthquake-induced downslope displacements were estimated by a procedure developed by Newmark (Ref. 3). The computed maximum displacement resulting from the double design earthquake was 10 inches.
The NRC reviewed the results of the analysis and concluded that the results were conservative.
In Supplement 7 of the DCPP SER (Ref. 4), the NRC staff updated its evaluation of the stability of the cut slope using ground motions from the Hosgri event rather than the double design earthquake.
In this evaluation the staff used the same Newmark-type displacement analysis procedure that was used in the FSAR. The static factor of safety and yield acceleration also were obtained from the FSAR. As input for these computations, the horizontal acceleration recording made at Pacoima during the 1971 San Fernando earthquake was scaled and its duration extended to correspond to a magnitude 7.5 earthquake. The computed maximum displacement from this evaluation was 19 inches along the potential slip surface.
The staff then concluded that since seismic Category I structures and equipment are about 30 feet removed from the toe of the slope, this revised estimate of slope movement would have no effect on safety.
Based on its review of such previous detailed slope stability evaluations by its own consultant and by the NRC staff as described above, PG&E concluded that the movement of the slopes during the occurrence of the Hosgri event willnot damage seismic Category I structures or equipment.
Present Evaluation of Seismic Stability of Cut Slope East of Unit 2 by PG&E In its current evaluation of the seismic stability of the cut slope east of Unit 2, PG&E reviewed available data and studies performed to date to assess groundwater conditions within the slope and to estimate the engineering properties of the slope materials. Ground motions were estimated for a magnitude 7.2 earthquake on the Hosgri fault zone, based on long term seismic program (LTSP) evaluations (Ref. 5). Using these parameters, and assuming there is a temporary perched water table on bedrock at its interface with surficial deposits in the slope, PG&E evaluated the seismic stability of the slope for ground motions representative of the LTSP 84th percentile evaluation spectra (Ref. 5). Earthquake-induced permanent deformations of the cut slope were estimated to be only 0.2 inches using Newmark-type analysis procedures (Ref. 2). This estimate was made assuming that the undrained strength of the sandy clays of the Pleistocene colluvium was not reduced due to earthquake shaking.
Makdisi and Seed (Ref.
- 6) have shown that stiffclays and dense sands do not suffer significant reduction in undrained strength due to cyclic loading, and that such soils would likely retain at least 80 percent of their initial undrained strength after being subjected to significant cyclic loading. Accordingly, the stability of the slope was also evaluated using an undrained strength that was reduced to 80 percent of its static undrained value. The corresponding estimated permanent deformation of the cut slope east of Unit 2 increased to about 2.5 inches from 0.2 inches.
On the basis of PG&E's current investigation and previous investigations of the cut slope, PG&E has concluded that the cut slope east of Unit 2 may sustain small deformations during a postulated maximum earthquake on the Mosgri fault zone. The maximum computed displacement is about 2.5 inches for the input time-histories representative of the 84th percentile ground motions from a magnitude 7.2 earthquake.
Because there is a distance of more than 60 feet between the toe ofthe slope and the closest Category I structure (the condensate storage tank), PG&E concludes, and the staff agrees, that potential movements of this slope under earthquake loading following periods of heavy precipitation willnot damage any Category I structures or related equipment.
This conclusion is valid even for the previously computed higher displacement of 19 inches (Ref. 4) in view of the distance of more than 60 feet between the toe of the cut slope east of Unit 2 and the closest Category I structures mentioned above.
N t ral I
PG&E's interpretation of aerial photographs taken prior to site grading indicates that there is no evidence of prior slope failure. PG&E has further reported that shotcrete applied along the excavated toe of the slope in 1995 to prevent surficial erosion and creep has performed well, and that there is no evidence of slope movement (Ref. 2) ~ The surficial soil above the shotcrete area, however, appears to be undergoing active surficial creep and localized surficial slumping.
Some distress of surficial pipes due to the downslope creep of the upper 1 to 3 feet of soil is noticed. Also a 5-to 8-foot thick, 160-foot-wide slump is present in the oversteepened cut above the paved cable trough where it traverses the upper slope. According to PG&E (Ref. 2),
this slump is confined to the uppermost slopewash soils in the cut slope face, and toes along inboard edge of the cable trough, and that no distress or evidence of displacement was observed in the cable trough or pavement below the slump.
The geologic conditions of the natural slope east of Unit I generally are similar to those of the cut slope east of Unit 2 (Ref. 2). PG&E states that this natural slope appeared to be dry, similar to the cut slope east of Unit 2, and that no evidence of shallow ground water was observed in the slope despite the heavy 1996/1997 winter storms.
However, for the purposes of evaluating slope stability, PG&E conservatively assumed a temporary perched water table on bedrock following periods of heavy, sustained rainfall. PG&E performed a dynamic slope stability analysis of this slope using the same procedure as was used in the analysis of the cut slope east of Unit 2, and found that the maximum displacement was about 5 inches for the seismic loading representative of the 84th percentile ground motions from a magnitude 7.2 earthquake.
This amount of seismically-induced displacement is not a safety concern, since there is a distance of about 30 feet between the toe of the natural slope and the closest Category I
structure (the condensate storage tank behind Unit 1).
SI it V'I PG&E performed geologic mapping and stability analyses for the slopes in the vicinityof the discharge structure and determined that the discharge structure is founded on competent, stable bedrock, and that it is not vulnerable to damage from sea cliffinstability under the combined effects of heavy rainfall and earthquake shaking (Ref. 2).
Similarly, PG8 E has determined by geologic investigations that the stability of the circulating water intake tunnels and auxiliary saltwater (ASW) pipes has been good; it observed no indications of past or incipient distress or damage from foundation or slope movements.
Recently, in connection with the installation of two new ASW bypass pipes, the staff evaluated the liquefaction potential and the slope stability of the ASW slope, and concurred with PG&E's conclusion that this slope is stable under the plant's SSE.
yean B
Wtr In response to the staff's query about a "large landslide scar" south of the raw water reservoirs (Ref. 1), PG&E has stated that the "scar" is, in fact, a large (900-foot-long, 300-foot-high) cut slope that was created on the hillslope above the reservoirs during excavation of borrow material for the large embankment fills beneath the 230-kV and 500-kV switchyards.
PG8 E evaluated the stability of this cut slope by performing geologic mapping and kinematic stability analyses, and determined that both the cut slope above the reservoirs and the natural slope below the reservoirs are stable under combined high rainfall and earthquake loading, and pose no hazard to the raw water reservoirs.
Vi i PG8 E's analyses of the slopes in the vicinityof these non-Category I structures indicate that they are generally stable, and that there was no significant hazard to switching facilities during the earthquake ground motion following a period of prolonged rainfall. However, PG&E predicts that transmission towers on the landslide mass above the switchyard could be affected by slope movements caused by the combined effects of high rainfall and strong earthquake ground shaking.
In such an event, PG&E expects to be able to repair or replace these towers.
I e
Alon Di Driv c
v'I B
h PG8E is responsible for the maintenance of the 7.5-mite long Diablo Ocean Drive that provides access to the plant. Since construction of this road in 1968-1969, it has generally performed well (Ref. 2). Slope stability issues typically have involved minor soil slips and sloughing along natural and cut slopes, and minor settlement of embankment fills causing localized cracking of the asphalt pavement.
Two significant slope failures leading to the temporary closure of the road is reported to have occurred since its construction in 1968-69.
Following heavy rains in 1997, pavement cracking arid offset of several inches in the roadbed occurred in an area of historic movement of this road (Ref. 2). This condition was repaired in February 1997 by construction of a large fill-buttress at the toe of the slide, and regrading the head wall area of the slide.
In general, PG&E's investigations have shown that both cut slopes and embankment fillsalong the access road have performed well since their construction, and natural slopes are also stable.
Existing landslides near the road pose no hazard to the road.
However, PG8E has identified three areas of potential natural slope instability along Diablo Ocean Drive that may impact the road under the combined effects of earthquake loading and heavy, sustained rainfall.
These areas are being monitored by PG8 E.
~
A Several landslides occurred in January 1997 along Avila Beach Road which is a county road leading to the DCPP entrance gate.
These slides caused the closure of the road for a few hours.
PG&E reports that the coastal bluffs along Avila Beach Road have experienced landslides in the past.
Slope stabilization measures have been undertaken in the past.
PG&E reports, however, that slope failures along Avila Beach Road represent an on-going issue along the road and that additional landsliding from the coastal bluffs can be expected.
Even though such landslides can hamper easy access to the plant site, past experience has shown that access to the site can be reestablished in a short time. For example, after the debris from two landslides temporarily blocked this road in the winter of 1996-1997, the road was reopened in about 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, and normal traffic reestablished within 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> (Ref. 2). It is further learned from the licensee, that there is an alternate unpaved access road leading to the plant, and also, that any essential plant personnel can be transported by helicopter to the site.
4.0 Based on its detailed investigations related to the landslide potential at Diablo Canyon Power Plant (Ref. 2), PG&E has concluded, and the staff agrees, that earthquake loading following periods of prolonged rainfall will not produce any significant slope failure that can adversely affect the Category I structures including the intake and discharge structures at the site.
In addition, PG&E has concluded that potential slope failures under such conditions willnot adversely impact other important facilities such as the raw water reservoirs, the 230- and 500-kV switchyards.
A close review of PG&E's submittal (Ref. 2) by the staff indicates that PG&E has assembled a
significant amount of data related to the previous landslides and soil properties, and has also performed several detailed analytical studies of the slope stability issue.
Therefore, the staff concurs with PG&E's conclusions described above.
5.0 2.
Letter from Steven D. Bloom, NRC, to Gregory M. Reuger, PG&E.
Subject:
Request for Additional Information - The Landslide Potential at Diablo Canyon Power Plant, January 31, 1997.
Letter from Gregory M. Reuger, PG&E,'to NRC.
Subject:
Diablo Canyon Units 1 and 2 Response to Request for Additional Informatio'n Regarding the Landslide Potential at Diablo Canyon Power Plant, April 28, 1997.
3.
Newmark, N.M., "Effects of Earthquakes on Dams and Embankments", Geotechnique, v.15, pp.139-159, 1965.
4.
U.S. NRC Supplement 7 to the Safety Evaluation of the Diablo Canyon Nuclear Power Station Units 1 and 2, May 1978.
5.
PG&E's Final Report to USNRC on Diablo Canyon Long Term Seismic Program:
Chapter 4 Characterization of Ground Motions, 1988.
4 A.
6.
Makdisi, F.l., and Seed, H.B., "Simplified Procedure for Estimating Dam and Embankment Earthquake-induced Deformation", ASCE Journal of Geotechnical Engineering, v. 104, No. GT7, July 1978.
Principal Contributor:
R. Pichumani Date October 15, 1998
I f 0
~
IL
~
4