ML19254D865
ML19254D865 | |
Person / Time | |
---|---|
Site: | Perry |
Issue date: | 10/31/1979 |
From: | Gudikunst P, Santamour W, Schultz L GILBERT/COMMONWEALTH, INC. (FORMERLY GILBERT ASSOCIAT |
To: | |
Shared Package | |
ML19254D864 | List: |
References | |
2063, NUDOCS 7910300269 | |
Download: ML19254D865 (400) | |
Text
I i
BEDROCK DEFORMATION :
IN THE COOLING L WATER TUNNELS t
i Perry Nuclear Power plant
[
Units 1 & 2 i l
l i
VOLUME I l P00R !.lPdGliQf THE CLEVELAND ELECTRIC !
ILLUMINATING COMPANY :
I Gilbert / Commonwealth 79103nstA ;
GAI REPORT No. 2063 Prepared for Cleveland Electric Illuminating Co.
BEDROCK DEFORMATION IN THE COOLING WATER TUNNELS PERRY NUCLEAR POWER PLANT NORTH PERRY, OHIO OCTOBER 1979 Compiled by L. D. Schultz Project Geologist Perry Nuclear Power Plant Gilbert Associates, Inc.
Reviewed by W. J. Santamour Geotechnical Section Supervisor Gilbert Associates, Inc.
Approved for Release by P. B. Gudikunst Project Manager Perry Nuclear Power Plant Gilbert Asociates, Inc.
124)002 -
GAI Report No. 2063 THE CLEVELAND ELECTRIC ILLUMINATING COMPANY PERRY NUCLEAR POWER PIANT BEDROCK DEFORMATION IN THE COOLING WATER TUNhTLS PERRY NUCLEAR POWER PLANT NORTH PERRY, OHIO OCTOBER 1979 Compiled by: M . 1 3
L. D. Schultz Project Geologist Gilbert Associates Review'd by: g_ -
W J. Santamour Geotechnical Section Supervisor Gilbert Associates, Inc.
Approved by: ,
P. B. Gudikunst Project Manager Gilbert Associates, Inc.
12(j003
TABLE OF CONTENTS Page_
1.0
SUMMARY
1
2.0 INTRODUCTION
3 2.1 STATEMENT OF PROBLEM 3 2.2 INVESTIGATIVE CHRONOLOGY 5 2.3 GEOLOGIC SETTING 9 3.0 METHODS OF INVESTIGATION 13 3.1 GEOLOGIC 14 3.1.1 LITERATURE REVIEW AND PERSONAL COMMUNICATIONS 14 3.1.2 EXPLORATORY BORINGS 15 3.1.3 SHORELINE RECONNAISSANCE 17 3.1.4 VIDEO EXAMINATION OF LAKE BOTTOM FEATURES 17 3.1.5 DETAILED GEOLOGIC MAPPING IN THE INTAKE AND DISCHARGE TUNNELS 18 3.1.6 MICR0 CRACK ANALYSIS 19 3.1.7 WATER ANALYSIS 19 3.2 GEOPHYSICAL STUDIES 20 3.2.1 EVALUATION OF PUBLISHED AND UNPUBLISHED DATA 20 3.2.2 MAGNETIC SURVEYS 20 3.2.2.1 Offshore 20 3.2.2.2 Onshore 20 3.2.3 BOREHOLE LOGS 21 3.2.4 IN-SITU VELOCITY MEASUREMENTS 22 3.3 EVALUATION OF LOCAL SEISMICITY ACTIVITY AROUND THE PERRY NUCLEAR POWER PLANT SITE 23 3.4 IN-SITU STRESS MEASUREMENTS 23 12O 004 TABLE OF CONTENTS (Continued) fage 4.0 RESULTS 24 4.1 GEOLOGIC 24 4.1.1 LITERATURE REVIEW AND PERSONAL COMMUNICATIONS 24 4.1.2 EXPLORATORY BORINGS 25 4.1.3 SHORELINE RECONNAISSLNCE 27 4.1.4 VIDEO EXAMINATION OF LAKE BOTTOM FEATURES 28 4.1.5 DETAILED GEOLOGIC MAPPING IN THE INTAKE AND DISCHARGE TUNNELS 28 4.1.5.1 Stratigraphy 28 4.1.5.2 Tunnel Structural Geology 31 4.1.5.2.1 Intake Tunnel Structure 31 4.1.5.2.2 Discharge Tunnel Structure 32 4.1.6 MICR0 CRACK ANALYSIS 34 4.1.7 WATER ANALYSIS 36 4.2 GEOPHYSICAL STUDIES 37 4.2.1 EVALUATION OF PUBLISHED AND UNPUBLISHED GE0'HYSICAL DATA 37 4.2.2 MAGNETIC SURVEIS 38 4.2.3 BOREHOLE LOGS 38 4.2.4 IN-SITU VELOCITY MEASUREMENTS 39 4.3 CALCULATION OF LOCAL SEISMICITY ACTIVITY AROUND THE PERRY NUCLEAR POWER PLANT SITE 39 4.4 IN-SITU STRESS MEASUREMENTS 40
5.0 CONCLUSION
S 41
6.0 REFERENCES
44 TABLE 1 ANALYSES OF TUNNEL FAULTING SEEPAGE 46 pf 005 LIST OF FIGURES FIGURE NO. TITLE 1 Location Map, Perry Nuclear Power Plant 2 Geologic Map of Tunne; Excavations (23 sheets) 3 Tunneling Plan 4 Bedrock Geologic Map of Northeastern Ohio 5 Glacial Map of Northeastern Ohio 6 Structural Contour Map - Top of Big Lime 7 Strucural Contour Map - Top of Packer Shell 8 Isopach Map of Big Lime and Niagaran Shale 9 Fault and Outcrop Location Map 10 Schematic Northwest-Southeast Cross Section, Perry Nuclear
~
Power Plant 11 TX-Series Boring Plan 12 Map of Shoreline Survey 13 Location Map, Video Survey Lake Erie Bottom 14 Longitudinal Section, Discharge Tunnel 15 Longitudinal Section, Intake Tunnel 16 Location Map, Offshore Magnetic Survey 17 Location Map, Onshore Magnetic Survey 18 Location Map, Seismic Spreads, Intake Tunnel 19 Location Map, Seismic Spreads, Discharge Tunnel 20 Schematic Map, Lake Bottom Fractures 21 Sketch of Facies Relationships among the Huron, Chagrin, &
Cleveland Shales 22 Detailed Stratigraphic Section, Intake Tunnel East Wall Station 10+30-10+40 23 Detailed Stra".1 graphic Section, Discharge Tunnel East Wall Station 11+40-11+46 24 Detailed Stratigraphic Section, Discharge Tunnel East Wall Station 13+22-13+28
. 006
-lii-
LIST OF FIGURES (Continued)
FIGURE NO. T_ I]'LE 25 Geologic Structure Map, Intake & Discharge Tunnel Faults 26 Intake Tunnel Wall Maps, Stations 10+25-10+95 27 Discharge Tunnel Wall Maps, Stations 13+00-12+00 28 Discharge Tunnel Wal; Maps, Stations 11+40-12+00 29 Geologic Maps, Intake & Discharge Tunnels 30 Detailed Map, Intake Tunnel Faul.
31 Photographs, Structural Details, Intake Tunnel 32 Detr.iled Map, Discharge Tunnal Fault 33 Photographs, Structural Details, Discharge Tunnel 34 Offshore for Shipborne Magnetic Profile 14 35 Offshore for Shipborne Magnetic Profiles 10 and 12 36 Offshore for Shipborn; Magnetic Profiles 6 and 8 37 Offshore for Shipborne Magnetic Profiles 2 and 4 38 Offshore for Shipborne Magnetic Profiles 0 and 3 39 Offshore for Shipborne Magnetic Profiles SA and 7 40 Offshore for Shipborne Magnetic Profiles 9 aal 11 41 Offshore for Shipborne Magnetic Profiles 13 and 15 42 Offshore for Snipborne Magnetic Profiles 17 and 19 43 Onshore for Land Magnetic Profile 1S-A 44 Onshore for Land Magnetic Profile IS 45 Onshore for Land Magnetic Profile IE 46 Onshore for Land Magnetic Profile 2S 47 Onshore for Land Magnetic Profile 3S 48 Onshore for Land Magnetic Profile 3S-A 49 Borehole Logs - Gamma / Sonic, TX Borings 2,3,4,5,6,7 50 Borehole Logs - Gauna/ Sonic, 1;* Borings 8,9,10 12yj007
-iv-
LIST OF APPENDICES APPENDIX A A Study of the Microcracks Associated with Faulting at the Perry Nuclear Power Plant Site APPENDIX B A Study of the Isotopic Composit. ion of Water from the Fault in the Intake and Discharge Tunnels at the Perry Nuclear Power Plant Site APPENDIX C Geophysical Methods APPENDIX D Independent Reviews of Cooling Water Tunnel Faulting - Perry Nuclear Power Plant APPENDIX E Stress Measurements Hydrofricturing Technique - Perry Nuclear Power Plant APPENDIX F TX-Series Geologic Logs APPENDIX G Consolidation Tests on Fault Gouge Samples, Perry Nuclear Power Plant APPENDIX H Evaluat. ion of Local Seimicity Around the Perry Nuclear Power Plant Site 12Dj008
-V-
1.0
SUMMARY
Geological, geophysical, and seismological studies were conducted on and in the vicinity of a fault observr a in the intake and discharge tunnels at the Perry Nuclear Power Plant .te of The Cleveland Electric Illuminating Company.
The Perry site is locate,on the shore of Lake Erie, approximately 35 miles northeast of Cleveland. The general location of the site is shown on Figure 1.
A comprehensive investi;ative program evolved as a result of the bedrock deformation exposed during the excavation phase of tunnel construction.
Deterministic fault study objectives, extent, origin and age, were realized as a consequence of a series c.. :nterrelated geologic and geophysical research and engineering. The nature of fault plane geometry and its gouge and mineralogical as well as chemical constitutents were studied. After site specific data had been assembled, the lo:alized anomalous deformation was interpreted in context of its regional geologic setting.
The extent of faulting was defined on the basis of the following: (1) planned, tunnel mapping program (scale 1:120); (2) detailed mapping of tunnel deformation segments (scale 1:12); (3) exploratory borings; (4) geophy..ical logging of borings; (5) shoreline reconnaissance; (6) offshore magnetic survey; (7) lake bottom reconnaissance mapping and review of seismic track line data; and (8) comparative isotopic analyses of Lake Erie water and fault seepage.
Fault zone gouge and fractured rock samples were obtained foe X-ray diffraction, clay-mineralogical determinations, SEM (scanning electron microscope) microcrack analysis, .ud miscellaneous engineering property determinations including consolidation pressure analysis. No radioactive isotopes, which could have been dated, were identified in fault zone samples.
With respect to the site area and locale studies, the following were performed or prepared: (1) in-situ borehole (TX-11) stress measurements to determine existing site stress field orientation and magnitude; (2) structural contour maps of " Big Lime" upper and basal (-50 ft) horizons and isopachous map of intervening interval for Lake and portions of adjacent Ashtabula, Geauga and 12$[f009 Cuyahoga Counties; (3) evaluation of microseismicity in northeastern Ohio; (4) literature and field review of area salt mines and interviews with mine personnel (Mr. Jaroslav Vaverka, resident mining engineer, Cleveland mine, International Salt Company and Mr. B. C. Cummings, resident chief engineer, Painesville mine, Morton Salt Division of Morton Norwick).
Independent opinions based on their field inspection of the tunnel deformation and literature review were obtained from the following geologists recognited for their expertise in the indicated disciplines:
Dr. Robert G. LaFleur Pleistocene Geology and Sedimentology Rensselaer Polytechnical Institute Mr. James Murphy Areal Geology and Stratigraphy of Northeastern Ohic Ohio Historical Society Dr. Barry Voight Structural Geology Penn State University It is concluded on the bases of data and interpretation of the aforementioned studies and other site and regional geological, geophysical, and seismological information that the last movement on the cooling water tunnel bedrock deformation was not tectonic. It occurred during Pleistocene time probably associated with deglaciation-rebound rather than ice advance compression. On the basis of geomatry alone it is possible that the initial deformation was a pre-Pleistocene event. The presence of the fault deformation intersecting the tunnels was considered during design review and redesign was not required.
12f j010
2.0 INTRODUCTION
2.1 STATEMENT OF PROBLEM Tunnel excavstions for the plant cooling water system exposed low-angle thrust faulting of minor displacement in the Chagrin shale beneath Lake Erie. The presence of faulting within the intake and discharge tunnels did not greatly hinder tunneling operations. Additional rock bolts and tunnel shields were installed in tunnel fault segments to insure crown stability. Methane and water, which had been stored within fault zone fracture porosity, were discharged upon intersection of the intake tunnel fault segment by a horizontal exploratory boring drilled in advance of the tunnel boring machine.
Both conditions were short term, within anticipated limits and controlled by norma! pumping and ventilation.
Geologic mapping (scale 1:120) of the tunnels was conducted concurrent with tunneling consistent with PSAR (Preliminary Safety Analysis Report) commitments (see Figure 2, 24 sheets). After faulting had been intersected by the tunnel boring machine and the bedrock mapped, preliminary interpretations and the mappir.g data were forwarded to the NRC (Nue'- r Regulatory Commission) in a timely manner. The extent, age, and origin of faulting were nct well understood following its initial encounter within the intake tunnel. More than four months elapsed between faulting exposed in the intake and discharge tunnels, respectively.
The fault plane exposed in the intake tunnel subsequently has been determined to have a strike of approximately N51 E which projects in the vicinity of the discharge tunnel deformation. Based on the similarity of structural style, flexural sli and brittle failure attributable to compression, and assuming minor warping of the fault plane along its strike, it is probable that the fault plane is contin'ious between both tunnels. This determination could not be concluded prior to completion of all tunnel excavations which was accomplished nearly six months af ter exposing the first deformation.
12Tl0i1 I
The origin of the deformation could not be readily determined on the basis of known regional geology. Results of site and regional studies (field and literature) conducted during the preconstruction phase are reported in the PSAR. On the basis of these studies and the opinions of university professors, bedrock in northeastern Ohio is not known to have undergone significant farlttug. Professors contacted are listed as follows:
Prof. Legeae J. Synuk, Kent State University, Prof. Murray R. McCosas, Kent State University, Prof. Tom Lewis, Cleveland Sta.. University, and Prof. Charles M. Somerson, Ohio State University.
In this area a nearly ubiquitous veneer of glacial and glaciolacustrine deposits obscure bedrock except where incised by stream erosion. Accessible outcrops do not reveal evidence of having experienced tectonism, either late Paleozoic associated with the Alleghenian (Appalacian) Orogeny or any other.
Subsurface data, geological and geophysical, do not imply the presence of a regional fault system which could have been interpreted to be genetically related to tunnel faulting. The general lack of information to the contrary suggested that this portion of the Central Lowland Physiographic Province is tectonically stable having undergone little if any tectonic deformation.
Shallow bedrock deformation, consisting of small-scale folding and faulting, had been revealed as a consequence of plant foundation excavation during 1975 and 1976. It has been demonstrated by field relationships that this deformation was caused by the direct action of late Wisconsinan glaciation.
Similarity of evidence of glacially induced deformation has been found elsewhere in the saae bedrock unit within northeastern Ohio.
Within this context, investigations were undertaken to determine the lateral and vertical extent of the tunnel fault, origin and age of deformation, and effect that this def emation could have had on the tunnel design. Many conventional age dating techniques could not be employed because of mineralogy and stratigraphy. Consequentially, in conducting the tunnel faulting study 12@ i' 012 innovative and conventional investigative techniques were employed in supplementing the existing state of site and regional knowledge available for analysis and interpretation.
2.2 INVESTIGATIV CHRONOLOGY A comprehensive investigative program evolved as a result of the bedruck deformation exposed during the excavation phase of tunnel construction.
Deterministic fault study objectives previ.asly outlined were realized as a consequence of a series of interrelated geclogic and geophysical research and engineering. Concurrent wit 1, and subsequent to, tunnel excavations, the nature of the fault plane geometry, gouge, and country rock mineralogical as well as chemical constituents were re.ealed. After the necessary site specific data had been assembled, the localized anomalous deformation was interpreted in context of its regional geologic setting.
Tunneling activities began in July 1977 after the main shafts and temporary access shafts had been excavated by conventional " drill and shoot" metheds.
" Drill and shoot" methods were also employed in providing sufficiant room at the base of the temporary access shaf ts for assembling tunnel excavation machines. A D0SCO Roadheader MK-2A tunneling machine excavated 426 li:eal feet of beorock mostly south of the temporary access shafts. The remaining 2600 feet of tunneling was accomplished with a Jarva circular bora tunneling machine. The excavation phase was completeu in November 1978. Tunnel advancement was documented during the geologic mapping program and is shown on Figure 2. Tunnel ar.d shaft components of the cooling water system are shown on Figure 3.
Tunnel heading advancement was initiated from the intake tunnel access shaft.
First, the segment between this access shaf t and service water pumphouse intake riser was completed. Then the connecting tunnei to the emergency service water pumphouse was excavated. Both tunnel segments were excavated with the Roadheader MK-2A machine, which was dismantled ani removed upon their completion, and reassembled in the discharge tunnel temporary access shaft.
Subsequently, tunnel segments between the discharge tunnel access shaft to the
+
12 b Di3
discharge riser in the discharge tunnel entrance structure and a connecting tunnel from this segment to the emergency service water pump house were excavated.
Bedrock conditions in these segments were quite good with minimal crown overbreak.
Predictably minor, discontinuous and closed vertical to near vertical joints, minimal groundwater seepage, and short-term gas emissions (predominantly methane) were experienced in these tunnel segments. None of these conditions were beyond an anticipated range.
Advancement of the intake tunnel heading to the north from the temporary access shaft began in September 1977 with the Roadheader. In April 1978 the runneling advancement rate greatly accelerated with the employment of the Ja rva . As a routine procedure for these tunneling operations, horizontal exyloration boreholes were drilled in advance of the heading. Probe boriras ,
whis h intersected the first tuanel segment containing bt : rock deformation yielded gaseous emissions and groundwater. In addition, the variability of probe hole drilling resistance suggested an atypical condition. During the week of April 17, 1978, the Jarva intersected the tunnel fault segment which could not be observed until April 25, 1978, subsequent to sufficient advancement of the Jarve.
A fault plane, oriented normal to the intake tunnel bearing and dipping approximately 16 degrees to the southeast, was identified by the site resident geclogists and confirmed by the Project Geologist and an internal project consultant. The apparent displacement, with a thrust sense of motion, was estimated to be less than two feet and the throw less than one foot. The fault plane width was estimated between 0.5 and 18 inches, although the latter was presumed more indicative of gently flexed and/or abruptly kinked or otherwise simply fractured rock. A gray-clay gouge of tough leathery consistency containing small angular shale f agments was observed vithin the fault zone. Inis descriptive information was communicated to the NRC.
Samples of gouge were collected and X-ray mineralogical identification conducted on the two micron and smaller size fraction. Results demonstrated a mineralogical assemblage typical of Chagrin shale as reported in the PSAR. On the basis af L
i2J a 0i4
the mineralogical data, proximity and physical resemblance between th.- intake tunnel fault and onshore deformation, and lack of ccatradictory evidence, a common origin for deformation exposed in the open-onshore and tunnel excavations was suggested. However, the intake tunnel fault exposure occurs more than 100 feet deeper than the deepest known onshore deformation. For this reason other deformational mechanisms, notably stress relief and tectonics, remained plausible ot % ins.
Advancement of the discharge tunnel heading in a northerly direction from the temporary access shaft began in January 1978 with the Roadheader machine, which was withdrawn in February after sufficienc room was provided for assembling the Jarv.. The Jarva began excavating in August 1978. In late August a tunnel segment was exposed and observed to have been only mildly deformed by ccmpression. A second discharge tunnel segment, more deformed than the first, wa:. ,ncountered. The two discharge tunnel segments containing deformation are separated by apprcxims.tely 200 feet.
The zones of warping with very small displacement ar.d thrust faulting, respectively, identified in the ?ischarge tunnel were mapped and reported to the NRC. The discharge tunnel fatlt lies on strike projection with the intake tunnel fault, suggesting that they are the same structure. The zone of minor compressional features preceding the main discharge tunnel deformation was presumed to be either en echelon or a splay of the .:ortheasterly striking thrur,t fault.
Concurrent with exposure of the discharge tunnel deformation, a lake bottom survey and shoreline reconnaissance were conducted. Neither revealed evidence of surface faulting.
More investigative work followed, including a series of exploratory borings, onshore and offshore geophysics, conventional and isotopic analysis of groundwater seepage discharged from the fault, and SEM (scanning electron microscope) analysis of fault gouge. Results of these studies demonstrated that the fault plane maintained a low-angle inclination beneath the intake tunnel more than 600 feet to the southeast. it is uncertain if a deep onshore boring, located at the crest of the shoreline bluff and offset 100 feet from the intake tunnel, intersected the fault. This boring was drilled sufficiently deep so that it 12f)015 should have intersected the fault unless the fault p.ane attitude changed or the fault zone thins and becomes conformable with bedding.
Borings located approximately one mile west of the plant area along the shoreline and on projection with the fault trace at the bedrock surface did not intersect faulting. Neither the onshore nor of fshore eagnetic survtys revealed evidence of faulting. Saline discharges from the tunnel fault we.e determined by isotopic analysis to be meteoric but not Lake Erie water. Preliminary SEM analysis of fault zone gouge showed that new mineral growth bridges had formed across microcracks interpreted to have for.ned syngenetic with faulting or at least the last fault movement.
A geophysical signature of the fault was provided by compressional wave low-velocity zones. Undeformed bedrork exhibited a relatively high velocity.
No low velocity zones were identified in the onshore borings, geophysically logged. It also appeared that a low level of gamma radiation was correlative with fault zone gouge. Longitudinal velocity measurements in the tunnels across the fault indicated that the bedrock was sound in spite of the discontinuity. The velocity values are within the range of those reported in the PSAR for preconstruction site exploration.
Three geologists independently reviewed the tunnel faulting prior to construction of the concrete liner. They determined that the deformation was brittle rather than soft sediment and was not penecoc:temporaneous with depositioa.
Other origins, including direct and indirect g_acial action and tectonic, were considered.
A very deep onshore boring was drilled slightly east of the discharge tunnel.
In-situ stress measurements employing the hydrofracture technique were performed within the borehole. Subsequent laboratory testing of core from the boring supplemented the in-situ test data. The field and laboratory testing programs were directed by Dr. Jean-Claude Roegiers (Department of Civil Engineering, University of Toronto). Dr. Poegiers also evaluated the in-situ stress and laloratory data.
ISB( ')0i6 Other aspects of the investigative program included discussions with local resident salt mine engineers, and geologists with knowledge of regional surface and subsurface geology; laboratory determination of gouge physical properties; continuing literature review; and various geological, geophysical, and engineering analyses. Very detailed mapping combined with photographic, video tape and sound track reproduction of the tunnel bedrock deformation serve as permanent documentation.
2.3 Gl:0 LOGIC SETTING The Perry Nuclear Power Plant site is situated on the northwestern ? lank of the Appalachian geosyncline in the Central Lowlands Physiographic /rovince adjacent to Lake Erie. Bedrock directly beneath the site is the Chagin shale member of the Ohxo Shale formation (Upper Devonian). Regionally, these rocks dip gently to the . atheast at a gradient of approximately 20 to 40 feet per mile. The Precambrian crystalline basement occurs at a depth slight / greater than 5000 feet. To the south the Devonian strata are overlain by successively younger Paleozoic sediments (see Figure 4).
Lake Erie, which lies several hundred yards north of the plant area, has a ma.:imum depth of approximately 210 feet and an average depth of 58 feet. The western end of the Lake is extremely shallow and is immediately underlain by resistant carbonate bedrock. From the general vicinity of Sandusky, Ohio, to the east beyond the Pennsylvania boundary, Lake Erie has been eroded into Upper Devonian shales which overlie the relatively more resistant rocks comprising the lake bottom strata of the western portion.
In northeastern Ohio glacial drift and glaciolacustrine sediments overlying bedrock reach a maximum thickness of 250 feet. The site is located on the Lake Plains Section, a physiographic subdivision of the Central Lowlands province formerly submerged during higher Lake Erie levels. Here, bedrock overburden deposits ranging in thickness from 55 to 60 feet consist of dense till and lacustrine sediments, respectively. A steep bluff contiguous to the shoreline exposes 40 to 45 feet of overburden stratigraphy (see Figure 5 for glacial deposits). ,
1 Oh '
(- 0' 9 \L.
Secondary structures demonstrative of bedrock deformation in the site vicinity, as well as throughout northeastern Ohio, are rare. This is attributable to the nearly ubiquitous veneer of glacia. deposits obscuring bedrock, the minimal effect of the Alleghenian (Appalachian) Orogany on Paleozoic strata in this region, and the attenuation of Alleghe lan orogenic stresses during their northwestward propagation beyond the Appalachian Structural Front.
Most of the subsurface structural interpretations for these regions are founded on deep well data. It is reported by Stone and Webster, based on personal communication with A. Janssens, formerly employed by the Ohio Geological Sursey, that the sedimentary sequence above the Middle Devonian Delaware Formation is affected by folding. Structural contours of the Delaware and Dayton Formations prepared by Stone e.nd Webster show persistent small structures, probably folds, especially in Portage County, Ohio.(1) Structural contour maps of the " Big Lime" top (Delaware) and a definable geophysical base (Packer Shell) approximately 50 feet below the stratigraphic base of the " Big Lime,"
and an isopachous map of the interveaing interval were prepared to determine subsurface structure in Lake, and adjacent Counties (see Figures 6, 7, and 8).
" Big Lime" is a shortened drillers' expression for the thick Silurian-Devonian carbonate and evaporate sequence knoun as " Big Niagaran Lime." The only anomalous structures revealed are located in central Ashtabula County. Appa rer.t thickening of the " Big Lime" in this region, due to faulting or folding, may be attributable to Appalachian orogenic stresses. No shallow deformation in that locale is known.
Salt mining has exposed deformation within the Salina beds. Heimlick describes minor folds, amplitude of six inches, and wave length less than twelve inches, locally overturned, in the production interval of the Internatienal Salt Co.
mine in Cleveland.( } Structural contours of the sa' produc. a interval for the Morton Salt Division of Morton Norwick mine in the Painesville area reveal northeasterly trending synclinal troughs interpreted by Jacoby to be salt flowage preceding faulting in response to Appalachian tectonism.(3) However, large scale folding in Lake County of either the salt or overlying shale strata is not present in surface or subsurface exposures, nor interpreted f rom sub .rface geological or geophysical data.
-[ f) } h
Faulting is nearly as anomalous as fold structures but does affect Paleozoic strata to the south and has been exposed in the International Salt Company mine in Cleveland to the west. More locally, Jacoby reports that a high angle thrust fault intersects the salt production interval of the Morton Salt Division of Morton Norwick mine in Fairport Harbor, approximately eight miles southwest of the Perry site.0) He does not believe that this fault is pervasive vertically through the Oriskany Sandstone of Middle Devonian age.
Rock cores from salt strata exploratory borings in the Painesville area occasionally intersect displacements within the " Big Lime" of a very minor nature, at.ounting to a few inches at most, which are comple; *1y healed.
Donald R. Richner, consulting geologist, has examined these discontinuities, which range from very minor to miniscule, consisting mainly of stylolites and minor slips with traces of slickensides but having observable displacements of two inches at most. He has not seen any evidence that these discontinuities were of a tectonic origin. Those observed above and below the Salina sap beds appear to result from penecontemporaneous deformation. }
Geologists are in agreement that the faulting and folding exposed in the International Salt Co. and Morton Salt Division of Morton Notwick mines in Cleveland and Painesville, respectively, are attributable to dissolutioning of the salt during sediment lithification. '"}
Subsequent failure of the overlying strata resulted in graben structures, slumping, and down-warping dependent voon overlying lithology. Locally, salt flowage into fractures and irregularly shaped cavities is evident.
The only well-documented fault near the site locale is a relatively minor localized overthrust with approximately one foot of displacement in the Bedford shale (Mississippian age), known as the Gabor Fault (Prosser).(5) The minor thrust fault described by Prosser was observed in the field on the east bank of Bates Creek, also known as Warners Creek. The strike of this fault is northeast. Three faults, not reported in the literature, were found on the we-t bank of the Paine Creek about one mile north of the Bates (Warners) Creek 1N lL~ ~ O}9
~
fault. These faults, two gravity faults, and a small bedding thrust fault, named Hell H'.iow 1, 2 and 3, were found to be associated with slumping.
These site locale faults are shown on Figure 9.
Field investigations and literature studies were completed to determine the characteristics, origin and age of both the Bates (Warners) Crcek and Hell Hollow faults. Thore faults were determined to be of surficial nature, limited in extent and unrelated to deep-seated faulting. Their origin is conclud so have been glacially induced at Pates Creek and related f.o '-i~ -k slumping at Hell Hollow.
Geologic mapping, inspection, and ev.luation of bedrock foundations, including excavation cuts and foundation grade, for the plant ares ~tructures s were initiated in August 1975. Several localized areas ot deformed bedrock were revealed as a conseg 2nce of the excavation. The deformation consisted of folds and faults within the Chagrin shale. Vertically, the lower limit of the onshore deformation was established at a horizon defined by the deepest foundatica excavations, specifically those for the condensata demineralizer and heater bay buildings. The upper limit of this deformation terminates at the base of a boulder layer, which maintains grade at approximate elevation 570 feet and is pervasive throughout the plant site. The boulder layer defines the base of structurelest lower till. Below the boulder layer and above competent shale, a six- to eight-foot thick transitory interval was mapped in which the lower till has been incorporated within contorted, blocky, and weathered shale. The relationship of the enshore deformation to tunnel faulting is shown on Figure 10.
\ L 3.0 METHODS OF INVESTIGATION The purpose of the studies was three-fold: first, to determine lateral extent of the fault observed in the intake and discharge tunnels at the Perry site; second, to analyze the type and degree of fracturing within and adjacent to the fault; third, to examine in detail the seismicity of the area surrounding the Perry plant; and fourth, to investigate the origin of deformation.
The following techniques were used in determining the extent of the fault on land and on the bottom of Lake Erie:
O Literature review and personal communications with geologists cognizant of area geolog3 (surface and subsurface) o Exploratory borings o Shoreline reconnaissance o A video survey of the bottom of Lake Erie in the vicinity of the updip projection o Detailed geologic tunnel mapping of deformation o Microcrack analyses of fault zone samples o Analysis of water from the fault and from Lake Erie o Evaluation of published and unpublished geophysical data o Magnetic (total field) profiling (both onshore and offshore) 124l) 021 o Borehole (in hole) logging of (compressional) wave velocity o In-situ seismic velocity measurements o An evaluation of the seismicity in the area surrounding the Perry plant was made on a very detailed search on period newspaperc and other document o
In-situ borehole stress measurements to determine stress field orientation, magnitude and gradient (vertical)
The detailed mapping, lake bottom survey, and geophysical and seismological studies were performed by the Weston Geophysical Corporation 3.1 GEOLOGIC 3.1.1 LITERATURE REVIEW AND PERSONAL COMMUNICATIONS Published and unpublished sources were reviewed in order to learn more about the surface and subsurface bedrock structure in northeastern Ohio. These activities were supplemented by personal communications with resident engineers at two salt mines (Mr. Jaroslav Vaveeka, Cleveland mine, International Salt Company and Mr. B. C. Cummings, Painesville mine, Morton Salt Division of Morton Norwick) and Mr. Robert G. Van Horn, Head Regional Geology Section, Division of Geological Survey, Ohio Department of Natural Resources. Two consultant geologists, Mr. Donald R. Richner and Mr. Charles R. Jacoby, with considerable experience of subsurface geology from exploratory drilling and mining operations in northeastern Ohio, were also contacted. Finally, three independent reviews of the cooling water tunnel faulting were performed by the following recognized for their respective specializations:
s 12F)022 Dr. Robert LaFleur Pleistocene Geology and Sedimentology Rensselaer Polytechnic Institute Mr. James Murphy Areal Geology and Stra*igraphy of Northeaster Ohio Ohio Historical Societ, Dr. Barry Voight Structural Geology Penn State University Mr. Murphy had been contacted previously to provide independent opinions during Applicant evaluations of bedrock faulting in the site locale and onshore plant area bedrock deformation exposed by excavation. He had also arranged for radiocarbon dating of comminuted plant material obtained from the site lacustrine deposits. Results of this dating (14,480 1 310 B.P.) established an age somewhat older than previously assumed for the retreat of Hiram ice and a minimum age for the onshore plant excavation deformation.
Data and evaluations of an offshore shoreline parallel survey, especially in the vicinity of the site, conducted by the Coastal Engineering Research Center, were forwarded by Mr. S. Jeffress Williams, marine geologist, Geotechnical Engineering branch. The survey consisted of high resolution seismic reflection profiling suitable for evaluating abrupt elevation changes in the lake floor or acoustic contrasts of sediments, both potential indicators of faulting. (6) 3.1.2 EXPLORATORY B0 RINGS The TX-test boring series was conceived for the purpose of tracing the downdip extent of faulting intersecting the cooling water tunnels at the site (see Figure 11). Drilling operations occurred within the intake tunnel excavation, on the shoreline bluff, and along the beach west of the site. TX-1 through 6 were in-tunnel, the first test holes of the series. Limiting conditions dictated the use of a powered drill, mounted on a steel "A" frame, stabilized by two expanding rods braced against the tunnel crown. This system implemented kj an NX-size, diamond bit, single-tube core barrel. Both air and water were supplied by existing utility lines used in tunnel construction. An additional air line was used to dissipate natural gas inflows encountered in drilling.
Drilling operations on the shoreline bluff (TX-7, 11, and 12) were conducted from truck mounted drill rigs. TX-8, 9, and 10 were drilled on the beach from an ATV (all terrain vehicle) mountea rig.
In-tunnel borings were laid out at progressively greater distances south of where the fault plane intersects the intake tunnel invert. TX-1 was located five feet south of the fault-invert contact, assuring advcnce through the fault plane would occur at a very shallow depth. In order to establish characteristic indicators of test hole advance through faulted rock, close ettention was paid to all aspects of sampling in the initial hole. After each core run a gas detection meter was used to measure concentratiens of me'.hane emitted from the drill water. This device was also used for safety purposes.
Cognizance of indicators from TX-l's advance aided in interpreting intervals in subsequent holes where faulted rock was projected to greater depths.
Test borings drilled from the shoreline bluff sought to encounter the fault at greater depths than those of the in-tunnel boring group (several hundred feet rather than 2 to 90 feet below tunnel invert elevation).
The group of test borings located on the be .h was designed to encounter a shallow southwesterly projection of the fault based on limited strike measurements attained from exposures in both tunnels and previous TX-series borings.
Several types of in-hole testing were performed in TX-series holes. The Weston Geophysical ' Corporation conducted gamma and sonic velocity logging to confirm fault zone idcatification. In addition tc, this, a long steel " feeler" probe (length, 10 feet) was implemented in shallow holes TX-1, 2, and 3. A hydrofracture in-situ stress measurement study was performed in the deepest boring of the TX series (730 feet). This efiort was planned and directed by Dr. Jean-Claude Roegiers (Project Consultant). Instrumentation was supplied by Serv-Ko r, Inc. and pressurized fluid capab'ility by Halliburton Services.
All rock core samples of the TX-series were logged in detail and photographed.
All pertinent and representative core was wrapped in clear plastic.
3.1.3 SHORELINE RECONNAISSANCE Continuous shoreline reconnaissance southwest and northeast of the site was performed with the objective of identifying evidance of offset or structural disturbance in the lacustrine and till deposits exposed by the shoreline parallel bluff contiguous to Lake Erie. Reconnaissance was conducted a considerable distance beyond the land surface projection of the intake and discharge tunnel faulting (see Figure 12).
3.1.4 VIDEO EXAMINATION OF LAKE BOTTOM FEATURES An underwater camera survey of the lake floor was conducted to permit close examination of the lake bottom by a diver, and provide visual aid and documentation for other technical personnel. The intent was to examine the i bedrock surface for the presence of structural features. The floor of Lake
'cie, offshore of the Perry Nuclear Power Plant, essentially consists of a bedrock surface with a very thin covering of silt. Locally, the bedrock surface is covered by concentrations of boulders and cobbles.
The video survey consisted of two parallel east-west traverse lines, labeled Lines A and B, approximately 800 feet ia length and 200 feet apart, previos -ly located and horizontally surveyed offshore of the Perry Nuclear Power Plar.t.
The lines were selected to cover the vicinity of the updip projection of the fault noted in the intake tunnel, and to cross the projected continuation of the fault to the east.
Each traverse line concisted of five relatively evenly-spaced stations. The video coverage was circular in fashion around each station to a maximum radius of approximately 75 feet. Figure 13 shows the location of the traverse lines and stations, as well as the area of, coverage around each station.
12@ 025 The diver, equipped with aa unde rwater compass, described and noted the orientation of bottom features as he moved relative to the lake floor. A two-way concunication system with surface monitor permitted the surface operator and other technical personnel to discuss the bottom conditions with the diver at the time of the survey, and to request detailed examination of specific features of interest. In all instances, the original videotapes have been retained in their entirety.
3.1.5 DETAILED GEOLOGIC MAPPING IN THE INTAKE AND DISCHARGE TUNNELS Four hundred lineal feet of tunnel wall rock exposure were mapped to study and document the nature of bedrock deformation encountered in the intake and discharge tunnels at the Perry Nuclear Power Plant. The field mapping was carried out in the period from February 15 through 27, 1979. One structure was mapped in the intake tunnel at Stations 10+25 to 10+95. Two bedcock structures were mapped in the discharge tunnel at Stations 11+40 to 12+00 and 13+00 to 13+70. Both walls were mapped in each area. Rock bolts, straps, and wire mesh on the crown, and muck and rails on the invert prevented mapping of these surfaces. Approximately 7 vertical feet of wall were mapped on each tunnel wall. Figures 14 and 15 show the location of intervals mapped in the intake and discharge tunnels, respectively.
Mapping was carried out subsequent to placement of stations every 5 feet, as well as three constant elevation lines at 2-foot intervals, along the entire mapped tunnel wall area. S2rvey control for the stations and elevations lines allowed all mappable features to be located by a standard 6-foot rule and transferred to cross section paper at a scale of 1 foot to 1 inch. The minimum resolution of the beds mapped was 0.5 inches or approximately 1.0 centimeter.
Photomosaics of the entire mapped areas were composed from professionally taken photographs. Closeup photomosaics of the fault zones in both tunnels provide detailed documentation of'these structures. The mapped areas of both tunnels were videotaped; approximately 3 1/2 hours of videotape were acquired.
In all instances, the ariginal videotapes have been retained in their entirety.
124 I 026 3.1.6 MICR0 CRACK ANALYSIS A microcrack analysis was performed on samples of gouge obtained from the faults in the intake and discharge tunnels at the Perry site. These investigations were performed by Dr. Gene Simmons whose complete report is included as Appendix A to this report. The following is a summary of the metho's of ir.vestigation employed by Dr. S % ons.
Microcrack samples were acquired by Dr. Simmons and Weston personnel from the fault zone in both the intake and discharge tunnels.
The samples were acquired in such a way as to minimize production of microcracks during sampling. Upon acquisition, the samples were carefully packed and transported to Dr. Simmons' laboratory foc analysis.
Laboratory analysis of the samples involved examination of individual microcracks with a scanning electron microscope (SEM) to determine crack length and extent of filling. An electron dispersive X-ray system (EDX) was used to determine the elemental composition of the material filling the observed cracks.
The details of the sampling procedure and laboratory analysis for the microcrack studies are discussed by Dr. Simmons in his report, Appendix A to this study.
3.1.7 WATER ANALYSIS Chemical analyses of intake and discharge tunnel faulting seepage samples were pe r fo rmed. Ionic concentrations were obtained for chloride, sulfate, and sodium. In addition, ,alinity and pH measurements were conducted on each sample. Comparative evaluation of data provided information on trends.
6 The isotopic ratios of D/H and 0/ 0 were measured with a mass spectrograph for three water samples from the fault in the intake tunnel, one sample from the fault in the discharge tunnel, and two samples from Lake Erie. A sulfur 12% 027 isotope analysis was attempted on samples of water from the intake an6 Jischarge tunnels and from Lake Erie. The sulfur analysis did not succeed bec4w,e of the lack of sufficient sulfur for analysis in any of the tunnel samples.
Appendix B to thi' report prepared by Dr. Gene Simmocs presents the basis of the technique for the hydrogen and oxygen isotype analysis.
3.2 GEOPHYSICAL STUDIES 3.2.1 EVALUATION OF PUBLISHED AND UNPUBLISHED DATA Published and unpublished geophysical data for the immediate vicinity of the Perry nuclear site were examined for any anomalies which could be related to the fault noted in the intake and discharge tunnels. The examination consisted of a review of published and unpublished geological and geophysical data, as well as federal and state government reports and data files.
3.2.2 MAGNETIC SURVEYS 3.2.2.1 Offshore Total field magnetometer data were obtained over the projected strike of the fault to determine whether or not an associated magnetic signature was present.
A shipborne magnetic survey of Lake Erie consisted of 17 lines, perpendicular to the projected strike, at 200-foot intervals (Figure 16). Coverage was essectially continuous along each profile, and the data were displayed by means of a strip charge recorder at a vertical scale of 200 gammas / inch.
3.2.2.2 Onshore For onshore coverage, four separate lines were traversed with readings taken every . feet (Figure 17). Lines were operated along existing roads, which resultec in profiles oriented at 45 to the projected strike.
k !-
All data were obtained with a proton precession magnetometer. For a further discussion of the magnetic survey metbod refer to Appendix C, Section 2.0.
3.2.3 BOREHOLE LOGS Gamma radiation and velocity logs were obtained in drill holes located in the intake tunnel and on the shore. Figure 11 is a borehole location map. The objective of these measurements was to locate geologic units to be used as markers in determining the offset of the fault and/or to provide means of locating the fault itself.
The velocity logger measures the difference in the travel time for seismic energy, moving up a drill hole from a common source to reach two geophones separated by a known distance. It provides a rapid, accurate measure of the in-situ seismic velocites ("P" and "S" wave values) in the material between the two geophones. For a further discussion of seismic velocity logging refer to Appendix C, Sectica 4.0.
Measurements were made at 1/2-foot intervals adjacent to the projection of the fault and st 1-foot intervals for a distance of 10 to 20 feet away from the projection. Selected boreholes (TX-6, TX-4, and TX-7) were logged at a 1-foot interval for their entire length.
The gamma logging was accomplished with a probe which measures the gamma radiation incident on an enclosed scintillation sensor as it moves up the hole. In sedimentary rock sequences, the instrument responds primarily to shale content, because radioactive elements tend to concentrate in shales and clays. At the site, the logs were obtained using two rates of ascent up the hole (20 ft/ min and 3 ft/ min). The slower rate provides a smaller sampling interval and, thus, greater resolution. For further discussion of gamma radiation logging refer to Appendix C, Section 3.0.
12$029 3.2.4 IN-SITU VELOCITY MEANSUREMENTS A seismic in-situ velocity survey was conducted to examine the condition of the tunnel wall in both the intake and discharge tunnels in the vicinity of the fault.
Seismic velocity values are diagnostic of rock conditions in tunnels and provide a comparison between rock in and adjacent to the fault and rock located some distance from the fault. For further discussion of in-situ velocity measurements refer to Appendix C, Section 4.0.
In the intake tunnel, a 6 geophone spread (each geophone has 3 components) and a 12-geophone spread (each geophone has one horizontal and one vertical component) were used. Geophones were separated by 10-foot intervals, and each spread was centered on the observed fault (Figure Id). In the discharge tunnel, data were obtained across the fault and across a fracture zone located 100 feet south of the fault (Figure 19). Two spreads were used across the fault, both with 12 (two-component) geophones. The first had 10-foot spacings with the fault located 10 feet south of the center of the spread; the second had 5-foot spacings for higher resolution and was centered on the fault. The velocity measurements across the fracture zone were obtained with a 12 geophone spread with 5-foot spacings centered on the fracture zone.
Three-component geophones, which detect vibration energy along vertical, radial, and transverse alignments, were placed on pedestals drilled into both the intake and discharge tunnel walls. Seismic energy was generated by a hammer blow against the tunnel wall adjacent to a geophone.
3.3 EVALUATION OF LOCAL SEISMICITY AROUND THE PERRY NUCLEAR POWER PLANT SITE A detailed examination of the local seismicity around the Perry site was performed with the purpose of evaluating the validity of each epicentral location and intensity for all reported events.
12e-
> o30 v
A parallel compilation cf all cataloged entries was made, and subsequently a local newspaper search was initiated to collect additional supporting evidence for each event. The details on the sources of the data base and the texts of all new material acquired are presented in Appendix H. A separate summary evaluation was then prepared for each historical event within the 50-mile radius of the site, taking into account cataloged entries as well as the entire file of supporting evidence.
3.4 IN-SITU STRESS MEASUREMENTS Hydraulic fracturing was performed in test boring TX-11 in order to determine the magnitude and orientation of the in-situ principal stresses. Eight intervals were fractured between a depth of 394 and 718 feet. TX-11 boring rock cores were subsequently tested in the laboratory in order to provide confirmatory tensile stress data of the field data. The in-situ borehole stress program was planned and directed by Dr. J. C. Rocgiers, Department of Civil Engineering, University of Toronto. The in-situ stress program methods results and coaclusions are appended to this report (see Appendix E).
\
4.0 RESULTS 4.1 GEOLOGIC 4.1.1 LITERATURE REVIEW AND PERSONAL COMMUNICATIONS As reviewed in Section 2.3, Geologic Setting, bedrock throughout northeastern Ohio is not known to have been significantly affected by late Paleozoic orogenic stresses or any other tectonic disturbance. F::ulting !n the sequence of evaporite and carbonate rocks comprising the Salina Group has been exposed in salt mines within the region. These faults are attributed to dissolution of salt beds followed by failure of the overlying carbonate beds. Structures of this type are generally believed to have been developed shortly af ter sediment lithification. Alternatively, late Paleozoic orogenic stresses may have been sufficiently high to have caused salt flowage which induced brittle deformation of the interbedded, more competent carbonate beds.
Salina Group strata begin at a depth of approximately 1750 feet beneath the site. Correlations between site borings and regional exploratory drill holes do not suggest the existence of any pervasive fault or fault system. Neither top and bottom structure contour nor isopachous maps of the " Big Lime" support the concept of a regional fault or fault system (see Figures 6 to 8). In the context of regional geology there is no basis for lateral extropolation or deepening of the tunnel faulting. Shallow bedrock deformation is exposed in outcrop seven to eight miles south of the site and had been exposed in plant area exravations. However, these structures terminate with depth on undeformed strata. The outcrop exposure deformation was the result of glacial shove and loading (Bates Creek) and slump (Hell Hollow). Plant area bedrock deformation was caused by late Wisconsinan glacial shove and loading. A minimum age of 14,4801310 B.P. (Hiram ice) for the plant area deformation is inferred from a radiocarbon date of the organic debnis interbedded within lacustrine sediments.
1 24!!, 032 Independent opinions provided by the three reviewing geologists are in agreement that the tunnel faulting is not pececontemporaneous but is most likely caused by localized stresses created during Pleistocene time by either the advance of the ice sheet (s) and concomitant depression of the crust, or in reaction to removal of weight of the overlying ice (glacial rebound). In addition, Dr. Robert LaFleur was requested to review the data, interpretations , and conclusions of earlier investigations regarding the stated origin of the Bates Creek, Hell Hollow, and plant area deformation. He concurs that the Bates Creek and plant area deformation are the result of glacial shove and loading (active glaciotectonics) and the Hell Hollow vertical faults were the result of post glacial slumping. These opinions had been stated in the earlier investigations by Mr. James Murphy. Dr. LaFleur does not believe the tunnel faulting is demonstrative cf either active or passive glaciotectonics. In his opinion the deeper tunnel faulting is a response to the state of stress imposed by glaciation during advance or subsequent to recession (glacial rebound).
Dr. Barry Voight considered other modes of Paleozoic deformation in addition to late Paleozoic tectonism including Mesozoic-Tertiary tectonics and miscellaneous Pleistocene - Recent faulting mechanisms. Opinions of the three reviewers are attached to this report in Appendix D.
4.1.2 EXPLORATORY B0 RINGS Fault zone endicators, revealed in the ad.ance of :X-1, 2 and 3, were s follows: (1) increased vibration in irill rods; (2) a creamy grey influx to a normally light grey drill wash; (3) platy clay particles in drill wash; (4) a release of gas when the core barrel was pulled after the run; (5) a 0 to 80 percent recovery in the cored fault zone (recovery in undisturbed rock was consistently very high); (6) highly broken, rotated rock frags speckled with remnant grey clay for those portions of the fault zone that were recovered; and (7) a change in the dip to normally flat lying laminae, above, and below the fault zone.
} 2h )
All indicators did not occur in each boring where a fault zone was suggested.
In fact, only the loss of recovery and the character of rock that was recovered from suspect fault zones remained consistent throughout those borings. Using these indicators, a fault zone was detected in all of the in-tunnel borings, TX-1 through 6. The boring locations and depth where faulted rock was identified revealed a constant fault plane dip of 17 degrees SE. Low gamma radiation levels and low P-wave (compressional) velocity values coincided wit h zones of disturbed rock at elevations where a fault zone was logged fr'm drilling program indicators in TX-2, 3, 4, and 6. TX-1 was too shallow to log geophysically and TX-5 caved at the fault preventing geophysical logging of the suspect zone.
The constant 17 degree fault plane dip derived from TX-1 through 6 aided in the location of TX-7 through 12. TX-7 was initially advanced to a depth of 395 feet from the shoreline bluff. Increased rod vibration, loss of core recovery, remnant clay on broken, rotated shale fragments, and a stuck core barrel (eventually retrieved and coated on the bottom three feet with a thin grey clay) indicated a disturbed zone from 371.3 to 372.4 in TX-7. Geophysical logging, however, did not confirm this zone. It is suspected that a lack of proper drill water circulation may have caused increased friction at the core barrel, falsely suggesting a zone of disturbance.
TX-8, 9, and 10 were drilled along the beach west of the site. Both TX-8 and 10 encountered zones of broken rock with what appeared to be minimal clay remnants from depths of 65.85 to 66.7 feet and 63.5 to 64.9 feet, respectively.
Minimal loss of recovery was mea:ured in both zones. Geophysical logging, however, did not recognize disturbed rock in either 'lX-8, 9, or 10.
TX-11 was drilled approximately 1060 feet southeast along dip direction of the intake tunnel fault exposure. This boring was the deepest of the TX series, drilled to a depth cf 730 feet down from the shoreline bluff. No naturally disturbed rock was encountered in the entire borehole length. Three runs of core were disturbed by uncontrollable core barrel handling because of gas inflows. Hydrofract stress measurements were performed in this hole in test intervals between 394 and 718 feet.
12T 034 Angle hole TX-12 was drilled from approximately the same location as TX-7.
The drill rig employed a wire-line, double barrel system. The double barrel was actually able to core the fault zone materials with very little loss in recovery from 376.4 to 380.0 feet. This boring confirmed the continuation of a 17 degree SE fault plane dip, approximately 230 feet horizontally southeast of the last confirmed fault zone occurence in tunnel-boring TX-4. Cored fault zone materials included angular shale fragments within several grey, clayey, gouge seams, broken fractured rock, and rock laminae with multiple dips.
TX-12 was completed at an angle depth of 480.0 feet.
After the completion of TX-12, TX-7 was reamed and extended 100 feet c;ing the double barrel wire line system. A disturbed zone was encountered fran 412.8 to 413.9 feet, vertical depth. Unlike TX-12 a gouge zone was not recovered.
Increased drill rod vibration, a 100 psi increase in drill water pressure, 50 percent loss in recovery, and broken rotated rock, speckled with grey clay remnants suggested a zone of disturbance. If this zone represents the fault, its location marks an increase in fault plane dip between TX-12 and the extended TX-7.
Geologic logs of TX-series borings are attached as Appendix F.
4.1.3 SHGll.INE RECONNAISSANCE Traverses northeast and southwest of the plant area along the shoreline and headward in stream cuts emerging at the beach revealed no offsets or structural disturbance of the expo ed lacustrine and till deposits. A boulder layer which occurs at the base of structureless lower till is not offset and maintains a constant elevat'en within the lake facing bluff. An absence of bedrock outcrops and maintenance of boulder layer elevation are demonstrative of ',he lack of surface faulting.
1,
{
4.1.4 VIDEO EXAMINATION OF LAKE BOTTOM FEATURES ,
The video survey of the Lake Erie bottom in the vicinity of the t.odip projection of the fault did not indicate the preseoce of any long cetinuous fractures parallel to the projected fault trace. Those frastures which are noted show no evidence of lateral or vertical offset and seem to close with depth. Figure 20 shows a schematic diagram of the fracturing on the Lake Erie bottom.
4.1.5 DETAILED GEOLOGIC MAPPING IN THE INTAKE AND DISCHARGE TUNNELS 4.- .1 Stratigraphy Chagrin shale at Perry, is on the order of 800 feet thick based on reported thicknesses of 500 feet at Cleveland and 1,200 feet at the Ohio-Pennsylvania border (see Figure 21). Accordingly, the sequence of strata exposed in the tunnels is assigned here to the stratigraphic center of the Chagrin and is considered representative of the unit. This placement ia consistent with the absence, within the tunnels, of marginal lithologic <,equences and fossiliferous strata.
In both tunnels, the strata dip westward to northwestward about 2 degrees (for the detailed mapping sections see Figures 30 and 32). Most of the units are quite persistent in down-tunnel directions, and because the tunnels are random exposures of the internal geometry of the Chagrin, there is reason to assume that the observed units and sequences are qually persistent from east to west, a td that a part of the mapped sequence should appear in both tunnels Inasmuch as attempts to establish a correlation between tunnels were unsuccessful, it is concluded that the strata exposed in the intake tunnel pac.s below the discharge tunnel, and that the described sections are separated by a very short interval of unexplored strata. These relationships and detailed descriptions of the mapped intervals are presented on Figures 22, 23, and 24.
12(')036 The Chagrin strata exposed in the tunnels were subdivided to provide a framework t .hin which fo? ding and faulting in the tunnels could be described and interpreted. Unit boundaries were selected according to their mapability across tunnel wall exposures smeared during excavation and subsequently stain?d and otherwise obscured by minor surficial weathering. There is no genetic significance implied in their selection.
Bedding characteristics and stratigraphic relationships were examined to determine depositional modes and the role of penecontemporaneous deformation in the genesis of the folds and faults. The characteristics considered most significant in their regard are: (1) the attitude of the strata; (2) their thickness; (3) their cor: position and texture; and (4) the detailed nature of their boundaries.
The near-horizontal attitude of the strata and their marked planarity indicate clearly that the immediate substrate during deposition was similarly flat and featureless, a relatively stable distal shelf environment considerably removed from a postulated northerly source of clastic detritus. Minimal sand-size material reached this part of the shelf, and sedimentary structures and bedforms related to sand deposition are nowhere apparent. The- is scant evidence, for example, of bedload transport of detritus, and none whatsoever of either outbuilding or proximal deposition from density currents, any of which would have produced a geometry significantly different from the planar parallel configuration of the tunnel sequence. Virtually all sediment exposed in the tunnel reaching the site area must have been deposited from periodic suspension clouds by processes of vertical accretion.
Bedforms and stratigraphic patterns indicate that sedirnentary cycles begin at the sharply defined upper boundaries of prominent siltstones or siltstone-shale bedsets. These commonly exhibit asymmetrical ripple rearks and, very locally, are truncated to a limited extent; overall, they suggest modification by bottom currents of low velccity and constant direction. These apparently were 120 037 effective in distributing the limited amounts of available silt over fairly wide areas, p obably through 'ople migration; but, for the most part, were not competent ;0 subs *2nti. 43 e ify the den osits or entrain the silt once deposited. The pr aeace of tnin suale laminas iithin many siltstone beds sugge.ts that eveu -tie- wing was at times an ineffective process. During such perio's ,f maximum arrent intensitf, suspended detrital clay and buoyant organic debris must have been carried farther basinward and incorporated in the more distal black shale equivalents of the Chagrin.
Although the siltstones lend :5emselves readily to megascopic and microscopic analyses and are revealing of process-related structures, shalt is everwhere the dominant lithologic-type comprising the lower, thicker part of each cycle.
These are mainly dark gray, clay shales with planar to broadly wavy, very sharply defined laminae, 1 mm to 2 cm thick, of purplish to brownish, clay shale. The laminae, reportedly sideritic in composition, impart a " banded" aspect to the beds; there is no discernible disparity in texture between the shale types to indicate fluctuations in depositional processes. Instead, the
" banding" likely reficcts oscillations in geochemical parameters and possibly detrital clay mineral composition. The shales, therefore, are considered simple beds deposited under uniform sedimentological conditions, and rapid, spasmodic, or uneven deposition of mud are essentially ruled out by their internal structure. Additionally, the general absence of load structures at shale-siltstone boundaries suggests that the mud substrate was quite viscous at times of silt deposition and also that the basin was seismically inactive.
Thickness variations in these strata are restricted to the attenuation and pinch-out of some of the more prominent siltstone beds. These are sedimentary in origin, locally modified by compaction of the section. Their effects on the thickness of the mapping units is negligible.
The indicated depositional setting, dominated by the process 01 slow vertical accretion, winnowing, sublevation, and bypassing, virtually precludes the possibility of rapid sedimentation at or near the site during Chagrin sediment deposition. Localized buildups of clastic sediment and primary slopes steep
- ^
f l 2_ fe!a 036 enough to generate adjustments by slumping are clearly inconsistent with the conditions postulated. Morever, had faulting occurred prior to total consolidation, the adjacent strata, given their clayey composition, would certainly have been thrown into a series of folds and pull-apart structures, lithologic boundaries would have been grooved and polished, and shale thicknesses would have been considerably affected. In particular, those strata between the main fault and the main splay (intake tunnel, Station 10+50 to Station 10+80) would certainly have been markedly distorted. None of these (teria for penecontemporaneous faulting are met in this instance. Instead, the strata are little affected to within very short distances of the fault itself where the bedforms exhibit brittle deformation as subsequently discussed.
4.?.5.2 Tunnel Structural Geology Tunnel excavation for the intake and dis:harge tunnel structures exposed three limited zones of bedrock deformation in the Chagrin shale (Figures 14, 15, and 25). This deformation is characterized by low-angle thrusting, fracturing, and small-scale folding. Deformation in the intake tunnel extends from Station 10+85 to Station 10+55 (Figure 26). Similar deformation occurs in the di-charge tunnel from Station 13+65 to Station 13+25 (Figure 27). In the discharge tunnel from Station 11+50 to Station 11+80 (Figure 28), an interval of disturbed rock is recognized. Figure 29 contains geologic maps of the intake and discharge tunnel deformation.
4.1.5.2.1 Intake Tunnel Structure Bedrock deformation exposed in the intake tunnel extends from Station 10+85 to Station 10+55 (Figure 30). Deformation consists of a low-angle thrust fault which strikes and dips approximately N51E, 18S (Figure 31). Stratigraphic offset is 1.4 feet with the strata to the southeast, upthrown. The throw becomes slightly less (i.e., 0.8 feet) towards the crown of the tunnel.
l n7Q U .) i ji
-31 \
The brittle nature of this deformation is exemplified by the development of fractured and broken drag folds, kinks, and angular / flaggy fragments of siltstone and shale adjacent to and in the prominent gouge zone and dip-slip striations (Figure 30 and 31).
The gouge is light gray, nlastic clay with angular fragments of siltstone and shale derived from the adjacent strata. (See Appendix G for laboratory testing of gouge samples.) Gouge development is greatest where the main fault component 4.s inclined and thinnest where the fault is bedding parallel. Associated with thrusting are numerous thin (J.1 feet) splays of gouge along which the strata have been offset. Offsete are somewhat variable but are on the order of 0.1 feet to 0.3 feet. In cil instances, these stringers / splays are initiated at the main fault zone and die into bedding planes away from the deformation.
Drag folding is both well developed and quite pronouncec. Locally, a faint axial plane cleavage is developed at the fold hinges. Drag folds are asymmetric, northwest verging, and exhibit a distinct bedding plane parting facility.
Thin seams of gouge occur in the hinge area and parallel to this facility.
Orientations of urag fold axes are parallel to the strike of faulting.
Numerous striations are recognized on both the hanging wall and foot wall (Figure 31). Striations indicate the fault movement is dip slip and does not exhibit any strike-slip component. Striations are primarily developed along the bedding parallel sections of the fault but are also recognized in the inclined sections.
To the immediate south of the intake tunnel thrust, an asymmetric syncline is exposed (Figures 26 and 31). Ba:,ed on limited exposure, deformation associated with folding dies out up section and increases down section. The east wall of the intake tunnel exhibits a greater degree of fold deformation than the west wall. This fold is characterized by bedding parallel flexural slip and minor northwest-dipping thrusting on the northwest limb of the fald (Figures 26 and 31). Offset is minimal (0.1 feet to 0 2 feet), with thrusts merging with bedding planes.
1 O [+ b Detailed examination of the intake tunnel fault (Figure 26) indicates that the hanging wall is apparently more deformed than the footwall; deformation is brittle in nature and appears to diminish up section.
4.1.5.2.2 Discharge Tunnel Structure Two zones of bcdrock deformation are exposed in the discharge tunnel (Figures 27 and 28). Both structures are the result of compression. The structure closest to the shoreline is very minor and essenially a kink fold with very minor displacement along the hinge line. The second and furtherest offshore structure is similar to the intake tunnel fault.
The nearshore structure is located approximately at Station 11+70 (Figure 28). Most of the deformation was accommodated by abrupt monoclinal strata bending. The hinge line (plane of deformation) has a strike and dip of N16E, 35SE (Figure 25). Stratigraphic offset dies out below the tunnel crown into a fractured / flexed zone immediately overlain by flat-lying strata. At the invert, the stratigraphic offset (mostly attributable to monoclinal flexure) is approximately 0.4 feet with the southeastern strata upthrowa. Distinctly zig-zag in character, the structure exhibits gouge, localized fracturing, and flexuring of adjacent strata (Figure 28).
The gouge is similar to that developed elsewhere in the tunnels but quite thin (0.1 feet). Apart from the variation in strike and displacement magnitude (Figure 25), the style and the sense of offset are similar to other zones of deformatica exposed in the tunnels.
The main zcne of deformation in the discharge tunnel exter.ds from stion 13+25 to Station 13+60 (Figure 27). Deformation is remarkably similar . .yle and nature to the intake exposure. The discharge thrust strikes and dips N61E, 13SE with the strata to t'ae southeast upthrown approximately 0.8 feet (Figure 29). Associated with faulting are drag folds, fracturing, and well developed gouge (Figure 32). The gouge is light gray, plastic, and contains angular, randomly oriented fragments of siltstone and shale derived from the adjacent strata. Gouge development, as in the intake tunnel, is a function i Ok 2
of the geometry of the fault plane. The thinnest gouge zones occur where the fault is bedding parallel while the thickest zones occur where the fault plane
'~
steepens.
Drag folds are quite prominent, with a northwest-verging sense and fold axes parallel to the strike of the fault. Hinge areas of the drag folds show a slight axial plane cleavage and the development of bedding parallel flexural-slip gouge.
Fracturing is intense in the vicinity of Station 13+40 where the fault plane is essentially bedding parallel. Associated with this fracturing are numerous small gouge-filled eftsets. Stratigraphic analysis indicates that the strata here have been overthickened slightly due to thrusting.
Numerous splays / stringers of gouge trend out from the fault zone and exhibit minor offsets (0.1 feet to 0.4 feet). These splays / stringers, which die into bedding, become more frequent toward the crown of the tunnel and account for the diminished offset along the fault plane.
Striations are recognized on both the hanging and footwalls. Striation orientations indicate a dip-slip motion with no avidence of a strike-slip component (Figure 33).
Based on structural style, orientation, and sense of offset, the two main thrusts exposed in the tunnels are apparently the same structure. Faulting is distinctly brittle with deformation confined to the immediate vicinity of the faulting (Figure 33). The small-scale thrust at Station 11+70 in the discharge tunnel may be an en echelon structure or a splay off the main fault. However, based on limited structural data, the latter is favored.
4.1.6 MICR0 CRACK ANALYSIS Dr. Gene Simmons performed an analysis of microcracks observed within gouge obtained from the fault zone in the intake and discharge tunnels at the site.
Dr. Simmons' complete report is included as Appendix A to this report. The following is a summary of the results of the Simmons' investigations.
Specimens of the gouge and the adjacent country rock were prepared in a form suitable for the examination of microcracks and elemental compositions of individual minerals by the SEM. Two types of cracks were observed. The first type is caused by unaeoidable desiccation of the sample. Desiccation cracks occur subsequent to sampling and are unrelated to tunnel deformation and are recognized as such on the basis of criteria developed previous to the present studies. The second type of crack appears to be related to the last movement on the fault and always contains new mineral growths that extend completely across the crack.
Approximately 350 cracks of the type produced by faulting were examined in six samples. Every crack examined contained approximately one percent new mineral growth.
On the basis of previous observations of several thousand microcracks in a wide variety of rock types, healed microcracks appear to be ubiquitous in rocks. Evidently, the microcracks begin to heal immediately on forming. The degree of healing can be a measure of the amount of time that has been available for the microcrack to heal. The exact mathematical description of th- function that relates degree of filling to elapsed time since the crack was formed is unknown, but is likely S-shaped and asymptotic to the zero and 100 percent values. Two data points have been obtained - one point at one million years (possibly two to five million years) from sandstone at the Satsop site,( the other at 18.5 million years from shocked rock at Ries Crater, Germany. (9)
The rate of healing of microcracks is very likely a function of temperature, pressure, mineralogy, and the composition and flow rate of pore fluids.
Fortunately, the conditions at the Perry site and at the Satsop site are quite similar, and the degree of filling of the cracks >t each site are comparable.
Therefore, the data obtained previously for the Satsop site form a suitable basis on which to estimate the age of the microfractures in the gouge zone at Perry.
{
On the basis of a thorough examination of the microcracks in six representative samples of the gouge and country rock from the fault, or faults, in the intake tunnel and the discharge tunnel and from the fracture zone in the discharge tunnel, it is concluded that the time of last movement of each of these faults is approximstely one million years and may be as old as two to five million years.
4.1.7 WATER ANALYSIS Chemical analyses of tunnel faulting seepages indicate a salinity concentration ranging from 14.4 to 8.4 percent during the period of April 17, 1978 to March 6, 1979. Both the intake and discharge tunnel seepages indicated decreases in salinity, chloride, and sodium concentrations with time. No apparent trend for relatively low sulfate concentrations was established. Measurements of pH were uniform ranging between 7.2 and 8.0. Table 1 contains the results of these analyses.
Salts within Chagrin shale groundwater are not uncomron for northeastern Ohio.
Compositionally, no salts are known within the Chagrin shale member of Ohio Shale formation. Salt bearing strata of the Salina Group occur more than 1650 feet below the tunnel. Even though tunnel faulting is not presumed to extend into the Salina s nt beds, the imperviour, character of the Chagrin shale including the tunnel fault zones would tend to confine the upward flow of salt-saturated groundwater from a great depth. It is more probable that sediment pore water residuum has been diluted by meteoric recharge water in a manner originally suggested by L.U. DeSitter in 1947.(10) This contention is supported by the isotopic ratio results subsequently discussed.
6 The isotopic ratios of D/H and 0/ 0 were measured with a mass spectograph for three samples of water from the fault in the intake tunnel, one sample from the fault in the discharge tunnel, and two samples from Lake Erie. The three samples from the intake tunnel differ insignificantly from each other and from the sample from the discharge tunnel. The two lake samples differ insignificantly from each other. However, the waters from the fault (s) differ 120 044 si;nificantly from the lake water. All three water samples have a meteoric crigin. A sulfur isotope analysis was attempted unsuccessfully on the waters from the fault and Lake Erie. The data obtained indicate a high sulfur content for the lake waters and essentially no sulfur i- the waters from the fault.
The interpretation of the present set of data is that the ' fault water' is not Lake Erie water. Appendix B, prepared by Dr. Simmons, presents the details of the results for the hydrogen and oxygen isotope analysis.
4.2 GEOPHYSICAL STUDIES 4.2.1 EVALUATION OF PUBLISHED AND UNPUBLISHED GEOPHYSICAL DATA A review has been marie of the available published and unpublished geophysical data for the immediate site area of the Perry site. These data include shipborne, high resolution, seismic reflection surveying,(I ' shipborne magnetic data,(12) aeromagnetic surveys,(13,14) and gravity data.(15,16)
The seismic reflection surveys indicate no evidence of eit her abrupt changes in the roieozoic bedrock surface beneath the lake or disruptions of the overlying unconsolidated lake bottom sediments.(15,11,6)
Several profiles which would have crossed the projection of the faults noted in the intake and discharge tunnels did not indicate vertical offset.(6)
A shipborne magnetic survey in the site area, which consisted of three north-south profile lines at 5-mile spacings and one east-west line, shows no evidence for any linear trends parallel to the projected trace of the intake and discharge faults.(11,12)
Similarly, the aeromagnetic surveys which were parallel to the projected trace of the fault and widely spaced (flight line separation on the order of 5 to 10 miles) do not suggest any linear magnetic anomaligs in the near-site area of the Perry plant.(I '
\ "t - Q, O[i h The shipborne gravity data reported by Wall consist of a single traverse in the site area.(15) The relatively widely-spaced shipbocie gravity data are interpreted by Wall as indicative of lithologic variations within the Precambrian basement and not indicative of structure.
Wall's interpretation is similar to Heiskanen and Uotila,(I who interpreted most of the gravity anomalies in Ohio as reflective or lithologic variations in the Precambrian basement.
4.2.2 MAGNETIC SURVEYS The magnetic profiles taken from Lake Erie traverses (Figures 34 to 42) display a generally flat signature. All of the significant peaks appear to be related to cultural influences such as drill barges and metal pipes. There are no anomalies which are associated with the fault.
The land magnetic profiles (Figures 43 to 48) show generally erratic signatures which are attributed primarily to cultural sources. There is no fault-related magnetic signature.
4.2.3 BOREHOLE LOGS Units which could be used as marker beds, as a result of either an anomalous velocity or radiation level, were not detected in the geoJogic section (Figures 49 and 50). This is probably because of the relative macroscopic homogeneity of the Chagrin shale as evidenced by the thinness of the individual beds within the Chagrin. Offset which could be associated with the fault could not be determined.
In borings TX-3, TX-4, and TX-6, velocity logs show low velocity values associated with the fault. No such velocity " lows" are observed outside the tunnel in either the down-dip (TX-7) or along the strike (TX-8, TX-9, TX-10) projection of the fault. Outside the fault zone, the measured velocity is 10,500tfps; within the fault zone, the measured velocity value is approximately 6,000 fps. This lower velocity value at the fault zone is most likely because of the (PVC) casing material.
In the tunnel drill holes (TX-3, TX-4, and TX-6), a low level of gamma radiation can be associated with the fault. However, the signature is not very marked.
It appears that certainly the low P-wave velocity values can, and possibly the low radiation levels may, be used as distinguishing characteristics of the fault.
4.2.4 IN-SITU VELOCITY MEASURL ENTS The velocities determined vary from 10,000 to 11,000 fps for "P" waves (compressional) and 3,900 to 5,200 fps for "S" (shear) waves and are similar to those determined in the previous cross-hole study at the plant area.
Velocities determined from measurements across the fault in the intake tunnel are 10,500 fps for "P" waves and 5,250 fps for "S" waves. These values indicate a relatively intact, sound shale.
Velocities across the fracture zor.e in the discharge tunnel are 10,700 fps for "P" waves and 3,900 fps for "S" waves; again, these values indicate a sound material. The values across the fault in the discharge tunnel are similar; the velocities are 10,000 fps for "P" waves and 4,000 fps for "S" waves.
These velocity values are also within the range of those reported in the initial studies for the plant area.
4.3 EVALUATION OF LOCAL SEISMICITY AROUND THE PERRY NUCLEAR POWER PLANT SITE A detailed study of the local seismicity around the Perry site was made with some significant observations (see Appendix H) . In brief, the local historical seismicity is low: less than 50 events over a period of a century and a half, and no intensity larger than Intensity V Modified Mercalli. In general, t
12lb047 assigned intensities can be considered conservative, and epicentral coordinates relatively uncertain. This uncertainty results, in part, from soil amplification and population distribution which make it difficult in many cases to delineate a clear epicentral area. As a consequence of this epicentral uncertainty, apparent alignments, or clustering of epicenters have no reliable tectonic significance. Details on local seismicity evaluations are presented in Appendix H.
4.4 IN-SITU STRESS MEASUREMENTS Data regarding the orientation and magnitude of the complete stress tensor were obtained for the test intervals between 394 and 718 feet in TX-11. The direction of o was g consistent with stress orientations over a regional basis.
The stress magnitudes (the horizontal stresses are the maximum and intermediate principal stresses) fall within the limits of stresses measured in other parts of northeastern and north central United States and in southern Canada. The vertical component corresponds closely to the anticipated overburden pressure.
At the shallower depths, the tendency for c ~ ~
- is wen Mnd and 3 2 3 extrapolations of existing measurements to the surface are reasonable. No high stress magnitudes were experienced in either the tunnel or plant area excavations or concluded from measurements of extensometers installed in the bedrock walls of the emergency service water pump house. These conclusions regarding stresses in plant structure excavations are consistent with the extrapolation of the deeper in-situ borehole measurements. Below a depth of approximately 600 feet, both a and a show an increase in gradient, with the gradient for a being larger. Data conclusions and an overview of the hydraulic fracturing technique are attached as Appendix D.
}2
5.0 CONCLUSION
S Based on structural style, orientation, and sense of offset, the thrust fault exposed in each tunnel is apparently the same feature or en echelon. Faulting is distinctly brittle with deformation confined to the immediate vicinity of the fault planc. The zigzag fracture pattern and accompanying evidences of flexure characterizing the more southerly discharge tunnel deformation may be an en echelon structure, but more probably represents a splay from tne main fault.
Paleozoic Tectonics, Mesozoic-Tertiary Tectonics, and Pleistocene-Recent faulting mechanisms were considered. Regarding mid-Paleozoic deformation, the concept of soft sediment deformation can be ruled out by the brittle nature of observed deformation. The tunnel fault formed following lithification of the shale sequence. Notwithstanding interpretation regarding age, pre-Pleistocene tectonics are evaluated primarily in consideration of geometric data on tunnel fault strike and shallow dip. Alleghenian (Appalachian) orogenic compressional stresses propagated northwesterly, employing Salina salt bed decollements, would be techr.ically feasible. Upward propagation of faulting at low dip angles, as with the tunnel faulting, would be compatible. Alternatively, southeasterly gravitational movement during late Paleozoic or early Mesozoic time was possible when overburden pressure and formation temperatures were about at peak values. Again, a majority of the lateral movement would be cxpected to occur upon the Salina salts. Relatively high loading conditions existing during glaciation with high stress gradients near ice sheet boundaries may have activated flowage deformation within the salt which resulted in underthrusting of the more competent overlying strata. Other mechanisms associated with deeper rooted deformation such as basement-block faulting and differential warping of Paleozoic strata would tend to produce normal faulting in overlying formations, not thrust faults.
Data regarding the age of faulting were derived from field and laboratory studies. An age determination from fault gouge mineralization could not be undertaken because none of the constituent minerals contained radioactive
\
isotopes suitable for dating. However, on the basis of syn and/or post-deformational mineral growth extending completely across fault zone microcracks related to the last movement on the fault, Dr. Simmons concludes that the time of last movement for each of the tunnel fault segments is approximately one million years but may be as cid as two to five million years or as young as 0.8 million years.
Comparisons of the Perry microcrack data to similar data from other locales were employed in age determinations. Allowances for variability in factors such as temperature, pressure, and chemical environment and uncertainty related to mineral growth rates could suggest a greater range in estimated formation time. Notwithstanding the foregoing consideration, it is not reasonable to postulate a Recent age for last fault movement. Microcrack mineral growth bridges, some of which are quite delicate, remained intact and unruptured during the period of historical seismicity.
During faulting, the orientation of the maximum principal stress was oriented normal to fault strike. In-situ stress measurements employing the hydrofracture technique demo.' strate that the stress field orientation has changed since faulting. The maximum principal stress consistent with the prevailing regional stress field is parallel to fault strike. The magnitude of vertical stresses measured is as expected for calculated overburden pressure. Reorientation of the stress field must have occurred during Pleistocene time in response to glaciation. Deposits of three major stages are recognized in northeastern Ohio. No Nebraskan stage deposits have been identified in Ohio. It is not known which major ice advance or minor recessional-readvance cycle altered the stress field prevailing during the last fault novement. This method of qualitatively dating the last fault movement is in agreement with the microcrack study.
\u-Dr. Voight hypothesizes on the basis of maximum past consolidation pressure of the fault gouge that the associated overburden pressure was not substantial but on the order inferred from an ice sheet considerably thinner than that estimated for northeastern Ohio at the Laurentide maximum. On this basis the last fault movement is more likely a::cciated with deglaciation-rebcund than an ice sheet advance. However, rock-to-rock contacts across the fault zone, as well as the step-like pattern of faulting, were documented during detailed mapping of the deformed tunnel segments. Furthermore, Dr. Voight suggests from extrapolations of fault displacement data that approximately 70 feet of undeformed bedrock overlie t'ae updip projection of faulting. Therefore, it is doubtful whether the fault gauge would have experienced maximum overburden loading during any of the major or minor glacial stage advances when ice thicknesses exceeded several thousand feet. Hence, the age of movement for the fault based on gouge consolidation tests is not reliable.
The most reasonable interpretation of all the data is that the tunnel deformation or at least the last movement on the fault was a Pleistocene event assaciated with glaciation. Candidate mechanisms include ice-sheet traction, differential down-bowing with glacial advance, differential rebound with glacial retreat, surficial stress-relief or " pop-up" and subsurface salt tectonics, the latter as previously discussed. More probable is glacio-isostatic uplift and surficial stress relief during deglaciation rebound. Recurrent movement on deeper seated pre-Pleistocene structures or faults, either by direct propagation or', en echelon deformation could have been possible. Both of the latter would have been activated by glacial ice loading or unloading. The conclusions of these investigations, the opinions of the independent reviewing geologists and lack of evidence to the contrary are consistent; the fault is not capable as defined in 10 CFR 100 Appendix A.
b 1
\2st 0b\
6.0 REFERENCES
(1) Stone and Webster Engineering Corporation, October 1978, Regional geology of the Salina Basin, Report of Geologic Project Manager - Salina Basia, Phase I August 1977 - January 1978, Vol. 1.
(2) Heimlick, R. A., R. W. Manus, and C. H. Jacoby, 1974, General geology of the International Salt Mine, Cleveland, Ohio: in Heimlick, R. A. and R, M. Feldman, (eds), Selected field trips in northeastern Ohio: Ohio 5epartment of Natural Resources, Division of Geological Survey, Survey Guidebook No. 2, p. 5-17., 59 p.
(3) Jacoby, C. H., 1979, personal and written communications.
(4) Richner, D. R., 1974, Minor discontinuities reported in the core description of Diamond Alkali Core Hole #202, Perry Township, Ohio, unpublished report.
(5) Prosser, Charles S., 1912, The Devonian and Mississippian formations of northeastern Ohio: Ohio Geological Survey Bulletin, 15p.
(6) Williams, S. Jeffress, 1978 and 1979, personal and written communications.
(7) Cushing, H. P., F. Leverett, and F. R. Van Horn, 1931, Geology and mineral resources of the Cleveland District, Ohio: Geological Society of America Bulletin 818, p. 33-35.
(8) Weston Geophysical Research, Inc., 1978, Feasibility of Dating the faults in the foundation of kWP 3 at the b3P 3 and 5 (Satsop) Site ot Washington Public Power Supply System: report prepared for EBASCO Services Incorporated and submitted to Washington Public Power Supply System, 26 pp.
(9) Padovani, E. R., M. L. Batzle, and G. Simmons, 1979, Characteristics of microcracks in asmples from the drill hole Nordlingen 1973 in the Ries Crater, Germany. Proceedings, Lunar Science Conference, 9th, in press.
(10) Blatt, Middleton and Murray, 1972, Origin of Sedimentary Rocks: Englewood, N.J., Prentice-Hall, p. 338.
(11) Wall, R. E., 1968, A sub-bottom reflection survey in the Central Basin of Lake Erie: Geological Society of America Bulletin, v. 79, p.91-106.
(12) Peter, G. and R. E. Wall,1961, Magnetic Total Intensity Measurements on Lake Erie: Lamont-Doherty Geophysical Observatory Technical Report, pp 9.
(13) Ahern, J. L., 1975, Aeromagnetic reconnaissar.ce survey of Lake Erie: Ohio State University, Columbus, Ohio, unpublished M. S. Thesis, 153 p.
(14) Meyer s, C. D. ,1977, Aeromagnetic reconnaissance survey of Lake Erie:
Ohir State Univ rsity, Columbus, Ohio, unpublished M. S. Thesis, 172 p.
(15) Wall, R. E. ,196, Geophysical Investigations in the Central Lake Erie 5asin: Uni <ersity of Ohio, Unpublished Ph.d. Thesis, Columbus, 66 pp.
(16) 0'Hara, N. W., F. Mequid, and W. J. Hinze, 1974, Gravity and Magnetic Observations from the Lake Erie and Lake Ontario Region: Geological Society of America Abstracts wit'i Programs, v. 6, p. 896.
(17) Heiskanen, W. A., and U. A. Uotila, 1956, Gravity Survey of the state of Ohio: State of Ohio, Department of Natural Resources, Division of Geological Survey, Report of Investigations, No. 30, 34 p.
120 053
!b !
I d) ud UniGa;b Lok
- mpEggy yyggggy N g,je POWER PLANT SITE F j
M X-Clevehnd PENNSYLVANI A OHIO !"
y fD MAP
-.EX IND f'
y _G'
, 7
..;lb
- P "
,, ,.t!(g, .
~,.
'O ,m %
,1,'-
~D
,g - A PERRY NUCLEAR A- POWER PLANT SITE
i
[d ' ' .
/,)h (l
' d>(py(,,
.\
'N
)>'
3
. Ar 6s./;,,i%e\h s h-Q >; !
~
se ,
4 y
l
< b S RESFRV4 TION ', ,A ,] _
M.,o . cM -d '%
m.
N cx[>
/..
. r.a . l .)-
\. r' %,w g,' .
\
Q
, c- t r ,1 }. . x(
,N g (. ~l
.s icamri
'. *d y* '
- \. )'
thosevelt dd?,' on
'b ,
.( )
(( &
' ,s t,
-( .( hl -
\
-I' x
fi ' ~l73
' .1 '<
[" o, g y\ .q '
. ', , ,[
t,
- % .1 . oa o is ,,%j ,r I
, O .
, Ni>rth perry.'I'l 8,'\
.c
> - . n. .
. % \
[ , , - **f" : ;.'~~
)j ,g.
b
- C 'k ' f . .,,,*,.,l;,[,
%.-l.
\- -
t ..
..+,
~
y O. . . . .
l/2 ',l MI L E :
1 054 LOCATION MAP, PERRY NUCLEAR POWER PLANT FIGURE 1
h' *
[
j k -1 Itti
,!W'75 ?L bI 4,* !] A J t
\
N;
)
w.V .
)
~
- h. e '.
P t ; IL P 1 %
- . N 4
f' o' o' O' o' o! o or o; o S I' h T 7f Si S
.m..__.__
u = = aj k~; a -
-o 2 m. ERT y) - _mA----. _ _ _ _ - - - " N 44
-- r J 8 7 "-U $s -N -Q' ;y
,)
Tib_-7=
-12 p -
-115 9 - -
's
$ !K.a L!NE
. q N w
!! '$ 4
_-]- -- __ _ -- . . _ . . _ _ _ _ ,,
I ", y; (99,9
,j - - . _ ,
4
_12_5_ ._ __ _11.
,, / / j iGHT P.;pg y
_ gg 7 g 1 _u 7 - 4W 33'F
-_T T / $ LINE
--r _1 .- f- - O M sNVEDT Horseshoe type cut cut intake $
Con struction_
Access Shaf t
~
h 2
-Mopped Separately Escovation by DOSCO (rood hoode-Tunneling Wochine Escoveted by drill and shoot method cowN - 448
- 446
_ _ _ _ _ __ - - - - __ __ d 4 44 1 -
442.3[442 cp 44o
-- -= 438 INVERT' 476 R E tC AV ATIO N P R O G R E ':S ee 4 no. 4 ESTIMATED ROCK Ine 5 ev erewe ne 5 et erewa CCNDITION (TERZ AGH1 NO )
k bee .a swee .63 es l3 eak betes ia see.1 esbo se 3 SE4 TEMPOR arf SUPPORT
- b****"" l he8 3 ISie' . _ . _ . . . _ _ , _ GeaMP"' 'P*aa'd by w med sySrEy
, ,4 . i .
- " - d '~' r. 3- e m NG suc v, betM. ence
- a-P.'***a' " **d ' e*W ee .pe,.; een, I'-lo' FRACTURE SPACING A 1 ** S 4 No aallees WATEM CONDITION 4 i
, , espesa.
e e- 2,n' -pre.of lecese,ne. oppeen. 33' O
- e. . e. ef f *T h CF CO'dR .
GEOLOGIC MAP OF TUNNEL EXCAVATIONS
} 2g'f _,0
' FIGURE 2 SHEET 1 0F 24
c. t lkbkl M5I
,t ,c. , .. k. u- ' t. ^ ? f. % 5 - ,6', ' ' V' ' ' t, '
d -0 g\g- g 7; i e, ,s 3 .* [ ---- r' l [ j__ues*_
.. - N , - o 7g',,, , g ge ?o A
- t. ; E-4 7 C, L' - sti'i;* hEf ' t. *, " Y
,n: #
g ., l , I sPrm- o.4 Lot? "' ~T > m d " ' g p'T
/ [.,' ,
y
,x .- ~
j s
- m. s .m .. ..
*g*
rAfii
' : ! > h"" 'l '
C"'
.. F H T ': a t .,97 ' / . j /* 75 NFL:NED J0:NT N.nte v...:--, ,3., ,,, ,. .
____/ _
,k) MEET I Y '
PINCHING IRON BANDS 4J
' O o' o o ol o!
o g, Ql g' o' o g g o. o
** 7 M 7 ?. T i. 6 7 l +!
oj
+e Ti + + 1 "' ,, o o' o . _ _ o: o o' -;
_1_ _ _8 *;L__ J __o _L__.m___L
! _ a_ -j -\i I i r, o [ '
2
-]6f~ l f -52 200/2)_ j._ k l p- -
4 ,
- 25:i _ {3,ni , _ __ _ __, ,_, _ g ho t,. ,_, __ 3, _ _ _qe ,, -- - ,,, _ y _ ___ _ _ ,,,, g ^;3j M d1 L 12 m -
is 3,4. ___3, i. , i , __ _ _ _ _ _ i ?- - - - - -- ^2- - 3s 70* th *oCO n 1 -[ < i la / 4 e t; f g .3/4-171s'.A___'s___-___-________p____-_________________a.7_,,_,,_.
- - - s.o 4 ist i i a- 0 lp y 2+ -22i1/2 9 o ,_ , --ff
_-\'4 - ret i . ',22_________,_ - n __
.- ec L .
1
%___.. _ _ _ _ 2tl2_ _ __,
_ _ _ _ _ . _ _ ___--____43___ l'4 " **h **8* **** Horseshoe typ
-e Excavation by 005CO Q
O
.* (rood header)
Tunneling Machine P L t. N F Vlf #
'w 447 ; cuoeN w ' 446-z _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ - - - - - - - - -
443- - 4 4 i .o , E z 44 l .--. o t
; 4,, . _________.____
C 437: d 435-INVERT Cy c. S ]Fq g
,-e .u .
u e.
- s. : :
l2 r.. . ,r ,s.- I I I
.l .....i. i .... .e,...,2..,.. .. . m . . .m _. _ __ _ _
1233 ')S6
I N3 E W;u 9 ATE
\, $.J.
faz u . .s, T
- v. 9 s
.f 'd O O O c' # On O O o o o o o h $ $ h 2 e i e i i T T. T. t t 9, 3p(3.tj4) a' -' - * ' > N E fi T 53 2 s 5 9 5- - - - ~10 5 (1/86 - - ~ 7 - IU3[4 1 [ - - --,; '1,m ,g g . , 1, 3h uNe - . - . _ h5 H . - . _(31 . -
l_ j 'I ( -- - . -sac ~ -- . --- o u . __ . ~__. ~ ~ CPOaN
- t ( 7 ' ?54[+
;I . . .} x s_e,J 4 i 4 _ AY h ,_, i ' d RNHT y - -- - - -14 3(3/4) - - --tTg3(374 ) ~ ~ ~ '--*-6C g SFHM 53 i ' O LINE (11/4 % / sI '.._q .) L INVERT Horseshoe , d Circular abd Encovation by DOSCO _ , I"si2' Escavation by e 1"s12' _
(rood header) ble. [ JARVA (mole) bla. L " Tunneling Machine y =s 20 Tunneling I" s 20' Cut Moe. hine Cut CRO*N g450 [448 y [446 1 1 ,, 443.61 444
-------'------.- - 442
-- L I on band h440
- N,Ea r L 438 E '*
g~- __ _ _._ _ _ . .g _ _ _ _ _ _ _ _ _ _ _g2 E XC AVATION DROGRESS i _ __ I no. 4-5 na 4 ESTiv4TED ROCK 4 CONOt TECN ( TERIAWI NO)
" TEVPOHARY $JF PGP T ...___.._._.a l 5(SIEV *- II4 I ' cE C M N'; S pa O N') . _ . . .. -_ ._ . _ _ . _ _I___._.,..____.. _ . _ _ _ _ _ . . _ _ _ . _j 4 - 10 Not pers pene FAACTUPE 5 PAC ! N'.}
WATER CONDITION _ _ _ _ . _ _ . _ _ _ .no %%l(h,)__ LEPTH OF COVER GEOLOGIC MAP OF TUN'iE EXCAVATIONS FIGURE . snEET 2 Or 24 12k] 05
gj l
. m a. . . ._
l, y INDEX m=__=.c MAP:.__
=
_ s, g y y g. t 5
" - = = - - - '
Sc c e . I 12001 m (g 4 ty I
,' h % L 1
N . i h.
-w
_ .(1g' H AR D, ' A N, SRO
- N, CH E R T Y,
;PCN SED OR LAMIN AE , ., , , MC stb 4E AL N3 SEC P4 9 7 : '. 5 1
4pb ~ -- CiOCO N TIN UCUS IRO N EED $ E E PA G E FACu rEATI
, _, j21,_ H AR D, LIGHT G9 E Y, SANDY q g y g g g g g g,,3 7 3 3 9 gg w CH ALE TO SILTSTONE PED
- a _21, 2.L.;21.,_
/ DI SCC N T iN UOU S IHON PED . , , ,
9
;g q7 E,PMEAK -
ZJsr MAi
,, ; g WITHiN GPEY BED Q' \ FRACTUPE ZONE V f 7 ' M 43 V A ' V / ' >c fNT .F: A 'T H rAULT 4* '
4
- c. e ' <
4 r i . .. , sew.m Tm . , . .
" ' '"'J'**"" "' '"
- Djv5 V E R T IC A L J 0!N T
_.i_..- 8 f.73 l )NCLINED J0:NT samte r e ,3 -
,p _, , . : rrj w , r.--n.. o.x .
ect 4 e I
; Y .: E E D!NG, STRik E AND ;!P
- e c s c e" er.'
O O O O o o O O N o o o o o o o o o
+
e
+
N
+
e+ e+
- o. -+ +- n. .
T, c. w,
+
N
+
c
+ e. o+ -
N N + . N N N M M M. M: m mf n m M M T T
,_'y-)'_.. ...w. i __
IE
- , m ,
39
- - - - - 2.1--h -* 4 . , - - - . 2.4 4,_ __,,_ _ _ __,, _ _ *_
4 Q*) to
,'I ,aaw.-- - _. 3 . ,- l. - -n 40)-
8 so I %
- - - - - - ~'--
l
&*LI so p . _ .
1 L L" ')**-- --- - . - ._.-._._1-._._
- : - - -* 2 -
y
.u ' 2 i
9 30* - 4 ? n 6 80* 13 3 4 s t
~
121/2(2 3/4).-- - . _ . , - .e_7(3j,)) _ . -8 5% *g g - *-85(25>+ .~.--4(2)_._._._. _ _ _~2ts e> 1
.2 -_-. ---. 2 2 s '_} L - 2gy,y _ _ _ _ _2 S_ _2 s _
39 4 40 s_ 95 415
~_ fy C Kg,_, __p - ..<, 9 ; u-.
47 47 X w P L. A N E. V.E n
;; 450 ,CROa5 Y 448 I I 4 \\ \ _. .-._.
- 446 44 L 2 4444
~ ~
2 442 ~ - - - -~ 7 yo
~
442.3
,u _ __ __ _____ q_._)t_ _ _ _ _ _ _ _ _ _ _ _ _'~__ _ _ _ _
g i 2 438 J:NVE ai (Aq11-;E(3]
- . . . - ~ . .x == ~ ,
e
. , -l - I.,j ....
m, -l-
. .z.........-.........
m.....m
- g. ,- ,,,,- r... ,;.-
l ,= .- p,s ._. l ,,. .- l ,... j _.c_ .... l
- t. . .. <. ..,0 l,,,.,.. . ... j _. .......
, Ag -- . .,
e
j l ,o E JUM TE t h.c k nes s. inches if ENLARGEMENT BELOW o o o o o, o o 0 o-I I T S 'I- Y 7.: I T. e e. o s s e- e N b-l m l l l I , s b 2 we_sm'l]3r_-,=r_ i!J b l _ _ _ _ _ - _ - + e
. . 40 5j ,, , r , - g -- M 6'0 w' LEFT SPR'N G LINE
- 36 , y 4 - -L. ,: , , - . - Q& crown
,..43 5 '32' 29 ,, a 60 $ RIGHT SPRING le QQH.120 "tl------- - . . -
r o L1NE a SPRING I 'NE light grey seltstone
-y?- __
iron band light grey sandy shale ENLARGEMENT CROWN - 450
- 448 4449 446 444 - 442 -440 INVERT 438 h I EXCAVATION PROGRESS l wil, .i:fa ., e .., .. .... . 70 g = g o g g g ,,, ,,, -" T E MPOR ARY SUPPCR T SYST E M BEDDING SPACING , f{
FRACTUPE SPACING 7 Slight me.s w na stew = on g beddeg bocevre WATER CONDITION 120 l' ivbsw oc ( e GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 3 0F 24
P00RBRIGIM 4__ . . - L_
'd' **19 4.2W llNDEX MAP jf- OE 1 \ f, l Scale : l* 12OO' dk l *' ( HARD, TAN, BROWN, CHERTY, " ' ' ' MO!STURE ALO NG BEDD k , - 4 {1[*_ --+ k t- IRON BED OR LAMINAE PARTING , f gSL - DISCONTINUOUS IRON BED y [ SEEPAGE FROM FEATU
(, , %2> T,- 12 h)* HARD. LIGHT GREY, S ANDY SHALE TO SILTSTONE SED e OV ER E XC AVATION OR OVER8REAK ZONE. WAX
, Q .f x\w $% _3M*.2]*_ DISCONT6NUOUS IRON BED AMOUNT SHOWN i WITH:N GREY BED -~_ -__.4___4 _ 7J j FRACTURE METHANE GAS, MAXIML l w 4 PERCENT L E.L. wlTH L I
- l 42e77
' g FAULT OF OCCURRENCE BEDDING PARTING
- Note:
N wm be r (s) in poreetheses sedico'en be
\ VERTICAL JOINT i
[t y3 Num t;er preceding poreethesis indicate-l I N C LI N E D JOINT obove or below spring ime at ste t .on
'kiai u BE DDING, STRIK E AND DlP rr e u sur e men t T
SE
$l 9 8 8 I 8' O! Oi 0: E $ E. O M O e 2: A s, a 2' t 2: A s ; a ; a s a 1;6OL ! 1 l L l l L- J 1-- l -
i asm.) h _ _ _ _ _ _ _ _ _ _ _ __1!!! 3_4 L _ _ _ _ _ _ _ _ 2'1114 1_ _ _ _ _c22__,_a,_,
;; O h - - - - - - *Bi - 3 4 ab #81-w-,, is------.,i;- ,
4 r 2
' 3; $ g ioth; ,,.
b + +l g ' 6.0 o # u - ,o a2 z ,, p .% RSS e
+ ,, gg 3 o- mg.
a o +u sum 8 is.5p 4) ________------ a4) , a _a
.....m.-----,20,--
4 2
$ 4251154) 60 --
y 4-4-in PLANE view [ 450 CROWN g 448 7 446 ,, ,
~
7 444 M45:,6, q _ _ q - 2 44, : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
$ 440: $438:inyrg7 --
CROSS- SECTION
$ $ no. 4 4E o.. . .m.. ....n. .. ._.. .. ... . . ... .. .... ..,. .d sedding fr.cteere. .nly
( ....... I% fl24.F's 6
,# . (.hes. s
4G f b E v.U W 4 ',T E L n c on.. it nectes g%
'N NN o o o o o o o o o \A 7 7 I T T T T T T e e e e' e e e e o j I s
f 44 44 - 4:
.o Q
gl! g a1 i
. . - - to _ .___.-___j . _. . . . gn *!aLEFTj SPF:NG '
LINE 43 41 l! 7Gl-, , 'gh . 3----
' 52 -q 4e . $,IjCRO*N ,i u "
4- o [ M ' _______--__'y9_3_--_._._____ __.--.
,_ ___ __ __. g o d L J
HIGHT S PR i'i G
-------- 3e A l gj LINE . -- - !% I ul 4S (H /4) N 5-75 #
l
!g ]L
__ _._ ._. A.. , . ler,yeobe from station 9t64 encountered briney groundwater and clay, 4/14/78. 2 demeter probe yielded a full bore of inflow.By 4:45A.M. Inflow decreased nume. At 7:45A.M. groundwater was spurtihg from bore at 3 second intervals. comw
' - " ^N }-450 , t l l ' ' ' - '- 448 - 1 446.2? 446 h444
_____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ r, 440 lNVERT' , lEXCAVATIGNPROGRESS
} no 4-5 ; ESTiu ATED FOCK
_._____l___ , CONDIT!ON (TERZAGHi NQ')
- ' T E MPOR A R Y SUPPCRT 1.] ,\ *s"
_ _ . _ _ _ _ _ _ SYSTEM lL li 2".10
- l BEcc:N3 SPAC'NG soddiae wwe. eaty SFActNG l
] FR ACTLGE ,,,$ **" . '* See Ne'. Abe r . .d , '"'l w AT ER CONDITICN + DEPTH OF COVER
t16 s' h ,% *sheiwe GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 4 0F 24
sn es -. _; . . h ilNDEX MAPI P-T
.:___a$_ (f <;v u g o ,jh .' i..n:_r t:.:s (9
Sco e: 1 =1200 p -
* 'I 4 .
i %( l' . [ k . { ._ -. - !
.._4___._
( , ___4 (tg*___ H ARD. TAN. B PO*N, CHE RT Y, t RC N BED OR LAMINAE ' ' ' ' Pa RTlN:;
. V01STURE ALO NG SECC i - DISCONTINUOUS IRON BED y [ SEEFACE FROM FEATU 'k, !.211),*_,,,, H A R O 1.(GHT GREY.SANCY CVEREXCAVATtCN OR t
f,. e l [ j f g. I HAtf ,0 S LTSTONE BfD 2' OVERB4EAK 70NE, M A) i ! '", y i _3tg-21' _ 5!SCON TINUCVS ' A CN BE ; V. lT h N OREY BED AMOUNT SHO*N I
.%g.._- A.__ _.
F R ACTURE METHANE SA S - M AXiML i e FtRCENT LLL *tT H ' l l C LI 4 & 77 Cr CC,.URhfNCE SHEET 4 4, t
) Bf DD;NG PAR T!N G eye, .
NSW in Wen &nn m'n be
\[ . **rd /. V E ATir AL v i
- m. T N um be r pr eced.: a parent hesis ind,co'#
gn
-g, - _ _ .
y INCLINED JOINT ec.. , or t e:cs. so. N li e W s'm SEDD:N1 ST RIK t. AND D!P **0'd""'"'
~~Fy; ,. l O O O O O O O u O O O O O O O O O e n. e e o -
ca n w e e k e a O
$ $ et $; 4 iI e$ e al $ ,
e $ - E $ $ $ $ i i- r g ( l I l I I e 60 N-0
';5 72.fl9L __.____ _ _
_ 31f!&idL -_ _ _ _ 1*24I38- 5 - - ~_ _1s.- _ _ _- - - 43 0 ) 3 I ! hu ,I , --- - TUD72T - - - ---- -- -- - -
,( --- - --- -
6p _11_- 442 l g m __ -.... ,
'3 M 49 $l'@ j
_V ,t p' . a
- s. . ._ t e;A 52. -
..('. 9 ' . _ __ _.. _ .__ - ~
q c o L -- ' __3__.=. 8i i W
-_--' ' Mirror Image of ^ ' ' ' ' ' ' ' ~ ' '
SFi S I depositional feature ' I ' Y. l illustrated below D \23 o
% _._./_.__g.__-.__________._._.
O _ - - . _ . _ . oI.Io -- rrcz(_ _) _33O/4) 26 2-i/2 \ s 36(i) a;1g / -------------- 3, asr )- _ __ _ _ _ _ _4a i . > f s
;!_j,g uc g 4e6 2) /
_ _._ _f _ _ __ _ _\ __.__ _ . _ _ _ _ _ . _ . _ . _ _* FM -. a k gr y sha e PLANE VIEW n - _ -- '# An 80' long f
% SPRING LINEg At 3:45 AM. t to half bore F- " 450 7 CROWN , , , ,,, , ,, , " i i - s - - is t,
448.. . . . 446 7444,9 _ _ ,,, q __
$ 444} - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
n
> 442-W a 440_- tNVERT CROS0 - SE CTION w *I. tI.j~ 9I. I. C 4)~ 4l~ 4f 4 no.4 5 cent a..d . poll.cg .t.ag b.ddag f.ec. e me., c.o.a = ,e..... ...,...n._~.......;.x.w.w-p ,r,= -
7777 7 t _ . _ , _ , _ .. s +,2.,m...,_...., _
P00R ElWAL NG
.E WUM AT E t h sc h e es s 'nches e
of k.
=
O $! o S k k h O'
= 1 te n n a n a - l l - l l l o & 'b ^ ^ ^ ^ ^'
7
'l 6 LEFT SPRING o .. __ _ __ , usE
" ~ ' _7 -
'z' 6-s , , - . . i . 3 -- 1 of -
CROWN wN MW - - [,- -
- c '___ w _; q _jggw472 g __----_- -- RIGH PRING
__] c o LINE
,J . .
- i. .
.ll52 p' ' ' , , ,a @
O z LT GEOMETRY displacement - 22" rtical displacement -12" 47'E r om .17' broken rock ( 1/2"- 18") i 452
- i 1_ i CROWN {450' . - - - - - . - _ . .' - m c 447.5 448 w
t . , [446
, , - - - 444 - 442 INVERT - 440 g R EXC AVATION PROGRESS no 4 5 5 rock bolts at 2' centers m r.bs of 3.5' ESTIMATED ROCK
{ centers. spanned by ...e meshm CONDITION (TERZAGHI NO )
' 3.5' centers spanned by wire mesh Full shielding ! "'
fy y 14' 8 ~ - 15' BEDDING SPACING tures only, spaced as above FRACTURE SPACING he me.oute along oppron. 20% of bedding fractur es and rod bolts WATER CONDITION P so9 T subs e, shole)+ D PTH OF COVER t GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SilEET 5 0F 24
, n ) ,U a% %gg l+
e -d hM33 p( ma llNDEX MAP F { I Sec.e: r. izoo l 5" B o's r 6,4 { k \ _.,_,, q 1g'* H ARD, TAN, BROWN, CHERT Y, " ' ' '. MOISTURE PARTING ALONG BEDE k- IRON BED CR LAMINAE g g$L DISCONTINUOUS IRON BED y [ SEEPAGE FROM FEATL
\g i#. , G, _ , _1211).*_, _,, H A R D LIGHT GREY, SANDY SHALE TO SILTSTONE BED OVER E XC AVATION OR % t- .2 OVER8REAK ZONE, MA) e., *p., ~ -iog-2)*~ DISCONTINUOUS lRON BED AMOUNT SHOWN % p W IT HiN GREY BED j FRACTURE /\ METHANE GAS, MAXIM- * # PERCENT L.E. L. WITH FAULT (irith sense of motion) l.101.ir l \ g f 42s OF OCCURRENCE l 1 BEDDING PARTING eNo, :
NamW in parenmenn endico'es t l
\ l b[ VERTICAL JOINT J
[el 73e INCLINED JOINT Num ber precedirig parenthesis indicate obove or below spring line at station
- (7g y mea sur e ment ! T BEDDING, STRIKE AND DIP O O O O O O O o o O 0, o o o o.
o.- +
- : e e i e e ~
8 e e e - ~ m - m 3 2 2 2 2 2 o O O :! 1 r, il 5 ii J 60 l
] l l ] I 1 l. I' l __ L_ _
1
$ 5 3a 50 5 - 51.5 - - -5 7 5 - - - - - - - - - .-
A. . n3 - 3(T2 gi)Z - .~.3) ~ -~~~ '~
~ ~ '- -
g ,sibt . _. _. . i - - se 2 f r u: u w O ~ yny_L..w _Npg [- _ e y-e
,3 mo - - _ - e t s_m - .__ _,___. _-, _
r__, 2 +j d 5'
"'* M '5" n , . -r- 20 r ,-4 4M, 21
- 4' 4 fE r v (N_, - ._ 'A f' t f-oy , C _ ,
74 8,., 5"- w G.0 c_,# 7 , i t~ 6~ ~6 y m z {
? 36 + % . i 24 5 ea ' m f_?- A ---- --m- - -N; .. _i ss. s__. _
2 _ _ _ . 33 s _ .._ u , ,s a , , j - - - - 33.5- - -330.v4) - . .,_33 - 32 9-- - - - - 3 d P- - - g 520; u
]j si.5 -- - - - - S t s gg e I l F w ( # o.) Appare SMN PLANE VIEW b.) Actual v Gas bubbling 6" trough C-) Sf fike -
through water d.) Dip -op e.) Wid t h [* 452- CROWN 450- " '
' - i s i s N _
i , , . g 448 .m._ __ _ _ _ _ _ _.,_ _ ,_,_,_ 446 - - - - 3. - y
- - s y4442- - - - - - - - - ~ ~ ~ ' ~ ~ ~ - " ' - - - -
w 442- - w 440" INVERT CROSS- SECTION ie n -6 {h je $ de iR no 5-6 0-24 of overbreak betweer. rock bolo no 2- 4 5 rock bolts at 2 centers in steel ribs at J - 3.6Teaters 5 rock botes av 2'convers in speel ribs as spanned by were niesh / 2"- 10 I"- 7 " 2" 9" II 'hIding -Rock botes in 1.ner plate' appron. 6' c"mple'ely o sheeld upper 3/8 of funcol 10~ Bedding frocevres & 1/2'to18%k, Bedding fr hiqNy fractured foult zone w/gge No inflows M '"A $1.ght iniIow from roch [ from fold No inflows WeGpp.ng from sh.elding /
\ / boles $ \ , /. wa'er -
L 2 inflows et opprom. .I gol. / min.
....,.bs.r...,s.,.,
I.. 'w-1238 064
9 4 G i 4 E UM TE t hic k n es s. inches N. f O v b k e $ h
' ' kk '
k kO 1
's 2
P 19 . -22(1-21/2)- - - - - - - 1 S _~'4 -' __~_ Z 2.12_r____ 7 p_ggq6 0 ' LEFT SPRING
^
LINE g I,$ - r^ - N Wa rg , ew q l'-3' overbreak
)
_) ['_ ._ _ _ _ _. _ 1 -y;y-
~ ' l'1' 6~O RIGHT SPRING E
7 7 '.%$k2' 2 5 = - * -- -'- a OS CROWN - 454
~ - 452 r 44882 -._._._.._._._._._._-.b.-.~--448 ~ - 444 INVERT - 442 1R $ R[R EXCAVATION PROGRESS ESTIMATED ROCK CONDITION (T ERZAGHI NO) ,
sGJJe TEMPORARY SUPPORT Cuttmo gt bur.ed pior.den overbrech SYSTEM Appron. 6* 10' BEDDIN^ SPACING dding spo<'a9 FRACTURE SPACING N'g *f,,*jn'" WATER CONDITION 107. ' subsur e shal )+ DENH & COM f GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 6 0F 24
P00R ORIGINAL 4_ . ___ _ 1 kU4 llNDEX MAP D 5"e*5 f h Le:i,izOO- LS
\ 2 Y s
l * .WOISTURE ALO NG BEDC Y, _g {t _ HARD, TAN, BROWN, CHERTY, ' ' ' ' PARTING w IRON BED OR L AMINAE g -- OlSCONTINUOUS IRON BED g SEEPAGE FROM FEATU
%., Q?- _, _12 Q)*,_,_ H ARD. LIGHT GREY. SANCY SHALE TO SILTSTONE BE-OV ER. 'CAVATION OR E hEAK gg* hE' k 3%0;L DISCONTINUOJS RON BED 7 yOUNT 5HO N METHANE GAS, M AXIMt I
i' 'E % g FRACTURE PE RC E NT L.E.L. WITH t l FAULT 7 OF OCCURRENCE j Shese 6 \* A BEDD!NG PARTING eNote: l Number (s) in parentheses indicates b
\ M /, VE RTIC AL JOIN T N;m be r preceding parentP,esis indica'e l
h h#3 INCLINED JOIN T cbove or below spring line at s t a tion
- g m e a sur e men t.
3' SEDDING, STRIKE AND DIP E S R R R 8 2 2 R ? 8 S R 2 2 8
! 6.0 k k k 5,
k k kk a k khk k k - k c, 2 A.
, . . . is s . *t_,
ff _,. --
._._,_,3__,_,_,,,,3o.
z g -- 43 75 o O m,y -- u- _ r m ~ ' } , - . ' g IQ K'-
$ + f 20 3 ' ~ ls ^ -- g y' " + g n., .. pW.2, j , i b 4 w 6.0 ; L,1, f;L-_ _
8 + M S- .. l k . 24 a . ,$9 ' y, ua
- w)h _' ,, ~ ' . . , , . $8 " _p.
g 0 __
'"13-'"- ---,,=:=
3 g r
. . gs J ~ T ~*~~ - - . a., a, .s. z g . \
p EO }
> Weg b
5.n.is , . , . l e -**w i lenses of light gray PLANE VIEW sg N o If Cfine grain sandy shale pinching in and out
$ 454: CROWN U 452; z 450;
- z 448- - - -
L g s , s_ -.-.-.- -._.-._._._.- =.- _ ,- t; __
> 444-w CROSS- SECTION ;;j 442 INVERT I :
R b bent ,*-.* between rock Z :: pRib bent 2"-6* between rock
$ I e Ih ! e no.4-5 boles no. 3- 4 "1 ( bolti na 2-3 a , b_ bent 2"-. " ber-sea rock boire no. 2 - 3 Appron. s*.is* - 6" between rock Appron. e*-is* y,, bent Bedding trac tures only, spaced ei APProa'mo'ely 20 % of bedding fractures and rock boles sho- emnimoE No inflows . _H--- W
- 11. water I ll. iv jyr ace iho t 4
m 1238 066
PDDR DilBINAL 1 NG IE MUM ATE th:c k nes s. inches '+ of O e S b b O k k $
- : C : : : :
u_ 1 i 1 t , i -, o <d P 2 ~ IN _._._p,3_._1_.__.-._.---.-(2) 3 gg. FO'g LEFT SPRING LINE 34 (2 $) 3ng a ( ha CROWN
-~ -
_ -.- ' _, g _ -.= - 4 ~' # --~---r--2 C RIGHT SPRING
= - - - - - - - - - - - - g' 6. 0 z LINE 22 ', ' + 23 ' ',
Q ri O2 CROWN r t454 h452 _ _.__._. . _ 450.lt
;448 -446 INVERT ?
- EXCAVATION PROGRESS e -
) no.4-5 ESTi"ATED ROCK COND.IlON (TERZAGHT NQ )
y wwe mesh T E MPOR ARY SUPPORT YSTEM 6' BEDDING SPACING FRACTURE SPACING Appron.10% of beddmg fractures show moisture WATER CONDITi( N ' DEPTH OF CO'<ER i 19 . h. . o . +[$oYe) GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SilEET 7 0F 24
f90R E M t
~#1%M A j C
l INDEX MAP' OLS k ScSe : l"s :200.
, +
l I H ARD, TAN, BROWN, CHERTY, f, ' - ' '. MOISTURE PARTINGALONG BED: 9(g f _ qi b \*
-t 4(h- IRON BED OR LAMINAE p - M' - DISCONTINUGUS 1RON BED y ,,[ JE E PAGE FROM FEATt
_, _12M*_,_ H ARD. UGHT GR E Y, S AN DY i l N. ,s l b. +e SHALE T SILTS TON E BED 2' OV ER EXC AVATION OR O ERB EAK Z NE, MA: lh l
;., J@Lo;2d_ DISCONTINUOUS IRON BED :
W IT H t N GREY BED SheetL _gg METHANE GAS, MAXIM
-s j '3 FRACTURE A fi '
- i O70 PERCENT L E.L- WITH
\ g FA ULT 4 2s-n CF OCCU RRENCE BEDDING PARTING . Note: \
{ V E R TIC AL JCINT N ombn (s) ai pareesn ind' cates b l N 88 /, Num ber preced,ng perenthes:s ind, cat
--(gg f 73 INCLINED JO!NT obose or t elow spring line of staticr ' ' O S d M ** "' ; y BEDDING, ST RIKE AND D1P C O: rj o o a O o o o o; O O: O O' e, 3 O -
c4 n e u* N e m o. - m o
-A A 3 : A A a a a a ! 1 a 3 ;
i 1 1 -L_ ~ i 1 L. 1 1 m 1 m 1 -: 1
- 60 o
Z
,, ja - - . , , - - - - - - -u tvo- - -- ,o z $ O & '~~l' - ' 'E $-~ ' ~' ,' ~ ~ ~ ~ 5.5 g, . ^
945)- - - - -9.5 - - --- -i 3 5 (2 3 6 6.5 - k ,+ f AC" y " ~
$, - > u \'
m a
+ m x w 60 3s .
- 3. .
-u . ~Js . . , eg_stn- , <a 2s /. / v< L-- /
y,- o + v m z n i s
._,',.____%_____- - / / - _ _ .2 _ _ _ .m y + 32 M 7 O -,.9__ _- 2i _[-
5(u t-- - - -
- 4.x2;'
e- --; . . . c 28 p L f60 - 4-l It gray sandy si!?slune~ / f---/ dark gray shale
> 19. gray fine groin." cherty" siltstone P- CROWN U 454; -' . i --' ,_,.m 452- __ --------------________________ ' - - = = = - -
E 450 2 4 z o 448-
--.--L--------------'- - - - - - - ' - - - -'--'-"-'-----Q- - - _ - - . . _ . _ _ _
k> 446-CROSS- 5E CTION INVERT 5R T 9
- R <
no 4-5 St.ght scaling from crown along entere !eng*h of e .rovat.on ( 0~-2*overbre 5 rod bolts et 2' centers in steel obs at 1.5' to 4.0 centers, sponned A ppros. 6~-10' 3~- 9 ~ 8edding fractures only Approx. 20 3 of bed.ling Iroctures show min mal me.sto,e {
\ 14.54=oter .10E 2%sur.fa'a.12 bole 1238 068
k G i 'E 4 MUM AT E tNc h r ess. oche? a a e. s e e 2 2 8 a 7, a a A A fa a
-i i , ; -i - ' -[- j ~ 'O r O -
2 y 14 -
.22 @
l --
, _ _ _ ._ ,- . 42)- - ' ^ , 3]l! '4- - -- , - - u s.c *w r,----
L EF T SPRING LINE 35 l g
- , , G.
CROWN M_ i $5 ' 'NY ' ' ' _ _ _ _ _ _ _ _ _ _ __ r,.o $ RIGHT SPRING
}
3s . . l u O - I "Fe" bands 4j 'i-o sprin gline d I 91 f CROWN r, 458
- 456 ~ ~ ~ ~
s i ' k "% t' t s' 't - 454 e s 451.45 452 450
, m i i , i a - 448 ~
INVERT - 446 f :" EXCAVATION PROGRESS St.g ht scohng from roof ' 0~-41 no 4 ESTIMATED ROCK CONDITION (TERZAGH1 NO.) Son, rebs ne 4.0' center, T E MPOR AR f SUPPORT SYSTEM e* no 9 ' BEDDING SPACING FRACTURE SPACING
$1.gke me.sture on oppros 60% of the escovation WATER CONDITION t o l .5 ivbior f o e ( + OEPTH QF COVER s
GEOLOGIC MAF OF TUNNEL EXCAVATIONS FIGURE 2 C; SHEET 8 0F 24
P00RBRIBl01 ( i
~nw w i n
_f 0? IINDEX MAP'_ _ _ _ r Scale: I"= 1200. (f ~ O SYMBOLS
\
kh- --
\ ----
_, HARD, TAN, BROWN, CHE R T Y: IRUN BED OR L AMINAE
,_jj-M0ISTURE aLONG BEN PA RTING , 5 --'I-- DISCO N TINUOUS IRON BED y [ SEEPAGE FROM FEATL '#. b; _ , _12h)*_ _ H ARD. LIGHT GREY, SANDY - OV E R E XC AVATICN CR 4
9- SHALE TO SILTS T ON E BED - 2' CE BREA ,MA) sheet 8 ,
, y, T@g !
_ .10Q)Lo;2]*_ DISCONTINUOUS iRCN BED WITH N GREY BED g g qy 5 W METP'NE GAS, M A Y ! nd -
~
{ j FRACTURE /\
) *f FERCENT L.E.L. WITH g FAULT 4Q-28 1 Of OC CU P RENCE j l 1 ) DEDDING PARTING eNote:
NNN' 00*"'8" *
- C U I
\ .='lB I VERTIC AL JC!NT N a rn be r precateg 00ren?Nats todicate
_1- . __ k
/)75, INCLINED JUIN T obeve or below sc r m ; fire of s'a t ion SE CCING, ST RIK E AND DlP "'0'***'
7 O o' o' o' o- o o o[ o,, Oi o O o- O e O Ol. -. cw c. e s- e o+ w eri. e, . 3,, e+ ,j o.
+ +- m+ + & * +
- f
+ +
m{ + +, +i
- E E b!
b; b: Di bi
- E i E, E. *_ f. E E .
6 0 -- L-~ L ' L ~- A I ^ llo i c,l l! l h-- 'o u ~ 28 ~ j H " LO g747374r - - - --- E IN -- 17 - is 5(21 1 7 i21 2 hh [ r'- r n= w---,9, gD 277 e 2a r ~ 'Q 4 - ^r-b(
+
g 6.0 ' lr.~__2G W _W 1 -hZ"A_W
** I Q\ ' J 4 I
nA1 - .__ _ _19 (3/41.-- -
' ^
S $$ ,
-A y
Wh o"---- ,- r-d_ ,_- -g 00/2)
5 g -
3 g E $b 2- - a I g u l+ p LO I - 4*
- 8 I \
, q- s-41a PLANE VIEW
{~ k mass in slight fold [ 4 58- CROWN U 456; g 454: _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ r 452- ^ o - 450.1 . L g 450 T.
$ 448 ' '
446 INVERT OSS- Smm 9, : 9 : e 4I: ing no 4 Slighe waling olong crown for ent.re lengeh (between rock bolt 2 & 4) 5 rod bolts at 2.0* centers in steel ribs of 3.S* to 4.0' centers, sponned by ..re mesh A pprox. 6' A pprox. 4' to 10" Bedding fractures only trace, of me.iture oloag a couple bedd ag froctores
.., _o...
X.4 luhWd2LR[1hMt 5 w.e 1238 070
A " NG i .E MUM ATE t hic k nes s. ie:bes .c of
\
k i ki Ol b $ k O k O. 1: N N 1. N E N. 64 s' N 4' CW A N
&I N ! l i i 4 1 1 i O d 3
a tS \1 11S(3/41 19(1-112 & ng 3- 154 ' LINE _ h Miltn_- - 26
- 27 2 w 0
i @O CROWN
- 19. S I D ' ' 15 5(II,2) Iklf ggg7 3pgg IKiT2) 'tS 5tW - g LINE
-37 tim y OS 456 452d 454
- 452
._450 ;448 - 446 INVERT -
5 R EXCAVATION PROGRESS >$ ESTIMATED ROCK CON 0lT10N (T ERZACHI NO ) es et 2* centers en steel ribs et 3.S' 4.0' centers spanned by wire mesh EM RARY SUPPORT 4*- 9" BEDD!NG SPACINO FRACTURE SPACING h WATER CONDITION
,__'_.a*'+ 7 8 '- DEPTH OF COVER GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 9 0F 24
u
~N
_Ai ]{G% llNDEX MAP 1 SY M B OL S
.3 N g ,.; =1200 (2 \ .
A [ _ g {1 _ HARD, TAN, BROWN, CHERTY, ' ' ' ', MCISTURE ALO NG BEDC titON BED OR LAMINAE PAPTING
, +
g - N' - DIS"CNTINUOUS IRON BED y g SEEPAGE FROM FEATL HARD. LIGHT GREY, SANDY
$heet 9 Y(, %
S ~ -'1)~* ~ SHALE 12 TO SILTSTONE BED " OV ER EXC AVATION *AR
%'; +g , ._'ex- g',s,c=Tggus goN s *o !!!o"a"si#N" ' " ^ ' FRACTURE A METHANE GAS, M AXIM' "w% $ FEFCENT L.E.L. WITH QM l \' h FAULT e 2s.n CF OCCU PRENC E ! 1* ' BEDDING PARTING . wte:
N u m be r (s ) en parmheses indectes b N N V ERi lC AL JOIN T Num ber preceding ;orenteesis ind. cote [)73 IN CLINL D JOINT obove ar beicw sprms hoe et votier 33 *'3*d'**"- i [t-. y BEDDING, STRIKE AND DlP Oi o' o! o; Ot of O! c' Oi c' O- o O- o' O o> w Ni N5 Y D h I. kl 7 7- h $I E 2 Si 2 2' l 2! 2 2 2! 2 2 s s' &;
& s' &!
8 I ' ! I I I 1-- I I I Omz 6.0 o
- 7- , .
,28 .
O - ,"f_I - _ _ _ s% - - .. __ _ _ . f " I + 3 n 22 d$ll/4)" - l'4 - 270i*= ' 4 a: L [ + w 6.0 o + v ' U2) ' ' ' - ' Q3 Z 3'4 5 (11/f' ' ' y t % . ., 2: 24,+ & .
. ^ 24.$(t/2) -
_ a sot gq"3 0 p py- S:: - - _ _ _
- 8 Sjtv2) _ '22 g a: ." 24 h-5 / \
p Go ; x I w I \ W PLANE VIE # crowr 456l, CROWN F W 1 , w , , - _r ' 454-E 4525 _ q _ , 45,.4 5_ 450 _
> 448--
W 446- CROS S - SEC TION d INVERT
$2 i 1 lje ne. .
5 r.ch bolts at 2* centers in steel ribs at 4' centers sponned by ..re mesh 5 c.ck 6"- 9
- Bedding ,r.ctures only
[ st .. , , ,.a
\ S,..ht m..s te ,,.m .p.n ..r t .n . o, ,.r .n t.,e i
s _m .
.t , , . .r .. .. u, J
1233 072
P00R~0Rl81NAL 1 NG
,E WUM ATE \ ,f hitit ees t. %,
enchei of } 2 R! Si S: S: R E! 2l 8: i $ t______L.__.- E i ii i' d i ii il O Np 2 2f( 3) 22% - 26f2-1/2 M 7 O
- ~ ~ Egir st1Til - -
-- -12(1) 71-@ - - s.o '* LEFT SPRING 11 to 23 41(1).
M Z
- ~~
( $ CROWN b 6 $PRING f- - - E 0 N'OH[l N E
<= 's L hw g o
O .t CROWN - _ 458
- 456
" ~ _
^
454
- 454.0-
-452 C450 44R lt '. ER T ' ; 1XCAVATION PROGRESS ESTIMATED ROCK ~
_ overbraokb e tw e en_ roc k holtret.2 JL Al".- 0_ Je*R(M96t_ CONDC #ON (TERZAGH1 NO-) by wire mesh T E M POR A RY SUPPORT 1 II '77 3" 9
- SYSTEM BEDDING SPACING l /}
FRACTURE SPAC:NG t No seepage WATER CONDITION { is.s' woi ' + OEPTH OF COVER I 98.C' su su@Se sho e e GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 10 0F 24
i I r B r. 7.3acm i
? J IN!iIX MAP' F SY M BOL S f,
Fxcle:l's12 %. kk k I
~
i e~ HARD, TAN, BROWN, CHERTY, IRON BED OR LAMINAE ~' ' ' ' PARTING M0! STORE ALONG BED' 12 (0-2F FEAT g g (4 DISCONTINUCOS IRON BED 7 g SEEPAGE FROM HARD. LIGHT GREY, SANDY OVE R EXCAVATICN OR
' %. %. 12h)*
- CVERBREAK ZONE, MA
. : C T N 10 R N BE AMOUNT SHOWN *, r... --
- lT H!N GREY BED METHANE GAS, M AXIM
* $ FRACTURE PERCENT L.E. L WIT H FAULT \ g BEDDING PARTIN G o Note:
42an CF OCCURRENCE l \p'k-Number (s ) in pareetheses indscotes b l V F R tic AL JOINT Num be r
-j lN CLINED J0.NT prece$cg parent hesis Indecc'.c above or below spring line at static g l 75, * ' 3 5 W r "'"'
j y SEDDING, S'RIKE AND dip o' of o' of of o o 01 o o g ol or w S. Tl d t; ei ?; ; 9 ':N ! ? o!'
? .i ol!
T !!
?l 3- 60
- l l '
2' :j q : Ej 0, I 0; R 2 I 0; E
-_L e
z
& 38 E g , y,3 ,2g-IL 14.5(3' y4' !! , as ' _- pr__ Mk" ; 2'f, 4l, 3 ,. g ^
ih - m b b w 6.0 __ o 4 v na z p x 131t 1/4- 10 - _D - -
, Y,3 '
g 3 0-- , y -- - - - --
,A Ter . =-7 ,
z Y 15 0-a m 29 '"# 3 > / g /
/ ,/
p LO x w f ' darksholegrey /
/ -- _-pdork brown ulty sandstone dark grey PLANE VIEW
_o - shale T. F CROWN II grey sandy siltstone lenses d458f 456.. . 2'- 454' 452.7 a z ' o 452' '- h 430
- 448 CROSS- SECTION w INVERT
*!* ? ! ? s. <lN <l*s e
no. 4 ** 4 _ ! (I nke n rk,.nh k.e ... ,n,6 L , am A4 nt A= A n ,nno #g - 4 5 rock bolts at 2' Centers in steel ribs at 3.5'to 4 o' centers, spane 3"- 7 " Moist beddmg froovre, above springline along east well m 8 ..... L.. t . L A i
.L f.
w-1238 0/4
L t E UM TE thickness. inches f O O S b E 8 E O 8 i e i FJ E s E e s
! I oW 4 I" . , ,2: E a
6.0 A LEF SPRING
/=
U -gg a &\ g T. h E CROWN }}!a -* ' ' mom kV2I - 1 _ _ * * - g' --- 3j RIGHT SPRING LINE fjL. \- ' ' ' tos g 1s(2 s) n i '4o ii2 CONCENTRATION
=
0F IRON BANDS _M pspringline CROWN ,460
. _' - 458 - 455.3 .456. , i
- 454
- 452
- 450 INVERT -448
{R EXCAVATION PROGRESS ESTlWATED ROCK alona e%re secten -6" CONDITX)N (TEPZAGHI NO.) i at 5 rock boles or 2' centers n steel ribs at TEMPORARY SUPPORT 3 5'-4 0' centers spanned by were mesh SYSTEM d' - 9
- BEDDING SPACING FRACTURE SPACING No seepose WATER CONDITION
,,7,,, u .)', D+ DEPTH OF COVER i
GEOLOGIC MAP OF TUNNEL EXCAVATIONS qq FIGURE 2 j2Cr 3 ' V > SHEET ll 0F 24
303 ORIBE zwn-e INDEX MAP a ( SYMBO Scale .1"* l2%* k HARD, TAN, BROWN, CHERTY, . MOISTURE ALONG BEDD
\ _, [1g IRON BED OR LAMINAE .'. ' ' ' PARTING \ g NY - DISCONTINUOUS IRON BED 7 g SEEPAGE FROM FEATU Sheet 11 0 _, _12.Q)*__ H ARD. LIGHT GRPY. SANDY OVEREXCAVATION OR g SHALE TO SILTSTONE BED - OVERBREAK ZONE, MAX
_.]ggXg;2)*_ DISC Tl US RON BED 4 FRACTURE METHANE GAS, MAXIM
\ PERCENT L.E.L. WITH 4-2s-77 \ g )
FAULT BEDDING PARTING . Note: OF OCCURRD4CE Number (s) in parentheses indicates be VERTICAL JOINT Number preceding parenthesis indscote
- n. INCLINED JOINT above or below spring line at station ggi g measurement.
7 BEDDING, STRIKE AND DIP j 8 9 2 2 ? 2 8 R 2 gl 8 e 2 R ? 8 yE N $ n o e o o N e N e N c N
$ N E
N E $ 5 N N N N N N N *4 N
- 3 6.0 o
z , p. m
$ _ _ _ _9g{h y______i[3 'f 21(1/C o
E k .- _C-e o --17. __ ___ { 4 s PZ__- w}de [ '- . . __ . ,. . ..- - --' , 26 :- n, p uG/2 f. e 4 @ ,
* 0 **' 7 f F 4 J W 6.0 /' " 7 0
f -b 3 -9?t . ; ik -idh A fO ~S" 5
,wW - ~ n-+ m , g- ,- - r - - '- - f,y>-
c ,
=
g
~ aw- -g 4 ,
32,
/ ,h u s- s-n / ,' 'N1 f' I y Silght warp to entire n *ss--
T C [ 460 ; CROWN : 7 7 458- ~
- - - ~~"
z 456 2 z 454 - 1 - ~ 1 ~ h 452
$ 4 50 -~
CROSS- SECTION 448 INVERT Q <n r- Slight overbreak olong center C , yjN
, e :0[
e N \ **ct.on 0 -6" deep enest :r 3 J> N ct 4
\ no. 4 4 *7 r 5'ShLertrhtt91 16"-6"overbreak en crown moest ) 6 /"overbreak en crown wer S rock bolts at 2' centers in steel r r _
d 0' cenlers soonnsdly-_wunae h 3"- 9 " 3~ C- Adda.onal rock bolts and ribs added t'_jo-to or.ghibit addit onal overbragh Bedding Iroctures only, spaced as above P Sipt drip from overbreak en crowa ' shght men % bdding %ctures IMorst en orethito 4 k is 5' e wa ter
/4 98 O'subsurf ace (shale 1238 076
1 i i. E UM
'TE ,t hic k nes s. ... .s ef 2 R: S R S Rj 2 2 # a! &! a' # a: A E $
d N N N N N, N. eq N W l i I I i 1 1 1 O d u . - _ .._.-- - .--. ---- . - . - -.s,w ,
?
y,l k. -3 _=T_=K=R= -=&js21=
? [s % ~= --E n ~~ = =3Alb 7o G , ~~~ M~ = ~
1'"Y"'"
' ' ^ ' ^
7 :_-,--- - _ ,. -- ,y , ( CR0WN
,f 3GtkU2; go y, w= _== =__i-- - - - te- go-va _ - - - - - -ggwa- _ --4 g -- -- -- RIGHT SPRING c.r- ~6 0 $ t.1N E o
w r) - - .~ . - - 4 s (s ) - .- . - . - . _ . -5xa [ O 1 CROWN ,462 _ _ _ _ _ _ _ _ _ _ _ . ,460
- 458 ~ ~
456.6 :456 454 452 INVERT - 450 R y = EXCAVATION PROGRESS ESTIMATED ROCK Intirt.Jtition m crowr CONDillON (TERZAGHI NO-) 5 rock bolts of 2' centers m steel ribs TEMPOR ARY SUPPORT ot 4' centers sponned by wve mesh SYSTEM
~
BEDDING SPACING FRACTURE SPACIN G Sl.ght me sture from ted- WATER CONDITION An9Jmuure_u1_trewn
??.5 water , m .m.'4mi.) DEPTH OF COVER GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 I
SHEET 12 0F 24 120 077
pit mo
?00FE8!ML F
l INDEX MAP-S'"" 'S
!, sme: c i2 J j
\ - k 1 \* HARD, TAN, BROWN, CHERTY, MOISTURE ALONG BED 4' _
'_ ' - - PA RTING % 2P IRUN BED OR LAMINAE She.t 12 g --
DISCONTINUOUS IRON BED p g SEEPAGE FROM FEATU ['(. # k4 _ J{,t)* _ t H ARD. LIGHT GREY, S ANDY SHALE TO SILTSTONE BED 2 OVEREXCAVATION OR OVERBNEAK Z NE, MAX g@& _ Q)(0-2)*_ DISCONTINUOUS WITH'N GREY BED IRON BED , FRACTURE A METHANE GAS, M AXIMt
- PE RC E NT L.E.L WITH FAULT .MO 2617
\ g BEDDING PAR TIN G . Note:
CF OCCURRENCE N a mt>e r (s) in parentheses indicates b
\ M / VE F(TIC AL JOIN T Q Num ber preceding porerithesis Indicate 4 ,,,' l'3 INCLINED JOIN T obeve or below spring line at station BEDDiaG, STRIKE AND dip re e d sur e men t.
I 7 2 S! E1 2' 2 8: 9 2 8 ? S! Sl R g R 8i
$j $' 4' $ - dP!
dl N dl N
$f N
4 CW d' N CW N N N N 4! ed N' N ed &
"60 I I I I I I I _1 1-- 1 I ' I y 6 3_._. _ . _ _ . _ . _ _ . _ . _ _ . _ .__ _ . _ . _ .asu.___. _ __ _ ,__ _ _ . _ -
n = 0 - -
- - - - - - . _ r_. - - :- n - _.c.,.= ,,7 3 j%m=mm% , y ,)
2 & 5 2M- - -LU2, o - 16 ~~ ~~- -l 7sh,iu- y- 7 _. en y a - . -- - p . . . 3, m b & w 6.0 _ - g - ,_7 ,_ - o + o e T 4%. e - - ( --1) _ v' 3
~ ~ 7[4 h 4 - ~_
w f g 'orh
- _ - - - - - - - - - - . --.- - w - ga _ --- -- - my- - -$)- - ,30, g 3 0 - -
gg. q - _m - ac f- to ' ' * * - - * - -' l Il-. - - . -- -- - - - - - - - -ms vn - . -. - - . E
$ grey, fine grain sandyf shole .I
_j I dark Grey shale A f' PLANE VIEW light grey shole I l gpringhne [W 462: CROWN 460 -
~
z 458 ____ z 456 - L 4553 [454f
=c > 452, _._ _ _ _ _ _ _ . _ _ _ _ _ _ . _ _ _ . _ . _ . - - . - - . ~ . -
CROSS- SECTION d 450 INVERT e . <-Ov.rbreok 6" d.ep 1.m.t.d by v.rtual p nis gl. 4 % { sr.ad ng poeall.i to tunn.1 bear.ng _ dN
\ n. 4 b 49hl_overbiecknlor 3 rock bolts at 3' centers in steel ribs at 4' centers spann.d by were mesh / '5 rod bolts at Tc.nters en steel r.bs
_oj. 4' centers sponned_by_ wire _me h 3-8' jI d - Beddirg f rcutures only, spo(ed as above Sligf.I moisture f rom all beddeg fractures [
\ ,20S woter 4 2_awbsdlacLhhale w
1238 078
P00ROPUBINAL I A i NG lWUM M AT E
$ fluckness.
inches of
- q. INTAKE RISER NO. 2 (MAPPED SEPARATELY)
O' O! o! o' O' O O. Oi O! o,i
; ol o- o o oi 0: O g.
c+\' + +i + + + +1 + ' ed. W v c' e N el, m. O' W l i I 42f1V2i J 603TT l l
' h T, O g-t i a 3(D 3(1/2) 55 T S%NG e 50)3 m nsn . 60 ' ~
LINE 1 35(W2i E )Ev4, g f, 44f O 45fM)- z
^ ^ ^ ^ ^ I. ^ < ,. , .,.,L E, $ -
CROWN sem
~ \ 5 )
- g. 35ft1 J 4)
P(2 [,5E6 -- - 8 $41_ _ _ 4 'i'o
-- - 6.0 0z - PIGHT SPRING 9(b LINE type cut ggw_ ._.. _. -.-.- --me, .----.I, , g _
Excavation by Dosco tunneling machine] l o ) , I"
- 12' above spring line ] l
] 1" s 20' below spring Hne ' PLANE VIEW B
.CROWNr C
[460
--- _ 7 - - 458
_457.8{ 456 -
- 454 = - - - . =
- 452 CROSS - SEC TION INVERT -
Re* _$ ER [ = R fa R fr EXCAVATION PROGRESS no. 4 ESTIM ATED ROCK CONDITION (TERZAGHf NO.)
=k b.se. .e 2' seatees ia steel rib. .e t. 4,5 * ,..,,,,, TEMPOPARY SUPPORT ,,,,ia d by . ire med SYSTEM 3"'I '
ib ".k' .U .' BEDDING SPACING
*e h=wees.a9.ap=*d .*.bm ,8*dd',*e,* FRACTURE SPACING N. seep se S t.sh' a'*.3 %,. fr bedding p., e;., 6. c , . W AT E R CONDITION l 20 s . DEPTH OF COVER l.1118 8* 91.2' wb..,4.<*hi['
6 FIGURE 2 SHEET 13 0F 24
P00fr0ggL, r
~s~ , , " ^
k l sw.e u lNDEX MAP SY M BOL S
, {_ Scose: i"= i2OO' \ ~. * \;~
h "
%* +
_g* HARD, TAN, BROWN, CHERT Y, IRON 8ED OR LAMINAE
, , , , MOISTURE ALONG BE PAR TING \ , \Y -N - DISCONTINUOUS lRON BED T SEEPAGE FROM FEAT g%, A- - *- ' 12
- 3) *~ ~ HARD, LIGHT GRE SANDY OV ER EXC AV ATION OR g
h
\*
g% SMALE TO SILTS
. Lop.E0.;2f_. DISCONTINUOUS lRON BED E BED h
OVER8REAK ZONE, MA AMOUNT SHOWN s{., , WITHIN GREY BED g % # FRACTURE METHANE GAS, M AXIit
* \\ PERCENT LE.L. WITH 4 as-77 w FAULT OF OCCURRENCE SEDDING PARTIN G
- Note:
Nomber(s) in parentheses Indicates M #)s VERTICAL JOINT INCLINED JOINT Number preceding potenthesis iedecot obove or below sprmg flee of statio
- l #3
[ T BEDDING, STRIK E AND DIP **0'*'"I* 8 2! 2 S S gl E gi g y5
- ~
Jm JI e e 6 e 3 n 6'
" gi e t +
6.0
. -,sm-.- . - - . - . -ua ,y .- .- i O d- f 6.0 su g - . -. , g E l 0 g -
- O 4 g
3 0 _c 6.0 ~
~ ~ - . - -
f IN E
~~~~~3 $ @ s z b b+ 6.0 * , 61 . p o m , Z o
g + w 60 62 62 -+ ( m - CROWN 4u o + o 'E 52 o x
+ r w m +
z s 35 tu) w i i 8 e 5 0 29( T3, y C - ==
- PMG z f
J
-l.'M- - - -
6.0 N
- RIGH LINE M Horseth Q
g g ;;
;-- 6.0 - .- -. - .- .- .- _._s.r.3__._ r. _ .3 0 x- w a b &nn PLANE VIEW A -lNT. RISER NO.1 (MAPPED SEP AR AT ELY ) +- _ a ' I F 462 7 CROW N h462 d 4604 462 u L 460 460'J z 4583 458 - ? 4581 z 456.7456.6 456 7-456 4 __
9 454- h454 456.3 452 -*--+-*--+-*--+-*--*-+--al 452 - w - 452- " d 450 INVERT CROSS - SECTION 450 5g
.$ #"EXCAVATION PROGRFSS ~
- n. 4 ESTIM AT ED ROCK
$hght overbreak along entire se(ten CONDIT!ON ( TERZAGHI NO.)
5 rock bolts at 2' centers m steel ribs at 4' TEMPORARY SUPPORT centers spanned by wire mesh SYSTE M d**I BEDDING S PACIN G 8.ddmg frocevr., only, spaced as above FRACTURE SPACING St.ght me.sture along all beddag froctures WATER CONDITION 91 5' sub ac shol. 9 .'2.+ DEPTH OF COVER +,fl7 [ 1238 0B0
P00R O!!IGINAl. NG ?E I MUM 1 [4 4 ATE thsc> ness. inches j!j of ' 9 f C 2 o; e o m 8, 2' 2, S, E S N
, + , , - 1 N N N N N N 2 ,1NVERT O
~~ 4t $, -----gn.nrg 2 5 3
-- ~. 6. N N3 14(3/4, 14(3/4; N N D /r$ $2 59(M/2) ,n --, ( c/ 6 c. h CROWN y 8 e f' w e!ao,,.*
14 r> 43 ( l /J Q RIGHT
= - - - - 'M, - " - 6.0 / -
R N3 f' ' ' ' 7 O h" INVERT by DOSCO
*['"f Emergency Service Water intake Riser - Mapped Separately %
I er) ochine
~
C POW N ,446
,444 ;442 , -440 ~
_ _ _ ~ 439.04 I 438
-436 INVERT - ~yR yq ~
EXCAVATION PROGRESS g 'f qr ESTIMATED ROCK L. I UV CON 0lTION (TERZAGHI NO.)
*** TEMPORA 4 f SUPPORT SYSTEM 2* 7* r.6a BE0 DING SPACING fy spaced 20' F R ACTURE SPACING i
WATER CONDITION oppe.n . 315'l=.or. 2 7.5'es.c 4 s,le DEPTH OF COVER lll'_aLahal-4 GE0 LOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 14 0F 24
PJORBRIB!NAL
,1 \' [-
IbibiX l Scale: I,,* l200 M Pl 4 [ R SY M BOL S
~
g k * -4 g ,
*~ HARD, TAN, BROWN, CHERTY, IRON SED OR LAMINAE ' ' ' " PARTING MO!STURE ALONG BED' -- DISCON TINUO US IRON BED p g SEEPAGE FROM FEAT ' -g/* _ ,3Q)",_ . _ HARD, LIGHT GREY, SANDY OVERE XC AVATIO N OH gf
- SHALE TO SILTSTONE BED OVERBRE AK 20NE' M A e -'
_1,.ql,%-j)*, _ DISCONTINUOUS IRON BED AMOUNT SHOWN
/ % WITHIN GREY BED ~g ~~
k*%~ "'
/ FRACTURE A VETHANE GAS, M AXIM \g M e.as.n FE'lCE NT L.E.L. WITH C# DCCU R RE'vCE g FAULT N
h g
/
BEDDING PARTING 4 N ue: N u m be r(s ) in F3(entheste indicates b
/ VERTICAL JOINT Num ber preceoing parenthesis indicat j ' en
_t_ ... f5 IN CLIN E D JOINT above or beMw spring line of stats BEDDING, STRIKE AND DIP "'0*'" f . . _ . _ _ _[ Shee,14 7
$ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N O 0 N -" o - N m o e e O - m e n + + + + + + + + + + + + + + +
O O O O O O O O O -, - - - - o 7 C -
" ~ ~ $ c1-? _'_ ~_YY-3-}~~.'Z ^^
o . - - - - y (....'h3/4) \ -..--. i430ps g jn: 8W 14 0)~ g f N \ 32 5hW31(1VZl- - 37 (1 ) qig F + W 6 '0 y ntoke -Tun nel-\ g e
/@* ' ,,e -" 3 g ,)
m 4 z M pped \ g ej 'j 3g g 33g3 -375(1)
+ s ' dr7;, em y0 - yeparatoir s x s .n. i35m -
g __ _ _ __ _ _ r_." 4 g - - - qp_ __ ___yg - - 1 c: 34th 31 A 9 s-w O , I"s 12' above sprwi4 line Excavatio
' (rood he 1"s20' below spring line PLANE Vf E #
Tu nneling C - CROWN w 446 - 444 _ z 442 1441 - - o440- - W 436 - -CROSS SEC TION w "lNVERT
% E h E E N I N $
5 rock belos se 2' < **** * 'a ** *l ',bs se 3' 3.5' < *eerg spaaaed by T-2 o- 2- i/r .- 3 .- r r- r i6 [ r s* g* Widely spaced 10' 20' 1* 10' W' ( ~. . .e R,e oppeem. 43I.3' et servier.n one, 2 7.5' si.cel eill I?tS' of she se i w 1238 082
h f i RT SPRIN G LINE N SYMBOLS T 5/ RING LINE _,g8_ HARD, TAN, BROWN, CHERTY, IRON BED OR L AMINAE RT __'2!1ll*-- CISCONTINUOUS IRON BED (3*' ~~ ~ HA R D, LIGH T G R E Y, S AN DY SHALE TO SILT S TON E BED _gg;g* DISCONTINUOUS IRON BED WITHIN GREY BED FRACTURS F AU LT BEDDING PARTING VERTICAL JOINT y
/l75* INCLINED JOINT BEDDING, STRIKE AND dip MOISTURE ALONG BE eDING ' ' ' " PA R TING SEEPAGE FROM FEATURE OVEREXCAVATION OR PROGRESS OVERBREAK ZONE, MAXIMUM ,. . ;, MOUNT SHOWN ROCK CONDITION (TENZAGHI NO.)
SUPPORT SYSTEM PACING
- Note:
Number (s) in parentheses ind<otes bed thickness. SPAC:NG Number precedmg parenthesis indicates inches above or below spring fine at station of DITION me asure m ent. COVER a GEOLOGIC MAP OF TUNNEL EXCAVATIONS i zj'. 083 4 , FIGURE 2 SHEET 15 0F 24
!k !_.- ,
'h 9 \
j PODRORENR xs
*o d\t ~4 V Ng(,
g M N \\
- s\' %i v %, -
g k b \
\\g ) - Sh._ ,, , :
T
+
0 OI O O O Ol O n O O O O g O n g O - a m O, e e e e 0 -
+ 4 + + + + + + + + + + + 4- + d o O O O O O O O O O O - - - - -
2 - - O 2 iyy KH o Discharge . _ _ _ . _ O 2' S O I"""'l
~~~ -----r--- * -
6.0 LEF 0 Proper - 1.M 1 w Mapped . _, . _3 gg _ _ _ . _ .50(2 ) - - S g . __ _ . _ 3,___._.. L_ w Separo y sf b + 6.0 i; had - 1 ;$m CR S lb l-;f, .__,3,3i_' - stSi 2 -- ' - 52 '-- - ~ 5 ~ ~'~'-~ iAI!1 5 6* 6.0 RIG O . mab4p
- - - _y e.ix - - _- - - 2r s ill!Ld__ _
o o -- - - - - -iis(0-i b - o j -54% -O h INV V Alternate Emergency Service water ?
$ Intake Riser-Mapped Separa te ly - ~_ E ncavation by DOSCO O e 1"sl2' above spring line
( roa d header) 1"s 20' below spring line Tunneling Machine PLANE VIEW C 452 - CROWN CROWN 452 E 450 - - -
- - - - - - - - - - - 2 450 z 448 5 44t5 z 4461 445.8 ,, - 446
_ - , e o - s u ' , p 444- ------ __ ------ _.,_ _ _ _ _ _ _ 444.9-444 4 -
> 442- - 442 w
w" 440 INVE RT CROSS SECTION INVERT - 440 2 I R 5 5 E X O AVATIC no 4 ESTIMATEI 5 rodi bolts at 2' cont 's, in steel ribs of 3.0'-15' centers, spanned by wire mesh TEMPORA o*-2" 1*-s* A P8 ' ' P " '" 9 ' BEDDING I/2" s* vogue 7* 14 " Bedding f racture. only, ipw 4 o. above FRACTURE Slight seepose from bedding porting along weit wall j Ne kilow. WATER CC f*
- A pprom n 5' t.c..m. 2 7 5' ei.< WI . I t ~Appros 3 i 91.< v.wia. 21 fee.TTiT
~~~
DEPTH Of i t i. i ' iti.i._ i i 3 i' .h.i. t l'
I N3 'E MUM i w ATE . thid ness. / of inches p h %4 2 2 C R 2 81 Bj ? 2 9 ___.2_ 2
- C I I
; N' ; w -: O , .NVERT
= = = = = =
=.=========== 'h ?g
- s. -
i% x.c E x z LINE
/ _. _ _ _ __ __ __ _.
___ OA*< . r. g t {5 - CROWN _ . _. e __. . - _ . _ _ . _ . . _ . _ _ . . _ _ . _ . . _ 1
,6.07/ .J RIGHT 5__~_ =. _Z._ _CZ i ^ ~ - ' '
E=_E-Z~_~= F:E-F -- - -
- R NG trsction . _l
'I-O bet INVERT (Mopped Separately)W "
2 1 I I
~
C RO W N
I s
_447.2 2448 ___.___.__e__ ___ _- - _- - _- _ _ _ = ====-.r=_m- , 446
;444 ;442 INVERT -440
- lR % 0 % k R h E XCAVATION PROGRESS ESTIM ATED ROCK ao 4-5 CONDITION (TERZAGHI NO.)
TEMPCRARY SUPPORT 1/2* 6* Es NG SPACING
,.g. Nov.ced sv . ne.al amount ol mo.sture FRACTURE S PACIN G
_'-a M8!'e d'Y_hl'**2'DL*L' [ f_ WATER CONDITION f APPres. 315'locuser.ae 27.q'91otioLt.16 & io9.em,1,-t DEPTH OF COVER ( GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 16 0F 24
p$yy [ dh b(DION "llif0NNLL 4ca ., 1. . - _ . e a INDEX MAP!
\ *d
- j I
Sco w: 1,.* l200 l {h b. p' SYuBOLS
=====2 E
b I
-r kg. _. '9 4, [ _ _ _ ._ ! . _ _ - S b. ~
HARD, TAN, BROWN, CHERTY, IRON BED OR LAMIN AE ' ' ' ' PAR TING WCISTURE ALONG BED
- lbA- ' DISCCNTINUOUS IRON BED t) p [ SEEPAGE FROM FE ATt k
N<, ((U*. . "'g*-----12 HARD. LICHT GREY. SANDY CVEREXC AVATION OR s# ' I $ HALE TO SILTSTONE BED 2 M M M M Z W ,MA.
'! ~ lo~ g.g-- )* DISCONTINUOUS f RON BED 4-\ Y{['
I OUN S N
'O IHllI 4 % WIT H!N GftEY BED . . . \ ll I - 'I ~~ ~ j FRACTUk METHANF GAS, M9 tM t
g . Y PE PCE NT Lf.t WIT H i
,, FAULT 4- i CF OCCUeENCE EEDDING PARTING e N?te:
j \g ,%J / VERTICAL JOINT N u m t>e r ( s ) in parenf % eses ind. cates ; GB N ure t>e r preced og parenthesis in dic o .
.g . , ,
_ _ + _ _ rl75, INCLINED JOINT oke or tielc e spring ime of state BEDDING, STRIKE AND DIP r e os c e me r) l 3 b To Alternate Emergency Service Water intone Reser w , o z o o o o o, o o o l o 'D' . o o o o of o o 3 o+ y N m o e b e e o - nl y N, c, O O
+
O
+
o
+
o
+
o
+
o o
+ +
o b,f O+ g
+ , + ~
z
-~
l~
== = - = - ==== _. ,m ig _m T - e_ .r(
e, a* Tsy - _ = = = = = = = = = _=_= === =. f, w 0 2 g t' g[ a g ._ ._.- -._.- .- - - -
~ , _ _ . _ _ . _ . _ . _ - _ _
f / _g*f _ ft %*_f__
$ "g 60 7 2 t - - --- -----
g_-___ o , o l m ',' -----_.-._.-._._._.__,.-._.-g--- . _ .__ _ _ _ _ _,_ w e o a 0 __;___________ _ _ __ ___ __ __ ___. _ _ __ _ _ _._____ _ 8 5 j - g ( -_ --- ,7_ _,_. _,_ ;___ ;;__ _, __ ___ _ so gy _ ;._f. g___ e (oTrT (o n a y _1 9
- Discharge Ce
$ N" - Escovatior. try DOSCO O I I"al2' above spr ng line Access Shaf t > (rood hooder) I"= 20' below spring line Tunneling Machin* PLANE VIEW lg ill I 'glt e'i i I
i li j C 452- CROc 'I t i ,' U 450 ~ z 448-r 446 445.8 _ _ T-L o -~-------=.=-=__:_==____.__==-=_.___.____=___._______ _
;~ 444--
I f442 CROSS SECTION d 4405 INVERT m t 01: # # $ g: @ no. 4 s ,.a w ,, . 2 . . .. ..,, .. ,,.. , r +, . , r.1,......,,,,....n,......, 2 *- 5* t/2*- 6" 1/2"- d* 2*- 6* 1* - 6* I". y* l' - S' 5'- 10' s....o. _, . ( ,,,,,,
.g ~. .. .
above spnngl .... .. on ..it .oll N. see 225' giocal till & 113.2 shot. f-
70BRORIB!NAL , ANG , RE IWUM
.. .....n.s.
s lpebes
'I grey springli - - ~8 "8 "
e.l a. :.
- . . \. e. b I.
. s, S t * * *
- g4 lt t en g j s-O d -lNVERT "d} 3) 52(lv2 -51(2); N 2 S(I
, ,,,,, , _ _ A _ g , _ ,_1 &M k P-ii5{,
60
\ O LEFT N , 12 26 26 . . .22
[
)\' w N 50 . -- $8-9( )h5 -
CROWN
'o ^ !!- aim $ \
31 " A' N C Y j RlGHT
- - - - - - - _ -_ - - - -[6.O e- -
S PRING (h31 -*. C *-7* N [hT*~ " -*d_ YO- LINE 490) SE 540,' / a-
/ O O -
INVERT
*ircular , I Escavation by . obv. 1 I"al2' _
L ~ JARVA (mole) blw. I 1"s 20' Tunneling Mochine Cut CROWN -454 452 44A1
-= =:= = =:= = a = = = = = = -j::: -444 lNVERT [442 ;R fR R{a EXCAVATION PROGRESS ESTlWATED ROOC CONDITION (TERZAGHI NO.) ,,h TEMPORARY SUPPORT SYSTEM 4"-9' 5"- I I" BEJOING SPACING mc ovat en Bedding frocevr es only, Spaced as abov, FRACTURE SPACING WATER CONDITION y A* pron. 2' of b wh + DEPTH OF COVER 82Lof_s%
GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 17 0F 24
P90R O N K
,mw , s INDEX MAP * < sf l m SYWBOLS Sc ole : I"* 1200 ~ \
E k d k %g l
,,,, _ e (If}'_ H AR D, TA N, BR O W N, CH E RTY, WOISTURE ALONG BE --]- k+ IRON BED OR LAM NAE ' i' ',,' PARTING g -- -- DISCCNTINUOUS 1RON BED SE EPAG E FROW FEAT
[. , , 12 (e)* HARD LIGHT GREY SANDY OV E R EXC AVATION OR Y2%f )., _$ (710
- D_'_ DISCONT NU S RON BE [WOUNT SHOW
_k..g . J(g .5 WITHIN GREY BED 17 g f ' FRACTURE FAULT a ' BEDDING PARTING s Note.
\ Number (s) in pa renthese s sedicates b / VERTICAL JOINT N umber precedrog paren$ests indico i . 73 y,p,., l lNCLINED JOINT above or below s pr;ng line at stati T BEDDING, STRIKE AND Clp moos sr o men t.
O O O' O O w @ N e e O
* + + +
- z O O
O W g O O O O-O N O O O A N M M T
] + + + #. + O.l + + M. T. C. ~ ~ ~ ~ ~ n n n . . ,r, g
I* ---___
&a %3,.,, _- - - - -Lqi >--, c => -- ._ _ 2 _ - g; _ _
ut, g 'A2M_-- m--------- _ 2+ 3 e + g .-. - - . _41514 W . - . _ . _. _ . 4 3g43 g- _ 479 _ , _ . l 2, 4, tt a sO y 8+ v -.-p'- . _ Q,y- p-.-413t4 _. ._. /-._ 47g3) /_. _4,o). . _ . _ _, '
' 47 D
a o 3gr-- -- --- -- --- e e fs _- _ - _ _ _ _ _ _ t3 C- g,_ _ ___.__i.
~_
j ,0
.h - - - ..i .ee ut u
{ Horseshoe Excavation by OOSCO ,b obv. 1 1"s 12' (rood hooder) bis 1 1". 20' PLANE VIEW Tunneling Machine Cut i1 l C452; 0 454;] CROWN, , , , _ ,[,,,,,_,__,_,,,_,_,_,1,_,_g z 450; z 448- 447.2 1 , g ,,p _ _ _ _._ _ _ _ _._ 2 _ _ _ _ - - _.__.-.-_._._-.-.5-_-.--._-_ _
> 444-w -----+---
d 442-INVERT CROSS - SECTION de de h de 2 !e 1: no. 4 S tod bolts or 2* centers en steel ribs se 3.0'.3.5' cen*ers. spanned by wire 8 1"- 9 " 1*- 5 " g*. g 3",9 " Appros. 5*-10' A pp,,,, s' Appron, i t' Jointing less opparent en smooth circular bore No inflows Appres. 24' loews.n. & 34' elac.el t ll 4 iu. m 1238 088
7 1 EDDING ATURE R MAX l MUM xlMLM TH CATE es bed tbckness. i C Qle s i f* C ht$ di so ft of pfiron band N _ s _ grey siltstone bond I 01 e ,I 0 !O O' O' O. O O'i
+ >l tj f e .l tl etj 0 +
O .i
+
u-t n Ol t-
+ m!
t 9 e -l e. / el e k > > > s si g . . . IC . ;_
-._.-._C 0 * -
yw-=C =w zm._ - _ f__ _ _so.2). - - -- -. -. g- -.-. =ma a . - I 6 LEFT SPRIN) q-7------.__-_-___.___.--- 6.0 gg
,2e-._-.-.__.-._
44 n 46 2 56 4 3g $. h CROWN 45 4 a
._.-._32-_
_.~_~.E l _T E' _ _~_ _ so 2 RIGHT S P R W) LINE Sir.__- 2 ffy_[_, 2
,,_ ,,,,,,,, , .,_ _ - . _ _ . _ ;,_ ., ,.,_28L4L_,,;.=_ . _ , ,,,,,,,, ,, _ ,,,,,, g . _ . _ . _ . __ _ . _ _ _,, o m,3. _._l g m - . _ . _3,c,,_ . _ .
J0 I
W*N 454 =
_ . _ _ . _ . _ - . - : 452
- , 449.9,,gg b.
__ .__ _ . _ . _ _ . , _ . . _ . _ . . _ . _ . _ . - - . . - . - - - - - - - - - * - 448 __._.__._.---.-.--.-.--.**446 L
.~.- $ 444 u INVERTL $I* 5I* .bi E X CAVATIO N PROGRESS ESTIM AT ED ROCK CONDITION (TEhlAGHI NO.)
ll '. Steel esbs at 3 0' 4 0* c.ni.rs TEMPCRARY SUPPCRT SYSTEM
.,4 7* 12" BEDDING SPACING Bedding froctures only , spoced as abo . FRACTURE SPACING f WAT ER CON DITIO N 9'watyr i s o.t 'subsve loce ( shal.)+ DEPTH OF COVER '.
GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SilEET 18 0F 24
i _ wwm , t 8
\*
1-INDEX MAPI
~ ,,
ff SYVBOLS p Scale: l s t2OO l } b+ ! H ARD, TAN, B ROWN, CHERT Y, ' ' ' ' MCISTURE s.LO N G
-k---+ --{
l {!i2
- IRON BE0 OR L AMINAE PARTING l N, .
- Ni - CISCONTINUOUS IRON BEO y [ SEEPAGE FROM F ! w% g,, gg 92 (1)* HARD LIGHT GREY SANDY 9 OV EREXCAV ATION l , y l Y( a ~ ~ - ~ ~ ' ~ ~ SHALE TO $1LTSThE BED 1 OVERBREAK ZONE D - M-J)*
H) DISCONTINUOUS IRON BED AMOUNT SHOWN
,i i Y . $,fk l j WITHIN GREY BED "" - ~ %- ~ I~ ~ ~ ~V FRACTURE A METHANE GAS, M i l t wf / FAULT W20 4 2e 77 PE RCE NT LEL i Of OCCURHENCE 4b BEDCING PARTING e Note:
i
\
j hd en
/ VERTICAL VOiN T Numoer(s) in parenthstes indigt Nurr be r precedmg pareMPet.s ir mj ___ 7l73, INCLINED JOINT obove or below spring ime of y DEDDING, STRIK E AND DIP "'0*"***'-
O O Oi O O O O O O O O' O O O O O - m . g . so P- e cm O - N . v O.
+ + +l + + + + + + . .,m_y____. . . . . .t .t .t . .e .& .t .t a l ,c3=" --o -sx3 ma-.e-e- - -.--suo mu $
j t ' w - -- - = --%%,===---24cm --- 2'
'
- lLc Cr--------------------------- - - - - - - - -
f dx v' u} f' 45 , 39 . 49 22 . . 44 _ 3, [ 3 _ _ . _- - - . - . - 2 7 45 46
$ i y 0Or - - - - - - --- -- - - - - - - - - ----- _.
{ 4 y ; 43 50 '
-43 -
45 '
'4 7J , 3' -- U'jj pi $r i i .
20 - .g3 i , - 23 -
- o. -.--._. - _ - --.. _._._ _ ._ _ __ _.____ _ __ __ __ _ __ __. _ ___ -.
2 h - _- _ _- g o_ _
!!M - - - - - - -- - - - - - - -"MP- - - - - - -P . _ . _ . _. . _ . _ . _ . _ . _ _ . __ . _ so,4 . e _ _ e _ e _ _m 3m.,,2 ). _ ._ _ _._,3 3 e .ol .I FLANE VIEW 454] '., 4 5 2 2 ' . - 450 _ 448.6 , ~ -
e j 448 ,3 _ _ _ _ _ _ _ ,__ _ _ __,,,__ _ _ _ _ _ ti 4 4 6 ---- - - - C 444 i._.__._.._._._._.-._._.-.-.-.-.-.-.-+-+---.- w VERT QO SS - S F CT [ON y, .--
~
a' _ a no 4
$ rock bolts at 2 centers en steel ribs of 15* 4 0* centers spanned by were mesh $*-11' a Bedd.ng irocevres caly, spaced os above No inflows /= # Approe.7 beech 121.9 subierface shole) s =_-
1233 090
- f. I BEDO:NG 1
ATURE >R MA X 1WUM .kiM W TH DATE se bed *Nc hress. s o 'N ecctes inches w gray . hole
. r. , . . .. ., nt - s:6qm /springlineN g Oj I'
oi O Oi o g g. ko o
- gi b
.I .l .
I
+
I ", . 1 .,, e
+ 9 - 8(ayo!3 )~_~Z .Ctrr._ f_- Soli s 20 * - - . - 41st2i vf31[o:i~2) - . - ~' . -. _--.- . _.4 .._ .._f . y - 29(3)- -. - - - - . -79(2 Ab - - --- - - 25(2 M- . M O W . _ . _ . - - - _ _ _ _ _ _ 2 _ _ __ _ _ _ } . b _ ~__ 24 g L LEFT SCH tNG e 53 , d'3 3 2) " d j
u 45= h_Q_._{l__( $ 3 _ (qcg3
~ 26 QS 3... ,
_~_ RIGHT S PR;NG to- - - u 6 0 _ _ . . - - - - - ~ ~ -
. -g(2 Vn - . _ . _ - . _. - - . - - - . _-- - - . -- - 26( 3) - - - 8 'I N E -56tli/2) - -- - - -37 3(lif 2)* - . - - . -56(n(0 D-- - cc O Sd CRCw N 456 D454 451.3E 452 P450 ----__ _____._ _ _._._.-.-.- - - -.- -3 448 ,,, y}
INVERT b 444 , EXCAV ATIO N PROGRESS ESTIM ATED ROCK __ CONDITION (TERZ AGHI NO) __ed ' / wwe mesh T E V F OR AftY SUPP($T _.____ _ _ SYSTEM BEDD N3 SPACtNG FR ACTU AE SPACING 4 WAT ER CON DIT IO N
- 7. 5
- take - _ _ . ,f li t.7'mhwetac . (shale)p CEPTH OF COVER l
b GEOLOGIC &\P OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 19 0F 24
I Jt u-a K*;en_4 DEX MAPi llN fsC e m.uoO-la._ R SV M80L 9
-==
g 3 m 4 44 4 4
~ _ .
4 k . I __ _ q% . . _-__( f~ ! I _.._
/ ge h 2i ~
H ARD, TAN, BRC AN, CHERT Y, IRCN BED CR L AMIN AE
' - - ' F A R T!N G v ois T URE ALON3 \ SL - d9.* 2[- DISCON TIN UO US t h 0N BED S E E PAG E FROM f Iv4 'D. '
12(O' HARO. LIGHT GREY SA NDY i C OVEREXCAVATION iy , W* j " * - * *-* ~ " S H A L E 10 SILTSTbNE BED ' ' OVERSREAK ZONE
* - - 'qk i i ! ~ (7X0,21'~~
to ~ DISCONTINUOUS WilHIN GREY BEu ON BED W' AMOUNT *HOWN ,
%v , - ~ */ FRACTURE /\ METHANE G A S. A , g $ PERCfNT L E L. .
y ,,, tg } g F AU LT 4 9e 77 0F CCCUPRENCE
$1 % BECDING PAR T IN G e NMe-.
A i b=) w
/ VERTICAL J0 INT '" **
N e t,'e r creceOr g s venthes.s b
.. [ _ ! _ _ _._._ _ - _ _ _ _ . . _ . .
rl'5' 1NCLINED JO!NT coove or b eio s opneg hee at (# # y BEDDING, STRIK E AND 0.P "*"***L o o o o o g- o'
*
- o s e o 8* ol ol
= o,, , , o i, e e ol ee o;
e, o W t + + t e, t' +1 + + * * *t + 5 8 s 8 8 y e e l
=; e e e ,j of = e' !
l el * [co - -- --_ _3,0 29 --sito W " A - 7x- 2 - -- - sav A "-
---hi:in)=
n z s z _= ram = = - - . mesr_ _ .-esy. s1.' = - - . = = -
-w-- r-w g____________________________,______hW $ t 3 f' 2'-2,8f - -
s_ ,( ' ' O.- m -- ,
% - sis _ s2 s 3 st . ,
5 I 0 CU
-w. ,so . o o , , , ,30 - s.
5 f
.-.-22 o-------.-----------__-_.._------. ---.---27(3)----.- -- 3my -i:Ed n, - C -.
E
] - * ._=.'.'ML*.- .
sem--
- - u .-- m-}'W- - - =- = -1.'Us'?b . -"_=R%
_ ._. _ __ _ _ _ssm _ _ _ i
-s..n_ ==Wii, =-
_ _ _s . - 3
. g c,0 . .__. _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ . . _ _ _ _ _ . . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ . _ _ _ _ _ _
uJ
~*
PLANE VIEW C 456 gCROWN w E 454 . z 452 : - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 450 4#" i.
- 5.
448 _ _. . .__ ___._._ _ . - -_,,;----_---.;---.,__'_~_"'-'-_~_. - 44s _ _ _ _ ___________- 3 ______ __ d444g INVERT Q ROpS - S E C T ION cf. d. x el' ]1~ J
- n. 4
$ FOCk belft of 2' Ceetert il Ste@l ribt Of 3'
- d' Cef stef t, %Dr*
4 ".12 " 8edding fractures only, spaced as ok. 3 No enflows
, g %,- _ _ _ . _
Ho.i'aubivefoc e ( shale 1 I 't a n n f k[ < t
j (p g l. i. . a , . ,, . A b q s (J b el A w 6 w s a *. h*= ( r I B EDDING
- ATURE R
M A XIW UM xlMUM ITH DATE tes bed thickness. IC 0ttl IRCheS 90f l0 A Of O O O O O O O O O O to P. en et O - N M g en
+ + + + + + + + + +
2 2 2 2 U U U " U U
,3 ' - - _-_ __. w.62L9 3, _ _
O W~
- 62 gjg - 6[2'- - - $ y,y. q1uu .__.__ - - - N41 __ - . - d?tyn.n - - - - - -- - - - - . - - - . - 3om - - - f, 6 LEFT SPRING - - - - - - - - - - - - ~ ~ - - - 6.0 LINE g .n _pi{_ ___ Z _"..~._- ' ' -' 3 o)-' -^ '
M
- , 3, , T. ho CROWN 3(5) - - - - 36(6.(76 ._ -. - 23 5(3)- - - -- J ---- - - - - - - - - - - - - - - 6'0 if2 RIGHT SPRING LINL =3iW s. ' 2 gy,2 - st>" -- - - - - - -amu - - -
3 8
- ._ m.s ik - _ _ .- = sF#2 =- a = , 8 FRACTURE ZONE C_ ROW N - 458 ,_ _._ __._.__. --.-.-- - _.___ _._.__ _ __.._ ;456 y 452.7 454 -~ - 452
_ . _ _ . ' _ _ . _ . . . , _ . _ _ _ . _:450 _ . _ _ _ . - . - - - . - - - . - . - -448 (NVERT - 446 S EXCAVATION PROGRESS ESTIMATED ROCK CONDITION (TERZAGHI NO,) y ..re mesh TEMPORARY SUPPORT
- SYSTEM 4*., a 3".,* BECDING SPACING Sagece N' n , ; ledding frocom only, spaced as above FRACTURE SPACING d.ng poings ,n WATER CONDITION %% ,n*. No .nflows f io, y a,, , , ,[j,[ + DEPTH OF CCNER f GEOLOGIC MAP OF TUNNEL EXCAVATIONS , FIGURE 2 j } } ,) SHEET 20 0F 24
-paw ~ ,' -INDEX MAP f-(R SYMBOLS
- g. - Scse: e, .iz00,
~ a ~
k 4 l MOtSTURE ALONG
$y ' - * /g -L/2 H ARD, TAN, B ROWN, CHERT Y, ', ' - - PARTING \ N IRON BED OR LAMINAE
__J1IO .7L* DISCONTINUOUS IRON BED [ SEEPAGE FROM wV. 12(1)* HARD, LIGHT GREY SA NDY OVEREXCAVATiON gr I **, SHALE TO SILTS E BED OVERBREAK ZON h, ss ,
% -(7V0;7)'-
to - - - DISCCNTINUOUS WITHIN GREY BED IRON BED ' AMOU N T SHOWN
* " / METHANE GAS, FRACTURE g Sirw m PE RCE NT L E.L.
NN FAULT 4 7 OF OCCURRENCE BEDDING PARTING Note; s k
%Mw VERTICAL JOINT "*"' '" '*"**'"N Number preceding parenthesis i 75 ' INCLINED JOINT above or below spring hne at
[ p BEDDING, STRIKE AND DIP **G"'**"'- o o o o o o o o o o o o o o o o W 5 + 5 Y Y Z O o o o o ~ O_ - - o_ o_ - o_ O_ _- _
~_ _-
o _ _l
?" _. _7. e_- %n .2 3 n - - 4M& == = = ="r -* "- - -mu=r rswrc -- 4 E
g m _ _ . _ ; yguo. yo_ . _ _ . _,, 3 .. =
.____g__._ __;--gg._. _ _gg_ _ _gt,% __
O 16 15 16 N 16 3
.p. 7 .
g +pg yg 2e 7s , g 48 46 , 48 48 . v- 4 r-.' 5 b Y w u 6D ~
-53 o 4 i , 37
( ' 36 Y$
,( ,' 33 - 20 38 i 'p a
t _'8_ d# o _ O ___--_ _ _ _ _ _ _ _ _- _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ E '*E'2'= = = =1"" = = = '"
$ _r. _gMS _. _- --Esm - _ _ - -
g w,.im _ _ . _ :3) _ . _ . - isi_-. - E _:--
- 6D 5
A e n. PLANE VIEW C 458 - CROVrN w - w 456 - - 2 454 -
, 4sa - 4si.s _ ; _ q -
w - _ . _ . . _ . _ . . - - . - - . _ . _ . =-s->=~'~'=-**=*== d 446
- INVERT CMSS - S E CTION 52 $R no 4 o'.4"of overbreoh spelling from crow
$ rock bolts et 7' c enters en steal ribs at 4 O'. 4.S' cer,ters, spon.ed 3* .10 '
Bedding froctores only, spaced as above No enf'iow, 5.epag+ from i '*
; 3 =aier f
i i i r ~b ~r t.. ,%ie 1238 094
v.s t.- f EDD;NG ' ATURE R M A X !M DM x lM oM TH CATE es bed t hic k nes s '.- tcotes inches otson of N 9+ C e 2+ 2
+ ?.
2 e
+
8+ e Rl
+
2+ 2+ 8,! e _ e_ e_ w_ _ e_lj _e e_ n_ 2r- - . aTC ' ~--- - . -- m4% . -- . _-- -- - . 5 r ar . - ,
- -29 (2.53 - - - -.- - _ - . -. 24(2)_.- E C ----l_._ h,,_3 - .Z - .._. _.-._y 3) . _.._.. _ _. _g h LEFT PRING O
- 51. - . . - 29 (4 s H.- - -- - -- - -- - - - - . - - - - - 2953)-- - U
.se ,4' . CROWN 88 ' . , ,.,,,W ' % . _ . 3i3g 3 ) - - - - - - . . 3 i(4) - . - . _ . - - . _30t 3 3. _ ;
6'0 R!GHT S PRING
- .- . .-.-_.---W -- u == =:r = . = = : 2 LINE g_- _-r9( r.25)- . - . - . - . - - -.T. q7153: ,7- _ --- - - _. . _ T.2 -
_ __ _ _- . 3 x 3 ) . - - . - -- - . -55 (41- - _ - . - . - -- 5 e - O bI f/ dLT GEOMETRY arent displacement-12" to 25" d.) Dip a p p ro x . 16. 58 uci vertical displacement-8 l/2"to ll f/2" e.) Width of broken rock (1/2" to 24") Ike - N 66* E 0 - 456 4
,_._.____.._.._._....----E--.54.5- - "454 452 ._._,._._._._._.---------.-.450 -
L 448 INVERT
= .
- EXCAVATIO N PROGRESS o % e n4 ESTIM ATED ROCK CONDITION (TERZAGHI NO.)
5 rock bolts at 2 centers m steel ribs ne 3S-4 censeri, spanned by wire mesh hsYEM 3 9- P.ED din G SPAC;NG 6eddiao fractures only, spaced os above FRACTURE SPAC!NG Id'ae. wAT ER CONDIT1ON each 34.5' w e' CEPTH OF COVER I W4uainathal:L+._ f GEOLOGIC MAP OF TUNNEL EXCAVATIONS 12h,095 FIGURE 2 SHEET 21 0F 24
n n s w) PD0il DRIGINAL
\ ~II W
[ I lINDEX MAPi { Scale: I
- 1200, i
(f# Q
- SYMBOLS (w- $} f
[ _ {tg'_ H ARD, TAN, O RO*C., CHER T Y, lRON BED F. L A MI N AE
, _ _ _ MO!S TU NE ALONG PARTINo Se p 2 - cisr% flNUOUS IRON BED g SEEPAGE FROM F \ j i %#g/ 1* ! _ . _ . h .~e_ . _ HARD. LIGHT GRE Y SA NDY SHALE TO SILTSTbtvE BED OVEREXCAVATION 2' OVER8AEAM ZONE i s $^., q@, ,
_J0WLO;2.L,_ DISCONTINUOUS IRON BED ' ' . AMOUN/ SHOWN wlTHIN GREY BED 4
.p _ _ .-.1 _ _4 .. A. _
i s PE RCE NT L.E.L.
^l s #
e2s47 e I
% py FAULT (with sense of motion) OF OCCURRENCE 'a BEDDING PARTING e Ncte: \ DM(5 / VERTICAL JOINT N"*DI'I '" Ph8 '"d'9 Number preceding parenthesis er - . _ ___ f3, !NCLINED JOINT obove or takaw spring line at BEDDING, STRIKE ANO dip mea s ur e me n t.
l _ _ - _ - _[ 7 O o o o o o o e o o o o; o o, o o u 7 ! t t t t 0 ! I t t t t ? E u u 5 5 m e e e t c c c{ c S c I o '- _ __,u,>D
= ==.21 L-L ._ . _
- .=-2,,-- - - .qq.g ~ anosu -_. .w. 1s =
o'"
=,
e E -. . - - _ - .__ .__ - --- -- c, fE ..mo
%. i. a i- _-- - .q Q ~ 3- 5 2?. _ ..__ _ f).
Z sn n __ __ J ' ~ $ Sh r.'r .. _~1.E.==, =. + w _ m - ,} ~~_.~, Q [_. f
+ g t=.. 6 5m =2x5m ~2ns -7 - -n$,.s+ _ , _a,g. - -mo>- . - . -- ~ r_iir u o o e . ,,s s1 s t + w 6.0 w ,- , ,
t 6-- m y r- ~ _ Q- -- me v-0 -
-y!&- .=:= py ._ = .:.=,);flf i -UW :~ -3* S('W - --f~p? . __ g, p;_ ._ . _ . _
m373 r r-M __4~),
~
3 0 0 Q 3 p---- '" C. _~_ ! E._ .- 2u5'.'*__-- _ _.,_.m m _. = q l u - 55 5 _ . ,_ _3- s.ga,g.55. : .ge n - . __hpur %ip.s . _W ?= ==* # - ~ ~ ~ 6M
~ ~ ~' 42!75; eHol 6.0 " FAULT io 2>rs ___ ,)g PLANE VIEW b.) A l l ll _ h 1, 'f.t6t1 '
w 4 56 qh . t _ , , z 454 5 45f 7 - q____ -- l j~- l '
. + - - __
pA - ---- M --^ =--- - c --- - 2 _ j
~^~ -~~~
448 - . - - - . - - - . - - - -
~
U CROSS-SECTION w c
..RT o E o E o n.. . n.a .s 5 rocli bolts at 2' centers in steel ribs at 3.5' - ( cen'ers, spanned by were mesh Full sheild.ng emp 3* Ribs at 2' centers) 8edding fractureg spaced as above, joints widely spaced a'*ture Ir c ' P
- s*ce'N"n 3 -9 at S - 1.5' oport. other..ie bedding erol d p I o Moest bedding Irocturu .here endgated o{ lt '
ng i p n lroclured creas _ , s
, . 3 .o.er 109 3'subsur ocg sha e 1238 096
in ECCAG ll AT.,R E R PJ A X f W UW X )VDM TH CATE es bed 'b a .. s o'ckress ,. anon of
\
i , C S R! 2 8 8 e 2' ? 8 e i E i i $ $ $ $ E E
. - - note _ . _ . _. . _ o 5 _ . __ _._ga t a _
_ _ . _ . ~ 63p) o
.L- dh'[_ - ' -_ 21_-- -l._-i-Z b'lI}' -P-Fi~5 d[d 60 -
g LEFT SFR tNG LINE
- . -26 30) . - 26612)-- - - -n5(3 in_ . . - - -2 5/N d 50 so m7 - - 4 52 h " " ' ' E' to CROWN 33 55 52
- A 8
- -27(4a/4 w - - --27.50 inh . - . 27: 3 H - - . - - - gl * =_ =,$f; _. _ d@%=- _ ._- Qif,g_.= = _ _ =- " *"b N E E
{-._ _ $g ( 4 ) - -
. -5 7 - ~ $9, $ [ 4 p . - . - . . _ . . , . . . , . , ,1 460
_ . _ . . _ . _ . _ . _ . _ _ 458
-._._.-.-._..4_E 455 456 r
454
- 452
_ . _ . _ . _ _ _._.- _. . - - - ~ ~
- 450 INVERT 448 R
{R EXCAVATIO N PROGRESS a4 ESTIM ATED ROCK CON DITION ( TERZ AGH1 NO.) by wir, mesh T E'M PO R AR Y SUPPORT SYST EM 3 " .10 " BEDDtNG SPA 0 LNG Mr*mol from 5"Po9'ortone bedding p g FRACTURE SPACING i Mo.swre from bedd ng parting J WATE R CON DITION
'55'*o4' DEPTH OF COVER 99 6'ev osur19sebb910 4 q 123 f 097 GEOLCJIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 22 0F 24
?00lF M INAL ,
t w"
- _ _ _ _1. . ..J_
E flWDEX MAP; $
# SYMBOLS lScoie : I"s l2 OO' C -
S Q , j
= \ l- R (
l + f' M0iSTURE ALONG _.__l g ._1._ _ %q ' i - e h2)e~ H A RD, TAN, IRON BED B RC*N, OR LAMIN ChER AE T Y, - - - - PA R T.N G j % h 22 N _ _ ,y(o.2)* DISCONTINUCUS IRCJ BED g SEEPAGE FROM F t% i - -12 ge-- HARD. UGHT GREY S A NOY OV ER EX CA VATION
^49, SHALE TO SILTSTbNE BED l OVER8REAK ZONE %p. , ; ,o g.g)--
- DISCONTINUOUS IRON BED "3 AMOUNT SHOWN l
6 -. ..A_ WITHIN GREY BED
- y, i sN l } PE RCE NT L E.L. '
l FAULT (with sense of motion) e.n.p r UF GCCL WENCE s BEDDING PARTING , Nota:
'- Mga VERTICAL JOINT N""" '" E 0 " ""' "" D N w t'e r p re ce d +g pa ree'he.'"is e INCLINED JOINT a ___p_
_/
/)7S' obove or below sprmg hee at meos ur e me nt.
[ BEDDING, STRIK E AND dip o o o o o o o o o ol o o o N o M o o W **
+
N M T C W
+
P'
+
e e
+
o -
+ + e-g o +
z o. + + + + + + .
] f f f f f f f f f f f f I f $ f 6.0 g _ Qg :- . _ . _,, _ ,, 60 (4)- - - - 60I11 .,._ _._. . _ __ __- ____ - .- -- - _-- g4g __ - -
m _ _23. ) _. _ __2 _ = - 20 m - -- - i &--- ~- ~'" L~~--g)_ T-C "^^ b ) "~ -- ~~-- ~~A~ ~- "'" Z g
- C 0 "'"~ ' - - - - - ~-' -
U ' "~" ~MI ~ ' - ~ ' ~ ~ ' --- 1 b 1 - -2 70W6 -
* + 2 o - - 28.5(4 H.- - .- M31/2) - - -24 (4) - -- --- - * * - -2 ?(3) * ^ ^ ^ ^ '- T% - - , . ' " - " ^ > , ,
6' q w
' "- -d y 6.O [' ^
t 53 6-
- S4 ' ' ~ Si 56 56 h 4 p $ +-- - 30(5 >---- - - -30(4W - - -30. 5f$)- - -
-30( 3)- - - -3C(llD)- --
- m u - O O
8L2; ' " o 9(2) E a
-8 . 5( 3 ) - - - - 8 (4 ) - -
- =/r- : : . -- - 25(0- l ::: C -- 25(1)- -
- - 6( 2 ) - -. . -. ~ ~ - - . _. . - - 2 Oi T, __..._2m3._.._._". _
h 6O"M20W 7 b F I 1- - M - - - - F"-'~--- 5 A e e is
'-' -- --- - NhA- A e is48 PLANE VIEW CROWN w 'w 460 l - " ' ' - - ' '
g 458--,,___.__,_._._. _.. . - .-. .- - - . -. _._.__. z 454 - o
,._..__._..._s._..-.-..__.--.._._._q.___.
454.1 - ,. G 452 :
> 450 -
U cpOSS - S E C TION w 448 INVERT Ol'e 2le al d { _f l* Slight overbrook olces entere escovation (O'-o')
$ rock bolts at 2' centers in steel ribs at 4D'.4 5' centers sponne 3 .,= 4a.e.
Bedding froctures only, spaced as above Mo.oure from 1 . Meesture from bedding par ting Mo bedding . par ting j e e'*t,ure from bedding / g 4.s. tai 9 - L
.o. -Lm a ( An tal e
t 1238 098
?
I S EDD:NG 0 ATURE .R WA x lMUM . x. i P.8,. M TH CATE es bed thc k ness. % - cotes inches iQtion Of
\
O O' O O O o o o o 0 (s
+
N
+
M
+ v.
en
. e.
p 80
+ m.
O
+ g f f f f fI f f f f b - -- - . _ - 36( y;__. . _ . _ . _ _ _ . _ . - 5s ( s ) -- - - -
oW 2 o
~ . ~ i~P~._ . _ '~'.~._._'.~'.~'.~-
- 42) -- -
LEFT SPR W ~ _. Z _~ 72P3)~-
~ ----7(3I-------~~-'"~~-'~ -
w L! N E o 50 z
,55 g ~ '. ". . , , Q CRO*H so - - -9'. . . 4:~ . . . _ f. 2 o
l
- - -14D ip) . _ . _ . __ . . -. -8 41)- -- - - ~ gf RIGHT S P hiNG uNL
_._ - i2 m - . _ . _.. _. _._ . _ -2r3.,n s_._.. _ .._.. _ g
- 8. 5< 4 > _. _ e,i 1
. -.- - --. - . - a_2(7 V2H - - -- -- - - 6 2 ( M - - . - - - . - - - - 0 0
d
. - -r r,- _s 2WN -._-_____.__._._._._.__.-._.-456.8f 458
__._._._._Y_._~._._._._. _--_-__._._._._.__.__.__._._.--_-.-456 C 454 [- 452
- 450 INVERT -
9 ll> EXCAVATION PROGRESS
. 6. R.
ESTiM ATED ROCK CONDITION ( TERZ AGHI NO.) by were mesh s TE VPOR ARY S'JPPORT Sy3n y 4*.i;* BEDDING SPACING fraowres only, apoed as obo. FRACTURE S PACING
' I q, Mo srure from beddng portMs WATLR CONDITION ' 8 # * ** * '
ft obey.e.ee M .I.)t DEPTH OF C3 LR { GEOLOGIC MAP OF TUNNEL EXCAVATIONS FIGURE 2 SHEET 23 0F 24
P00R E BINkL pm a -_ a i , L !!NDEX MAP, 3_ I SyuROLS ( Q ,, - lSc$e: I = 1200,j \ p y,
~ - g i 4
g O i I (! r + ; h ' ', M OIS T U R E ALCNG _.. k _ _ _ !_..__ _ _ _ _ ' . .- _ o ha- b A R D, '.IRON A N, BED O ROOR WLAMIN N, CHER AE T Y, ' ' NRTM
--.O9.*1' 7 g" SEEPAGE FRCW 7
N , DISCONTINUCUS IRON BED 4 H ArtD. LiGH T GRE". SANCY t% i ga'-"- OV E R E A CAVATION j %< '6 SHALE TO SILT SlCA E BED 3 C V E R 8 R E.A K ACHE Qi. (&'m
--(7XC IC - ,2)*-- D?SCONTINUOUS IRON BED WITHiN GREY BED AWOUNT SHOWN *~ - *' t - / METHANE GAS, A , FHkCTURE h - ),/ 10 Pf ROFNT L E L.
FA ULT 4 24.n or OCCURNtNCE h w [ i i B E D DIN G PART NC e N0 fe;
\ ! Ys-# .
VERTICAL JOINT " ' d 'I * # ' " * "," 'N k u rn t e r pceced.nq pare ttes<s v INCLINED JO!NT cteve or bem= scring ime of _4 . . . - ____. _.
) 75' U** .Y ! BEDDING, STRIKE AND DIP N 38 Vf 8 * *
- 4 I Y o o 0
- o. of
.i o ~ .o o . ol o
o o
~ = o,
- o. . ol s
o'
.I ol a, + + + + + + + + + +l * +
- 5 s
+j s
s_
+ = = e_ l . =l E; E a _ s_ l _ _ _ - .l
_, =l
"- -l a
r 6.0 k - - -61.3 - - - - -62CL--- - - - -63131- - - -e4f2I- - - - -o M41- ~
~ - 3 ' 2' h'""-
I 2 - - - 4 On - -- o . _ _1 M4 ) - . - -
="[;!;g== = %L ~~2 h t= = = = m E2 - - -- i. h)t y = ~=v y 0. - L ,,=== = -=== '3 -,,,, ._~ - +g l . _ . -7<g3 :17). - - 2M - - -2 2(3)-- ' 3 8 l2I -- ' - II 3 } ~ ~ ~" ~~ ~ ' ~ I ?I 3I~ '
v) 4 x a 56 ,58 v $f 54 f 6 .O ---
- - . .- c ,--; -v - - - - - -
o p u Z 3 Se - So So ' 50 ' cc -. t h ~26.Sf3 If2) - -25(1-if 2)-- - - - 2 4 - - - ~ 22( 3) - - -17 50)- - - - - ---2001/2) cc t o g c.) w m -_- _mm 84 ))- -
, iu )y- - - - ---9 i,01/2>w (33 --- - - -_ . _.h=_=_=-n r2)-- -
21.50 irr - -- 2pe -.- id a)
==, -~m =__=-__=w ).
am% - . -- . -- wm ew
- - - - - 10 <t u w _ . _ l,g(3)_ . _ . _ . - 60(3 )-- -- - - - - - -.6(2 lM - - - 6F 31_-- -. _ --.- 6W - - ~ 58(1) - -
y io.i2-PLANE VIEW k ' C - C R O *' N _ __ ._ ._ _ _ __ . _ . '. .- s s w 460 458h ',.,_'_'_I_._._..-.---------.---._.-._..-.-----
-' 4 5 6 - 455.4 ___ _ ___ e __ __,_,_______,__,__L_,____, _
g454. -
-- - ; _ . _ _ . _ ,_ ;-- ,; , _ . n _ - . . . __ __.__
E452 C 4 d,n3_.__._...--.-.-.-.-.-.--.----------------- d lNVERT C RO SS -- S Egg e
.R ?"
o no. 4 5 rock bolts of 2' Centers en aseel tibs of 40'-4.S' Centers, spann i Mo,sture from bead ng portmgs Bedd Mo.stvre froen beddmg portmos
,s. s. _ .
dN&MrIetf.I.db -
\DD 12SB
P00RORIBINAL t Nc w R!NG L!NE EM10_g}
'R :NG LlN E -- -6 ('M* H A R D, T AN, b PO W N CH E R TY, RON BED OR l AMIN AE DISCO!4 T IN UOU S IRON BED
_. .l? M__. H A RD, LIGH T GREY, SAN DY SHALE TO SILT STC N E BED 8/D A, D:SCONT!NUGUS IRON BED WITHIN GREY BED FRACTURE
, FAULT BEDDING PARTING
[ VERTICAL JOINT f INCLINED JOINT li BEDDING, STRIKE AND DlP
- M3;sTURE ALONG BEDD NG ' ' ' " PARTING y [ SE E P. IGE FROM FEATURE " OVE R ExC AVA TION OR GRESS - OVE R BG E AK ZONE, MAXIVUM k- '
AMOUNT SHOWN CONDITION ( T ERZ AGHI NO) PORT SYST E M Note; fdamber(s) in parentheses in6 cates bed thickness. G Number preceding potenthesis m dica tes inches cDove or beiow spring hee of st a tion of 4 m sasu r e rr-e nt. ' GEOLOGIC MAP OF TUNNEL EXCAVATIONS 1 24 $101 v1cUaE 2 SHEET 24 0F 24
rm~ m - I _s INDEX MAP % D' ' t = = . - - __ 7 m
; Scale: 1"s 12OO, 2 l l j . -
s ( %w \ Sb"'24 yh 1($ m%s w%, ,. NN v s %
% M '%w-4 -ww 7 i / I 8 e 2 g 8 g g 8 9 2 w
z O
+
O O O
?.
O O O
. e.
O O
. R.
O
. R. .
N N N N N N N N N N N N N N N e, 6.0 - - - 6c(ab- - - -- 3 6 (6)- - - - - - 5 9 ( s i - - - - -5 7.s tsi -- - -- - - - - - T6 W 0 z 510.SX 2 [ _c TJPMr___ .m.2M2f_ _.- c_-2n4* - - - - - mm
-m e p r o _ ,, __
e- __ _ _
- - - - s ou 'un ini 2uone 60 m
w LEFT
- 4 + >
% ?w __.so..,--._.____ - -.---===->um.., <o- m - - - - - -
55,5 -i _. _ _ _ _ _ ,., sc s dito-o y
<a -+--+ x 3 b t y 6o . . ~.7 p -e . ,iT'-' _ -- 3pr. E $ CHOWN --- Slo-ink 2'- 'D! -N 3ES(75i-Nh-- -
6l N T o C C .21 = ' ~ _ _ E7_H._ = = -- ; = -(232) - - '-M-~~ ~ - 60 RIGHT
- ~- _-._.j_3$
S
] C ~~ h'A.ini C CZ${il5 ~ C - -24(s>- -. - - . aso im . - . - 29r,u. - 55(4 3r - -- - 5s(6) _ - = =: =57M u= : =_5gpy_____;_;r _ _- 3 ij
_ y g () - - -58(6)- - 0 I s cc w PLANE VIE W C - CHowN w 462 . . _. -s. . . . . . . . . . . CRChN [ 462 460 _________.___._.___._.-.-.---- h460
- 458- _ _ _ _ _ _ _ _ _ _ . _ _ _ _ r 457 5 C 458
_ _ - - - - - _ - - - - - -i_
~ ~ -- ~ - - L '
O 4561 457.8 ._ - _ _ _ . _y456
< 454- 454 w 4sg-_ _ _ . _ _ . _ _ _ _ _ . - . _ . - . _ - _ - . - - - - - - - - . _ - - _45g d -INVERT INVE RT T CROSS SECTION E .
h h
- EXCAVATION PR no 4 LLght overbrenk along ent,re excavor.on (0"-3") E STIM ATED ROC 3 rock bolts at 2 centers, in steel ribs at d'- 4.5' centers, spanned by wire mesh T E M POR A RY SU
*-3 BEDDING SPACIN 6eddeng partings only, spaced as above FRACTURE SPAC Shght moisture olong bedding portings, where indicated WATER CONDITIC + 5.2 vb gse f shole) 93.5'subsur 5c (h e)+ DEPTH OF COVE 1 ; 1 1238 102
O O O P00R02BK 1 i O i Xo N
~
5 % ,y g .. s ', k
/
4 p' . (+ ' E' m!!r'!M 4/'-L ,'X
\yajQ""""hl.A ' O,(f ,. a ".) f n D.. < 4 N a,aa, p < -) fmh'" )h* ;p&
t *. nl h
** '< '*>2 D '
M u ' e lf SERVICE W{ ,
- ,' n .d W 4 ',k , ^ M.~Ik* y . 5$Y '$:,C w(/q.,
- c g O y,f ;;\y ;
*Q'ggi:w Oh Qi 'gu a m^ -. y' ,8 x.. '8 ai 3'h',T%,,jy;* % 7fQg$ ,e /n A9e:q$Q g
sa 4, n&v 4 we jg \ ** *
- m #, i +
j' y ' O 'u ' }$5g's i. p ' kth) \ !
%k 'I .! , , Jt '
u { s
, \ M, "T .
q 9.. '
'p\k#7 sdIO s _
% _gg@ ,e m\ ... s Q ,\ %%, '.,',,h # //i 4
\ ' / 'N ~
h h' 3 \\ s .' _/TI }- % \\ g" N
, (%A - 'i i 1
TUNNELING P N
P00R OPJBINAl. 8 8 8 8 8 g STR TURES 1 : = = = = TRUCTURE ( '\ N 78 3,000 g
\ \
v 'o-LAKC_ Crit N 782,000 'kr< y SE"E""sAr"r
-__2 '
Ey?OuT'" SCALE IN FEET [ N 781,000
$ #b }s g -
- 1. .
#$ 0 / # :.' %
N 780,000 h ' - , r?. k N ,,8.- gj,g3 ! f. j,4 sV., 1 s . 4 0
) ,[ -{ j $ k $ c.?,
s
?g:~ ; \ l*W.
l
%** 4y) % , '... T >
N 778,000 _
,, , 3 o, ; ', - l Q %J+,3 f j
( 9 Amtm4 , h, "N ,
.M" . W h.
l 1238 104
%, Kingsyt'fe o h ok A thtobula K8IIM9'III' o
_3 ( 1 osaybroot 1 X
~
h 2m :, Pier
, w sat eJEFFERSON LEGEND 1 %p 0'" 'o POTTSVILLE 8 ALLEGHENY (COAL, S.S, SH, L.S.)
og R1:n PE N NSYLVANI A N '-
/ . ' WAVERLY 8 M AX VI L LE - '
i- ; ( S H , S.S., L.S.) g ? MISSISSIPPI er
~
i S 0.LENTANGY Q OHIO (SHALES) DEVON I A N O. o Orwell o Colebroo 0 1 3 . eld OA o Bloo field ' to ia Kinsm , G , j e. o Bristolville f o Mecco
\ I 12($105 y "
a i fW f" jr e *
- - -- BFDROCK GEOLOGIC MAP OF v I J
NORTilEASTERN 01110 FIGURE 4
.7 l
PLANT SITE _ g .-c; 4,
~r r .
y h Mentor-on-the L 0 -=
% r, Will ghby
- g
*g, r , .p:j Brotench ] c.- >b l
v Huntsburg O
, esterl i I
CLEVEL D:,: j
- h. ( -
y a3-
~ , y r -'
[.) ( , _O Y g
&ex
(,t. 's e..g. _ n ... (.m. N e.4 , e i,- t,- Xy N. C:) h 1 1, l -i Bron..ica 1 yg g 3 ,. ( O Dor roweville Both O Brody LO y .- 8 0 8 16 AFTER- BOWNOCKER 1965
i 5 $ , 5 e
!e ?
g i 8 s a e i! - 2 7 , y M V 8
~
C C i n e i 8 y %i-O i o i g G & 7 E ' I i 8! ! "a ,
! ! : ! ! E ! ? s 5!l i r =;5 !.
e s s : e
- sb o "
o e g a s ? I 5 g s o W W d so a e 9 m E
$ a s < x o w = = -8 % } e-n 5 x *i INI .[ , / .
u- 3 L(
/p( bl % ;.s=
h P0BilOfilBliM
~ ,. 1. . , .. < $;. X s }.9=5 - ]l s t p3 4 &Is j6 >%
r I 3 i *.. i e 8 -. n .a
/ s , ,3 . s l} 4! . ;',; df pF ,
j i .s i s - FC j s'
! 3 4 g' 5 ,
t ?-- i ; o W e)-' 3 g . . i
> ig N
Jl
. l. '\ \ _
GLACIAL MAP OF NORTHEASTERN OHIO FIGURE 5 3 % in7
?DBR OR'i@SN j y , ~ T ' ' l,,, ,, J , :..'.,0$ W # ., ) ,
[ Y ~~~ ~ d({, c = =t a u r - p v, [f] -. ' . . . ._
/y r-pVc- ..
in .. .. n
,y $k;/ _),,g* .g / .. ~7 f, / /. $_ / [p/m":y. ':' : ; - /<e. /]..):$ ;- ./ h :e; - .'-
s
/ . /
/ .
,se=rvv. _,__ . ,/ a, e -
savanok - gu f s !s;,cnieto -
- 7. ,
/ g[ow oc ,' , - 9s \ >
7 . ./ f' ,
$s . 3y .. ..,, _. . . ..: .,,,,<. \\ .. . . .
(< e @#,- i.. 7 m ua \ - . ry i ,
,j i -
u..
-(.);os )
A. . , ,, -
'v f ; u,,- .. w .
_r.~ - y . . ' ' ' . .~.' i ' ~ ..;11**/e,_.. -
\. . . .t 2 s :J. t; . e:.r.iJi.ii :-
m/l i y l .. I. / i
.. ./
l +- ._,-_)7" 9 ;', ,
-- , - f .: - L y A -- -
_ L"u "' A S - !
, / \ / p/ / ' ,/
t u-
! /.\ y /. /c j '.t I. /
s' ( i i
/ / . /. j / l 3.pA - - k?p' . '. .: c , m c ,L(\,j/ */pi r y ) x - - , I - / Isuco ' 'I / /
m, ..i.
.. j / //-
i .
-t-?lISO i / / /,//
1.,uec. -
,- l
_m
/ // i m, m / t.../ 1.m _,m .
s STRUCTURAL CONTOUR MAP - g TOP OF BIG LIME 12 c 108 FIGURE 6
LEGEND LOCATION OF WELLS EVALUATED f FOR CONTOUR DATA
- i. CONTOUR LINE
-~~ CONTOUR IN AREA OF INFERRED DATA b
- 9. \
/ / /
PERRY PLANT SITE / s / .
\....
y / . L / i CllY 0."
, r'AINES VIL
- i
/ /
D ... , o,' ' 1 1/ J'
,i.. ' . 's /l uar>my_[_, .lLLE . >PE *st / / / / ..:- .F- ]c mc sl( /
1
/ /. / /
( /
/ . . g..#
s vt a n
/ 'h / /l / /
_pe'/ \- y! / xyl Aleyl- eGK. ._ 7'// /
/ /. ? 9 : / , / / / p o A l/ ce + .J I / / / / ei # 2 l .2 augn l I I ,,clas j / 3/gtat; t
(/ e A_/ he, __ _]i {p'/1
~~tc L .J (q \\
I
.I I / / n:P. f t /
i I k
/pA/idGA I / \ / / / // )
j
/
I
) /
(
\
j' / :
/i uArieto /
cA_s n/ / wurg/._ _/ ict adow _f_ . in
- suas t
4 0 4 8 SCALE IN MILES 1238 109
k s'jf., . A $ f )'
, ..c any cir c/ /- s n t . u_. .: '_
y.$? l/ f
/y%$u G \
g d,,vt -- ) . . . . . (.T2050 ' j'
. " ' * *
- v 1 L .- m - , , . -:
'/ ' ,, s " J 48 .-
~
g s_,,. / )./' / ,,q/ 220
' 9'pfa. +mo** ,s~- ' \ /.h.,. ~j , ,.
- bo*l .'-
j~i'":
/ /'z.t ' ,s- ,-U"'~ ,
Y}f'
;?L "f .\ g$ ~} l , ) ~~~ '
d* r,. -
..in ge. ricto !. m . a ,,r so= .-
ee .p50n
.io ~ . 3 ._ _ ,_ *2 r.. .
l
/l l . . . ..
I l A. $ if ' T/ . 6 ',.l/ - l, (,95 .
, I ',. .; U . , ; gq600 f ,'
l 19 ' ,_= _ -- _dik"-,- h'". ~ ' '
' --l ' * * * ' ' - f hV'"-# s s 15
[k y/
/@1 7 ,/ f@ / .
l}ps*9l' anr, f% _s 2 .... e ... 4 i.. m f 6g yN V:- (' ( - Fq,s5 o, -
) \'
A1'O -
)) ) l /gs9o
- c. M ,.y \ , 4 ; ? ' }W C d / 1 ..., e _
i STRUCTURAL CONTOUR MAP - TOP OF PACKER SHELL FIGURE 7
. - , ~ .
P00PEDRIS!El LEGEND LOCATION OF WELLS EVALUATED (' . :.243-FOR CONTOUR DATA
^ CONTOUR LINE - CONTOUR IN AREA 0F INFERRED DATA \
b B E _/- PERRY PLANT SITE f .
~
h
- j y>Y i
/ i ' -
r or / N raistsviar I t'U 2 lSo l O
/ /
L L,)8$ sc
~
p, , go / 2 -
\- w3 g ~x:e _ /
(ff {/ acsmte / j ; N-n25 / l ,
/ !p .. *g.
j \ ( (-)2300 /
/ / / . ,LA .s K\ 1 Y(-)2350 .
j/j
/ . / / -re.
ge- c- x ' N-)2 4 00 - - L tenov
', g.
1 c.,so.
/ l7
( I/ l-
\ \
g f / f843 - I \
} -) f' a l -
l ( f. =~~ f Thool-)2500 A ---- VIL L A GE
~~ '
( ~~
'3,,
c UY H pO,d A ,[ ,/ , ,H2600
/ / / / i f, ,,v 21 y1.. ,/ _Je.se._ _ _,e m ..,_ _ i e m .v.,
( [_ l. 4 0 4 8 SCALE IN MILES 1238 lil
i J l i Clf Y CF
. . . . . .. c.- r 1 . =% 3 , ) I # \ l '
icc navr - g 1 -
- cpiG{u\ , .
/ .- .: - ...Rn.o #s,yiLkf.:,4
- t. -.'. :t - i
- s. -. .:..-
.f . . .,. l .- .o s -.g @. . . . . r' . . . . . .
L I~ 'N N. ss i . 1 , N j ,, y . . e t R e v4
.-r --
s a v e't oc a
,! % Y e) --y TH su y'#.> ,- w " 04
- l ".. .
. 1 = . i
(.I) #.... . . :.. -(;..,rgyor
.f,; . , . . -
i .. no it i ..,
-)
ty .
\ \
eR,Rsh y
,. .v ,1, N o v a
( ,.
;,,,,Ra,,
m .
. x. . .-
- . : -l:ii .Rs
" . e , R.o, .- r_ : .- = . _ . \
0 g%o n 5,' ,.:.....;....'. 1: s' is..f.r. ... i:. E .'.:'2 ' Ni TRUWBULL i . : ; .:. . ii-?
. -l: . . Goo ~-
M O R IA M
--..,, _ b 'Y .- !DORst. 'NM .t-
[r
.J. -m "i, , N L. .- \
r .' /- 4 r400 .
-- l i ci i .'
l . I. pso H f H AR TS 3 ROVE MMC NEW WC CHERRY VAL L (Y ANDOVE R
~%
[ . l
/ .I.-
14 5 l/ t
. Ooo ~ I, ' - ~ ~ . 1 ~~ ,
w'ND$CR . iCRwfLL ~ -- COLFBRons
-- * = ,. w atea V N.C W il t. l*A W S Fi E.L D. i 6 )
ISOPACH MAP OF BIG LIME i AND NIAGARAN SHALE FIGURE 8
LEGEND POUR. OEth2W H6IL i
...- LOCATION OF WELLS EVALUATED \ FOR CONTOUR DATA
( ^ CONTOUR LINE _ _ _ CONTOUR IN AREA OF INFERRED DATA b tB
\ / /
A
/ %
PERRY PLANT SITE 4 b / IN40'oy b b / !
.g /
e~x . an o, nusm I I
\
i
\
y>N \
, g jRav , f' \
y
.i pu _
1 3 f
)'f u\,c _ _ \
i
/,60 -
N is'o
~ .' \
8 n %, i - LJ \ ' g K E WENTOR
/ , \
CONCCRD_ fLfROY _ THOkPSON _ _ t
.- I \ ,.- GjE A I[GA ., 1 , . .I
_.J / i
% -l ~ < unoox / 3600 I
gu er , WILLOUG H8 Y
/ RTLAND CHARDON HAM 80EN ! WO N TVil L E /l . I i ! , . g I .' l - - \ l I
CUYA HOGA I I \/ . I
; g ' = AmE LD Cs)_ga IC,saioc=
wu scN_ _ \.- su=TsiuRs k l 4 0 4 8
'~ '
SCALE IN MILES 1238 \\5
. g ,',ff_
y3 - ur u 7 :;
,g
- f ;y j ' n 4 !
l
) . - ., )
\[E
- p8 R - . n. y y. <
q-
<;qb -? [, 1;<-55h p,,, -
6 , tys. -d$ h 1 I: , . IU, U l
~
t f tr _'"FgJ(ih';G:%'ij.
# g maky.N, I ,f f '. W I,- - I k *. % ? w l .... (t_- '
fh,* f ?... ~ s
)
fL =:c G y h ?
**F Tw;< ,.'
n 9L:, ', el'~ c .~ ' ' '!
~7':: : 7, n s_U 1, y y . sbr - -q 7s -
4 e L.- .
- ; v3 f." p;,CC T-w ,
wa - ;r k ,; .,. ). O
- 5. -
6000 0 6000 12000 W_g _w;w.N]4, &.."
},v' Q(. 3 '- . +
%~$N' M. .' ' R 7 . [' -
,, 4 3
SCALE IN FEET Q ;t- . M i, [ b
~ ?D ; M h); I t:.2 "i. M3". , ((.. ,v- 7. - 1/
& -a^
q~i(R . 20 Tgly_ ;Q.;;QQf? l\ +1 y -
y e yt~ f.. u
-m ' f. -
Y' , LEGEND
? l ,' " -f p ',*,. 't- -. ru -: .!' v.1 % OBSERVED OUTCROFS , . < g' tt ,*., t 'e -
EW M- -- APPROXIMATE OUTCROP OF
-4, ^
% < -~
-.- 4-3:1 BEDROCK SHALE 14 ( w' J/
17,
$~" , ,- 4y_->, d*p-OJ , ,[R,. _ r APPROXIMATE OUTCROP OF g'- * -t m .-- <
BEREA SANDSTONE {- w ,
.q q ,-
q .1,,_ ',4 4 ' .i3
- - - gv , '-r f'; f
- J-km . . f,e
.7~. ~~U* .
'* / v ~,{ < k
&,. R ] (ri myl ,: M4 m/< ,
2 ' - by 9t]pWsf;.: [ D'; ; .% F 2.e. gAtR. ': n ] 7y4, s - pg?rMT Jr, r o rg
- e. , 4, h+yf 'n g tY L ^'
as yhp. g nd: . . . .< r L gt r,2 7 n. m. hyhw nwN 12L4114
~ ~ -m..
f nGREM .AULT,( F
, = . . -
a (' ', m,i' ,
)d '# -
c #
~F/ "n a .' Of [ ?^ %j )g \
j'
% ,4., @0 m \p dr , a '- '
k~,x[gn'*rhjn, Dhx y' a)g FAULT AND OUTCROP LOCATION MAP
-h(! ; nth-N, .
j FIGURE 9 w.un2+E.9 c; mm%t
- i - .
\ ^
s 1. + 4 s . ; 1 s , , , g
-(
g a j #. 2 1 ERR Y.J ELEAR 4 W _[
- POyER J4AN}
+ ,_ x I w r h _ > .g 7% N >f.-gI ,. e ' %. ri j'N ' . .y ...m (m . ., }$ _ l_' . ., .- -N, 1
3 : '%A
\\ .g /g + ,
m m;m! " nc ! MGLACIALs
'f- C ._% TI
_L J 2 ./ p .
-rd., .- (,/ph8 ~, '. [ '
p.---' Q'S 4Al.E EXPOSURE 52EKDEPT
- = - IVER: " . - '
f-f /, p
- 4. [%71' 9[; ., 'r' y}jM'QN(,j,O ' 5 GRANDxR12.a
- ,. f%:-A~l - .&q x - # D ..;.... . ~ 1~
n,&. t.-- . ; %f.sJX . M W u% 3 .,- 1% */.ME:Awr.c
;j % . . n/"- -
y k:g' W cce j V - - 7' R,- : gt..a';kW
,V 4y ' u W s +.
u g-
-_k-a n 1^ f A ~ ,a, ~ ' . l,. . l /,,', 6l ",' w . Q. . . ^ . . ; i. c' -
{ ,5 *' zi _~~~...
,,Q'vd '~ ..s.< ._.- ~\ N- ;N ;,- .?"
t - g'e Q- 3.;.] t'gg
./ W "' / ; & ~-.v.e- g -.n- pe %
y n .~ . % ./ . f..r J~
- e. ,_x~. ,; --,, a ws
J', a* m' m . . .% m p'_\,p,y '& ', , s g .s'r' + f, ,1.,v~ga<. py '~}(~ <g% 's .+ ~m' , ' ~ ~~,
- g. SA m,b>-i-fi,,- ,.at, D ,_j py, J g r
- ,, y _, w ~ UAGRIN SHACh ~ m y , . .a,3_u . ,- a. - , , ~t,. - >' GRh ~
V R N? / t .%^- ', ' l p;d.c .g ' g 5 gy'sy& A - 7 - N. a .->p , g- ~
- , . a , ,; 3t ,2 . . N(a_,y_
x .
,,, pap 4 ,;.-4 ? g- g . .;'<
m AER% M,7,]c t h f 1 % g f $ .. ry:. , e -
.-."g"p/ + ,- Q W V. .e' g4 } y'j j yq gyhi ,
q,;-',i% q.c3; , k- - ' kKi [pjpghh$5h fh.y.p, ai g?,Q. e., 1fn.;w
@JG%
- x. ; , , .
- a. \l.. dy% w.)f,g . .-k:'Y 'hd,qOSU h; '
t(s_Y.a nerir4 . =~7m .:.awmr~c . 1233 115
NOTE: PLANNED EXCAVATION LIMITS ARE , SHOWN BY DI AGONAL PATTERN, a AND OVEREXCAVATION BY STlPPLED f PATTERN. an
* \
R 4 b qd E 4 g V s h} . _ (
. NORTH 3 r
b a* $ $ $ SOUTH PLANT SITE FIN. GR. EL.620 INTAKE HEAD LANE ER/E g4
-, ._ ooooc aco a oc j/ /
yb l
/
ks
/l/ fl hV hll/ 2 a a a a os 620 llll / R -520 10' DlA. CONC. INTAKE TUNNEL I I N ,
s LIMITING DEPTH OF 420 mM N DEEPEST ONSHORE
$0 HO DEFORMATION AS DOCUMENTED IN o5 -
OVEREXCAVATED AREAS, 320
%do -\ DRILLED CAISSONS, EXPLORATORY TEST BORINGS yz f=j Qg s Np AND TRENCHES AS WELL g AS PLANNED EXCAVATION. - 220 5 *y 5E o k' e
pa ,WA -- s g M & 600' O, 600, 1200, o o 3 HORIZONTAL SCALE Ty Figure 2: Schematic northwest-southeast N9 3x VERTICAL E X AGG ER ATION Cross section, Perry Nuclear Power [U Plant, North Perry, Ohio. E9 __ M8 m
A
.c a . gg_g . .
I m M V 1.7 }_ $< ~ \ h - pa m
-,3 \b\,f b' c.
s ce.sg ,,. 42
.n l'o %,,i o ,r g'\ . ',p p -h , .. . s a k l 4 rv e -
3 t < ,,, ,e eL) ,
,. _ s ,8 , ,m,u,dm< ; z In,I, _
l{ . 5 O, e.
?' . .j {q 3 t' lI c h * t 1, 'a . i hl,it l, . ll' t-aiz$'h ia w ?' ' <.& ; "jh. .~ ;.j :.y).. [g. ,p.;.g fs.i h?%s}Q.
1 $ $ \$ ..: : kbhk$Ri$OYbb - bh ewse~
$$ M6h~ hcl $.hYgh
^ il EA' ' E d%d' W'lhl@M
$$$$$en ee* M@i$${$$g$$$$$@
k sages 3 ; a g ao s o
-s o i 3 g.3 Q {5S . , ~
s m y
;; w "z:3 c
f f ( w
- ;; r;g;; - o es-N zA M C l'*""
z
~ - ,
s' P90R BRIGINAL I !- 4 3
- u W
\
l JNTAKE STRUCTURES R T k Y\
'N *\\
NN Ng\s
\ ' *4 h Con
[MERGENCY Q 4 SERvtCE waygg \ E
\f+ PUMP HOUSE 3 "!!3er.
CONSTRUC T/0g \\
\ .y / e SERV /CE ggygg \N ,,.
rump nouSE As'Ess skarr T rx-4 3 s eTSql;. $ p 6 Q, s . o c
- %.9r, ,
j A i 4 44%esbr1 w..- a ,
;s[gyj4 g$Q> ... w%i t
P/m 'C \ a" y, L,wnsk.;s g ; -r
- pt;. ;,yr- v . .rp* ., f \ . . ow ,e, 'y, s-/ \\-
kQ&>',-Qp'"jg*i &
- '%g
- x.
{ ~ i ( Yl k \ {
.1~mgq:gy k$ !
y . 1238 118
P00R GRIGIML 8 8 8 8 O O O O i ,i i i n n .,, a a a c> \ a a f w RGE w w 4 w TUREo u - l a
.r<w '*S^** -- ,. p: n .. o~ ,. / c. , J. ' LL- ,C , 9 %d. ! 'A C C N'iT R UC T IO N . At ?', -t 'S ACCES S SNAF T gf '" . ./ ~ ~ ** [ $.1 O!5 CHARGE TUNNEL ENinAu E Ma 9s h $- \f '" -
- cp
-staucTusE pg hy p}f. f +; 4... ~ . f / ;.
Tx-s --
;;Tx-2 Tx-s - ~ ~ -- i .eSedpa , e - . .;.,,.~ < t ., ~
s f i7cestN:1 M,VMsN'" w t- n Ww bd,.L4-. J
-..w we. 'fj '}.
- 4- -# -
~ T u ; .-
I*4 ,jg
- b .h '; ha.) p' ss k *p ER ,
.* , g# ,lh p gj ,y; i, _ p +r .*"*1 . -).
s.. ,, U . .- , m
/ ~N 'i - , et .fWY .p, Rp e w,3 '. 9 ;c,' e .5 .N.,,Nsao
- p M'7% Y* :L . e.
1.. 4, ' (i +;ce - r
- 5. g%gp x,cr
. w')gh g ,e o{4 .A n
m _ tas-- ..o{. . #2 nv .- - emwav .
~
7 .% .. ; .~,, \ -'
- n Qbi Q., g.e )}. . .w %tpff \p, "::} s& c_'q, ;n7 m ' \bP L ?)^ -q%
E ..$ ' _w: % ;.>:
!; # y 4 *' ^>
- Jib 0 # 6 \ "
,4 _- v c
LA A 9,#,e l ' '.y.. q }, :q: f .Em^
+ s i * \.h.
9.. e .< e,. e . ; ? y fh .:-
\. o -,
- 4. 4 ,
.y _..,-
o t -:<.4 \ si . . . 'i . ,. f-
% 1 11 .p h*, , N' ,, , [ g ). , , , :'[,4.s[E ?s( ,
a & ?C4-<rzib NW AWlAMM S, , 4M p- t @i }.'h .. h\(M . 3.# m ! d,6* i e h [s'h[r$; :-] i A ~ w. MhQj d w '1 sre- .
\\ ,( N 3 j p,. ' ,z :;- ,g ., .
4 . mb, 1 u. a 9 s. - no e-
. f s ', i e'u d+ - c a o y v1 .Q;. ' c. r;(g e t .8 *\ @*g;Q; , - y ;y_ , :.
p M )k 3 ?
~p \ k < ,3 &cgl/5 \ x \ , \gg g h1 "' o yAe j. m,e- -e 4 jn- ,d f\\ )
A N ,- a# .
'\[M-:..; g* j A S
[e!%Jg i 2W 119
. 3, .
s . , ,s .\
,. x gys s
MAP OF SHORELINE SURVEY FIGURE 12
8 P00R ORBINAL g g! 8
@ o R e
3 e R
* "J !
INTAKE 0 k N STRUCTURES M.
\. DISC .As STRU
( 1 N 78 3,000 l _w. 4 1, . _ _ l ! ! l l i i l
! kb j
kW
; LAKC '
CRIE e
! ! C 2 ! x N 782,000 - -- -- - - . - .
TX
, ! CONSTRUCTI j ACCESS SH- ,* 00 0 !
500 1000 i 6 EER WE W, ______i ' i , PUMP HOUS 9
! SC ALE IN FEET N 781,00 0 _ _ . _ .
INTERSECTION OF
- WS$
hkt PROJECTED INTAKE TUNNEL FAULT AND *f .g' i ELEVATION 440 FT. A.g '. y' l /
.g g'f,, f. ,p' .% -j/
l{ . Ahk1 hi'% l / 6 N 780,000 ___}. , M_5D- - ^ - --
' J l
i
/ ,/ '*?$~1T;e[ ) q, I QL % /'
3
/ . <- i m
INTERSECTION OF
' ,,+.;, p, p%,X'rp4,y PROJECTED INTAKE ~, l k.' . ";_s ,- ,p / g '4 TUNNEL FAULT AND - -
f' ' ELEVATION 570 FT.-
'+a.->J1:rQ
- th -
w
. , .. ' j%f-t . - ( , . eu ,{J,#h'-~~,J/s '.h g %J - f:j ~f.;g]l m f,'
e e ,- 1 3
, s W f.. istog M' iy $' *lh . gf.4Q_ 2'7' N 779,000 * -
N
.(
s' .h .. b $$$?- 5 Y C...b C v.e'. 3 fgn.7 h,/, ; ../ ,',sf,,w$> b p,_)??M;;h
,ugos:
c .
,v '"' '*7,."y5 '- Ml~ 5h -- < fjfp ^lNT ${k? m.cup,, "$$m@ ,' _x gd( nk gQg<:i m d ,- a yA c. M yIk hpw N T ,8.eeO e. a~ . , ( +$$db'." [ M M 5 E d.h '% N ,1 h ( 1*230d I"c0 N / / /"' "** / A-0 % A-4 A-5 / A-1 * .) ', PROJECTION OF THE FAULT A-2 ON LAKE ERIE BOTTOM B-1 [4j8 B-3 s 4 B-4 B-2 B-5 -+- STRIKE OF THE FAULT / IN DISCHARGE TUNilEL STRIKE OF THE FAUL / IN INTAKE TUNNEL N O LAKE ERIE S 4 INTAKE TUNNEL " DISCHARGE TUNilEL 7 N N'87 % 0 300 FT. 6 * = STATION LOCATION A t 0[ 121 LAKE ERIE BOTTOM FIGURE 13 m o m m o H C z 2 ENTRANCE STRUCTURE a a o m m o Q me E "8 2 G .E. o zo UNIT 1 oz ~ COOLING TOWER oo z mm UNIT 1 o , n TURBINE BUILDING n" E ww mm m om N o o o g"o - L%3. y o' W g __ UNIT 1 y3 CONTAINMENT E ri - g z mm m z O a rn rm r-N C
- e $.
o w$ c UNIT 2 z CONTAINMENT 3 > 2 F w v1 m O m H UNIT 2 0 0 TURBINE BUILDING b m w O ? V V) Ch b o 5 x C M q 5 5 5 5 l c a m cn w o ~ o o o o o d) o o o o o M F FEET ABOVE SEA LEVEL a x i s DISCHARGE TUNNEL HEADING N27*06'21"W d 800 - C [ 700 - m w 600 - LAKE ERIE E 574.0 v 0 ~~ N APPROXIMATE LAKE BOTTOM 500 - *-DISCHARGE RISER " 'N Of w (457.5 '- S'rg N 12 FT. DIAMET 400 - % % MAPPED INTERVALS EXPLANATION h k3Y-THRUST FAULT, DASHED AND QUESTIONED WHERE PROJECTED. ARROWS INDICATE DIRECTION OF RELATIVE MOVEMENT. DISPLACEMENT IN FEET AND TENTHS. 1238 123 I LOCATIONS OF CENTERS OF POWER PLANT Bu!LDINGS PROJECTED ONTO TUNNEL ALIGNMENT " Y s' : _ e _ ;; ~ a ~ e S - 5 2 - t - 5 M I ; w ; .' ' 2 ; <c 5 ; w -800 - r 2 2' .; % 5 5 3 3 5 5 E S = 3 0 o m w m " " * $ j u $ - 700 1' ir~ p , i, , ,~ m / ( t20 -600 " INTAKE RISER y -500 (441.0 h 200 L O O 200 SCALE IN FEET 12% 124 LONGITUDINAL SECTION, INTAKE TUNNEL FIGURE 15 8 s \ L INTAKE TUNNEL HEADING N44* l2'41"W a 800 - 5 700 - 4 $ LAKE ERIC ( 574.0 600 - h ~!. APPROXIMATE LAKE BOTTOM E 500 - %s 0 C (456.8 b
- 12 FT. DIAMETER BORED TUNNEL 400 - INTAKE RISERS MAPPED INTERV EXPtANATION
-r_+ i o' THRUST FAULT. CASHED AND QUESTIONED N -WHERE PROJECTED. ARROWS INDICATE DIRECTION Of RELATIVE MOVEMENT. DISPLACEMENT IN FEET AND TENTHS. I. t 1233 125 P08RORiBINA_ yy, e -\\ ' s.gtpy xs';;c7.. .. / 3 N yb - oh'O gD.v 7 . = a / / 'i.akyJ.f Vgs v .. .e 7~ A .;.\ ' & ,n4# ti . 1 u l1 l %.ma.3"is .y. j;! .. ' / j ' , g; $$ \ h l'- j e . s c '.? h-v. wp e \ p o J>, l . '.g ;, . 'k.' '
- s -
f._ _ f r i *M _. .: i l~.~*l -j%g //* \,/ - / Nbf{h 1 %.. .., / ,/c u f ;'y. / 8 7 / /'/;,*:.~.' ~ ~ l( syQp o ).1 l ~ /i; [ ' ;! ' . ' l . /,r ,/j;' l y I //,/ I y$ / , 3_ . -- w , 1 g- d ; : s i 4 __m., , l _2 12 8 126 LOCATION MAP, OFFSHORE MAGNETIC SURVEY FIGURE 16 Ot9CHAPKsE Tue64EL \ __ INTAKE TutteEL '" W OIP PRMCTION o or m r m (sp ,.;, / f - ,, ' AREA OF OFFSHORE MAGNETIC COVERAGE F" ' \ - s _. ,s s. f ,5 g - . - . ~ or , 6 EftRY PLAN.,T SITE; 3 4 \ '% s ^ \.. .sw g' $ _g[,,,, M,(N g ,l .. ._ 3 'g.Y ,,, ' . ,s .- <[ - LINE'tS.,.- .- A _ . !o - ( ' I - M /, O,+0 h _ gl p)gh*g% [ &,,,, .-! (, 5, f\( W _ , . , l p f.* _, . ,/ , ,/ - ' :\' h . %; ,/ Cf , f\ .f i'" r f .' ,V t;_ n , 'N s I O' ~ 4 J j ., ,s , t ,' , n 7 . ;q ~ \ ^%' , .* /g J y .I- s /. , %ga 1 L p - ~ . $& '. ) ~ -~ QW Q&< :: :;@%.A. -'& ' ; 4 ( / . Q >. 52-y /* / .n 4 ,,, ,- " p)} . '/ I y '" :~ . . . ~
- 3. rw - Q 2-n A '
9-. . </ . . _ _c i # s l . _ ,,, .* .n s .- . w id ; . .' Cd-. d /, ag~4 L. f 4 a l MMy ', r ; & .,. dQ' j- i [ , . _ . , .T.- g g . gg h_~ _ _g/ I I P l- .: in ) *,' '~ i- - -" k o :':*E % ': i.i..
- LINE 35 ;-
- l e' O .$,
- y y i 3 ~~ s H- , ' .O \
g: *'- a v% r $y,%l0:' a! M M g fV;. - Q% 22* * ] .'* ' [ .,1. '" T~ *' * *' .g;d ' .A M: [ffwn 'T cd N 'o ,g. N ',- *I - q[ ;gdg ~., \ . o+o [~f,
- r j
/ ' ( N AN --# M' - M "f p,ui if;'_\. i - y S ,, ,f' . ~ p - QN H x I .- LIN E,M -W o p*. ' y NNE ($ V I LLE 3 / i .i *:o g n Soon f . x ~' !2S[h:# 3 i , . .j~ , i ;: ' n EAST WALL CEILING NORTH #4Ul? GEOPHONE SOUTH TO 216 TO 10 INLET - PLANT A A A A STA. ll +10 STA. STA. STA.10 +00 10+60 10+50 N ~ go' , S N J m i A A STA.10 +90 STA.10+20 3 ELEMENT GEOPHONE 12' AT EACH STATION _t , 3 }2 ) LOCATION MAP, SEISMIC SPREADS, INTAKE TUNNEL FIGURE 18 EAST WALL CEILING ULT GEOPHONES SOUTH TNORTH TO 216' 10 TO OUTLET . PLANTj . : x : - A STA.14+05 STA. STA. STA.12+ 95 13+35 13+25 SHORT SEISMIC SPREAD Ul? CLOSE GEOPHONE SPACING N 216,/, GEOPHONESf , 3 i t- . g - J A A STA.13+60 STA.13+05 s .SMALL SCALE THRUST 1 w 1 GE0 PHONES 1 12' N 216 . S _12' J '--* [ ^ A ' ' ^ l i A A STA. ll*90 STA.11 + 35 12H 129 LOCATION MAP, SEISMIC SPREADS, DISCHARGE TUNNEL FIGURE 19 / / 0300g / A-0 A- f A-1 '4 ['y 2 & -5 ' '3 *C -R A-4) /
- A /
' PROJECTION OF THE FAULT ON LAKE ERIE BOTTOM >gg ~' kh)."y B-4 B-5 %/B-h s Y STRIKE OF THE FAULT IN DISCHARGE TUNNEL STRIKE OF THE FAULT A / IN INTAKE TlINNEL g o LAKE ERIE INTAKE TUf!!iEL [ % DISCHARGE TUNNEL x ' O log N 0 300 FT. 1 I O * = STATION LOCATION M= FRACTURE g = S ALLOW DEPRESS Of SCHEl!ATIC MAP, LAKE BOTTOM FRACTURES FIGURE 20 12 0 130 ?BORDE BL , e .i SITE Y . s . .s , . . . . , . . P.jr. .-
- u . .. . . ..... , ,.t.s .. ,.,.
. . . . ~ * ' ' - * .,..ql.. N 5HA.LE.' : ..: . . . , . .
- q. .. . . . . .
- s. - .. .
. ..'s,. .t. .. x .f... . E r-r--r- ~=t-r--r-r-- ~ -=-=-=r-=--- r-r = -z-r= -- =-r~2 =-=-=2 -- ,s r- - _ _--- _ _ _ = . = - - _ , - - - - . _ .-_- ---=-_r--r__--- =- =- . -z=_-r- - ~z~. . z- = ~ _= - _ - ==. = - - -E = = s . - = = = -. . _ ._ = . .-=- - - _ -. = __-. = = --- _ _ - . = . = = - - - - = . = . = - . r =-r -z-=-= xZ-r- = --=- ==..=== -z-z--a z-..:- - . E3= z- =3= g===~r-=_._. _:- _ - =r-~=~-r--- - - - - -- ---__ ----------=---_===_.--4= m-k b 1 o SKETCll 0F FACIES RELATIONSilIPS AMONG THE HURAN, CilARGRIN, & CLEVELAND SilALES FIGURE 21 ( E00000$$l. L CLEVELAND f CLEVELAND SHALE y .pyww.mc. := ? lE__, " a d .wc.w y I E .5 $Y $. m n w$ ) D j N N N 8 . +,u . - . _- - _ : . - ;- . : m . . : m....-.:.,.v _=__ ___.__ _ ___= _m- _ =_-_ _- .__ _- = = == _ _=~ = _= = = .. .. _ _ _-_ _ _ _ _ _ -_ == _ = _ _ - -.==3-=-=-=-=-=-= _=_ ,= ---- C --=-=-=-=-=-=.=~==-=me=e=_=e =-=-- =-=- : -=- -Z3-=-= r=- - ==ee= i-=-=-=-=-=-=-=-=-=C-=-=-=.-.=.=____-_-=_- ._=:=_-=.=-=.:.=______-__-=_-_=_:C:=-=_-=-=-- _= =.== _ _- U _R__- -=-=-=, H = _ -=_ = _ _ _ _ __-=_-.=- ___=- =_ _== _ ==e=_=====-=_=e=-=_=e=__e=___=-=-____=_ _ =__=_=___ = __=_ _ _=_ = = _ _ _ __=__ _e_ -==__ =_ _ _ _- _ _ i-=-__-=_ =- = _ _ =_- ___=.______- ____ rC - Z__'C _ _ -----':-=_- J ":-=-- -=-= -:-= _C-=E-=E-=_-=~"'='~~C-=-=-=-=-=--- C --_ = = _ = ========================= = = - = = = = = = = = = = = = = = = = = == = = = = = = = = = = = ======-========================== = = = = = = = = = = = = = = = = = = - = = = = = = = = = = - =-============================================================= ===============+=+=============e=============e==-==========- NOTE : SECTION ORIENTED NORTHESTERLY. \ I. ( 1238 132 i - - hij - M'dn UnidllW'k EXPLANATION IWZ1 SILTSTONE. SILTSTONE LAMINA. DASHED WHERE ! . BEDDING PLANE CONTINUOUSLY MAPPED ! BUT SILTSTONE LITH 0 LOGY PINCHED ! OUT. i i 448.0 n.. etu u.a: un ..u..n..tu. rec c.u..< Q SHALE. ~ .'l*l'U ui.a.l..d.';i.= u d@ .,l'd.':l.ullu."*"
- u. ess. .mp IR0NST0NE C0NCRET10NS.
SIL157 t.
- n. ca,s..a: u l
.no., n .n. .n.,u.tu .u, u .un.is.
- u. .s.< :
c.,,...s = .. u i ,.1. uu i I i .u., . . .tu. .o.. . .u. .. n.s..o u u,i.s .r u . . u n u 0 , i if C: C t .f 5..it ..& Sit f CL.T 5 7 .f - pt .f.t.L..C. , Ed.t v 79E .. ] . s. t ., :..sru uuou ou .OE% n .u .=i .61 ..c.u s.t. .. e-t.. .n..u' . 91.C.u.6. n i f n.. . [ 0.. 1. ... . T.E o.u, u.0.{ .c 0 u i .u s .u. . s0.t so.u.ns..can.nus u i .n u ues,. 4 k h , ) 4 17 C t.' 5..( f . 5..f 5 u 0t .t .t . , ! ? t.t..uu.r.f ( t ill t... Pt .513 %.ff l f .T P.t . , 416.Y . ..f 1 l L it t 0.f p I I.t l . 6 .v 1 $..T11. E .L. ..!5.f4 t . . i t.MLilf t Ut 0.1T g i.t... . . b.11 t , CL.9 1. 41. .... Suf. 50f 0.- UP8( .f .O.L E t.1.. 1't .iCEo.i 11.. 1 .ful.%%.f d ou L E .t t.f T 6.f .F4.l.el St .!( .g i%7f..u Of Y llt i g UM 9 T. l .v0u.l T.O.E lf 15 a tl t .. : ..(r .. t v , et t 7. i c v ..a. i
- 1. ' t .L .1.f.0.f : L I6,....Tf f -6..e -
5 6. ' litiv h . t i. t .1.L '1 ' O.i ; F O..t(0.. n.(o. 0 0 .T o.p . i . t. n ic is a
- k. l i 8 - Ci . e 5..L E ..D $] t t17 0.f - ...&..v, tg$tF.t f 15 l
- n. ., u u . ..o
.u.n s..a.f t 11 . C.- l IC u. . : 20.u ( et .F v (0.sct at t ,70 t : . . '( . L s.( T.' ti t .t' f l.t . . 6.l k..Y f .t -16... 051f.% f L f.5 6..v T trLto ! 5..t t . ( 3 1 % .0 1(0*9s . 710 5 4 0< .u f .444.0 cu.u.."1."s 1.o ..st o.e".: m.,- c o u ao
- i .....i.....
. . .. n ..o- i . , *a, o..a. . . . ...n o.: ..s .. . .u.o.v.m,,a, . = ,.o.u u .
- n. u. ..a . . , . . . . . . . . . . ,. .u.. . . n. .. u. . , [
- u. nn.. ..( .o.. . .. u...o.o.n ...u.. n.. .. t wi-s ...tn... a. .:.u u.a cn.u,nn..u..nu.o un u...
a.:. s - .. nuc.n.un.un...o.u r .w u ru ..an - s u c n p.i f . (t .9 S..(1, I upp 11t .fi f G.E 1 . .( e 'l . % v' t u0.f.Dt 29 . t .605 08 7 u,...., .f. .u. .. . . . .. , u..t en.. s 10.i . . n.. u. .i..u.ns.a . .o n u.s .s. .u . .,o n ( .1 5 0 ... I d.%Q......%..g F f (0.t s ro.. CL0tt4 9 ( o.( g'.i lg .% g.g 442.0 e u .a.o.5. . '. .U ia ...t c.s i '. n c f u.t i= =s i L I i t ,i f ! P DETAILED STRATIGRAPHIC SECTION, :. INTAKE TUNNEL & I , EAST WALL STATION 10+30-10+40 k
- 2. > b FIGURE 22 s
K'. _ _ - _ - . . - - - - - - . . . ._. .. . , . . . . .. . ..R n j k j l e INTAKE TUNNEL EAST WALL \ f (9 O sn o 4 M m + + + O o o v'4 m j y c , . . A . . . 5. . .s ,+ . 3- }: , ' ., " <'Q < i e ' < e f4 4 . e * * , . . . > . <
- W 8 ,( "
4 - I I i t ' N '. Oh .s..N 'M *N ' f
- f ,'fM*
~ ?. * * " * **** 3 B '* h. . ?N- .- A , .. ;.-. - u 9 .x . - <- . p. .y - - < . 3- a _ Mf) ** Ap ~ ' 1 ( L .. j " a - Y l . ' e . Y. . s g' s s > ' f, m y: ,,g.;,.;v.. g.g, .5w. .rp -: .... . . ... .:....- p ....._ ..g. ,'y .- = .. - . a . ;;; . . .w o. . . .. .. ,v' . Ws)~+ 3 y *.'$e .. n $s e(. 4[r. .c. 4fe) . . n . , ,m,e,, - q g ,I '* p'^4. . . $y, .' A > 9: A *n. ;[E &g '?.a**gv ' .. . . , av r n + . ;; *. . ._ t . a n . . l W'P*'** ':-- (, ' D J ' %w' '. g, E - - * 4. < t'u' '
- f Q: ' +
. ^: - .5; %"y~. 2 D , ~ ?}_.;i?) kd:l W, s: A i _ h*M - 'OS X'.E ; ' . - ]. , ll, v . -- g, '~~' ? 'i .u, . . .> .s.. . n' . . +n ,,.> 3 ,r ++,g,7 sv r, , n.g e, y * * ' ~ ' ' ' _ u j b ?y'$ ,EEN Bh*#,ha . N E bk$*k$h, ~-SX e>w -- ,1 &.ldkh:kY'A~A:~-- r-vg:- . ,~ > u . ~ '4 13 h ' O 3- ,. . . ' ~ -- n :. w: . ; e .. . .:7..-r~=--- . :4,-r n'? - - ? 'W:.v:' . , - ,r..,-.. , , .. .a e'{ + ' ~
- a g. m ys o
'. u o -t- ;~w3:w I m w- m:c .. .
- ; c , m -
_g = w ,.,_ . ' H i ^* j j wa \ i 3 i V 1 I 4 i t - 1238 15 l i s i II I .. Mk " $ uhtum EXPLANATION
- A
. G.5EA,i . AL. PL , . A.i f A. .L O.C Aaau,, ,Et E A C TU. E , j$81 SILTSTONE.
- IsAE Or sa0mei5n-eL AC (Sleta:TICT) c Il'!!'100'Oi'inntn:.Afi. SILTSTONE LAMINA, DASHED WHERE Pli"!.U!ti't",'"nk.UDATE. .... . . . BEDDING PLANE CONTINUOUSLY MAPPED fngg; (% !%thniS"'- BUT SILTSTONE LITH 0 LOGY PINCHED f CL&f STALE. SIL T 5 f o#15 la
-10 raaf P t 4CM A40 5mELL, RAeGE OllT vb'e R 2 TG 0 mp IN Tw!CenE55; bPPEs- . -T SIL T int. TITOst00.F T47pI5 0FatSTAn0 OSE.T BEC.Ont n "- 5
- t GRav, atin BLotav 70 ConCn0!DAL SHALE.
CTUR(; WITN Oht P R 0m l e E 4 T mtDI AL .'unta'; ': i'J"id'"- ' O 2 on T* s C E . OF LIGNT-Ga A f, ===. .a IRONSTONE CONCRETIONS. T5TOWE AND BR0m415# BL ACs AtuGla0uST) CLAT SnAtt UEat ^#0 4 A80sE REDI AL SILT 5 foot; 5!tGLE CORTinuou$ $1L T5fC=E L AnleA #140 .E OF U41 T . U417 N-Ut!T m .ROART 15 4 SE 00!a6 50RF ACE AT
- 5 5f Afl0s, h0STueARD FRON
^ T ids 11*55 A Pe0=Iulst $1LT570ht AS ATE 5 041T5 4 A43 # TST0st ; caos Ga Ae, mA551st , ef f e TO C0sCm0ICAL FE ACTURE ; J'C I ALE20at T OF SIL T5T0et C C 4515 T 5 B A C A DL T eAUT 1 mT( BL A414 A T !0t5 L I GwT -GR A T SILT 5T0nt 843 Stout!5#- tv slLTT SM ALE , CTEsta te tv A -GLE 015COtTI#uous (LE4Titutas) 3 0F LIGwT-Ga AT S I L T 5 T04E . c a uGl t005 ( T ) CL AT SpAtt A5 UEsf ,a OI5CC#TieU005 Pan AttEL ltAE Atovt atCI AL S I L T 5Tont . ' A5 PL A44R C0tfinuGUS PAR ALLEL
- lRAE 14 L0e[ A PatT OF un i t .
_ AL Ott MALF MA5 "9 A90ED* ASPE CT. . T5T04E : L0m(t 041 #ALF 15 CLAY 'L E , D ARE GE AT , ulTN 40#f R0b5 u gaow415n.GaAT, F E ta uG I #005 ( T ) , , ADL T WAUT L Ault4E , WITN 'lA40E0* 'ECT , SLOCav Fe Actual . UPPER
- -MAL F 15 GeAT CLAT Sn&LE tifw ita AL Ba0ADLT WAf f $1LT5T0tE SED 5 TO 3 Cm It fulCatt$5; UPPER SED 5
.C0tTIrvou5, .Jef e SE05 C04Tlquous, tLL CMA44ELS PRESE4T A5 $N0ut. 'Ott , AnD !#045f CtE t D As t-GE A T ,
- GwTLv $1LTv Claw Sn&L E m!Tm
,C NCI DAL veACYuat ; viTg numE 8 00 5 ADL f mAf f L Ag: NAE OF anl;n-se A,f alt F E a st G l h 3u1( T ) CLaf AL E . uPPE8m05T 6 Ca 15 mayy DE P 2 D4 E OF FL AGG', f altti 'E ttadIt&TE D SIL T5T0mE A43 5=AL T 4 T$70tE e!Te P ART]hG5 0F DA AR-6B AW H E , 50uT=w ARD Y*E T OP T*!$ 5IL T 5704E 15 B lp*L E maar T3. 'I AL 51LT5 YP45, UP TO ) C# ;CE, >AS 5f B AIGmT.f 8E 5f E D CL48tti >PLE S AT TOP, BASE OF utIT gta tes GED OF CLO5ELv SPACED 45f C4f C0bCRE T[Ct5 400bT l ,5 (p 'Cd. 7 5 tont - D abs Ga AT CL AW SW At t , 'N 4Lm150e5 Twin L An t ent , I t o 2 ** 'CE, OF DAEL 8tCut!54 Ca Av - Eb 6140 u 5 l ' ) CLAv snat t . U P P( an J 5 7 C# 15 unt? 810C10 ZCtf 0F FL AGGY , 'st V IgYrat amin A f t: L IGaf-Ge a v . T5fC4E &#D 50FT, C AB8-GR A T 5=4L E e 'LL LCAD CA$f 5 COVER 3ASE cr mate T5TCat Laele&E ; =lte0-elPPLE5 un LOCatl7 04 TOPS CF S ELT5f C4E 14 A E . 53uf nm&B D
- 7 J C04TAlh5 A mEDI AL FittuttG
- StELL!tG SED OF 5:L TST04t .
'CRE, As0 IRC45f Cet : S A=E SE QuitCE metT a, EnttPT A5 ItalC ATED "PhlC ALL T. et Asusta � DCSC#1BtD AT , s ,T75AT s 104Twe0u.Ga13 25 O mE RE et A5uaED * - l DETAILED STRATIGRAPHIC SECTION, DISCHARGE TUNNEL EAST WALL STATION 11+40-11+46 FIGURE 23 hh I"!N ('Udn fWO[lA6lL Jhc DISCHARGE TUNNEL EAST WALL k e 4 O Unit 0: CL Af sn4LI: 05 u + q ~ + r. M S: 0 T
- i.
~- ' ':-. , _ q g_ - , -a t u h C' u.n . : Ct4, s.Au o _. 1^- -454.0 5 O N X l .....................%.'............. ...............*....* ................ y 8 Q. e-u ,.....................,,.......,,..................... . . . . . . . . - . . . . ... .. ) i s-5' [ .' Unif M: CL Af 5 MALL &#D 5' _ _ _ w .- .-, ~ ~ r - '; c' M , o. 6: .easese. seats -~ s! s - , f: s-u , . . . < . . _ 452.0 [. a a u ~ ,' C ,4,,, , , Lhlf L: CL AT SMALE AmO S' r B-A L i: r - . . . = , ........ . ......... s WRi? E: taf SaALt. 51L T 5 cemeier ime masas I ' W + % ._ N - - - r-e.- o . . . .g . ,- _.7,. s. p:.amr......eener...............p.............. ,...,.... g ,,,,, ,,,, g - - K I, _ __ . _ _. i_L 450.0 u.it a ct4f s,.Au Ac ~ ~ J . W &' f; *. , 2-. : WZ _ ._ m f__. ^ ' t, ' j , T-o ._4 - s [ r _u._',- w ah. . . .W~- ji ~.m ~ ~ '"- I Y kem -: r - ~ A-bhlf 1: (LAf SMALE. SILT , N* - _ ' ~ "_448.0 I I" ! c t $ ~ \\\\\\\\\\N\\\\\\\ $,b\\\\\\\\\ 0\\d\\\0 0 0\\\\ , ms m EXPLANATION FAULT GOUGE ZONE. GRAY, PLASTIC CLAY GOUGE MATRIX WITH AGGREGATE
- OF RANDOMLY ORIENTED SILTSTONE AND SHALE FRAGMENTS.
,.. .. A. , . i r. . LOC., F. AC Tu.o
- fN SE VE R AL PL A4&B P AR AL LE L I T/M SILTSTONE.
imisAE " OF 380wel5#-OL ACE (51DE RIflC T ) -I"n! 10'!'0!'iU!'tni.AYE0 SILTSTONE LAMINA, DASHED WHERE l','!!'!.UM't",'"n'i.Ul'! ATE D BEDDING PLANE CONTINUOUSLY MAPPED ' ' ~ t'51 Dun',',!" EnliiinT-
- 49 Clev SmALE SILTSTONE5 It BUT SILTSTONE LITHOLOGY PINCHED uEt Pas? PitCM ARD SWELL , san 6E nemy On 2 TO O se le Tutta4E55; UPPER- vUe .
;5.T .C fS.ILi C .75f i . ott0.GF.T4,15 f O 5 7 AOE DSE f i 0. n."- T BE cone 5 .st GaAy, m!Tg gl0 Cat TO C04CuoIDAL SHALE.
- 4CTUGE ; elf M Chi P RO#1 NE t t #EDI AL 4"MishanfNt'L1".s!"A!',
TO 2 *# Twlts, OF L ic=f eAv .s IRONSTONE CONCRETIONS. . L f 5T0tE &#D SP0m415#.0L AC E E ARJGl40u5 ? ) CL A T SaALE tEBf NN04 A8 0eE *E 01 AL 51t f 5704E ; 5 8tGLC 5C04fleucus SILf 5f 04E L A#ttA RE AR ~5E OF ut i f . utIT m. 641 r m 04:507 15 A BE00thG SueF ACE Af !5 ST AT104, hcsim=&s0 Feou ' A T Ion 11*5 5, A PRCaletti SIL T570ht P A8 AT E 5 b91 f 5 # A43 N L T170eE - DAas se Av, mA55 t y*, elfe CCtf TO ConcnolcAL Fa AC iunt ; DI AL Z 0t t 0F SILT 5f ont Coh51575 BROAOL T eAVT 14TE #i Ael4 AT IQ45 L TEnf.6aAv SILT 5T34E AeD anows!5a-At 5itTV 5m AL L, Ou t RL A14 St A 4GLE DISC 04Tituous (LE4TICUL AR) 0 0F LIE =T-Ga Av SiL T 570ht . 580 git 00$( ? ; CL Af 5 mat t A5 Uttf
- In 015CC4 Tie JOU5 P AR ALL E L ell AE ABOVE ME DI AL 5 tLf 5T04E ,
3 Al PL Atas C0gT1400u5 Pas ALLEL *IR AE It L0mit 8447 0F L4 t f . 5AL 04 E -hA L F MA$ "S AhDE 0= ASPECT, i f 5 704E LC=Et OmE kAL F !5 CLAT ALE , 04st seat, ult = auste0v5 it t R 0m915 N - E R A T , F E tey<995( ? ), 043L T wae ? La#l4Ak , elTM "B A4DE D* *E C T , ILOCs e Fa ACTLAE . UPFf e E ma6 F !5 ERAT CLAT Sa AL E uf f M # E R AL Is0&OL v hatt $lt 75f C4E GE OS TO 3 f u 14 T*1Ct kE 55, uPPEG BEDS SCCRTie s ou5, L0mER SE05 Confituous, AL L CuangE LS PRE 5Enf 35 tutet. TCtE, Ann le045TChE . C Ast. Es a v , I Emit t 51LT' Ciat $na E m;Tu 4C d0 l C A t Flat Tuef ; ult
- humi e 00 5 O AOL T WAff ' ult L AGItAE OF 90415#-584V F E 80'J 6 f t005( ? l CL Af ALE, U P8E RMOST 6 C# 15 4 A V V CCEO 20tf 0F f ( AG;V , f a t ti v
-'E#t A* !4 A T E D 51s T 5f C4E An0 5 = A L e L 7570et wif e P A ATins5 0F C4e a-CG A f AL E ; 10dT*maa n '=E TcP 'hl5 Slif 5T0tE 15 t!#Pi t .en t al 5 DI At SILf 5f cht . LP TO 1 (# -ICR, WA5 57taiGWT.CsE STE D C usain' PPLE5 AT TC#. SA%E OF 041T 4T A!45 SEO 08 CLO5ELv SP A' E 3 OW5T0sf C04CRE TI045 A80JT 1. 5 C 4 !CE. L 15 f G4 E - DAtt 68 Av CL Av SM AL E , in test a 005 Talt i Am:n AE, I t o 2 ** -!CE, OF DA85 840 st l 5M-Et a f 88J6!n0u5( ? ) CLAv smALE. u 'E 8805 7 - Co is = Ave sEc0E0 20st or 'FL Acqv, I tL T lt?t eL Amt q ATED L l6af-Ga Av L T 5T 0tt AsD 507 ', Dant.Gaar gna E; A(L LGA3 C A5:5 C0vEn B ASE OF =Ang Lf 570tE La#14 AZ , plC RO alPPL E 5 CWG LOCatty en TOPS 08 5ttf 570ht al#AE. 500f amAa0 If J Cont Ala5 A mEDI AL PIgCnigg 0 ShELLIOG SED OF Silf 5f 0tt . , . t 70ml, A#0180h570nt : 5 Ant S E Qu t eC E y . W41 f E , E ACEPT A5 It0!C A'E3 , y F A PHI C AL L 7, E sE A5uaE0 ARD DESC81BE0 af i b -IT5 t TmROUGs 0 meat *E A5ueE0 DETAILED STRATIGRAPHIC SECTION, DISCHARGE TUNNEL EAST WALL STATION 13+22-13+28 FIGURE 24 I DISCHARGE TUNNEL EAST WALL ! N CO ~ \ ~ + + M f M M % M talf 0: (L AY SMALE : n 'l
- : .y.lv.W
.g- ;.;y: .. +* : m, a.. ;;# g .: e-;- c' s, sm u->. r .c - . ~ --s .y , M'f ., - 454.0 , .Jons g < s r . .. LJelf #: CLaf smALE: -- - ....m..m...........;... v.....-...4.... . . . ................a........-...........,....... ir f'
- ......................... . . . . ; . . . . . . . . . . . . . . . . . : . . . . . . . . /g . . . . . . . . . d .
.( N . ~ . )S MTi _ . ~' L917 M CLav 5 male A40 _"a .- .... . . . . . . . . 2.... . . . . . . , . . . . . . . . . . ......,...........'??... s - 452.0 (;. . m _ . ---- _ _ --v---- - s ,4 _
- i. -
.*GAS $EM. MI , 3 <, '94 + ~ ic , ~.. s ; unit L: cLav Saatt Ano a w_ M 4y stF - - dL. L i > '. gelf A - Clay saat t , 51L I' ' -T : (:. : ,7q ,- . - gso o '. ^' , . 9.. M. - ' D. ., ,y.,. JL 3- += ****f.t** - 3 , , ,. y, , K ****d ~ '* W .l . % m = %.- < %**"' y .k=. . . ,o " 17; ~ - m, 1f , AL " . U411 J: Ctaf SMALE atD J v - 448.0 nut s j unit 1: CL AT SnaLE, SIL l ROTE: UNIT 5 ! A40 J W 57 4 T 104 ' 1 *4 ); anD DESCE10E0 A 12% 15g i v f IRT Att ST AuCTUR(5 \\ \\ \\ \\ \\ . 's . \\ DI5CNARGL STRUCivat LAKE ERIE \\ \\ ll H \\ \\ \\ s. w TN res # CISCHARGE TUNNEL \\ SHOREllPE \\ w > IAast $ttp g. PUM9 H0uit s Tual!N SU LDING COOL G TOWit Ta a b unti t suli 2 Unif 2 \ U'N I (I PL 44 A T 104 TURBINE pull 0rns % C00Litt 195tt T T- Ag ((TW bu R A
- $;;;1 ;? """om s
AT ELiv4T10s 450.0 W O W W pd , N) 3r; ,s SCALE IN FEET . GEOLOGIC STRUCTURE MAP, INTAKE & DISCHARGE TUNNEL FAULTS FIGURE 25 P00R ORIGINA_ u .oy ey em ea _ a . a w y=- = = - - s.~2 n - . .x-**^- \ __,4 - ..~7 < Th% . _ . _ . .+ 3i '5 g I -% e ' t [.' V c .= m - - n = ____ns ; . l I _ , c, -- ~ c=.:.-- c_ a e7-pr-?m}w w e :- en.~ y *~ .A ~ ' ~ ' ~ ~ ~ ~ ~j ; L' ^" ? _ --~------''~~~~: ' ~*'" ---~__..._.___.____._ 3 m:w---< 3-gg. - ; Nae- - - _. _r= . .,,,,4 ---~~'4-N - ___ _..____._; f ' -J'" 2 , . , . . . . - 0999 - _ -- - E K,,- - - o ** r gw.=, u . - _ _ _ _.__ _ _ - - - - - w..- - __ _ __ w awm Ea n.'a so ub i INTAKE TUNNEL WALL MAPS, STAf10NS 10+25-10+95 FIGURE 26 . v (f I t INTAME TLMEL EAST uMa so.go so.co m.m 10+40 1 * -% . . ~ . NON" ' ' * ~ ""'"I','** M _ ' s ' * " . _ , . .'N* ' ~ ~ ' ' "'-' 4 ' "-~=*- ~' w e ~ 4400i - _ a f*p" "J' h j . - ,wnis - 3 .y gM. s++%- --- - - w_ m_ - _C~ - - g , y .-a, '- _. q w yeI=sas r<- - -) ' * * * " (( I~~~~ ~~ ~~~~ * **** " ' l e440 fI ~ \~ _. +..,., J a4 , - - - W e440 _u } --- - - - - ~ _ _ . . _ - - - _ _ - - _3 I x %_- .m -- + _n <> &n r u .w e en - ., -f - j -- '" , , wa .- _ , _ - - n a- _u__ . Q_.w j f , I / ~ '~20. 30&
- F
...e /.'.X.**."..'"2, ,-. C3...'....a. e , , n, "a.,.,,. . ~ ~ ~. .. . . . . . . .;- . .,n. . ,.. ... . ga . . . g -3 w.
- .'. ~ iE*+C9- .
o ne - ; - - I ~- - - - - - . - _ . _ _ . - . . . . . .___ . __ _. l :t _ _ _ _ _ _ _ . . -we C ~ 0" .7 f .g, . f% "lrT "7 , 4 & -sm - . _ - - . , . _ . ;[ ) ' 5 %6-*.E'~~ --f -' - m. *
- G ., p. ,: M,.,_ ' , - . - .
_ ~ +g," "_,' ' ~ ',,s e). s.r _p:. . . . , . _. . _ 4-g;one : ~, ~ - w.A- ~ W '^~~' *~=~.,.__,,,_*",R.(,,m.*,*%w,. , A' =-. . L &l * ' ,*y / p ..:. , _-==--======---.-==~~~w Y ~ ~ ~ ~ -^~ -- -- d --- 3 owe - ', ,w 4 -- [ m wr%.w - __ -- _ u p, - weg. % p .. ._A I t y - - - M*01 Op 01 0s 0e 4.cs 09.eus Tfvm 153m 1M, I }238 ik 6 * *G. g P00R BREE P m mau. " 13+4 , f -. / / i ,_,g 4 y _ _ _ _ . _ _ _ _ _ _ _ _ _ f l o ~ ~ ~ ~ ~ Lj g g j ~- -- om. =,- e = - - - - _ _ _- _ uf .m. _ _;__]_ - _ ~ ~ ~ s j ,, e l l a, -' .A. 6 , I 12h142 DISCHARGE TUNNEL WALL MAPS, STATIONS 13+00-12+00 FIGURE 27 P30!! GRIG!!UI \ ( -- is.,o is.ao *** . '5;S0 . .. N A ,t . t g y '* y ,g en . . se4, g __ c_ - . _ - . - - - .- -- w. , _ . - ~ _ . . _ _ _ _ _ _ _ . _ _ . , 95 py - . . g*** % =+ /, =. ,p+; '. . Arti* M.t "'&, ... n 1 48,888
- m.
- r d. g4 g 44j E' : . . . .. ' ' " ~ "
#75 ~ ~ 4 :. . .e.. :~..~~k. . o ~ ~ .:~.~. 7.:3.. 55 .. ,...r.- * ,-:=... .. e .. / , ..
- o. ,
- ~ . , . ._,_j._p._.__._-_n_. - .-c"' -- "m. _ _ _ _ ,_ n - ia_ O ' f~itrA"-x~,yjs'-ru., - m _ .. _ . . _ _ AN +" g one. ; . . _ _ - _ 4MeM. . ~- _ . . . _ . _ p . # y. - m - - - . . - - - . .----l-.----- n w__ -a- -_- - 0%9 - y .. i s i I C44: (s.ts OG4s M3 T1Tu AS3m 1M i j '} } 93 }k N # 6 EAST WALL i 38*6 0 31 30 g.40 8 e i ..a. . % sd-)W;,;.4( Q ' . f.','.'~/- ' ' , . - b , . I. i7' - bS w.4* :iw) (3 * $ -4540 -- I - 452.0 -.-. .. _.--_ __~_ ~. g .A . _- m _S*_----___m~__,__.._._..--~---_--.__2.C....._.....,____._.....__.5 -4500 ,. F' ---w .==-cr -- _ li k __ -- l #Aggwe - , - =65W'.- - - r# : - . ie asessus m e.e.m M ~ ' 7
- 4> W O M . ,y e pame .. side b _Q _ M h ' ende=_ &
~ %*= ,, . y , p :r._ .; .g. . -.a.. ,;:: a. *,,.,=.,,,. p =.. ..=...a -= = = = y;,,=;a .=oi. . S LJ *,."J;*,:.= , = *
- C
- 3 f -Ottt
- r________. - - . . . . - . . _ _ - -
ge* .2v f, = w3r t ', , , , g - -- mm.m.- .c. . -- 2.===2-5kgj .gM alu - ---- -g,. _, __ _ m_,, g - - -; .- . 4 _ _ y- - 4 A -__ _ - _f, %.- i . . .sy C4 "% am.a= . ===. ? g ..,Z MJ r4- , _ "*- = = = . ti , -o rst r gW* . - ._ __. l -09G9 a , i 09 11 o;.is op.n 304NH3510 i 12O 14L l DISCHARGE TUNNEL WALL MAPS, STATIONS 11+40-12+00 FIGURE 28 P00ROfiGK 1 \ \ DISCHARGE TLNC E*00 lie 9o ll *00 18
- i t i i 4540 -
9 f - "b ,u p% ~ .* f "" ~ ~ _ - 1 - - - - - - ... y I - ~~ <$k. .... T % ,,. . n- *% o - f _ ..: . . . _ . ._ .. ._e___ w r - #= -~ hpg s , .___.__a..----*--==-----=-----~===~~---**Q n.___-_w- ----. . -o- .1-~~====== --- = -~== ,-r _----;---_; me 4490~ ' y -. p: ; , __. .g, ;;# . s ,1 - .w . * ~ - *n gn o gy, r r:CiY _ _ . _ _ _ _ . . _ _ _ _ _ _ _ _ . . . . _ _ _ _ _ _ _ , _ _ I mm5=ga = ge._ _ .. ~ _a._,. .=_=w_ _ , _,n - ==mel-res ^ _ h'- , -- -- -w,= mime.__w . % as~W'; F6 . p L-gg- g L _g_---w, ,g , emw- - - e_.-=.
- 4 0 059 -
________--._----.-.__.....__~r--~ ~ ._ . . . ~ a ~ , m _r = __7 ,_g--__-- _-------- - - - - g 0259 - p == - .- h . ~ 0956- , i t i 00*2l 0 6 *18 0 0 *18 Odelt TWm IS3M 13% ( 1233 145 e 10+40 10+25 l l l 1 EAST WALL 1 % l %. s k l i 's N E s i D 's s i s s s t 2 l 1 l WEST HALL 1t+40 l I FAST WALL WEST WALL EXDLa%ATION STRATIGRAPHIC UNIT 5: A, B. C, D. E. K. A L. UMITS K AND L ARE SUSDiv!DED IhTO M UPPER AhD LO ER SECTIONS 45 IhDICATED By SUBSCRIPTi 13+00 ,inRUST FAULT SHCWING DISPLACEME%T AND O!P OJ , - STRATtGRAPHIC 041T C04 TACT. t" g SEVERELt FRACTURED ROCK. AST WALL SviCLINAL FCLD A115. h0TES: 1. S'E FIGURES 22, 23. A%D 24 FOR STRAT! GRAPHIC Lh!T RELATIC%5 HIPS.
- 2. MAPS CCmPILEn rR0m IgTA(E Aga DISCHARGE TUN %EL WALL MAPS. 5EE FIGURES 26, 21 AND 28.
WEST MALL l t GEOLOGIC MAPS, INTAKE & DISCIIARGE TUNNELS 1 12$146 FIGURE 29 11+00 10+80 [ 10t60 i - l 1 1 I l l I INTAKE TUNNEL 1 i, - \ \ l% \ II ._ 5, * \ B i !# $. TUNNEL HE ADlhG R440 }2'4g=g A I ' l 18' k l D l3;g i \ . l l- , \ l, PLAhE OF MAP AT EL. 446.0 y I l al " '.- Ull ) zl li+80 lI+60 l I 1 \N l I \ DISCHARGE TUNNEL \ \ . \ l' s %. Tunnit ntsa:sc m.7o 06 22 w \ -2' ' 35' 4 Li \ Kg Ku \ p ,, 0.1, \ g~ \ \
- s. .
PLANE OF MAP AT EL. 450.0 \4' \ \ 13+60 13+40 l 13 +20 1 I I I 1 I I I f T - E.: .. .... ....... , W . g 4 1.3' 1 PLANE OF MAP AT EL. 450.0 ( ml L . El 1 \- l . = l EAST WALL o+ss 'l ~ k -*4ao I o+eo /4 %%"m,6 I /' & +55 D Y - , s % __ m ***O 9r ahim- X g s '^ 5 ?g,g;. ..= = =_. .._- . ,y g ,n ,w g n ,gxxNQ'f35i ks\busMN5\N \S' 1 ..=r..m. I I
- =
- :
- \%Ng(N ygk'NN [ b DETAILED MAP, INTAKE TUNNEL FAULT FIGURE 30 00 O! o,.D..!.P,llj.F 1 IJ l s. 4 t u s i n \ l l 6, siens ~ ' 44&O-i I ) _, %?q *- ..g" ] c+70 l ; ~,_s _ - y m ..V~ ? . J =~ e =_ .:.. _.=.:. ,a s =- .. -. = .:u = e . . a. .g;= . . .rm a . - . g~ --
- t. ..
- . : .:=..-.r:.:
.4 E' **r s , I. 1238 149 i NI' g 3. EL. 444.0' CNTT' ' ' " ~ ^ ~ ' ~ ~ ~ ~i' ' ~ ->" ~ ~ 2 ~ ]R '[ ~~ ' - ,[_ , g4
- c. q' , :s t , ..;* s;.
A . . , y ,. . .. ; 'Q c . :p ,, ;. . +. ~+w W , h% - >! 1 . e9 e -m v+ y ,.,.R ; - u . . 4 ,i 7 %. .1 se_ n .. .. .. s ;; + ,4 97 . g ~ i
- a A _ S3 y j, "g $
. .. . 9 4_ 3 l . . i . ~s.
- g
.. [i . e ,t 4 V' ,l l % 4 J* , m .y w
- w. -
s J y A. a(>" r, T' # .z~ . n. % 4 ; _. . m *g -e g, y~ - , * +m t, , e , , en . ,, , . I* lg g, .. lC . , - . , , .... # . .y ,. 3 m "a * ..."'1 .j p u ; . .. -p. A i; , 4 ' . ,, .,'~ , f , . _; , . ,_ h 4[ _._ A _ . . . _m_c :./s- ggi,, : .o
- s 2
j s -- r l/w q g _ . ~m '. " . . ~ rt " *A s .- m . ,r w, %
- e. . ,.c__
_, x _.-2_..._ .+.... w..., -
- a. . 4.
~y. g .-._..._ , = . . , . . _ 10+65 TUNNEL, STATION 10+61. FRACTURED AND DRAGGED STRATA; EAST WALL INTAKE TUNNEL, N370W PARALLEL TO NAIL. STATION 10+65. 12H i 150 PHOTOGRAPHS, STRUCTURAL DETAILS, INTAKE TUNNEL ,, t FIGURE 31 { l e
- 1. 2.
448.0' # , ,. t 4 . , , , e Mwee.w*** I f.* #A pw i EL. ,. . e. - , : .g . . . , d f-D.D ; yl .ff,y. M ~1 -'e3yM
- n,4 {}f;l die
%sy.qs!Sid@?WIJ2 mgeq;p;.fg WI9*g.** # h j- e A -- % ;{- p"?%Qg4 i **r %4 $ h* *; '" 's , W;khg,[ .] **7*[" [ +[ t.'}#y [,,al ," WCWd A i: ~ ~ ,:c: 44,,4 h( ~ .f y y.. m.a+ ., p- ,y., w - 3; ~ vag 9 ;em - jz. . $~.g pd . . . y , g ,% g :k y y q Q "..
- g. .
7.. _ . . . . . . . : gt, 44s,g. ' *g .' 7 x * ~ % R y k,7 - .~ m' ' 'k. ,. fy . p . a '.. :% i' .$+ '**Y _l'f _5 . ,( __ _ d I 10+51 10+61 ASYMMETRIC FOLD; STATION 10+51, EAST WALL INTAKE TUNNEL. STRIATIONS; WEST WALL INTAKE NOTE FAULTED NW LIMB 0F FOLD. STRIATIONS ON F00TWALL, TREN I4 4 \ t .y n,. w<c0:n nn,m .w 1d<; _ b u n a f cv vatun i, EL EAST WALL WO ..?-- ' ..i.- .+40 ... ... " q p y 1 0 f t.;. %, r -- L --- 'Up_a ,. a...=/" i d %E71*"1%mewWp . - i ' yyQl!" 4 #dn ' Ei"d"9..T m!a'a- m ' % 3 h@ @!! M ISI'.n[-[FF"? ,'~ nui;gd"Es>M E D $E W 4 500 -. . .re4 ~'4 3WY$_. ~ ~
- IUM
&p-' gS=']TJsj ~ ~1. . ~"- -- y, '.}:IMT.": g w 3 ,- 62I W _; ~\;hkl4xxs \ ~" -~ U'+'I~ cc e" - ,s\, . s\ ' 4Ngx\.\\13
- E =j .m =:. :. , .V,\ \\b\' ' \ s\
ouo ~= = r.::=..-.,::::.- } , \, .k\\k.\{.. EE25.h'L:"d"I 12kSI152 I DETAILED MAP, DISCHARGE TUNNEL FAULT FIGURE 32 P00ff0!?iBINE i ( 13+40 __ . . _ f 'J.. I 13+50 DtCCHARGE TU _j n/ ,/ A-2: \ 4 54 o- -454o j 13j45 ~ L f*%7- w Gl m -f. - b 4sz o- ; (g ~ ~ l ut.mc / , "; s.m..*.. ep # g u. ...- ag;;a, a - 4 . ~' "'. il::I: f "" N."2.I y, 1238 153 0 r, \ l P00R DniSINA 3. ffrf,+3j ' -Ew. -: r.u - A c. _.,.y --- % e v : . L -. c ,- ., __ v .: - . - EL, a-13+50 1DISCHARGL TUNNEL, DRAG FOLDS, K;NKING AND FAULT SPLAYS; STATION 13+do WEST WALL DISCHARGE. SCALE GIVEN BY STATIONING AND RULE. l ) ~ l PHOTOGRAPHS, STRUCTURAL DETAILS, 1 DISCHARGE TUNNEL l FIGURE 33 o , , , - i e M %n t a $ j l ,, i l r I F i
- 1. !
.c 2 ,,.4;, u U* , v. , '* 4 * . 3;
- g
. *. y a- - g[
- ,,.t , .- ,
41 + s' }L ? l [ ry 1 .w; s* , . g_.r, ,7- , .v' u . ;'d ;,[ I, , .s p., .I5- ra c s E ;c - . n, a, . - 'I . y .1, e.
- a. >
?g *
- r w'
.gR. , , +
- . , e .
[. . 45 'h'y y - + 13+36 13+53 : STRIATIONS AND GOUGE; WEST WALL DISCHARGE TUNNEL, STATION 13+36. DRAG FOLD; STATION 13+53, E AST WA.i STRIATIO15 TREND N220W PARALLEL TO NAIL (TWO INCHES LONG). I NOTE ANGULAR SHALE AND SILTSTONE FRAGMENTS IN G0UGE. A i ') . 1238 15r3 I, N- - __ NOTE: MARK SEPARATION APPR0XIMATELY 1000' UNKNOWN CAUSE -57,600 N -- -e_ r' _9 -- % -57.400 en M LINE I4 k PROJECTED FAULT PLANE g & LOCATION ON LAKE BOTTOM -57.200 e i i i . . . . . . , 57,000 il 10 9 8 7 6 5 4 3 2 i LOCATION MARK O N e8 mw H e w o2 mw th ., m= "8 G m NOTE: MARK SEPARATION APPR0XIMATELY 1000' -57,600 -~ -57,400 E F LINE 12 $ o -57,200 ' , , , , . S7.000 io h h 7 6 S 4 3 2 I LOCATION MARKS N -s7,sOO 4 -esw ~ m , -s7,400 M Q F LINE 10 a 37 , ao TWO PIPES STICKING zm OUT OF WATER mm HM , , , , . . . 57.000 h h 7 6 5 4 3 2 o LOCATION MARKS 8 3 ;8 g e2 EXPLANATION m E* u Ci s a PROJECTED FAULT LOCATION ON
- ~y LAKE BOTTOM.
>W 8 b C NOTE: MARK SEPARATION APPR0XIMATELY 1000' -57,600 UNKNOWN CAUSE -57,400 Y F j LINE 8 5 -57,200 . . . . . , , , . , 57,000 10 9 8 7 6 5 4 3 2 i LOCATION MARKS , -57,600 r0 -g _- - __ * -57,400 < Q c r. r E cc 3 LINE 6 _3, 5o am . , , , , , , , , , 57,000 y] 10 9 8 7 6 5 4 3 2 1 Hg LOCATION MARKS ox m 8 x 5o mx EXPLANATION m ~ u EE ^ PROJECTED FAULT LOCATION ON *4 LAKE BOTTOM, %E em CD NOTE: MARK SEPARATION APPR0XIMATELY 1000' -57,600 46 -57,400 Q 3 500' FROM DRILL RIG , o ~ ' LINE 4 57,000 9 8 7 6 5 4 3 2 i LOCATION MARKS -57,600 -57,400 a " ^' _ z - F g LINE 2 -57,200 w RlLL RIG 3o am zm , , . 57,000 MM 7 6 5 4 3 2 I d$ LOCATION MARKS m N 8 2m 5 @S EXPLANATION am O O! 5 n PROJECTED FAULT LOCATION ON ws LAKE BOTTOM. >S EM NOTE: MARK SEPARATION APPR0XIMATELY 1000' ,3,,, m - -57,400 F DRILL RIG LINE O , , , , , . . 57,o00 7 6 5 4 3 : I LOCATION M ARKS -S7,600 N SE' r.na,. ~__ - -57,400 g o O r LINE 3 roo rr. rRoM s7,em STEEL 8ARGE $o kb d i i i i i LOCATION MARKS 4 i i i m E E 5m EXPLANATION %S Em " PROJECTED FAULT LOCATION ON $ E$ LAKE BOTTOM. 8 %E em w NOTE: MARK SEPARATION APPROXIMATELY 1000' -s7,soo n 37,400 LINE 5A y , NEAR STEEL $ BARGE 57,20o l 57,000 8 7 6 S 4 5 2 i LOCATION M ARKS ~ - s S7.400 N e g LINE 7 F g h NF4R STEEL 3 o P\RGE S7,200 f C' a - 5% 87* Nh 9 8 7 4 3 4 3 2 ; LOCATION MARKS 8 8* m ?s EXPLANATION p := 8 "E o PROJECTED FAULT LOCATION ON yG LAKE BOTTOM. >8 sEm ~ N NOTE: MARK SEPARATION APPR0XIMATELY 1000' 57,600 ^ - ^ M n % -57,400 LINE 9 r E E 3 57,200 9 8 Y 6 5 4 3 2 LOCATION *AARKS UNKNOWN CAUSE f % 57,400 N LINE il O !T* I 57,200 S%
- 4 ;2 N gg 8 7 6 5 4 3 2
I 57,000 m x H M LOCATION MARKS E e2 m pW EXPLANATION e t; hPROJECTED FAULT LOCATION ON >8 LAKE BOTTOM. 5E* W NOTE: MARK SEPARATION APPROXIMATELY 1000' 3z,oo Q LINE 13 5 5 57,200 , 57,000 7 6 5 4 3 I 8 LOCATION MARKS N M Mesa = - - - _- ^ 57,800 # t F C' LINE 15 V 3 o 57,200 hO *l E %?a O . , - 57.000 LOCATK)N MARKS O %es E ES EXPLANATION z~ iA' w ~5 ^ PROJECTED FAULT LOCATION ON "E LAKE BOTTOM. Y m %.n NOTE: MARK SEPARATION APPROXIMATELY 1000' 37,soo
- n. _, _
57,400 F E 2 LINE 17 2 3 -57,200 . 57,000 8 7 6 5 4 3 2 i LOCATION MARKS Q r - :- - - g j -s7,4oo UNE 19 F - 2 3 m p -57,200 So OE :7,ooo Eso s 7 s 6
- LOCATION MARKS 3
1 3 EE e m E ES EXPLANATION n d en ~5 o PROJECTED FAULT LOCATION ON "5 >o LAKE BOTTOM. d w" m mm m mm m W w w w ww wz z x x x w xx xe a e o o z oo ro-w w w - we awm a w e nao 58200- wz za aza aa e- -e-m-- a-< 2e <-44 44 o<o $ m-4 r Imr e rr use 58000-57800-57600-m r r e # f4 57400- d Kg N[ ) b 57200- i 57000-56800 , , , , , , 0 10+0 20+0 STATIONS 1241 165 1"=1000' ONSHORE FOR LAND MAGNETIC PROFILE IS-A FIGURE 43 GAMMAS $ $ $ C C C $
- 8 8 % 8 o
8 8 8 o 5 o o o o o O e i n I e a a a a a a e o. _ POWER LINES - POWER LINES C _ DRAIN GRATE POWER LINES g _ PHONE LINES + %---- f Wa ,, ,_ TRANSFORMER T' o _ W - CULVERT ~ SIGN N C w T-5 U 8 . 8 m - p 'g T- POWER LINES S o <,- DRIVEWAY m - "' N - SIGN S MO' e m m 2 - S- - Abhh ink cn RR TRACKS _ m Rh g g POWER LINES E m s ?- * ~ SIGN 8+= o C $ 5 >- 5 sw m $ 8 $"55 5 $$ $ $ $$"$ $ = E8 o- r ; e dCas s 2 o e u reu u m 58000-57800-57600-57400-l s J 57200-57000-56800 . . . . . . . - 0 10+0 90+0 30+0 40+0 STATIONS 1"=1000' j' )b ONSHORE FOR LAND MAGNETIC PROFILE lE FIGURE 45 GAMMAS w w w w w w w cn u w w w w w a ~ w o o o # w w ~ o o o o o . ? . ? . 9 . ? . ? Z CAR CULVERT _ _e S~ o GAS LINE POWER LINES m 5 CULVER' = -H ?~ o ( MAIL B0XES u > - w -4 o o o" o = w - m o. + o g . -b E DRIVEWAY k $. + MAIL BOX 8 ~ k g rv - h54 E $ RR TRACKS M O s POWER LINES _, + P 5 } W _ ~ ~ B: m ?- o th C z L x mz zd s x nam < x<xos>- 58000- * " "* 57800-57600-r ~ E i 57400- ~ y # , 6 57200-l 57000-56800 . . > ' ' 0 10+0 '>20+0 STATIONS 1"=1000' j }69 ONSHORE FOR LAND MAGNETIC PROFILE 3S FIGURE 47 m m e m w w > z = o a e ss - >- ez w- "z o-or E ds 25%" 58000 " '" " "* 57800-57600-m E e 57400-57200-57000-56800 . . - ' - 10+0 20+0 STATI0liS 1"=1000' 12Yi'170 ONSHOBE FOR LAND MAGNETIC PROFILES 3S-A FIGURE 48 N j4 y e) ~$ ',. g i %#3 . 'an teo * ;k , V edMEM w &l m, 4 TM-7 u ..a e n o. u.s , u, .t. ..., las ELEvaT m 506 ,,_ kh Ee 3.- x .n .n% n g INTAKE TUNNEL f~ L - - _ _ .o 4, _Ig- y - -- + 2= 5-@ 1. T_,_ M+ ettvanoN _42d .._ ps= g ., s3r 3- K--
- ... 9y v T-g-
m _. , z, . .- 2 +- ~ r &q_ f *s ~, >p
- i. m a g ,-
_%.h = Ettvarm 3ee ,x_. 4% .m, t4 f ,, 2?' , .. >.i.o....,,,-a.....u. d, x . .,. e m. .,m.s..., ..n., ,,, 5 ' :"": 'L: . ' , ;",*,J,' .".' ,:, p: .n 4:,- .'u- u Is_i-p- 1 , *O - . . . . . . , . . , m. o, .. . . e r [ [+7 n v. . 56.it t 06
- i !
2 - et . 0t ' ' s t ge t , b BOREHOLE LOGS - GAMMA / SONIC, TX BORINGS 2,3,4,5,6,7 FIGURE 49 1 = \ c k Tx-3 c eu T x -2 <n Tx-6 Tx-4 .- .ma i . i o. g ,,,,,,,,,,2. " as-.**** 2a t 6 I '? h . u ? l ,....o. .n . . , .. . . , .w.s ., .. ,,- .1 .. e _P _s _ ", _' p < ,,,:"' "_'7-g y# -.. .3 __ % O "5"& ~ y T ==-- ~..y a~ ~- 7; ,- " -d _ 4._ L 1 e weir us [.T 4 y e =K- i~ So=rc in g ~,,
- f. - p ,-
%. g > 'ia? %
- x a , F es # +
F 0 g, ( ,, ,. ut
- M^ --
4 ,_ ~ $ f .m.m ,e,.uo ,, r s_ - C 5"~. 4-
- 8 :
- +-
2 = .e.u ; s. < e s su n x j~~ ~ l. ,7.=^% y ' > 5. ne IO"i 3. ? i et t 1e so ; es to 4 he W SCAS I, t 1238 172 xa Tx-9 VW kA A 4 l' AMW A & , j & " l i,4 l '; f $ g,. ' <') 1 1 a I a a p , gy - 3 %i G 4 - Y ,. es c_r-x- S _ ~ ,. - -g w& gs :-- ,'. x g_ 5f ~ .7 ,. s p g f,- g ":r- f- 5 ~f z*- Nw. g m- _e -{ o,a i s c,- - -_ / (4 $0%IC L(G .l. I'v 18' il( i s i t 'i; 2 's if 7 sf L ) %'t5 1 Fw: F [i t a; ith6 D ADi a a t ua ![( v 5a.gE f6 'lJ $' AIL GRlDj
- 1. ,, u " 4 6 A;l a i l,% touGE D s' I f t :mi % ,
at i ,1 b T e t 1 5 4m it A%3 $r nt { S i, T I l g g 3 > NAE ns FECJt(Tl04 Cf T*L T AcL 1 BOREHOLE LOGS - GAMfA/ SONIC, TX BORINGS 8,9,10 FIGURE 50 g 5 e **=A sa :;ar g g , , , , , i :. . , / ELE'vATION (,co' (, 'J tu . . , s,. % y .. ~ 4' %" - F s* sic - N k i % '91 in e n ,3 , , , ,, ,,3 Wq ~ so > sn 50%iC L c,G ELEVATsON 500' "o e a r, su , s ES 10' 9 \ 1238 174 s = ~ 12f/ 175 a ? APPENDIX A A STUDY OF THE MICR0 CRACKS '"0CIATED WITH FAULTING AT THE PERRY NUCLEAh .vWER PLAh'T SITE by Dr. Gene Simmons April 1979 ) { A STUDY OF THE MICROCRACKS ASSOCIATED WITH FAULTING AT THE PERRY NUCLEAR POWER PLANT SITE
1.0 INTRODUCTION
A small fault was discovered during the excavation of the intake tunnel for the emergency cooling water at the Perry nuclear site. Samples of the fault gouge and adjacent shale were collected in July 1978 by Dr. Gene Simmons and Weston personnel. Those samples were examined briefly with the scanning electron microscope (SEM) using techniques for the study of microcracks that have been recently developed by Dr. Simmons and colleagues. A report on the preliminary findings of that investigation was submitted to the Nuclear Regulatory Commission on November 1, 1978. During the excavation of the tunnel for the discharge of emergency cooling water at the Perry nuclear site, a fault was intersected at approximately the location of the projection along strike of the fault present in the intake tunnel. In addition, a small fracture zone was recognized approximately 200 feet south of the fault. Samples were obtained in October 1978, January 1979, and March 1979 by Dr. Simmons and Weston personnel. 2.0
SUMMARY
AUD CONCLUSIONS Specimens of the gouge and the adjacent country rock were prepared in a form suitable for the examination of microcracks and elemental compositions of individual minerals in the SEM. Two types of cracks were observed. The first type, due to the desiccation of the sample, appears to be unavoidable, but is readily recognizable on the basis of objective criteria developed previous to the present studies. The second type of crack appears to be related to the last movement on the fault and always contains new mineral growths that extend completely across the crack. Approximately 350 cracks of the type produced by faulting were examined in six samples. Every crack examined contained approximately one percent new mineral growth. On the basis of previous observations of several thousand microcracks in a wide variety of rock types, healed microcracks appear to be ubiquitous in rocks. Evidently, the microcracks begin to heal immediately on forming. The degree of healing can be a maasure of the amount of time that has been available for the microcrack to heal. The exact mathematical description of the function that relates degree of filling to elapsed time since the crack was formed is unknown, but is likely S-shaped and asymptotic to the zer and 100 percent values. Two data points have been obtained - one point at 1 million years (possibly 2 to 5 million years) from sandstone at the Satsop site, the other at 18.5 million years from shocked rock at the Ries Crater, Germany. A-1
\ .Q,d A ~J
The rate of healing of microcracks is very likely a function of temperature, pressure, mineralogy, and the composition and flow rate of pore fluids. Fortunately, the conditions at the Perry site and at the Satsop site are quite similar, and the degree of filling of the cracks at each site are comparable. Therefore, the data obtained previously for the Satsop site are a suitable basis on which to estirate the age of the microfractures in the gouge zone at Perry. On the basis of a thorough examination of the microcracks in six representative samples of the gouge and country rock from the fault, or faults, in the intake tunnel and the discharge tunnel and from the fracture zone in the discharge tunnel, it is our conclusion that the time of last movement of each of these faults is conservatively estimated at approximately 1 million years and may be as old as 2 to 5 million years. 3.0 BASIS OF METHOD Displacement of rock along fault surfaces, in the laboratory as well as in the field, appears to produce microfractures. For examples of representative laboratory studies, reference is made to the trork of Griggs and Handin (1960), Conrad and Friedman (1976) , Jackson and Dunn (1974). The examination of natural specimens from faults by Engelder (1974), Swain and Jackson (1976), and Stearns (1972) demonstrates the applicability of the laboratory results to rocx in situ. Work done during the past decade on microcracks (Simmons and Richter, 1976; Richter and Simmons, 1977; Simmons et al., 1975; Bat:1e and Simmons, 1976, 1977; and Wang and Simmons, 1978) has shown clearly that healed and partially healed microcracks in rocks are ubiquitious. Apparently, the microcracks began to heal immediately upon forming. The degree of healing, as measured by the volume percentage of new mineral growth that fills the microcracks, is an indication of the amount of time that has elapsed since the formation of the microcrack. The general form of the function that relates degree of filling to elapsed time, shown on Figure 1, can be deduced in the following manner. The initial rate is low because of the difficulty of nucleation. The final rate is low because the transfer of fluid from residual cavities (i.e., fluid inclusions) must occur by diffusion of water through solid phases. Thus, the functional form of the curve is asymptotic at both zero percent filling and at one hundred percent filling. During the intermediate phase, the rate is controlled by both the availability and fluid phase. Because the rate of many processes is described adequately by an Arhennius-type relation (Kingery et al. , 1976, Chapter 9), we suggest that the rate of sealing of microcracks is described satisfac-torily by in (c/co) = K (t - t o) K =, Aexp(-Q/RT) (1) where c/co is the volume fraction of fi' ling K is the reaction rate ( }fh A-2
t is time A and Q are experimentally determined constants R is the gas constant T is absolute temperature. At the present time, we have two data points that appear to lie on the curve during the intermediate period. They are shown on Figure 2 and are connected with a straight line. Both data points lie in the inter-mediate region because in each case the new mineral growth had extended completely across the open microcracks, but an open channelway still exists throughout the microcracks. Additional confidence is derived from the observation that apparent degree of filling of a 0.2 mybp crack shown by Swain and Jackson (1976, Figure 4) is very small. The data for the low end of the curve were obtained on a sample of sandstone from the Satsop site (Weston Geophysical Research, Inc., 1978). The cracks were produced during the compaction phase of the sandstone. The stratigraphic unit (the "^ntesano formation) that was deposited above the sandstone was datea on the basis of fossils (Rau, 1967) as at leact 1 million years and possibly 2 to 5 million years. Because the creation of compaction fractures must have ceased when the unit began uplift, the youngest age for any compaction-induced fracture must be the age of the youngest overlying formation, approximately 1 million years. The minerals that were examined in the study of the sample from Satsop included quartz, feldspar, and pyroxene. These minerals, as a group, contain Al, Si, Fe, K, Ca, Na, and Ti. The maximum depth of burial was approximately 3,000 te 3,500 feet. The thermal gradient at the site was probably 150 to 250 C/km. Therefore, the maximum temperature to which the sample had been exposed was probably 20 to 30 C, an estimate that is consistent with, but somewhat higher than, the temperature estimr.ted from the metamorphic grade of the organic material contained in the sandstone. The data for the high end of the curve are based on data reported by Padovani et al. (1979) for a series of core samples from the 3,500-foot hole drilled in the Ries Crater, Germany. Tne Ries Crater and the microcracks in the rocks from the Ries Crater were produced when a meteorite hit the earth 18.5 million years ago. The age was obtained with radiometric techniques. Figure 3 shows a typical crack in the mineral amphibole partially filled with grains of the mineral chlorite. Cracks were observed in quartz and feldspar also. The degree of filling was highest in quartz, intermediate in feldspar, and lowest in amphibole. The host grains for the partially sealed microcracks contained the elements Al, Si, Fe, Mg, K, Ca, Na. The thermal gradient at the present time in the Ries Crater is 150 to 25 C/km. Thus, the maximum temperature at present to which the samples in situ have been exposed is approximately 20 to 25 C. The time zequired for nucleation in the cracks in the rocks from the Ries Crater may have been very short. The meteorite impact produced a A-3 }-[k jT79
high temperature associated with the shock waves that lasted a few microseconds to perhaps a few milliseconds. In addition, a significant volume of the rocks in the vicinity of the impact and sampled by the drill would have been exposed to a temperature that might have been as high as 100 to 200 C for intervals of time of the order of hundreds or perhaps thousands of years. The higher temperatures would likely have shortened greatly the amount of time required for the nucleation of the new mineral growths in individual microcracks. We have included the uncertainty of this effect in the error bar that is shown for this data point on Figure 3 by indicating that the degree of filling might appear to be too large for a sample whose age is 18.5 million years, but which used 5 million years for the nucleation time. 4.0 PROCEDURES 4.1 SAMPLE COLLECTION The samples for this study were collected with methods designed to minimize, or perhaps prevent completely, the creation of open micro-fractures in the material which had very low strength. Two different techniques were used. In the first technique, we used a jackhammer to line-drill a large block of rock. The concept for this procedure was that the jackhammer would damage material relatively near the drilled holes which could then be removed and discarded. The procedure, illustrated on Figure 4, appears to have been successful for several samples but was not successful f>r all samples. Some samples simply disintegrated within a few days after collection. In the second procedure, we use' a small masonry saw to remove completely, the specimen from the rock mass in situ. A series of photographs on Figure 5 illustrates the second technique. This procedure, though rather time consuming for large samples, was highly successful. 4.2 SPECIMEN PREPARATION The rock and gouge while _in situ contain free water in the cracks and pores. The examination of the material in the SEM requires that the free water be removed. Therefore. a major problem in the preparation of the sp2cimens for the examination with the SEM is the removal of the free water without creating open microfractures or destroying any deli-cate structures that existed in the microcracks while the material was still in situ. This problem appears to have been overcome completely in our specimen preparation (as judged on the basis that no open microfracture without new mineral vrowth was observed and that many microcracks with delicate structures of new mineral growth were observed). We used Buehler isomet diamond saws operated at very low speeds, drying furnaces kept at temperaturas below 45 C, and epoxies that cure at room temperature. 4.3 SEM PROCEDURES The procedures for the examination of specimens in the scanning electron microscope are described for general specimens by Hearle et al. (1972) and for rock samoles by Simmons and Richter (1976), Richter and Simmons dm
(1977), and Batz1c and Simmons (1976, 1977). We include here only a brief description of the procedures. The SEM consists of an electron source, focussing and rastoring coils, a moveable stage for supporting the specimen, various detectors, and associated electronics for amplify-ing, displaying, and recording the detected signal. The major systems of an SEM are shown on Figure 6 schematically. A typical image is shown on Figure 7 Unlike a photographic image, the SEM image is generated sequentially in time by the detection and recording of the intensity of the image at individual points. The intensity is controlled by the composition of the material at the point, the topographic rougimass of the surface of the material at the point, and (to a lesser extent) by the crystallographic orientation of the material at the point. The detector in the scanning electron microscope may be sensitive to secondary electrons, backscattered electrons, or x-rays. Most of the work done on the Perry samples was done with secondary electrons or with the x-ray detector. With the x-ray detector, one also uses associated electronics to measure the energy spectrum of the x-rays that are emitted by the specimen. Because each element produces x-rays with characteristic energies, the spectrum of energies can be used to obtain semiquantitative estimates of the elemental composition of the specimen. Typical spectra are shown on Figure 8. 5.0 SAMPLE LOCATIONS Representative samples of the various faults were collected from the intake tunrel and the discharge tunnel. Samples of the fracture zone in the discharge tunnel were also collected. The sample locations are shown on the intake and discharge tunnel wall maps (Figures 17, 18, and 19) of the main body of Weston Geophysical's text. 6.0 RESULTS
6.1 DESCRIPTION
OF GOUCE The gouge zone contains lithic fragments set in a matrix of clay-sized (1 to 4 microns) grains. A typical image is shown on Figure 9. The texture and minerals of the lithic fragments are identical to those of the adjacent country rock. The gouge matrix contains the same clay mineral (illite) as the country rock and also contains gypsum and felds-par. Crystals of Nacl, observed in the gouge zone but not in the country rock, are believed to have crystallized from saline water after collection. 6.2 MICROCRACKS Two types of cracks were observed in the samples from the Perry site. One set, termed desiccation cracks, was produced during the drying of the specimen and appears to be unavoidable. The other set, termed fault-cracks, was not produced during the drying of the sample and appears to have been produced by the last movement of the fault. Desiccation cracks had been observed previously in other samples. On Figure 10, an example of desiccation cracks in clay-like minerals (chlorite in this case) are shown for a specimen
- described by Wang and Simmons (1978). These cracks developed during examination of the specimen
) \
with the SEM. They were actually observed during the time that they formed; hence, their origin is known unequivocally. Desiccation cracks have distinct enaracteristics: (1) they are relatively wide in comparison with their lengths; (2) their walls are very irregular, but opposite walls would fit exactly when restored to the contacting position; (3) they are relatively short (typically a few microns); and (4) they are often - curved. The criteria for the recognition of desiccation cracks are unambiguous. An example of desiccation cracks in the Perry samples is shown on Figure 11 and may be compared with the cracks on Figure 10. Examples of the other type of cracks observed in the Perry samples are shown on Figures 12, 13, and 14. These cracks are typical representatives of approximately 350 cracks that were examined in the Perry samples. Every individual crack in the set of 350 cracks contained new mineral growths that spanned completely the fracture. No open microcrack without new mineral growth was observed - except, of course, the desiccation cracks. 6.3 AGE OF MICROCRACKS The age of the microcracks can be obtained from the degree of filling, approximately one percent. The value is the same as the value for the cccpaction fractures in the sandstone at the Satsop site. If the factors that control rate of fracture filling are approximately the same for the two sites (as they are), then the ages of the cracks are the same. The factors are compared in Table 1, and we conclude that they are quite similar for the two sites. Therefore, the age of the microcracks associated with faulting at the Perry site is approximately 1 million years. Although our estimates of the several parameters that ar'ect the rate of healing of microfractures are similar for the Perry and Satsop sites, they are not identical. Therefore, some possible error exists in the estimate of the date of last fracturing for the Perry site. In our opinion, and based on our experience of working on microcracks in a variety of rocks during the past 10 years, the date might be as young as 0.8 nillion years and as old as 5 million years. 6.4 SLICKENSIDES 2..ples that contained slickensides were examined with the SEM. A typical image is shown on Figure 15. The grooves appear to have been created by grains of pyrite that were embedded in a surface that moved with respect to another adjacent surface. The mineral pyrite was identi-fied on the basis of elemental composition (FeS) and crystal morphology (octahedra). 7.O REFERENCES Batcle, M. L. and G. Simmons, 1976, "Microfractures in Rocks from Two Geothermal Areas," Earth Plante. Sci. Lett., 30, 71-93 Batcle, M. L. and G. Simmons, 1977, " Geothermal Systems: Rocks, Fluids, Fractures," in The Earth's Crust: Its Nature and Physical Properties, Geophys. Monogr. Ser. , Vol. 20, edited by J. G. Heacock, AGU, Washington, DC, 233-242.
\0
Conrad, R. E. and'M. Friedman, 1976, " Microscopic Feather Fractures in the Faulting Process," Tectonophysics, 33, 187-198. Engelder, J. T., 1974, "Cataciasis and the Generation of Fault Gouge," Geol. Soc. Am. Bull., 85, 1515-1522. Griggs, D. and J. Handin, 1960, " Observations on Fracture and a Hypothesis of Earthquakes," in Rock Deformation JA Symposium), Geol. Soc. Am. Men. 79, edited by D. Griggs and J. Handin, G.S.A., New York, 347-364. Hearle, J.W.S., J. T. Sparrow, and P. M. Cross, 1972, The Use of the Scanning Electron Microscope, Pergamon Press, New York, 278 pp. Jackson, R. E. and D. E. Dunn, 1974, " Experimental Sliding Friction and Cataclasis of Foliated Rocks," Int. J., Rock Mech. Min. Sci. & Geomech. Abstr., 11, 235-249. Kingery, W. D., H. K. Bowen, and D. R. Uhlmann, 1976, " Introduction to Ceramics," 2nd Edition, Wiley, New York. Padovani, E. R., M. L. Batzle, and G. Simmons, 1979, " Characteristics of Microcracks in Samples from the Drill Hole Nordlingen 1973 in the Ries Crater, Germany," Proc. Lunar Sci. Conf. 9th, in press. Rau, W. W., 1967, " Geology of the Wynoochee Valley Quadrangle" Wash. Div. Mines and Geol., 56, 51 p, 1 pl. Richter, D. and G. Simmons, 1977, "Microcracks in Crustal Igneous Rocks: Microscopy," in The Earth's Crust: Its Nature and Physical Properties, Geophys. Monogr. Ser. , Vol. 20, edited by J. G. Heacock, AGU, Washington, DC, 149-180. Simmons, G., R. Siegfried, and D. Richter, 1975, " Characteristics of Microcracks in Lunar Samples," Proc. Lunar Sci. Conf. 6th, 3227-3254. Simmons, G. and D. Richter, 1976, "Microcracks in Rocks," in The Physics and Chemistry of, Minerals and Rocks, edited by R.G.J. Strens, Wiley-Interscience, New York, 105-137 Stearns, D. W., 1972, " Structural Interpretation of the Fractures Associated with the Bonita Fault," in Guidebook of East-Central New Mexico, edited by V.C. Kelley and F. D. Trauger, New Mexico Geological Society, 161-164. Swain, M. V. and R. E. Jackson, 1976, " Wear-like Featuras on Natural Fault Surfaces," Wear, 37, 63-68. Uang, H. and G. Simmons, 1978, "Microcracks in Crystalline Rock from 5.3-km depth in the Michigan Basin," J. Geophys. Res., 83, 5849-5856.
\'l{ % \0 A-7
Weston Geophysical Research, Inc., 1978, " Feasibility of Dating the Faults in the Foundation of WMP 3 at the WNP 3 and 5 (Satsop) site of Washington Public Power Supply System, report prepared for EBASCO Services Incorporated and submitted to Washington Public Power Supply System, 26 pp.
\ >
A-8
TABLE 1 COMPARISON OF PERRY AND SATSOP SITES WITil RESPECT TO FACTORS AFFECTING RATE OF FRACTURE IIEALING Factor Perry Satsop IIost Minerals Illite (based on EDX) Quartz, Feldspar, Pyroxene Y Elements in llost(s) Al, Si, K, Fe Na, Mg, Al, Si, K, Ca, Fe Elements in Growth Minerals Al, Si, K, Fe Not measured Maximum Temperature 288 to 293 K 288 to 293 K Maximum Lithostatic Pressure 300 bars -300 bars Width of Microcracks 1 to 5 microns 1 to 5 microns A
+
J
O Cc Nc % r4%~ E: o&R la 8 c$$o D' o O g F i g u r e 1 hsiH e hs e aa a lp s l e d .ehn eg ci c m oTao mhtf pei l cm ec O i trac eanr lcdo yk c sir a na T I tatc rek M ctensd E
- cev rde e r aos tnu els dy t
ati tom e ts. ih mo ewT h tg e oe nc aeu nrr dav le a r e O
zBlfi2! O DEGREE OF FILLING a I o C N e MO (D C o X9 0 M5 %3 OM 5 a Q. Cp O
- M M CW B: em o :r m -
3 (D _ mm O ow MC % m- O. (D o 4 tu M 4 sa 4 OO g 'gs on W7 1 CU M &
- X (30 to 50% )
m 3 - cD CD O
; $O ; n c-D:J. . ;,e I - ?
he; ~ x, L.
...-cge.
N
\ .- ?
as Figure 3 Partially healed microcracks from Ries Crater, Germany. SEM micrograph. Host grain is amphibole. New mineral growth is chlorite. The sample is described by Padovani et al. (1979).
~_.
M n
,, ' R :Q ':
_1~ n
@E ;rd st-g;)
fi$ '~l[;()t wu - m 31 ll
- 4. y,. .c - -
.u . < _M % : , + .
Figure 4 Sample number 4, partially outlined with holes that have been drilled with a jackhammer, still in situ. Note the gouge zone that is contained in - this sample. The webs between the individual holes were later removed. A chisel was used to split a bedding plane at the base of the sample, freeing the sample completely.
) i \
k w n.v
.., g , 6 .y ' '.
x + e
, o4 , Y{k.'> />
ZG
~
gv? s,'k' ' p .-
-, r, : . , .s , ::' :c.
- d. w .y Figure 5 Sample number 25, partially sawn, still in situ.
After the rear cut had been made, th e s aiiiil e wa s freed by splitting gently along a bedding plane. Note the gouge zone that extends diagonally across the sample.
-. so m o. .m us siss .un.m,x ,m,, si l._
O~.-k rn ._ D- - -
~ ,
re -gliFh_\/.. t i s_ -._ i w~ It I
>l s-. - 'Q-
_ao - m.u.,_._. I233 I8Cl Figure 6 Block diagram of a typical scanning electron microscope. The image on the cathode ray tube is recorded photographically. An x-ray detector may be substituted for the collector.
g $
, . f%.
a -+;. 7 ' %
, -~, , g ~
se s k. 1 f,, f Figure 7 Typical micrograph obta ined with a scanning electron microscope. PNPP sample 1 This image was made with a specimen from the gouge zone in the intake tunnel at the Perry Nuclear Site. The deformed crystals near the center of the micrograph are probably gypsum. 7., . Q,,-& T [ 4 :l
$[
s h ifi <
'l
- d!
" l I{ ! , . J 8 ,i ~
j l \, \ *
- i h '
4 . , t j f.Y0- ~ . _ - - Figure 8 Typical x-ray spectrum obtained with energy dispersive systems. This spectrum was obtained from clay minerals in the gouge zone. The abscissa is enercy (of x-rays) and the ordinate is counts per channel. The identification of the individual peaks is shown on the figure. The peak for copper is due to contamination within the system and not to the presence of copper in the specimen. The mineral is probably illite. 12 190 6
j 9a _ COUNTRY ROCK
<'a c ..- i G0UGE . .: . - . . + ZONE r.! - #-k. :I. 3 .+: , jjp;..
k'p% ) -
"I //-/j. k * .a !!!!j g; ,,lllIlit ggp, yf, g., . /2 I" inops, ,'npu o p
itlol.l o 12 9b ,
, j, O' COUNTRY , / . ROCK ,' ._ , , j. , o. ;; ; .. ; U : a m -. - ];;+;;Zi.v = =1(.:l:C ^ ,y. p -Q)yj%7;tDG;g;. ;
Mj!(/N's NITSSN v:
. '. i F G ?, w 497h' tcr M,e d. n-( G0UGE g5~ - Z0NE 7;:g$[I,9 dh. 5c kN -
1 Figure 9 Figure 9a is an enlargement of an optical photograph and shows in small scale the many features that are present in the gouge and can be readily recognized on a freshly sawn surface. Note the abundant lithic fragments of shale that a re set in the fine-grained matrix. Figure 9b is an SEM microgra h of the gouge (area differs from 9a .
\
P00RBRMAl-7 mr& &2 vi M.,qs*.. s;:N WE5;;WrQ,[W- v.y
,- -C a . - p m j p s,. g g:77 3 ; .1.u ~, k .: ~.
men:n
- -:ui:g~L;w. 4' -@ Q'%p
,, 7-i % = ,M,'
ew w:~y; .,s St
%.w s p:
s -<
~s~ ,;..x A:.-m w ,,,,v." r< . p mwa,,p .<*" %.c -l
- Q,.
- + . ,' fi$
yam mt,_ys , . ..,o , 3m s;g :. g.' ;, , g' . pf m.cm u. : W,. m .ce n . . e .
~ .s;%2l> c%: y, g. y.s,. . '? ' %,<<,. .' , ,%. w 4, n. ,Q2L, ti .. o ,'~ '
_ - - - , f ' ,% 3,,',$%g ;%( g- a*
' ' - - i w ay 'p' m %ae-p ; 7 [, , %: j ygg ..;p' 1m. . <.- ~a :-;g 3:, ..
W.. : = . ... . w,
~ ,s Oc .'UE N. :yg y ; C,V .4 , e.QO.$- :.AR h, ,[z-;wr" o u ~, .'.~.c <w,f:' & ._ y ; 'y, ', %s. . . , .. *g # :'%;, .g', ,;' **. i ;w. q4lPrdgg4;.,, ..,, ,
b _ .m ._ Figure 10 Desiccation cracks observed in a sample of chlorite. These cracks were observad in the SEM during formation. p -y -, . . ;a s- +w. n . Aw ~~ <
~
I. . 3 f &g gv C 13 % $&?$ , 1
.g . < e_ . ,,.
Q1- g 4.- y- -
,. . 4 * . 's , *"gr 'a > .. i . . .
o ;f,9,>, , .a -. t Et ( 8-?/0 & ~
'$)t, -
i 4 f ,1
) ,}y R}
1nn7L Figure 11 Crack produced during collection or specimen preparation.
ld - 12a 12b < -
. c_ y 6
_% .m ,, Q nye,, 'f ps -
- M_., }' .em 12b ,
41>
. o . ; ._~ '% 'y G , k ^ ~ ~g __ . = . gggpf A 'gj % c+e- -
wq-
,v, n - .,-ord J4 % 4,;.
Q'- J
. , ~ .
(%3.,# & z , _ .. -. ,
- f. . y .-
% ~2J g \
Figure 12 A typical microcrack in the Perry samples. This crack occurs along the boundary between the gouge zone and the adjacent country rock. The enlargement (12b) shows that new mineral grawth has occurred with the crack, an indication that the crack was not disturbed during the collection and specimen preparation.
P00R ORIBEL 13a y- - ny& -
)
n ,, .~ .. Q! ' w- .' . '
-,'2 13b ~ -n -
t, wwyy . N kl. Y. 51sl.
.3b y u gc. . . ". ,, ,. 9' . ??) ' ,,' '
i ' ,. ,.f
.L
{
%O '$ > 2 .q 4 Figure 13 Microcrack in Perry sample. This variant for the microcracks in the Perry samples is relatively short and contains new mineral growths that span completely the microfracture. Note that many crystals can be seen projecting into the crack from the walls'(136).
12Q194
P00RORlBlhA-14a .w . . I
^~
2, 14b ma f cat on 14b . 1233 195
P13R:PJGINAL , L 15c
' ' ;p j
f . ,
+
(y. t
'Q) .; ! , ' ,;c . , u . . m ,.. : ,F q,' *l, , Vt f( . , ' t, - ., ,jf 'dyd{ ~* '# ; ,{ , .. )Q s 'j'?gy.
y ;; ,.
.) a t
yj. '. - st . .,, 1}. y N.,, , q, ,.' c %. {f. ,, ' ' ' '
^
j
.+ '. ,., ,;} ' . , ,. , 6 . i f ~ ,, , j ', . "q. . ' .;.,'_,,i..
Figure 15 Slickensides in the Perry samples. The slickenslides appear to have been formed by pyrite grains that are stronger than the shale. In 15b, and 15c individual grains cf pyrite can be readily observed. I 12kk I% t
300R ORGIL \ l
. . r , ;.-Ff ' i f l*g. .c-* * '
15a . . Lm 6.M jff h l?[ ~'{:' ef*, f e m. xya e ,. .-. -
#-a-...
py , v ,.,, ;; pg/ * * . ,- . . ' . . 15b ~ e '- 1 . 1 - .
$.s%<:W+;.mg;D,.,*'*
i?,* r , , ,n '.s.i.V,,L. . F _efi,75 g.-y ~,..r. . . -. .a . . . I # 4 ' g + .[' .* .
, en A W'M,Qir$N%y: ' jtn . "
(,?Yfm hyA*: m y.:'._
- 4. ~*p l
~
- . Ti&.f :f ' :
15b f.Ow .' <9:
%Q'" -. .
O ' '" JcJ.;.f:f _
, i, \ .* b..;,G, , ;.J .
- .'o,. .
p , J , *!,' .l
, A ~ .8.% , . . . . a r. e'-
A 15c ' ' ' "* * 'f - ' . ';. *' A ,i' ' an' ?
~
G
, e r 9' ~ , V,V'*
f 7,'.
,f 'q '
N. ) 1 .s . g' a 9 . e. . .- ' .M. , - -
.Y.,,lQ .$..y;v'i. ;f1'r $%'! .f,p ? ,1 j . ' . Y$ = 7 L ,1' '[.
((a ,5 b/ i,.- */,
,..s (iM.','4[>I.'3.[, / ' , ,/, f 3' N .' Lg (* ...., f J yg,,*jc.. '?
a., k
.. ;c l' ' [' .'s I4 j ] iT i .) b *I7 fT 7
124$198 l
~
E
APPENDIX B STUDY OF THE ISOTOPIC COMPOSITION OF WATER FROM TIE FAULT IN TIE INTAll AND DISCHARGE TUNhTLS AT THE PERRY NUCLEAR POWER PLANT by Dr. Gene Sinunons April 1979
\
APPENDIX B STUDY OF THE ISOTOPIC COMPOSITION OF WATER FROM THE FAULT IN THE INTAKE AND DISCHARGE TUNNELS AT THE PERRY hJCLEAR POWER PLANT 1.0 INTRODUCTICN A small fault was intersected by the intake tunnel for emergency cooling water at the Perry Nuclear Power Plant site. A small fault was also intersected by the discharge tunnel at the approximate location expected from the projection of the fault in the intake tunnel. Water enters each tunnel in the vicinity of the fault and its isotopic composition may be a useful guide to the vertical extent of the fault. 2.0
SUMMARY
AND CONCLUSIONS The isotopic ratios of D/H and 180/100 were measured with a mass spec-trograph for three samples of water from the fault in the intake tunnel, one sample from the fault in the discharge tunnel, and two samples from Lake Erie. The three samples from the intake tunnel differ insignificant 1y from each other and from the sample from the discharge tunnel. The two lake samples differ insignificantly from each other. However, the waters from the fault (s) differ significantly from the lake water. All three samples are meteoric. The interpretation of the present set of data is that the ' fault water' is not Lake Erie water. 3.0 BASIS OF TECHNIQUE The isotopic ratios of Deuterium to Hydrogen (D/H) and of Oxygen-18 to Oxygen-16 (180/16 0 ) in water have been shown to depend on the source of the water (e.g. , Epstein and Mayeda,1953; Craig,1961) . The ratios are measured with a mass spectrometer. Experimental details of the measuring techniques are given by Epstein (1959). The ratios are normally reported by differences relative to a standard defined by Craig (1961) and termed SMOW, an acronym derived from standard mean ocean water, where 16 18 16 (18 / I spl-( O/ 0)3ggg 30fog 618 0= x 10 (160f
/ 6 ) 0SMOW (D/H) spl- (D/H) SMOW 6D = X 10 30/ 00 (D/H)SMOW $
and the subscript spl indicates values of the sample. 9OO b [ B-1
Craig (1961) showed that the isotopic variations in meteoric waters could be represented by the equation 6D = 8618 0t 10 Figure 1 is a plot of his data. Clayton e_t al. (1966) examined the isotopic ratios of saline waters from several sedimentary basins. Their data are summarized on Figure 2. 4.0 DATA AND DISCUSS G The isotopic ratios relative to standard mean ocean water, SMOW, are given in Table 1. They are also shown on Figure 3. TABLE 1 6DSMOW 6 O SMOW SAMPLE (0/00) (0/00) F1 -73.3 0/00 -11.5 0/00 F2 -73.5 0/00 -11.4 0/00 L1 -54.0 0/00 -7.4 0/00 L4 -52.3 0/00 -7.6 0/00 IF4 -70.6 0/00 -11.7 0/00 FD10 -79.3 0/00 -11.4 0/00 The isotopic ratios of all three water samples are near the Craig (1961, curve for meteoric water. Therefore, the water from the fault is meteoric water. The ratios for F1, F2, and IF-4 are very close to each other. If we take the differences to be an indication of experimental precision, then the isotopic ratios for the water from the fault in the discharge tunnel differ from the vTlues for the intake tunnel by approximately the experimental error. We therefore conclude that the waters from the fault (s) in the two tunnels have a common source, which is not Lake Erie. The data are consistent with a single fault intersecting both tunnels. { L E-2
The ratio of the water from the fault differs significantly from the ratio of the sample of Lake Erie water. Sample L1 was collected near the lake surface, L2 near the bottom. Both samples were obtained near the projection of the fault in the intake tunnel dip to *he lake bottom. On the basis that the isotopic ratios of the waters from the fault in both tunnels differ greatly from the ratio for water from Lake Erie, we conclude that the fault water is not Lake Erie water.
5.0 REFERENCES
Clayton, R. N., I. Friedman, D. L. Graf, T. K. Mayeda, W. F. Meents, and N. F. Shimp, 1966, "The Origin of Saline Formation Waters," J. Geophys. Rer., 71, 3869-3882. Craig, H., 1961, " Isotopic Variations in Meteoric Waters," Science, 133, 1702-1703. Epstein, S., 1959, "The Variations of the 018 79 16 Ratio in Nature and Some Geologic Implications, in Abelson, P. H. (editor) ," Researches in Geochemister, John Wiley and Sons, New York, 217-240. Epstein, S. and T. Mayeda, 1953, " Variation of 018 Content of Waters from Natural Sources," Geochimica et Cosmochimica Acta, 4, 213-224. Faure, G., 1977, Princieles oi Isotope Geology, John Wiley and Sons, New York, 464 pp. t 12i! 202 B-3
P00RO!!iG!Na d a . .g a a a g ....,e aa .,.e ..g ... , , , . ,
/* 1 p .-.,.i , _
q s
.aoc w g ,f o '%,,,,/
t - - IOC P- - h , 33'S10' 53 ., b : Soc f-- - F. ................ . . . . . . . . . . . . -
-to ea 33 la o o e -23 8O't%)
FIGURE 1 Deuterium and oxygen-18 variations in rivers, lakes, rain, and snow, relative to ' standard mean ocean water' (SM0W). Points which fit the dashed line at the upper end of the curve are rivers and lakes from East Africa. (After Craig, 1961) s , . . . . . , . . . . s
=
4 - -
~
2- .se
~
n .
} ye d' WW N ,
3e g*a 9' _ : *
- g -2 -
g_ . - o - -
- e n
** -4 - , ~ ,~,+ d "_G q s.p p . _- , ~* , gy o3 s . -6 - , e . - 6.- , 3 , o Imrcs . -C * * ##7 o, e 1,:t C,0 cst .
12 -- ,e,
- AM3 -
t, , , . . . , . . ,, 16 12 -lo 8 -6 -4 -2 o 2 4 6 8 D 3 o'
- FIGURE 2 Isotopic compositions of brines. The ' meteoric waters' line is the line determined by Craig (1961) and shown on Figure 1. (After Clayton et al., 1966) l
\ 'l .
90b B-4 '
8D( /00) -4 0- *L4
- LI .IF-4 FI,F2@ FDIO . 8D = 88 '8 0+10
-13 -12 - 11 -10 -9 -8 -7 8'8 0 (0/00)
Figure 3. Isotopic compositions of the waters from PNPP. F1, F2, and IF-4 denote samples from the fault in the intake tunnel. FD10 denotes water from the fault in the discharge tunnel. L1 and L4 denote water from the top and bottom, respectively, of Lake Erie. r B-5 {tL-
') * -I ~
1 i
D e
APPENDIX C GEOPHYSICAL METHODS prepared by WESTON GEOPHYSICAL CORPORATION April 1979 12fp 206
APPENDI.Y C GEOPHYSICAL METHODS
1.0 INTRODUCTION
The following sections discuss the geophysical techniques employed during the investigation of the fault discovered in the intake and discharge tunnels at the Perry Nuclear Power Plant sita. These tech-niques include magnetics, gamma radiation, logging, and in-situ velocity measurements. 2.0 THE MAGNETIC METHOD The magnetic method is a versatile, relatively inexpensive, geophysical exploration technique. Magnetje data can be acquired on land, over water, or in the air. Aeromagnetic surveys and deep water marine studies are commonly used as a reconnaissance tool for tracing large-scale geologic structure, especially basement depth. Land and coastal water marine data are more useful in tracing smaller, more localized geologic structures, such as mineral and ore deposits, and for detailed geologic structural modeling. Land and coastal water marine surveys yield more detail and higl.er resolution, since the measurements are taken closer to the anomaly source. Land magnetic data can also be used to locate buried, man-made structures such as pipelines and tunnels, and for archaeological prospecting. 2.1 EARTH MAGNETISM Magnetics, like gravity, is a " potential field" method. For a given magnetic field, the magnetic force in a given direction is equal to the derivative of the magnetic potential in that direction. The source of the earth's magnetic potential is its own magnetic field (F) and the inducing effect this field has on magnetic objects or bodies above and below the surface. The earth's field is a vector quantity having a unique magnitude and direction at every point on the earth's surface. This magnetic field is defined in three dimensions by angular quantities known as declination and inclination. Declination is defined as the angle between geographic north and magnetic north, and inclination is the angle between the direction of the earth's field and the horizontal. The earth's total magnetic field is measured in " gammas" (Y) (where 1 gamma = 10 -5 Oersted) and varies from about 25,000 gammas near the equator to 70,000 gammas near the poles. The earth's magnetic field is not completely stable. It undergoes long-term (secular) variations over centuries; small, daily (diurnal) variations (less than 1% of the total field magnitude); and transient fluctuations called magnetic storms resulting from solar flare phenomena. L.
\
C-1
The earth's ambient magnetic field can be modified locally by both naturally-occurring and man-made magnetic materials. There are two types of magnetism involved: induced and remanent. In the case of induced magnetization, the earth's ambient field is enhanced by materials which can behave like a magnet when an external magnetic field is applied. Crustal rocks become " magnetic" due to the presence of magnetic parti-cles, usually magnetite or related iron oxide minerals, in their ccm-positional structure. These particles act as small dipoles, which can be uniformly oriented by an external magnetic field, making the host rock " susceptible" to magnetic induction by the earth's field. These " susceptible" rocks (or any magnetic objhet) will thus receive an
" induced" magnetic field (E) , which represents a local perturbation in the main earth field. The net field (f t) in the vicinity of this perturbation is simply the vector sum of the induced and earth fields.
Although the induced field is not necessarily parallel to the ambient field, for cases where lHl<.25 lEl,whichisgenerallytrueformost geologic applications, the directional difference between the net field (ft), and the ambient field (f) is negligible. Thus, the induced field really serves to enhance the ambient field. The degree to which the ambient field is enhanced is a function of the " susceptibility" of the material, or its ability to act like a magnet. Remanent magnetization is produced in materials which have been heated above the Curie point allowing magnetic minerals in the material to become aligned with the earth's field before cooling. The remanent field direction is not always parallel to the earth's present field, and can often be completely reversed. The remanent field c3mbines vectorially with the ambient and induced field components. The contribution of the remanent components must be considered in magnetic interpretations. 2.2 INSTRUMENTATION At present, the most widely used magnetometer is the " proton precession" type. This device utilizas the precession of spinning protons of the hydrogen atoms in a sample of fluid (kerosene, alchohol, or water) to measure total magnetic field intensity. Protons spinning in an atomic nucleus behave like tiny magnetic dipoles, which can be aligned (polarized) by a uniform magnet'a field. The protons are initially alfened parallel to the earth's field. A secord, much stronger magnetic field is produced approximately perpendicular to the earth's field by introducir.g current through a coil of wire. The proto.:s become temporarily aligned with this stronger field. When this secondary field is removed, the protons tend to realign themselves parallel to the earth's field direction, causing them to precess about this direction at a frequency of about 2,000 hertz. The precessing protons will generate a small electric signal in the same coil used to
\ '2 s. ?08 C-2
polarize them with a freq,3ncy proporti,nal to the total magnetic field intensity and independent of the coil orientation. By measuring the signal frequer.cy, one can obtain the absolute value of the total earth field intensity to a 1 gamma r ,c" The total magnetic field value measured by the proton precest n atometer is the net vector sum of the ambient earth's field and any local induced and/or remanent perturbations. The total field proton precession magnetometer is portable er . not require orientation or leveling, as,is required with vertic__ instruments. There are a few limitations associated with the .assion system, however; the precession signal can be severely degraut2 .n the presence of large field gradients (greater than 200 gammas per foot) and near 60-cycle A/C rc<er lines; also, interpretation of total field data is semewhat more cor.s icated than for vertical field data. 2.3 FIELD TECHNIQUES In the field, the operator must avoid any sources of high magnetic gradients and alternating currents, such as power lines, buildings, and any large iron or steel objects. The operator should also avoid carrying any metal articles. Readings are taken at a predetermined interval which depends on the nature of the survey, the accuracy required, and the gradients encountered. Base station readings, if required, are usually made several times a day to check for diurnal variations and magnetic storms. Depending on survey recuirements, one shoald deternine tne magnetic susceptibility and remanent magnetism for the rock units in the survey area. If this information is not available, several representative rock samples should be collected and analyzed. One must properly mark the in-situ orientation of these samples with respect to north direction and horizontal plane. Susceptibility and remanent field measurements are obtained using standard laboratory techniques. 2.4 INTERPRETATION Lateral variations in susceptibility and/or remanent magnetization in crustal rocks give rise to localized anomalies in the measured total magnetic field intensity. Geologic structural features (faults, contacts, intrusions, etc.) which correlate with susceptibility and/or remanent magnetization variations will cause magnetic anomalies, which can be measured and interpreted to quantitatively define the geometry of this causative structure. After Ciurnal effects and regional gradients have been removed, magnetic anomalies can be studied in detail; derivative operations and frequency filtering can be employed. Because it is a potential field method, there is an infinite number of possible source configurations for any given magnetic anomaly. There is also an inherent complexity in magnetic dipole behavior. Remanent field
'i c-3 .12! j 209
effects further add to the complexity. But if the various magnetic field parameters (inclination, declination, and susceptibility) are well defined, and some reasonable assumptions can be made regarding the nature of the source, an accurate source model can generally be derived. Magnetic anomalies can be analyzed both qualitatively and quantitatively. The physical dimensicas of an anomaly (slope, wavelength, amplitude, etc.) often reveal enough to draw some general qualitative conclusions regarding the causative source. Precise interpretation must be done quantitatively, however, and there are two basic approaches, each ideally requiring prior knowledge of earth and remanent magnetic field parameters. Modeling can be performed by various approximation methods, whereby one reduces the source to a system of poles or dipoles, or assumes it to b'e one of several simple, geometric forms (vertical prism, horizontal slab, step, etc.). The magnetic properties for this simplified model can be rather easily defined mathematically. Simple formulas can be derived which relate readily measurable anomaly parameters, such as slope, width, and amplitude ratios, to the general dimensions of the anomaly source, including depth to top, thickness, dip, and width normal to strike. Since these methods involve very limiting geometric assumptions, the results can only be treated as good approximations except for very simplified sources. The second and more accurate quantitative method utilizes computer iteration techniques to directly calculate the resultant magnetic anomaly for a two- or three-dimensional geometric model constructed to fit the expected geologic source. This method allows one to develop by trial and error a model whose calculated magnetic field anomaly matches the observed anomaly as closely as possible. In both two- and three-dimensional computer modeling, the source body is spatially defined by one or more n-sided polygons. In the two dimensional case, a vertical polygon of infinite length in a direction normal to the magnetic profile is used to define the source. Each polygonal segment then represents the vertical edge of a rectangular prism, which is infinitely long in the profile direction. The magnetic effect of each of these prisms is computed and summed with appropriate sign convention to give the net magnetic effect of the body circumscribed by the polygon, and thus, the magnetic anomaly. In three dimensions, a series of horizontal polygons are stacked verti-cally to define the source. The net magnetic effect for the total volume is then obtained by computing the effect of each polygon, integrating it over the vertical extent of the body, and summing the results for all of the polygons used. The polygonal geometry allows a great deal of flexibility in defining an anomaly source and can encompass a wide range of geologic forms. Ui
\L.h oL0 f
C-4
3.0 GAyy.A RADIATION LOGGING 3.1 PURPOSE AND EACKGROUND Gamma radiation logging can provide an efficient method for correlating geologic units in uncored boreholes. The logging probe measures gamma radiation resulting from the natural radioactivity of the uranium (U), thorium (Th), and potassium (K40) in nearby bedrock or soils. Although the radiation from either the U or Th series is much greater than that of K40, the background radiation from each element is approximately equal because the potassium isotope is far more common. The intensity of gamma radiation decreases rapidly as it passes through a material. This attenuation is exponential a_nd dependent on the energy of the radiation and the absorption coefficient of the particular material. For the average energy of natural radiation, the range of penetration in sediments is roughly 1 foot with about half the gamma rays detected in the borehole originating within 5 inches of the borehole wall. The natural radicactivity in sedimentary rocks and metamorphosed sediments is generally higher than that in igneous and other metamorphic types, with the exception of potassium-rich granites. In sediments, the gamma ray log reflects mostly shale content because radioactive elements tend to concentrate in clays and shales; sands and carbonates usually have low radioactivity. Subtle changes in rock composition not readily apparent to the inspecting geologist may be revealed by the gamma ray log. 3.2 EQUIPMENT AND PROCEDURE The logging system consists of a probe containing a scintillation crystal and photomultiplier tube, an electronic counting unit, a strip chart recorder with variable scale settings, and a power winch. Gamma radiation incident on the scintillation crystal is converted to light through interaction with the crystal. This light enters a photo-multiplier tube where it is converted to a pulse of electricity which is conditioned and transmitted thro ~ ugh the cable to the counting unit. The average number of pulses per time unit (seconds or minutes) is plotted versus depth on the strip chart recorder. In logging, the probe is lowered to the bottom of the borehole and measures the radiation as it is raised. Boreholes are generally logged twice to determine the " repeatability" of the data. Statistical variations in radiation emission, significant at low counting rates, can generally be smoothed out by integration over a short time interval. If the hole is logged too quickly, however, the smoothing effect leads to erroneous results, and data are shifted in the direction n.
\' h_ / \\
C-5
of logging. The logging speed must be adjusted for the bed thicknesses and radiation levels. The length of the detector (the scintillation crystal) with respect to the bed thickness also affects the shape of the resulting log. Optimum resolution for thin beds is obtained with a short detector and a slow logging speed. 3.3 INTERPRETATION The interpretation of gamma logs is relatively straightforward. The interface between beds of different natural radioactivity can be located with reasonable accuracy if it is assumed to occur halfway between the two count levels for thick beds (<6 ft.). For thinner zones, the location of the maximum count rate can be taken as the center of the zone. In making correlations, all available geologic information is taken into consideration. This includes unit thickness and composition, and position in the geologic column. The ganna ray log displays this information in the form of the radiation level within a particular unit, as well as the gamma ray signature for that unit (the frequency of minor deviations from the average radiation level). If other geophysical information is available, it is also considered in the final interpretation. 4.0 IN-SITU VELCCITY MEASUREMENTS 4.1 PURPOSE In-situ velocity measurements provide a reliable determination of material properties. The velocity measurements together with known or estimated densities are used to determine the dynamic elastic moduli of the material. It is necessary to obtain the data on material in place; velocity measurements made with laboratory samples may be strongly effected by alteration of the material in obtaining the sample, and by differences between the in-situ and tcst-imposed stress conditions. 4.2 EQUIPMENT AND PROCEDURE In-situ velocity measurements are based on the determination of the time required for elastic waves, generated at a point source, to travel to a series of vibration-sensitive devices (geophones or seismometers). For in-situ velocity measurements, usually the geophones contain three orthogonal seismometers, one vertical and two horizontal. These three components allow the seismologist to estirate the mode of vibration of the material in the vicinity of each geophone. Seismograms are obtained using a portable 12- or 24-channel seismograph system which amplifies and filters the seismic signal detected by the individual geophones and provides a photographic record for each of the 12 channels (Figure 1). Timing lines are provided across the entire recording at two-millisecond intervals allowing direct reading to one C-6
millisecond. The seismograph is equipped so that the background noise level can be observed for all geophones simultaneously, enabling the operator to determine if the noise level is sufficiently low to minimize trace interference. Depending on the requirements of the survey and specific site conditions, in-situ velocity measurements are acquired in a number of w ;s, depending upon the deployment of source and geophones (Figure 2):
- 1. source and receivers in different boreholes (cross hole;;
- 2. source in borehole and receivers on the surface (uphole);
- 3. source on the surface and receivers in borehole (downhole);
- 4. high frequency source and receivers in the same hole (sonic logging);
- 5. scurce and receivers in tunnel; or, *
- 6. source and receivers on surface.
4.3 INTERPRETATION The interpretation involves picking the arrival times of two forms of seismic waves at each geophone and determining the relationship between arrival times for each wave type. The two waves are the compressional ("P") wave and the shear ("S") wave. The "P" wave is transmitted as a series of compressions and rarefactions, and the particle motion is parallel to the direction of propagation. The "S" wave, on the other hand, exhibits a particle motion perpendicular to the direction of propagation. Therefore, the information on particle motion given by the three-component seismometers can be used as an aid in determining the wave type of arrivals. When the arrival times are plotted against distance from the source, the velocity of the material is determined by the interse slope of the best linear fit to the data. o* i 2bd)b l \ C- 7
P00R BRIGD!AL
& Vibration From Source Produces Small Voltoge
& 3-Component U Seismometer fkD ( v y ;i ' . l
) i l %- l j . ; ..w -i?. ;r .::.ds .
O: l l
! 4 mplifier.TQ G d M8..,5.M, . i Y:??. D. W. + 5 5. ~ ,
_a _.i
.-:.,n.__.:.:.3:.-x.....O ~m %. . a .: ~
6:a:.-w.cR.' l.;. .p:y;'.L- l I$.;bhE
~w : A- q l
N x < J l
' ' -i.i... - '
l
'N ,
l hNiler -' I
\ , ; Seismograph ' n ? ,. ' I l
h s I l if g Seismic Energy :,b .ic.j; ,
- I Source -Q . . .nq g .
l x I-#
.-a . _ "_ic.d , 6 Camera-- .s l
Q y.. (. ' 3 : .. 2.
~ - p ' .: m .-, . ,f / l g , ;;._ . - -
3 . u ____ __ _.._ ______.-__ _ _ _l 7 SeiS'" SEISMIC INSTRUMENTATION FIGURE 1 C-8 \ -4 h 1)&
A. CROSSHOL E l
. 1 il, , , ,
B. UPHOLE C. DOWNHOLE D. SONIC LOGGING
~ -
c ,
, /
v y c e wx ar x , Y ,
- s e
E SURFACE _ Y Y y 7
'/) Y y l e sessuic enenar sounce \ \
y sElsMOMETER E TUNNEL x \ v gy%%%%g$$%Np M c,suic cuency IN-SITU VELOCITY
/ j /
MEASUREMENTS FIGURE 2 C-9 12@k215
~
s ya h I i 0
=
5 E e
APPENDIX D INDEPENDENT REVIEWS OF COOLING WATER TUNNEL FAULTINC 128I217
Mr. James Murphy Ohio Historical Society 12(0 2iB
Bedrock Deformation in the Water Intake Tunnel, Perry Nuclear Power Plant, Lake County, Ohio James L. Murphy The Ohio Historical Society Columbus, Ohio 43211 On January 19, 1979, I examined bedrock exposures of the Chagrin Shale exposed in a water intake tunnel at the Perry Nuclear Power Plant, Lake County, Ohio. Details of this arid a similar exposure in the outlet tunnel are described in GAI Reports 1986 and 1997, which have been avail-able for study. . Based upon my exanination of the actual outcrop and supplementary evidence presented in the above-mentioned reports, I believe that the low angle thrust fault and related small anticlinal fold are essentially identical with similar features found nearer the surface during excav-ations for the power plant (Gilbert & Associates,1975). It is my belief that such bedrock deformation was caused by the horizontal component of localized stresses created during the Pleistocene by either the advance of the ice sneet(s) and concomitant depression of the crust, or in reaction to removal of the weight of the overlying ice (glacial rebound). This would mean that the deformation occurred some time during the last one million years. I am inclined
- believe that it is related to the last (Wisconsinan) glaciation but conclusive proof of this is lacking. The defomation could be related to any one of the major
\
2 glaciations that covered nmethern Ohio, and different faults and folds may owe their origin to different glaciations. In any case, further movement along such features is not to be expected, and these are not, therefore, classifiable as capable faults. In reviewing the original reports (CAI 1986, 1997) on the deformation exposed in the cooling system tunnels, I would make the following additional statements:
- 1) Based upon my knowledge of similar faults in the Chagrin Formation of northeastern Ohio, I believe that in all probability, two separate faults are represented, one in each tunnel. The chief evidence for this is the considerable difference in strike represented in the two exposures and the rather local nature of similar faults exposed elsewhere in the Chagrin.
- 2) I believe that the hypothetical subsurface projection of the fault (s) shown in Figure 2 (GAI 1997) is incorrect and doubt that the fault (s) extend quite so far, either laterally or vertically. (It should be noted that the vertical exaggeration used in Figure 2, though stated in the figure, gives a somewhat misleading impression of the magnitude and dip of the fault.) Presumed evidence of the extension of the fault seems somewhat equivocal and cannot be taken as conclusive proof of the existance of the fault at the distance ani depth projected. Even were the fault of the size and extent presumed, I believe the proposed glacial mechanism still the most probable cause of the deformation.
- 3) The possibility of penecontemporaneous defcrmation of Une unlith-ified Chagrin sediments is completely out of the question in these instances
>2 12F 3 220
3 and, I think, would immediately be dismissed by any geologist who examines the exposure in the intake tunnel. In this re6ard, I suggest that detailed close-up photographs be taken of the lower portion of the fault as exposed in the (north) east wall, a few feet above the base of the tunnel, where rather large (approximately 3 inches in diameter) fragments of detached Chagrir. shale occur in the fault "6ouge", conclusively demonstrating that the deformation occurred subsequmt t' lithification. ~
- 4) Deformation by deep-sea'ted late Paleozoic tectonism cannot be entirely ruled out of the question as a possible cause of some Chagrin deformation, but it is considered an unlikely possibility in the present instance, particularly in view of the fact that similar faulting and foldin6 (notably in the on-snore NPNPP excavations) rapidly diminishes and disappears with depth. Such is also believed to be the case with the water system tunnel faults. Although they are deeper than previously studied examples in the Chagrin Formation, they are nonetheless comparatively shallow "surficial" phenomena unrelated to deep-seated tectonism.
James L. Murphy February 19, 1979 126 221 D-3
Hewspaper Account of 1818 "Kingston" Earthquake Michael Hansen of the Ohio Division of Geological Survey has given me the following information regarding a newspaper account of aa 1818 earthquake that has been believed to have had its epicenter at Kingston, Ross Co. , Ohio. The Cleveland Berister of March 16, 1819, reprints a news item from the Quebec Carette (no date) stating that "Two severe shocks of an earth-quake were felt at Kingston and its vicinity on the morning of the 7th December. They were accompanied with a runbling noise. The disturbance was not as long as those of 1812 but were equally violent." Since the newspaper iten originally appeared in a Quebec newspaper, it is evident that the earthquake occurred at Kingston, Ontario, rather than Kingston, Ohio. The only known copy of this issue of the Cleveland Recister is at the Western Reserve Historical Society in Cleveland. James L. Murphy
) E. -
D-4
fir . Robert G. LaFleur Rensselas.c Polytechnical Institute 124j223
/
Robert G. LaFleur o r p -l" .' y W 34 .' - Taborton Road FEs !.n - ' Sud La. t, New York 12153 (,;Vil E R .... u... DEPT. - gal
.anuary 30, 1979 tir . I.ane D. Schultz .:: :13. r t Associates Inc. ' . " . I.ancatter Avenue
- p. a.! t ng , Pennsylvania 19603
- . .i r !.a ne :
on January 19, 1979, with L. D. Schultz and J. Murphy, I inspected the reverse fault which intersects the intake tunnel of the Perry Nuclear l'ower Plant, and subsequently reviewed GAI Reports No. 1986 and 1997 describing this feature and other near-surface bedrock deformations. The following comments summarize my impressions of the tunnel fault.
- 1. I see no evidence which suggests the deformation occurred.
while the Chagrin shale was in a poorly consolidated state. Soft-sediment deformation is usually indicated by the presence of flow structure, wispy sediment tails, mess-bedding, deformed and pulled-apart plasts, etc. Early Paleozoic slope clastics and carbonates in the Taconics commonly show such features in deep water rocks - by comparison the Chagrin deformation, with the exception of bedding irregularities attributable to compaction and ninor sole marks, is devoid of such features. Brittle fracture is represented by the tunnel fault. Drag and adiacent open folds maintain good parallel banding. Gouge breccia is angular and untorn. I would conclude from this the tunnel fault deformation occurred after consolidation was completed.
- 2. The depth of active influence, observed elsewhere (200m.),
of overriding ice may be enough to permit inclusion of the tunnel fault in the same glacitectonic category as the shallow features. Ilowever, 1) the fault sole shows no clear sign of passing into bedding plane orientation at reasonable depth; 2) the fault dip direction is considerably at variance with the usual direction of 124Q4224 D-5
~
Dr. L. D. Schultz Jan. 30, 1979 Erie lobe movement (from the NE or N); 3) it seems difficult to see how glacier movement alone would produce a deep structure at all (when the ice can tear up the surface rocks instead) unless there were an existing weakness plane which override could acti-vate. There is no indication in the tunnel that the fault might have a multiple movement history. Elevated methane pressure in the Chagrin would enhance movement along a deep fault, but there is little proof abnormal pressures existed during glaciation. I think it is unreasonable to expect the Chagrin was frozen deeply enough to permit ice expansion along such a weakness plane to motivate faulting. I agree with the proposed glacitectonic origin for both the tunnel fault and shallow deformation, but I am not completely persuaded that active ice, ground-coupled in the presence of permafrost, is necessarily responsible for these features. I cannot rule this process out on the basis of the evidence at hand,' but would point out that it is certain that glacier loading and unloading, glacial quarrying of the Erie basin, and episodic glacial lake development caused vertical stresses and might also have permitted horizontal stress development sufficient to pro-duce the structures. In this sense a more passive role of glaciers in regional crustal movements is indicated. It may be important to this notion that the strike of the Chagrin deformations agrees well with the regional trend of- the Erie basin axis and south edge more than it appears to agree with a direction normal to common ice flow. In addition, deep permafrost, to my knowledge, does not appear to have been widely developed at this latitude during the Late Wisconsin - these glaciers rather were temperate, wet-based, and often advanced through proglacial lakes.
- 3. Although the upper and lower limits of the tunnel fault are not determined, I would expect the fault to intersect the bottom of Lake Erie. The gouge water chemistry does not rule out 126 3 225 D-6
s Dr. L. D. Schultz Jan. 30, 1979 hydraulic connection with the lake. Water movement along the fault should be directed, as recharge, toward the lake. Interpretation in the TX borings of intersections with the fault trace projected' to depth appear reasonable. Absence of mineralized gouge suggests the fault is confined to the Chagrin, but one might also attribute this to a younger (than Paleozoic) age for the fault. 4 One can only conjecture what role che Salina salt played in glacier-induced crustal warpings, and particularly its influence in maintaining and cumulating abnormal horizontal stress. I point this out only to convey the idea that oscillating Pleistocene glaciers may have triggered more complicated " late" tectonic settings in which a ductile substrate influences development of faults in overlying rocks, perhaps like the one exposed in the intake tunnel. In any event I supporc the conclusion that the tunnel fault is related to some manifestation of glaciation - not necessarily as young as Latt Wisconsin. In view of its movement sense, it may be related to crustal loading (down-warping) of the Erie basin while near-surface rocks were in a state of horizontal stress. I see no reason to consider the tunnel fault active, capable, or of post glacial age. Yours truly, W Robert G. LaFleur k2 RGLaF:vb D-7
At the request of Dr. Lane D. Schultz, I visited the office of Gilbert Associates on April 12, 1979 and inspected documents describing the shallow deformations at PNPP and these exposed at Warners Creek and Hell Hollow. I support the conclusions reached by several others that the shallow structural features are the result of glacial ice drag - those exposed at PNPP and also the compressional folds and related thrusts shown in the creek sections. The correctly-oriented fold asymmetry, thrust sense, shallow depth, and participation of bedrock with till are all persuasive features indicative of an active glaciotectonic origin. There is little one can add to the carefully documented and considered opinions offered by C. E. Herdendorf, J. L. Murphy and the Gilbert Associates Staff. As a minor point one might note the occurrence of rare near vertical faulting, illustrated by Hell Hollow faults #1, 2, and 3 which appear to post-date the compressional structures. Comparable relations are not apparent at PNPP; compressional (glacially induced) movement there is the terminal event. If high-angle faulting at Hell Hollow is the result of slumping, one might conclude there is no evidence for fracturing during application or removal of glacial ice load. That is, post-glacial uplift has no structural manifestation at PNPP. If, on the other hand, the Hell Hollow faults are due to post-glacial uplif t, PNPP is still free of such features. My impression that the intake tunnel structure is neither of active or passive glaciotectonic origin remains - although the notion seems plausible that a million-year-old crack along a much older fault zone might be a manifestation of a passive, early glacial event. I see no reason to relate the tunnel fault to the surface structures. It is also clear that the shallow deformations do not resemble pop-ups. Stability of the bedrock since the glacial override seems apparent.
, hf Roberg/C'. LaFleur 4/12/79 t ,
126 3 227 D-8
Dr. Barry Voight Penn State University 12fkl228
BARRY VOIGHT CONSULTANT GF.0 LOGY AND GEO TECHNICS .,7, y n . , , , ,, g , g p gsos s i g t , , si \IE< o!Ie4.t 1 t Ns sii V \ N s te.mi
~1titriloNE mi4> 39- s til L's 1 INVESTIGATION OF COOLING k'ATER TUNNEL FAULTS, PERRY NUCLEAR PO'a?.R PLANT, CHIO 12[{229 D-9
INVESTIGATION OF COOLING WATER TUNNEL FAULTS, PERRY NUCLEAR POWER PLANT, OHIO Barry Voight PAGE
- 1. S umma ry o f Re p o r t - -- -- ----- -- -- - -- -- - --------------- --- 1
- 2. Introduction ------------------------------------------- 4
- 3. Tu n nel Fa ul t De s cri p ti o n ------------------------------- 5
- a. Intake Tunnel -------------------------------- 5
- b. Discharge Tunnel ----------------------------- 8
- c. Extent of Faults ----------------------------- 10
- d. Mutual Geometric Relationships --------------- 16
- 4. Age of Faulting ---------------------------------------- 19
- 5. Rock Stress Investigations ----------------------------- 31
- a. Orientation and Magnitude of Stresses -------- 31
- b. Possibility of Future Slip on Existing Fault-- 41
- 6. Regi onal Tec ton i c Framewor k ---------------------------- 48
- a. Structure under Lake Erie -------------------- 48
- b. Structure of Southwest Ontario --------------- 50
- c. S tructure South of La ke Eri e ----------------- 52
- d. Relationship of Inferred Structure to Seismicity -------------------------------- 56
- 7. O ri g i n o f Tu n n e l Fa u l ts -------------------------------- 66
- 8. References --------------------------------------------- 73 124 230 D-10
FIGURE CAPTIONS
- 1. Drillhole log, TX-4; fault intercept.
- 2. Drillhole log, TX-7; possible fault intercept
- 3. Drillhole log, TX-11; possible fault intercept.
- 4. Fault slip vs. distance along fault plane from tunnel base.
- 5. Direction of a max at the FNPP site in comparison to regional measurements.
- 6. Sketch of PNPP structural trends with hypothetical stress orientations (a) before and (b) after faulting, and (c) measured stress orientations in TX-11.
- 7. Joint orientations PNPP foundation exposures. 220 Measurements.
Plot by WGC.
- 8. og and a2 vs depth, TX-11.
- 9. Stresses in vertical planc perpendicular to tunnel fault.
- 10. Mohr diagram comparing calculated stresses in vicinity of tunnel fault to minimum strength envelopes.
- 11. Structure in the Dover field, Ca1ada.
- 12. " Graben" in salt production shaft, Fairport Harbor, Ohio
- 13. Regional seismicity (WGC base) and structural anomalies.
- 14. Regional seismicity and selected structural anomalies.
12fl 231 D-ll
- 1.
SUMMARY
OF REPORT The tunnel thrust faults represent a single fault with splays or a closely-associated en-echelon set of faults that extends at least 750 ft along northeast strike and at least 600 ft along a 15 dip angle to the southeast. Slip gradient information suggests that the faults die out at elevation 450 ft within about 20 vertical feet above the tunnel crown. If so a toe buttress of " solid" rock about 70 ft thick lies between the terminated fault and the lake bottom. Insufficient information is available to conclusively establish whether or not the faults terminate to the southeast between elevations of 300 ft and 150 ft, or continue to a deeper level. There is no evidence to suggest an increase in dip angle toward the southeast, but the possibility has not been eliminated. Consolidation tests on two samples of the fault gouge suggest a maximum vertical effective consolidation pressure of about 9 + 4 tsf. This value is consistent with vertical compression of fault gouge by a somewhat greater thickness of overburden than exists today, or by minor late Pleistocene ice sheets associated with deposition and com-pression of till deposits recognized at the PNPP site. The gouge consolidation pressure is not consistent with compression by the four or more Pleistocene ice sheet maxima. The latest of these events, associated with the Kent Till, cccurred about 21,000 YFB, with an end moraine 70 miles or so south of the PNPP site and an inferred overburden pressure on the order of 100 tsf. Local arching effects are r.ot considered so severe as to preclude such an event from leaving a marked imprint on gouge consolidation characteristics. It is therefore considered likely that the last movement of the tunnel fault occurred not more than 20,000 YBP. D-12 12$ I232
An ENE aaximum compression stress field orientation exists at the PNPP site, as determined by the hydrofracturing method. This corresponds to a regional orientation of stress that extends through-out Ohio and across much of New York State and southern Canada. Inasmuch as a northwest orientation for causative maximum compression was associated with the tunnel thrust fault, the tunnel fault is considered to be older than the age of the existing system. A lower bound age for movement on the fault is thus suggested, viz. about 10,000 YBP, giving a rather restricted estimated age range,10,000 - 20,000 YBP, and cn estimated age of 15,000 YBP + 5,000. Magnitudes of rock stresses were measured for the depth range of 394-/18 ft, giving the following rounded-off average values: maximum horizontal stress = 1500 psi minimum horizontal stress = 900 psi vertical overburden stress = 400-800 psi Similar values have been recorded throughout the midwest, New York, and southern Canada. Measured stresses were resolved for the vertical plane perpendicular to the tunnel fault, and the question of recurrent slip was examined. The results show that below about 200 ft depth (elevation about 300 ft), the fault plane may be con-sidered to be strongly clamped by frictional resistance, and no recurrent motion seems possible. Accordingly, it may be academic whether or not the fault tenninates at 150-300 ft elevation or continues in a down-dip direction. At shallower levels, the fault plane is apparently less strongly clamped (stresses are inferred
\2tA/ 233 D-13
by extrapolation), but slip is not considered likely because it would require def' mation of the inferred toe buttress. On balance the data suggest that che tunnel fault should probably not be regarded as " capable" despite its relatively young age. The last movements on the tunnel faults were apparently generated by northwest-orientated compressive stresses associated with a rebounding crust during deglaciation of the Laurentide maximum ice sheet. Nucleation of the fault at some earlier time is not precluded by the available data. l:
)} 'h b D-14
- 2. INTRODUCTION The writer was retained by Gilbert Associates, Inc., in February 1979 as a reviewing consultant with the principal task of establishing the origin of the tunnel faults, as considered in relation to the Perry Nuclear Power Plant. This report presents the results of the investigation which followed.
I am grateful to L. D. Schultz and R. Wardrop of Gilbert Associates, Inc., for their cooperation, assistance, and courtesy in many matters related to my investigaticn. At my recommendation Gilbert Associates, Inc. approved Edditional drilling, rock stress investigations, and consolidation testing of fault gouge, and the cooperation in these endeavors of the Pennsylvania Drilling Conany, of J.C. Roegiers and J. D. McLennan, University of Toronto, end of A. Dvinoff, Woodward-Clyde Consultants, is hereby acknowledge. I also appreciate the cooperation of the Weston Geophysical Corporation in providing data from their tunnel mapping and regional seismicity programs. 126 235 D-15
- 3. TUNNEL FAULT DESCRIPTION
- a. Intake Tunnel The deformed zone begins at about station 10 + 40 and extends to station 10 + 90. The principal structure is essentially a low-angle thrust with approximate attitude of 050/17 SE (su ike N 50U E, dip 17U SE). In detail, the fault zone is comprised of a series of irregular steps, with local dips varying from zero, parallel to bedding, to 50 SE on one o~ the riser surfaces. Dip slip, which virtually co1ncides with net slip, ranges from about 1.6 ft near the Crown to about 2.5 ft near the invert. The slip difference is taken up by splay faults and minor structures of various kinds which dis-tribute the strain within a volume of rock adjacent to the main thrust surface.
The zone of observable deformation extends locally as much as 10 ft above and 6 ft below thn fault, as measured perpendicular to the fault surface, but is ordinarily much less. Splay faults are best developed in the footwall above the spring line. The splays are them-selves thrust faults, with dip slip on the order of an inch. Like the main thrust they are influenced by bedding-controlled anisotropy. Their attitude varies from " horizontal" (i.e., parallel to bedding) to an U inclination of about 20 (average of 14 measurements) to bedding. They appear to die out in bedding planes at horizontal distances of 13 f t or less from the fault plant. Curvature of layering occurs adjacent to the main thrust and splays. Some of the curvature may be attributed to displacement along a fault surface of upward-increasing dip. Normal drag folds are locally well developed, affectil g layering within a foot or two of the thrust. D-16 120 3 236
Fold hinge lines are nearly horizontal and trend approximately 0 050 , parallel to the strike of the fault surface. Hinges are often rounded, and most folds are approximately parallel (bed-normal thick-ness about constant) and locally concentric. Angular hinges occur locally, most often in close association with the fault boundary surface (e.g. East Wall, Station 10 + 63 - 65). Axial planes are not always 0 well-defined but seem to strike about 050 , parallel to the fault sur-face. Flexural slip is indicated by thin gouge zones parallel to layer boundaries on fold limbs. Most folds are fractured, intensely so adjacent to the rain thrust where folding, splay faulting and fracturing are closely associated. Systematic small-scale open fractures appear locally on fold hinges at high angles to the deformed layers. No mineralization was observed in fractures. The fault zone is comonly filled with a grey breccia-gouge, about half of which is comprised of particles in the clay-silt range, with the remainder angular sand- to gravel- size fragments of shale and siltstone. Rock 4agments contained within the brecciated or gouge-filled fault zone are not randomly orientated, but are preferentially orientated such that their mean strike azimuth is approximately parallel to that of the fault. This suggests rotation of the fragments about an axis normal to fauit slip. Gouge is irregularly distributed along the main thrust, with the thickness range varying from about half a foot to less than a i inch. The splays also contain gouge, to a maximum thickness of aoout half an inch. Gouge thickness appears to be a function of fault offset (slip), the relative attitudes of bedding D-17
}2 237
and the fault surface, roughness of the fault surface, and deformability of fault boundary layers. Physical properties of the gouge are dis-cussed subsequently. Under low-angle illumination, striations and gro]ves were discovered on bedding and riser fault surfaces and ca gouge adjacent to it. These features were produced by frictional wear associated with faulting. Groove lengths appear to be ataut 0.1 ft or more, adjacent to the main fault. Striation orientatit,s are parallel to the fault dip azimuth. The orientations of striatione, minor folds, and tabular fragments in the fault zone all require the dominance of dip-slip in faulting. A small right-lateral componenc is indicated by striations orientated at 15 to the fault dip azimuth at Station 10 + 58, west wall. A minor syncline with a steep axial plane appears in the hanging wall at Station 10 + 51. The hinge is rounded on the bottom and continues below the invert muck. The fold dies out toward the crown through a zone of conjugate shears and bedding plane slip, with offsets on the order of 0.1 ft. Local gouge on layer boundaries throughout the fold suggests deformation by flexural slip. The fold probably reflects the inficence of a local shear force on the buckling of a multilayer under axial (horizontal) load. It could reflect a dip change in the thrust surface, located below the tunnel at about Station 10 + 50.
\ -
D-18
- b. Discharce Tunnel Two deformed areas are present. One scch area extends from about station 13 + 24 to 13 + 62. The princ'. pal structure is a low angle thrust, with approximate attitude 060/15 SW. In detail the thrust surface is comprised of connected bedding plane fault and riser segments, with local splays. In places a single fault zone is present, sometimes characterized by breccia-gouge as much as 0.3 ft thick, and sometimes by intensely fractured rock; in other places the fault zone is comprised of a " nested" sequence of a half-dozen individual faults, with thin gouge layers separated by fractured rock. The zone of significant deformation is rarely more than 3 ft thick. Conjugate splay faults are best developed between Station 13 + 50 and 13 + 60. The mean angle for riser faults (including splays) from bedding is 260 (13 measurements). Dip slip on the principal fault ranges from about 2 ft near the invert to about 1.5 ft near the crown, with splay faults and other minor structures accommodating the strain (associated with the slip difference) over a larger volume of rock adjacent to the fault surfaces.
On the whole the structure and its associated minor structural elements closely resembles the intake tunnel fault. Striations on the fault surfaces are normal to the fault strike, indicating dip slip motion. Drag fold hinge lines have negligible plunge and have azimuths approximately parallel to the fault strike. Folds are essentially parallel, with hinges that vary from sharply angular 1 239 D-19
_g. to rounded. The strike of the axial planes parallelsthe fault strike, as do the strike of folded limbs. From Station 11 + 50 to 11 + 80, a small thrust termination is exposed. Strike is about 020, with irregular dip, roughly 20SE. Vertical offset is less than half a foot near the invert. The fault terminates in a cluster of conjugate thrusts (displacement on the order of 0.1 ft) with NW and SE dips, which pass into bedding planes. The layering takes the approximate form of a monocline with axial plane attitude 015/25SE. Offset near the crown is virtually negligible. 12 1, 240 D-20
- c. Extent of Fault Most of what is known about the faults is based on the tunnel exposures. In map view it is known that the faults extend at least 750 ft along strike. The extent of the faults beyond tunnel exposures to the southeast, along the dip, and southwest and northeast along strike is unknown. It might be inferred from the
" splay" that the main fault will terminate toward the southwest, and increase in size toward the northeast. How far it goes, and how large the slip becomes, are purely matters of conjecture. In profile, limited additional infonnation on extent of fault is available from boreholes TX-1 to TX-6 in the intake tunnel, with the most distantof these holes penetrating the fault TX-4 at Station 7 + 44. Based on the assumption of a linear slip gradient, approximately 14 ft of slip was predicted for the fault at the TX-4 location. This figure corresponds to 4.1 f t of predicted vertical off-set. Two ironstone Lands (key beds) in the TX-4 hole suggest (but do not prove) an actual vertical offset of 3.8 f t, corresponding to about 13 ft of slip (Fig. 1). A tentative identification of the fault in TX-7 was reported by GAI (Gilbert) (Nov. 78) based on core recovery loss and clay atelevation 245 f t (Fig. 2), although no anomaly was later observed on the WGC (Weston) velocity log. This depth was consistent with a straight-line extrapolation, using the observed fault dip from tunnel exposures and previous borehole data. No fault was later distinctly recognized in the nearby TX-11 hole to elevation
'h k, 2k D-21
I k ~11" CILBERT A33CCIATES, INC. SOIL AND RCCX CLASSIFICATICH SHEET 13 CF 19 SHEET O PRCJECT: P.N.P.P. w.C. site AREA DRILL HCLE NC. N CONTRACTCR: CCCRDINATES ELEVATICH DRILLER: CWL 0 HR$ CLASSIFIED BY: DATE: 9/13/73 24 HR$
$PT Seil Cr Rec's d slows / J . C E3CRt.')T!CN q g Y }e 6 in.
E a
- ' O*"*Y (*' Consistency), Caior
. U d
j Range Size G,ein G*elesic Dete, Shane C'evad We. d .*. Rock Cr Seil Type - Accesseries Core Res. Construction Pre' elenes, 85 6 12 1a Run Core
. 85.0-86.45 - Med. hard, ned. gy. - Lt. gy. wash .-_ shale and siltstone, tr. it. gy. -
siltstone - (m "Fe" band'@ 85.35' 9/15/78 Man cage bj , accident
.-[ ,,7 h4 @ 12:33 p.m.
prevented oo '
. afeernoon
- ~ .. d,A $ < N{\ U 8 - eoring g, -
Coring y stopped g - Top of Fault Zone - 86.45' G 87.0' -Core barrel raised
- , _, , several inches and water left
_ running in si 15.5" of recoverv over 30.5" of , . ' hole pviT ~
.run - _,n.Lig n.- Coring 4" lt. gy. sandy shall -
resu=ed - 9/18/78 4" ned. gy. shale - Driller notes starting
- 3 1/2" Broken, it. gy. siltstone 30% 5.0 3.65 @ 87.0 - - ~~ ~
continuing to
~
4" lt. gy. sandy shale ~ 88' - Milky se . gray wash
. Clay re=enants in partings and - .around broken prices -
89 3ottom of Fault Zone - 89.0'
.89.0-90.0 - Med. hard, =ed. gy. -
^- shale, sene it. gy.j/siltstone in -
-1/4" bancs V ._ _ WP - "Fe" band ? 89.15' - Lt. 37. wash . G - ~
90 ~T Lent eiece - la" w .zr s/-i m22 {P g
\ 2g,j 242
cit. SERT A55CCIATES. INC. SCIL AND RCCK CLASSIFICATICN SHEET 8 -a 5HEET CF P R CJ E CT-w.C. 51TE ARE AN rth Shoreline 31uf%gigtgeggge, 3-7 Herren Testing '
- ELEVAT;CN 6l3*l CONTRACTCR: COCRDINATES Joe Minarchick E 9, 095.96 3,L g gg3 o pig.,g g .
CLA551FIED SY: R. '". Wardrop c ATE: 10/12/7S 484, g ns H.P R EMAR K3 5PT Soil Cr Rock e. g Chemical Camp, C Sle=s/ . DESCRIPTICH
. .J '. o. s,m e logic O m ,
menge I 6 in. 9 -i Density (or Ceasierency). Color d Ground water, i.C !
?
Rec's Cr 5-il Type - Accesseries
~
d Size $hese q .c. Construction Presleens, core 35r 6 12 la non i Cm Q. 338.2'-363.6' - Med. hard, =ed. gy. shale w/some dk. gy. brn. siltstone lam. lit:le 10. gy. 29. 2*: u :: - - candy shale bands, tr. thin 1:. 10.0' 10.0- Long piece - 5-1/2" gy. siltstone lam. .
,, a) High concen: ration of .
siltstone las. fres .
--. 350.2'-351 .
_.. b) Thin fiss11 shale sea =s .
- @ 362.8' and 364.0' '1. 7 + - - :) Clay re=nants in par:ings 10.0' 10.01 - 9 352.65', 362.8', and -
m 361.6' - Long piece -
) d) 1/8" seam of pyrite @ -
10" En - - 356.5' ~
-.- 363.6'-371.3' - Med. hard, dk. .
_ gy. shale to med. gy. shale, w/ . Long piece -
- some dk. gy. bra. siltseene la=. 36. M . 6 1/2" M 1/S"-1/2" thk. :r.1:. gy. sand 10.0' 9.1;' 365-375' run b ; shale, tr. thin sil: stone la=.
a) X-bedded sandy bands 0 cored s:cothly ye: 0 fast 365.95 and 366.3' - rate b) Clay re=s. in partings Driller has ve:7 difficul:
@ 368.5' '--'I ) Seam of thin fissile -
time pulling w - shale @ 368.4' barrel after FAULT ZONE -371.3'- . ,
. 365'-375' run 372*4' a 10,, c :. core =1ssine (2) 1,, core 10.0' 9.96' So::o= 3' of - - barrel coa:ed =55, pieces w/clav re=nants - pieces -
v/1/16 thk. Jo: .__ do no: inter 1cck w/each other or ~ ,,
"Y"# *
r..n.1
- w/ core above and below - ~v* c'ayev - 372.a'-395.0' - Med. hard, dk. 9,,s
- .. when
. gy. shale w/so=e dk. gy. b:n. -
d siltstone lam., seme it. gy. - b sandy shal( bands, tr. thin'1 . - Long piece - a gy. sil:s:ene las. - 2 S,, a) 4" :hk. sandy band 1 - Drill water
^
380.6' and 6-1/2" 3 386.4' 19. 2': - ge::ing plussed m R30T-*CM CF ECLE 395 ' (elev.223.1') 10.0' 9.75-' up in bo:::= " of hole (b) clay re=s. in cartings 3 376.5', 379.0', 382.55', 382.65', 34ec=s 383.6', 391.35'. .*; - ' /-"- (c) 3roken sea =s of rock frass. 3 375.a' 385.0, 392.45' (i")' 7 associa:ed w/ clay re=s. 3 392.45' anc 393.3 T q3"? h,
- 3* 3 C..
C (d) Thin fissile sea =s 3 386.l' _e :5,de " p"", Ie) Possible "ye** band 3 394.3' w.= ,
-100 (depth 730 f t), despite the fact that improved multiple-tube boring techniques were use.d so that core-recovery loss would not have been the necessary basis for fault identification. If the fault indeed passes through TX-11, it must do so along a thin bedding plane segment associated with little damage to hanging and foot wall s.
A possible bedding plane fault in TX-11 (Fig. 3) may be interpreted on the basis of thin clay seams observed at 470-425 ft and 485-490 ft depths (elevation approximately 140-160 ft). Un-fortunately these segments of core were disturbed, e.g. bf impact of the flying gas-propelled core barrel on the drill platfoi m, so that the interpretation of broken rock here is not unambiguous. A gas pocket at this elevation would not be inconsistent with a fault inter-pretation (indications of gas pressure were sporadically observed in TX-11, especially between depths of 310-510 ft). The intercretation is strengthened by the fact that the 155 ft elevation in TX-11 corresponds exactly to a straight-line extrapolation from the known location of the fault in TX-12 and its inferred possible location in TX-7. If this interpretation is correct, the fault extends at least 1150 ft in the dip direction. Core loss in TX-7 is possibly explicable by drilling technique, and because of the uncertainties associated with TX-7 and TX-11, an alternative interpretation was considered, namely that the fault surface steepens toward the southeast. Drilling of an inclined borehole (TX-12) using a multiple-tube wireline technique was recommended in order to assess this interpretation. THe TX-12 hole was drilled from approximately the TX-7 site, but angled 30U toward the northwest. A :one of broken rock and gouge (three seanis,1.5-3 inches thick) was found between depths
>24 ,,2 4 fi }2 4
h' GIL BERT ASSOCI ATES. INC. t' SOIL AND ROCK CLASSIFICATION SHEET SHEET 10 oF
# 'r t a o bd CRILL HOLE NO. D // ,: [ //- [ w.c. SITE ARE A Il I E LE V ATION 0 O/' '"* O. CCCRDINATES A/ 7 6l S #C.'77 . ACT OR:
l E 11M 9 1B C G I t- GwLonRs u t R . ' '- '..!,c JL Adamt ASSIFIED BY: [ W8 R e7 D ATE: #!l8!77 24 HR5 l l 5.;l or Rock
$pT Chemical Comp, j _
c'*i* 'i' D t
- f, Blows / $ .* d O. Reage G'*'a
_e _* cc j, Densi,y (or Coa sis'*acF}e C'I*' v{ . 5;se Shece G'* W a d * *e Cons,ruc, ion Problems, N Rock or Soil TyP* - Accessories Core Rec.
^ *'**
Run Core g 6 12 18 ety,' 3k elt Intede n. A, ,- yy, , ,,,,, y ,,
% "/!,/tlecl.h a , -- tile 1,&rt! tr. e. sittsf.ns, httle g 5.0 5. 0 '-' buMfueyhY v4 ~ dKbr.d, ale --fasil sha!e re<~s WM,,eK ", g,',' I', _ff,,, ,, ~ c 4st.S '-toed cnt homo-s e.vst.as, e ,41 ,,
- y. a._ - ta~y prece -ts "
QSV.1' i 5.0 g S.0 [ - S a,,,e - p a m e h.rel lrgy, ,,, ;
,,,,,,0*'
Q ~_ 311ls' tar,e brn. theIs 4 m (o - ! "fL.. " Yo'n.
'*"O g % .
Se=d y aheIt ha-cl (Q 458.o'-458.C'b l,d A (y-b ede - O
't - 5*ttsfe-e b=-elz ys s.es'-yst.ts' x.
Somt. it b 5.0 $b -
~ ~
_ gy.b w .o*It So.e'fa'e/s/t.{fa,.k'soe c ener fr= % ru 11.e fa Y' n, I 15943.75-<(.$s,.re,.*if 0 y Co
,,F l,e er -- /4" t s == - *n t'S - ,So n,z .
Some ,,ir,;e.,e ba.,J k- '
- e W-
) L n, y {. $ 2 - u s 7 y e w.ss k'( q SO ~
q (,,, , ,,,ee t - /S ,,
,sny,,5 ,.
5,'" t ete k akatier<.el -fre.n w ,- V <s blo~s ceta-. eC .4 s G.ter ou t afhote. , g d 973.7-dv.3.' 76n ciy ae. s1 a cl & $o' 50-_ %'.,1eJ '/Qf2,'.,
.,1, yn r , i. kite sk.te e v70.7sp12.e, y R** . <erer 'C'iil'X..,' [
55~ 47sSC'4'). Ye7 n,<., clay .reo , e 97/.('
~'T - drallers ?lat4rm & greaf ty d,g.,s W. ~ '
So**1e 3h.le IS med. 'b (1. r m': 4',sy'7r et gy br.,. la .gy,Y 6.0 47.y sith her b..Jq 4779' ene es* f - ,g , [) f,.,4f,&. y gy .g gg
<,q,e d,, h3'/*/e e 4 7C.2 %. thie) _
p ,,,, .; 3 * "]M Some fr./t.yy.rinstene lam.y Q , , D 5' O f'.C __/tI.*r.j"5
--tnctir..,.1e e b...I ;e(f6 , c s 5,3 'l,-7r.ofirve31ar.<y k u i.ed cast n.rr ea., q t, ret.3 , - Icy prece -W - , - - vaarrel put/4 g- So.n e, ntit//, try sitts tc-e inI G -
_ fe i,.if -y retre 1'
-11Ute arbcn .s c i
i-5c.? 5.0, 5. w~
; tt yy.,x. h.-.. 14 te I..,*)sa Ji
- y. t..!, b .,d e Q " _ ,4 }an U 75 p ,,,fjty,Mlen klect ,as aauses g,rt ru,, N m
y85. 9 '-. y a c.cs ' as b
-c u,1,I' $na. 4P** leeds. . nett.~J ex.y chy(inh. sty purse,eck y', ,
g" ' N "' 6 '3
.Ge-. ka e,-.- .M f)e Sg.cs74 .
g*()' 6.0 p',"3 "P*" lYg
~~'
C7 T[W),I y 884(S'S) &c Sam % fr. si/ts1tne le~k.1el,*.skel, y, ,n ,,,,.... . roTnhem blowu~de r
'irt.C' . futtke be ,d k y 4.*r ' :s per: sum
_ Shf.tt (,I,,# ) e,,f- [ ' b5 54 ole t 10.0 g d ."~0 MScD.4 't1.Ia-. >d) g te ]sifrm,ur.p e j, r. s c 3 3^'. :21 s.'72
'L ' } [4 N". /[
o-25 -L i 8
of 376.0 and 380.4 ft (elevation approximately 300 ft) which undoubtedly represents the fault zone. This depth corresponds exactly to a straight-line extrapolation from tunnel exposures through TX-4, and therefore no significant curvature of the fault surface is indicated to the 300 ft elevation. Despite excellent core recovery, local stratigraphy could not be used to determine offset. Drilling continued to 420 ft with no further structural disturbances noted. From the data of TX-12, the fault extends along dip with certainty at least 600 ft. The drill data available at present permit thne interpreta-tions: (1) The fault terminates between TX-12 and TX-11. (2) The fault passes through TX-11 along a cedding decollement perhaps at elevation 140-160 ft. (3) The fault steepens between TX-12 and TX-11 (indeed, probably between TX-7 and TX-11) and passes beneath TX-11 giving a minimum average dip angle between TX-12 and TX-11 of 36 . Hypothesis (3) is weakened (but not ruled out) by the lack of any significant concave-dcwnward curvature betwtan the tunnel exposures and TX-12. Hypothesis (2) is enhanced by the straight-line correspondence of fault elevations between the tunnel exposures and boreholes TX-1 to 7 and TX-12. and by the offset suggested by TX-4. 12ff 246 D-26
- d. Mutual Geometric Relationshios The three tunnel fault structures display similar deformation-al style, magnitude of slip, slip gradient, moderately brittle de-formational mode, and are clearly genetically related. The main dis-charge tunnel fault is very nearly on strike (0440 ) with the intake tunnel fault. The two exposures are in many respects virtually identical, and interpretation in terms of a " single fault" model is reasonable. (In an alternative model the two structures are considered as separate elements in an en-echelon system). The Station 11 + 50 discharge tunnel structure strikes so as to inter-cept the main discharge tunnel fault. Because of its smaller slip magnitude, it is interpreted as splay fault to the main discharge fault.
Similar slip gradient on all three fault exposures (Fig. 4; data were taken from the tunnel maps prepared by Weston), and the ob-served termination in the discharge tunnel, suggest that the structures have propagated from some lower elevation. This conclusion has a bear-ing on genetic interpretation. Furthermore, the slip gradient (about 4 ft of slip per 100 ft of fault) suggests that the principal intake and discharge tunnel faults will terminate within about 40 ft or so of the tunnel crowns as measured along the fault surfaces, or within roughly 20 vertical feet above the crowns. The faults therefore should not reach the elevation of the lake bottom. In this light the Lake Erie bottom video survey results seem understandable. The discharge tunnel " splay", if projected eastward, intercepts the intake tunnel . No such structure was observed in D-27
?0M BMINiml , 1 I I I e \ 9 ' -
- Ea s
.l e *3 T
a. aI E wt j A-I [** N 1..--?a (
. , ~3 u
m 1 :, m
-+,. ,. - - -vy. ;p
( .
.s,a l .e
- n. , t
. g , ;., .o -- - ~ ~ . - . _. .
- ,E. yg g.,
;{v-- .. . .. . . , 4j.
f' f . p ,3 O ,.C c.
/. . . . . . . .;*w .y -Myf a i . .: .
l s 3 "",' gr
.?. .? 'L 1
_ ~- O Al y iagv.g4., L6' Q ; E '
-_ es,;. , /
O h-T W e m m I d%Q - a m.%, p% hC 11 I tw mN ci 1E F " " kN W d i b " i w w
+ u j , ,j - *"*- 4 a ,
m
%*- s - u y y . v +- a * .
- M %" g t 2 A J y w .-, n C . y 1 - .
% -t . x- ~e w
41 2 L_s a mo cs. T_ d .. 4 1 T' w g wl b .E' N
- - I E O , v. ,. 1 te ".
w 89
" f >3 1 1 ji '
aW L e g , , l' I U f*I
, t-2 .f . . . . -
T 1 r
*B' b *b5*- \ _/ %E. I a
I 3 E F I4 I g n L ^ g I m W . I A y t* MI v t i 25
*1 ! ' t -a 1 L ~
i en ' e' ' f E M L_ F I E I ___I__ I r II O T g L3 Ll I M i 1 l L N Ld L W f , ME m og g - ., - L f 1 4e u=='* a 7 y ni
~
_3 - S D-28 i LJ p{ L
the intake tunnel, which indicates that either the entire splay dies out towards the southwest, or that it is present below the intake tunnel but has terminated on an elevation below the intake tunnel invert. Either interpretation is consistent with observed evidenca.
/ ~120cl 249 D-29
- 4. AGE OF FAULTING Based on test results and visual observation the gouge is classified in soil mechanics terminology as a gray, stiff to very stiff silty clay with abundant sand and gravel-sized sof t friable shale fragments.
Consolidation tests were conducted on two relatively un-disturbed samples from the Intake Tunnel, and on one remolded slurry specimen. Plasticity limits and compression indices were similar for all three samples. . Details are given in the Woodward-Clyde report of July 5,1979, in Appendix VII. Maximum past consolidation pressure (Pc) was estimated for two samples of the gouge by the standard methods of Casagrande (1936) and of Schertmann (1955). The results are summarized as follows: P c (tsf) Samole Casagrande Schmertmann Cc' (unit strain) PL LL I-2 8.0 12.0 0.110 18 27 I-4 4.5 6.0 0.112 19 28 The agreement of the two methods is considered satisfactory, and on the basis of these results the maximum past consolidation pressure of the gouge is taken as about 914 tsf (say 125 1 55 psi). For comparison, consider that the tunnel depth at the fault locality is about 110 ft. Ignoring the 15 ft of lake water above the top of rock, the corresponding total vertical pressure is about 119 psi (8.6 tsf). Average effective vertical pressure, assuming a standard
.32 3 250 D-30
fluid pressure-overburden ratio of about 0.4 is 71 psi (5.1 tsf). (The fluid pressure gradient assumed is about ;.43 psi /ft). This value falls near the lower limit of the estimated range of uncertainty for maximum past consolidation pressure. On these grounds, while one could not conclude with certainty that the fault gouge was sub-jected to greater vertical pressure than that existing at the pre-sent time, the results suggest such a possibility. If it is assumed that, because of erosion, present overburden thickness at tunnel level is less than the maximum value of overburden to which the fault at tunnel level had once been sub-jected, a vertical pressure of perhaps 6-9 tsf can be postulated for tunnel level under lake level conditions similar to those at pre-sent. If a prehistoric decrease in pore pressure is postulated, e.g. associated with lake drainage prior to the establishment of Early Lake Erie (470-ft le.al) at 12,000 Y3P. maximum overburden pressure can be increased to about 9-12 tsf. The entire range of values (6-12 tsf) is consistent with gouge data. The maximum past consolidation pressure estimated by the Casagrande method for upper and lower tills at the PNPP site is 4.3 tsf (average of 3 tests; range 4.0-5.0 tsf) and 6.0 tsf (average of 10 tests; range 4.3-10.0 tsf). (Appendix 21, Foundation Investigations and Design Analyses, PNPP). The results indicate that both tills have been consolidated in the geologic past to pressures well in excess of the pressure imposed by present overburden (about 1 tsf). The probable loading mechanism is glacial ice. 12((251 D-31
Assume for the moment that the tunnel fault was present at the time the lower till was subjected to its maximum consolidation pressure of about 6 tsf. This corresponds to an ice sheet at least 200 ft thick. Pressure at tunnel level was about 5 tsf more, and eroded rock and till could account for abM i tsf, for a total of 12 tsf. This is within the range of consolidation test results for the gouge, and it could be argued that the gauge and lower till were subjected to maximum consolidation loads by the same event. The argument is strengthened by lending more weight to the Schmertmann-method calculations (which seems reasonable), or by assuming a higher fluid oressure-overburden ratio for the gouge. A consistent argument can also apparently be given in regard to the max inum past consolidation pressure sustained by the upper till (4 tsf) to which must be added 2 tsf for assumed inter-vening till and 5 tsf for rock overburden. The estimated total of 11 tsf at tunnel level falls within the range of uncertainty for P c f the gouge. These arguments are summarized as follows: (1) Hypothesis: Maximum consolidation pressure for fault gouge corresponds to present overburden. Result: Pressure estimate at tunnel level is 5 tsf, near lower limit of range of uncertainty for P ' c Interpretation: Hypothesis cannot be rejected but additional pressure mechanism seems likely. 12k252 D-32
(2) Hypothesis: Maximum consolidation pressure for fault gcuge corresponds to conditions of pre-existing overburden or pre-existing groundwater conditions. Result: Pressure estimate at tunnel level is 6-12 tsf, consistent with estimated values for Pc. Interpretation: Hypothesis cannot be rejected. (3) Hypothesis: Maximum consolidation pressure for fault gouge corresponds to maximum pressurization of lower till. Result: Pressure estimate at tunnel level is 12 tsf, near upper limit of data range for Pc. Interpretation: Hypothesis cannot be rejected. (4) Hypothesis: Maximum consolidation pressure for fault gouge corresponds to maximum pressurization of upper till. Result: Pressure estimate at tunnel level is 11 tsf, within range of uncertainty for Pc. Interpretation: Hypothesis cannot be rejected. The radiocarbon date of 14.480 YBP i 310 derived from the lacustrine sediments over the upper till suggests that the upper till is at least as old as Hiram Till (14,500 YEP). (GAI Report No.1997, Nov. 7, 1978). Compression of the Hiram Till could be accomplished by an ice sheet associated with the Hiram advance or by a younger ice sheet, corresponding to the Ashtabula Till (13,000 YBP). The lower till may represent the first part of an advance-retreat glacial deposition couplet, in which case it couid correspond to the Hiram 120,253 D-33
advance, or it may represent a separate late Wisconsinian movement. In the latter case, it could correspond to Lavery Till (16,500 YBP). The late Wisconsin maximum is associated with Kent Till about 21,000 YBP with an end moraine 70 miles or so south of the Pt4PP site. Sugden (1977) suggests a thickness for this Laurentide Ice Sheet of 1 km at the Pf4PP site. Comparable advances also occurred during the early Wisconsinian (Titusville Till, ca. 40,000 YPB), the Illinoisan, and perhaps pre-Illinoisan (Lessig and Rice, 1962) times. The increase in overburden pressure associated with a 1 km thick ice sheet is on the order of 100 tsf. It is difficult to con-ceive of circumstances that would prevent such events from leaving a marked imprint on gouge consolidation characteristics, even grant-ing uncertainty in the selection of appropriate fluid pressure-over-burden ratios and some redistribution of stress in the vicinity of the fault. I conclude that the formation of fault gouge was to a large extent, and perhaps exclusively, associated with faulting younger than the Kent advance. For similar reasons exclusively Paleozoic or early Mesozoic faulting can be rejected; several thousand feet of overburden corresponds to an effective overburden pressure on the order of 100 tsf. The possibility of incremental fault propagation is not excluded, but this discussion is focused upon the last fault movement capable of forming new gouge or significantly disturbing pre-existing gouge. 12Q254 D-34
On the above grounds, assuming the data as representive, the tunnel fault reflects significant movement younger than about 20,000 YBP. The data are consistent with compression of fault gouge by a lesser ice sheet than that associated with the Laurentide maximum. Three candidate ice sheets are associated with Lavery, Hiram, and Ashtabula Tills. The youngest of these is about 13,000 YBP suggesting that if the fault is related to a glacial mechanism, its age is probably in the range 13,000-20,000. But the mechanism of faulting is uncertain, so the 13,000 age is not a firm lower bound. The hypothesis that maximum gouge consolidation pressure corresponds to present overburden and fluid pressure cannot be wholly rejected by consolidation test data, but the data suggest the operation of additional effective vertical pressure mechanisms. Drainage of the rock mass at about 12,000 YBP yields a more consistent predicted pressure, as does the assumption of a greater prehistoric thickness of overburden. But lacking adequate data on erosion rates it is not possible to be very precise in the matter of a lower-bound age on these grounds. I would judge the minimum age to be on the order of several thousand years, but this is merely a guess. However, rock stress cricatation information (to be discussed in the following se cion) suggests that the fault developed under different stress conditions than that in evidence today. On these grounds a lower bound of about 10,000 YBP is prcoosed. Cinally, it would not seem surprising if, over the past ten t;ousand years or so the gouge developed : few cracks, and mineralization in extremely small amounts (such as re-ported by WC) occurred within them. 17_$4255 D-35
This estimate of the age of the last movement of the fault differs by two orders of magnituoe with a " minimum age" estimate of 1,000,000 yr offered by WGC, based on rate of microfracture " healing". However, the lack of agreement is not disturbing to me because I do not believe that there is an adequately demonstrated basis for the " mineral grcwth vs. time function" proposed by WGC for the PNPP site. Inasmuch as this function forms the foundation for the WGC age estimate, the accuracy of the WGC inferred age is open to serious doubt. By the same token, the age of faulting as based on the con-solidation tests reflects certain specific assumptions regarding boundary conditions and material behavior. Error is possible to the extent that actual behavior differed from that assun.ec. These aspects are discussed below: (1) There is considerable precedent in the use of consolidation tests to establish past consolidation pressure. The adequacy of the method has been tested in civil engineering practice (e.g., Casagrande and Fadum, 1944; Zeevaert, 1953; Schmertmann, 1955). There is also precedent in the interpretation of past consolidation pressure in terms of geologic history, and in instances in which the maximum past consolidation pressure has been reliably determined by geologic evidence or other independent means, agreement between the actual maximum past consolidation pressure and that determined by consolidation tests on " undisturbed" saaples has been quite satisfactory (Terzaghi and Peck, 1967, p. 77). There is also precedent for quantitative determinations of ice sheet thickness 12f i256
from consolidation test data, both in Europe and in North America (e.g., Kogler and Scheidig,1948; D'ucker,1951; Harrison, 1957,1958). (2) The " sealed" block samples, from which the test specimens were prepared, sustained moisture loss during storage. The effect of water loss is commonly to produce intergranular stresses within the samples, which could lead to an overestimated value of past consolidation pressure. In the present instance no interpretive problem arises from this possible effect. (3) Lateral strain and squeezing of gouge at the time of faulting seems likely. Therefore the early strain history of the gouge may be described as complicated. However, the strains associated with sub-sequent vertical loading conditions, such as burial by ice, meet the standard assumptions associated with consolidation testing. The assumption of zero lateral strain associated with ice sheet compression seems valid, at least to a reasonable approximation. (4) Because fault gouge exhibits a complicated strain history, it is possible that its past-fault consolidation characteristics are not necessarily identical to those of similarly-graded sediments of different. origin. There is little information in the published literature to directly assist interpretation of the matter of fault gouge con-solidation. On the other hand, silts of similar grain size gradation which have been c.ontorted by the directional drag of overiding ice have been subjected to consolidation testing, and glacially-induced distortion of this kind seems reasonably analogous to disturbance by l 2 D-37
faulting. The directional stresses and associated strains in such disturbed silts were shown by Harrison (1958, p. 77) not to have affected the maximum past consolidation-pressure value induced by the thickest over-riding ice sheet. (5) Pore water under pressure must be permitted to drain away during consolidation. The hydrostatic pore pressure distribution observed in most boreholes in shale in and near the PNPP site lend support to this assumption. Because of the drainage factor, there may also be an effective upper limit to the distance ' rom a glacial margin over which past consolidation pressures can be accurately determined (Harrison,1958,
- p. 77). But in Indiana, this distance seems to be no smaller than about 30 miles (associated past consolidation pressures are about 50 tsf) (Harrison, 1958, p. 81, 83), suggesting that this factor does not pose a problem.
(6) Is tha gouge so old that soil mechanics tests are no longer applicable, e.g., has the bulk material sustaineJ changes due to aging such that consolidation characteristics have been altered? The answer appears to be, no. Successful preconsolidation estimates by consolidation tests have been conducted on materials of Tertiary age. Tills and lake silts overidden by four oscillations of the Wisconsin ice margin were subjected to consolidation tc)ts by Harrison (1958), and the past consolidation pressures thus est3blished were used to re-construct a paleoglacier map of the vanished East-White sublobe of central Indiana. There is no indication of diagenetic changes or 12d 258 D-38
significant chemical changes in the gouge material that would significantly alter consolidation properties. Further indication is that compression indices for undisturbed and slurry samples are identical. The consolidation behavior of the surface tills (which are also are comprised mainly of comminuted shales) is similar, and the past consolidation pressures established by consolidation tests of tills are consistent with the data obtained from tests on fault gouge. (7) The bulk laboratory samples were not specifically orientated, but the prepared consolidation test samples are considered to be approximately horizontal (+ 15 ) based on bulk sample shape and size and location sampled. (8) As described in Section 3 of this report, the fault itself is not horizontal, but is comprised of a series of irregular steps with local dips varying from zero (parallel to bedding) to about 20 on riser surfaces. Gouge thickness is not unifom. One may therefore question whether or not the maximum pressure exerted by overburden and an overlying glacier is transmitted everywhere to the gouge, because of " arching" (stress concentration) effects. My personal opinion is that severe arching effects associated with the distribution of vertical pressures in this case are extremely local. The slight average dip of the fault surface (15-17 ) does not favor the development of vertical stress arching over large domains. (Horizontal stresses may be a different matter entirely). The shale strata of the hanging wall have been disturbed (fractures, splays,
etc.) to distances as great as 10 feet as meas",ed perpendicular to the fault surface. Thti shale is wholly thin bedded, and thes e 1s evidence that the shear strength parallel to bedding is small. Evidence for bedding place slip is observed where minor bending nas occurred. Splay faults and fractures are common; thin gouge seams are associated with the fractures. The hanging wall rock mass is therefore weak and very flexible. Therefore the capability of the bulk material to sustain significant horizonti ' 3 hear stresses as required in order for significant arching to occur seems slight. Those portions of the fault zone characterized by broad patches of gauge, several feet long and several inches thick, are thus likely to be subjected to, at least to a first approximation, full overburdan pressures. The consolidation results themselves lend some support to this view. The gouge consolidation tests are internally consistent in that two separate samples from different locations produced results that are in good accord, with respect to consolidation behavior and past con-solidation pressures. They are externally consistent in comparison to calculations considering present overburden pressure, and to pressures inferred from extensive consolidation testing of near-surface glacial tills. The burden of proof would seem to reside with those who might doubt the gouge results because of the possibility of non-representative behavior associated with arching. Further sampling and testing is of course possible, although only at considerable effort and expense. 12h 260 D-40
To conclude this section, it must be acknowledged that not all possibilities for error have been absolutely eliminated. Still, on balance, in my opinion the best available estimate of the age of the last movement on the fault is that provided by interpreta-tion of the consolidation test data for the fault gouge. Accordingly, the last movement of the fault probably occurred no more than 20,000 yr ago , and the age for this last movement is estimated at 15,000 YBP + 5000. 1244 261 D-41
- 5. ROCK STRESS INVESTIGATIONS
- a. Orientation and Magnitude of Stresses A program of stress measurements was strangly recommended, because I considered it incautious to select or render judgment on design details to ensure safety against possible fault displace-ment without adequate information on rock force fields.
The test prograi.' was carried out at my recommendation by J.C. Roegiers and associates using the hydrofracturing technique. I was at the PNPP site at the time the measurements were carried out at TX-11 and I am satisfied that the results obtained repre-sent a state of the art capability. This discussion is based on the data contained in the preliminary report by J.C. Roegiers and J.D. McLennan, dated July 1979, and subsequent telephone conversations. Details of the stress investigation are given in Appendix IV, and a sumary of results is provided in Table 1. The direction of maximum compression is east-northeast. The result on stress orientation was not wholly unexpected because stress orientations in western New York and southern Ohio were known to display similar trends (Fig. 5). Specific tests at the PNPP site were nonetheless considered necessary in the interests of safety. Figure 6 is a sketch map which illustrates the relation of the tunnel fault and other structurcs to various stress fields. Stresses at (a) and (b) refer to stress orientations theoretically associated with a northeast-striking thrust with (a) the condition just prior to faulting, and (b) the condition after faulting has occurred, D-42 j
Table 1. llole TX-11: Suninary of Stresses for Ilydrofracturin9 Data. Ilydraulic Fracture P o a 01 02 "I 3 2 Fracture llorizon T= Lab T= Field T=1000 (selected average) 01 "lla f Iden ti fica tion ( f t) (psi) (psi) (psi} (psi) (psi) (psi) (psi) (psi) (deg) (psi) 1 718 311 796 1061 1971 -- 1931 1061 1951 -- 1230 3 654 283 733 1023 1943 1413 1643 923 1367 080 1030 823 1343 813* 1043 4 614 266 686 906 -- 1281 646* 906 1281 067 - 5 574 249 634 849 929 1033 29* 849 981 100 930 6 511 221 586 921 1246 -- 1826 821 1526 94 1150 721 646* -- 1226 7(a) 454 197 577 1137 1987 1917 1947 1137 1950 37 1170 [ (b) pre-existin9 joint assumed: 730 min 730 min 1450 max 085 970 min 8(a) 394 171 411 971 1881 -- 2096 761 1358 -- 870 551 681 -- 836 (b) 971 1988 -- 1150 Note: o g7 indicates calculated total horizontal stress in plane perpendicular to tunnel fault. Ot is azimuth of o l xis; 076 assumed for Fractures 1, 8 in calculating o H F. 8(b) based on assumption2 a = 971. Asterisk indicates o gvalues impossibly low. Mee-N CD e U U.
33-P00R3RIGINAL :
- __ *o .--
'7 L
h :.1
~ ~b * \
1_ m\' . ... . - *
- i g s\. .re,j$.
j% f ,, *
- N.i 's, 77. an r
* ., , 4 's. !, v' , *
- 12 J g ' 5[g -
- o xz -. t ,
sf N 1,
,%* , ' N. ;
- w 5 A.$
~ \
q gc% N..
* ' 9 ' ,'s, ,
r
' - .. e.~~2 -- 5k .. J -fj o e * <l .--S 6 I.
1
% ,k i . s k g(= 3 . *9 , ,o
- r .-* ' s. : ' h.':. f %- [r ';-i 1 ,.- g , ,
- ; e S
, . j. _.-*" o *
- e ~~ .
i t" i -1
'\.
Y. s .. t. ,~.- -
- [sue%l l
- c 4 '
- -3*$ '4 ;)
i x o---~~ a
.< .. r. D,*
t y' . I , 3 =T
- m
..__....., / . . - .- ' ~5 .: > ._. . _ f
- aj*
4 ,
.. \,,A, ' ,x s
- 4 ' /'s ,
- {l3 3 cmn 19:', 5 ,
- l r-
'd;..:..----- ~ ~ ~ '
- rT.i . -., :-
e t . 's k .-$, 3.5 L $.s.'i N .- O,iE- T I z s 1 4
. r m -s*
o i , u., ? .A98ytd 3
- e: xa
^ 3 2 i
[?$$ e s e*
.y .d., . ,, e./ . j 52
- Q%p . ---: ^* /
g-3 d .~. 3
; ~^ ,/ .! -f / n d U * ; I z
- n: ^
., . f 3, . =
j-? g h. ,
//a 1 '5
- j ' * . .
s
- i s '.,.f
.. *p . d
[--J---,J'.d.o!.Pl:F ., .., ~ % y . <= { ,
/ .
- \
/ A GS-v4
- _i _2 _ _ _ _ _ h_ --
12h y 264
- D-44
with the northwest stress system diminishing to some residual value. The northent stresses remain relatively enchanged, but because in (b) they are greater than the relaxed northwest stresses, the assign-ment of prindpal stresses changes. As a result of faulting, the assignment of og is changed from the northwest to the northeast. Still, the two principal stresses in map view are orientated per-pendicular and parallel to the fault strike, for both conditions (a) and (b). The measured stress orientations in the hydrofracturing pro-gram suggest an average azimuth for go of 0760; neglecting the measure-ment of Fracture 7, the mean value is 085 , and the range of four values is 067 to 1000. At face value the 085 orientation of c1 is evidently not compatible with the formation of an 050 thrust either by the analogy in Fig. 6 of (a) or (b), by directed pressure or stress-relaxation.There is no evidence of strong anisotropy in the rock mass which would per-mit structures to form at high obliquity to principal s,.resses. The o present 085 orientation of a lthus suggests that the local stress system formerly associated with the development of the tunnel fault nas been altered. The stress field at the PNPP site closely corres-ponds now to a regional field that apparently ex .nds from the upper Mississippi Valley area to New York. The tunnel fault is therefore considered to be older than the age of this regional stress system. Without doubt Pleistocene ice loading prof' idly altered the stress systems in the upper crust, and the present stress system is considered 126 265
es .
. . ) 4 f ~ ~- . . . . . . . ~ . - - - * < d ' ~
f . _ .; =
.a ' I. > .s. . ]O i M.6 c
- g. . . . .. . . . . - . . . . - - - . = . . ~ .
,f _ .. ', -l. .. . . ., . 5G . . '. . . :O.
c s\ . . . . . . .
..2.'.'.^ .. .- . - ...a......_...-._2.. . . :. . . . .. --.: .. .. D. ... .. . .. . . s . x _
s
..,.,...c.:.-.... > ... f . . r ..;..., .. . .
_. . . __...3.... .
.. s y - . /..cw..._.__.. ' i. , z, /. _ _ m. ,9\)fo S,hc[ . fr . .1. ..:y_ _. ,
- s. . . Str' DSS Pic %.
?.,. ', ,
3 . .2.- : be n
.' ^y CTj .. ~ . * .
Q p {-u r.y . J. . > 3 , _.
- y . . g .. . . y ,s -. . . -;.. .. . _. '; } f** , . - - . 2.L.
- . ,. L - . N. .
N: ' ..:
$* . 6 ,. /
- V ..n... . . ..
b
~
r ., '.. _,.9
~[ .. .-
OW 500.'
. _.7 ; . -. . . g g g f'; . .j . .y - __ . . .,. .
w . _ r..._._
...;i. . . m .. J . , '. . . .. ~ . f% ri c .a.... . _ _ . . . _
k.g.o .$$6,6 _..s . , . . . ; '.
'..-g . am6M # 4v - . aN.y E+ .[== g. y_ - - w e og e -
I. .- ..m..= g
~ ". . ?. . . . ,' ~ i k. ,.;, Y> I? O . ,Y. $,s&. . - .hl, .k. .&.' ', .- .f- ' *- c -n. . .-.,.4..... g n.e. ., *",.~ -- . N %. r y9._.. .w.s.N.,..'..*/."hy"'.' 's. 4. - . _ c' . . .J. ' *O
- C.U.' t .
,.s. . .v.*- *4 1**. ' f ' . .'y.'.".,,'.'> ..? ^ . ** tr ' ,' . ~ * }, .c , . .
h .. s b . %
. * . A. -4 . . .s n.. . >. 's s._.%- 'V.- - r . .e . , . . ;* f *. . .r. .7 . .,. - .
c.._;......,- . ..J._.... ~
' t%
T
,..P,"* . . . __ ,. . _ . . . .
17 4*
~.
z*
.w.. .- .
s ,.
^ , , - - . -
1
-h ' r s , ;s G .-r . .. f, ,.; g L .s...... . .. _ - . . . . . , .. . .
_... :~ *
. . . -. . .1 .-. >, . D-46 .
to have developed following retreat of the ice sheet. A stress system associated with ice-deformed crust seems consistent with that inferred for the tunnel fault. A minir.um age for faulting is therefore suggested, viz. on the order of 10,000 YBP. This is consistent with the interpretation of fault age based on gouge consolidation tests, and leads to an estimated age of 10,000-20,000 YBP. The poorly defined fractures at 037 indicated for Fracture 7 differs from the 085 average from Fractures 3-6. This orientation pennits an interpretation in terms of Figure 6(b), with 037 not greatly different from the strike of the tunnel fault. Fracture 7 lies above the proposed intersection point of the tunnel fault with TX-11, so that it may be possible to formulate an argument in regard to behavior of the hancing wall as distinct from the foot wall. On the other hand, it may be simplest to interpret Fracture 7 as in-fluenced by pre-existing joints. The pole maximum of 220 founda-tion joints as compiled and plotted by WGC is associated with 0440 , with 0370 lying within the range of significant pole concentrations, e.g. 025-0540 (Fig. 7). Details concerning stress magnitudes must be interpreted with caution due to complex frac aring sequences associated with hydrof racturi ng. These sequences renders difficult the estimation of instantaneous shut in and breakdown pressures. Some uncertainty must therefore be attached to che individual principal stresses ey and c2 calculated fran these selected critical pressures. c1and s 2are assumed to be 'n the horizontal plane. ' g7 D-47
~ ,Jg, yT C A.CA> Tis Ta oalj
- O' in~ foaa> CATS oa) deO4c:K uuat O!isdldfIbr ' = ' ...
' v ' .,
feer l Nff g C.I. A Y #/o
% n: ;50 d
l N l A =
, -- + , &
a N 4 _f , e gau E 9c' T f d 9
~
v' :..
/ / .P t, 1- * /' , g /
4 0 sl' G3 o?
# '*P (f.5A . . c,?,s . \ ; ',
L E h- ~
. N ' 's $ ~' e _
s
. Ns \
Or , b x, s . 12h 268 i S F;3 t - I D-48
The average values probably give a true indication of average stress conditions at the site. For the full depth range of 394-718 ft, and rounding off values to the nearest hundred psi: maximum horizonta' stress = 1500 psi = o g minimum horizontal stress = 900 psi = c 2 vertical overburden stress = 400-800 psi = c 3 Similar values have been determined in other engineering and mining sites (including nuclear )ower plants) in Ohio, New York, and southern Canada. Furthermore, despite uncertainties associated with individual measurements, certain trends seem to possess validity. The gradient of c 2below about 500 ft seems to parallel that for overburden pressure, such that approximately 2 * #v + 250 psi (Fig. 8). Some uncertainty must be attached to the Fracture 7 calculations; tabulated values based on c2= 1137 are considered as upper bounds. If the orientation of Fracture 7 is considered controlled by ~ e-existing fractures, wish actual c3 oriented at 0850, a range of values seems compatible with the data, viz. c = 730-1137, o g = 2 1450-1137. Despite this uncertainty, Fractures 7 and 8 suggest possibly greater values of c2 than at lower levels; higher than average ey values are also evident for Fractures 6-8. To a certain extent c reflects the selected values for c , so that trends 1 2 exhibited by the two principal stresses are not wholly independent. Extrapolation of stress values to higher elevations is uncertain because of the apparent increase in stress between 511 and 394 ft. Estimation of c2above 394 ft based on extrapolation of D-49 j' 2bh
.-bg -g, y, f
- 7) .f f* ... . _ . . ..
l --- u-d000 .- - SOQ . - .
. . ~
g'
^ , - l , w g j . . . . . .. ,. . . .s s d',.
g I. --V
.-. -- _ . .r .. - ^
A. 3.. . .... . . r
,..W'?. v . . . . _ _ .q,2 . .
4 .. _. . .
.. . . . -.t. _.., . . . .. _ . .. G~ ~Q'i 3 I o ,40 0 0 . . . . . . . a ._ q. . . . . .. . . . . - ~. .q . .. . (f->;- Q , ,,,,,:q;g y < .. \. .
_.- / - y. .
- s. . r. .
,Q/ w. . .. : ; . . .. . u . _ . . . - . .L._..
1, . c .-
-Q l ) g .%2.( .}.- -. .- . r_.. .s . . . .* . ... . . .r Q ,,, V . ._ . c \ .; .. .n > .
a . _... ..;. ..
.l'. - . .
4., .. s .. . r- -" -
. i. - . . .. . \ f_F h .
I f *> h M k.+ .. . L'_ , I U fe rw' k h , g'q.h. . . . , . ' . . ~* .. J
. . . b.c. h._rvi b,7 ..
J(M' . . . .. - . . b GId cM
...s._..._.g \. ' . .. .i. .- -. . . _ ...-. . . , . . . . - 9... ..m; .
_. .a .~ . x. . . : .g,...._...%=.......,..,.
.. . 5y. .,. -. _ m . .
g -. n .%
-- t (S . V . ,. , . . _ i. . ..+ e ... . . . .
- 3. . ...Y .
.. c. .;, ~, i ..._..,,., .. ,4 . ..,.,:- . . . . . , , . . ....w. . . - . . . . . ~ . .
n [>
.._ . ,. _p....< a x
m_ _. ._m_. , . ._. ,
.. : . .c .
3 ..., . .. a.. .:
. ,1. m ,._-a..._. .. . - . = - ,, 3 . . . .,,. 7. ,.,, ,.a .-.,.w.~ 4 .i;^, . , . - w ,c. i Q _c,.t .vga. - ~
__ _ . - , j 2 _ _ g.L _ __ . . . . _ _ _ . . . . . . . Tk2 >---._ . . .x -
,.Q-Q : ( ., . 4T l 3... ~. = :.. ,. .. ..*
< m, . -
;.4..J . .a.... .. . ~~..ct . y __. .x. o. o .%,2. \.,, ,
t.
.:.. .. ..... .m , .... q._. . . 4 ,. , u ~ , l . g ..~ ,. . . , - 'n . . . '1-*--S .. t, '.g. $ %.,.- g.
yp w z ' g, .g ; ,, y 4 r, e. 1.'_ , . _. -. w , .
..,w m . - ". - - m --- ' . - - " -
_.,e . r ; v. * - - . -- - 3 . .. .r.. --- t- . w..- - .. ~ =--t y - '
. 5 , . .h ,.t , , . . 3.,4 .g . . .. .,.... . m.?'.,..,. . . . . p.; . . . . . ..y,.., '
W. . ;. . g. . ..,. .q, .-. . . . . ,
.,s-
- g ~ j.,'.. ,[, N. N j, . I-' t' l4 . 'i * '
' y. . . m, ..~ .' . } : . .~. . %m - . . . ...,,; . .c . . a w: e , . . . . .. .. . , . '..3 * ^ .e , .n- L. . . p m.: :- . V. _ n.g , w . - - - , . . Jy,, .n, n. . . . 4 4/~',; ya."
e . . * .
. 1 h..,4. y ;;;; . ..;.'. .;
i .. w p.j w..pp ,.
- .g < - , . g? .. . .. . j. a.
wl .
, g- ; -
y~ . .;7 w ,w - v n v w h: u s / H / 7/ / /./ << e / f___ . T, a u f f. - _ _ _ l y--W "- , + .f~e 7 T,' =- - _ : w~ ..-_, . *?.%n--- . .'. ,. .G f_n4 _ e\ y-%m. . . + .; -e & s, A.,. - _ 3 .QL .,. p.
.7 . . .c n.. . . . ~ . ..n t- U u : ;.-;. m, u = . .n, <4 . , . .- .s. . .; . a. ... . . .<...,.*.-,-.s- . .; > : . :. 7. .~. . w.:.... .,.,c~--.:. .-y. n w._ m. .w- .. -: A .9 7..Ym_. -
_., '.a . . . - - /
- s** b7 m e , g . . > - . . + '- g. *
..*b.
f< . '[ s ' 3
+r*,,- *? ,, > * - .*
k-1~. q . . ;;. .. . .x .'%' . . . f , . .:.J,- .2. n .: s.:4 , . .n . ..
- y .-. g . g-O. - p4.n'3r-*.'$h
-, ; ,, . ..y , ., . .. J *s,' .u' ,....-. m, . J.s - .4 e -a; '.. - . , . . ' , , . , $, . *, ' ' .w'.,. --.j , . s*~i ' Vf , "..#...,....* W ,.u .' ' .e' * , y ;- ^
3 , & L g m g,.- . s'. 7'.*N's rC- , . s.. .c n~,.,. l- 'J, ~ ba*-., , s,
*e,*'.'., .4',.u. %*
w% 4y4-. .'" W .. . ; - . '.gy .eMo Jw *
. - . .. _Aa.we, . . .'sw' .M,r.4.+'*, . i.. ... . i
- 2. % '# ."5 .4 u. 5m o . - . -,- (..,., .
,6 .
NQ 'o*~.4 ** ' . . ** ?N ' f,.-=5 - g ' ' *
- m. *s".? .
g
,. f. ,
i Y, .- O, ... . . _ . . _ _ '..:s .. .,"
-.
- t..c,- s.*: t :,: . W ?* p'.-r %., _a_ '.
~ ,.r. __s".2.
- f"."g**w! %.
.;s'.' 'ks. - } .5***P- . :. . p #. . .l , a .&., 'es- ..,.~..-b.,1 <- , 'l
[** t {._tf at - '8. - 4 .- . .k, ."K. ~.g k,
. -. .s: .gw., 4 . *0 .7 g - .;a .'n- - : ,.'.,5* -w ' ,, p.g t
i . ,e.g'
., 5 , m . . 3.
a r.- . . m . . .c ..c
. ._. . .$. .*. e.sk.1 . ~, ,.. .w . '. p. .; ..,.._-.,.c.,..kH 1. . ' . #,..-. ' .*i. -.u.. . - . .* 4 . . _ ,. ,.Q i k b, C .'4 i.e% *. :(j * * ^*# 7 G -* *~
g"g'O. T. .* .' ~. .*. * ,'*
~M..., _,' .s. .. .. L =-.y .~,.f h . 4 ,. ,a, ..e-,y.
s - ; , .,g,
-.,y,.,, .[ , _. : .
c
.. :R_,;. _ - _- ,f . . . . . - - 3 ,. I' ._ . n .: , ; ,'-w. j .q:.. .,..:.,
84 m,e 'f i . * .Ac *
, : :.n_
c--~..W ', m .,-.--..-s.. . , .
> ,~ . .. .
- m
.- . ~. a n . . m. 2. ... .b , .*. . :d., =: + . . i " 'T. m., .r- - y . ~- - . -
ame ,
, N
- y C. * ,. . . *.,'3
..r .'"';- . * . . .W. p 6'..4, ' ,---..?-- ' ~
l
. . , . . hI % . G. * ., da, a, a. > ; - *e L . NN"m.n. ,. a.S.'4 D.gl.' *[.b . -[*b.-: m N,,.,.:__. . -- L . J ' ., . c w ;-. . .y. .} . . 'L,[; *, .. - d ' ... a ,.s ~L.i. f,f .. . , - .O ..,,,-T.+9 W >*.-. .- e . ~. - . , . .
9._-.. ka..'-.--. 7) s . .. r r -T.3,.:?w. ,.y.. ,. - . .'- . V .g.. y , , . p< , .,.5 . . -, -
. q .A su.-- .,,r , . . . .y ;v g- ,3 .
- r. >e
,.Q . , .->e.- y e .y ..i-. .
n . .
.,e .,* .+.% . . ,w ._.e, g. . ~ - e,.. . ' -P .J.'1 .g '.s. . , , " , . . . - . . , , ' _y. . -',....*.A3%,4 .-gr f, ; g ,. ..-_rs. . f.i.$.~,
r L.
.---c., ..$- -,..'".
e
, i . , . , . * *e v * *
- T ;a 'E -* . '
p
,m. .2..y,p. c.*. *c *m: ,., -. ; .- x.,g
_ :. . w
%f .. ,
- .r . - cA% .e_.f .. .,.. . ;. ,.* . t ., .* .n "
,P-g*. dy.'. ..,'?.,.w-'.. .. I'.y.:..'. g.R..y/g. a-..
g
= . . ...c,... $~ -d - .+
7,..-.
=',..'v.-_ . . - . - . - . . - ' .., ~ ~ ~ .. ,3 r. .. , o .6.,'
s m-+z..,-C ,~, ..;_ .. .wI- . mi-
-4.a -- - Me ,- g_5Q m- 4 m, ,,#%. - - ' ' '
40-the data trend from Fractures 1 to 6 is considered to be a lower-bound. Upper-bound values are not clearly defined. The reason for the apparent increase in stresses at and above 511 ft is not clear. One possibility, however, is that the tunnel fault indeed passes through TX-11 between Fractures 6 and 7. Higher horizontal stresses could therefore be interpreted as stress concentrations associated with this fault. One alternative possibility is to consider the high values as stress concentration effects below a downward-terminated stress-relief fault.
) ?.
D-51
- b. possibility of Future Slic on Existino Fault Consideration of this important matter is examined by comparing rock stress information to rock strength.
The value of horizontal stress in the vertical plane per-pendicular to the tunnel fault (og g) was calculated from selected stress values and 9 1 rientations as given in Table 1, assuming an 0 050 azimuth for the tunnel fault. Subtracting out fomation pressure, effective stress values for the vertical plane perpendicular tothetunnelfault(cy,o{p)aregiveninTable2andplottedin Figure 9. The average value of the horizontal effective stress obF is about 800 psi. Stresses for Fractures 6-8 are greater than those for Fractures 1-5; the trend appears similar to that previcusly discussed for principal stresses. The specifics for Fracture 7 are uncertain, depending on interpretation of the 037 fracture orientation. Accordingly, obF f r Fracture 7 could be as low as 669 psi. Extrapolation of stresses to shallow elevations is uncertain. Data for Fractures 1-5 permit a lower-bcurd estimate. A reasonable estimate would appear to be the average of stresses calculated for Fractures 7 and 8. An upper-bound is not well defined. Vertical effective stresses are given by average overburden pressure in psi (taken as 1.1 x depth in ft), subtracting out forma-tion pressure (Table 2). Consolidated-undrained triaxial compression tests with pore fluid pressure measur<nents, or other test methods appropriate for measuring the effective stress strength parameters, were not D-52 124 272 9
Table 2. Sunnary of Effective Stresses in Vertical Plane Perpendicular to Tunnel Fault Hydraulic c' c(p Fracture Identification (osi) (osi) 1 479 919 3 436 747 4 409 664 5 382 681 6 341 9.1 7(a) 302 973 (b) 668 8(a) 262 699 (b) 979 TX-12 226 836 Average of 7,8 Tunnel Level (a) 69 836 Average of 7,8 (b) 400 Lower-bound extrapolation
)5 D-53
x.
= 9 .. . . \ e &
e . . . . . . .d Q
. . . . . . .. = 44 = .. . +. ,.a + . . . . . ,d b.'.2 _
S * #'
. . e , . . . _ . . . . , . 8
- b .
. i.. . ~ . . ./* ' . . . .. .-/ - . ' . . . .,-. ;+.. ,.
e *.
.m .
4
%r , * ? '; .> "- * * ' '~. , .c -. . , . y ,.2 ,{
- s. , , .
. . , - . _ , . . . , -P . ~' - ~ * ,i's ,}*r- -
1" * - ; -
,..'g a~ *, ^
s
, ..b. . 1 ., ' .x O . .., wl . ._. .m ,.. , .. . . . . . .
m .. . . . . ..:. a
. ~ . . ._ _ ,, .
- m. ,, .... -
. -.. . - . . . .. . . . _? .s.. . . . . ._,. _ , . . ~ . _ :. .
7.....e.
.'. 4. . -m. . .,p.m.q y: . M ,. .. .. .:.
o O ~
. . ::v. . . . .= . ... . m . _ 4.. .Q V. . 7 '. .. . x n.. .. ...- ...: . . .,.m .. a :s. ..,.+.m.. , ~ . . .w : .- .- . Q .- m. . b.y. Q*'}.] ,. F. , . , ...y.. . .- ., .
e
;. lew d.' boM b/; WA.I .J .l. ..~ ! _. . ,,.,*. - .erMpd1M-4/fQ g o .c .e x. .W. 1 o..Ixt.m. . s f-&. ] g , . :. . . .t_. ., j- - . . , .. e. .
s. h 1 - 6
.4- =.3 s y p .. ,3 s-- . .
% .. +~. ....e 9..'!'.**.*:...
. . ,a :. . G 14. 2. ., . ; . ._ a . ,- , .% . '.'*: . . r -r. . , . h. u X :'. ;- : ,. .:.J . - - - ...,. . _._ .. ....u.. _. . .a.. c. -. , . , _.,.%. - . . , . . , . . <. ~ . . . +.. e. .,i__..
m...,...c.., ..
.. . . rs .. - .
w . 1 o
.g-l..:.',,,9.- ~ ,*'4.
- .m p Ty,..
d..h.Q g p... ..,' ... J. , s .. - ..< y
..], ; . j -
L .,. .:..
...-g.,.s..:
a
,,s .;..,
c..y . .; . . . s
..> t c .-' . % &~; .s,:- e -r ,.v. . t. w .,. .: - .: A ..r.,g ...=s- 4 %. .
- W <- . ,,
p.., 5 s. { _*:t .; m.
' ~ . y.v.: ',~.~,' f_-P , .I w . -l D., rr.T ,L.'p. . ~ . >2.- . . - * ', .* . 4 f.z : . t .2, . , . , 1 ,y . .s. .a. >...-a3..., , . . , . n. , . .. . e. .n. _. , .
1 s *- a. . w. ..v' - . e e* - g. -*t s ,, .;r y . . . y = or ;,
- ...'%..,,- , " u y . : . t =.. , .* . ,y . ' . , 'f
. * . o* . . .*~ - . . , . ., e:...
t;.wn,
. , '. L, 7. . .,,.,+- g * :w... 4 w.'s u. .. . .J ,7 . . . . .%....... -. . >, . . . . . - L n>, . e. s ......: . ' .R-.- s.. .s.y. , ,. a.- n.. s . -. . . u y a. g~,y d s. -. . . +.. c . .. . , g .- - .s;. .q. p<.q :,_.,y. . . . . fm. . # ) . y f. <'.; . . ;g. . , nq.:. .. -?. . . . .
y __ . . , _ _.
,_, _ a. , -m . v . r.. e . j.. r; - ,n_, . 5 .:*., ,M , s j e c p . !,. a . t. .c : s . ,- . ,, m, .. .
j - ... . . , - . ~ .... .~ . . . .....7 . *.
. s .,.<:, % *y..
. . .. . s :
t : L ** -: s ~: Q** j?~;' , y ' -9 * * * . y. . u .A-w%.n.~ e.~.'G ..:< -t H. ;r
- s. .*.&* )s** .....p s,.i mp
, -3 . y .,o .,; . ,.* .z, s .,.: ',. _.y o,Q yya . s~ n g o ,. . .n, , .. ' -4 - ,1, ._-.-*a. :.* r, t ' e.n g . . . ~ y ..; n , . u. ..a .. ,s._. ..g ,. ,. -,a . ' . .y. . . Q , .a. c .., . . n . *-.. ..- -4. ..; , =.- .~. y... ; n.s - *^ . . s
- n. .~. ; ' u' g' ,'. j ., q. ~- %.
=.._, M. fi.a.. u , i ..y .- _.; - - .., ,~ ,. ~ ~ =. s. . . ~, w;pn, m' - 4 . -;,w .,.a , , ,. s ".ty ~ e, %" -.s. **, .s, . . . ; .4:=, .. . < , r , ..%j,*w.v; _9. <,:. ~, .L._~'., -eL
- y. r. .x..y. ._
. . ~.. . ,* , . . , p+.y ,,- a., .r.w. . - ~.q.. . e. . s. r ; . ,,e .*s- o s s. 2~.-. ' .r . , . . . . ; .; . - e ,,.e _._s. .s. , ~ m'. ,. g. '* n ,~./ .37. *.", .,._;... . , : . r.n_, : w.
- D .P..; .4_. . .g.$..; - . . .-- > .
v- . a . y I ', M ,,** ..,.u'm..,.,.. . . +~). . ., . _ 1. g . . . . . . . . . , . ... _. - '5 * * *' ~k- 5. ^, *'5 ai . '""# y*'.- [ 2. . 7- . h.Pr* 2 * .~ ,'vW. r .,-r cp. ,. J,. :...=. 3.f,s,. .-'"? hu. **' . . ..- ,. 7,ic, . ; 'c .,.#'. " G. ;x., .; , ' g , ,r.> ,n >1, ., - .,. -.f. r . m.p.1 ...L e.a,'W} .g; . .1.,,f,'. 1 e(. . ;t -s' ', .*.g. a 4 4 N,.Q @ f. .,. 4. ;;,,,, g,,Q.,,. "% s s ..e 4 . . u l 1 -gs.}} . .e>. -_,..f..,,,.p. ,;v=* . sey, ' a,. ;*
- a . .,; . .,a.,Q . V . l *s.p m.
, 4 J *. .%.. ~ , .s J . ~ , v. , a a m.,...,.%.W.._ ,y _ . .t e ,,, -- . .,g a .u u s n . - m...,4.. m, P .. . ., l n. 3-, ,v,.r, e.g .. . . ..p..,. .=.; .. A. .a . .e ..z., ,,a . -1...<. g. .** . .. . u . . . . g . g , .- s.an . - . eft.p , ;-, jg * . , - ,ag , . ., -. .. y w. M*., p .s.. >+i. . . , e . .. -('~ 4--4/.,4 p . mJ . ,p-,,<y,.%, .
- s. t ,
b . g.4g - .N
- _ . ., . *;f
, m ar =e-=== "',b,I k , . [a '. . . - sg.. wg. .,,, ,..y . r ..t ; , .; g [ _ . .. -. . . . . _.%. . -- t... 4;f. p . . w n .,m,... -, - s M .. .c.p ,- i.. b % 2,,%lJ ,
- j. .. .. e .y.s ,.~ +. ., e, u cw. gm,.g.
,. - . , . .,a.. I .,. ,. J".. .. 3.w, .-; 7 . ,. .,r %. m. , ..-. . .,,. .s* , u: - .xg. , ,.:,4g.,,.. ~.,. . - - r ; : yg' .a - . - .p,.,'L,.. . .n.g . _- .: . : f % ,+ ., , +;. . n . m r.~M,vJ .. . .y... , , . ..w.. .m. '.~. ~W 'l 'k: > . . ~ f .2 Q .fjf:., . *W. ~f A * ^l h. * --....u..- - g' V ..,.y + m',.$*f mm.;w jKTf,f**},VN ww.ay -p m',.'.'f',,f.**:[M+ ,,; . . ,. _ y. ';, ~~~. ~ . --' . ...- x .. y ; = y :. e +. y.n ... . . , %.* M;,.,g ;44 t, w m..,1 . .n. ,...g p.. . . . . . y.. u. ,...... . v..,. a2g.^ .. - 4 A ia , . 4... ,,1+ .v. , . .='w- W.:,: .- . . . . " --3 i.e' * **b = ., ; y . ==s.*.eg.*c- ;,.,,e;g* ,f e. 5.% - -- . . ,p . - - -e. : ',: ~ . M.*?.. *r - _~<e.-.'.. .% . a a. .. - .,. -t .e -t W, h..3 .4.~.R- . '21, . a&, . .M.Q.-Y. .. .,.,*.. ~-*ws .s, ^ *;c , ' ' * * ~
- t. 3. . r
- y*
s , . ..y n.m'P.W. %.g s'y 3. .;. ~.%.%. . c.W. .s,. %%.. J C-..s f.a<: j ,.. y. . . 4 f . ..c-., r..1 %-4. +%.... [, *. w . 4. _. .; , 1..,.- . ,-3._A,Q 4.*.gg. ,N p8 "pM. ; v s p . . - 4 ,e . . .r , , . ; rn. U '.~, _ W j-*L. s.~..T.& V. %.+. == A~ l'.,.$.,..... %'* t.+- ~ * = ..'.[ Q J/ i '. '. .s n. ;'.,D. ,.> ,6.pm'r .Qt.l,f ?-W....'C, s ; - - ...,,,.', - . .s .g-..'.. . . s .s . (y,? . r = .c .c %s>... w..~ e .. m ~. . m c.- . ..t'**. - . , , '..) ~ d. ,$. c'M ,-];N. ** h %* Owg g@ m .[ @. ,..gemywh[M.ww; .w e, $4;.3_' .'3$.hhb E . A bO."ils ~.jf1 N.'g.k N 4f.j.y,,g.Jhr*EU.k j-pQg]s ,* . N,.,2* J . . . ~. .q p Q[.h. . . < . w* m md *: ': .. m . w:w* n,-;.-1 4 . . -l :< s.'.&e ;L. *-...t,ya.1 .1.w.g. 4 r . .te.g . wg- s. %m*.'.. .L .=*v W- . s, - n.sv G >L m"w.*'r*;&.k + . .f wf.A,4 xRs~g'.e.j g ya v.$ ..mmmm.e3. s.f
- 3,.-g.,g f.. h, p
i ,Q a.5*l:q[e'h.d.n*'.f.m. .*. p. g.%%g ,M * .d.4,y%. .. + esy u s 4 E=~A X _, -*A g'. .f ; . - i r m~ms y% . i [ _4..g. . .' . .' . , ;7p' . . .p:.,9 gg%. , ,4w, , y8*- g 3.*y.pi. w y f(e ' y e .y.w , - ~~ n - 7..t. . c, .=.a:, y .,g.,,7 a.: .e 43% ; e4 s ,- gd e _ch>.g.. .p >.*,,e a-r ".V*D ., , Sw -p.q , .r ...w .,.,m m ..#i.,. .s .,n, m w}g
- p. e. a . . .s .-..y -.
.- 3 _w. 4 4,..-._a o.- .e,L..c. n g g. w ,: p , ...cp:.a. . .- 4.j, r. . ..
- m. v
- r. . . . ,.xo s o,e ,. e,,.,,m.w. . . .-.t.
s " p..sea . . t. v . .
- av<N- . . pa . . _.. s.,. .*w
- r. y- w,%..a.. y ,.,4. ~ ,,.,,r..s.g S.. <
- m.. y ..4
*****q.M* 7 .* .e...w ** d... ff &. pF** ,V, e,. . g . .45 .u. ,., P*- . . t. I-it- > = .,.,g j* a - > d O '*%=".4 upe - 4]af s . *.e * . M **=*.%-= *a ' lM 5g... W, Y M -4= A- -'. . - . .e..p&_ygv "W'*.,S,. c m .we. -.,N. .. # m,.. ~0 $ h P. S. ~p. I mm { $&_ %_M.k"'"o-* A ,;;.v.c.. y - % p% . - -.. p. ,, .,_ r- . 7 .3-%_.g ..*S .$" --- _ i hbhM M .h h -.i .%_N h kbk l ..A '.r_.N .5 * . i.-c ' b-'4g . &_ ., A$~m.a~.y O ~-. :.. u W%p:Q.d:.@.. wp.4 /.u.a %.._c. h.2 4N d+. N-f h n' -rs. .M9.!J. g t.: w i'.:.yy w t@M. WS.*<...2T.f.n/.: M,%LR1 D ~9 N v.,fs dMG~. .~ .~ h,M5 ., .ga . .. .~r m --: --. ~,-,. ,w w- h-V .- 5 w M.at
- r .s .L.-s ..
+-W : . .w. a .m n-. w w w m. r m. . .u. . a.,. ., .m. .. u._:. 2; po : - w cx. . .m- -.g.~c., 4_ w w, ;:i.n.c.w. qc...v g.n,,,_4.4 .-y, .. ._, :s,. . . .. ., ... a. - ..a , ~+ - x. %.y. .~ u -. ;W. .u; < - . - w.. .g .,w, . . .%.. . sv n. .- e- c wn e- - - - D-54 _, W ,,s .v.: %%-- - .. ; n _,,. ,'. .s .m n .;= n .- A _', , + . , . 44-conducted owing to lack of suitable samples. The effective angle of internal friction for the fault gouge has been estimated at 30-37 U based on published correlations (Woodward-Clyde Consultants, letter of November 20,1978). This is al.;o consistent with plasticity limit correlations (Voight,1973). The increase of apparent friction angle at low confining pressures associated with rougnness of the fault surfaces is estimated at 10U. A conservative estimate of strength for a given segment of the fault zone is given by zero-cohesion envelopes inclined at 40-47 in a shear stress-normal stress diagram. These envelopes are lower-bound estimates inasmuch as additional strength may be obtained, e.g. through cohesive resistance. These strength envelopes are plotted in Figure 10 along with Mohr circles which represent assumed conditions in the vertical plane normal to fault strike. For each circle, the overburden stress and an estimate of the c{p horizontal stress is plotted. All stresses are " effective" values corrected for fluid pressure. Numbers attached to the stress circles are hydraulic fracture identification numbe-s. In addition, stress circles are estimated for the 335 ft level, corresponding to the tunnel fault positively identified in the TX-12 borehole, and for tunnel level. Minimum normal stresses which correspond to observed and inferred fault depths (in various boreholes) are noted on the horizontal axis. Results are as follows. Stresses associated with Fractures 1-5 permit construction of a stress envelope well below minimum strength. The stress circles associated with Fractures 6-8, which 126 275 D-55 i 100R BREA / / P' ' .- '/ A ;i j / f-l / .p ..e
- v. 3 s
. .p- ,e.- -t = -/ / ,, .. , ,- / .. .._l , t/ /- . .: '? Y ~ ~ ..'. 9. j'/ z../ I ~ ~ ' ,f ~ A.,,. *}}