ML032730608

Site Safety Analysis Report for the Exelon Generation Company, LLC Early Site Permit, Appendix B, Seismic Hazards Report, Attachment 1, Table of Contents Through Figure B-1-6

ML032730608
Person / Time
Site:	Clinton, 05200007, PROJ0718
Issue date:	09/25/2003
From:	Exelon Generation Co, Exelon Nuclear
To:	Office of New Reactors
Shared Package
ML032721596	List: ML032721594 ML032721599 ML032721600 ML032721601 ML032721602 ML032721603 ML032721604 ML032721605 ML032721606 ML032721607 ML032721608 ML032721609 ML032721610 ML032721611 ML032721613 ML032721615 ML032721617 ML032721619 ML032721620 ML032721621 ML032730608 ML032730609 ML032730610 ML032730612 ML032730613 ML032730614 ML032730615 ML032730616 ML032730617 ML032730618 ML032730619
References
+ReviewedClintonESP, +reviewednvg DEL-096-REV0
Download: ML032730608 (47)

Text

Attachment 1 to the Seismic Hazards Report for the Exelon Generation Company, LLC Early Site Permit Site Safety Analysis Report Appendix B

Contents

1. Paleoliquefaction Investigations....................................................................................B1-1-1 1.1 Previous Investigations in Southern Illinois Basin Region.............................. B1-1-2 1.2 Completeness of Paleoliquefaction Record ....................................................... B1-1-5 1.3 Estimating Source Areas and Magnitudes of Paleoearthquakes.................... B1-1-5 1.4 Paleoliquefaction Investigations in the Site Vicinity ........................................ B1-1-6 1.4.1 Criteria for Identifying Clastic Dikes of Seismic Origin....................... B1-1-8 1.4.2 Possible Paleoliquefaction Features ........................................................ B1-1-9 1.4.2.1 Locality SC 25 ................................................................................ B1-1-9 1.4.2.2 Locality SC 16 .............................................................................. B1-1-11 1.4.2.3 Locality SC 19/SC 18..................................................................B1-1-12 1.4.2.4 Locality M 6 .................................................................................B1-1-13 1.4.3 Evidence for the Absence of Paleoliquefaction.................................... B1-1-15 1.4.4 Conclusions............................................................................................... B1-1-16

2. References...........................................................................................................................B1-2-1 DEL-096-REV0 B1-i

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT Tables B-1-1 Liquefaction Evidence for Prehistoric Earthquakes in the Southern Illinois Basin ................................................................................................................................ B1.T-1 B-1-2 Summary of Deposits in Bank Exposures............................................................... B1.T-6 B-1-3 Criteria for Differentiating Origins of Liquefaction Features............................. B1.T-7 B-1-4 Characteristics and Estimated Ages of Potential Liquefaction Features Identified in Site Vicinity ............................................................................................ B1.T-9 B-1-5 Localities of Older Alluvium ..................................................................................... B1.T-10 B1-ii DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT Figures B-1-1 Epicenters of Historical Earthquakes and Estimated Energy Centers of Large Prehistoric Earthquakes in Site Region B-1-2 Sites of Paleoliquefaction in Southern Indiana and Illinois B-1-3 Map of Vicinity of the Springfield, Illinois, Earthquake Showing Approximate Limit of Liquefaction B-1-4 Moment Magnitude versus Maximum Distance to Surface Evidence of Liquefaction B-1-5 Map Showing Bedrock Structures, Geophysical Lineaments, and Paleoearthquake Energy Centers B-1-6 Map Showing Surveyed Stream Banks and Evidence for Paleoliquefaction in the Site Vicinity B-1-7 Areal Distribution of the Wedron and Mason Groups (Wisconsin and Hudson Episodes) and Deposits of the Illinois and Pre-Illinois Episodes in Illinois B-1-8 Areal Distribution of Moraines and Boundaries of Formations and Principal Members of the Wedron Group B-1-9 Geochronologic Units, Chronostratigraphic Units, and Diachronic Units in the Lake Michigan Lobe B-1-10 Photograph of Riverbank Exposure on Salt Creek Showing Fluvial Terrace Deposits of Two Ages B-1-11 Map of Quaternary Deposits in Site Vicinity B-1-12 Flow Chart Showing Seismic and Nonseismic Mechanisms That Create Deformation Features in Sediment B-1-13 Photographs of Dike 1 at Locality SC 25 B-1-14 Photograph of Dike 2 at Locality SC 19 B-1-15 Photographs of Parts of Dike 1 at Locality M 6 B-1-16 Photograph of Thick Silt Layer Overlaying Fluvial Deposit at Locality S 14 DEL-096-REV0 B1-i

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT Exhibit 1 Radiocarbon Dating B1-iv DEL-096-REV0

CHAPTER 1 Paleoliquefaction Investigations In the central and eastern United States (CEUS), seismogenic faults are uncommon or difficult to identify. Strain rates are relatively low, and recurrence intervals for large earthquakes are usually longer than the historical record. In this tectonic environment, identification of liquefaction or other features resulting from strong ground motion has proven to be an effective tool in identifying and characterizing seismogenic sources that generated moderate- to large-magnitude earthquakes in prehistoric time.

Earthquake-induced liquefaction is understood to be a process by which saturated, granular sediment temporarily loses its strength in response to earthquake ground shaking.

Relatively cohesionless sediment that is water-saturated and loosely packed will tend to compact, leading to an increase in pore water pressure. If pore water pressure increases to the point that it equals overburden pressure, the sediment can behave as a viscous liquid.

Under certain conditions the resulting slurry of sediment and water will tend to flow toward the ground surface, forming a number of distinctive sedimentary features, including sand blows (or boils), dikes, and sills.

Recent papers by Obermeier (1996), Obermeier et al. (2001 and 2002), Tuttle (2001), and Olson et al. (2003) review the criteria and multi-step process by which features of seismically induced liquefaction are identified and used to evaluate the timing, location of source area, and magnitude of the causative earthquake. These papers summarize and critique issues concerning field methods for identifying paleoliquefaction features, as well as the interpretation and back-analysis of strength of shaking and earthquake magnitude using geologic and geotechnical/seismologic procedures.

Recent studies in the region of the Exelon Generation Company (EGC) site that is the subject of this Early Site Permit (ESP) application have identified energy centers for several prehistoric moderate- to large-magnitude earthquakes in southern and central Illinois based on evidence of paleoliquefaction. Those studies show the usefulness of paleoliquefaction investigations in demonstrating a longer seismic record than provided by historical seismicity alone. The results of those studies were used to evaluate the earlier expert assessment of seismic sources in the CEUS performed by the Electric Power Research Institute (EPRI) Seismicity Owners Group (SOG). In particular, the results of recently published studies suggest that the range of maximum magnitude for the regional background sources defined by the EPRI-SOG teams may be insufficient. Paleoliquefaction studies previously had not been conducted throughout most of the region within a 25-mile radius of the EGC ESP Site. A reconnaissance field study was conducted as part of the current ESP application to supplement previously published studies and to provide more site-specific information that could be used to better evaluate the estimated magnitude of a background earthquake. The results of previous and current field studies are summarized below.

DEL-096-REV0 B1-1-1

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT 1.1 Previous Investigations in Southern Illinois Basin Region The Illinois basin encompasses most of Illinois and nearby parts of southwest Indiana and western Kentucky. The southern Illinois basin region is characterized by persistent, scattered seismicity that includes several moderate-magnitude historical earthquakes. The field investigation of liquefaction features at several sites indicates that multiple paleoearthquakes having magnitudes significantly larger than historical events have occurred in the region (Figure B-1-1).

Mapping and dating of liquefaction features throughout most of the southern Illinois basin (in parts of Indiana, Illinois, and Missouri) have identified epicentral areas for at least eight Holocene and latest Pleistocene earthquakes of estimated moment magnitude M 6 to ~ 7.8 (Figures B-1-1 and B-1-2) (Obermeier et al., 1991; Munson et al., 1997; Pond and Martin, 1997; Tuttle et al., 1999; Obermeier, 1998; McNulty and Obermeier, 1999). The energy sources (and inferred epicenters) for the identified paleoliquefaction evidence occur within Indiana and Illinois, except for the youngest features, observed in Cache Valley in southernmost Illinois, that probably were induced by the great New Madrid, Missouri, earthquakes of 1811-1812 (Obermeier, 1998). Evidence for the location, size, and timing of those prehistoric events is summarized in Table B-1-1. Discussed below are the identified events that are closest and most significant to the assessment of seismic hazard at the EGC ESP Site.

Lower Wabash Valley EarthquakesThe magnitude of the largest paleoearthquake in lower Wabash Valley (the Vincennes-Bridgeport earthquake), which occurred 6,100 +/- 200 yr.

BP1, was estimated to be M 7.5 using the magnitude-bound method (Obermeier, 1998).

Using a more recently developed magnitude-bound curve for the CEUS based on a value of M ~7.6-7.7 for the largest of the 1811-1812 New Madrid earthquakes (reduced from the higher M 8 used in the older curve) (Olson et al., 2003) and a distance of 89 km (Obermeier et al., 1993) gives a lower estimate of M 7.2 to 7.3. Based on geotechnical studies using the energy acceleration method, this event is estimated to have been a M 7.8 (Pond and Martin, 1997). This earthquake was re-analyzed by R. Green, S. Olson, and S. Obermeier using the more recent attenuation relations of peak ground acceleration for the central United States (Somerville et al., 2001; Campbell, 2001; Atkinson and Boore, 1997; Toro et al., 1997);

reviewing approximately 50 boring logs presented by Pond to select appropriate Standard Penetration Test (SPT) values for the re-analysis; and using the most recent magnitude scaling factors, suggested by Youd and Idriss (S. Obermeier, written communication, January 10, 2003).

Using the cyclic stress method, the best estimate of the magnitude for the Vincennes-Bridgeport, Indiana, earthquake based on all these solutions ranges from M 7+ to 7.5. The energy-based solution developed by Green (2001), which circumvents the use of the magnitude-scaling factor (an uncertain factor) in applying the cyclic stress method in the CEUS, gives a value of M ~ 7.5 for each of the four newer attenuation relations. The more recent evaluations by Green and others have considered the influence of aging effects on liquefaction susceptibility and concluded that for moderately susceptible sites like those in 1 Ages reported are uncorrected radiocarbon years BP (before present), except where noted (i.e., for calendar ages reported as BC or AD).

B1-1-2 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT southern Illinois, the small changes expected given the types of sediments would have little influence on the interpretation of paleomagnitude (Obermeier et al., 2001; S. Obermeier, Personal Communication, August 9, 2003).

Skelton-Mt. Carmel Earthquake - The next-largest earthquake (the Skelton-Mt. Carmel earthquake) occurred 12,000 +/- 1,000 yr. BP (Hajic and Wiant, 1997; Munson et al., 1997; Obermeier, 1998). This earthquake is estimated to have been M 7.1 to 7.2 by Munson et al.

(1997) and M 7.3 by Pond and Martin (1997). Both this and the Vincennes-Bridgeport earthquake were close to one another and took place in the general vicinity of the most numerous and strongest historical earthquakes (M 4 to 5.5) in the lower Wabash Valley of Indiana-Illinois (Obermeier, 1998). The inferred locations of the energy sources for these events are approximately 15 miles west and 25 miles southwest of Vincennes, Indiana, respectively (Figures B-1-1 and B-1-2).

Springfield and Shoal Creek Earthquakes - Paleoearthquakes of lower magnitude, as indicated by paleoliquefaction features, are more randomly distributed, commonly having struck in regions having no significant historical seismicity. Smaller paleoearthquakes recorded by liquefaction features in Indiana, described in Table B-1-1, include at least three events that occurred east of the lower Wabash Valley source region in central and southern Illinois. In this area two strong, mid-Holocene earthquakes, referred to as the Springfield and Shoal Creek earthquakes, are documented by paleoliquefaction features such as clastic dikes, sills, and detachments of fine-grained sediment that sank into liquefied sand (McNulty and Obermeier, 1999).

Springfield EarthquakeEvidence for a prehistoric earthquake near Springfield, Illinois, was first discovered by Hajic in 1994 (Hajic et al., 1995). Subsequent field investigations documented at least one moderate-sized earthquake (M 6.2 to 6.8) and probably a second smaller event (minimum magnitude of ~ M 5.5) in the region between 5,900 and 7,400 yr. BP (McNulty and Obermeier, 1999). Based on the areal extent of paleoliquefaction and the location of the largest dike observed (15 inches [37 cm] wide), the source of the larger event was centered about 22 miles northeast of Springfield, Illinois (Figure B-1-3). Both events may have had approximately the same source location and extent of liquefaction, but data are insufficient to confirm this conclusion.

Estimates of the magnitude of the Springfield earthquake are based on the maximum distance of paleoliquefaction features from the inferred energy center (approximately 22 miles [35 km]) compared to curves that relate moment magnitude versus maximum distance to surface evidence of liquefaction (Figure B-1-4). McNulty and Obermeier (1999) state that M 6.8 represents an upper-bound estimate for this earthquake, citing evidence for a high water table at the time of the earthquake and relatively shallow depth to bedrock that would limit amplification of bedrock shaking.

There are no readily apparent geologic structures that can serve as strong candidates for the causative source for the Springfield earthquake(s). Only two small structures are mapped within the recognized source region of the Springfield event(s) as defined by the observed paleoliquefaction features. These structures are a small (< 6 miles-long), generally northwest-trending domal structure and a 10-mile-long north-northwest-trending fold and associated fault (Figure B-1-5). The apparent localization of liquefaction features may relate more to thickness and susceptibility DEL-096-REV0 B1-1-3

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT of local sediment, so that the energy source for an event may not be coincident with the larger liquefaction features, but rather with a more distant, larger earthquake (R.

Bauer, Illinois Geological Survey, personal communication, November 21, 2002).

Shoal Creek EarthquakePaleoliquefaction evidence shows that a strong earthquake occurred in southwest Illinois at about 4,520 BC +/- 160 yr., or 6,500 years ago (also 5,670 +/- 80 radiocarbon years BP) (McNulty and Obermeier, 1999; Tuttle et al., 1999). McNulty and Obermeier (1999) infer the energy source to be about 50 miles east-southeast of St. Louis, Missouri, in the vicinity of lower Shoal Creek where there are dikes as wide as 1.6 feet (0.5 m), and there is a concentration of dike sites (Figure B-1-2). Dikes from this event as mapped by McNulty and Obermeier extend approximately 22 miles (35 km) from the inferred energy source. Tuttle et al. (1999) report prehistoric Holocene dikes at sites immediately south and east of St. Louis (see discussion below). They suggest that the energy source for all liquefaction sites near St. Louis and Shoal Creek was east of St. Louis, likely at the location inferred by McNulty and Obermeier (1999). However, they also suggest plausible sources nearer to St. Louis, or on the Centralia fault east of the Shoal Creek center shown on Figure B-1-2. McNulty and Obermeier (1999) believe that the source may be nearer to St. Louis, but based on the absence of dikes immediately east of the Kaskaskia River, they discount a source centered there.

Evidence for at least two earthquakes, widely spaced in time, is represented by the dikes within the approximate limit of liquefaction in the Shoal Creek area, as shown on Figure B-1-2. Although multiple events are indicated in the Shoal Creek region, the level of ground motion appears to have been significantly different during the two events, suggesting that the younger event was either of significantly lower magnitude or from a more distant source. The younger dikes are rare and small within deposits younger than mid-Holocene, indicating the absence of very strong ground shaking since that time. These features may have been caused by the 1811-1812 great New Madrid earthquakes (McNulty and Obermeier, 1999).

The magnitude of the Shoal Creek earthquake probably exceeded M 6. Based on comparison to the magnitude versus maximum distance of paleoliquefaction curves, a reasonable lower limit seems to be about M 6.5 (McNulty and Obermeier, 1999).

The closest major mapped bedrock structure to the energy source for the Shoal Creek earthquake is the Centralia fault, which is associated with the Du Quoin monocline approximately 12 miles to the east (Figure B-1-5). The distribution of paleoliquefaction features associated with this event suggests an energy source west of the Centralia fault.

Other small folds (< 6 miles long) are mapped at similar distances northeast, south, and southwest of the inferred energy center.

Southeastern Missouri (St. Louis area) -- A sand sill, associated dikes, and other liquefaction features along the Big Muddy River formed since 9,070 BC and possibly prior to 4,240 BC (Tuttle et al., 1999). In addition, sand dikes along the Meramec River near St. Louis appear to be prehistoric and to have formed since 13,210 BP. It is possible, but not necessary, that these features formed as a result of the same earthquake responsible for the middle-Holocene liquefaction features in the Shoal Creek region. Tuttle et al. (1999) consider three alternative scenarios for the formation of these features: a M 7.5 earthquake B1-1-4 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT on the Centralia fault; a M 7.0 centered near Germantown, Illinois, near the largest features on Shoal Creek; or a M > 5.2 near St. Louis.

Based on the spatial distribution of prehistoric liquefaction features, mapped faults, and modern seismicity in the region northwest of the New Madrid seismic zone, Tuttle et al.

(1999) suggest that possible paleoearthquake sources include the Valmeyer and Waterloo-Dupo anticlines; Du Quoin monocline; Centralia, St. Louis, and New Madrid faults; and an unidentified source near Shoal Creek and Germantown, Illinois. They conclude, however, that poor age constraints make it difficult to draw regional correlations of liquefaction features and to estimate magnitudes of causative earthquakes, and that any of the alternative scenarios described above could explain the distribution of paleoliquefaction in southern Illinois and southeast Missouri.

1.2 Completeness of Paleoliquefaction Record Previous workers have noted that additional moderate- to large-magnitude earthquakes may have occurred within the southern Illinois basin that cannot be detected with present methods. Obermeier (1998) discusses several factors to be considered, such as the few known liquefaction features that cannot be assigned to one of the eight postulated 2

prehistoric earthquakes, the few known liquefaction features older than 12 ka , the areas that have not been studied, and the lack of liquefiable deposits or local sedimentalogic and hydrologic conditions that affect the magnitude threshold for liquefaction. He concludes that the record is most likely to be complete for earthquakes that are most recent, particularly larger earthquakes. He suggests that probably no prehistoric earthquake larger than M 7 has been overlooked in southern Illinois or southwestern Indiana, but that ten or more Holocene or latest Pleistocene earthquakes of M 6 to 7 might have been overlooked because of lack of nearby liquefiable deposits.

1.3 Estimating Source Areas and Magnitudes of Paleoearthquakes As discussed above, the size distribution of liquefaction features commonly is assumed to reflect the source region of a paleoearthquake. Tuttle et al. (1999) note that empirical data from historical earthquakes (e.g., 1988 Saguenay, Quebec; 1989 Loma Prieta; and 1994 Northridge) show that the distribution of liquefaction features can be irregular and not necessarily centered around or even within the meizoseismal area. Factors that influence the distribution of liquefaction features include earthquake characteristics, such as directivity and focusing of seismic waves, and site conditions, such as liquefaction susceptibility of sediments, local ground motion amplification, and topography. Wheeler and Cramer (2002) note that for hazard calculations, inferring energy centers based on the size and distribution of paleoliquefaction features appears to involve uncertainties of a few tens of kilometers. They consider factors such as the non-uniform distribution of liquefiable deposits, the length of ruptures, particularly for larger events that likely would not result from a point source, directivity effects, and data from prehistoric events in Illinois and 2 ka (thousands of years before present)

DEL-096-REV0 B1-1-5

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT Indiana that show that the widest dikes for individual events are found 6 to 15 miles (10 to 25 km) from the approximate centers of the corresponding liquefaction fields.

1.4 Paleoliquefaction Investigations in the Site Vicinity Both field reconnaissance and detailed field investigations in the vicinity of the EGC ESP Site have been conducted along sections of several larger streams, including the Sangamon River, South Fork of the Sangamon River, and Salt Creek, as well as such smaller streams as Sugar Creek, Kickapoo Creek, Deer Creek, and Lake Fork (McNulty and Obermeier, 1999)

(Figures B-1-2 and B-1-3). These studies provide reasonably good coverage of the region south and west, and to a lesser degree east, of the site. Geomatrix performed reconnaissance-level investigations of rivers and streams north and east of the site to provide more complete coverage of the site region (within a radius of approximately 25 to 35 miles) (Figure B-1-6). Approximately 41 miles of river, including parts of the Mackinaw River, the Sangamon River, and the North and South forks of Salt Creek, were investigated for this study. This fieldwork was performed during a two-week period from September 23 to October 4, 2002. Dr. Stephen Obermeier, a recognized specialist in identifying and characterizing paleoliquefaction features in the central and eastern United States, participated in the field investigations and the interpretation of results.

Formation of seismically induced liquefaction features such as dikes is optimized by the following conditions: (1) liquefiable sediment (preferably a clean sand that has a thickness of 3 feet or greater); (2) an overlying thin, low-permeability soil capgenerally a cap of silt and clay at least 3 and less than 30 feet thick (optimally about 6 to 16 feet); and (3) a shallow water table (optimally less than 16 feet deep). Liquefaction-induced features can form even where the top of the overlying cap is submerged by many feet of water. Obermeier et al.

(2001 and 2002) discuss the basic mechanisms that induce seismic liquefaction features.

Dikes in a fine-grained cap form chiefly by three ground failure mechanisms: hydraulic fracturing, lateral spreading, and surface oscillations. These three mechanisms can occur independently or in combination.

The drainages in the study area are incised into till plains and moraine systems associated with the late Wisconsin-age Lake Michigan lobe (Hansel and Johnson, 1996) (Figures B-1-7 through B-1-9). Deposits of the Wisconsin Episode of glaciation in Illinois record migrating proglacial and glacial environments that are classified into two intertonguing groups. The diamictons, including till and ice-marginal deposits, are classified in the Wedron Group.

The sorted proglacial sediment deposits, primarily loess, eolian sand, lake sediment, and outwash, are classified in the Mason Group (Figures B-1-7 and B-1-9). The larger streams cut through broad valleys where the highest terraces are composed of Wisconsinan glaciofluvial and early Holocene braid-bar deposits of thick, clean sand. The sands are capped by fine-grained overbank and loess material that can reach 10 to 15 feet in thickness.

These highest terraces are generally about 15 feet above the low flow level of the modern streams. Inset into these terraces are younger deposits that include point-bar sediments (primarily sandy gravel, gravelly sand, and sand) from streams having a meandering character (Figures B-1-10 and B-1-11). The sand and gravel deposits, both of braid bars in low terraces and of point bars, usually are capped abruptly by 6 to 15 feet of fine-grained (sandy silt to clayey silt) alluvium, primarily from overbank and channel-fill deposits. The B1-1-6 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT degree of soil development in the fine-grained cap generally reflects the age of the sediments.

In the study area, deposits older than mid-Holocene (> 6 to 7 ka) are recognized by weathering profiles that exhibit oxidation in fine-grained sediments to depths at and below the modern depth of the water table. Obermeier (1998) notes that these weathering profiles were imposed chiefly during the early to middle Holocene, when the climate was generally warmer and drier than at present. This period of warm, dry weather commonly is referred to as the hypsithermic.

The following general approach was used in this study to identify sites having conditions favorable for the development and preservation of paleoliquefaction features.

1. The Quaternary geologic map for the State of Illinois (Lineback, 1979) was used to identify reaches of the larger streams within the study region (~ 40- to 56-mile radius of the site) where deposits of the late Wisconsin Henry Formation have been mapped (Figure B-1-11). The Henry Formation includes generally well-sorted sand and gravel deposited as glacial outwash in outwash plains, as valley train, or as ice-contact sediments. In areas of high groundwater, these sediments would be susceptible to liquefaction. Where the sediments are preserved at or within a few feet (meters) of present low water along current drainages, they provide the most complete geologic record of the presence or absence of late glacial and post-glacial liquefaction.

2. A secondary tool for identifying locations where deposits of latest Pleistocene to early Holocene age may be preserved along present drainages were the U.S.

Department of Agriculture Soil Conservation Survey (Soil Survey) maps for DeWitt, McClean, and Champaign counties. These maps were used to identify more specific locations where well-developed soils might be found in appropriate parent material (loess, glaciofluvial outwash terraces, or early Holocene alluvium).

3. U.S. Geological Survey 7.5-minute quadrangle topographic maps also were reviewed to identify geomorphic surfaces of latest Pleistocene to early Holocene age and to identify gravel pit operations that would provide accessible subsurface exposures.

The Geomatrix field reconnaissance was performed by examining banks of streams and exposures in local sand and gravel pits. Recent paleoliquefaction investigations demonstrate that search of eroded banks of steams, generally from a canoe, is the best means of documenting the presence or absence of paleoliquefaction features. Optimal conditions for finding liquefaction features depend on the size of the stream. For larger streams and rivers (e.g., the Mackinaw River and the lower reaches of the Sangamon River and Salt Creek), the low-water conditions of late fall/early winter after vegetation starts to die generally offer the most continuous exposures. For smaller streams that typically are overgrown with brush and vegetation (e.g., the upper reaches and North Fork of Salt Creek), ideal conditions occur soon after flooding (early spring), when undercutting of the stream creates new bank exposures. The larger streams examined during this study had relatively low water levels, which provided good exposures at and slightly above stream DEL-096-REV0 B1-1-7

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT level. The low water, however, limited access by canoe to the upper reaches of the drainages. Continuous reconnaissance along limited stretches of these reaches was conducted by walking. These relatively continuous reconnaissance-level investigations were supplemented by additional reconnaissance at targeted stream and gravel pit locations easily accessible from roads or bridge crossings.

In an average day, 3 to 5 miles of river bank could be searched. Access by canoe was limited along some heavily wooded sections where log jams were common. The degree of exposure of susceptible deposits varied along the stretches of the rivers studied (Table B-1-2).

Typically, approximately 5 to 7 percent of the banks revealed pre-hypsithermic deposits of mid-Holocene (at least 5,000 years old) to earlier Holocene age (~ 6 to 10 ka). Exposures of older latest Pleistocene alluvium were much more limited. Many bank exposures, particularly the youngest (late Holocene terrace deposits), were masked by a thin veneer of colluvium or slope wash and vegetation. However, the relatively low water levels in larger streams generally provided sufficient exposure of the lowermost 1 to 2 feet, particularly for older (pre-hypsithermic) deposits. The degree of exposure along larger streams and rivers in the study area was comparable to that observed in the Springfield study region. Along smaller streams, however, exposures in the Springfield area, which were investigated shortly after record floods, were more frequent and complete than those observed during this study. Overall, the reconnaissance performed for the EGC ESP project provided a reasonable opportunity to find clastic dikes that had formed throughout the Holocene, and especially since the mid-Holocene (post-hypsithermic) throughout the study area.

The objective of the field reconnaissance was to document the presence or absence of paleoliquefaction features in the vicinity of the EGC ESP Site. Liquefaction features of latest Pleistocene/early Holocene and mid-Holocene age were identified at isolated localities in the study area. The following sections present the criteria for differentiating seismic from nontectonic origins of liquefaction features, followed by descriptions of the features observed and interpretations of the likely origin or triggering event for each feature.

1.4.1 Criteria for Identifying Clastic Dikes of Seismic Origin The clastic dikes and possible sills identified during this investigation exhibit many characteristics in common with seismically induced paleoliquefaction features. However, nonseismic mechanisms may produce clastic dikes or other deformational features in sediments that may be difficult to distinguish from those that represent seismically induced liquefaction. As noted by others (e.g., Tuttle, 2001; Obermeier, 1996 and 1998), nonseismic mechanisms that should be considered include dewatering due to rapid sedimentation and compaction, artesian pressure, and, during floods, piping and diversion of runoff that may produce sand boils (Figure B-1-12).

In addition, in glaciated regions, clastic dikes comprising fluid escape structures, glaciogenic injections, and fluidal and viscous hydraulic expulsions may form in response to subglacial melting and dewatering (Broster, 1991; Dreimanis and Rappol, 1997). Alternative mechanisms postulated to explain formation of these features include Brosters (1991) model involving expulsion of pore water from confined layers and consolidation and fracturing of subglacial sediments by overriding grounded ice. He notes that most deformation structures formed by expulsion reflect initial overriding and compaction of saturated sediments, with this compaction producing hydraulic conditions that favor release of pore B1-1-8 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT water. Correlation with directions of glacial movement, association with glacial facies (e.g.,

termination at the base of overlying till), and infilling by glacial sediments are considered to be glaciogenic signatures useful in differentiating glacial from nonglacial (e.g., earthquake) origins for similar features (Broster, 1991).

As synthesized by Obermeier (1996) and summarized below, criteria for identifying earthquake-induced liquefaction features include:

1. sedimentary characteristics indicative of sudden, strong, upwardly directed hydraulic force of short duration;

2. sedimentary characteristics consistent with case histories of earthquake-induced liquefaction (e.g., the meizoseismal zone of the 1811-1812 New Madrid earthquakes);

3. occurrence of similar features at multiple locations;

4. occurrence in geomorphic settings where the hydraulic conditions described in No. 1 above would not develop under nonseismic conditions; and

5. age data to support both contemporaneous and episodic formation of the features over a large area.

Specific criteria (sedimentary and stratigraphic characteristics of host deposits and material source, conduit morphology, distribution pattern, etc.) that may be used to assess the origin of possible liquefaction features identified in this study are listed in Table B-1-3. These features are described below.

1.4.2 Possible Paleoliquefaction Features Possible liquefaction features were observed at four localities in the study area, three along Salt Creek (Localities SC 25, SC 16, and SC 19/SC 18), and one along the Mackinaw River (Locality M 6) (Figure B-1-6). The characteristics of these features and the stratigraphic relationships and ages of the associated deposits are described in general below. Table B-1-4 summarizes the estimated age and preferred interpretation for each feature.

1.4.2.1 Locality SC 25 Locality SC 25 lies approximately 5-1/2 miles northeast of Farmer City, Illinois, on Salt Creek.

At Locality SC 25, there is an approximately 11 feet-high terrace exposure that extends approximately 300 feet along the left stream bank, providing varying degrees of exposure.

The exposure generally consists of an ~ 8-foot-thick silty loess cap overlying medium to coarse sand and gravel of probable glaciofluvial origin. Within this 300-foot reach, the contact between the sand and overlying silt was exposed intermittently for a cumulative 100 feet, and the upper 6 feet of the silt cap was exposed for a cumulative 180 feet. The Soil Survey of McLean County maps this area as Tama silt loam, a moderately well-drained soil that typically forms on uplands, loess plains, terraces, and outwash plains. They describe a typical Tama silt loam as a very dark gray, friable silt loam from a depth of 0 to 12 inches; underlain by brown friable silty clay loam to a depth of 16 inches; underlain by dark yellowish brown, friable, silty clay loam that is mottled in the lower part extending to 46 inches; followed by a dark yellowish brown, mottled, friable silt loam that extends to 60 DEL-096-REV0 B1-1-9

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT inches and beyond. Field observations noted that the soil at Locality SC 25 is similar to that described as a typical Tama silt loam.

The most recent glacial flooding on the main fork of Salt Creek likely deposited the gravel and sand exposed at the base of this sequence. Leon Follmer of the Illinois State Geological Survey (Personal Communication, October 15, 2002) estimates this event occurred around 17 ka, when the Bloomington moraine (labeled B on Figure B-1-8) was breached near the town of Saybrook, Illinois (Figure B-1-11). Follmer suggests that during this event most of the floodwater drained down the Sangamon River, but overflow likely would have flowed down Salt Creek. The age of 17 ka is consistent with numerical ages and relative ages of various members of the Tiskilwa and Lemont formations from Hansel and Johnson (1996).

The loess that constitutes most of the upper 8 feet of exposure at Locality SC 25 likely was deposited soon (within several hundred years) after the final flood event. Based on the above information and the relatively deeply developed soil, the deposits in exposure at Locality SC 25 is estimated to be approximately 16 to 17 ka.

The silt cap at Locality SC 25 is intruded by 6 sand dikes (Dikes 1 through 6). These dikes range in width from ~ 1/4 inch to 4 inches, and in height above the sand-silt contact from ~1 foot to more than 6 feet. The dikes are described briefly below.

• Dike 1 protrudes approximately 6.3 feet up from the gravelly sand (Photo A in Figure B-1-13) into the silt cap, approaching within 20 inches of the ground surface, where it measures ~ 2 inches wide. This complex dike strikes nearly parallel to the exposure, but also bifurcates, with some branches oriented nearly perpendicular to the wall. It also includes apparently disjointed segments at its lower end (Photo B in Figure B-1-13). Dike 1 can be traced for ~ 13 feet along the stream bank. The dike fill material ranges from clean, oxidized, medium-grained sand near the top, to oxidized gravelly sand near the base.

• Dike 2 occurs high on the stream bank, extending to within 3 feet of the modern ground surface. It consists of two narrow (approximately 0.5-inch-wide),

tabular, subvertical dikes that extend 2 feet above and 6 inches below a horizontal sill that measures 6 inches long by 2 inches thick. The entire feature is filled with oxidized, uniform, fine-grained sand. Dike 2 is exposed well above the sand-silt contact, and therefore is not associated directly with the gravelly sand at the base of the exposure.

• Dikes 3, 4, and 5 each consists of thin (0.25- to 0.5-inch-thick), tabular dikes that extend a short distance (1 to 1.5 feet) up from the gravelly sand into the overlying silt. Each has sharp contacts and is filled with fine- to medium-grained, oxidized sand.

• Dike 6 intrudes the silt unit to within approximately 2.5 feet of the modern ground surface. This complex dike reaches a maximum width of 2.5 inches and is exposed clearly for more than 4 vertical feet. It consists of a prominent subvertical dike oriented at an ~ 60-degree angle to the exposure wall and two narrower (approximately 1-inch-wide) dikes, oriented roughly parallel to the exposure wall, that join the main dike and extend out from it. These secondary dikes can be traced along the exposure wall for a distance as great as 3 feet.

B1-1-10 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT Additionally, a narrow (~ 0.25-inch-wide) dike intrudes the silt approximately 3 feet downstream of the main dike, striking parallel to it. Although the dikes at this locality are roughly tabular in shape, their walls are very irregular. Near the base, the dike filling consists of oxidized gravelly sand, which grades upward to fine sand near the top. The filling includes irregular, angular clasts of silt (wall rock) that measure as much as 1 inch in diameter. This dike system could not be traced clearly into the gravelly sand below, because much of the lower part of the exposure was covered, but it does approach to within 1.5 feet of the lower unit.

The dikes observed at Locality SC 25 display many characteristics common in seismic liquefaction features (Table B-1-3). They are filled with sand; have sharp, irregular side contacts; widen with depth; occur in clear association with the underlying gravelly sand (except for Dikes 2 and 6); and are generally tabular in plan view, although Dikes 1 and 6 bifurcate in multiple directions. The high degree of weathering of the material within the dikes is interpreted as evidence that they were formed during the latest Pleistocene (between approximately 10 and 17 ka).

A seismic origin is preferred for the clastic dikes observed at this locality. The locality is not in an area where high artesian pressures are known to occur (Leon Follmer, Illinois Geological Survey, personal communication, October 4, 2002), nor in a topographic setting that could produce localized artesian conditions. The features post-date the most recent glacial advance at this location and clearly are not related to glaciotectonic processes. The features at SC 25 meet criteria 1, 2, and 4 outlined above. However, the available data do not fulfill criterion 3 (the occurrence of similar features at multiple sites) or criterion 5 (contemporaneous and episodic formation of features over a large area). The relatively limited extent of exposure of older deposits in the study area cannot provide an amount of evidence comparable to that found for the Springfield event, for example, which is evidenced by multiple dikes at ten localities and shows a well-defined pattern over a distance of at least 22 miles (35 km) from the inferred energy center.

1.4.2.2 Locality SC 16 Locality SC 16 is approximately 2.5 miles northeast of Farmer City on Salt Creek. This Locality provided approximately 20 feet of intermittent exposure of a terrace that is approximately 8 feet above creek level. The stratigraphy measured down from the ground surface includes ~ 4 feet of very dark brown silty clay to clayey silt that grades downward into dark brown sandy gravel with clay. Maximum clast size in the sandy gravel is ~1 inch.

Sandy gravel continues to a depth of 7 feet, where it sharply overlies bluish gray clayey silt that contains minor fine sand. A hand-auger hole revealed that this unit extends to at least 11 feet depth and coarsens downward, grading to sandy silt with clay and gravel at that depth. The entire unit is stiff to very stiff. The blue-gray color suggests that this silty unit is gleyed (reduced). The Soil Survey of De Witt County depicts the soil at this Locality as the Sawmill silty clay loam, a poorly drained soil that typically forms on flood plains that occasionally are flooded for brief periods from March through May.

This exposure is interpreted to be an overbank flood sequence and estimated to be middle to late Holocene in aged, based on the lack of significant soil development and lack of oxidation in the part near the modern water level (Table B-1-4).

DEL-096-REV0 B1-1-11

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT This clayey silt unit is interrupted by two possible liquefaction features, Dikes 1 and 2 (Table B-1-4). Dike 1 extends from the water surface to the upper contact of the clayey silt (approximately 1 foot) and ranges from 1 to 1.5 inches wide. A bulbous protrusion extends laterally 6 inches from the main dike at the top of the clayey silt. Irregular pockets of clean sand in the overlying sandy gravel suggest that Dike 1 may continue upward through this unit. The clayey silt observed for approximately 1 foot on either side of Dike 1 is black. The margins of this black clayey silt are sharp to very sharp.

Several characteristics suggest that Dike 1 represents a sudden injection originating from below: its side contacts are sharp and irregular; its width remains relatively constant with depth (over the short distance that could be observed); and it is filled with clean, medium-to coarse-grained gravelly sand. These characteristics are consistent with seismically induced features formed during liquefaction by hydraulic fracturing (Table B-1-3).

Dike 2 is a very thin (averaging 1/4 inch wide), tabular fracture filled with clean, coarse-grained sand. The lack of clay in the fracture fill suggests that it may have resulted from liquefaction. The clayey silt found for approximately 1 foot on either side of Dike 2 is black.

The margins of this black clayey silt are sharp to very sharp.

An approximately 12-foot-long section of the stream bank immediately downstream of Dike 2 has slumped, bringing grass and the uppermost brown silty clay down to stream level.

This section overlies an ~ 1.5-foot-thick lens of gravelly sand with cobbles at the upper contact of the blue-gray clayey silt. The proximity of the slump block and Dike 2 suggests that Dike 2 may be related to the slumping, either because sand was dragged along the slide plane during slumping, or because the slumping resulted from liquefaction. The black clay surrounding Dike 2 could be interpreted as part of the slide plane. These interpretations do not, however, explain the presence of Dike 1 approximately 8 feet upstream.

The preferred explanation for Dikes 1 and 2 is that they both resulted from seismic liquefaction and are unrelated to bank slumping. The slump block at Locality SC 16 appears young (between a few months and a few years old), whereas Dikes 1 and 2 appear to be at least several hundred years old. The preferred interpretation for the origin of the black clayey silt surrounding both dikes is chemical alteration (oxidation?) by water flowing through the sandy dikes.

1.4.2.3 Locality SC 19/SC 18 Locality SC 19 lies approximately 1.5 miles northeast of Farmer City on Salt Creek. This locality includes approximately 100 feet of high-quality exposure that was as much as 6 feet high above stream level. The exposed sequence includes approximately 4.5 feet of predominantly clay to clayey silt overlying interbedded sand and silt that extend to the water surface and below. The Soil Survey of DeWitt County shows the soil at this site to be Sawmill silty clay loam, a poorly drained soil that typically forms on flood plains that occasionally are flooded for brief periods from March through May.

A general field description of the soil profile at Locality SC 19 includes a 0.3-foot-thick black A horizon overlying a dark brownish gray silty clay loam AB horizon that extends to a depth of 1.9 feet. Below this layer, a dark brownish gray blocky clay Bt horizon continues to a depth of 2.8 feet, followed by a red, structureless clay Bt horizon to a depth of 3.6 feet.

Below this layer, a slightly weathered silty clay loam B horizon having a weak prismatic B1-1-12 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT structure and very thin clay coatings on peds extends to a depth of 4.0 feet. A light-brownish gray silt loam Cox horizon having abundant filled animal burrows extends to a depth of 4.7 feet, followed by medium-gray silt with sand lenses that continues below the water surface, which was at a depth of 5.8 feet.

This exposure is interpreted to be an overbank flood sequence, and its age estimated to be mid-Holocene based on its moderate degree of soil development and lack of deep oxidation near the modern water level (Table B-1-4).

The upper section of the exposure (the 4.5 feet of clay and clayey silt) represents a nonliquefiable cap. This cap is interrupted by two possible liquefaction features, Dikes 1 and 2 (Table B-1-4; Figure B-1-14). Both dikes are similar in size and shape, extending approximately 12 inches up from the interbedded sand and silt at the base of the exposure into the overlying silty clay loam. Both dikes reach a maximum width of 1.5 inches.

Several characteristics suggest that both Dikes 1 and 2 represent injection features. Their side contacts are sharp and irregular; they widen with depth; they are filled with clean fine-to medium-grained sand; the sand filling them fines upward; they include clayey silt clasts likely torn from their walls; and they extend upward from the sandy source material (in clear association with a source bed). These characteristics are consistent with those exhibited by seismically induced paleoliquefaction features (Table B-1-3).

Locality SC 18, located approximately 1,000 feet upstream of SC 19 on Salt Creek, includes approximately 15 feet of poor to moderate exposure that in places extends ~ 8 feet above stream level. The upper 4 to 5 feet of this exposure are covered with overhanging grasses and brush, which were not cleared. Below this section, a dark grayish brown silt loam extends to and below the water surface. The Soil Survey of DeWitt County, Illinois, maps Orion silt loam at this Locality. The Orion silt loam typically forms on floodplains that are somewhat higher than those where Sawmill soil is found, but that may be flooded for brief periods from March through May. The Orion silt loam typically includes a buried soil at an average depth of 38 to 40 inches.

This exposure is interpreted as an overbank flood sequence. Initially its age was estimated to be middle to late Holocene based on its lack of significant soil development and lack of oxidation in the part near the modern water level. However, dating of two radiocarbon samples from these deposits (7935_SC-19-2-1 and 7935_SC-19-2-2) yielded radiocarbon ages of 9,550 +/- 40 yr BP and 10,230 +/- 40 yr BP (Cal BC 9150 to 8750 and Cal BC 10,390 to 9780),

respectively, suggesting the deposits are early Holocene (Table B-1-4; Exhibit 1).

Several irregular bodies of clean fine sand interrupt this sequence between 6 and 12 inches above the water level. The largest of these bodies reaches 3 inches in thickness and approximately 4 feet in length. These features may reflect liquefied injection, or may represent deformation of a fluvial sand lens, possibly due to sediment loading.

1.4.2.4 Locality M 6 Locality M 6 lies approximately 6 miles west of Colfax, Illinois, on the Mackinaw River. At this locality approximately 150 feet of continuous exposure that was approximately 12 feet in height above river level was found. The stratigraphy includes approximately 3 feet of soft clayey silt loess overlying 3.5 feet of loose fluvial sand, with a gravel lag at the base.

DEL-096-REV0 B1-1-13

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT These fluvial deposits rest unconformably on a thin (0.5- to 1-foot-thick) bed of very dense, fissile, oxidized, cross-bedded sandy silt. The fissile structure of this unit indicates that it was overridden by a glacier at some point. This sandy silt overlies a very hard, dark grayish brown, pebbly sandy clay till that was present down to the present water level of the river.

Pebbles and cobbles, scattered sparsely throughout the till, sometimes reach 0.5 foot in diameter. The Soil Survey of McLean County shows the soil at this Locality as Fox silt loam, a well-drained soil that typically forms from thin sheets of loess overlying sands and gravels on stream terraces.

A field description of the soil developed at this exposure includes 0.5 foot of a very dark grayish brown, friable, organic, silt loam A horizon at the surface, followed by a friable, grayish brown, silty clay loam BE horizon that extends to a depth of 0.9 feet, followed by a firm, dark yellowish brown, silty clay loam to a clay loam Bt horizon that has a weak, fine, prismatic structure that extends to the base of the loess at a depth of 3.2 feet. Below the loess, a friable, dark brown, sandy clay loam Bt horizon having a massive structure and abundant manganese oxide staining extends to 4.8 feet depth, followed by a friable, dark brown loamy sand that extends to the unconformity at the base of the fluvial deposits, at 6.4 feet depth. Below the unconformity, a firm, fissile, yellowish brown, oxidized, sandy clay loam Cox horizon extends to a depth of 6.9 feet, followed by a hard, dark grayish brown pebbly sandy clay to clayey sand C horizon that extends to the base of the exposure at a depth of 10.5 feet.

Two prominent dikes intrude the till in the lower half of this exposure. Dike 1 comprises a vertical dike that reaches a maximum of 4 inches in width and strikes parallel to the bank exposure. A secondary dike extends outward in both directions from the southern (downstream) end of this dike, striking approximately perpendicular to the bank. This secondary part of Dike 1 reaches 2 inches in width and extends from near the water surface

(~ 6 inches up) to approximately 2 feet above the main body of Dike 1 (approximately 5 feet in total length). Another secondary dike extends outward from Dike 1 toward and approximately perpendicular to the bank (resulting in an apparent vertical dip). This vertical part of Dike 1 can be traced for 4 feet up the bank exposure, until it is truncated by the dense, fissile sandy silt unit (Figure B-1-15, upper photo). This dike maintains a fairly constant width of 2 to 3 inches, with short (3- to 6-inch-long) branches extending outward from it (Figure B-1-15, lower photo). Dike 1 has very sharp, very irregular contacts and is filled with very fine- to medium-grained sand. The fill material is banded parallel to the walls of Dike 1, with bands defined by changes in grain size and color (caused by differences in the lithology of the sand grains).

A hand-auger hole advanced near Dike 1 showed that the silty clay till extends to a depth of at least 17.5 feet below ground surface at the top of the exposure (10.5 feet below the top of the till).

Dike 2 intrudes the till approximately 100 feet south of Dike 1. Like Dike 1, this dike reaches a maximum width of 4 inches and is truncated at its top by the dense, fissile sandy silt unit.

Dike 2 is filled with medium-grained sand, has sharp contacts, and can be traced approximately 14 feet along and down the bank. This feature is tabular, strikes an approximately 20-degree angle to the bank exposure, and is vertical in cross section.

B1-1-14 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT Both Dikes 1 and 2 exhibit several characteristics typical of seismic liquefaction features (Table B-1-3). Both clearly resulted from liquefied sand injected into the till from below.

However, both Dikes 1 and 2 are truncated by the dense sandy silt deposit, which owes its fissile structure to compaction beneath a glacier. This relationship implies that both dikes were overridden by a glacier. Although clastic dikes commonly are formed by seismic liquefaction, they may also result from increased pore pressure, increased loading, drag, and compression associated with the advance and deflation of glaciers (Broster, 1991; Dreimanis, 1992; Dreimanis and Rappol, 1997). Therefore, glaciotectonic deformation is as likely a cause for Dikes 1 and 2 at Locality M 6 as is seismic shaking.

1.4.3 Evidence for the Absence of Paleoliquefaction Obermeier et al. (2001 and 2002) discuss the major factors that identify the severity of liquefaction and describe how to consider those factors when evaluating evidence for the presence or absence of strong shaking in a region. They note that throughout the meizoseismal area of a very strong earthquake, for which the Modified Mercalli Intensity (MMI) value is IX or higher, liquefaction features should abound even where liquefaction susceptibility is only moderate. Some dikes almost certainly exceeding 1 foot in width, and many smaller ones, should be discovered during a reasonably thorough search of the region. They suggest that in areas of moderate liquefaction susceptibility affected by earthquakes of MMI VII to VIII, small liquefaction features may be sparse, but should be numerous enough that some would be discovered during an examination of tens of miles of stream banks.

Moderate liquefaction susceptibility implies a medium relative density of the soil material, a water table within 10 to 15 feet of the surface, and a cap less than 25 to 30 feet thick (Obermeier et al., 2001 and 2002). A lower limit of moderate susceptibility requires a bed of silty sand, sand, or gravelly sand (generally less than about 40 percent gravel) that is at least 5 feet thick, capped by a minimum of approximately 1.5 feet of lower-permeability sediment. Where a cap is underlain by medium-grained sand or coarser sediment, the water table would have to be at or above the base of the cap at the time of the earthquake to produce liquefaction features. Otherwise, the high permeability of the material beneath the cap would permit dissipation of pore water pressure, leaving no evidence of paleoliquefaction. These authors, however, cite examples of liquefaction features produced from source beds much thinner than 10 feet in the meizoseismal zone of an ~ M 7 earthquake.

In the study area, deposits of latest Pleistocene and Holocene age were laid down by moderate to large streams and generally fit the criteria used to define moderate susceptibility. This level of susceptibility applies to stream deposits of both glaciofluvial braid-bar and Holocene point-bar origins. These deposits should have been at least moderately susceptible to liquefaction throughout the time since their deposition, with the possible exception of the hypsithermic period (middle Holocene; Obermeier, 1998), when the water table was several feet lower than in modern time. During our field reconnaissance, particular care was taken to identify and examine early Holocene and latest Pleistocene deposits. Exposures of pre-hypsithermic deposits were noted on field maps. In addition to the distinctive oxidation mottling observed at and below modern water levels, these deposits generally appeared to be more resistant to erosion than late Holocene DEL-096-REV0 B1-1-15

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT deposits, forming a blockier appearance at and below the water level. Based on previous experience, Dr. Obermeier judged that even small clastic dikes would be visible in exposures of these deposits and that sufficient deposits were observed to document the absence of a Springfield-type event centered in the study region since the hypsithermic period (in the past 6 to 7 ka).

Deposits of latest Pleistocene age were noted on field maps if judged to have been moderately susceptible to liquefaction throughout much of the Holocene. Deposits at some localities likely pre-date the loess deposits at SC 25, which were intruded by clastic dikes (Table B-1-5). Evidence for the absence of paleoliquefaction features at these localities is described in Table B-1-5; the locations of the localities are shown on Figure B-1-6. The reported absence of paleoliquefaction features in the Mahomet gravel pits (Locality S 14; Figure B-1-16) offers the strongest evidence for the absence of significant ground shaking in this part of the study region. The limited number and small extent of the exposures of late Pleistocene fluvial and loess deposits observed elsewhere in the region make it more difficult to evaluate the significance of the features observed at SC 25. However, no evidence comparable to that produced by the Springfield eventthe multiple dikes at ten localities and a well-defined regional pattern over an area 22 miles (35 km) in diameter -

was found.

1.4.4 Conclusions Field reconnaissance conducted for this study obtained additional information regarding the prehistoric record of earthquakes within the near region (approximately 25- to 30-mile radius) of the EGC ESP Site. That information is summarized below.

1. No evidence for a post-hypsithermic (post-mid-Holocene) earthquake comparable to the postulated Springfield event (McNulty and Obermeier, 1999) was observed in the study area. Sufficient exposures of pre-hypsithermic (> 6 to 7 ka) deposits were observed to demonstrate the absence of paleoliquefaction features indicative of an energy source for a comparable event (estimated M 6.2 to 6.8) in the site vicinity. Radiocarbon ages from samples originally estimated to be mid-Holocene yielded ages of approximately 10 ka (see Section 4.2.3),

suggesting that the absence of significant events may extend back even further into the early Holocene.

2. Isolated features of mid-Holocene and latest Pleistocene/early Holocene age were observed in the study area that could be interpreted as evidence of seismically induced paleoliquefaction. Features of probable mid- to early Holocene age were observed at two localities along Salt Creek (SC 16 and SC 19/SC 18), approximately 1.5 to 2.5 miles northeast of Farmer City and approximately 11.5 to 13 miles from the EGC ESP Site. Characteristics of the dikes exposed at these locations are consistent with seismic liquefaction features.

Assuming that these features are seismically induced, their small scale and the lack of evidence for similar features elsewhere in the study area suggest either a more distant source or a low-magnitude event (at or close to threshold of paleoliquefaction, estimated to be MMI VI or VII). Radiocarbon ages for samples from Locality SC 19 indicate that these features formed after 9550 +/- 40 yr. BP (CAL BC 9150 to 8750).

B1-1-16 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT

3. Older features (clastic dikes that cut the post-glacial silt cap [probably early post-glacial loess deposits]) were observed at Locality SC 25, approximately 5.5 miles northeast of Farmer City and approximately 17 miles from the EGC ESP Site.

Those features post-date loess deposits estimated to be ~ 16 to 17 ka. Based on weathering and soil development of the clastic dikes and silt cap, and the height of the water table at the time of formation (~ 3 feet higher than at present), the dike features are inferred to be latest Pleistocene to early Holocene (< 17 to 10 ka). Sedimentary and stratigraphic characteristics of host deposits and material sources, as well as conduit morphology, are consistent with a seismic origin for these features. It is estimated that, if seismically triggered, clastic dikes observed at this location would imply Modified Mercalli Intensity values of at least VII to VIII.

4. Clastic dikes observed in till deposits at Locality M 6, approximately 29 miles north-northeast of the EGC ESP Site, appear to have formed during the latest glacial advance in that region (~ 17.7 +/- 1 ka). The event that triggered injection of the clastic dikes at this location is uncertain. Both dewatering related to glacial processes and seismic shaking are possible mechanisms.

5. No evidence for paleoliquefaction of an age similar to that observed at Locality SC 25 was identified at any other locality, although the possibility that clastic dikes at Locality M 6 formed contemporaneously with the features at Locality SC 25 cannot be precluded at this time given the uncertainties in age estimates. The limited extent of exposure of older deposits makes it difficult to document the well-defined regional pattern needed to estimate a magnitude and location for this event. Susceptible deposits of estimated latest Pleistocene age at Localites M 2, S 6, S 14, and NSC 1 show evidence for no liquefaction. These localities should have been favorable sites for liquefaction throughout much of the latest Pleistocene and Holocene, with the possible exception of NSC 1, where the fluvial deposits may not have been below the water table throughout the Holocene. Deposits at these sites therefore provide reasonable evidence for the absence of significant ground shaking since latest Pleistocene/early Holocene time, and may limit the geographic extent of liquefaction that can be correlated with the features observed at Locality SC 25. The extensive Mahomet gravel pit exposures (S 14), in particular, provide evidence for the absence of strong ground motion that would produce significant liquefaction since deposition of the upper silt approximately 17 to 18 ka.

The results of this study suggest that there have been no repeated moderate- to large-magnitude (comparable to the postulated Springfield-type) events in the vicinity of the EGC ESP Site in latest Pleistocene to Holocene time. The late Holocene record in particular is sufficient to demonstrate the absence of such events in the past approximately 6 to 7 ka. The significance of the latest Pleistocene/early Holocene features recorded at Locality SC 25 is less certain. There is insufficient information to accurately estimate a location or magnitude for a postulated seismic source. The features, however, suggest that the range of maximum magnitude assigned to a random background earthquake in the probabilistic seismic hazard analysis for the EGC ESP Site should include events comparable to that estimated for the Springfield earthquake (i.e., M 6.2 to 6.8).

DEL-096-REV0 B1-1-17

CHAPTER 2 References Atkinson, G.M. and D.M. Boore. Some Comparisons between Recent Ground-Motion Relations. Seismological Research Letters. Vol. 68, No. 1. pp. 24-40. 1997.

Bauer, R.. Illinois Geological Survey. Personal Communication. November 21, 2002.

Broster, B.E. Glaciotectonic Deformation in Sediment and Bedrock, Hat Creek, British Columbia. Geographie Physique et Quaternaire. Vol. 45, No. 1. pp. 5-20. 1991.

Campbell, K.W. Development of Semi-Empirical Attenuation Relations for the CEUS, USGS Annual Technical Summary. Available: http://erp-web.er.usgs.gov/reports/annsum/vol43/ni/g0011.pdf. 2001.

Chester, J. S., and M.P. Tuttle. Paleoseismology in the Cache River Valley, Southern Illinois. Technical Progress Report Submitted to U.S. Geological Survey National Earthquake Hazards Reduction Program. U.S.G.S. External Grant Nos. 1434-HQ-98-GR-00013 and 1434-HQ-98-GR-00015. 2000.

Dreimanis, A. Downward Injected Till Wedges and Upward Injected Till Dikes, in Robertson, A.M., M., B. Ringberg, U. Miller, and L. Brunnberg (eds.). Quaternary Stratigraphy, Glacial Morphology, and Environmental Changes. Geological Survey of Sweden Research Paper. Serie Ca81. pp.91-96. 1992.

Dreimanis, A., and M. Rappol. Late Wisconsinan Sub-Glacial Clastic Intrusive Sheets along Lake Erie Bluffs, at Bradtville, Ontario, Canada. Sedimentary Geology. Vol. 111. pp. 225-248.

1997.

Folmer, Leon. Illinois Geological Survey. Personal Communication. October 4, 2002.

Folmer, Leon. Illinois Geological Survey. Personal Communication. October 15, 2002.

Green, R.A. Energy-Based Evaluation and Remediation of Liquefiable Soils. Ph.D. Dissertation.

Virginia Polytechnic Institute and State University. 394 pp. 2001.

Hajic, E.R., and M.D. Wiant. Dating of Prehistoric Earthquake Liquefaction in Southeastern and Central Illinois. Illinois State Museum Society. Report to the U.S. Geological Survey. 57 pp. 1997.

Hajic, E.R., M.D. Wiant, and J.J. Oliver. Distribution and Dating of Prehistoric Earthquake Liquefaction in Southeastern Illinois, Central U.S. Final Technical Report Submitted to the U. S. Geological Survey National Earthquake Hazards Reduction Program. Contract No.

1434-93-G-2359. 33 pp. 1995.

Hansel, A.K., and W.H. Johnson. Wedron and Mason Groups: Lithostratigraphic Reclassification of Deposits of the Wisconsin Episode, Lake Michigan Lobe Area.

Department of Natural Resources Illinois State Geological Survey Bulletin. Vol. 104. 116 pp. 1996.

DEL-096-REV0 B1-2-1

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT Harrison, R.W., and A. Schultz. Tectonic Framework of the Southwestern Margin of the Illinois Basin and its Influence on Neotectonism and Seismicity. Seismological Research Letters. Vol. 73, No. 5. pp. 698-731. 2002.

Hildenbrand, T. G., J. H. McBride, and D. Ravat. The Commerce Geophysical Lineament and its Possible Relation to Mesoproterozoic Igneous Complexes and Large Earthquakes in the Central Illinois Basin. Seismological Research Letters. Vol. 73, No. 5, pp. 640-659. 2002.

Illinois State Geological Survey. Structural Features in Illinois: Line Features (Axial or Flexure) of Anticlines, Synclines and Monoclines: ISGS GIS Database

• path_name_suppressed*/structclines, Illinois State Geological Survey, Champaign, IL.

1995. Available: http://www.isgs.uiuc.edu/nsdihome/outmeta/structclinesf.html.

[08.08.2002]

Indiana Geological Survey. Structural_Features_SW: Structural Features of Southwestern Indiana (Indiana Geological Survey, Line Shapefile). 2001. Available:

http://igs.indiana.edu/arcIMS/southwest/Metadata/Structural_Features_sw.html.

[29.05.2003].

Lineback, J.A. Quaternary Deposits of Illinois. Illinois State Geological Survey Map, scale 1:500,000. 1979.

McNulty, W.E., and S.F. Obermeier. Liquefaction Evidence for at least Two Strong Holocene Paleoearthquakes in Central and Southwestern Illinois, USA. Environmental and Engineering Geoscience. Vol. 5, No. 2. pp. 133-146. 1999.

Munson, P.J., S.M. Obermeier, C.A. Munson, and E.R. Hajic. Liquefaction Evidence for Holocene and Latest Pleistocene in the Southern Halves of Indiana and Illinoisa Preliminary Overview. Seismological Research Letters. Vol. 68, No. 4. pp. 523-536. 1997.

Nelson, W.J. Structural Features in Illinois. Illinois State Geological Survey Bulletin 100. 144 pp. 1995.

Obermeier, S.F. Using Liquefaction-Induced Features for Paleoseismic Analysis. in McCalpin, J.P. (ed.). Paleoseismology. Academic Press Inc., San Diego. pp. 331-396. 1996.

Obermeier, S.F. Liquefaction Evidence for Strong Earthquakes of Holocene and Latest Pleistocene Ages in the States of Indiana and Illinois, USA. Engineering Geology. Vol. 50. pp.

227-254. 1998.

Obermeier, S.F., U. S. Geological Survey, Emeritus, Reston, Virginia; EqLiq Consulting.

Written (electronic mail) Communication to Kathryn Hanson. January 10, 2003.

Obermeier, S.F., U. S. Geological Survey, Emeritus, Reston, Virginia; EqLiq Consulting.

Written (electronic mail) Communication to Kathryn Hanson. May 13, 2003.

Obermeier, S.F., U. S. Geological Survey, Emeritus, Reston, Virginia; EqLiq Consulting.

Personal Communication. August 9, 2003.

Obermeier, S.F., N.K. Bleuer, C.A. Munson, P.J. Munson, W.S. Marin, K.M. McWilliams, D.A. Tabaczynski, J.K. Odum, M. Rubin, and D.L. Eggeert. Evidence of Strong Earthquake B1-2-2 DEL-096-REV0

SSAR FOR THE EGC EARLY SITE PERMIT ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT Shaking in the Lower Wabash Valley from Prehistoric Liquefaction Features. Science. Vol.

251. pp. 1061-1063. 1991.

Obermeier, S.F., J.R. Martin, A.D. Frankel, T.L. Youd, P.J. Munson, C.A. Munson, and E.C.

Pond. Liquefaction Evidence for One or More Strong Holocene Earthquakes in the Wabash Valley of Southern Indiana and Illinois, with a Preliminary Estimate of Magnitude. U.S.

Geological Survey Professional Paper 1536. 27 pp. 1993.

Obermeier, S.F., E.C. Pond, and S.M. Olson, with contributions by Green, R.A., T.D. Stark, and J.D. Mitchell. Paleoliquefaction Studies in Continental Settings: Geologic and Geotechnical Factors in Interpretations and Back-Analysis. U.S. Geological Survey Open-File Report 01-29. 75 pp. 2001.

Obermeier, S.F., Pond, E.C., Olson, S.M, and Green, R.A. Paleoliquefaction Studies in Continental Settings. in Ettensohn, F.R., N. Rast, and C.E. Brett (eds.). Ancient Seismites.

Boulder, Colorado. Geological Society of America Special Paper 359. pp. 13-27. 2002.

Olson, S.M., R.A. Green, and S.F. Obermeier. "Geotechnical Analysis of Paleoseismic Shaking Using Liquefaction Features: Part I. Major Updating of Analysis Techniques." U. S.

Geological Survey Open-File Report 03-307. 33 pp. 2003.

Pond, E.C., and J.R. Martin. Estimated Magnitudes and Accelerations Associated with Prehistoric Earthquakes in the Wabash Valley Region of the Central United States.

Seismological Research Letters. Vol. 68, No. 4. pp. 611-623. 1997.

Somerville, P., N. Collins, N. Abrahamson, R. Graves, and C. Saikia. Ground Motion Attenuation Relations for the Central and Eastern United States. Final Report to U.S.

Geological Survey. 2001.

Su, W.J., and J.H. McBride. Final Technical ReportStudy of a Potential Seismic Source Zone in South-Central Illinois (abs.). Technical Report Submitted to the U.S. Geological Survey under USGS External Grant Number 99HQGR0075. 1999.

Toro, G., N. Abrahamson, and J. Schneider. Model of Strong Ground Motions from Earthquakes in the Central and Eastern North America: Best Estimates and Uncertainties.

Seismological Research Letters. Vol. 68, pp. 41-57. 1997.

Tuttle, M.P. The Use of Liquefaction Features in Paleoseismology: Lessons Learned in the New Madrid Seismic Zone, Central United States. Journal of Seismology. Vol. 5. pp. 361-380.

2001.

Tuttle, Martitia P. M. Tuttle & Associates. Written (Electronic) Communication. February 11, 2003.

Tuttle, M.P., J. Chester, R. Lafferty, K. Dyer-Williams, and R. Cande. Paleoseismology Study Northwest of the New Madrid Seismic Zone. U.S. Nuclear Regulatory Commission Report NUREG/CR-5730. 96 pp. 1999.

University of Missouri, 1991. Department of Geology. State Geologic Faults of Missouri.

Available: http://msdis.missouri.edu/html/sfault.html. [15.08.2002].

DEL-096-REV0 B1-2-3

ATTACHMENT 1 TO APPENDIX B - SEISMIC HAZARDS REPORT FOR THE EGC EARLY SITE PERMIT SSAR FOR THE EGC EARLY SITE PERMIT Wheeler, R.L., and C.H. Cramer. Updated Seismic Hazard in the Southern Illinois Basin:

Geological and Geophysical Foundations for Use in the 2002 USGS National Seismic-Hazard Maps. Seismological Research Letters. Vol. 73, No. 5. pp. 776-791. 2002.

B1-2-4 DEL-096-REV0

TABLE B-1-1 LIQUEFACTION EVIDENCE FOR PREHISTORIC EARTHQUAKES IN THE SOUTHERN ILLINOIS BASIN Seismic Hazards Report for the EGC ESP Site Name of Size and Distribution Estimated Earthquake(s) Location of Features Age1 Magnitude Reference(s)

Lower Wabash Valley Vincennes-Bridgeport ~ 15 miles (25 km) west of 0.5-m dike width; 150- ~ 6,100 +/- 200 yr. BP M 7.5 to 7.8 Munson et al. (1997);

Vincennes, Indiana km maximum distance (magnitude-bound, Obermeier (1998);

from inferred energy cyclic stress, and energy- Obermeier et al. (1993);

center stress methods)

M 7.7 to 7.8 Pond and Martin (1997)

(energy-stress)

(see comments below by S. Obermeier, 10 January, 2003 and 13 May, 2003 M ~7.2-7.3 (based on Olson et al. (2003)-

recently developed magnitude-bound curve CEUS magnitude-bound Obermeier et al. (1993)-

curve using Mmax of M distance 7.6-7.7 for New Madrid events and distance of 89 km)

DEL-096-REV0 B1.T-1

TABLE B-1-1 LIQUEFACTION EVIDENCE FOR PREHISTORIC EARTHQUAKES IN THE SOUTHERN ILLINOIS BASIN Seismic Hazards Report for the EGC ESP Site Name of Size and Distribution Estimated Earthquake(s) Location of Features Age1 Magnitude Reference(s)

M 7.0+ to 7.5 (cyclic S. Obermeier (person.

stress method using site- commun., January 10, 2003) specific geotechnical (results from studies by R.

data and most recent Green, S. Olson, and S.

magnitude scaling Obermeier to be published in factors, suggested by E. US Geological Survey Open-Idriss and L. Youd) File Report, in preparation M ~7.5 (2003 reanalysis using energy-based solution and NEHRP recommended procedures for calculating ground motions)

The energy-stress S. Obermeier (person.

method used by Pond commun., 13 May, 2003) and Martin (1997) is (conclusions from studies by flawed (energy R. Green, S. Olson, and S.

attenuation relations Obermeier to be published in used should not be used US Geological Survey Open-for liquefaction analysis) File Report, in preparation) and the results are not reliable.

Skelton-Mt. Carmel ~ 25 miles (40 km) 0.5-m dike width; 50- to ~ 12,000 +/- 1,000 yr. BP M 7.1 to 7.2 (magnitude- Munson et al. (1997); Pond southwest of Vincenees, 60-km maximum distance bound); and Martin (1997);

Indiana from inferred energy M 7.3 (energy-stress) Hajic and Wiant (1997);

center Obermeier (1998)

B1.T-2 DEL-096-REV0

TABLE B-1-1 LIQUEFACTION EVIDENCE FOR PREHISTORIC EARTHQUAKES IN THE SOUTHERN ILLINOIS BASIN Seismic Hazards Report for the EGC ESP Site Name of Size and Distribution Estimated Earthquake(s) Location of Features Age1 Magnitude Reference(s)

Single site near Iona, 22 miles (35 km) southeast Very small and restricted 4,000 +/- 500 yr. BP Near threshold Munson et al. (1997);

Indiana of Vincennes, Indiana, (probably limited to 5 km) (M ~5.5 - 6.0 to < 6.7) Obermeier (1998) near Iona (location not shown on Figure B-1-2)

Central and Southern Indiana Vallonia East Fork valley ~ 60 miles 0.5 m; 36 km maximum 3,950 +/- 250 yr. BP M 6.9 (magnitude- Munson et al. (1997);

(100 km) east of the distance from inferred bound) Obermeier (1998);

Wabash Valley seismic energy center M 7.1 (energy-stress) Pond and Martin (1997) zone Martinsville-Waverly ~ 18 to 30 miles (30 to 50 0.15 to 0.5 m; 28 km Between 8,500 and 3,500 yr. M 6.8 (magnitude- Munson et al. (1997);

km) southwest of maximum distance from BP bound); Obermeier (1998)

Indianapolis, Indiana inferred energy center M 6.9 (energy-stress) Geotechnical analyses (location poorly demonstrate that these constrained) features are not associated with the M ~ 7.5 Vincennes earthquake Single site near Elnora, 37 miles (60 km) east- Limited areal extent 2,000 +/- 500 yr. BP M 5.5 to 6.0 Munson et al. (1997);

Indiana northeast of Vincennes Obermeier (1998)

(location not shown on Figure B-1-2)

Single site along Indian ~ 30 mi (50 km) south- Single site ~ 20,000 yr. BP Unknown Munson et al. (1997);

Creek, Indiana southwest of Indianapolis Obermeier (1998)

(location not shown on Figure B-1-2)

Ohio River regionAbsence of paleoliquefaction in Ohio River sediments along the Indiana-Kentucky and Illinois-Kentucky borders suggests that this area has not experienced severe ground shaking in the past 4,500 years (Munson et al., 1997). However, those authors suggest that a 5- to 6-m-thick clay cap may have kept sand dikes from penetrating to levels above the current maintained water level of the Ohio River.

DEL-096-REV0 B1.T-3

TABLE B-1-1 LIQUEFACTION EVIDENCE FOR PREHISTORIC EARTHQUAKES IN THE SOUTHERN ILLINOIS BASIN Seismic Hazards Report for the EGC ESP Site Name of Size and Distribution Estimated Earthquake Location of Features Timing1 Magnitude Reference Central and Southern Illinois Springfield ~ 22 miles (35 km) Maximum dikes width 0.4 One, possibly two, events M 6.2 to 6.8 Hajic et al. (1995); McNulty northeast of Springfield, m; 35 km maximum between 5900 and 7400 yr. M 5.5 (second event) and Obermeier (1999)

Illinois distance from inferred BP energy center Shoal Creek Centered in vicinity of Maximum dike width 1.55 ~ 5670 +/- 80 yr. BP M 6.5 (lower limit) McNulty and Obermeier lower Shoal Creek near its m; 35 km maximum (4520 +/- 160 BC or 6500 yr. (1999); Tuttle et al. (1999) confluence with Kaskaskia distance from inferred BP)

River: energy center

~ 40 miles (65 km) east-southeast of St Louis, Missouri.

Alternative location:

Centralia fault (Du Quoin monocline)

Cache River Cache River from Dike width 1 to 9 cm Two ages: maximum age AD Unknown Chester and Tuttle (2000)

Sandusky to the 1020 to 1250 for younger Mississippi River and possibly older event(s)

B1.T-4 DEL-096-REV0

TABLE B-1-1 LIQUEFACTION EVIDENCE FOR PREHISTORIC EARTHQUAKES IN THE SOUTHERN ILLINOIS BASIN Seismic Hazards Report for the EGC ESP Site Name of Size and Distribution Estimated Earthquake Location of Features Timing1 Magnitude Reference Bg Muddy River no new sites identified. All paleoliquefaction Su and McBride (1999)

Geologic conditions not dikes in the region could favorable for liquefaction. possibly be induced by paleoearthquakes that occurred near the potential seismogenic sources identified by reanalysis of seismic reflection data. The maximum possible magnitude for a basement-involved fault in the region is between M 6 to just above M 7.

New Madrid Big, Cache, Kaskaskia, and Past 4,000 years (possibly -- Tuttle et al. (1999)

Marys rivers during the AD 900, AD 1530, or AD 1811-1812 earthquakes)

Southeastern Missouri Big Muddy and ~ 20 miles (30 km) Dikes 0.1 to 1 cm wide; Big Muddy River: post-9070 3 scenarios: Tuttle et al. (1999)

Meramec Rivers southwest of St. Louis, sand diapir that reaches 20 BC and possibly prior to Local, M > 5.2 Missouri cm wide 4240 BC Shoal Creek, M 7.0 Meramec River: -post-13,210 Centralia fault, M 7.5 BC 1 Ages given in yr. BP (years before present) are uncorrected radiocarbon ages DEL-096-REV0 B1.T-5

TABLE B-1-2

SUMMARY

OF DEPOSITS IN BANK EXPOSURES Seismic Hazards Report for the EGC ESP Site River Salt Creek Sangamon Mackinaw Total distance observed 11.7 miles 12.3 miles 17.5 miles Estimated total length of till 150 feet 4,950 feet 8,200 feet observed Estimated total length of 360 feet 135 feet 440 feet liquefiable latest Pleistocene alluvium observed (glaciofluvial outwash)

Estimated total length of pre- 1,200 feet 2,000 feet 5,850 feet hypsithermic (early Holocene) alluvium observed Notes:

1. Limited reconnaissance (approximately 1 mile walked) on the North Fork of Salt Creek revealed ~ 20-foot exposure of latest Pleistocene alluvium.

2. Sections of riverbank not accounted for in this table generally consist of younger Holocene terrace deposits.

Approximately 70 to 80 percent of the riverbanks observed were covered.

3. The estimated total lengths of liquefiable latest Pleistocene alluvium shown in this table do not include the exposures at Locality S 14 (the Mahomet gravel pit).

B1.T-6 DEL-096-REV0

TABLE B-1-3 CRITERIA FOR DIFFERENTIATING ORIGINS OF LIQUEFACTION FEATURES Seismic Hazards Report for the EGC ESP Site SC 25 SC 16 SC 18 SC 19 M6 Seismic Dike widens at depth or remains constant in width Yes Yes No Yes Yes?

(injected from below)

Dike fill includes silty sand, clean sand, or gravelly Yes Yes Yes Yes Yes sand (liquefiable material)

Dike fill fines upward or remains constant in grain Yes No N/A Yes Yes size (injected from below)

Dike is tabular in plan view Yes Yes No Yes? Yes Dike wall contacts are sharp and planar or irregular Yes Yes Yes Yes Yes (injected suddenly)

Dike is observed in clear association with source Yes No No Yes No material Source material is observed to be loose Yes N/a N/A Yes N/A Source material is located at or below the water table Yes Yes? Yes? Yes Yes?

Nonliquefiable cap present Yes Yes Yes Yes No Evidence of recurrent events is found nearby Yes Yes Yes Yes No i ii ii ii Dike fits a regional pattern based on dike width and No No No No Noi abundance Glaciotectonic Evidence of basal drag, ice push, meltwater expulsion, and hydraulic fracturing is found Temporal association with glaciers No No No No Yes Spatial association with glaciers No No No No Yes Associated or nearby evidence for subglacial No No No No Yes deformation of soft-sediments Terminates at base of overlying till No No No No Yes Other Nonseismic Mechanisms Artesian conditions are recorded historically in the No No No No No vicinity Dike is located in a setting prone to artesian No No No No No conditions (e.g., at the base of a hill or near an artificial levee)

Dike is associated with features that may indicate No Yes No No No nonseismic landsliding i

The paucity of latest Pleistocene to early Holocene deposits identified during this field investigation makes it difficult to assess this parameter. The absence of evidence for liquefaction was noted at several locations in the study area (Table B-1-5).

DEL-096-REV0 B1.T-7

TABLE B-1-3 CRITERIA FOR DIFFERENTIATING ORIGINS OF LIQUEFACTION FEATURES Seismic Hazards Report for the EGC ESP Site ii The proximity of the three sites where possible mid-Holocene liquefaction features are recognized (SC 16, SC 18, and SC 19) and the small scale of the features at all sites do not define a clear regional pattern for a mid-Holocene event centered in the EGC ESP Site area. Sufficient exposures of pre-hypsithermic deposits throughout the area are available to demonstrate the absence of larger late Holocene liquefaction features in the site vicinity.

B1.T-8 DEL-096-REV0

TABLE B-1-4 CHARACTERISTICS AND ESTIMATED AGES OF POTENTIAL LIQUEFACTION FEATURES IDENTIFIED IN SITE VICINITY Seismic Hazards Report for the EGC ESP Site Maximum Locality Coordinates No. of Width Material Interpretation Estimated Age dikes in. cm Intruded Salt Creek 365962 E 6 4 10 Silt cap Preferred origin Latest Pleistocene/early SC 25 4460489 N (loess) for dikes is Holocene based on seismic weathering of dike liquefaction. material, soil profile development, and estimation of paleo-water table levels.

Salt Creek 363816E 2 1.5 4 Clayey silt Dikes likely Mid- to late Holocene SC 16 4459593N (fluvial) caused by based on soil profile seismic development.

liquefaction.

Salt Creek 363376E 2 1.5 4 Clayey silt Preferred origin Mid- to late Holocene SC 19/ 4457299N (fluvial) for dikes is based on soil profile SC 18 seismic development.

liquefaction. Sill may indicate seismic liquefaction or soft sediment deformation.

Mackinaw 355764E 2 4 10 Lodgement Origin for dikes Latest Pleistocene (17.7 River 4493685N till is uncertain; +/- 1 ka) capped by M6 may be seismic glacially compacted or glaciotectonic fluvial deposits.

fluidized injection.

Note: Coordinate system is UTM (meters) zone 16, NAD 83.

DEL-096-REV0 B1.T-9

TABLE B-1-5 LOCALITIES OF OLDER ALLUVIUM Seismic Hazards Report for the EGC ESP Site Locality Coordinates Unit Description Estimated Age Comments Sangamon River S2 368766E Silt cap (~ 5 feet Middle Holocene Good exposure (~.50 feet) having no 4476983N thick) overlying apparent liquefaction features or oxidized pebbly fractures.

gravelly sand (fluvial)

S3 368874E Older alluvium Pre-hypsithermic No liquefaction features observed.

4476926N overlying till, (early Holocene) Good conditions for liquefaction, but overlain by ~ 6-foot- limited exposure.

thick silt cap at upstream end of exposure S6 383748E Silt cap (~.2 feet Latest Pleistocene to Contact between silty clay to clayey silt 4454301N thick ) overlying ~.10 early Holocene. Based unit and underlying coarse sand feet of interbedded on weathering and soil exposed for 40 feet. The sand below the gravelly sand, sand, development, this unit is fine-grained unit, which lies ~ 2 to 3 and silt (fluvial) estimated to be feet above present water level, should equivalent in age to SC have been below the water table at the 25 or perhaps older. time of the SC 25 event (comparable Radiocarbon analysis of age and position). There is good charcoal from deposit evidence for the absence of liquefaction (sample 7935_S-6-2) in susceptible deposits at this site.

yielded a radiocarbon age of 35,550 +/- 1,200 yr. BP. This age suggests the charcoal was re-used from an earlier time and does not represent the age of the deposit S8 No reading Fine-grained Middle Holocene No liquefaction observed along 30-taken at this alluvium (~ 9 to 10 foot-long exposure. Larger location (see feet thick) paleoliquefaction features should have Figure B-1-6) been apparent.

B1.T-10 DEL-096-REV0

TABLE B-1-5 LOCALITIES OF OLDER ALLUVIUM Seismic Hazards Report for the EGC ESP Site Locality Coordinates Unit Description Estimated Age Comments S9 383723 E 20-foot-high stream This unit appears to be The exposure at this site is poor, 4453574 N bank: interbedded older than S 6 or SC 25 covered by slump colluvium and loess and water-laid based on geomorphic vegetation.

fine sediment ( ~ 12 position and soil feet thick) overlying development (shown on thin gravelly sand, Soil Survey map as soil silt, and poorly sorted unit 570C2) gravel ( ~ 8 feet thick)

Sangamon River S 10 382937E 14- to 15-foot-high Latest Pleistocene to Poor exposures of terrace deposits 4452756N stream bank: silt cap early Holocene along meander loop. Exposures (~ 300-(3 to 5 feet thick) foot-long ) insufficient to assess overlying gravelly presence/absence of paleoliquefaction.

sand S 11 382525E 9-foot-high stream Pre-hypsithermic ~ 200 feet of exposure. Very good 4452432N bank: alluvium (early Holocene) evidence of no liquefaction in susceptible deposits.

S 12 no reading ~ 20-foot-high stream Latest Pleistocene/early Stream bank mostly covered. No taken at this bank: gravel Holocene based on information observed regarding location observed up at least geomorphic position. presence/absence of paleoliquefaction.

half the bank S 13 379041E 9-foot-high terrace: Pre-hypsithermic ~ 100 feet of good exposure (left bank 4449465N silt cap (4 to > 9 feet (early Holocene) adjacent to a rifle range) of susceptible thick) overlying deposits showing no evidence of strongly oxidized liquefaction.

fluvial deposits (clean, medium-grained sand)

DEL-096-REV0 B1.T-11

TABLE B-1-5 LOCALITIES OF OLDER ALLUVIUM Seismic Hazards Report for the EGC ESP Site Locality Coordinates Unit Description Estimated Age Comments S 14 379879E Upper section: Slackwater deposit in Prairie material gravel pit (Mahomet 4449215N glacial outwash upper part of section is Operation,Yard 58). Excavation started (cross-bedded sand correlated to flooding in 1971. Gravel pit is > 70 feet deep, and pebbly gravel) event (breaching of the covers ~ 150 acres. 100+ acres of overlies 4-foot- Bloomington moraine at continuous exposure observed since (average) to 6-foot- Saybrook) ~ 17 to 18 ka. operation began. Upper section well (locally)-thick silt exposed in older part of gravel pit.

bed (slackwater Present water table is just below floor deposit), which of older gravel pit (probably within a overlies glacial few feet of base of silt layer). Lack of outwash (sand and oxidation observed in deposits below gravel) silt layer suggests water table generally has been relatively high since deposition. No anomalous features (clastic dikes, fractures) in silt layer noted during excavation. Site conditions are very susceptible to liquefaction, providing good evidence for lack of significant paleoliquefaction since deposition of the silt layer.

Mackinaw River M2 361731E Older alluvium Based on geomorphic Deformation features observed in 4491708N (stratified sand) position and soil profile poorly sorted sand unit are likely underlying 7- to 8- development, the fluvial related to glaciotectonic or periglacial foot-high terrace that deposits underlying the (frost-wedge) processes. The onlaps hill, underlain terrace probably are hummocky terrain associated with these by poorly sorted latest Pleistocene in age. deposits suggests the deposits are ice-gravelly sand unit contact drift or ablation till (deposited having subvertical during stagnating ice conditions). The filled fractures and overlying fluvial deposits appear to be deformation features only slightly younger based on the soil profile developed across the contact.

These deposits, which would be susceptible to liquefaction, show good evidence of no disruption (paleoliquefaction) since deposition.

M3 360046E ~ 10-foot-high Pre-hypsithermic 25-foot long exposure showing good 4492330N terrace: silt cap (4 (early Holocene) evidence of no paleoliquefaction in feet thick) overlies susceptible deposits.

fluvial sand and gravel (4.5 feet thick); which overlies? silt (2 feet thick)

B1.T-12 DEL-096-REV0

TABLE B-1-5 LOCALITIES OF OLDER ALLUVIUM Seismic Hazards Report for the EGC ESP Site Locality Coordinates Unit Description Estimated Age Comments Mackinaw River M 101 351920E 10-foot-high terrace: Pre-hypsithermic 20-foot-long exposure showing good 4495543N silt cap (9 to 11 feet (early Holocene) evidence of no paleoliquefaction in thick) overlies gravel susceptible deposits.

M 102 350494E Alluvium Pre-hypsithermic Good 60-foot-long exposure showing 4496546N (early Holocene) no evidence of paleoliquefaction.

Deposits probably not as old as those at SC 25. Additional 200 feet of good exposure downstream showing no evidence of paleoliquefaction.

Additional No evidence of significant exposures of paleoliquefaction features or anomalies pre- noted during visual examination of hypsithermi these exposures from canoe.

c deposits noted on field maps Salt Creek SC 21 361929.6 12-foot-high terrace: Latest Pleistocene Approximately 30 feet of poor to 4456133 5-foot-thick silt cap moderate exposure. Sediments probably overlies 7 feet of would have liquefied, although features fluvial sand and may not have been preserved in the gravel (glacial relatively thick silt cap.

outwash)

North Fork Salt Creek NSC 1 350937E 11-foot-high terrace: Latest Pleistocene 20+ feet of exposure. Sediments would 4464273N thin silt cap (2 feet have liquefied if water table were 2+

~ 500 feet thick) overlies fluvial feet higher than at present. It might be upstream of deposits (glacial difficult to see small clastic dikes in confluence outwash) (~ 8.5 feet these coarse sands. No larger dikes or with West thick), which overlies anomalous features observed. These Fork of North 1 to 2 feet of till at deposits are likely old enough and in a Salt Creek base of exposure favorable position in relation to the water table to have recorded the SC 25 event.

Note: Coordinate system is UTM (meters) zone 16, NAD 83.

DEL-096-REV0 B1.T-13

Site s:\7900\7935\7935.000\03_0303_sectb1\_fig_b1-01(02).ai (2003-05-29, 10:09) 0 60 120 miles A star represents a magnitude of 5 or higher. A solid circle represents a magnitude between 4.5 and 5. A plus sign represents a magnitude between about 2.3 and 4.5. Historical earthquake data are from USGS/NEIC Global Hypocenter Data Base CD-ROM (Version 3.0).

Concentric circles show estimated energy centers of large prehistoric earthquakes. The estimated moment magnitude, M, for a prehistoric earthquake is located near the circle.

Note:

Epicenters of historical earthquakes are shown for the time period 1804-1992 From McNulty and Obermeier (1999)

Seismic Hazards Report for the EGC ESP Site Figure Epicenters of Historical Earthquakes and Estimated Energy B-1-1 Centers of Large Prehistoric Earthquakes in Site Region

s:\7900\7935\7935.000\03_0303_sectb1\_fig_b1-02(03).ai (2003-05-29, 10:10)

Site 0 30 60 miles From McNulty and Obermeier (1999), and Tuttle (electronic communication to Kathryn Hanson, February 11, 2003)

Seismic Hazards Report for the EGC ESP Site Figure Sites of Paleoliquefaction in Southern Indiana and Illinois B-1-2

s:\7900\7935\7935.000\03_0303_sectb1\_fig_b1-03.ai (2003-05-29, 10:11)

Site 0 15 30 miles From McNulty and Obermeier (1999).

Seismic Hazards Report for the EGC ESP Site Figure Map of the Vicinity of the Springfield, Illinois, Earthquake Showing B-1-3 Approximate Limit of Liquefaction for the Earthquake

s:\7900\7935\7935.000\03_0303_sectb1\_fig_b1-04(05).ai (2003-05-29, 10:29)

From McNulty and Obermeier (1999).

Seismic Hazards Report for the EGC ESP Site Figure Moment Magnitude Versus Maximum Distance to B-1-4 Surface Evidence of Liquefaction

EXPLANATION Site State boundaries Bedrock Structures SITE Mapped faults Location Map ILLINOIS Mapped folds INDIANA Vincennes Earthquake Dome Structures Mahomet Clinton (5,900 to 6,300 years BP) Data sources: Nelson (1995), University of Missouri (1991),

Indiana Geological Society (2001),

Illinois Geological Survey (1995)

Geophysical Lineaments Geophysical lineaments Data sources: Harrison and Schultz (2002) &

25 Miles Hildenbrand and others (2002)

Springfield Earthquake Waverly Eathquake Paleo-earthquake Energy Centers (5,900 to 7,400 years BP) (3,500 to 8,500 years BP)

Paleoliquefaction sites and dike widths (m):

>0.5 EN T

0.15 - 0.5 AM NE

< S:\7900\7935\arcmap\_fig_05 structure.mxd T I EN L <0.15 AM AL IC INE YS Square indicates young (1811-12?)

L S PH and ancient dikes RLE EO A G  ? Approximate limit of liquefaction for E

. CH RC paleo-earthquakes. Queried where ST CO poorly constrained.

MM Shoal Creek Earthquake E Data source: McNulty and Obermeier (1999), and (5,700 years BP) Tuttle and others (1999) and Tuttle electronic communication to Kathryn MISSOURI Hanson, February 11, 2003).

s:\7900\7935\7935.000\03_0303_sectb1\_fig_b1-05.ai (2003-05-29, 10:31) 0 25 50 Kilometers 0 20 40 Miles Scale 1:1,267,200 Figure B-1-5 Seismic Hazards Report for the EGC ESP Site Map Showing Bedrock Structures, Skelton Eathquake Geophysical Lineaments, and (11,000 to 13,000 years BP)

Paleoearthquake Energy Centers

Woodford Livingston EXPLANATION Peoria Ma ckin M102 aw Site location M101 R iver M6 Fulton M3 County Boundaries M2 Ford Paleoliquefaction sites and dike Tazewell widths (m):

Location Map San Gibson City McLean ga mo H 0.15 - 0.5 S3 nR ive r

S2 <0.15 Downs k H Cree reek alt Identified this study ar C Sug kS k Fo r ree r th NSC1 lt C Significant exposure of latest No  ? Sa SC25 Mason Pleistocene alluvium with no k paleoliquefaction noted ree Two liquefaction oC SC21 SC16 (see Table B-1-5) apo Farmer City H S6 Kik De Witt SC19/18 S8 events widely S10 S9 Exposure of older alluvium S12 S9 spaced in time Mahomet S11 (Table B-1-5)

Sangamon River k  !

H ree S13 rC S14 (Mahomet e Clinton De H Gravel Pit)  ? Approximate limit of liquefaction for ve r paleo-earthquakes Ri on am Inferred energy center Logan Salt Creek ng Sa Data source: McNulty and Obermeier (1999)

Champaign S:\7900\7935\arcmap\gmx_investigation.mxd Menard Piatt Surveyed Stream Banks Springfield Earthquake Lake Fork (5,900 to 7,400 years BP) Reaches investigated in McNulty and Obermeier (1999)

Reaches investigated by canoe or Possibly two Macon foot for this study.

liquefaction events Sangamon River Reaches investigated by road widely spaced in time Vincennes Earthquake reconnaisance for this study.

Springfield (5,900 to 6,300 years BP)

H s:\7900\7935\7935.000\03_0303_sectb1\_fig_b1-06.ai (2003-05-29, 10:38)

Douglas Sangamon 0 5 10 20 Km Surveyed Pit 0 5 10 20 Miles Moultrie Flat Branch Christian ork Figure B-1-6 S. F Lake Shelbyville Sa Coles Seismic Hazards Report for the EGC ESP Site ng Shelby am Map Showing Surveyed Stream Banks and on Macoupin Montgomery er Evidence for Paleoliquefaction Riv in the Site Vicinity