ML20199L513
| ML20199L513 | |
| Person / Time | |
|---|---|
| Issue date: | 06/30/1986 |
| From: | Minogue R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| RIL-146, NUDOCS 8607090460 | |
| Download: ML20199L513 (28) | |
Text
{{#Wiki_filter:. g# 8 h*^ n UNITED STATES NUCLEAR REGULATORY COMMISSION h j u WASHINGTON. D. C. 20555 e
\ *****
/
JUN 3 01986 l MEMORANDUM FOR: Harold Denton, Director Office of Nuclear Reactor Regulation FROM: Robert B. Minogue, Director Office of Nuclear Regulatory Research
SUBJECT:
RESEARCH INFORMATION LETTER NUMBER 146, RESULTS OF NEW MADRID SEISM 0 TECTONIC RESEARCH PROGRAM ABSTRACT Ti.e New Madrid earthquakes of the Central Mississippi River Valley (1811-1812) were'among the strongest known seismic events in North America. Their intens-- ity has been estimated at XI-XII on the Modified Mercalli scale and the felt area extended eastward as far as Charleston, SC, Washington, DC, New York City, and Hartford, CT. The earthquakes are unusual in that they occurred within a continental plate rather than at a plate margin as is the case for most large earthquakes. Using only surficial geology and historic seismicity, it was not possible to relate the New Madrid earthquakes to a specific geologic structure or capable fault. To resolve uncertainty in the causative mechanism of these earthquakes, a multidisciplinary program of investigations was begun by the NRC in 1976 called the New Madrid Seismotectonic Research Program. This program is now at a stage where a conceptual model can be constructed that explains the recurring intraplate seismicity of the Central Mississippi River Valley. The earthquakes are spatially associated with fault movement within an ancient rift complex called the New Madrid Rift complex. This rift complex can be delineated based on geological, geophysical, and seismic evidence. This is significant to NRC regulatory activities because it means that seismicity in the area can be linked to a tectonic structure of finite extent. The rift complex formed in late Precambrian to Cambrian time (1,500 to 500 million years ago). The contemporary stress regime is quite different (compressional as opposed to tensile) from the one that produced the original zones of weakness. Several generalizations caa be made that would be significant in drawing a seis-mic hazards map for the New Madrid area: (1) the Reelfoot Rift is seismically the most hazardous portion of the rift complex; (2) the recurrence interval for events large enough to cause liquefication in the area is about 600 years; and (3) the remaining portions of the rift complex (i.e., the St. Louis and Indiana arms and Rough Creek Graben) should be considered seismically active but to a lesser extent than the Reelfoot Rift. 8607090460 RES 060630 PDR PDR 8607090460
Harold Denton, Director 2 rg INTRODUCTION
Background
The series of earthquakes which occurred in the Central Mississippi River Valley near New Madrid, Missouri, in the winter of 1811-1812 were among the strongest known earthquakes in North America. Three major earthquakes, plus 203 damaging aftershocks, occurred in the 3-month interval from December 16, 1811 through March 15,1812 (Nuttli,1982). They are unusual in that they happened in the interior of a continental plate (North American Plate) rather than at a plate margin. Thus we are dealing with "intraplate tectonics." The intensity of the earthquakes was extremely high. Nuttli (1982) estimates an epicentral Modified Mercalli Jntensity (MMI) of XII and a body wave magnitude of 7.2, 7.3, and 7.5 for the three major earthquakes. The zone of greatest deformation (i.e., fis-suring, sandblows, landslides, subsidence) extended over an area of 48,000 square kil weters (approximately the area of West Virginia), and the earthquakes were felt as far away as Hartford, CT, Washington, DC, Charlsston,.SC and New Orleans, LA. Since 1812 there have been at least 20 damaging earthquakes of MMI intensity VI-VII to XI (Nuttli,1982, p.15) in the Central Mississippi River Valley (see Figure 2 for their location) making this one of the most seismically active areas in the United States east of the Rocky Mountains. Although the frequency and severity of earthquakes in this area are high, until very recently, the earthquakes could not be directly linked to geologic structures as can be done
- in some areas such as California. This is due to: (1) deep burial of the
$ causative structures, (2) the fact that most of them do not extend to the earth's surface, (3) vegetative cover which obscures faults and rock exposures, and (4) the fact that until about 1974 there was very little geophysical data available for the Central Mississippi River Valley that could provide a basis for interpreting subsurface structure. The lack of definitive evidence linking seismicity with geologic structures was a matter of serious concern for, in the absence of this evidence, Appendix A to Part 100 requires nuclear power plant license applicants to design the facility to withstand the largest earthquake which could oCCJr in the tectonic province where the power plant is located. In licensing actions going back to the early 1970's the AEC/NRC licensing staff dealt with the New Madrid earthquakes as an anomalous zone of high seismicity within the Mississippi Embayment portion of the Gulf Coastal Plain and the adja-cent Wabash River Valley. In the early 1970's the staff had to rely princi-pally on the pattern of seismicity which centered on the Central Mississippi River Valley,.the frequent recurring earthquakes in the area, and surficial geology. With this limited evidence the licensing staff treated the New Madrid earthquakes as being linked to buried unidentified structure. ~The zone of New Madrid seismicity was presumed to extend from Memphis, Tennessee, to Vincennes, Indiana. To resolve the causative mechanism, that is, to see if the New Madrid earthquakes could be linked with geologic structure the Office of Nuclear Regulatory Research initiated in 1976 a program of confirmatory research. The purpose of the research was to provide the licensing staff with information to support regulatory decisions. The research program was a cooperative multi-disciplinary effort called the New Madrid Seismotectonic Research Program and
Harold Denton, Director 3
'AIN 30 686 funding was shared by 7 State Geological Surveys, and several academic institu-tions in cooperation with the U.S. Geological Survey, National Science Founda-tion, and Tennessee Valley Authority.
Description of the Program The New Madrid Seismotectonic Research Program was a coordinated program of geo-logical, geophysical, and seismological investigations of the area within a 200-mile radius of New Madrid, Missouri, which concluded in FY 1986. The loca-tion of the New Madrid Area is shown in Figure 1 and the contributions of the participants is summarized in NUREG/CR-4632 (Buschbach, 1986). The program was designed to define the structural setting and tectonic history of the area in order to evaluate earthquake hazards, particularly to nuclear facilities. Specific program objectives are: (1) to identify the earthquake source and source mechanisms for the numerous recurring earthquakes in the Central Mississippi River Valley; (2) to delineate the zone of high seismic activity; and (3) to determine the recurrence interval of potentrally damaging earth-quakes in the area. Participants in the program have included scientists from the State Geological Surveys of Illinois, Indiana, Kentucky, Tennessee, Alabamac Arkansas, and Missouri; faculty members of Earth Science Departments at Saint Louis Univer-sity, Vanderbilt University, Purdue University, Memphis State University, Eastern Kentucky University, the University of Pittsburgh, the University of Kentucky, the University of Wisconsin at Milwaukee, and the University of Texas at El Paso; and staff members of the U.S. Geological Survey, U.S. Army Corps of Engineers, and the Tennessee Valley Authority. RESULTS OF THE PROGRAM Integration of Results into a Conceptual Model The program has produced significant results which led to the formulation of a conceptual model that explains the New Madrid seismicity based on many lines of evidence. For an historical review of key developments in the formulation of the conceptual model see USGS Open File Report 84-770, pp. 6-10. Among them are: seismic history, contemporary seismicity, geophysical exploration tech-niques (particularly gravity and aeromagnetic data and seismic reflection pro-files), subsurface geology, surface geology, stress measurement data, and hydro-geochemical data. The conceptual model relates the New Madrid earthquakes to the reactivation of basement structures within a rift complex that formed in late Precambrian to Cambrian time. (A geologic time scale is provided in Figure 13.) The model relates the seismicity to a contemporary compressional stress regime greatly different from the tensional stress field which originally produced the rift complex. Thus the research has reduced uncertainty for the licensing process by relating seismicity to geologic structures which have a finite extent. The conceptual model is shown schem'atically in Figures 3, 4, and 5. Braile, Keller, Hinze, and Lidiak (1982) named overall rift complex the New Madrid Rift
Harold Denton, Director 4 JUN 301986 Complex and the term is accepted by many researchers in the area although other terms such as Reelfoot Rift, New Madrid Seismic source zone, and Mississippi Valley Graben are used by others. As can be seen in Figure 3b, according to Braile et al. (1982), there are four structural components of the New Madrid Rift Complex: (1) the Reelfoot Rif t, (2) the Rough Creek Graben, (3) the Indiana Arm, and (4) the St. Louis Arm. Their presence has been inferred from seismic reflection profiling, aeromagnetics, gravity, and subsurface geology as revealed by deep boreholes. As with many conceptual models, one must realize that it is an early approximation which fits the available data. As more data become available, details of the conceptual model may change. Summarized below are the lines of evidence which led to the formulation of this model which specifically proposes a reactivated rift as the causative structure for the New Madrid earthquakes of 1811-1812. Specific Research Results Seismicity Data: Until the early 1970's, the seismicity data base for the New Madrid area was limited to: (1) that gathered by instruments located far from the epicenters of the 1811-1812 events, and (2) human observations which were recorded in news-paper accounts and journals. Hadley and Devine (1974) collated the available data from 1800-1972 and produced a seismic frequency map (Figure 6). The seis-mic frequency contours from the Hadley and Devine map show a Y-shaped pattern which matches the New Madrid Rift Complex identified by this research. It follows the Mississippi River to southern Illinois where one branch follows the Mississippi northwestward toward St. Louis. The other branch follows the Ohio and Wabash Rivers northeastward. In part, the Y-shaped pattern may reflect location of observation points since population density was highest along the rivers, or it may reflect amplification of ground motion in the alluvial filled valleys of the Mississippi, Ohio, and Wabash Rivers. However, even with a possible bias due to population distribution, seismicity was indeed censider-ably higher than in the remainder of the Central Stable Region. Beginning in the early 1970's the NRC, USGS, NSF and TVA in a coordinated pro-gram funded installation of seismographic networks in the New Madrid area. Their purpose was to determine levels of seismicity, recurrence intervals, location of epicenters, and ground motion characteristics of the earthquakes. The results are summarized in Figures 7 and 8, which show epicentral location of detected earthquakes. Their gross pattern is similar to the Hadley and Devine map (Figure 6). In the Reelfoot Rift (see Figure 8b for location) there are two principal northeast-southwest trending zones extending from Cairo, Illinois into northeastern Arkansas. These two zones are offset from one another in the region between New Madrid, Missouri, and Dyersburg, Tennessee, by a zone of northwest-southeast trending seismicity. The seismic zones are thought to represent subsurface faults. Composite focal mechanism studies on the two northeast-southwest zones indicate probable right lateral strike slip movement and the focal plane studies on the northwest southeast zone suggest reverse faulting (Herrmann and Canas,1978). There may be another northeasterly trending zone of high seismicity parallel to the postulated western margin of the Reelfoot Rift but outside of the rift zone. However, the pattern is not
ata6bM==m a-a auM- m an - mA6 M4 -o Amdeuk.,m, b.a .,a 4 -Am.m -.+a,M e6 Harold Denton, Director 5 JUN 30133s distinct. Nuttli (1982, p. 15) notes that the three major earthquakes of 1811-1812 occurred in the central portion of the Reelfoot Rift. In the St. Louis Arm (see figure 8b for location) the seismic activity trends northwest from Cairo, Illinois towards St. Louis. Seismicity appears not to extend beyond St. Louis. Seismic activity is low in the Indiana Arm and Rough Creek Graben. Earthquake focal depths in the Reelfoot Rift are generally 3 to 15 km (Buschbach, 1986, p. 58). This puts them either within a zone of disturbed seismic reflec-tions (Figure 10b shows this disturbed zone) or within the magnetic basement rocks. In the Indiana Arm, several earthquakes had focal depths greater than 16 km. This makes them the deepest earthquakes recorded thus far in the Central Mississippi River Valley. Buschbach (1986) notes that these earthquakes appear to be associated with the eastern edge of the Fairfield Basin, a depression in the Illinois Basin, rather than with the Wabash Valley Fault System which is located 30 km to the east (see Figure 9 for location of the Fairfield Basin and Wabash Valley Fault System). Aeromagnetic and Gravity Data: - The pattern exhibited by aeromagnetic and gravity surveys of the Central Mississippi River Valley correspond strikingly well with seismic data (see previous section). In particular the data have delineated a northeast striking basement depression about 70 km wide and more than 300 km long (Kane et al., 1981; Hildenbrand et al., 1982). The depression, the Reelfoot Rift portion of the New Madrid Rif t Complex, is remarkably linear over its entire length, has i nearly parallel sides, and is at least 2 km deeper than the surrounding base-ment (Kane et al., 1981; Hildenbrand et al., 1982). Both gravity and magnetic anomalies are aligned along the boundaries of the depression. The geophysical signatures along the rift boundaries are characteristic of mafic plutons. Esti-mated magnetization directions for the postulated plutons indicate a Mesozoic age (Kane et al . ,1981). Similar to the Reelfoot portion of the New Madrid Rift Complex, the St. Louis Arm, Indiana Arm and Rough Creek Graben can be delineated on their aeromagnetic and gravity signatures (Braile et al . ,1982). The St. Louis Arm is the broadest of the four arms. The southwest side has the strongest aeromagnetic gradient of any part of the New Madrid Complex. .However, its northeast side is not as distinct. Its presence is deduced primarily from interpretations of the patterns exhibited on aeromagnetic and gravity maps. Unlike the other arms of the New Madrid Rift Complex, there is no significant thickening of Paleozoic sediments in the St. Louis Arm (Schwalb, 1982; Buschbach, 1986, p. 26). From stratigraphic evidence alone one would not deduce its presence. Gravity and aeromagnetic data (Braile et al . ,1982; Braile et al . ,1984; Gori and Hays,1984) indicate that the Indiana Arm is restricted to the southwestern part of Indiana south of N39.5 latitude (see Figure 7). This is significant because Wollard (1958) using only seismicity called attention to an apparent alignment of the earthquakes of what is now called the Reelfoot Rift with the earthquakes of the St. Lawrence River valley. The Anna, Ohio earthquakes fall within this apparent alignment. Seismicity is low in the Indiana Arm and the most seismically active portions of it appear to be limited to the region south of the Kentucky Fluorspar District.
Harold Denton, Director 6 DUN .S i Ud5 As to the cause of the Anna, Ohio earthquakes, Braile et al., (1985) have investi-gated the possible connection of the Indiana Arm and Anna, Ohio. Based on an integration of aeromagnetic, gravity, and subsurface geology, they have not found sufficient evidence to warrant extending the Indiana Arm northeastward to Anna, Ohio. The Anna, Ohio earthquakes appear to be localized in a structurally complex area near the intersection of the northeast trending Grenville Front and a postulated rift complex, the Fort Wayne Rift, which strikes northwest-southeast. The Fort Wayne Rift has an orientation and geophysical signature similar to a Keweenawan Age rift complex found in the Michigan Basin. From the geological / geophysical data available Braile et al. (1985) interpret the Anna, Ohio seismicity as being caused by one or more of the following: (1) lithologic discontinuitie's associated with the contact between rift type volcanics of the Fort Wayne Rift and low density granitic rocks of the Grenville terrane; (2) reactivation along graben type faults on the northeast flank of a mafic volcanic body within or on the margins of a -60 milligal gravity anomaly which is thought to be a low density granitic body; and (3) basement inhomogeneities within Grenville basement rocks. Seismic Reflection Data: _I_ndiana Arm: Seismic reflection data is available for the Indiana Arm and the Reelfoot Rift. Work on the Indiana Arm was done by Purdue University under NRC funding (Hinze et al., 1983; Sexton et al., 1985). The Purdue seismic lines detected faulted layered Paleozoic sediments. More importantly, that data indi-cated a thickening of pre Mt. Simon layered rocks in the Wabash Valley Area. (Note: the Mt. Simon Sandstone is late Cambrian in age or approximately 500 million years old.) Approximately 2 kilometers of sediment are found in this area but none in the area to the west. This is significant, for the seis-mic data indicate the presence of a down-dropped fault block in the Wabash Valley Area, which received a thick accumul~ation of sediment prior to the late Cambrian. Reelfoot Rift: The seismic reflection data for this area are proprietary infor-mation. The USGS purchased over 250 km of multichannel, common-depth point seismic-reflection profiles (Crone et al . ,1985). A more complete presentation of the seismic profiles is found in Howe and Thompson (1984). There are several deep wells (i.e., Dow Chemical No. 1 Wilson and Houston Oil and Minerals No. 1 Singer) which allow tentative correlation of the principal seismic reflectors. The seismic reflection data confirm the existence of a fault bounded down dropped basin, i.e., the Reelfoot Rift. The Reelfoot contains a sedimentary section greater than 30,000 feet thick. This is considerably thicker than the area outside the Reelfoot Rift where sediment thicknesses are around 5,000 feet or less. The thickest part of this sedimentary sequence is found below late Cambrian sediments. Based on sediment thickness patterns, the Reelfoot rift must have formed prior to the late Cambrian. The seismic profiles in the area show a variety of structural and stratigraphic features which allow a reconstruction of the history of the Reelfoot (see the section on Subsurface Geology which follows on page 9). Noteworthy structural features are the faults bounding the down dropped basin, numerous other high
Harold Denton, Director 7 Jt/N H sg angle faults, and a prominent zone of disturbed seismic reflectors found along the axis in the southern portion of the Reelfoot Rift for a distance of 120 miles from Caruthersville, Missouri, to Marked Tree, Arkansas (see Figures 10a and 10b). This disturbed zone does not show on aeromagnetic or gravity maps. Its presence was ravealed by the seismic reflection data. The origin of the dis-turbed zone is not known. Howe and Thompson (1984) postulate that it originated from late Paleozoic compressional forces which uplifted and reactivated movement on late Precambrian-late Cambrian growth faults. Crone et al. (1985) argue for the disturbed zone being caused by felsic intrusive of post-Paleozoic, pre-Cretaceous age. Regardless of the disturbed zone's origin, the correlation between earthquake hypocenters and it are so pronounced that Crone et al. (1985) conclude that the mapped extent of the disturbed zone defines the source zone for the Reelfoot's axial seismic zone. Note in Figure 10a the close correspondence between earth-quake epicenters and the disturbed zone. The dashed line signifying the outer edge of.the disturbed zone encloses most of the epicenters. In addition, Crone et al.-(1985 p. 549) points out that hypocenters in the axial seismic zone occur within the disturbed zone. As to its extent, Crone et al. can trace the dis-turbed zone only 25 km southwest of Marked Tree, Arkansas, although geophysical data suggest that the Reelfoot Rift continues a considerable distance farther to the southwest. Thus, the axial seismic source zone does not appear to extend to the southwestern end of the rift. At the northeastern end of the axial zone near Caruthersville, Missouri, the reflectors do not appear to be upwarped; however, Crone et al.,1985 report that i in this area, narrow zones of disrupted reflectors, which they interpret as fault zones, coincide with the seismicity. The northeast end of the source zone probably merges with the Dyersburg, Tennessee-New Madrid seismicity trend. The latter trend is a northwest trending band of seismicity (see Figure 8b for loca-tion). Zoback and Hamilton (1980) postulate that the seismicity of this trend is spatially associated with igneous bodies of various ages, presumably emplaced along upper crustal zones of weakness that formed during the initial rifting. The zones of weakness probably formed in a regional extensional stress field rather than the current east-west compressive stress field indicated by fault plane solutions. Thus, from the seismic profiles coupled with seismic network data, one can define the extent of a major intraplate seismic source zone in the Reelfoot Rift on the basis of geologic structure. The profiles also emphasize the role of pre-existing rift related structures on the location of intraplate earthquakes. Hydrogeologic Data: Thermal data: Stearns and Reesman (Buschbach, 1984, pp. 145-162 and Buschbach, 1985, pp. 105-128) made ground water temperature measurements in 80 water wells in the Central Mississippi River Valley with the objective of locating zones where there is rapid ground water flux from depth. Such a zone is postulated to signify an unhealed fault. Vanderbilt located two significant anomalcus areas. The principal one was in the Missouri bootheel and adjoining s e
Harold Denton, Director 8 JUN 301995 County, Tennessee (see Figure 11). This is in the area where many microearth-quakes occur, and also where the Reelfoot Rift and Pascola Arch intersect (see Figure 9 for location of the Pascola Arch and Figure 8b for the location of the microearthquakes). A second area occurs along the Tennessee-Kentucky border, possibly along the east edge of the Reelfoot Rift. There was a remarkable correspondence between the thermal anomalies, some geologic structures in the area, and a zone of numerous microearthquakes. The thermal data can serve as a reconnaissance tool to locate areas worthy of further detailed investigation. Hydrogeochemical data: Stearns and Reesman also studied the ground water chem-istry of the 80 wells in which temjerature measurements were made. The objec-tive was to de'termine if chemical patterns could be linked with seismic activity and geophysical anomalies. Chemical patterns are difficult to draw from the sampled water wells because the data must be segregated by aquifer. However, some interesting relationships were identified. (Details are found in Buschbach, 1985, pp. 106-128.) The chloride level is high in alluvium near the Mississippi River in Missouri in the area of high earthquake activity.along the Pascola Arch. Apparentl'y there is vertical leakage'along faults, from the McNairy Sandstone-aquifer, from a depth of approximately 2000 feet to the land surface. Barium concentrations relate partially to the buried structures which were interpreted as plutons from aeromagnetic and gravity iata. Highest concentrations of barium occur on and near the Missouri and Arkansas plutons on the west side of the rift. Lithium concentrations also are higher on the west side of the rift, notably in water above the Bloomfield pluton. Some of that lithium may origi-nate from ground water flow along vertical fractures extending from the Bloom-field pluton to the ground surface. Strontium concentrations are high within the rift. Both lithium and strontium may well relate to the sediments in which the aquifer occurs. In addition to the lithium which may have come from verti-cal flow of ground water along fractures, igneous minerals containing lithium may have been washed or blown into the depositional area (Eocene volcanic activ-ity). Strontium could be leached from aragonite-rich shells in marine sediment that might concentrate more in the rift if water'was deeper there. As was the case with the ground water temperature data, the hydrogeochemical data also proved to be a useful reconnaissance tool. However, the chemical patterns can be complex and they require careful interpretation. Surface Geology: Investigation of major fault systems: Under sponsorship of NRC's New Madrid Seismotectonic Research Program, the major fault systems (see Figure 9) within a 200 mile radius of New Madrid were examined in dct:il to see if there was evidence of geologically recent offset, that is, movement during the Quaternary Period (approximately the last 2 million years). Results of that work is found in Kolata et al., 1981, Nelson and Lumm 1984 and 1985, and Ault, 1985. In the limited number of places where Quaternary materials cover these faults, no evidence of Quaternary reactivation was found. Trench investigations: The U.S. Geological Survey excavated a trench across the Reelfoot scarp, within the Reelfoot Rift, about 20 miles southeast of New Madrid, Missouri (Russ 1979). The scarp is a prominent linear feature 3 - 9
-w == -r,-
Harold Denton, Director 9 JUN 3 01986 meters high in an area which has little relief. It strikes in a northerly direction and forms the eastern boundary of the Tiptonville Dome. Deep seismic profiling in the area has shown the presence of a buried fault beneath the Reelfoot scarp which offsets Paleozoic, Mesozoic, and Tertiary strata. The Reelfoot scarp is thought to be an upward continuation of faults identified in the subsurface. The trench revealed a number of small faults and zones where there was ductile deformation of sediment layers similar to what one sees with seismically induced sand boils. Radiocarbon dating of mollusk shells in the trench indicates a maximum age for the sediment exposed in the trench of 2000 years. Field relations of the faults and sand boils indicate three epi-sodes of movement within the last 2000 years. That movement is presumably related to New' Madrid type events and it suggests a recurrence interval of about 600 years for earthquakes large enough to produce ground motion great enough to liquefy sand in the alluvium of the New Madrid area. Farther east in the Kentucky River Fault system, about 300 miles from New Madrid but close to the location of the Sharpsburg, Kentucky 1980 earthquake (body wave magnitude 5.1), nine trenches were excavated at four sites under a grant by NRC to the Kentucky Geological Survey (Van Arsdale and Seargeant, 1986) across mapped faults that were partially covered with Pliocene-Pleistocene age sediment. In four of the nine trenches Van Arsdale and Seargeant (1986) identified faulted or folded sediments. In their opinion the faulting an folding of the Pliocene-Pleistocene age sediment is tectonic in origin and that the Kentucky River Fault system has been active within the last 5 million years. Subsurface geology: There have been 434 deep holes drilled within a 200-mile radius of New Madrid, Missouri. Much of the geologic history of that area can be deduced from lithologic and stratigraphic study of the materials encountered in these holes. Details of the geologic history can be found in Schwalb (1982) and Buschbach (1986, pp. 37-54). See also Figure 5 of this Research Information Letter for a short summary of the geologic history. Thirty-eight of them pene-trate the Precambrian crystalline basement. The basement rocks are a mixture of granites, rhyolites and metamorphosed sediments. Age of the crystalline basement varies depending on location and the range is from around 1.0 to 1.5 billion years before present. The time of first rifting within the New Madrid Rift Complex is not known. However, sediments older than the late Cambrian are significantly thicker in the Rough Creek Graben and Reelfoot Rift than on the margins of these structures. That information plus the data provided through gravity, aeromagnetic, and seismic profiling point to the existence of linear troughs bounded by high angle faults dating from before late Cambrian time. Thus the troughs formed over 500 million years ago. For the remainder of the Paleozoic Era, . sediments thickened in the Reelfoot and Rough Creek areas indi-cating topographic depressions but thicknesses were only slightly greater (a few hundreds of feet vs. thousands of feet prior to the late Cambrian) than the surrounding areas. Apparently af ter an initial episode of rifting a topo-graphic depression was created. Although tectonic activity ceased, this area continued to subside at a slightly greater rate than the surrounding craton. The longevity of the Reelfoot Rift as a topographic depression is indicated by the fact that according to Potter (1978), the lower Mississippi River, i.e. , south of the confluence with the Ohio River, has existed in approximately its present position for at least 250 million years and that its course has been structurally controlled, probably by the Reelfoot Rift. This correlation of l 1
Harold Denton, Director 10 JUN 301986 large river systems with rift structures is a common geological occurrence according to Potter (1978). There was a long hiatus between the last Paleozoic sedimentation in the New Madrid area and the first Mesozoic deposits of Cretaceous age. From early Permian to late Cretaceous the sedimentary history of the New Madrid area is unknown because any sediments deposited have been completely removed. This interval is probably represented by uplift and erosion. Drilling in the area indicates that a karst topography developed on the land surface with sink holes later filled with late Cretaceous and younger sediments. In Early to Mid-Mesozoic time Eastern North America underwent rifting as North America and Africa were split apart by plate tectonics (see Figure 4). The tensional forces which produced the rif t basins in Eastern North America (e.g. , Newark, Gettysburg, Richmond, Culpeper and Scottsville Basins) were also felt in the New Madrid area. They resulted in reactivation of the rift complex and igneous bodies were intruded along the margins of the ancient rift complex (see Figure 5). Subsurface information on the igneous bodies is derived primarily from geophysical data, principally gravity, aeromagnetics, and seismic profiling. The seismic profiling points to their emplacement before the late Cretaceous (Crone et al., 1985, p. 549). Surface geology supports this interpretation because to the west of the New Madrid area in Central Arkansas at Magnet Cove, and near Little Rock, there are surface outcrops of early Cretaceous syenites which have a similar gravity and aeromagnetic signature to circular features seen in the gravity and aeromagnetic maps of the Mississippi Embayment. As North America and Africa continued to drift apart for the last 200 million years the regional stress field in Eastern North America has become a compres-sional one. It is this compressional stress field which is believed to have reactivated some of the old buried faults associated with the New Madrid Rift Complex.
SUMMARY
AND CONCLUSIONS The New Madrid Seismotectonic Research Program was a program of confirmatory research to provide information to support licensing decisions as to the cause of the large damaging earthquakes which.have been observed in the Central Mississippi River Valley near New Madrid, Missouri. Prior to the program's inception in 1976, the NRC licensing staff had to rely primarily on surficial geology, a limited number of deep drill holes, and the pattern exhibited by earthquake epicenters. At that time it was not possible to link the earthquakes to known surface faults. . To resolve the origin o.f.the New Madrid area earth-quakes and the extent 'of the seismically active area the New Madrid Program drew on many scientific disciplines. It was necessary to integrate surficial geology with subsurface geology through the use of improved earthquake locations, deep drill holes, geophysical exploration methods such as gravity, aeromagnetics, electrical resistivity, and seismic reflection and refraction profiling. These data were then linked with the seismicity and seismic history of the area. Through this integration of many lines of evidence a conceptual model was devel-oped which provides a framework that relates the New Madrid area earthquakes to geologic structure. In this model the earthquakes are associated with an
Harold Denton, Director 11 DON 3 01986 ancient rift complex. The New Madrid area earthquakes are the result of reac-tivation of structures associated with the formation of the rift complex such as deeply buried faults which extend into the earth's crust, or inhomogeneities associated with a buried zone of disturbed seismic reflectors found along the axis of one part of the rift complex and buried igneous bodies along the margins of the complex. Summarized below are significant points to be noted in licens-ing actions dealing with the cause and extent of the New Madrid seismicity:
- 1. The New Madrid area is located within the interior of the North American tectonic plate. (Areas of high seismic activity within crustal plates are rare.)
- 2. With respect to both size and number of earthquakes, the New Madrid area l is currently the most seismically active region of the United States east of the Rocky Mountains (Herrmann, in Gori and Hays,1984, p. 276; Johnson and Nava, in Gori and Hays, p. 283)
- 3. The New Madrid earthquakes of 1811-1812 are associated with tectonic struc-tures which are part of the New Madrid Rift Complex. The New Madrid Rift Complex is a major tectonic feature within the North American Continent.
It represents a " capable structure" within the meaning of Appendix A to 10 CFR Part 100.
- 4. Integration of geological / geophysical / seismological data has resulted in a conceptual model which relates historic and contemporary seismicity to geologic structures in the area.
n
- 5. The conceptual model consists of a rift complex, called the New Madrid Rift Complex, which has four arms--the Reelfoot Rift, the St. Louis Arm, the Indiana Arm, and the Rough Creek Graben.
- 6. The rift complex was formed in late Precambrian to Cambrian time (1,500 to 500 million years ago) in a tensional stress regime. There is suffi-cient geological / geophysical evidence to show that the Reelfoot, Indiana, and Rough Creek portions of the rift complex are down dropped fault blocks bounded by growth faults.
- 7. In late Paleozoic time (approximately 290 million years ago) the rif t com-plex underwent compression. This resulted in: (a) uplift and subsequent erosion of sediments previously deposited in the rift complex, (b) folding, (c) renewed movement on pre-existing faults, and (d) igneous activity pre-sumably along. pre-existing fractures.
- 8. In late Mesozoic to early Eocene time (approximately 130 to 40 million years ago) there was renewed tectonic activity in the New Madrid Rift Com-plex as evidenced by renewed igneous activity and faulting of sediments up through the Eocene Epoch.
- 9. Contemporary stress measurements show that the New Madrid area is now in a compressional stress regime where the principal compressive stress is oriented approximately eastnortheast-westsouthwest.
l i
Harold Denton, Director 12 ggg g gg
- 10. Of the four arms of the New Madrid Rift Complex, the Reelfoot is by far the most active seismically.
- 11. Based on USGS trenching across the Reelfoot Scarp, the recurrence interval for earthquakes in the Reelfoot Rift large enough to produce ground motion great enough to liquefy sand in the alluvium of the New Madrid area is about 600 years (Russ,1979).
- 12. In the southern part of the Reelfoot Rift there is an axial zone extending southwest from Caruthersville, Missouri to Marked Tree, Arkansas which shows as a " disturbed" zone in seismic reflection profiles. More than 90%
of contemporary earthquakes in the region coincide with this disturbed zone (McKeown, in Gori and Hays, 1984, p. 11). The disturbed zone also underlies the area containing liquefaction features (sand blows) associated with the New Madrid 1811-1812 earthquakes.
- 13. The origin of the disturbed zone in the southern portion of the Reelfo'ot Rift is not known. It appears only in seismic reflection profiles and is absent from aeromagnetic and gravity maps. Crone et al. (1985) postulate that it represents an uplifted, faulted zone associated with felsic igneous bodies. Howe and Thompson (1984) propose that it is an uplifted, faulted zone of former growth faults that were reactivated in a compressional stress regime.
- 14. Immediately to the north of the disturbed zone, contemporary seismicity strikes approximately 330 , in contrast to the disturbed zone, which strikes 050 . Zoback et al. (1980), using seismic reflection data, related seismic-ity in the area to small faults and igneous plutons of various ages.
- 15. Based on geophysical data, the New Madrid research indicates that the Anna, Ohio seismic zone is not an extension of the New Madrid Rift Complex. The Anna seismicity is strongly localized on and around two geophysical anomalies. Whether they are the cause of the seismicity or just control its pattern is not known.
- 16. The New Madrid Rift Complex can be bounded by geological / geophysical data used in conjunction with historic seismicity. This is significant, for it means that a tectonic structure can be delineated and that seismicity in the area can be linked to that tectonic structure.
RECOMMENDATIONS
- 1. The New Madrid Rift Complex should be treated as a major tectonic feature within the North American Continental Plate.
- 2. The New Madrid Rift Complex should be treated as a seismically active area with the potential for earthquakes up to Modified Mercalli Intensity XII.
- 3. Seismic design criteria should assume a 600 year recurrence interval for earthquakes large enough to liquefy sand in the alluvium of the New Madrid area.
---p n - - y
Harold Denton, Director 13 JUN 301986
- 4. The following generalizations should be considered in drawing seismic hazards maps for sites in or near the New Madrid area.
- . a. The zone of highest seismicity is located within the New Madrid Rift Complex and on the flanks of mafic plutons situated on its margins.
- b. The Reelfoot Rift is seismically the most active area. Within the last 2000 years it has experienced at least three intensity XI to XII events.
- c. The other three arms of the New Madrid Rift Complex, i.e., the St. Louis, Indiana, and Rough Creek, should be considered seismically active but to a lesser degree than the Reelfoot Rift.
^d . The seismicity of the New Madrid Rift Complex is distinct from that of the Anna, Ohio area. Based on geophysical evidence, the Indiana Arm of the New Madrid Complex appears not to extend north of latitude N39.5 .
gel 1/$ L d~a Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosure:
List of Selected References
-,-,w--- . - - , - - - - - - -
. . - ,-.---w- -,-3 .-.
SELECTED REFERENCES NUREG and NUREG/CR Reports Ault, C.H., D. Harper, C.R. Smith, and M.A. Wright, 1985, Faulting and Jointing in and near Surface Mines in Southwestern Indiana, NUREG/CR-4117, 27 pp. Braile, L.W.,.W.J. Hinze and J.L. Sexton, 1979, An Integrated Geophysical and Geological Study of the Tectonic Framework of the 38th Parallel Lineament in the Vicinity of Its Intersection with the Extension of the New Madrid Fault Zone, NUREG/CR-1014, 191 pp. -- Braile, L.W. , W.J. Hinze, J.L. Sexton, G.R. Keller, and E.G. Lidiak,1981, An Integrated Geophysical and Geological Study of the Tectonic Framework of the 38th Parallel Lineament in the Vicinity of Its Intersection with the Extension of the New Madrid Fault Zone, NUREG/CR-1878,131 pp. Braile, L.W., W.J. Hinze, J.L. Sexton, G.R. Keller, and E.G. Lidiak, 1982, A Tectonic Study of the Extension of the New Madrid Fault Zone Near Its Intersection with the 38th Parallel Lineament, NUREG/CR-2741, 70 pp. Buschbach, T.C., 1977, New Madrid Seismotectonic Study - Activities During Fiscal Year 1977, NUREG-0379, 61 pp. Buschbach, T.C., 1978, New Madrid Seismotectonic Study - Activities During Fiscal Year 1978, NUREG/CR-0450, 129 pp. Buschbach, T.C., 1980, New Madrid Seismotectonic Study - Activities During Fiscal Year 1979, NUREG/CR-0977, 149 pp. Buschbach, T.C., 1984, New Madrid Seismotectonic Study - Activities during Fiscal Year 1982, NUREG/CR-3768, 166 pp. Buschbach, T.C., 1985, New Madrid Seismotectonic Study - Activities During Fiscal Year 1983, NUREG/CR-4226, 137 pp. Buschbach, T.C., 1986, New Madrid Seismetectonic Program-Final Report, NUREG/CR-4632, 62 pp. Hinze, W.J., L.W. Braile, J.L. Sexton, G.R. Keller, and E.G. Lidiak, 1983, Geophysical-Geological Studies of Possible Extension of the New Madrid Fault Zone, Annual Report for 1982, NUREG/CR-3174, Vol. 1, 87 pp. Hinze, W.J. , L.W. Braile, J.L. Sexton, G.R. Keller, and E.G. Lidiak,1985, Geophysical-Geological Studies of Possible Extension of the New Madrid Fault Zone, Annual Report for 1983, NUREG/CR-3174, Vol. 2, 46 pp. 4 -
Johnson, R.W., C. Haygood, T.G. Hildenbrand, W.J. Hinze, P.M. Kunselman, 1980, Aeromagnetic Map of the East-Central Midcontinent of the United States, NUREG/CR-1662, 12 pp. Keller, G.R., D.R. Russell, W.J. Hinze, J.E. Reed, and P.J. Geraci, 1980, Bouguer Gravity Anomaly Map of the East-Central Midcontinent of the United States, NUREG/CR-1663, 12 pp. Kidd, J.T. 1980, General Geology, Geophysics, and Seismicity of Northwest Alabama, NUREG/CR-1519, 79 pp. Kolata, D.R. , J.D. Treworgy, and J.M. Masters,1981, Structural Framework of the Mississippi Embayment of Southern Illinois, NUREG/CR-1877, 72 pp. Nelson, W.J., and D.K. Lumm, 1984 Structural Geology of Southeastern Illinois and Vicinity NUREG/CR-4036, 127 pp. Nelson, W.G. and D.K. Lumm, 1985, Ste. Genevieve Fault Zone, Missouri and l Illinois, NUREG/CR-4333, 94 pp. Schwalb, H.R., 1982, Paleozoic Geology of the New Madrid Area, NUREG/CR-2909, 61 pp. Stearns, R.G., 1979, Recent Vertical Movement of the Land Surface in the Lake County Uplift and Reelfoot Lake Basin Areas, Tennessee, Missouri, and Kentucky, NUREG/CR-0874, 37 pp. Stearns, R.G., S.K. Towe, V.L. Hagee, S.J. Nova, and S.L. Wilson, 1984, Description and Significance of the Gravity Field in the Reelfoot Lake Region of Northwest Tennessee, NUREG/CR-3769, 39 pp. Stearns, R.G., 1980, Monoclinal Structure and Shallow Faulting of the Reelfoot Scarp as Estimated from Drill Holes with Variable Spacings, NUREG/CR-1501, 37 pp. Stearns, R.G., 1981, Influence of Shallow Structure, and a Clay-Filled Mississippi River Channel on Details of the Gravity Field at the Reelfoot Scarp, Lake County, Tennessee, NUREG/CR-2130, 61 pp. Van Arsdale, R.B., and R.E. Seargeant, 1986, Post-Pliocene Displacement on Faults within the Kentucky River Fault System of East-Central Kentucky, NUREG/ CR-4685, 31 pp. Other selected references Braile, L.W., G.R. Keller, W.J. Hinze, and E.G. Lidiak, 1982, An Ancient Rift Complex and its Relation to Contemporary Seismicity in the New Madrid Seismic Zone, fectonics, Vol. 1, pp 225-237. Crone, A.J., F.A. McKeown, S.T. Harding, R.M. Hamilton, D.P. Russ and M.D. Zoback, 1985, Structure of the New Madrid Seismic Source Zone in Southeastern Missouri and Northeastern Arkansas, Geology, Vol. 13, pp. 547-550. 2
l Ervin, C.P. , and L.D. McGinnis,1975, Reelfaot Rif t: Reactivated Precursor to the Mississippi Embayment, Bulletin of the Geological Society of America, Vol. 86, pp. 1287-1295. Fuller, M.L., 1912, The New Madrid Earthquake, U.S. Geological Survey Bulletin, 494, 118 pp. i Ginzburg, A., W.D. Mooney, A.W. Walter, W.J. Luther, and J.H. Healty, 1983, Deep Structure of Northern Mississippi Embayment, Bulletin of the American Association of Petroleum Geologists, Vol. 67, pp. 2031-2046. Gori, P.L., and W.W. Hays, editors, 1984, Proceedings of the Symposium on the New Madrid Seismic Zone, U.S. Geological Survey Open File Report 84-770, 468 pp. Hadley, J.B., and J.F. Devine, 1974, Seismotectonic Map of the Eastern United States, U.S. Geological Survey Miscellaneous Field Studies Map MF 620. Heigold, P.C., 1976, An Aeromagnatic Survey of Southwestern Illinois, Illinois State Geological Survey Circular 495, 28 pp. Herrman, R.B. and J.A. Canas, 1978, Focal mechanisms studies in the New Madrid seismic zone, Bulletin of the Seismological Society of America, Vol. 68, pp. 1095-1102. Hildenbrand, T.G., M.F. Kane, and J.D. Hendricks, 1982, Magnetic basement in the upper Mississippi embayment--A preliminary report, in F. A. McKeown and L.C. Pakiser, eds. , Investigations of the New Madrid,' Missouri earthquake region: U.S. Geological Survey Professional Paper 1236, chap. E, p. 39-53. Howe, J.R., and T.L. Thompson, 1984, Tectonics, Sedimentation, and Hydrocarbon Potential of the Reelfoot Rift, Oil and Gas Journal, November 12, 1984, pp. 179-189. Kane, M.F. , T.G. Hildenbrand, and J.D. Hendricks,1981, Model for the Tectonic Evolution of the Mississippi Embayment and its Contemporary Seismicity, Geology, Vol. 9, pp. 563-568. Keller, G.R., E.G. Lidiak, W.J. Hinze, and L.W. Braile, 1983, The Role of Rifting in the Tectonic Development of the Midcontinent U.S.A. Tectonophysics, Vol. 94, pp. 391-412. l McGinnis, L.D., P.C. Heigold, C.P. Ervin, and M. Heidari, 1976, The Gravity Field and Tectonics of Illinois, Illinois State Geological Survey Circular 494, 28 pp. McKeown, F.A., and L.C. Pakiser ed., 1982, Investigations of the New Madrid, Missouri, Earthquake Region, U.S. Geological Survey Professional Paper 1236, 201 pp. Mooney, W.D., M.C. Andrews, A. Ginzberg, D.A. Peters, and R.M. Hamilton, 1983, Crustal Structures of the Northern Mississippi Embayment and a Comparison with other Continental Rift Zones, Tectonophysics, Vol. 94, pp. 327-348.
.3 4
1 Nuttli, 0.W., 1982, Damaging Earthquakes of the Central Mississippi River Valley, in the U.S. Geological Survey Professional Paper 1236, pp. 15-30. J. Pennick, 1976, The New Madrid Earthquakes of 1811-1822, University of Missouri Press, 181 pp. Potter, P.E., 1978, Significance and origin of Big Rivers, J. Geol., Vol. 86, pp. 13-33. Russ, D.P., 1979, Late Holocene Faulting and Earthquake Recurrence in the Reelfoot Lake Area, Northwestern, Tennessee, Bulletin of the Geological Society of America, Vol. 90, pp. 1013-1018. Schilt, F.S. and R.E. Reilinger, 1981, Evidence for Contemporary Vertical Fault Displacement from Precise Leveling Data Near the New Madrid Seismic Zone, Western Kentucky, Bulletin of the Seismological Society of America, Vol. 71, pp. 1933-1942. Sexton, J.L. , L.W. Braile, W.J. Hinze, and M.J. Campbell,1985, Seismic reflec-tion profiling studies of a buried Precambrian rift beneath the Wabash Valley Fault Zone, Geophysics. Trace, R.D. and D.H. Amos, 1984, Stratigraphy and Structure of the Western Kentucky Fluorspan District, U.S. Geological Survey Professional Paper 1151-D, 41 pp. Wollard, G.P., 1958, Areas of tectonic activity in the United States as indicated by earthquake epicenters, Transactions of the American Geophysical Union, Vol. 39, pp. 1135-1150. Zartman, R.E., 1977, Geochronology of Some Alkali Rock Provences in Eastern and Central United States, Annual Review of Earth and Planetary Sciences, Vol. 5, pp. 257-286. Zoback, M.D. ,1979, Recurrent Faulting in the Vicinity of Reelfoot Lake, North-western Tennessee, Bulletin of the Geo1 physical Society of American, Vol. 90, pp. 1019-1024. Zoback, M.D., R.M. Hamilton, A.J. Crone, O.P. Russ, F.A. McKeown, and S.R. Brockman,1980, Recurrent Intraplate Tectonism in the New Madrid Seismic Zone, Science, Vol 209, pp. 971-976. 1 I e 4
w- .w .e -. in== IN it .... .
,,,,,_,,,,,,,,, g.g g egity h41.tl ett.K4 km.Et .tfflemes uo - . . . ..,
= =. = ' = = . . , , = . = . = = . = = ' ' ,,,,,,,,,,,,,,,,,,,,
'-"- = ~ -'fd" 200 miles ~ ~'~~
ug,_gg,o P-- stesestl Slag g.I.teSift in.usus amne i.e. asvist AR
't
- . . . us AL
".1'"T,*"'".'..,
/
. . - . , - 4 Figure 1. The New Madrid Area, showing State Geological Surveys. Universities, and 88* _86' gos Federal Agencies cooperating in the New 94 97 ,
Madrid Seismotectonic Research Program. i i i gi ' \'.,*,".'.c./
., j"
(NUREG/CR 4226) 47 _ ,. IOWA
- 1 ILLIN OIS 1
- i INDtANA 40' -
N. .
- 1. J.
MISSOURI l
..-*V' 38' -
' ( \ KENTUCKYs Epicent'ers of damaging earthquakes .
Figure 2. ~ in the Central Mississippi River Valley. .. .._ .. m "" ' -
/
Large circles are the epicenters of the 36' - three major earthquakes of 1811-1812. Epi-
/ TENNESSEE centers of the 203 damaging aftershocks are ARKANSAS . . -
not plotted. Small circles are the ept- - l centers of damaging earthquakes which f occurred subsequent to 1812. o so ioo m o (modified from Nuttli,1982) 34' ~ MISSISSIPPI , ALABAMA
! -t L-
- g. g i I
MICHIGAN RIVER SYSTEM N (PRE.au cIAt> - ILLINOIS INDIANA jt /
/ OUTLINE f*.*w?,\. 45s.,.*/.y ad' W
<*f ) Ob.%
OHIO ' 0F BURIED \, $n ,, . v RIFT COMPLEX *
- j ,*.
f.f* .*.,h
****D q
> MISSOURI ARKANSAS . . ..),;i.c
- ..4.,,) * * *h: ,
..n g.. ...}
KENTUCKY TENNESSEE MOW
/
...:::g W .* Eoce of ja. :.*.f. f'. ..e EMBAYMENT COMORESSIVE SmESS
'; , a; ,,g*
..e b gd >4 '9 ? C y 4 , I. d 9 4
, ANCIENT 3 g y gg'g g * < < < ," 4 s'*'s *'.
FA UL TS .
. "~.^<,
RIFT' 'a , UPPER CRUS T ' " '
' ~
ss'e, "
<w u>u, 4 4 a w'.c,..c ,, ,
v'* , *c) <d
><<<> ,.~>[
f , ,< ,, '
',*s'v v EA R THQ UA KES ' < ' ' ' v " ' * ' >
f'3' .,N I* ,"c'>< LOWER CRUST (Brail e,1984) Figure 3 a. Block diagram illustrating the present configuration of the buried New Madrid Rift Complex. The structurally controlled rivers, Paleozoic rocks in cratonic sedimentary basins, and the Mississippi Embayment, all associated with the buried rift complex, are also shown. Dark areas indicate intrusions near the edge of the buried rift. An uplifted and possibly anomalously dense lower crust is suggested as the cause of the linear positive gravity anomaly associated with the upper Mississippi Embayment. Indiane Arm
/
St. Louis Arm -. Reelf 001 Rif t N d ' s Rough f ' creen c Groben p Ny 4 ,, . # ced Ot**** Figure 3 b. The New Madrid Rift Complex is subdivided into four arms
a i A. D.
'#CO8 80' a * * ***
+ EC o , y*,.d a * ' 51L BLIN f.lS'* . ; a,8 ~/;e[;;.,(
ASIN* % ; [ 4o% . ! * * *, ' * ~'",a'",, dY[ g,b FO f -, g *,9' ,#.*
- 4 ,
E,
' 8,ASIN s , , J/ l. _
. g,
% \ a e' g> 's e e , <
1g l / s/ t s ,. .*,. - i<r',* p# . '!, '.
+30' t30 )p g.g , f gh 30, gl LATE PRECAMBRIAN - INCIPIE T RIFTING, A /
r"pygsggd.$.9
.,a, M
5 E* W. S 80*.1n* a l.. . FORbATION OF TRIPLE JUNCTI S' LATE PALEOZOIC - CONTIN N T A,U, %' a*,, j >. , AND AULACOGENS. hO C B A S I N S. '[, * . ? a "
- 7 - '
B. E.
'.'[. ' > ' . h ,
.e e. '%
'E FT.
COMPLEX.
#'[\ '
t.
^
,' ' ] ','^
,,c.... ' f',[,y n' f
*s R*< ! l>p'5,*'>'-
',a
*
- l[Y *:'Z' ' %
. .. > . e ,,*.c 7 ,,',' ,/ gx,s,g/ h:.x,. ,~
!a.v. *=
,> .. pg e .. ,,.
,y %
W % e d ,)
~s L' Q Ui //x QW
%p;:@3,m.o@ :%
+N -
+
\\o*/
# s N,
30 g :.
-s , , , - TTTP
.J..s..' T ' 7,,
+\v 30' E as LAT PRECAMBRIAN - EARLY CAMBRIAN-
\ .
MESOZOIC-RIFTING AT ATLANTIC M ARGift CONTINENTAL SEPAR ATION. '\ } IGNEOUS INTRUSIONS AND REACTIVATION FAILED INTRACONTINENTAL RIFTS./ OF FAULTS IN AUL ACOGENS. \/ C. F. 6 ,',
' > 5 , .
, ",4f i
. * , . -c O N TIN E N T A L
, , . C O L L I S IO N T O
\
n. e
/ . ',, *
- I, * **
- C e
*,h~c* %
[
' g J
A' * ,' ' ,, ' l .' - .,* ,40 **'**. T H E ,N O R T H
- Ok'a h'd '
' 2 [, ,' . , , ,' gI
- e. s n.'...,1..
. I ' ,' "[2 .
3 [ *'. M. "'/ ~. ' , f -,
. . . . ' < .* s s .- '+a ,- , * . - .
. **** . 'g'm 'a w ~
.';*/ $0so < *
- MlSS!$$1 ppt
*/ .
g Q\ [g ,
'*' [ . ,' ] / fi .<-g-'
EMBAYMENT
. . ., y N ,- - ,. .j f N
'.,f.h'Y,',s.'6(,
, ' ,;o \ fog \ o
.Y 4;. h y i V ', V y p$2Mf MeeV 2 '$A
~
b
\
. E ARLY TO MIDOLE P ALEOZOIC- TACONIC CRETACEOUS TO PRESENT EMBAYMENT OROGENY , SE0lMENTATION IN AND COASTAL SUBSIDENCE. COMPRESSION AL CR ATONIC B ASINS. STRCSS RE ACTIVATES FAULTS. Figure 4. Schematic diagrams illustrating the plate reconstruction of the North American craton and interactions with adjacent plates and geologic activity of the New Madrid Rift Complex since the late Precambrian. The outline of the State of Missouri is shown for location and approximate scale. (Brail e,1984) .
A. LAre rRecAusRIAn: sucnenEnT RnFTnnG
*4 y > a** v *Y n i4f d' ,* w> m 6 **
- EXTENSION C UST A T [2 ^""6
..:,,J: M +1; mMM5g .
UPPER.
\\N Md N h B. LATE PRECAhBRIAN-EARLY CAMBR AN:
. <g_- y , RIFTING
,a o v ',, '4 ,*4 , c.
*s Ad * ,;f t4 a t- 'n jh?l',(;,g;,y
..*T T 'E,'
f( :T!'*' j'
;;;,*h',[T,~; ; ,
C, D. PAtsozo C: Sues \oence
~
<,a v 3 . ,;". . y v v , . COMPRESSION
. 4 2 -s s-
. , . ,, g
^
h '.t:* M:t 'cetat4,f . . . . -^,[<7dd :K . 'y; y:g V c:.- w~-tm Ei Ti:n:n..:.E.,:n. L. xkh $ h.S.h N E. MESoZolC: UPLIFT, REACTIVATION w -- 5 . gw
^ '
e,.[ O ., d
. F. tare ussozoic-pReseur: SUBStoence, f sississione aivta COMPRESSION
e a, v4 , , > . ,
..w--
.. 4 a a
'4 a 6 , A
; :; . y I' ,7'fi$'tfA'k 1 N .Q 9,Y.'((*:' ;;t:1~dI.NI g:gi y : ~
v~
\\\\ \ ,,_ (Braile,1984)
Fiaure 5. Schematic cross sections for a northwest-southeast profile through the New Madrid Rift Complex illustrating the evolution.of the rift complex and associated cratonic basins through time. Stages of development shown in parts A through F are related approximately to the map views illustrated in the corresponding diagrams in Figure 4.
I. o ll.LIN0l$ INDIANA t 1
'\
- n- -
. . ( }
t< g % l /* MISSOURI 8 1
\ ,
/ l
) '
~
/ .
I f m - 37 l
\ )g l
{ . ( % MENTUCKY
. 9 ,
' %s.
TEhNESSkk
/ l ARKANSAS ,
g 4 35' - p . . q.---+ 7-(41 Y ft ALABAMA g . 92* So* 86* 46' t....t i o co Zoo ke Figure 6. Map showing the contours of the numbe'r of earthquake epicenters of 4modjffed Mercalli intensity greater than III from 1800 to 1972 per 10 km for the New Madrid area (after Hadley and Devine,1974). Heavy dashed lines are inferred boundaries of the rift complex as shown in Figure 3. (Braile et al ., NUREG/CR 2741).
.P
, g w w w
- Ye
.k i . :
- ix.
ep >m 3 - w
~ ~ . n. . .
/O .
+ /*
-.$ .\ -
h .* . 3 N7,' Qg /
- y. '.hs g N.
+
a% . ;q,-4 I, o ,5
\ ,
w ,{ . l gj r-} (Y f m ""* / .' .Q- C. *\.], y
. . . . g 'w g .
4 .r
- g+.-' ~ W * - s g .C
[%
~'
A /
./
- 1. ,oen d._,
a . .
*E N Y
.. . 7
. ' k ** 3. ._. _' . -m,ee
. . T jy 'wano *
.. . .R 9 .3 -
"I*/. -c .cs. i .. . . ,
. :NUREG/CR 3174)
/
- f,0 D . rws/ "! .
."ssL "*"* * ' '
/ m sas . .
- m. .
e-w=r
.N. 3 w
w w w
~
Figure 7. Earthquake epicenters, tectonic features, and inferred boundaries of the rift complex in the New Madrid area. The shaded regions are areas of positive gravity and magnetic anomalies which have been used to infer the boundaries of the rift complex. (NUREG/CR 3174). 1
O O o 40 G St. Louis, Missouri O gSe o o q *&& e e o o New Madrid- , O o :
-Dyersburg, Tennessee o
o o' Marked Tree, Arkansas
, @ 6g
$ c' O
O Figure 8 a. t .ts .r rtao.m..a.t..s.4ans 1 s.4 ut= u. 4 regional e.tv.rk f.r th. rsporttag p.rt.d 1976 = 1982. (Hermann,1983) . e v a go 0
- Oe
$ e @~ Oc>
pe Je o g e eo o New Madrid, Missouri No e e
# so o -
5 - o
~w# V
- e. . .
0- 's
<e e
fc8*e , Caruthersville, Missourt fe' q% $g
- O Dyersburg, Tennessee _
h [O o . Marked Tree, Arkansas o ek o e bco _@ N
*.a F1gure 8 b. t ts.o .c 4.t t.4 .ma 1 t.4 = tw es.
c.c .naz t.orm rtno,mi
- t. ta. c. ,or s t= es.4 ins -its: a u.
t=.4s.t. ,s.tastr er s. nause atuars. (Hennann ,1983)-
manoi. inda.n. ohi,
)
o >C 4 5.E ~
'Es FAIRFIELD w I
- t. Louis e
5 BASIN ?,. 'b )' . v.nc.nn.. - Wa terloo-Dup
, SPA RTA b " T' V '"*a
#r, SHELF d f /
'"' '* k Palmer , M Rough-C;e$** Sh'*!)e n
~
OZARK
- ~~
l .y Eaglevart,y rman syncline 3 DOME
-{e Fluorspar s,
-/ -5
.Mg",' Pennyrite Fault 6#
Poplar Bluf f/ f Area Fault s y;,, ,, ,' N,;4 Complex 's, k a'ucky 4,, . n . . , 4,, 4 )e
. Neshvule f' 4,,
'*g
) :
i NASHVILLE I. ult; tic on I downthrown ti.1e ',' DOME
,nonocian.
a -- i ...... (From Nelson and Lumm,1985, NUREG/CR 4333) Figure 9. Principal tectonic features in the New Madrid area.
- 1 p fmq
/. .. n/JC n, g%~g 1-- - jy 'f
.; ..c *
=
36' oo'
' - n - hf y
- f. --
;. j,,,
4 :*
- h.* / ..
,.,j;;;ig
- s. .
k j u f . p \' [ jI. f ./. ,' '$ ,. . n.<wuz, ..s.'.,
/
..(*v /,, ; %
~
~
pp 2% j? . . , . . . . . f; .
/ . ,h. {
r ' *
- h. .* ,*/ f .i *-.^.
c.. h e
$:'=/ . .
an % ( y4 y'*ED/
- sc.as pf# <0[ .. .., I'.?1. **
. ..'t ww -
,&...-V.....;,u,s t.>.'s ! - - ~ ~ ~ --
m.
;.v s' .I , ,.<- .a.s s e ,.a. . . s e s
i i .:' f/
! . W it'2:' K 4 '* a o
'g(.v* '
. . ge,y,g g,,,,
, . .u....
, (~ ~~ , e"J ::'.:a' v .
(Crone et al . ,1985) Figure 10 a. Prest-taarr aae sho-sa9 relattoaship of seis=icity to disturned soae mapped at the depth of magnetic basement in t%e southern poet of the me. Madrid seismic sone. Epicenters are for the period June 29. 1974 through March 28.1981. f rom R. 8. Herrmann (writtee conrwn.. 1981). om. ,
, 00
....a.s . ~ .. r ' . ;<-. , ' * .. ~
3 p( f.;.;.: ~'t.:.: W Q ; #3 e.M.
..; . a .., . . v. *:'Q: n *d
- .: .7 a - .- : a
"- .**-r-- - . ' . - - - ---------u .. / o.s s
) o s.fc.T.W 3 :. .t. i fGri.. *w .tPr . .:'" Y - J
=====_ .w -
g 2-- - - -_
- s m+ . . . = s
- .. ;- : . -. .~. g. .~ =~ . :-+ .-- +- z 4 .
.y.
f" - MhAk "!
-h'a'7'm.s.. #ff$ . . . .g. N.e kw .-
1 y Figure 10b. Seismic reflection profile c._, % % '?
,. 3 .t*
E.. 6. .c. . ,. . and interpretive line drawing showing PL ,,,. . #. M. ._- . EF.6.M,-2. .3.. ., ".. .3. . . 4= . %hr;_ details of one edge of the disturbed b.-
-e .. .c zone shown in Figure 10a. Note the
' ' structural divergence between the magnetic
, l , ,
5* * * * . ,,, basement (MB) and the slightly warped top of the Paleozoic (PZ). Specific location nf these data cannot be identified W os. - gg -
.o.S s I because they are proprietary.
5 _ ..
-e . . . .-
.g, -
I . - . - -- 3 legend: K = Cretaceous I **- . .
- .u 5a PZ= 'Ibp of Paleozoic sediments 4
.~;
_ , . . ., Q, .6 ."
. * -' S = BC= Base of marine carbonates T . .
.C MEk Magnetic basenunt
- 33. * +
. . * .. . * ~.* .
- h. N. .
esc ;.. . _.m. . g.4
~ . . -; - .~
(Crone et al . ,1985) - - - l
._. - ~ _ _ _ _ __ _ . - .
- . . - . . _ . . - - .-. -- . . - - - - _ . _. . _ - - - - - - _ - . - - = . _ .-
i i ' r~. - 90
'Ni'4(
89 %.) . 88 gi go '\g \ (, 89
% ., _)
37 i l GP g . ' n' 0 atop @ / f , - ~ 'tLD p *'* *= == ) y
, ':::.,_ _mg -s i n - - - g
- S =
o 4 . s --~% \i g
~
s' ' sI ' ' '~ ~'
,1l' ' a lk so
-* #
- I 4 "r. .\}'"
- -- -J O g- -8 ',
,d -
,- moins
', , s .,.c
i
- i o,,+
g / p j 4 s
- ? y,y,11
? , ,
d s
- g I f77 g'
# . e .
- *j l
,, ? ? f _< c _
o utlios -
$ 'm,
^
P
)[
* =
0 o bk
, sotor
#a:
-
- h J. o
" )
i V , j.a 1/ eg/y}~" 3 de -- .- -- __ . - -- de 1 -- -- ~~ ~~ 35 ~ ! o so us. o, so ut
- 9 s_o m 9 s.o nu i
a Figure 11. Temperature gradients in Figure .12. Temperature gradients in relation to the edge of the Reelfoot relation to the edge of the Reelfoot Rift and location of earthquakes. Contours Rift and buried plutons. (Modified from are in degrees c/km. Note the correspond- NUREG/CR 4226). ance of contours with the Dyersburg Line and earthquakes. (Modified from j NUREG/CR 4226)
MAJOR GEOCHRONOLOGIC AND CHRONOSTRATIGRAPHIC UNITS Age estimetes'd el Subderisions in use by the U. S. Geological Sunroy bounderies in tend their map symbolst million years Im.v.! Quaternary Holocene Epoch or Ser'es , Pered or System 0 010 101 Pleistocene Epoch or Seres 2 11.2 211 - Neogene Piecene Epoch or Series Cenosoic Subperiod or g4.g .5.31 - Erae' Tertiary Subs tem
, 3,g
[tathem Pered o, 24 123 261 - grf,] Paleogene Oligocene Epoch or Series System 38 I34 383 ~ Subgened or gig [ocene Epoch or Series Subsystem 55 154 561 - IPgl Paleocene [poch or Seres 63 163 661 - late Cretaceous (poch or Upper Cretaceous Series Wesozoic Cretaceous Pered . 96 195 971 - or. System (Kl. Early Cretaceous fooch or [on o, I'* ' lower Cretaceous Seres (rathem . 138 l))$.14ll-tanothem gwg Jurassic Pered or System IJi 205 1200 2151 - Triassic Penad or System th i
~240 -
Permian Pered or System IPl
" d "'
Carboniferous '[,"$'$"s,, Periods or ~330 - Patectoic Systems ICI M'55'55'CD'an Pened or frao,
' ' ' '"' 360 1360 3651 -
Devonian Period or System 101 (Pil Silunan Pered or System 151 435 1435 4401 - Iledovician Pered or System 101 500 1495 5101 - Cambnan Pered or System ICI
~ 570U -
Protercioic 1 til'J Proteccioic 800 - [on or Protercioic Y lYl'/ [onothem (El 1.600 Protercioic X IXIU 2.500 Archean (onor fonothem lAl Oldest known rocks in U S 3.600 - U anges R reflect oncer amties of isotopic and bestratigraphic age assignments Age of boundanes not closely bracheied by einstmg data s'own by ~. Decay constanis and esotope rates employed are cited in Steiger and Jager l19771. U ocks R older than 5 0 m y also calied Precambnan (pCl, a time term wahaut specific rank
& Time terms wahou specific rank Geologic Names Commater.1980 edition Figure 13. Geologic Time Scale. (U. S. Geological Survey)
, t' '. Harold Denton, Director 13 JUN 301986
- 4. The following generalizations should be considered in drawing seismic hazards maps for sites in or near the New Madrid area.
- a. The zone of highest seismicity is located within the New Madrid Rift Complex and on the flanks of mafic plutons situated on its margins.
- b. The Reelfoot Rift is seismically the most active area. Within the last 2000 years it has experienced at least three intensity XI to XII events.
- c. The other three arms of the New Madrid Rift Complex, i.e., the St. Louis, Indiana, and Rough Creek, should be considered seismically active but to a lesser degree than the Reelfoot Rift.
- d. The seismicity of the Nes Madrid Rift Complex is distinct from that of the Anna, Ohio area. Based on geophysical evidence, the Indiana-Arm.of the New Madrid Complex appears not to extend north of latitude N39.5 .
Os%1aal signed byr 201ER2' 3..,gIjegg Robert B. Minogue, Director Office oj . Nuclear Regulatary__B.e. search n
. c
Enclosure:
List of Selected References ' P d : c t. .'n e n . MD[_,_ JQL .19.k.. Distribution /R-2811: _ , , _ Circ /Chron Econti ?- ro. , _ _ _ , . DCS/PDR. LBeratan '/. : ah Request No. _ _ _ , . , _ [ ESB Sbj/Rd AMurphy . ru so, RBMinogue D/N: Memo Denton/146 22G No. ____ Dross E0'Donnell Docket No. KGol1er Rulemaking No. ME7[E. Other Return NRC-3L8 to RES Yes_No_
- Note: See. Previous Concurrences 0FC:ESB:RES:Jk :ESB:RES :ESB:RES :DRPES:DD :DRPES:D :RESDD :RE 3 g NAME:E0'Donnell?AMurphy *:LBeratan*:Econti* :KGoller* : Dross * :RBVinogue DATE:6/ /86 :6/ /86 :6/ /86 :6/ /86 :6/ /86 :6/ /86 :6/30/86 0FFICIAL RECORD COPY
Harold Denton, Director 12
- 4. The following generalizations should be considered in drawing seismic hazards maps for sites in or near the New Madrid area.
- a. The zone of highest seismicity is located within the New Madrid Rift Complex and on the flanks of mafic plutons situated on its margins,
- b. The Reelfoot Rift is seismically the most active area. Within the last 2000 years it nas experienced at least three intensity XI to XII events..
- c. The other three arms of the New Madrid Rift Complex, i.e. the St.
Louis, Indiana, and Rough Creek, should be considered seismically active but to a lesser degree than the Reelfoot Rift.
- d. The seismicity of the New Madrid Rift Complex is. distinct from that of the Anna, Ohio area. Based on geophysical' evidence, the Indiana Arm of the New Madrid Complex appears not to extend north of latitude N39.5 .
Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosure:
List of Selected References. Distribution /R-2811: Circ /Chron RMinogue Econti D/N: Memo Denton/146 DCS/PDR Dross LBeratan ESB Sbj/Rd KGoller AMurphy E0'Donnell l (fk ESB:RES: # Q
-ESB:RES Yb
.ESB:RES DR hh
- DD DRPES:D R S DD'. RES:D E0'Donnell AMurphy LBeratan EC ti _KGoller DRpfss RMinogue 05/Ap/86 05/t.7/86 0 786.p8 05/ / 05/2 7/86 Obb y86. 05/ /86
[ o,, o UNITED STATES NUCLEAR REGULATORY COMMISSION t j WASHINGTON, D. C. 20555
~s ...../
JUN 301986 MEMORANDUM FOR: Harold Denton, Director Office of Nuclear Reactor Regulation FROM: Robert B. Minogue, Director Office of Nuclear Regulatory Research
SUBJECT:
RESEARCH INFORMATION LETTER NUMBER 146, RESULTS OF NEW MADRID SEISM 0 TECTONIC RESEARCH PROGRAM ABSTRACT The New Madrid earthquakes.of the Central Mississippi River Valley (1811-1812) were among the strongest known seismic events ~in North America. Their intens-ity has been estimated'at'XI-XII on the Modified Mercalli scale and the felt area extended eastward as far as Charleston, SC, Washington, DC, New York City, and Hartford, CT. The earthquakes are unusual in that they occurred within a continental plate rather than at a plate margin as is the case for most large earthquakes. Using only surficial geology and historic seismicity,. it was not possible to relate the New Madrid earthquakes to a specific geologic structure or capable fault. To resolve uncertainty in.the causative mechanism of these earthquakes, a multidisciplinary program of investigations was begun by the NRC in 1976 called the New Madrid Seismotectonic Research Program. This program is now at a stage where a conceptual model can be constructed that explains the recurring intraplate seismicity of the Central Mississippi River Valley. The earthquakes are spatially associated with fault movement within an ancient rift complex called the New Madrid Rift complex. This rift complex can be delineated based on geological, oeophysical, and seismic evidence. This is significant to NRC regulatory activities because it means that seismicity in the area can be linked to a tectonic struc ure of finite extent. The rift complex formed in late Precambrian to Cambrian time (1,500 to 500 million years ago). The contemporary stress regime is quite different (compressional as opposed to tensile) from the one that produced the original zones of weakness. Several generalizations can be made that would be significant in drawing a seis-mic hazards map for the New Madrid area: (1) the Reelfoot Rif t is seismically the most hazardous portion of the rift complex; (2) the recurrence interval for events large enough to cause liquefication in the area is about 600 years; and (3) the remaining portions of the rift complex (i.e., the St. Louis and Indiana arms and Rough Creek Grabe ) sFould be considered seismically active but to a lesser extent than the Reeifrot Rift.
Harold Denton, Director 2 rdlM 30 BB6 INTRODUCTION
Background
The series of earthquakes which occurred in the Central Mississippi River Valley near New Madrid, Missouri, in the winter of 1811-1812 were among the strongest known earthquakes in North America. Three major earthquakes, plus 203 damaging aftershocks, occurred in the 3-month interval from December 16, 1811 through March 15,1812 (Nuttli,1982). They are unusual in that they happened in the interior of a continental plate (North American Plate) rather than at a plate margin. Thus we are dealing with "intraplate tectonics." The intensity of the earthquakes was extremely high. Nutt11 (1982) estimates an epicentral Modified Mercalli Intensity (MMI) of XII and a body wave magnitude of 7.2, 7.3, and 7.5 for the three major earthquakes. The zone of greatest deformation (i .e. , fis-suring, sandblows, landslides, subsidence) extended over an area of 48,000 square kilometers (approximately the area of West Virginia), and the earthquakes were felt as far away as Hartford, CT, Washington, DC, Charleston, SC and New Orleans, LA. Since 1812 there have been at least 20 damaging earthquakes of MMI intensity
-VI-VII to XI (Nuttli,1982, p.15) in the Central Mississippi River Valley (see Figure 2 for their location) making this one of the most seismicelly active areas in the United States east of the Rocky Mountains. Although the frequency and severity of earthquakes in this area are high, until very recently, the earthquakes could not be directly linked to geologic structures as can be done in some areas such as California. This is due to: (1) deep burial of the 3 causative structures, (2) the fact that most of them do not extend to the earth's surface, (3) vegetative cover which obscures faults and rock exposures, and (4) the fact that until about 1974 there was very little geophysical data available for the Central Mississippi River Valley that could provide a basis for interpreting subsurface structure. The lack of definitive evidence linking seismicity with geologic structures was a matter of serious concern for, in the absence of this evidence, Appendix A to Part 100 requires nuclear power plant license applicants to design the facility to withstand the largest earthquake which could occur in the tectonic province where the power plant is located.
In licensing actions going back to the early 1970's the AEC/NRC licensing staff dealt with the New Madrid earthquakes as an anomalous zone of high seismicity within the Mississippi Embayment portion of the Gulf Coastal Plain and the adja-cent Wabash River Valley. In the early 1970's the staff had to rely princi-pally on the pattern of seismicity which centered on the Central Mississippi River Valley,.the frequent recurring earthquakes in the area, and surficial geology. With this limited evidence the licensing staff treated the New Madrid earthquakes as being linked to buried unidentified structure. The zone of New Madrid seismicity was presumed to extend from Memphis, Tennessee, to Vincennes, Indiana. To resolve the causative mechanism, that is, to see if the New Madrid earthquakes could be linked with geologic structure the Office of Nuclear Regulatory Research initiated in 1976 a program of confirmatory research. The purpose of the research was to provide the licensing staff with information to support regulatory decisions. The research program was a cooperative multi-disciplinary effort called the New Madrid Seismotectonic Research Program and
l Harold Denton, Director 3 ' M 3 G 1986 l funding was shared by 7 State Geological Surveys, and several academic institu-tions in cooperation with the U.S. Geological Survey, National Science Founda-tion, and Tennessee Valley Authority. Description of the Program The New Madrid Seismotectonic Research Program was a coordinated program of geo-logical, geophysical, and seismological investigations of the area within a 200-mile radius of New Madrid, Missouri, which concluded in FY 1986. The loca-tion of the New Madrid Area is shown in Figure 1 and the contributions of the participants is summarized in NUREG/CR-4632 (Buschbach, 1986). The program was designed to define the structural setting and tectonic history of the area in order to evaluate earthquake hazards, particularly to nuclear facilities. Specific program objectives are: (1) to identify the earthquake source and source mechanisms for the numerous recurring earthquakes in the Central Mississippi River Valley; (2) to delineate the zone of high seismic activity; and (3) to determine the recurrence interval of potentially damaging earth-quakes in the area. Participants in the program have included scientists from the State Geological Surveys of Illinois, Indiana, Kentucky, Tennessee, Alabama, Arkansas, and Missouri; faculty members of Earth Science Departments at Saint Louis Univer-sity, Vanderbilt University, Purdue University, Memphis State Universit.y, Eastern Kentucky University, the University of Pittsburgh, the University of Kentucky, the University of Wisconsin at Milwaukee, and the University of Texas at El Paso; and staff members of the U.S. Geological Survey, U.S. Army Corps of Engineers, and the Tennessee Valley Authority. RESULTS OF THE PROGRAM Integration of Results into a Conceptual Madel The program has produced significant results which led to the formulation of a conceptual model that explains the New Madrid seismicity based on many lines of evidence. For an historical review of key developments in the formulation of the conceptual model see USGS Open File Report 84-770, pp. 6-10. Among them are: seismic history, contemporary seismicity, geophysical exploration tech- ' niques (particularly gravity and aeromagnetic data and seismic reflection pro-files), subsurface geology, surface geology, stress measurement data, and hydro-geochemical data. The conceptual model relates the New Madrid earthquakes to the reactivation of basement structures within a rift complex that formed in late Precambrian to Cambrian time. (A geologic time scale is provided in Figure 13.) The model relates the seismicity to a contemporary compressional stress regime greatly different from the tensional stress field which originally produced the rift complex. Thus the research has reduced uncertainty for the licensing process by relating seismicity to geologic structures which have a finite extent. The conceptual model is shown schematically in Figures 3, 4, and 5. Braile, Keller, Hinze, and Lidiak (1982) named overall rift complex the New Madrid Rift s
Harold Denton, Director 4 JUN 301986 Complex and the term is accepted by many researchers in the area although other terms such as Reelfoot Rift, New Madrid Seismic source zone, and Mississippi Valley Graben are used by others. As can be seen in Figure 3b, according to Braile et al. (1982), there are four structural components of the New Madrid Rift Complex: (1) the Reelfoot Rift, (2) the Rough Creek Graben, (3) the Indiana Arm, and (4) the St. Louis Arm. Their presence has been inferred from seismic reflection profiling, aeromagnetics, gravity, and subsurface geology as revealed by deep boreholes. As with many conceptual models, one must realize that it is an early approximation which fits the available data. As more data become available, details of the conceptual model may change. Summarized below are the lines of evidence which led to the formulation of this model which specifically proposes a reactivated rift as the causative structure for the New Madrid earthquakes of 1811-1812. Specific Research Results Seismicity Data: Until the early 1970's, the seismicity data base for the New Madrid area was limited to: (1) that gathered by instruments located far frorr the epicenters of the 1811-1812 events, and (2) human observations which were recorded in news-paper accounts and journals. Hadley and Devine (1974) collated the available data from 1800-1972 and produced a seismic frequency map (Figure 6). The seis-mic frequency contours from the Hadley and Devine map show a Y-shaped pattern which matches the New Madrid Rift Complex identified by this research. It follows the Mississippi River to southern Illinois where one branch follows the Mississippi northwestward toward St. Louis. The other branch follows the Ohio and Wabash Rivers northeastward. In part, the Y-shaped pattern may reflect location of observation points since population density was highest along the rivers, or it may reflect amplification of ground motion in the alluvial filled valleys of the Mississippi, Ohio, and Wabash Rivers. However, even with a possible bias due to population distribution, seismicity was indeed consider-ably higher than in the remainder of the Central Stable Region. Beginning in the early 1970's the NRC, USGS, NSF and TVA in a coordinated pro-gram funded installation of seismographic networks in the New Madrid area. Their purpose was to determine levels of seismicity, recurrence intervals, location of epicenters, and ground motion characteristics of the earthquakes. The results are summarized in Figures 7 and 8, which show epicentral location of detected earthquakes. Their gross pattern is similar to the Hadley and Devine map (Figure 6). In the Reelfoot Rift (see Figure 8b for location) there are two principal northeast-southwest trending zones extending from Cairo, Illinois into northeastern Arkansas. These two zones are offset from one another in the region between New Madrid, Missouri, and Dyersburg, Tennessee, by a zone of northwest-southeast trending seismicity. The seismic zones are thought to represer.t subsurface faults. Composite focal mechanism studies on the two northeast-southwest zones indicate probable right lateral strike slip movement and the focal plane studies on the northwest-southeast zone suggest reverse faulting (Herrmann and Canas, 1978). There may be another northeasterly trending zone of high seismicity parallel to the postulated western margin of the Reelfoot Rift but outside of the rift zone. However, the pattern is not
Harold Denton, Director 5 JUN 301333 distinct. Nuttli (1982, p. 15) notes that the three major earthquakes of 1811-1812 occurred in the central portion of the Reelfoot Rift. In the St. Louis Arm (see figure 8b for location) the seismic activity trends northwest from Cairo, Illinois towards St. Louis. Seismicity appears not to extend beyond St. Louis. Seismic activity is low in the Indiana Arm and Rough Creek Graben. Earthquake focal depths in the Reelfoot Rift are generally 3 to 15 km (Buschbach, 1986, p. 58). This puts them either within a zone of disturbed seismic reflec-tions (Figure 10b shows this disturbed zone) or within the magnetic basement rocks. In the Indiana Arm, several earthquakes had focal depths greater than 16 km. This makes them the deepest earthquakes recorded thus far in the Central Mississippi River Valley. Buschbach (1986) notes that these earthquakes appear to be associated with the eastern edge of the Fairfield Basin, a depression in the Illinois Basin, rather than with the Wabash Valley Fault System which is located 30 km to the east (see Figure 9 for location of the Fairfield Basin and Wabash Valley Fault System). Aeroinagnetic and Gravity Data: The pattern exhibited by aeromagnetic and gravity surveys of the Central Mississippi River Valley correspond strikingly well with seismic data (see previous section). In particular the data have delineated a northeast striking basement depression about 70 km wide and more than 300 km long (Kane et al., 1981; Hildenbrand et al., 1982). The depression, the Reelfoot Rift portion of the New Madrid Rift Complex, is remarkably linear over its entire length, has nearly parallel sides, and is at least 2 km deeper than the surrounding base-ment (Kane et al . ,1981; Hildenbrand et al . ,1982). Both gravity and magnetic anomalies are aligned along the boundaries of the depression. The geophysical signatures along the rift boundaries are characteristic of mafic plutons. Esti-mated magnetization directions for the postulated plutons indicate a Mesozoic age (Kane et al., 1981). Similar to the Reelfoot portion of the New Madrid Rift Complex, the St. Louis Arm, Indiana Arm and Rough Creek Graben can be delineated on their aeromagnetic and gravity signatures (Braile et al . ,1982). The St. Louis Arm is the broadest of the four arms. The southwest side has the strongest aeromagnetic gradient of any part of the New Madrid Complex. .However, its northeast side is not as distinct. Its presence is deduced primarily from interpretations of the patterns exhibited on aeromagnetic and gravity maps. Unlike the other arms of the New Madrid Rift Complex, there is no significant thickening of Paleozoic sediments in the St. Louis Arm (Schwalb, 1982; Buschbach, 1986, p. 26). From stratigraphic evidence.alone one would not deduce its presence. . Gravity ano aeromagnetic data (Braile et al . ,1982; Braile et al . ,1984; Gori and Hays,' 1984) indicate that the Indiana Arm is restricted to the southwestern part of Indiana south of N39.5 latitude (see Figure 7). This is significant because Wollard (1958) using only seismicity called attention to an apparent alignment of the earthquakes of what is now called the Reelfoot Rift with the earthquakes of the St. Lawrence River valley. The Anna, Ohio earthquakes fall within this apparent alignment. Seismicity is low in the Indiana Arm and the most seismically active portions of it appear to be limited to the region south of the Kentucky Fluorspar District.
. . - _ m
Harold Denton, Director 6 h .'li M S As to the cause of the Anna, Ohio earthquakes, Braile et al., (1985) have investi-gated the possible connection of the Indiana Arm and Anna, Ohio. Based on an integration of aeromagnetic, gravity, and subsurface geology, they have not found sufficient evidence to warrant extending the Indiana Arm northeastward to Anna, Ohio. The Anna, Ohio earthquakes appear to be localized in a structurally complex area near the intersection of the northeast trending Grenville Front and a postulated rift complex, the Fort Wayne Rift, which strikes northwest-southeast. The Fort Wayne Rift has an orientation and geophysical signature similar to a Keweenawan Age rift complex found in the Michigan Basin. From the geological / geophysical data available Braile et al. (1985) interpret the Anna, Ohio seismicity as being caused by one or more of the following: (1) lithologic discontinuitie's associated with the contact between rift type volcanics of the Fort Wayne Rift and low density granitic rocks of the Grenville terrane; (2) reactivation along graben type faults on the northeast flank of a mafic volcanic body within or on the margins of a -60 milligal gravity anomaly which is thought to be a icw density granitic body; and (3) basement inhomogeneities within Grenville basement rocks. Seismic Reflection Data: Indiana Arm: Seismic reflection data is available for the Indiana Arm and the Reelfoot Rift. Work on the Indiana Arm was done by Purdue University under NRC funding (Hinze et al., 1983; Sexton et al., 1985). The Purdue seismic lines detected faulted layered Paleozoic sediments. More importantly, that data indi-cated a thickening of pre Mt. Simon layered rocks in the Wabash Valley Area. (Note: the Mt. Simon Sandstone is late Cambrian in age or approximately 500 million years old.) Approximately 2 kilometers of sediment are found in this area but none in the area to the west. This is significant, for the seis-mic data indicate the presence of a down-dropped fault block in the Wabash Valley Area, which received a thick accumulation of sediment prior to the late Cambrian. Reelfoot Rift: The seismic reflection data for this area are proprietary infor-mation. The USGS purchased over 250 km of multichannel, common-depth point seismic-reflection profiles (Crone et al.,1985). A more complete presentation of the seismic profiles is found in Howe and Thompson (1984). There are several deep wells (i.e., Dow Chemical No. 1 Wilson and Houston Oil and Minerals No. 1 Singer) which allow tentative correlation of the principal seismic reflectors. The seismic reflection data confirm the existence of a fault bounded down dropped basin, i.e., the Reelfoot Rift. The Reelfoot contains a sedimentary section greater than 30,000 feet thick. This is considerably thicker than the area.outside the Reelfoot Rift where sediment thicknesses are around 5,000 feet or less. The thickest part of this sedimentary sequence is found below late Cambrian sediments. Based on sediment thickness patterns, the Reelfoot rift must have formed prior to the late Cambrian. The seismic profiles in the area show a variety of structural and stratigraphic features which allow a reconstruction of the history of the Reelfoot (see the section on Subsurface Geology which follows on page 9). Noteworthy structural features are the faults bounding the down dropped basin, numerous other high
Harold Denton, Director 7 gg angle faults, and a prominent zone of disturbed seismic reflectors found along the axis in the southern portion of the Reelfoot Rift for a distance of 120 miles from Caruthersville, Missoeri, to Marked Tree, Arkansas (see Figures 10a and 10b). This disturbed zone does not show on aeromagnetic or gravity maps. Its presence was revealed by the seismic reflection data. The origin of the dis-turbed zone is not known. Howe and Thompson (1984) postulate that it originated from late Paleozoic compressional forces which uplifted and reactivated movement on late Precambrian-late Cambrian growth faults. Crone et al. (1985) argue for the disturbed zone being caused by felsic intrusive of post-Paleozoic, pre-Cretaceous age. Regardless of t'he disturbed zone's origin, the correlation between earthquake hypocenters and it are so pronounced that Crone et al. (1985) conclude that the mapped extent of the di.sturbed zone defines the source zone for the Reelfoot's axial seismic zone. Note in Figure 10a the close correspondence between earth-quake epicenters and the disturbed zone. The dashed line signifying the outer edge of the disturbed zone encloses r.ost of the epicenters. In addition, Crone et al. (1985 p. 549) points out that hypocenters in-the axial seismic zone occur within the disturbed zone. As to its extent, Crone et al. can trace the dis-turbed zone only 25 km southwest of Marked Tree, Arkansas, although geophysical data suggest that the Reelfoot Rift continues a considerable distance farther to the southwest. Thus, the axial seismic source zone does not appear to extend to the southwestern end of the rift. At the northeastern end of the axial zone near Caruthersville, Missouri, the reflectors do not appear to be upwarped; however, Crone et al., 1985 report that i in this area, narrow zones of disrupted reflectors, which they interpret as fault zones, coincide with the seismicity. The northeast end of the source zone probably merges with the Dyersburg, Tennessee-New Madrid seismicity trend. The latter trend is a northwest trending band of seismicity (see Figure 8b for loca-tion). Zoback and Hamilton (1980) postulate that the seismicity of this trend is spatially associated with igneous bodies of various ages, presumably emplaced along upper crustal zones of weakness that formed during the initial rifting. The zones of weakness probably formed in a regional extensional stress field rather than the current east west compressive stress field indicated by fault plane solutions. Thus, from the seismic profiles coupled with seismic network data, one can define the extent of a major intraplate seismic source zone in the Reelfoot Rift on the basis of geologic structure. The profiles also emphasize the role of pre-existing rift related structures on the location of intraplate earthquakes. Hydrogeologic Data: Thermal data: Stearns and Reesman (Buschbach, 1984, pp. 145-162 and Buschbach, 1985, pp. 105-128) made ground water temperature measurements in 80 water wells in the Central Mississippi River Valley with the objective of locating zones where there is rapid ground water flux fron depth. Such a zone is postulated to signify an unhealed fault. Vanderbilt located two significant anomalous areas. The principal one was in the Missouri bootheel and adjoining Lake
l i Harold Denton, Director 8 l JUN 301986 I County, Tennessee (see Figure 11). This is in the area where many microearth-quakes occur, and also where the Reelfoot Rift and Pascola Arch intersect (see Figure 9 for location of the Pascola Arch and Figure 8b for the location of the microearthquakes). A second area occurs along the Tennessee-Kentucky border, J possibly along the east edge of the Reelfoot Rift. There was a remarkable i correspondence between the thermal anomalies, some geologic structures in the area, and a zone of numerous microearthquakes. The thermal data can serve as a reconnaissance tool to locate areas worthy of further detailed investigation. Hydrogeochemical data: Stearns and Reesman also studied the ground water chem-1stry of the 80 wells in which temperature measurements were made. The objec-tive was to de'termine if chemical patterr.s could be linked with seismic activity and geophysical anomalies. Chemical patterns are difficult to draw from the sampled water wells because the data must be segregated by aquifer. However, some interesting relationships were identified. (Details are found in Buschbach, 1985, pp. 106-128.) The chloride level is high in alluvium near the Mississippi River in Missouri in the area of high earthquake _ activity al.ong the Pascola Arch. Apparently there is vertical leakage along faults', from the McNairy Sandstone aquifer, from a depth of approximately 2000 feet to the land surface. Barium concentrations relate partially to the buried structures which were interpreted as plutons from aeromagnetic and gravity data. Highest concentrations of barium occur on and near the Missouri and Arkansas plutons on the west side of the rift. Lithium concentrations also are higher on the west side of the rift, notably in water above the Bloomfield pluton. Some of that lithium may origi-nate from ground water flow along vertical fractures extending from the Bloom-field pluton to the ground surface. Strontium concentrations are high within the rift. Both lithium and strontium may well relate to the sediments in which the aquifer occurs. In addition to the lithium which may have come from verti-cal flow of ground water along fractures, igneous minerals containing lithium may have been washed or blown into the depositional area (Eocene volcanic activ-ity). Strontium could be leached from aragonite-rich shells in marine sediment that might concentrate more in the rift if water ~was deeper there. As was the case with the ground water temperature data, the hydrogeochemical data also proved to be a useful reconnaissance tool. However, the chemical patterns can be complex and they require careful interpretation. Surface Geology: Investigation of major fault systems: Under sponsorship of NRC's New Madrid Seismotectonic Research Program, the major fault systems (see Figure 9) within a 200 mile radius of New Madrid were examined in detail to see if there was evidence of geologically recent offset, that is, movement during the Quaternary Period (approximately the last 2 million years). Results of that work is found in Kolata et al. ,1981, Nelson and Lumm 1984 and 1985, and Ault,1985. In the limited number of places where Quaternary materials cover these faults, no evidence of Quaternary reactivation was found.
. Trench investigations: The U.S. Geological Survey excavated a trench across the Reelfoot scarp, within the Reelfoot Rift, about 20 miles southeast of New Madrid, Missouri (Russ 1979). The scarp is a prominent linear feature 3 - 9
Harold Denton, Director 9 JUN 301986 meters high in an area which has little relief. It strikes in a northerly direction and forms the eastern boundary of the Tiptonville Dome. Deep seismic profiling in the area has shown the presence of a buried fault beneath the Reelfoot scarp which offsets Paleozoic, Mesozoic, and Tertiary strata. The Reelfoot scarp is thought to be an upward continuation of faults identified in the subsurface. The trench revealed a number of small faults and zones where there was ductile deformation of sediment layers similar to what one sees with seismically induced sand boils. Radiocarbon dating of mollusk shells in the trench indicates a maximum age for the sediment exposed in the trench of 2000 years. Field relations of the faults and sand boils indicate three epi-sodes of movement within the last 2000 years. That movement is presumably related to New' Madrid type events and it suggests a recurrence interval of about 600 years for earthquakes large enough to produce ground motion great enough to liquefy sand in the alluvium of the New Madrid area. Farther east in the , Kentucky River Fault system, about 300 miles from New Madrid but close to the location of the Sharpsburg, Kentucky 198J earthquake (body wave magnitude.5.1), nine trenches were excavated at four sites under a grant by NRC to the Kentucky Geological Survey (Van Arsdale and Seargeant,1986) across mapped faults that . were partially covered with Pliocene-Pleistocene age sediment. In four of the nine trenches Van Arsdale and Seargeant (1986) identified faulted or folded sediments. In their opinion the faulting an folding of the Pliocene-Pleistocene age sediment is tectonic in origin and that the Kentucky River Fault system has been active within the last 5 million years. Subsurface geology: There have been 434 deep holes drilled within a 200-mile radius of New Madrid, Missouri. Much of the geologic history of that area can be deduced from lithologic and stratigraphic study of the materials encountered in these holes. Details of the geologic history can be found in Schwalb (1982) and Buschbach (1986, pp. 37-54). See also Figure 5 of this Research Information Letter for a short summary of the geologic history. Thirty-eight of them pene-trate the Precambrian crystalline basement. The basement rocks are a mixture of granites, rhyolites and metamorphosed sediments. Age of the crystalline basement varies depending on location and the range is from around 1.0 to 1.5 billion years before present. The time of first rifting within the New Madrid Rift Complex is not known. However, sediments older than the late Cambrian are significantly thicker in the Rough Creek Graben and Reelfoot Rift than on the margins of these structures. That information plus the data provided through gravity, aeromagnetic, and seismic profiling point to the existence of linear troughs bounded by high angle faults dating from before late Cambrian time. Thus the troughs formed over 500 million years ago. For the remainder of the Paleozoic Era,. sediments thickened in the Reelfoot and Rough Creek areas indi-cating topographic depressions but thicknesses were only.slightly greater (a few hundreds of feet vs. thousands of-feet prior to the late Cambrian) than l the surrounding areas. Apparently after an initial episode of rifting a topo- I graphic depression was created. Although tectonic activity ceased, this area l continued to subside at a slightly greater rate than the surrounding craton. I The longevity of the Reelfoot Rift as a topographic depression is indicated by J the fact that according to Potter (1978), the lower Mississippi River, i.e. , J south of the confluence with the Ohio River, has existed in approximately its i present position for at least 250 million years and that its course has been structurally controlled, probably by the Reelfoot Rift. This correlation of
Harold Denton, Director 10 MN 301986 large river systems with rift structures is a common geological occurrence according to Potter (1978). There was a long hiatus between the last Paleozoic sedimentation in the New Madrid area and the first Mesozoic deposits of Cretaceous age. From early Permian to late Cretaceous the sedimentary history of the New Madrid area is unknown because any sediments deposited have been completely removed. This interval is probably represented by uplift and erosion. Drilling in the area indicates that a karst topography developed on the land surface with sink holes later filled with late Cretaceous and younger sediments. In Early to Mid-Mesozoic time Eastern North America underwent rifting as North America and Africa were split apart by plate tectonics (see Figure 4). The tensional forces which produced the rift basins in Eastern North America (e.g. , Newark, Gettysburg, Richmond, Culpeper and Scottsville Basins) were also felt in the New Madrid area. They resulted in. reactivation of the rift complex and igneous bodies' were intruded along the margins of the ancient rift complex (see Figure 5). Subsurface information on the igneous bodies is derived primarily from geophysical data, principally gravity, aeromagnetics, and seismic profiling. The seismic profiling points to their emplacement before the late Cretaceous (Crone et al., 1985, p. 549). Surface geology supports this interpretation because to the west of the New Madrid area in Central Arkansas at Magnet Cove, and near Little Rock, there are surface outcrops of early Cretaceous syenites which have a similar gravity and aeromagnetic signature to circular features seen in the gravity and aeromagnetic maps of the Mississippi Embayment. As North America and Africa continued to drift apart for the last 200 million years the regional stress field in Eastern North America has become a compres-sional one. It is this compressional stress field which is believed to have reactivated some of the old buried faults associated with the New Madrid Rift Complex.
SUMMARY
AND CONCLUSIONS The New Madrid Seismotectonic Research Program was a program of confirmatory research to provide information to support licensing decisions as to the cause of the large damaging earthquakes which.have been observed in the Central Mississippi River Valley near New Madrid, Missouri. Prior to the program's inception in 1976, the NRC licensing staff had to rely primarily on surficial geology, a limited number of deep drill holes, and the pattern exhibited by earthquake epicenters. At that time it was not possible to link the earthquakes to known surface faults. To resolve the origin of the New Madrid area earth-quakes and.the extent of the seismically active area.the New Madrid Program. drew on many scientific disciplines. It was necessary to integrate surficial geology with subsurface geology through the use of improved earthquake locations, deep drill holes, geophysical exploration methods st.ch as gravity, aeromagnetics, electrical resistivity, and seismic reflection and refraction profiling. These data were then linked with the seismicity and seismic history of the area. Through this integration of many lines of evidence a conceptual model was devel-oped which provides a framework that relates the New Madrid area earthquakes to geologic structure. In this model the carthquakes are associated with an
Harold Denton, Director 11 DON 3 0 IS86 ancient rift complex. The New Madrid area earthquakes are the result of reac-tivation of structures associated with the formation of the rift complex such as deeply buried faults which extend into the earth's crust, or inhomogeneities associated with a buried zone of disturbed seismic reflectors found along the axis of one part of the rift complex and buried igneous bodies along the margins of the complex. Summarized below are significant points to be noted in licens-ing actions dealing with the cause and extent of the New Madrid seismicity:
- 1. The New Madrid area is located within the interior of the North American tectonic plate. (Areas of high seismic activity within crustal plates are rare.)
- 2. With respect to both size and number of earthquakes, the New Madrid area is currently the most seismically active region of the United States east of the Rocky Mountains (Herrmann, in Gori and Hays,1984, p. 276; Johnson and Nava, in Gori and Hays, p. 283).
- 3. The New Madrid earthquakes of 1811-1812 are associated with tectonic struc-tures which are part of the New Madrid Rift Complex. The New Madrid Rift Complex is a major tectonic feature within the North American Continent.
It represents a " capable structure" within the meaning of Appendix A to 10 CFR Part 100.
- 4. Integration of geological / geophysical / seismological data has resulted in a conceptual model which relates historic and contemporary seismicity to geologic structures in the area.
- 5. The conceptual model consists of a rift complex, called the New Madrid Rift Complex, which has four arms--the Reelfoot Rift, the St. Louis Arm, the Indiana Arm, and the Rough Creek Graben.
- 6. The rift complex was formed in late Precambrian to Cambrian time (1,500 to 500 million years ago) in a tensional stress regime. There is suffi-cient geological / geophysical evidence to show that the Reelfoot, Indiana, and Rough Creek portions of the rift complex are down dropped fault blocks bounded by growth faults.
- 7. In late Paleozoic time (approximately 290 million years ago) the rift com-plex underwent compression. This resulted in: (a) uplift and subsequent erosion of sediments previously deposited in the rift complex, (b) folding, (c) re'newed movement on pre-existing faults, and (d) igneous activity pre-sumably along pre-existing fractures.
- 8. In late Mesozoic to early Eocene time (approximately 130 to 40 million years ago) there was renewed tectonic activity in the New Madrid Rift Com-plex as evidenced by renewed igneous activity and faulting of sediments up through the Eocene Epoch.
- 9. Contemporary stress measurements show that the New Madrid area is now in a compressional stress regime where the principal compressive stress is oriented approximately eastnortheast-westsouthwest.
Harold Denton, Director 12 JUN 3 61986
- 10. Of the four arms of the New Madrid Rift Complex, the Reelfoot is by far the most active seismically.
- 11. Based on USGS trenching across the Reelfoot Scarp, the recurrence interval for earthquakes in the Reelfoot Rift large enough to produce ground motion great enough to liquefy sand in the alluvium of the New Madrid area is about 600 years (Russ, 1979).
- 12. In the southern part of the Reelfoot Rift there is an axial zone extending southwest from Caruthersville, Missouri to Marked Tree, Arkansas which shows as a " disturbed" zone in seismic reflection profiles. More than 90%
of contemporary earthquakes in the region coincide with this disturbed zone (McKeown, in Gori and Hays, 1984, p. 11). The disturbed zone also underlies the area containing liquefaction features (sand blows) associated with the New Madrid 1811-1812 earthquakes.
- 13. The. origin of the disturbed zone in the southern portion of the Reelfoot Rift is not known. It appears only in seismic reflection profiles and is-absent from a'eromagnetic and gravity maps. Crone et al. (1985) postulate that it represents an uplifted, faulted zone associated with felsic igneous bodies. Howe and Thompson (1984) propose that it is an uplifted, faulted zone of former growth faults that were reactivated in a compressional stress regime.
- 14. Immediately to the north of the disturbed zone, contemporary seismicity strikes approximately 330 , in contrast to the disturbed zone, which strikes 050 . Zoback et al. (1980), using seismic reflection data, related seismic-ity in the area to small faults and igneous plutons of various ages.
- 15. Based on geophysical data, the New Madrid research indicates that the Anna, Ohio seismic zone is not an extension of the New Madrid Rift Complex. The Anna seismicity is strongly localized on and around two geophysical anomalies. Whether they are the cause of the seismicity or just control its pattern is not known.
- 16. The New Madrid Rift Complex can be bounded by geological / geophysical data used in conjunction with historic seismicity. This is significant, for it means that a tectonic structure can be delineated and that seismicity in the area can be linked to that tectonic structure.
RECOMMENDATIONS
- 1. The New Madrid Rift Complex should be treated as a major tectonic feature within the North American Continental Plate.
- 2. The New Madrid Rift Complex should be treated as a seismically active area with the potential for earthquakes up to Modified Mercalli Intensity XII.
- 3. Seismic design criteria should assume a 600 year recurrence interval for earthquakes large enough to liquefy sand in the alluvium of the New Madrid area.
Harold Denton, Director 13 JUN 3 61986
- 4. The following generalizations should be considered in drawing seismic hazards maps for sites in or near the New Madrid area.
- a. The zone of highest seismicity is located within the New Madrid Rift Complex and on the flanks of mafic plutons situated on its margins.
- b. The Reelfoot Rift is seismically the most active area. Within the last 2000 years it has experienced at least three intensity XI to XII events.
- c. The other three arms of the New Madrid Rift Complex, i.e., the St. Louis, Indiana, and Rough Creek, should be considered seismically active but to a lesser degree than the Reelfoot Rift.
- d. The seismicity of the New Madrid Rift Complex is distinct from that of the Anna, Ohio area. Based on geophysical evidence, the. Indiana Arm of the New Madrid Complex appears not to extend north of latitude N39.5 .
kek1 t$ L& Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosure:
List of Selected References l
i i 4 SELECTED REFERENCES NUREG and NUREG/CR Reports Ault, C.H., D. Harper, C.R. Smith, and M.A. Wright, 1985, Faulting and Jointing in and near Surface Mines in Southwestern Indiana, NUREG/CR-4117, 27 pp. Braile, L.W. ,.W.J. Hinze and J. L. Sexton,1979, An Integrated Geophysical and Geological Study of the Tectonic Framework of the 38th Parallel Lineament in the Vicinity of Its Intersection with the Extension of the New Madrid Fault Zone, NUREG/CR-1014, 191 pp. Braile, L.W., W.J. Hinze, J.L. Sexton, G.R. Keller,'and'E.G- Lidiak, 1981, An . Integrated Geophysical-and Geological Study of the.Te~ctonic Framework'of.the. 38th Parallel Lineament in the Vicinity of Its Intersection with the Extension of the New Madrid Fault Zone, NUREG/CR-1878, 131 pp. Braile, L.W., W.J. Hinze, J.L. Sexton, G.R. Keller, and E.G. Lidick, 1982, A
; - Tectonic Study of the Extension of the New Madrid Fault Zone Near Its Intersection with the 38th Parallel Lineament, NUREG/CR-2741, 70 pp.
Buschbach, T.C., 1977, New Madrid Seismotectonic Study - Activities During
, Fiscal Year 1977, NUREG-0379, 61 pp.
Buschbach, T.C., 1978, New Madrid Seismotectonic Study - Activities During Fiscal Year 1978, NUREG/CR-0450, 129 pp. Buschbach, T.C., 1980, New Madrid Seismotectonic Study - Activities During Fiscal Year 1979, NUREG/CR-0977, 149 pp.
, Buschbach, T.C., 1984, New Madrid Seismotectonic Study - Activities during Fiscal Year 1982, NUREG/CR-3768, 166 pp.
Buschbach, T.C., 1985, New Madrid Seismotectonic Study - Activities During Fiscal Year 1983, NUREG/CR-4226, 137 pp.
. Buschbach, T.C., 1986, New Madrid Seismotectonic Program-Final Report, NUREG/CR-4632' 62 pp. ,
~
l Hinze, W.J. , L.W. Braile, J.L. Sexton, G.R. Keller, and E.G. Lidiak,1983, Geophysical-Geological Studies of Possible Extension of the New Madrid Fault 2 Zone, Annual Report for 1982, NUREG/CR-3174, Vol. 1, 87 pp. Hinze, W.J. , L.W. Braile, J. L. Sexton, G.R. Keller, and E.G. Lidiak,1985, Geophysical-Geological Studies of Possible Extension of the New Madrid Fault
> Zone, Annual Report for 1983, NUREG/CR-3174, Vol. 2, 46 pp.
w w - . , - -,.v.,-w ew - - - *- ~----w---v-i= ----v-- w- - v---4-- -' - -rWr -- " - " ' * -=- -
Johnson, R.W., C. Haygood, T.G. Hildenbrand, W.J. Hinze, P.M. Kunselman, 1980, Aeromagnetic Map of the East-Central Midcontinent of the United States, NUREG/CR-1662, 12 pp. Keller, G.R., D.R. Russell, W.J. Hinze, J.E. Reed, and P.J. Geraci, 1980, Bouguer Gravity Anomaly Map of the East-Central Midcontinent of the United States, NUREG/CR-1663, 12 pp. Kidd, J.T.1980, General Geology, Geophysics, and Seismicity of Northwest Alabama, NUREG/CR-1519, 79 pp. Kolata, D.R., J.D. Treworgy, and J.M. Masters, 1981, Structural Framework of the Mississippi Embayment of Southern Illinois, NUREG/CR-1877, 72 pp. Nelson, W.J., and D.K. Lumm, 1984 Structural Geology of Southeastern Illinois and Vicinity NUREG/CR-4036, 127 pp. Nelson, W.G. and D.K. Lumm, 1985, Ste. Genevieve Fault Zone, Missouri and Illinois, NUREG/CR-4333, 94 pp. Schwalb, H.R. ,1982, Paleozoic Geology of the New Madrid Area, NUREG/CR-2909, 61 pp. Stearns, R.G., 1979, Recent Vertical Movement of the Land Surface in the Lake County Uplift and Reelfoot Lake Basin Areas, Tennessee, Missouri, and Kentucky, NUREG/CR-0874, 37 pp. Stearns , R.G. , S. K. Towe, V. L. Hagee, S.J. Nova, and S. L. Wilson,1984, Description and Significance of the Gravity Field in the Reelfoot Lake Region of Northwest Tennessee, NUREG/CR-3769, 39 pp. Stearns, R.G. ,1980, Monoclinal Structure and Shallow Faulting of the Reelfoot Scarp as Estimated from Drill Holes with Variable Spacings, NUREG/CR-1501, 37 pp. Stearns, R.G. ,1981, Influence of Shallow Structure, and a Clay-Filled Mississippi River Channel on Details of the Gravity Field at the Reelfoot Scarp, Lake County, Tennessee, NUREG/CR-2130, 61 pp. Van Arsdale, R.B. , and R.E. Seargeant,1986, Post-Pliocene Displacement on Faults within the Kentucky River Fault System of East-Central Kentucky, NUREG/ CR-4685, 31 pp. Other selected references Braile, L.W., G.R. Keller, W.J. Hinze, and E.G. Lidiak, 1982, An Ancient Rift Complex and its Relation to Contemporary Seismicity in the New Madrid Seismic Zone, Tectonics, Vol.1, pp 225-237. Crone, A.J., F.A. McKeown, S.T. Harding, R.M. Hamilton, D.P. Russ and M.D. Zoback, 1985, Structure of the New Madrid Seismic Jource Zone in Southeastern Missouri and Northeastern Arkansas, Geology, Vol. 13, pp. 547-550.
'2
\/
s Ervin, C.P. , and L.D. McGinnis,1975, Reelfoot Rift: Reactivated Precursor to the Mississippi Embayment, Bulletin of the Geological Society of America, Vol. 86, pp. 1287-1295. Fuller, M.L., 1912, The New Madrid Earthquake, U.S. Geological Survey Bulletin, 494, 118 pp. Ginzburg, A. , W.D. Mooney, A.W. Walter, W.J. Luther, and J.H. Healty,1983, Deep Structure of Northern Mississippi Embayment, Bulletin of the American Association of Petroleum Geologists, Vol. 67, pp. 2031-2046. Gori, P.L., and W.W. Hays, editors, 1984, Proceedings of the Symposium on the New Madrid Seismic Zone, U.S. Geological Survey Open File Report 84-770, 468 pp. Hadley, J.B., and J.F. Devine, 1974, Seismotectonic Map of the Eastern United States, U.S. Geological Survey Miscellaneous Field Studies Map MF 620. Heigold, P.C., 1976, An Aeromagnatic Survey of Southwestern Illinois, Illinois State Geological Survey Circular 495, 28 pp. - Herrman, R.B. and J.A. Canas, 1978, Focal mechanisms studies in the New' Madrid seismic zone, Bulletin of the Seismological Society of America, Vol. 68, pp. 1095-1102. Hildenbrand, T.G., M.F. Kane, and J.D. Hendricks, 1982, Magnetic basement in the upper Mississippi embayment--A preliminary report, in F.A. McKeown and L.C. Pakiser, eds. , Investigations of the New Madrid, Missouri earthquake region: U.S. Geological Survey Professional Paper 1236, chap. E, p. 39-53. Howe, J.R., and T.L. Thompson, 1984, Tectonics, Sedimentation, and Hydrocarbon Potential of the Reelfoot Rift, Oil and Gas Journal, November 12, 1984, pp. 179-189. Kane, M.F. , T.G. Hildenbrand, and J.D. Hendricks,1981, Model for the Tectonic Evolution of the Mississippi Embayment and its Contemporary Seismicity, Geology, Vol. 9, pp. 563-568. Keller, G.R., E.G. Lidiak, W.J. Hinze, and L.W. Braile, 1983, The Role of Rifting in the Tectonic Development of the Midcontinent U.S.A. Tectonophysics, Vol. 94, pp. 391-412. McGinnis, L.D., P.C. Heigold, C.P. Ervin, and M. Heidari, 1976, The Gravity Field and Tectonics of Illinois, Illinois State Geological Survey Circular 494, 28 pp. McKeown, F.A., and L.C. Pakiser ed., 1982, Investigations of the New Madrid, Missouri, Earthquake Region, U.S. Geological Survey Professional Paper 1236, 201 pp. Mooney, W.D., M.C. Andrews, A. Ginzberg, D.A. Peters, and R.M. Hamilton, 1983, Crustal Structures of the Northern Mississippi Embayment and a Comparison with other Continental Rift Zones, Tectonophysics, Vol. 94, pp. 327-348. 3
v (
- Nuttli, 0.W. ,1982, Damaging Earthquakes of the Central Mississippi River Valley, in the U.S. Geological Survey Professional Paper 1236, pp. 15-30.
J. Pennick, 1976, The New Madrid Earthquakes of 1811-1822, University of Missouri Press, 181 pp. Potter, P.E., 1978, Significance and origin of Big Rivers, J. Geol., Vol. 86, pp. 13-33. Russ, D.P., 1979, Late Holocene Faulting and Earthquake Recurrence in the Reelfoot Lake Area, Northwestern, Tennessee, Bulletin of the Geological Society of America, Vol. 90, pp. 1013-1018. Schilt, F.S. and R.E. Reilinger,1981, Evidence for Contemporary Vertical Fault Displacement from Precise Leveling Data Near the New Madrid Seismic Zone, Western Kentucky, Bulletin of the Seismological Society of America, Vol. 71, pp. 1933-1942.
. Sexton,gJ.L., L.W. Braile, W.J. Hinze, and M.J. Campbell, 1985, Seismic reflec-tion . profiling studies of a buried Precambrian rift beneath the Wabash Valley Fault Zone, Geophysics.
Trace, R.D. and D.H. Amos, 1984, Stratigraphy and Structure of the Western Kentucky Fluorspan District, U.S. Geological Survey Professional Paper 1151-D, 41 pp. Wollard, G.P., 1958, Areas of tectonic activity in the United States as indicated by earthquake epicenters, Transactions of the American Geophysical Union, Vol. 39, pp. 1135-1150. Zartman, R.E. ,1977, Geochronology of Some Alkali Rock Provences in Eastern and Central United States, Annual Review of Earth and Planetary Sciences, Vol. 5, pp. 257-286. Zoback, M.D., 1979, Recurrent Faulting in the Vicinity of Reelfoot Lake, North-western Tennessee, Bulletin of the Geolphysical Society of American, Vol. 90, pp. 1019-1024. Zoback, M.D., R.M. Hamilton, A.J. Crone, D.P. Russ, F.A. McKeown, and S.R. Brockman,1980, Recurrent Intraplate Tectonism in the New Madrid Seismic Zone, Science, Vol 209, pp. 971-976. 4
... m
,t ,
,,,,,,,,,,_,,, .n
. . . . . - . .m uo - . . . .
au i a =ec w m a usesm , ,,,,,,, ,,
--n,-- . , , , , ,
-- - E' 200 mi1es ue n,o
~ - ' ~ * ~ ~ ' "
F-
......,m
.mapas, sfast us,.elift (M.utlME LABI.e.EE SW9. geftet AR
'I c . .. yS .
AL 1 ==.' , /
.. . j Figure 1. The New Madrid Area, showing State Geological Surveys, Universities, and @,
Federal Agencies cooperating in the New 88' p, 97 90s ,
/
Madrid Seismotectonic Research Program. i i i gi ' (!'a'.4 5
., j*"
(NUREG/CR 4226) 47 _ IOWA
- 1 ILLINOIS !
- i c40 TAN A 40' -
s J.
'~1 MISSOURI 38' -
\ '
'k-a 'J
l' ( KENTUCKY Figure 2. Epicenters of damaging earthquakes ., # __ _ in the Central Mississippi River Valley. ._... ..___..m.
/ "" -
Large circles are the epicenters of the three major earthquakes of 1811-1812. Epi- 36'- / TENNESSEE l centers of the 203 damaging aftershocks are ARKANSAS / not plotted. Small circles are the epi-centers of damaging earthquakes which 7
,,,,,_,_,,[..-*
l o' so occurred subsequent to 1812. ' ' im m ~ o (modified from Nuttli,1982) M' - MISSISSIPPI ,, . ALABAMA L i i t - _1
MICHIGAN RIVER SYSTEM N (PR,E-ctA Crat) - ILLINOIS INDIANA 4 ' OHIO f
- E 4@ d2 y 0 OUruNe r*. a g . .v./
OF BURIED \. f* * . h / . **
- V RIFT COMPLEX (* .*.f. . .*ag._ ..'.~. Q
' MISSOURI ). **
KENTUCKY ARKANSAS . q.% TENNESSEE REGION L Jk$$.*'$%h ' EMBAYM T COMPRESSIVE
'c,,
-- ;- .,. v, - .
6 ," , d ,, 'c 3'
< *[,d JJ 6 ,4,*'^2 ,
,,(*>
, A NCIENT y g y'gg'go'" < n <* *' ' , 4 s'"'s*'.
FA UL TS - '~ * '* ns'e
<,*, *4
*L, e~ '^<,
a -RIFT * ' > U'PPER CRUS T>^' a w '. *
- c) '*[
4
, . , ' , ><,,,c '*
,<,.>,3 EARTHQUAKES ' < ^ q'.>
' *s'v v ,
v ;s ' *. , > fhfbYi$ $ ' ' ,e' ' > < a^ ' s > _ LOWER CRUST (Braile,1984) Figure 3 a. Block diagram illustrating the present configuration of the buried New Madrid Rift Complex. The structurally controlled rivers, Paleozoic rocks in cratonic sedimentary basins, and the Mississippi Embayment, all associated with the buried rift complex, are also shown. Dark areas indicate intrusions near the edge of the buried rift. An uplifted and possibly anomalously dense lower crust is suggested as the cause of the linear positive gravity ano=aly associated with the upper Mississippi Embayment. ladene Arm *$ { St. Louis Arm - Reelfoot Rift '\ s Rough f[ cr k .+ Groben e
.._ N<*,,,.#
ced Oc'*"' 1 Figure 3 b. The New Madrid Rift Complex is subdivided into.four arms i
i A. D.
,800' So e 20* [a * ' * * *6 * * ' 9 8 ,L *L BI N O I S' . , ,' < '/ [.,(
b.
+
- g,p>4o%
.'. ; ; . 's fc: .,,% - ' *, : :,,,' :g...t ;
A SIN
. ,' 4-,g.,
V ' -' ./: 1-
\ *',' 's,,** *.* *2 'REELFOO 7 # 8ASIN f , ' ['/j. -~
x x . . , 4* ',, . ', ~ .I f 'c
~ -, . ,
Q f % - elh; '
- j. ,' , , '
- t
.s.:6.9. % %.,..' .t- n . ', =
g@ ,
+3 O' t30
$ $ :M 3 O*. s,'
Me r ,'[g g ,l k e*** O' . > , ' LATE PRECAMBRIAN - INCIPIE T R,lFTING, ,pf #.4,aurarc ,a;.9 ,a ' . , ,. FORbATION OF TRIPLE JUNCTI LATE PALEOZOIC - CONTIN N T A,(.',I. ,a **
,,j AND AULACOGENS. S) O C B A SI N S. 'l * '.? a "' 7 - '.
B. E.
. [. ' ' . [..e e.. 5'-NiFT."$h'/
C0MPLEX, N' [4~ ,
^
f ', e' l[ 5 ,
<90 c ,,e.... -
'/',',v-gg.g ,, :- ., . ca ,
! ' g -'*:*
;y ~<:'.
~.-
- yen %g , s k'
h~ "(% ' O h r h$$h.hnb (30 Y_, . Q , 4&.NN P, u w p 40 ' '- 'v
,, -rfT"T +
NLAT PRECAMBRIAN - EARLY CAMBRIAN- MESOZOIC-RIFTING AT ATL ANTIC M ARGIN, 30*yw CONTINENTAL SEP AR ATION, h IGNEOUS INTRUSIONS AND REACTIVATION FAILED INTRACONTINENTAL RIFTS. OF F AULTS IN AUL ACOGENS. \/ C. F.
>w * ',15 v ~ c *
* " +*.
p8 . ] 1 * . T,, * +* c' ', ',*','"'1**[
.,t '
* * . . CON TIN ENT A L[ J
/
'>5 ,w g* ' , *C OLLISION T O e -
> ", e 2 ,",c,.,,,O, THE NORTH i\
- ON,' av d $ **# * ~~
. ' , .'l ,* w , ,' k ,* E a' ', .'
=
,e'"h</
,a{*'d '
,'**,' ;.1 9 < [> , ,? . N ,, , , )'
** a .*(*/ *
- O s. <a *
*/ ,,S '
'g ~ '. MISSISSIPPI
. . , $ M lo # \p , 8'*'"'
d.N. [ if ~ N cfr ; D 9 n Ih ,g., b .
.E ARLY TO MIDDLE P ALEOZOIC- TACONIC CRETACEOUS TO PRESENT EMBAYMENT OROGENY , SEDIMENTATION IN \\ AND . COASTAL SUBSIDENCE. COMPRESSION AL CR ATONIC B ASINS. \/ STRESS RE ACTIVATES FAULTS.
Figure 4. Schematic diagrams illustrating the plate reconstruction of the North American craton and interactions with adjacent plates and geologic activity of the New Madrid Rift Complex since the late Precambrian. The outline of the State of Missouri is shown for location and approximate scale. (Brail e,1984) .
i A. LAre rRecAuBRisu: unCsPsenT Rsersna
- 3 - v , v Y r i f ' ,w v%
<.,>a** 5 4
>** * ' EXTENSION dv CRUSTA , f,4 6 **
Epsm:* ^ f ' #em:qqD .
\NMAgQ
\ UPPER. s B. LATE PRECA'Ef8RIAN*eARLY A CAuBRIAN:
a y , RirTING A 4 V q a & Y4 ) P'
> s n4 * , ;,- ,,; e,o e
- e, jT..** .
.' ~
T :x~ -' ,h',[T,/j,,"g,-:z.,.';,,
~v -:;;.
h b C, D. pateozoic: SuBSIOence
- <,a6, y
v2 v .,
-o
,-g- u y v a
v
., c . - 6 COMPRESSION .
1~ t'~'td2t6E?{lTO~{2E;5,;T{;T(;T"L
'tx~x~x~ m - .~. , ~k: -
N .s, h E. uesozoic: uPtirT. REACTIVATION v o' >s < -;<3.;._ f,.p-
. 4
, u ,Z n ,"
" ra< , ,
1 4a
, , v ., c-
.vw A#, y '*(;
'II)T~ ';T(2,~{ '*( 9,-') ~
a yy -
. F. tare ussozoic-pReseur: SueSsOence.
pississippi aivta
e V
a y e.Rs . .c . , 4 A a 4w&
. , COMPRESSION
$$hy4kobm M' kNNNNNNxTJERNN1
,, (Braile,1984)
Fioure 5. Schematic cross sections for a northwest-southeast profile through the New Madrid Rift Complex illustrating the evolution.of the rift complex and associated cratonic basins through time. Stages of development shown in parts A through F are related approximately to the map views illustrated in the corresponding diagrams in Figure 4. l I i
4 ILLINOIS INDIANA
\
\
- N g
/ y l /*
MISSOURI 8-
\
i
, 52 \
. / l l
s . / J 'f
~ s, /t- t p -s -
-{
g I g % , KENTUCKY 37* -
. I l g, l, -s+
4 64 i TENNESSEE j [ ARKANSAS , a 4 ss' - g __._ . g ._. . -+ 7-(4, d f ALAsAMA y
, z. so. se* as*
I....t l o too tooke Figure 6. Map showing the contours of the number of' earthquake epicenters of4mod {fied Mercalli intensity greater than III from 1800 to 1972 per 10 km for the New Madrid area (after Hadley and Devine,1974). Heavy dashed lines are inferred boundaries of the rift complex as shown in Figure 3. (Braile et al ., NUREG/CR 2741). M t ? 9 e
1 j - w a - "
,l -
.t. i. -
,n.
,.Q*"
* - J u.os .
J n - '
% ~s E) n<*
Y l -
. (\~'
3 \- w
- to -
' / ,
O .P. ' ., 0'
/
(i , b .st t,.f?s*
% .2. Q , J4 t , ,,.
w ,( \.. m , ,.] gv
/-
c/ p _ y c-@1gy -
,, .;.g ;
~ ~' ,c
'Q% r s ,
-)
v-(.x
...,T" e
"#V. -
r ' a, .. l
$ ps *
- :~
_ _ _ _ _ _ = _
.. .. m e 3. .,, . L .. .
. w
,6* .
- Fa. .. - *
/
h . .. **fe r .=c, . _ _ . . . . . . . .
. 'NUREG/CR 3174) e . -i ~.: .
=,_ ,
--/ ~>-
-p r / muusas c mar , _ 3
- m. . .-
~
w w a Figure 7. Earthquake epicenters, tectonic features, and inferred boundaries of the rift complex in the New Madrid area. The shaded regions are areas of positive gravity and magnetic anomalies which have been used to infer the - boundaries of the rift complex. (NUREG/CR 3174). 1
O O O O o 40 c. St. Louis, Missouri E*O A e N% O e *o
.e g. .
% c o o :
Dyersburg, Tennessee o o c' Marked Tree, Arkansas
, @ 6g
$ e o'
O Figure 8 a . w.tr .r urutum. 4.t..t.4 4 1.us.4 = tas a. 4 ' regional network for the report 1ag period 1976 = 1982. (Hermann,1983) ,
= w a go 0 *
(D e ho e e" O CI g e o e g o New Madrid, Missouri _ , g8 o D"-
- b o 8. e o 0- 'g , ,e 4*cs 4 o Caruthersville, Missouri p p- ..o o
O %p ,, O e c
- Dyersburg, Tennessee O
*o ,~
/ t o
Marked Tree, Arkansas -o O oo bc G N l Figure 8 b. region.1 r a ts .c ru e=m u 4.t.et.4 4 1 t.4 ut: a. e.=. metwork ror the reporttag perica 1976 a 1982 1. La* to.41.t. ,s.aity .c s m. arse, nsu s. (Hemann,1983) i i v' .. _ --
l
=
e lilinois Indiana Ohka r-O e p
*o RE
*' FAIRFIELD (*g c - BASIN ?,. .v;nc.nn., -
- t. l.ouit *6 Waterloo Dup
, SPA RTA 5* 7 V""*a f
c SHELF $
~
c 3 [
'V 0p l
Palmet
% Rough.c;,Q.Shawn ,!
Nou n O*O fgW
- - ~~ r Eagtey,,,,y OZARK ,. ----
'*8n Svacline 4, .
3 DOME
,e-p' ~~ Fluorspar s,
-/ ,,.
Pennyrile FaultSO Poplar Blufff t .Mg','4 Area Fault s y;,,,, ,' . ,;d Complex 's, k'ntucky
' ~~-
,,,n,.,, 4,, 4 i.t e Nashville f 4,,
8
) !
's NASHVILLE Iault; tic o't I downthrown side DOME 5,
rnonociine ; x -"- \ . . . (From Nelson and Lumm,1985, NUREG/CR 4333) Figure 9. Principal tectonic features in the New Madrid area.
ym p* g'
.+j+g%w.,di / *T
- ';%, f. .,
.. .~
_ q- -- gy e
. , f .
[ M 'h,,-%' $* *
""". f; 6; . i. / ; , if$i@ (/
N : - e-4 . p ' N
/ $- ;e..4, . .n .
mm, .
.l_',.
- if ..,' *:! f
~
*).' -u. ... .<,
jww. * ...s .s
.p g,,_,1 h?
fkl$:<. s-
/ ~*
.fi:
fyv. t D .: yl h( p
- _.^,
r.i .e. si; .- y ,* . . v. ..# ..
. [aMED *
- w so, ..
/
[N I ' 4s*
. . . ;;x.,s.
, . . . -- p ww, .w .- ;f.s.. i,
* * * = = . -
w
,,/ / i- v- are s.cse. o .. s c
\) :: '
! .v . mt'n nr.y~'o
.. i- *g;~f
,... go;n g.,,
't . . ~ . . . . .
, ( ,C ::' A fx'c:P, (Crone et al .,1985)
Ficpre 10 a.preti-iaers map sho-tas relatieaship or seis icity to disturned tone mapped at the depth of magnetic basement in the southern part of the New Madrid seis ic zone. Epicenters are for she pertod June 29. 1974 through March 78. 1981 f ro R. B. Herra. san (written cown.. 1981). om. .
, .o.o
.~w.---
5:.M, s. .~ :+.iw
.--.a,
. ': .:=-
. ; .?,.
;4*e",......%.,+.,.
m+ > .- =<=:: E bi$$FNkN' 'Nc" : . -$. . o.s fs gos wr- - ,.. ---'d-Y c;- f 2- _ -- _ - - - - - -_-- . W { io.; ;f .h& .4.N-C-uM . Q .$ '. M .*f.?-~: F D r.t-Wm
~- E
.i. -& .i.e.. _
i
- Q-k %:4 Y:i - - ;i&&s,;i-tg.%%e'-%r:S= @M. - @.M%'-W.' i 8 Q Figure 10b. Seismic reflection profile N :ssatsFar g *-a s .[*
$ is.F@%)f!- c 9 : 'G4rg.2.F. and interpretive line drawing showing E%@%M.-[;tQ'3T %y=:r@M;$-T3M
- 5C .~,~.c. ; details of one edge of the disturbed
- - - .co. zone shown in Figure 10a. Note the structural divergence between the magnetic
, l , ,
O.o -
$*" " i
- 0.0 basement (MB) and the slightly warped top of the Paleozoic (PZ). Specific location nf these data cannot be identified 5... , _, . .. 3 because they are proprietary.
l .. --... -n.. . .. y, - I - , - - --- E legend: K = Cretaceous E '8-
.e4
< PZ= 'Ibp of Paleozoic sedinents 3: ! BC= Base of marine carbonates t
T .
.f.-
. ., Q; .C*
.-g=c -- *
{. MB= Magnetic basement o is - , . .: ,, - _ _ - _ 3.. ; E . . (Crone et al . ,1985) -- l 1 l l
e
- f. , .
90 'q q (s' 89
., h 88 37 9; go '
e. ( o' 89 ..) Mg s './ 3 eLOW 'LO
/
p ,~ 'tLD ) y
, ":::.bw s
~s .
3 - ,,,.
- ==-== ~ g
=
o 4 . % s ' -- %%, g
,g
'* , s' *
,' ; '^
,1,II *, lk a }~
-e4k , 4 ' 'p , ,* . * * .*3, I
.- . _ _ KY__g MO,_ ,j I
,, at x , ,i *
, ,' i TENN ARK i #
- J # j
',,' INSIDE
*p' ' N ~. ,
N 3 / o . s , e
.r. .- ', I
?? gt1 s N '
f I f f yt i
- 2
't. #
A j t9
< ( UTalDE
.. __i- H ' ' . o i
j
,i I.' ' I #
, k pt
=
a 0 o
*f Y a , J.
- O )
- A Y '
/
d6 v -. o so us. 35 29 v -- *- -- -- - o s.o us. ! 9 so nu 9 s.onu Figure 11. Temperature gradients in Figure 12. Temperature gradients in relation to the edge of the Reelfoot relation to the edge of the Reelfoot l Rift and location of earthquakes. Contours Rift and buried plutons. (Modified from l are in degrees c/km. Note the correspond- NUREG/CR 4226). ance of contours with the Dyersburg Line and earthquakes. ~(Modified from . NUREG/CR4226)
MAJOR GEOCHRONOLO'GIC AND CHRONOSTRATIGRAPHIC UNITS Age estimatesU el Subdivisions in use by the U. S. Geological Sunroy boundaries in land their map symbolsi million years (m.v.I Quaternary Holocene Epoch of Ser'es , Pered or System 0 010 101 Pleistocene Epoch or Series 2 111 2.21 - Neogene Cenotoic Piecene Epoch or Series Subperiod or 5 14.9 5.31 - Eraor yerinary Subs tem g ,gg Erathem Pered or 24 I23 20I ~ 10r1 Paleogene Ohgocene Epoch or Series System ubpered or 38 134 381 - ITI Eocene Epoch or Series Subsystem 55 154 561 - IPal Paleocene Epoch of Series 63 163 661 - taie Cretaceous Epoch or Upper Cretaceous Senes y,3nga,g Cretaceous Pered 96 195 971 - Phaneratoic or System IKI Early Cretaceous Epoc'h or g, ,, . Eraor lower Cretaceous Senes (rathem . 138 1135 1411 - Eonothem Jurassic Pered or System IJi 205 1200 2151 - inassic Penod or System th I
~240 -
Permian Pered or System IPI Pennsylvanian Pened or i9036 - Carbonderous System IPl Pereds or ~330 - Paleoroic Systems (CI M'55'55'8D'an Pered or frao, 360 1360 3651 - Devonian Pened or System IDI IPrl Sdunan Pened or System ISI 435 1435 4401 - Vrdovecian Pered or System 101 500 1495-5101 - Cambnan Pered or System IEl
~ 570D -
Protercioic Z lis'J Proterciot 800 - ! Eon or Protercioic Y (YlD Eonothem les 1.600 Protercioic X IXID 2.500 Archean Een or Eonothem lAl Oldest known rocks in U S 3.600 - U Ranges reflect uncertaenties of isotopic and bestratigraphic age assignments Age of boundaries not closely bracketed by esisting data shown by ~. Oecay cons: ants and rsotope rates employed are cited in Steiger and Jager (19771. D Rocks older than 570 m y also called Precambnan IpCl, a time term withou' specific rank U inme terms wahout specific rank. Geologic Names Committee.1980 editen Figure 13. Geologic Time Scale. (U. S. Geological Survey)
o . Harold Denton, Director 13 yy 301986
- 4. The following generalizations should be considered in drawing seismic hazards maps for sites in or near the New Madrid area.
- a. The zone of highest seismicity is located within the New Madrid Rift Complex and on the flanks of mafic plutons situated on its margins.
- b. The Reelfoot Rift is scismically the most active area. Within the last 2000 years it has experienced at least three intensity XI to XII events.
- c. The other three arms of the New Madrid Rift Complex, i.e., the St. Louis, Indiana, and Rough Creek, should be considered seismically active but to a lesser degree than the Reelfoot Rift.
- d. The seismicity of the New Madrid Rift Complex is distinct from that of the Anna, Ohio area. Based on geophysical evidence, the Indiana '
Arm of the New Madrid Complex appears not to extend north of latitude N39.5 . signet tiy: ROBERT 3 XIg)gug RobertB.Minogue,Dbector Office of Nuclear Regulatory Research l -~ I m riun
Enclosure:
List of Selected References .f Cabj ect Fils No. f; i /1 ?l( v Distribution /R-2811: $-d If Alb Circ /Chron EConti
,DCS/PDR LBeratan Mk No.
ESB Sbj/Rd AMurphy Ecserch Request No. RBMinogue D/N: Memo Denton/146 N No. ,, Dross E0'Donnell N No. KGoller D eket Tc. l Ruleunking No. _ _ Other , Keturn IGC-318 to ICS, YeQNo
- Note: See. Previous Concurrences -
0FC:ESB:RES:jk :ESB:RES :ESB:RES :DRPES:DD :DRPES:D :RESDD :RE ___________________________________________________________________ 3 - _ NAME:E0'Donnell?AMurphy *:LBeratan*:EConti* :KGoller* : Dross * :R h nogue DATE:6/ /86 :6/ /86 :6/ /86 :6/ /86 :6/ /86 :6/ /85 :6/30/86 0FFICIAL RECORD COPY
O Harold Denton, Director 12
- 4. The following generalizations should be considered in drawing seismic hazards maps for sites in or near the New Madrid area,
- a. The zone of highest seismicity is located within the New Madrid Rift Complex and on the flanks of mafic plutons situated on its margins,
- b. The Reelfoot Rift is seismically the most active area. Within the last 2000 years it has experienced at least three intensity XI to XII events..
- c. The other three arms of the New Madrid Rift Complex, i.e. the St.
Louis, Indiana, and Rough Creek, should be considered seismically active but to a lesser degree than the Reelfoot Rift.
- d. The: seismicity of the New Madrid Rift Complex is distinct from .that. of the Anna, Ohio area. Based on geophysical evidence, the Indiana Arm of the New Madrid Complex appears not to extend north of latitude N39.5 .
Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosure:
List of Selected References. Distribution /R-2811: Circ /Chron RMinogue EConti D/N: Memo Denton/146 DCS/PDR Dross LBeratan ESB Sbj/Rd KGoller AMurphy E0'Donnell
- DD DR S
~
ESB:RES ESB:R S ESB:RES DR .R S DD. RES:D
'E0'Donnell AMurphy. LBeratan 'EC ti. KGoller . Drys- RMinogue-05/r//86 05/t.7/86 ~ 05 786p05/'/ 05/A7/86 '0 4 y86 05/ ./86 l}}