The basement structure, tectonic history, seismicity, and seismic hazard potential of the Floridan Plateau region

MISSING IMAGE

Material Information

Title:
The basement structure, tectonic history, seismicity, and seismic hazard potential of the Floridan Plateau region
Physical Description:
x, 221 leaves : ill. ; 29 cm.
Language:
English
Creator:
Lord, Kenneth M
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1993.
Bibliography:
Includes bibliographical references (leaves 194-219).
Statement of Responsibility:
by Kenneth M. Lord.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001933253
notis - AKA9320
oclc - 30784199
System ID:
AA00003247:00001

Full Text









THE BASEMENT STRUCTURE, TECTONIC HISTORY,
SEISMICITY, AND SEISMIC HAZARD POTENTIAL
OF THE FLORIDAN PLATEAU REGION














By

KENNETH M. LORD


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY












ACKNOWLEDGMENTS

This project was generously funded by the Florida Power and


Light Corporation.


I am indebted to John Luke, Jerry Burford, Russ


Gouldy, Orin Pearson, and David Chaney, all of F.P. & L,


continuing support and enthusiasm.


for their


Additional support was given


by Walt Schmidt and the Florida Geological Survey


who provided the


drill rig necessary for the installation of the seismograph bases.


Over the past years,


Doug Smith's enthusiasm, professionalism,


and friendship have made all aspects of my graduate studies both


rewarding and enjoyable.


In addition, numerous other people


provided advice, assistance, and encouragement to me during the


course of this project.


Tony Randazzo, Paul Mueller, Neil Opdyke,


Tom Crisman, and Ann Heatherington have each provided me with


valuable guidance and helpful suggestions.


Ray


Thomas spent many


long hours helping me to debug various digital systems.


Joe Smyth


and John Roche of the Florida Division of Recreation and Parks and
Robert Arnberger of the National Park Service each went to a great
deal of trouble to make their facilities available for this project, as
did Wes Greenman and Sam Langley of the Department of
Astronomy.







helped in the difficult tasks of installing and maintaining the
network of seismographs. In addition, lengthy conversations with
Jon Jee proved to be very fruitful. I am grateful to each of these


people for their time, interest, and assistance.














TABLE OF CONTENTS


Page


ACKNOWLEDGMENTS

LIST OF FIGURES ....

ABSTRACT ........


CHAPTERS


INTRODUCTION


Purpose . . . . . . . . . 1
Methodology .................................. 4


THE STRUCTURE OF THE FLORIDAN PLATEAU


Introduction . . . . . . . . .
Methods of Investigation . . . . . ...
Results and Interpretations. . . . . . ...
Summary ..... . . . . . .


TECTONIC EVOLUTION OF THE FLORIDAN
PLATEAU ........ . .... . .. . . 64

Introduction ................ ......... .. .. ... 64
Regional Tectonic History ......................... 64


SEISMICITY AND STRESS IN THE
SOUTHEASTERN UNITED STATES ...............80


Introduction. ........
Methods ............
T?^Clrt-


. . . . . . . 80
. . . . . . . 81
51








5 REGIONAL SEISMOTECTONIC PROVINCES........... 104

Introduction. .. . . .. .. .. ..104
Background and Methods ........................105
Results . . . . . . . . . 111

6 SEISMIC ATTENUATION IN THE
FLORIDA REGION ............................ 123

Introduction . . .. .. .. .. . . . .. .. 123
Background and Methods.........................124
Test Cases for Intensity-Absorption Method ....... .128
Results from the Florida Region ..................133

7 SEISMIC HAZARD IN FLORIDA. . . . . .139

Introduction... ............... .................139
Methods . . . . . . . . .139
Results . . . . . . . . . . .143

8 DISCUSSION AND CONCLUSIONS ...................148

Introduction ................................... 148
Discussion . . . . . . . . . 148
Conclusions.. .. .. ... ..... ..... ...... .. ... 158

APPENDIX SEISMIC HAZARD PROBABILITIES FOR SITES
IN FLORID A .. ... .. .. ... .. .... 161

REFERENCES CITED ................................... 194

BIOGRAPHICAL SKETCH............................... 220














LIST OF FIGURES


Drill hole locations and basement lithologies within the
study area. . . . . . . . .. .a. .


Geophysical survey locations within the study area ......... 11


Depth-to-basement of the Floridan Plateau.


. . . . .. 12


4. Bouguer anomaly map of the Floridan Plateau region.


. . .13


Magnetic anomaly map of the Floridan Plateau region........16

Bouguer anomaly field of the Floridan Plateau downward-


continued to basement level........


. .17


. Bouguer anomaly field of the Floridan Plateau upward-


continued to 10 km.


. . . .18


Second vertical derivative of the Bouguer anomaly field


of the Floridan Plateau.


. . ..19


Map of drill hole distribution and basement lithologies


with the designated features discussed in the text.


. . .24


Depth-to-basement map with the designated features
discussed in the text. . . . . . . ...


Magnetic anomaly field map with the designated features


discussed in the text.


. . . .26


- S S- a a.


Page


d a Q d q -


I









Upward-continued Bouguer anomaly map with the


designated features discussed in the text.


. . . .29


Second vertical derivative of the Bouguer anomaly field


with the designated features discussed in the text.

Various previously published configurations for the
Mesozoic structure of the southern and western


Floridan Plateau basement.


. . .30


. . . . .46


Pre-Mesozoic lithotectonic features of the Floridan


Plateau basement...


. . . . .60


Mesozoic structural features of the Floridan Plateau


basement.


. . . . .6 1


Hypothesized sequence of events during Alleghenian
closure resulting in the translocation of Gondwanan
terrane fragments along transform faults in the
Floridan Plateau basement. . . . .....


Hypothesized sequence of events during the formation
of the Late Triassic and Jurassic extensional features of


the Floridan Plateau basement.


Epicentral distribution in the southeastern United States
and northwestern Caribbean from historical and


instrumental records.


. . .72


. . . . 82


Pre-Mesozoic allochthonous terranes of the


southeastern United States.


. . . . .93


Selected Mesozoic seismotectonic features of the


southeastern United States.


. . . .95


The University of Florida Seismograph Network.


110


- a









Modified Mercalli intensity isoseismal map for the
northern Kentucky earthquake of 27 July, 1980.


. .. .. .130


. Modified Mercalli intensity isoseismal map for the


central Idaho earthquake of 28 October, 1983.


Modified Mercalli intensity isoseismal map for the 1886


Charleston earthquake.


Modified Mercalli intensity map for the Florida
earthquake of 27 October, 1973...........


137


The SEISRISK III input file for the calculation of
seismic hazard in peninsular Florida using the
format described in Bender and Perkins (1987)......... 141


The distribution of annual probabilities of exceeding


0.08 g inFlorida.


The distribution of annual probabilities of exceeding


0.02 g in Florida.


...... .. .. .. 145


. . . . . 146


..... .... 131


. . . .134












Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE BASEMENT STRUCTURE, TECTONIC HISTORY,
SEISMICITY, AND SEISMIC HAZARD POTENTIAL
OF THE FLORIDAN PLATEAU REGION

By

Kenneth M. Lord


August 1993


Chairman:


Dr. Douglas L. Smith


Major Department:


Geology


This study seeks to reduce subjectivity in probabilistic seismic
hazard assessments in Florida by providing refined characterizations
of the basement structure, tectonic history, regional seismicity, and


seismic attenuation.


Interpretations of digitally filtered gravity


maps supplemented by previously published data reveal a basement
configuration reflecting the probable translocation of fragments of
Gondwanan crust during Alleghenian continental convergence and
the formation of extensional basins along the southwestern half of


the Floridan Plateau during Mesozoic extension.


The Jay Fault zone


may have originally formed as a dextral transform during
cnn ti n Inrcrtl s -lli inn ht wra c clhchloi onlrtl; ooratr\/a1 tar







Fault.


The Jurassic Tampa Basin is larger than previously mapped


and the South Florida Basin region probably consists of 2-3 separate


northwest-trending basins.


The orientations of the Jurassic


extensional basins suggest two episodes of formation, one during the
rotation of the Yucatan block away from the Floridan Plateau and


another during the separation of North and South America.


There is


no evidence of fault reactivation or other tectonic activity on the
Floridan Plateau since the Middle Cretaceous.
This tectonic quiescence is manifested in the unusually low
level of seismic activity on the Floridan Plateau as demonstrated by a
review of regional instrumental and historical records and the
deployment of a network of digital seismographs accompanied by


four years of monitoring.


The nonuniform distribution of regional


seismic activity is probably related to the distribution and
orientations of preexisting mid-crustal faults and to their
propensities for slippage.
Seismic attenuation in Florida appears to be slightly higher


than in the central United States


western United States.


, but significantly lower than the


Although seismic hazard throughout Florida


is extremely low, the greatest probability for damaging ground
motion (> 0.2 g) exists in northwestern Florida and is calculated to be


8.0 x 10-5


/year, indicating a recurrence interval of 12,500 years.


Throughout most of Florida, however, the probability of exceeding
0.2 g is significantly less than 1.0 x 10-5/year, suggesting recurrence













CHAPTER 1
INTRODUCTION


Purpose


Large construction projects, such as dams, bridges, and nuclear


power plants,


are designed to withstand specific maximum levels and


durations of strong ground motion.


The projections of maximum


levels and durations of strong motion are made through the use of


site-specific seismic hazard analyses,


of criteria.


which are based on a number


These criteria include the levels of historical and


instrumental seismicity


the identification and characterization of


seismotectonic sources, expected maximum magnitudes of
earthquakes, and expected resultant ground motion at particular


sites (e.g.,


Yegian, 1979; Johnston, 1981


Reiter, 1990)


Because the


Floridan Plateau has a historically low level of seismic activity and a
rather enigmatic tectonic setting, little of the information needed to
assess or quantify the criteria used for the analysis of seismic hazard
is available.
The most reliable indicators of seismic hazard are the regional


levels of historical and instrumental seismicity


The paucity of


reported seismic events in the historical record suggests that the


--c_- 1


I T -_. L---- -- -A --





2

escarpment, suggests that the historical record may not provide a


reliable measure of seismicity.


In addition, because there has been


no regional network of seismographs until recently, there is little
documentation for the level of instrumental seismicity.
Consequently, neither historical nor instrumental records provide
reliable gauges of seismic activity for use in hazard analyses in
Florida.
Another potential indicator of seismic hazard is the
identification and characterization of seismotectonic sources and
source regions. Seismicity in the eastern United States is generally
attributed to the reactivation of Mesozoic or older faults in a
moderate stress field (Sykes, 1978; Hamilton, 1981; Wentworth and
Mergner-Keefer, 1983; Zoback, 1992); however, the structure of the


Floridan Plateau basement is deeply buried.


The crystalline


basement of the plateau is overlain by Coastal Plain sedimentary
deposits, which range in thickness from about 900 meters in north-
central Florida to over 5000 meters in southern Florida (Wicker and
Smith, 1978). As a result of this thick sedimentary sequence, there
are relatively few drill holes intersecting the basement (Barnett,
1975; Smith, 1982; Chowns and Williams, 1983) and regional
potential field interpretations are difficult (e.g., Oglesby et al., 1973;


Klitgord et al., 1984).


Consequently, current models for the structure


of the Florida basement are necessarily characterized only by
generalizations of large scale features and lithotectonic units, while





3

The amplitude of ground motion at a particular epicentral
distance from a seismic event is dependent not only upon the seismic


moment of the event


, but also upon individual site characteristics


and seismic attenuation. Seismic attenuation in the eastern and
central United States has been studied on a regional level (Nuttli,


1973A; Jones et al.,


1977


Bollinger, 1979; Campbell, 1981


Singh and


Herrmann, 1983), but there have been no independent studies of
attenuation in Florida. Accordingly, Nuttli's (1973A) generalized
attenuation coefficients for the eastern United States are employed in


calculations of seismic hazard in Florida.


However, the Floridan


Plateau has been shown to be an allochthonous terrane with respect


to the North American craton (e.g., Smith, 1982)


. This suggests that


the mechanical response of the plateau may differ from that of the
remainder of the eastern United States, which would invalidate the


use of Nuttli's attenuation coefficients.


As a result, estimations of


expected ground motion in Florida may be invalid.
One consequence of the limitations inherent in the historical
and instrumental records in Florida and the enigmatic nature of the
tectonics and structure of the basement is that the criteria used in
regional seismic hazard analyses (i.e., seismicity, seismotectonic


characterizations


, and seismic attenuation) are poorly documented.


Previous seismic hazard analyses in Florida have assumed, with little
substantiation, that the risk of local seismic activity is negligible and
that Nuttli's eastern United States attenuation coefficients are





4

analyses and to reexamine the tectonic history of the region.
Specifically, the intention is to provide improved characterizations of
the basement structure, tectonic evolution, regional seismicity, and
seismic attenuation in the Floridan Plateau region, and to incorporate
these into an empirical assessment of seismic hazard in Florida.


Methodoloev


Several geophysical methods have been employed and


incorporated into this study.


methods used are discussed in detail.


In each of the following chapters,


The purpose of this section is


to provide a general overview of the sequence of these individual
methods and how each has been utilized in this study.
The first step of the study was to digitize and filter potential-


field data for the entire Floridan Plateau.


The resultant filtered


gravity maps for the Plateau, when interpreted in conjunction with


previously published magnetic,


drill hole, seismic reflection


and


seismic refraction data, permitted an improved resolution of the
locations, sizes, natures, and spatial relationships between the
various crustal blocks and structural features of the plateau


basement.


A tectonic interpretation based on this improved


structural model prompted a general interpretation of the seismic
history of the plateau, and existing seismic reflection studies of the
Cenozoic sedimentary accumulations overlying the basement allowed
a more specific determination of recent levels of seismic activity.







which only encompass the past twenty years in this region.


Because


instrumental coverage in Florida was insufficient, a network of
digital seismographs was established for the purposes of attempting
to detect and characterize any previously unknown local sources of
seismic activity and to characterize the local mechanical response of


the crust to regional events.


The information compiled from these


historical and instrumental sources was then examined within the
context of previously published regional stress field indicators to
develop a lithospheric model for the distribution of earthquakes in
the southeastern United States.
Following this, each of the seismic source provinces in the
southeastern United States, the Caribbean, and the Gulf of Mexico in
which activity could potentially generate significant ground motion


in Florida were identified.


Each province was characterized based on


the compiled earthquake information, the lithospheric models for the
Floridan Plateau and the southeastern United States, and previously


published material.


Included in the characterizations were estimates


of maximum magnitude and recurrence intervals.
Subsequently, a method was developed to estimate a local


value of seismic attenuation in Florida.


This method utilizes


isoseismal maps as a means of approximating the decrease in peak
ground acceleration with distance from an event with a known or


reasonably well-approximated magnitude.


These parameters


(acceleration, distance, and magnitude) are then used to







where attenuation coefficients are well known.


Once validated, the


method was applied to the isoseismal maps from the 1886 Charleston
earthquake and a 1973 earthquake in northeastern Florida in order
to estimate attenuation coefficients for the southeastern Coastal Plain
and Florida.
Finally, the source zone characterizations and estimated
attenuation coefficients were used as input for a probabilistic seismic


hazard assessment for Florida.


This analysis is based on the method


of Cornell (1968), which addresses the effects of all of the possible
earthquakes likely to affect Florida for each site on a grid of sites


around the state.


Based on the distances to each seismic source zone,


recurrence intervals, and seismic attenuation coefficients, a value is
calculated for each site describing the probability that a particular
acceleration will be exceeded at that site within a given time frame.
The resulting grids were contoured to produce probabilistic seismic
hazard maps for Florida.












CHAPTER 2
THE STRUCTURE OF THE FLORIDAN PLATEAU


Introduction


All regional models describing the structural features of the
Floridan basement have been founded primarily on deep drill hole


data from wells penetrating pre-Cretaceous crystalline rock.


Because


of an intrinsic bias in the spatial distribution of these drill holes,
most of the regional models have been limited in scope to peninsular
Florida, rather than to the entire Floridan Plateau. Similarly, more
localized, yet more detailed, structural models based on seismic
reflection profiles tend to be spatially biased towards offshore areas.


As a result, with the notable exception of Klitgord et al. (1984)


there


has been little effort to integrate previously published studies into
an examination of the Floridan Plateau as a singlar feature.
Since the work of Klitgord et al., a number of pertinent
geochemical, geophysical, and drill hole studies have been published
(e.g., Nelson et al., 1985A; Ball et al., 1988; Heatherington et al., 1989;
Dobson, 1990; Heatherington and Mueller, 1991; McBride, 1991);


these have yet to be examined within a regional context.


The


intention in this chapter is to utilize digitally filtered regional
S,- I C 1 .. .. .. .... .-- ... 1 ..l .J. t, I .ta.a.f I...S t.,. .%C





8

Methods of Investigation


A series of digital geophysical and structure contour maps for
the Floridan Plateau were produced for this investigation using the


methods described in Lord (1992)


These maps cover the rectangular


area encompassed by 24 to 31 degrees north latitude and 80 to 86
degrees west longitude. All of the digitizing, digital filtering
operations, and contouring were done on a IBM 386 compatible,
equipped with a Houston Instruments digitizing tablet. Additional
drafting and figure preparation were done on an Apple Macintosh
IIci equipped with a scanner and laser printer. Commercial software
packages employed include Rockware's Digitize, Golden Software's
Surfer, Lotus 1-2-3, Silicon Beach Software's Superpaint, and a series
of potential-field geophysical programs available through the USGS
(Hildenbrand, 1983; Grauch, 1984; Godson and Mall, 1989).
The first step was to produce a detailed depth-to-basement


map for the plateau.


In this study area, basement is defined by the


unconformity between pre-Late Jurassic rocks and younger
sedimentary accumulations of the Coastal Plain. All available drill
hole depth-to-basement data, shown in Figure 1, were plotted on a
grid representing points with a 9.2 km spacing over the study area.


Sources for these data include Barnett (1975), Smith (1982)


and Williams (1983), Schlager et al. (1984),


Chowns


Ball et al. (1988), and


Dobson (1990).
A number of additional depth-to-basement determinations















* U


* r


"N(* .."4 +++
+
.Y "+'


Florida


Platform


* Granitoid Plutonic Rocks


* Diabase


o Felsic Extrusive Rocks
o Metamorphic Rocks
(Amphibolite Grade)


+ Quartzitic Sandstone or Shale


U Quartzitic Sandstone/Shale with Diabase
* Triassic Redbeds with Diabase
x Mafic Extrusive Rocks
E Intercalated Felsic and Mafic Extrusives


iCnwir 1


ritill1 h1 r d nl mn-l n Iltwvt Hlll;r lnn nalic wAnth;inr rho





10

al., 1981; Schlager et al., 1984; Shaub, 1984; Dillon et al., 1985; Nelson


et al., 1985A; Lord, 1987


al., 198
Figure


8


Ball et al., 1988; McNeely, 1988; Mullins et


Winker and Buffler, 1988; Dobson, 1990; McBride, 1991).


2 is a location map compiled from these geophysical surveys.


In those seismic studies in which the original authors had not made
time-depth conversions, drill hole depth data were used to provide


controls for these conversions.


Where this drill hole control is not


available, such as in deeper areas, time-depth conversions are


largely subjective.


These depth determinations were then plotted on


the appropriate points on the grid.


Depth-to-basement values were


interpolated to each remaining point and the grid was subsequently


digitized and contoured.


well.


The final map is shown in Figure 3.


A Bouguer anomaly map of the study area was produced as
This map, which is shown in Figure 4, is essentially a digitized


and recontoured version of that produced by Klitgord et al. (1984),
although an older map by Oglesby et al. (1973) was extensively


utilized as well.


The original Klitgord map was constructed through


the compilation of several different maps from several different
studies and, consequently, there are minor inconsistencies at the


borders between these study areas.


In addition, their map has a


small data-less area extending along the northwestern coast of
peninsular Florida (Klitgord et al., 1984, pg. 7760, fig. 8).
For the purposes of digital filtering, it was desirable to produce
a single, continuous data set for the study area. In order to










































Figure 2.


Geophysical survey locations within the study area.


(Compiled from Antoine and Harding, 1965; Sheridan et al., 1966;











14

introduce some error, it is believed that any such error will be


minimal.


As with the depth-to-basement map, the Bouguer anomaly


map was digitized on a rectangular grid with a 9.2 km spacing, which
was chosen to conform to software limitations while maximizing the


number of data points.


The resulting 6696 data points were then


contoured with a 5 milligal contour interval using the Surfer
software package.
A magnetic anomaly map (Figure 5) was also constructed, using


data from King (1959)


Gough (1967), and Klitgord et al. (1984).


The


data were digitized from these original maps and contoured using a


50 nanotesla contour interval.


Unfortunately, there are relatively


large areas of the plateau for which there are no data.


The digital


filters used in this study employ Fourier synthesis to transform the
data from the space domain to the frequency (or wave number)
domain. As a consequence of Gibbs phenomenon, that is, the error
introduced during Fourier synthesis near a discontinuity, it was
determined that the data-less areas in the magnetic anomaly map
would result in unacceptable errors should the magnetic data set be
digitally filtered. As a result, the use of an unfiltered magnetic
anomaly map for the plateau is considered to be more desirable for
interpretive purposes.
Because this map represents a compilation from a number of
magnetic surveys (e.g., Klitgord et al., 1984), the anomaly values
from different sections of Figure 5 may be derived from several





15

magnetic anomaly patterns presented in Figure 5, there is
undoubtedly some error in the relative magnetic values between any
given area and values in other sections of the map.
The Bouguer anomaly data were digitally filtered using several
methods and a series of maps were created from the filtered data.
These digital filters included a downward continuation (Figure 6), an


upward continuation (Figure 7),
(Figure 8) of the Bouguer anoma


filters are explained below.


and a second vertical derivative
ly field. The purposes of these


In each case, several rows and columns


of extrapolated data were added to each side of the grid to reduce


the near-boundary distortion of Gibbs phenomenon.


These additional


rows and columns were then removed prior to contouring.
The downward continuation of potential-field data was done
from a level to an irregular surface using the method of Grauch


(1984)


. This method is used to approximate the relative values of


gravitational potentials at the basement surface.


to-basement is quite variable,


Because the depth-


ranging from about 850 m to over


6000 m, the removal of resultant depth-related deviations in the
observed field through this method of downward-
continuation might be expected to reveal previously unobserved


trends or features.


In addition, downward continuation has the


effect of sharpening the field,


which better outlines source features


at depth.
The application of downward-continuation filters, either in the



















198


. This necessitates the use of a low-pass filter prior to


performing these operations.
In order to apply this low-pass filter, the gridded data set is
transformed into the frequency, or wave number, domain through


the use of the two-dimensional Fast Fourier transform (or FFT)


. The


FFT is simply an efficient method of calculating the Discrete Fourier


transform (or DFT),


which, in turn, is a digital method of


approximating the continuous Fourier integral function (Cooley and


Tukey


1965


Bergland,


1969)


. The two-dimensional FFT is given by


G(k,I)


m=O0 n


=o g(m,n)


-2ni (


where G(k,l) is the potential-field function, with k and 1 representing

grid point locations in the frequency domain, m and n are integers
representing grid point locations in the space domain, M is the
number of rows, and N is the number of columns (Hildenbrand,


1983)


. Once in the frequency domain, a tapered low-pass filter was


applied; this progressively filtered out wavelengths from 36.8 km to
18.4 km, while wavelengths below 18.4 km (high frequencies) were


totally squelched.


The data set was then transformed back to the


space domain through the application of the inverse FFT (or IFFT).
For the actual downward continuation, the method of Grauch
(1984) utilizes a Taylor series expansion to approximate the


potential-field data, f(z)


at some depth, z, from the observed


Ln)
+ N


,









f(0) +
2


F(z) = F(O) +


The


i2 f(O) +
OL


z value for each grid location was taken from the depth-to-


basement data already described.


This calculation was repeated for


each point on the grid, the added boundary data were removed, and
the resultant data set subsequently contoured (Figure 6).
Another filter applied to the Bouguer anomaly field was an


upward-continuation to an elevation of 10 km.


This has the effect of


removing short-wavelength features of the field, which are often


attributable to shallow crustal sources (Figure 7)


. The upward-


continuation is accomplished by taking the FFT of the data as
described above, giving the Fourier coefficients G(k,1l). As described


by Fuller (1967)


these Fourier coefficients are then multiplied by


H(k,1)


the wave number response of the filter.


H(k,l) is given by


H(k,I)


2rrz(k


where z is the continuation distance.


The application of the IFFT to


the gridded data and the removal of the added boundary data then
permitted the construction of the final upward-continued map.
In a similar manner, a second vertical derivative map was also


produced.


The second vertical derivative effectively removes long-





22

frequency domain and the Fourier coefficients subsequently
multiplied by the wave number response of the filter as given by


H(k,I)


=411


(Fuller, 1967)


Again, the data set was then transformed back into the space domain
using the IFFT, the extra boundary data removed, and the grid
subsequently contoured.


Results and Interoretations


In general, these maps complement and augment the results of


many previous investigations.


The depth-to-basement map


accurately depicts the Peninsular Arch, the expected southward
increase in depth, the series of Mesozoic basins in the southern and
western regions of the plateau, and a small previously undescribed
basin under the northeastern edge of the plateau (Figure 3).
While the differences between the original Bouguer anomaly
map (Figure 4) and the downward-continued map (Figure 6) are
relatively subtle, the downward continuation is preferable as it more
accurately delineates the boundaries of anomaly-producing features.
The upward-continued map (Figure 7) delineates a number of long-
wavelength features that are clearly associated with large scale
crustal features, such as the Mesozoic basins, that have been


previously described, but poorly located and characterized.


On the





23

accumulations of the Florida Platform are relatively homogeneous


with respect to density.
shallow crustal features


Furthermore, it suggests that smaller
, such as plutons, are not particularly


abundant on the plateau, although this does not preclude the
existence of such features with low density contrasts or in the middle
crust.


Pre-Mesozoic Features of the Plateau Basement
The lithologies of the pre-Mesozoic basement terranes of the
northeastern portion of the plateau have been moderately well


characterized.


Basement lithologies from the remaining portions of


the plateau are unresolved; however, more structural detail is


known.


As a result, this section will necessarily be focused on the


pre-Mesozoic lithotectonic terranes of northern peninsular Florida,
while the next section will be focused primarily on the Mesozoic


structural features of western Florida.


In the following discussion,


the location of each of the lithotectonic features or regions within the
study area is designated by a letter on Figures 9-15.


Osceola Granite--reeion A.


The Osceola Granite, previously


known as the Avalon terrane, is the oldest identified terrane in the


plateau basement.


Although described as a granite by convention, it


is better characterized as a relatively undeformed granitoid batholith
composed of a range of felsic lithologies (Milton and Grasty, 1969;


Barnett, 1975; Smith, 1982)


. This granitoid complex underlies the




























been obtained by Dallmeyer et al. (198


7) using 40Ar/39Ar


techniques on a composite biotite sample from several specimens.
Sm-Nd isotopic data for the Osceola Granite are suggestive of a mixed
source containing both Proterozoic and Archean components
(Heatherington et al., 1989).
The 8 drill holes intersecting granite in east-central Florida are
generally centered around a pentagonal negative gravity anomaly


(Figures 9, 13


14--A), although several of these holes are outside the


boundaries of the anomaly.


The magnetic signature of the terrane is


more complex, as the easternmost of the holes are located along a
north-south oriented, elongated positive magnetic anomaly, while
most of the others are located within the confines of a proximal


negative magnetic anomaly (Figure 11--A).


The north-south


oriented boundary between these magnetic features bisects the
negative gravity anomaly.
Barnett (1975) modeled the Osceola Granite as constituting the
block-faulted end cap of the Peninsular Arch, a hypothesized
structural high under north-central Florida. Although the Peninsular
Arch has since been described as an erosional feature (Smith, 1982)
and the Osceola Granite shows little petrologic evidence of
metamorphism (Milton and Grasty, 1969), the possibility of block


faulting remains.


The randomness and variability in depth-to-


basement for the 8 drill holes is suggestive of block-faulting, as is the
variability in both the gravity and magnetic fields in the area





32

suggested by the number of drill holes that have bottomed in
weathered diabase throughout the area (Figure 9).
The contacts between the Osceola Granite and the juxtaposed
terranes are not well constrained. A primary reason for this arises
from the inherently poor resolution of isotopically determined ages
for the North Florida Rhyolite to the north (Mueller and Porch, 1983)
and the St. Lucie Metamorphic Complex to the southeast (Bass, 1969).
As a result, the fundamental age relationships between the three
terranes remain undetermined and age-based inferences concerning
the nature of the contacts between them are ambivalent. A number
of contradictory suggestions have been made.
For example, Chowns and Williams (1983), citing the apparent
concordance in ages between the Osceola Granite and the
amphibolite, suggested that they could comprise a conterminous
terrane. Accordingly, the rhyolitic terrane would then overlie both
the granite and amphibolite, and be separated from them by a thrust
fault or unconformity. Alternatively, Bass (1969) believed that the
St. Lucie amphibolite, by reason of metamorphic grade, must
necessarily be separated from all other Florida basement rocks by a
"fundamental discontinuity." This was supported by the presence of
thin layers of unaltered basaltic rock, granite, and diorite
immediately over the amphibolite, which implies a horizontally
faulted contact (Bass, 1969). Another possibility is that the Osceola
Granite represents an intrusion into an older terrane consisting of the





33

Granite that it is bounded by steeply dipping contacts with the
surrounding lithologies.
Samples from several drill holes show that the Osceola Granite
extends northeastward and northwestward beyond the boundaries
inferred from the potential-field gradients; however, to the south,
other holes penetrate rocks indicating that the steep gradients mark


the boundaries of the terrane (Figures 9, 11, 13--A)


southwestern boundary


Along its


the Osceola Granite appears to be truncated


by the Jay Fault, as has been suggested by Smith (1982).


To the


southeast, the presence of both mafic and felsic extrusive rocks in
drill holes near to those intersecting the Osceola Granite clearly


delineates the boundary, which coincides with steep,


northeast-


trending gravity gradients separating the pentagonal negative


anomaly from a smaller ovoid positive anomaly.


It is unclear


whether the extrusive rocks overlie the granitoid or rather have


been juxtaposed against it.


The eastern boundary of the Osceola


Granite is unresolved, although the steep north-northeast trending
gravity gradient along the eastern side of the pentagonal negative
anomaly suggests that the granitoid may not extend beyond the coast
of peninsular Florida.


St. Lucie Metamorphic Comnlex--reeion B.


A single drill hole


has revealed the presence of an amphibolite grade metaigneous
assemblage at a depth of about 3760 m immediately southeast of the
Osceola Granite, underlying St. Lucie County (Bass, 1969; Milton and







approximate age obtained for the Osceola Granite.


The assemblage is


comprised primarily of amphibolite, with some quartz diorite gneiss


and chlorite schist (Bass, 1969; Thomas et al., 1989).


Although


several names have been applied to this terrane, in this work it will
be referred to as the St. Lucie Metamorphic Complex.
The drill hole is located near the northern edge of an ovoid


east-west oriented negative gravity and magnetic anomaly


. This


anomaly is relatively small


approximately


100 km long by 5


wide


, and may delineate the boundaries of the terrane (Figures 11


and 13


--B).


It is probable that the St. Lucie Complex is bounded to


the southwest by the Jay Fault.


To the north


the terrane extends


only a short distance, as demonstrated by drill holes and inferred


from the gravity and magnetic anomaly.


To the east, the anomaly


extends only to the coast of peninsular Florida and, in an
interpretation of an offshore seismic reflection profile, Sheridan et al.
(1981) found no evidence for the presence of this terrane off the


Florida coast.


As a result, it appears that the St. Lucie Metamorphic


Complex encompasses only a relatively small area and may therefore


be a fault-bounded block.


An alternative possibility is that the


metamorphic rocks represent a roof pendant in the proximal felsic


igneous rocks (P


Mueller, 1993


personal communication).


The metamorphic grade of the assemblage is indicative of
regional activity, which suggests that this small terrane represents


an isolated fragment of a once larger unit (Bass, 1969).


Another





35

Metamorphic Complex and are approximately the same age


(Dallmeyer, 1984).


Based on these similarities and because there are


no other similar metamorphic terranes known in the Gulf of


Mexico/Caribbean region


Schlager et al. (1984) have inferred the


two to be cogenetic.


North Florida Rhvolite--reeion C.


Overlying the Osceola Granite


to the north and west is a terrane of slightly metamorphosed calc-


alkaline rhyolitic to andesitic volcanic rocks.


The upper surface of


this terrane in northeastern Florida dips southeastward and ranges


in depth from about 1180 m to about 1650 m (Smith, 1982).


The


calc-alkaline nature (Mueller and Porch, 1983) and trace element


signatures (Heatherington et al.,


1989) of samples from this terrane


each suggest eruption in an ocean-continent convergent margin


setting.


In attempting to date two separate samples,


one rhyolite


and one andesite, Mueller and Porch (1983) identified disturbed


40Ar/39Ar release patterns from each--despite this,


concordant ages of 412


apparently


15 Ma and 418 9 Ma were obtained.


Although the disturbed release patterns suggest these age
determinations to be minima, stratigraphic relationships suggest a
Late Precambrian to Early Paleozoic age for the terrane (Arden,
1974; Chowns and Williams, 1983).
Few structural inferences have been made about the North


Florida Rhyolite.


From a seismic reflection survey in the Florida


panhandle, Arden (1974) distinguished faulting and gentle folds in





36

the South Georgia Rift, as suggested by Arden (1974) and Klitgord et
al. (1984).
The sublateral extent of this terrane is uncertain. Although the
distribution of drill holes intersecting rhyolite is generally coincident
with an elongated negative gravity and magnetic anomaly (Figures 9


and 13-C)


previous geophysical and drill hole studies suggest that


the terrane continues northward under the Suwannee Basin terrane,
described below, and extends into southern Georgia (Barnett, 1975;


Wicker and Smith, 1978; Chowns and Williams, 1983).


To the south,


two drill holes intersecting rhyolite have been interpreted to indicate
a subcrop of this terrane immediately south of the Osceola Granite


(e.g., Chowns and Williams, 1983


Dallmeyer, 1987)


Several undated


rhyolitic samples have also been obtained from the south-central
portion of the Floridan Plateau in the South Florida Basin; it has been
suggested that these samples are either from horsts of this same
felsic volcanic terrane (Thomas et al., 1989) or from part of a
separate southerly Mesozoic bimodal suite (Barnett, 1975; Mueller


and Porch, 1983)


. To the west, felsic basement rocks have been


encountered in the Florida panhandle area; these rocks are undated,
and have been variously correlated to the North Florida Rhyolite


(Arden, 1974; Thomas et al.,1989),


1975)


the Osceola Granite (Barnett,


and to the Permian (?) granite of the Wiggins Arch in


Mississippi (Smith, 1983).


Suwannee Basin--region D.


Overlying the north Florida igneous







1983; Thomas et al, 1989).


This terrane extends northward into


southern Georgia and westward at least to the Florida panhandle, as
well as southwestward to the Middle Ground Arch in the north-
central region of the plateau (Ball et al., 1983; Dobson and Buffler,


1991)


. The Suwannee Basin has been intersected by numerous drill


holes at depths ranging from 720 m to 1730 m (Smith, 1982) and is
composed of a massive Early Ordovician quartzitic sandstone overlain
by intercalated Middle Ordovician to Middle Devonian sandstones


and shales (Arden, 1974)


. The topographic high of the terrane, in the


north-central portion of peninsular Florida, is an erosional remnant
known as the Peninsular Arch.
The structure of the Suwannee Basin is somewhat better
defined than that of the underlying North Florida Rhyolite, although
the nature of the contact between the two is rather enigmatic. Arden
(1974), from an interpretation of a seismic reflection profile,
interpreted the contact to be a northeast-dipping, Middle to Late
Paleozoic age thrust fault that emplaced the Suwannee Basin over the
volcanics, while others (e.g., Thomas et al., 1989) have suggested the
contact to be an unconformity. Although drilling into the Suwannee
Basin terrane has never penetrated more than about 700 m, gravity
modeling indicates that the eastern Suwannee Basin has a maximum
thickness of about 2500 m (Wicker and Smith, 1978), while seismic
reflection profiles suggest the western Suwannee Basin may be even
thicker (Arden, 1974; Thomas et al., 1989).





38

profile and numerous drill holes have shown the western section to
have been extensively faulted, mildly folded, and intruded by


diabase during the Triassic evolution of the South Georgia Rift.


This


difference has prompted some workers to separate the Suwannee
Basin into two distinct terranes, the East and West Suwannee Basins
(e.g., Klitgord et al., 1983; Ball et al., 1988).
The relatively smooth potential-fields in the region occupied by
the Suwannee Basin terrane are bisected by two linear NE-trending
negative gravity anomalies separated by a distinct, northeast-
trending negative magnetic anomaly and low-amplitude positive


gravity anomaly.


These extend from the northeastemmost Gulf of


Mexico to the eastern Florida-Georgia border (Figures 11 and 1


feature D1)


Based on an onshore, local, high resolution gravity


study, Coleman and Stewart (1982) believed the Suwannee Basin
sedimentary deposits to be underlain by two northeast-trending
Paleozoic (?) troughs coinciding with these linear negative gravity


anomalies.


Offshore, however, Dobson and Buffler (1991) have


interpreted basement faulting in a seismic reflection profile and the
presence of Triassic red beds in proximal drill holes as indicative of a
moderately-sized, northeast-trending Triassic rift basin extending at
least to the northwest Florida coast and coinciding with the central
positive gravity anomaly and the associated negative magnetic


anomaly of feature D1.


The absence of corresponding red beds in


onshore drill holes suggests the possibility that feature D1, although





39

Although the boundary between the Suwannee Basin terrane
and the superimposed Triassic South Georgia Rift is reasonably well
constrained from drill hole data, it is likely that the Suwannee Basin
extends northward and westward under the rift basin and for a


significant distance beyond it.


COCORP seismic reflection profiling


(Nelson et al., 1985A) and gravity modeling (Lord et al., 1992) both
indicate that the Suwannee Basin extends into southern Georgia along
most of its border with Florida, although neither the nature nor the
location of the northern or western boundaries of the terrane have
been resolved.


lay Fault--feature E-E1


. A variety of evidence suggests the


presence of a major crustal boundary, known as the Jay Fault, which


may divide the Floridan Plateau crust.


Northeastern Florida is


generally characterized by northeast-southwest trending gravity and
magnetic anomalies, while those in southwestern Florida are
generally perpendicular to this trend (e.g., King, 1959; Oglesby et al.,
1973; Smith, 1982). Although the boundary between these two
trends is not well defined, it extends through the center of the
plateau in a northwest-southeast direction, close to a landward
extension of the trend of the Bahamas Fracture Zone.
Drill hole evidence and seismic reflection profiles suggest that
the implied extension of the fracture zone also constitutes the
boundaries of several features, including the South Georgia Rift, the
north Florida igneous rocks, and the Suwannee Basin terrane





40

and in the northeastern Gulf of Mexico (Dobson and Buffler, 1991)


suggest normal faulting along this same trace.


Consequently, the Jay


Fault is generally considered to be a morphological extension of the


Jurassic Bahamas Fracture Zone.


The origin, character, and age of this


extension, however, are unresolved (Barnett, 1975; Smith, 1982;
Klitgord et al., 1984; Smith, 1993).
It has been suggested that the Jay Fault is a Jurassic left-lateral
transform that served to connect spreading centers in the Atlantic
and Gulf of Mexico, while bringing several disparate lithotectonic
units into juxtaposition (Klitgord and Schouten, 1980; Smith, 1982;
Klitgord et al., 1984). Using this model as a starting point, several
authors (Smith, 1982; Chowns and Williams, 1983; Whitelaw and
Smith, 1989) have hypothesized an extension of this proposed
transform boundary across the Floridan Plateau and into Alabama,
where it may merge with the Middle Mesozoic Gulf basin marginal
fault zone (Smith, 1983; Klitgord et al., 1984).
Another possibility for the origin of the feature is that it is a
right-lateral feature formed during the Late Paleozoic closure of the
Iapetus Ocean (Smith et al., 1992; Smith, 1993). The premise of this
model is that the Jay Fault originated in two stages. During the
Paleozoic, the boundary was a right-lateral, strike-slip fault that
served to accommodate differential motion within Gondwana during


closure with Laurentia.


Later, during Triassic and Jurassic extension,


the boundary was reactivated as a normal fault.


This would be







panhandle region, also suggested a right-lateral fault model,


although


in this model, the fault formed in response to north-south
compressive stresses during the Jurassic.
Inherent in a third model for the nature of the lithotectonic
boundary across Florida is the implication that the Jay Fault has
never been subject to significant lateral motion (Heatherington et al.,


1989; Heatherington and Mueller


1991).


Results of trace element


analyses of basement rocks from north and south Florida exhibit
similar patterns of high field strength element depletion and large
ion lithophile element enrichment (Heatherington et al., 1989), as
well as similar Nd model ages for the lithosphere from which these


rocks were derived (Heatherington and Mueller, 1991)


. This


suggests that the volcanic suites of north and south Florida may have


been derived from a single contiguous underlying unit.


It is possible,


then, that this geophysically and lithologically inferred fault may not
be a transform that juxtaposed two entirely disparate lithotectonic
units, but rather that it is simply a normal fault zone that
experienced little or no lateral movement (Heatherington and
Mueller, 1991).
The linear nature of the Jay Fault is evident on the depth-to-
basement and upward-continued Bouguer anomaly maps (Figures 10


and 14--E-E1)


. The upward-continued Bouguer anomaly map


strongly suggests that the fault acts as boundaries for the Osceola
Granite (A), the St. Lucie Metamorphic Complex (B), the South Georgia





42

is the Jay Fault linear, but apparently continuous across the Floridan
Plateau as well.

Along the southeastern edge of the plateau, Sheridan's (1981)
seismic reflection profile clearly shows a series of faults with varying
amounts of horizontal offset along the proposed trace of the Jay


Fault.


Hence, the Jay Fault is more accurately described as a zone of


faulting, rather than as a single fault. An abrupt southward increase
in depth-to-basement marks the zone along most of its length. An
exception to this is in the northwestern part of the plateau where the
South Georgia Rift extension intersects the zone. At this point, a
seismic reflection profile (Dobson and Buffler, 1991) indicates the


presence of a single northeastward-dipping master fault.


Thus,


although the Jay Fault zone appears to be continuous and linear, the
nature of the Mesozoic faulting changes along strike.


Mesozoic Features of the Plateau Basement


The Floridan Plateau basement is characterized by a variety of


Mesozoic structural features.


Lithic accumulations associated with


these features (i.e., red beds, hypabyssal rocks, etc.) often obscure
the pre-Mesozoic units of the underlying crust and, as a result, the
pre-Mesozoic nature of the basement under these features is largely


unresolved.


While the general locations of Mesozoic features on the


plateau basement are reasonably well established, specific structural
details and subaerial boundaries are poorly constrained.





43

northeast-striking Jurassic basins and ridges. This series extends to
southern peninsular Florida, where a large Triassic-Jurassic basin
occupies the southern third of the peninsular basement (Klitgord et
al., 1984).


South Georgia Rift--reeion F.


Underlying the north-central


portion of the plateau is a 95 km wide, northeast-trending section of


the Early Mesozoic South Georgia Rift.


The South Georgia Rift is a


large, complex graben system extending from the Florida panhandle


area to eastern Georgia (Daniels et al.,


1983; McBride,


1991).


The


section in Florida overlies the western block-faulted edge of the
Suwannee Basin terrane (Smith, 1982) and has been previously
referred to as the Tallahassee Graben (Smith, 1983) and the


Southwest Georgia Embayment (Barnett, 1975; Miller, 198


. The rift


is filled with arkosic sandstones and mafic igneous rocks with a
maximum thickness of about 1800 m in Florida (Arden, 1974) and it
has been pervasively intruded by diabase sills (Barnett, 1975;


Chowns and Williams, 1983; Smith, 1983).


Milton and Grasty (1969)


acquired three K-Ar ages for samples of diabase sills from northern
Florida. Although these ages are now generally considered to be
suspect, they were found to range from about 180 to 200 Ma.
The onshore boundaries of the South Georgia Rift are
reasonably well constrained from the presence of red beds and
associated hypabyssal rocks encountered in deep drill holes. A
circular negative gravity anomaly with a diameter of about 50 km





44

accumulations of low density salt in the Jurassic Apalachicola
Embayment, which overlies the South Georgia Rift in this area.
The north-central Floridan extension of the Triassic-Early
Jurassic South Georgia Rift system is a moderately complex feature.
The presence of short-wavelength gravity and magnetic anomalies in
the area of the rift (Figures 11 and 15--F) may indicate it to be
comprised of an assemblage of secondary horst and graben


structures (Barnett, 1975


Smith, 1983).


Limited seismic reflection


profiling shows extensive normal faulting in the rift floor and the


surrounding area; however, 1
definitively detected (Arden,


no large secondary horsts have been
1974; Dobson and Buffler, 1991).


Because there are few seismic reflection profiles in the area
and, as shown in Figure 14, the long-wavelength potential-field
signatures associated with the rift have relatively low gradients, the
lateral boundaries of the rift extension are not definitively mapped


or characterized.


Ball et al. (1988), in a seismic reflection profile just


off the coast of the Floridan panhandle, interpreted a northeast-
dipping master fault with about 1400 m of throw to mark its


southwestern terminal boundary


Subsequent reflection work has


shown the offshore extension of the rift system to consist of two


separate grabens,


at least one of which probably extends to the


southwest as far as the Jay Fault zone (Dobson and Buffler, 1991).
Analachicola Embavment/Analachicola Basin-reeions F. G.


Extending from northwest to southeast across the Floridan Plateau





45

the Apalachicola (or Southwest Georgia) Embayment (F) and the
Middle Ground Arch (H), a basement high that may extend


undisturbed across the Jay Fault zone.


These are located to the


northeast of the Jay Fault zone and are part of a series that extends
northwestward and includes the Chattahochee Arch, the Covington
Embayment, the Conecuh Arch and the Wilcox Embayment, all of
which lie out of this study area (e.g., Miller, 1982; Mitchell-Tapping,
1982; Smith, 1983).
On the southwestern side of the Jay Fault zone are the


Apalachicola Basin (G)


also known as the Northeast Gulf Basin, and


the Tampa Basin (I), also known as the St. Petersburg Basin.


are separated by the Middle Ground Arch (H).


These


Continuing to the


southeast, the Sarasota Arch (J), also known as the Pinellas Arch,
separates the Tampa Basin (I) from the South Florida Basin (L) (e.g.,


Klitgord et al.,


1984; Ball et al.,


1988; Winker and Buffler, 1988


Dobson, 1990; Dobson and Buffler, 1991). As shown in Figure 16,
there have been significant variations between previous publications
with regard to the locations, sizes, and spatial relationships of these
various Jurassic features.
The Apalachicola Embayment (F), which is bounded along its
northwestern edge by the Chattahoochee Arch, is a Jurassic basement


trough overlying the Triassic South Georgia Rift.


On the basis of drill


holes and electric logs, Mitchell-Tapping (1982) modeled the
Apalachicola Embayment as a graben flanked by two peripheral








47

discussed circular negative gravity anomaly, there is no definitive
potential-field signature attributable to the Apalachicola Embayment.
The Apalachicola Embayment is contiguous with, and
essentially indistinguishable from, the Apalachicola Basin,except that
the two are separated by a prominent basement hingeline at the
apparent trace of the Jay Fault zone and the Apalachicola basin is


characterized by a much greater depth-to-basement.


Southwest of


the hingeline, dipping intrabasement reflections suggest the presence
of Paleozoic strata with a synclinal structure underlying the


Apalachicola Basin (Dobson, 1990; Dobson and Buffler, 1991).
eastern section of the basin in the study area is marked by a


The


moderate amplitude positive gravity and magnetic anomaly (G)


. The


basin extends southwestward to the plateau escarpment and may
reach a basement depth of 12 km (Buffler and Sawyer, 1985; Buffler,


1989)


. The eastern margin of the basin is marked by a major


bounding fault with a throw of at least 1000 m (Ball et al., 1988).
This fault separates the Apalachicola Basin from the Middle Ground
Arch.


Middle Ground Arch-region H.


Extending across the Jay Fault


zone, adjacent to the Apalachicola Basin, is the Middle Ground Arch.
This is a broad, topographically positive feature that extends from
the southwestern flank of the Peninsular Arch westward to the edge


of the plateau (Figure 10--H)


. The crest of this ridge dips generally


westward and ranges in depth from about 1500 m under peninsular





48

Apalachicola and Tampa Basins immediately west of the Jay Fault
zone, the arch has been classified by some as two separate features--
the Middle Ground Arch to the east and the Southern (or DeSoto)
Platform to the west (Winker and Buffler, 1988; Dobson, 1990;
Dobson and Buffler, 1991).
The Middle Ground Arch is marked by one of the northeast-
trending, long-wavelength, low amplitude negative gravity anomalies
(Figure 14--H) previously mentioned as flanking feature Dl.
Although there are few short-wavelength features in the region,
seismic reflection profiles have shown the crest of the Middle Ground
Arch to have been moderately faulted. As suggested by stratigraphic
perturbations, these faults have horizontal throws on the order of
100 m and became inactive after the Late Jurassic (Ball et al., 1988;
Dobson, 1990). In addition, graben structures and fill associated with
the South Georgia Rift extend along the crest of the northeastern end
of the Middle Ground Arch from peninsular Florida to the Jay Fault
zone (Dobson and Buffler, 1991). An intrabasement igneous
intrusive body, possibly Mesozoic in age, has been identified on
seismic reflection profiles and is marked by an oblong low amplitude
negative gravity anomaly (Figure 15-feature HI) (Ball et al., 1988).
Near the southeastern flank of the Middle Ground Arch, a single drill
hole intersecting Paleozoic siltstones (Ball et al., 1988) signifies a
possible westward extension of the Suwannee Basin terrane.
Martin (1978) suggested that the Middle Ground Arch is simply







Basin implies a significant degree of structural control.


The single


seismic reflection line extending across the southeastern boundary of
the arch, on the other hand, does not show any analogous faulting


(Ball et al., 1988).


In addition, the Bouguer anomaly gradient across


the strike of this southeastern boundary is uniform and relatively


low (Figure 1


. These observations suggest the Tampa Basin-Middle


Ground Arch margin to be a broad transition zone with little
structural control, which reinforces Martin's proposal of formation by
differential subsidence for this particular boundary.
Tampa Basin--region I. Southeast of the Middle Ground Arch is
the Tampa Basin, which is also known as the Florida Elbow Basin
(Pindell, 1985). It is essentially an analogue to, and coeval with, the
Apalachicola Basin. The Tampa Basin extends updip northeastward
from the western edge of the plateau although, unlike most the other
Jurassic features in this series, it has not been previously mapped as
extending eastward as far as the Jay Fault zone (e.g., Klitgord et al.,


1984; Buffler, 1989)


. It reaches a depth-to-basement of at least


5500 m (Figure 10--I) and Dobson and Buffler (1991) have modeled
it to be significantly deeper (greater than 9000 m).
Seismic reflection profiles are characterized by dipping
intrabasement reflections in the basin that, as in the basement of the
Apalachicola Basin, imply an underlying extensively faulted Paleozoic


synclinal structure (Ball et al., 1988; Dobson, 1990).


None of the


three offshore wells intersecting basement in the Tampa Basin,







intercalated with Jurassic diabase (Dobson and Buffler, 1991). In
addition, some of this faulting perturbed the Cretaceous sediments


immediately overlying these igneous rocks.


Thus, it is likely that at


least some of the basement faulting and structure originated during
the Jurassic evolution of the basin, rather than during the Paleozoic.
The Bouguer anomaly signature of the Tampa Basin is quite
distinct (Figure 14--1). It is marked by a 50 mgal long-wavelength
positive gravity anomaly. This positive anomaly averages about 160
km wide and 275 km long, suggesting that the Tampa Basin is
significantly larger than shown on previous basement maps;
however, a corresponding positive magnetic anomaly appears to be


somewhat smaller (Figure 11--I).


As discussed above, the


northwestern boundary of the basin is probably a sloping margin.
That margin appears to crest approximately 40 km south of the
northernmost expression of the gravity anomaly, although shallow
dip interpretations are difficult from the available seismic reflection
profiles.
Pindell (1985) has speculated that a large strike-slip fault, the
Florida Elbow Fault, extends northwestward from the Jay Fault zone,
across the Tampa Basin, and into the northeastern Gulf of Mexico.
Although there is no gravity or seismic evidence for such a fault, the
steep eastward-striking magnetic gradient which bisects the
northern half of the basin may be the result of a structure along that


trend.


The relative absence of short-wavelength gravity anomalies





51

diameter moderate amplitude negative gravity anomalies with
corresponding positive magnetic anomalies may indicate the
presence of mid-crustal intrusive bodies similar to that under the
Middle Ground Arch (Figures 11 and 15--feature II).
The northeastern boundary of the Tampa Basin, as suggested
by the positive gravity anomaly, appears to correspond extremely
well with the presumed trace of the Jay Fault zone (Figure 14). This
places the easternmost section of the Tampa Basin under western
peninsular Florida and shifts the boundary significantly more
northeastward than had been previously modeled, although Ball et
al. (1988) identified a basement monocline only 30 km off the west
coast of Florida that was suggested to be the boundary. In addition
to the positive gravity anomaly, the presence of Jurassic diabase in
several drill holes in the same area of western peninsular Florida
(Figure 9) and a corresponding positive magnetic anomaly (Figure
11) are also suggestive of a northeastward extension of the basin as
far as the Jay Fault zone.
The southeastern boundary of the basin, which separates it
from the Sarasota Arch, is better constrained with respect to location.
Along this boundary, both of the potential-field gradients are
relatively high, which is suggestive of a morphologically distinct
feature. Although not confirmed by seismic data, stratigraphic
correlations from well logs also suggest the presence of a large down-
to-the-north fault or series of faults along this boundary; this fault







Tampa Basin and the South Florida Basin.


It extends from the Jay


Fault zone under peninsular Florida westward possibly to the edge of


the Florida Escarpment (Figure 10--J)


although Klitgord et al. (1984)


have suggested that the arch may be truncated by an extension of


the Cuba Fracture Zone just east of the escarpment.


The depth-to-


basement of the Sarasota Arch averages 4 km and there is little


evidence of any significant dip along the crest.


There have been four


drill holes intersecting basement on the Sarasota Arch--three
bottomed in granitic rocks and one intersected diabase (Figure 9).
Although seismic profiles in the southern quarter of the
Floridan Plateau tend to be poor quality as a consequence of the
depth and the attenuation of seismic energy by the overlying karstic
limestone, a number of vertical faults have been observed on seismic
profiles crossing the Sarasota Arch (Shaub, 1984; Ball et al., 1988).
These faults are small, with throws on the order of 10s of meters,
and, as suggested by perturbations in the overburden, became
inactive by the Early Cretaceous.
Klitgord et al. (1984) interpreted a northeast-trending zone of
circular gravity and magnetic anomalies along the northwest edge of
the Sarasota Arch to be the result of a series of shallow
intrabasement intrusive bodies. Although several short-wavelength
circular negative gravity anomalies exist in this area (Figure 13--
region J), not all of these correlate well with the circular magnetic


anomalies (Figure 11-region J).


Therefore, while two or three of the





53

intrusions, a linear chain spatially associated with the Sarasota Arch
seems improbable.
The Sarasota Arch is marked by a distinct long-wavelength
negative gravity anomaly (Figure 14--J), although there seems to be
no particular associated magnetic signature (Figure 11). As
described above, the northwestern boundary of this negative
anomaly correlates well with the large bounding fault inferred by
Ball et al. (1988). Although the anomaly appears to be partially
truncated near its northeastern boundary, it generally extends
southwestward from the region of the Jay Fault. The southwestern
boundary of the negative anomaly could be interpreted to occur
slightly east of the escarpment; however, a seismic reflection profile
along the crest of the Sarasota Arch (Ball et al., 1988) shows no
evidence for truncation at an extension of the Cuba Fracture Zone. As
a result, it is likely that the southwestern boundary of the arch is
coincident with the Florida Escarpment.
The southeastern boundary of the Sarasota Arch, as interpreted
from a seismic reflection profile near the escarpment, is a
southeastward-dipping, relatively steep, block faulted margin
(Shaub, 1984) that corresponds to the boundary of the negative

gravity anomaly. However, this gravity boundary is not continuous.
As discussed below, a 60 km wide negative gravity anomaly extends
southeastward from the boundary (Figure 14 --feature L4) and is
consistent with the probable presence of a large basement horst





54

recognized a possible basement high along the Florida Escarpment
extending southeastward from the Sarasota Arch. This high, the
Sheffield Arch, was identified from a limited number of profiles with
no deep well control and is, therefore, not well characterized.
There is a positive gravity anomaly oriented generally along
the axis of this proposed arch (Figures 14 and 15 -K). This positive
anomaly is approximately 180 km long and 120 km wide. It has no
concomitant magnetic signature. Although the positive gravity
anomaly could be the result of a morphologically positive feature
such as the Sheffield Arch, all of the other positive basement
features of the Floridan Plateau have been characterized by negative


gravity anomalies.


Consequently, the characterization of a large,


continuous basement high in this area is unresolved.


South Florida Basin--region L.


Underlying southern peninsular


Florida is a Triassic-Jurassic structural feature known as the South


Florida Basin.


Based on a series of two-dimensional gravity profiles,


Wicker and Smith (1978) determined that the basin ranges in depth
from about 3000 m in south-central peninsular Florida to more than


5500 m at the southern edge of the peninsula.


The basin is generally


considered to underlie much of the southern portion of the Floridan
Plateau and to extend to its southern and eastern edges (e.g., Barnett,


1975; Smith, 1983; Klitgord et al., 1984).


The western extent of the


basin and the presence of any significant intrabasinal structures are
unresolved.





55

Triassic alkali basalts and rhyolites (Barnett, 1975; Mueller and


Porch, 1983; Heatherington and Mueller, 1991)


it has been suggested


that the occurrences of rhyolite represent horsts of an underlying
southerly extension of the Late Precambrian-Cambrian North Florida


Rhyolite (Thomas et al., 1989)


Isotopic dating of the rhyolites


(Milton and Grasty, 1969; Barnett, 1975) and the basalts (Milton,


197


Mueller and Porch, 1983) suggest that the two are coeval, with


ages of about 180 Ma.


This supports the model that the basalts and


rhyolites comprise a Mesozoic bimodal suite and almost certainly
discounts the premise that the rhyolites in the region of the South
Florida Basin are part of the North Florida Rhyolite.
There are only a few drill holes intersecting basement within
the South Florida Basin and all of these are in southwestern and


south-central peninsular Florida (Figure 9).


In addition, there is only


one published seismic reflection profile of the basement in the basin
(Figure 2). As a result, the basement model for this area is
necessarily based almost exclusively on potential-field data and is,
therefore, rather speculative.
The northeastern boundary of the South Florida Basin is the Jay


Fault zone.


In their seismic reflection profile along the eastern edge


of the Floridan Plateau, Sheridan et al. (1981) found this boundary to
be a large, down-to-the-south, normal fault with an associated series
of smaller faults at the landward extension of the Bahamas Fracture


Zone.


The boundary is marked by a northwest-trending zone of





56

southern margins of the South Florida Basin are not morphologically
well defined; however, the basin is considered to extend at least as
far as the Florida Escarpment.
Although the entire area bounded by the Jay Fault zone, the
Sarasota Arch, and the Florida Escarpment is commonly considered to
be a singular basin, there are four or five (depending on whether
feature K is included) individual long-wavelength potential-field
features within these bounds (Figures 11 and 14-L1-L4). For
example, the northeastern corner of this region is marked by an
oblong positive gravity and magnetic anomaly oriented along the Jay


Fault zone (LI).


Immediately southeast of this feature is a similarly


oriented negative gravity and magnetic anomaly (L2) which, in turn,
is separated from a second negative gravity anomaly (L3) by a


second positive gravity anomaly (L4)


. This long-wavelength


variability and the shapes of these features suggest the probability
of several northwest-southeast oriented basins within the area
traditionally considered to be the South Florida Basin.
All of the South Florida Basin area drill holes which have
successfully intersected basement have been located proximal to a


negative gravity and magnetic anomaly, specifically, feature L2.
suggests that the anomaly may mark a region characterized by a


relatively shallow basement.


This


In addition, the southern end of


Sheridan's reflection profile, near the southeastern corner of the
plateau, was located on a positive gravity and magnetic anomaly and





57

gravity and magnetic anomalies within the South Florida Basin
denote high depth-to-basement areas, while the negative gravity and
magnetic anomalies are characteristic of shallow basement arches.


Blake Plateau Basin--feature M.


Along the northeastern edge


of the Floridan Plateau is an area that, as suggested by a series of
seismic refraction profiles (Sheridan et al., 1966), is characterized by


an increase in depth-to-basement (Figure 10--M).


This feature may


represent a shallow salient of the Blake Plateau Basin, which is the
southernmost of the four major Jurassic-Early Cretaceous basins of
the eastern United States continental margin (e.g., Grow and


Sheridan, 1982; Dillon et al., 1985).
signature attributable to this salient.


There is no particular gravity
The western boundary of the


main body of the Blake Plateau Basin is located just east of the
eastern continental margin of the Floridan Plateau, slightly out of this
study area. Although the data are limited, there is no evidence that
the salient is fault-bounded. Rather, it appears to have formed
simply by differential subsidence.


Cenozoic Structural Features of the Floridan Plateau
There are only a few reports of Tertiary structural features of


tectonic origin on the Floridan Plateau.


These include small faults


resulting from halokinesis off the panhandle of peninsular Florida
(e.g., Addy and Buffler, 1984; Ball et al., 1988), as well as slump
faults and related features near the western Florida escarpment (e.g.,





58

tectonic origins; however, in each of these cases, subsequent workers
have failed to confirm these origins.
The most predominant of these reported features include
several surficial anticlinal and synclinal structures in northwestern
peninsular Florida and southwestern Georgia. Patterson and Herrick
(1971) summarized the reports of these features, which conflicted
with respect to locations and descriptions in many cases, and
concluded that there is no evidence to support the presence of any
surficial structural feature in this area, with one possible exception.
This exception is a small dome in southwestern Georgia, slightly
outside of this study area, known as the Gordon Anticline. The other
reported features were attributed to the differential erosion and
dissolution of the near-surface Tertiary carbonates.
A Tertiary feature that is regularly mentioned in the literature
is the Ocala Uplift, a reported anticlinal structure in central
peninsular Florida. As in the previous instances, a tectonic origin for
this feature was dismissed by Winston (1976), who attributed the
apparent anticlinal structure of the surficial carbonates as having
been caused by differential erosion and dissolution.
The only other purported Tertiary feature is the Bronson
Graben, described as a small northwest-trending structure in north-
central peninsular Florida with an associated series of surficial
normal faults paralleling it. This was proposed by Vernon (1951) on
the basis of lineations in aerial photographs and discrepancies in drill





59

(1989) to extend Appalachian fault provinces along the length of the


plateau.


Since 1951


, there has been no field recognition of these


features, and it is unlikely that Vernon's observations are indicative


of lithospheric movement during the Tertiary.


It appears, then, that


with the possible exception of the Gordon Anticline, there is no
evidence for any Tertiary structure on the Floridan Plateau
attributable to regional tectonic stress.


Summary


A compiled map of the pre-Mesozoic lithotectonic terranes of


the Florida Plateau basement is shown in Figure 17


In northeastern


peninsular Florida, southern Georgia, and the Florida Panhandle, the
subcrops of Precambrian and Paleozoic rocks apparently represent
isolated pieces of more extensive terranes that were fragmented
during the collision and subsequent separation of Laurentia and


Gondwana.


Although these fragments were not significantly


metamorphosed during this process,


there is evidence allowing the


suggestion of thrust faulting, block faulting, and faulted contacts.
Potential-field gradients, lithologic contrasts, and the limited existing
seismic reflection data suggest that all of the contacts between these
pre-Mesozoic terranes may be faulted, with the possible exception of
the boundary between the Osceola Granite and the North Florida
Rhyolite.
Figure 18 represents a compilation of the Mesozoic structural

















U. -- 0 *********n* -ee e- ** -* **



nne-ae a *'\ ** ---- *. at* n C..
... .... ........ ....T rram .. .... .
----- --- N

..... rn e ... ..
lay Fault -- "'"rt" Fla""
Zo"e a..e--* R- h lite

-- .e


..Undated Felsic/Maflc
i.jExtrusive Rocks


'- St. Lucie
Metamorphic
\k Complex


(I-\
4 (
.,At A
's's' ? S

.,,', ,, ~
1^\ 1 1
- \-\9 \
S-- -


Mesozoic Yolcanic


Complex


Rhyolite and Andesite

Felsic and Mafic Extrusive
Rocks

Sandstone and Shale

Felsic Plutonic Rocks


Amphibolite-Grade
Metamorphic Rocks

Indicates Unknown Boundary

Possi ble Faulted Contact


I1F"








62

plateau. Although the linear nature of the fault zone suggests that it
originated as a strike-slip fault, the present structure of the zone is
independently determined by each of the bordering Mesozoic


extensional features and consequently changes along strike.


In the


southeastern region of the plateau, the fault zone primarily consists


of a series of southwestward-dipping normal faults.


To the


northwest along much of its length, the zone alternates between
accommodating significant vertical offsets at southwestward-dipping
Jurassic basin boundaries and having little or no vertical offset


where intersected by basement arches.


In the northwest, the zone


changes to accommodate the northeastward-dipping normal
bounding faults of the South Georgia Rift.
The southwestward extension of the South Georgia Rift into
Florida resulted in block faulting in the western Suwannee Basin
terrane. Although there is some suggestion of intrarift horst and
graben formation, neither these structures nor the lateral master
bounding faults of the rift extension have been definitively located


or characterized.


There is a small ancillary rift basin (or basins)


southeast of the main rift.


As mentioned above, the southwestern


boundary of the South Georgia Rift is a large northeastward-dipping
normal fault at the Jay Fault zone.
Overlying the South Georgia Rift in this area is the Jurassic
Apalachicola Embayment, which, as suggested in previous studies,


may also be bordered by lateral master faults.


The Apalachicola





63

extensional basins bordering the southwestern side of the Jay Fault
zone.


These basins


, the Apalachicola Basin, the Tampa Basin, and the


South Florida Basin, are separated by two wide basement ridges, the


Middle Ground Arch and the Sarasota Arch.


Each of the basins


with


the probable exception of the northwestern edge of the Tampa Basin,


is fault-bounded along its lateral boundaries.


In addition, it is likely


that each extends from the Jay Fault zone southwestward at least as


far as the Florida Escarpment.


Although there is little corroborating


seismic reflection data for the South Florida Basin, potential-field
signatures suggest the probability that this area is underlain by


several isolated basins


, rather than being a singular feature.


basement low under the northeast edge of the Floridan Plateau is
interpreted to be a subsided landward salient of the Jurassic Blake
Plateau Basin.
With the exception of gravity-induced slump features near the
Floridan Escarpment and small faults in the northeastern Gulf


produced by halokinesis, there is no evidence of any


structural feature on the Floridan Plateau.


Tertiary


There may be a small


anticlinal structure, the Gordon Anticline, in southwestern Georgia
just north of this study area.












CHAPTER 3
TECTONIC EVOLUTION OF THE FLORIDAN PLATEAU


Introduction


As a consequence of the depth of burial of the Floridan Plateau
basement and the lithotectonic complexity of the region, past
interpretations of tectonic history have necessarily been based on an


inadequate knowledge of basement structure.


The intention in this


chapter is to provide a reinterpretation of the tectonic history of the
Floridan Plateau within the context of the new interpretations of
lithologic and structural relationships presented in Chapter 2 and to
incorporate the findings of recently published geochemical,
geophysical, and tectonic studies into this interpretation.


Regional Tectonic History


Pre-Mesozoic


Tectonic reconstructions for the continental blocks) underlying
the Floridan Plateau and the adjacent Bahamas Platform are based
primarily on sparse information from deep drill hole samples and, as
a consequence, the origin of these blocks has been the subject of


some debate.


The Suwannee Basin terrane and, by association, the







assemblages to have Gondwanan affinities (e.g., Cramer, 1971


Pojeta


et al.


,1976)


In addition, paleomagnetic data and ages of detrital


zircons (Opdyke et al.,


1987


Mueller et el.


,1993) are also


inconsistent with a Laurentian origin for this terrane.


As a result, a


correlation between these terranes and Gondwana prior to the
Alleghenian closure of the lapetus Ocean has been reasonably well
established; however, the specific position of the Floridan Plateau
with respect to Gondwana during the Paleozoic has not been
definitively established.
A number of investigations have suggested that the pre-
Mesozoic basement terranes of the northeastern Floridan Plateau


originated along Africa's western margin near Senegal.


These


investigations include continental reconstructions (Bullard et al.,


1965; Wilson, 1966


Dietz et al., 1970; LePichon and Fox, 1971


Pindell and Dewey, 1982; Klitgord et al.,
1983; Van Siclen, 1984; Pindell, 1985; V


1983; Roussell and Liger,
enkatakrishnan and Culver,


1988), paleontological analyses from Florida and Africa (Cramer


1971,1973


Pojeta et al.,


1976),


comparisons of isotopic abundances


and ages


(Dallmeyer, 1987, 1988; Dallmeyer et al.,


1987;


Heatherington et al., 1993), paleomagnetic data (Van der Voo et al.,


1976; Opdyke et al., 1987),


and stratigraphic correlations between


Florida and west Africa (Smith, 1982; Chowns and Williams, 1983).
Although a proposed tectonic history for Florida based on any one of
these lines of evidence would be inconclusive, the cumulative effect





66

There are specific units in western Africa that can be
correlated to their apparent counterparts underlying the Floridan


Plateau.


These units provide the sole record of the pre-Mesozoic


tectonic history of the Floridan Plateau.


The extensive North Florida


Rhyolite of north Florida and south Georgia may be correlated to
similar felsic volcanic rocks which are present throughout the
western margin of Africa (Dillon and Sougy, 1974; Dallmeyer, 1987


Dallmeyer and Villeneuve, 1987


Heatherington et al.,


1993).


Geochemical analyses of the Florida suite suggest emplacement in a


convergent plate margin setting,


with an ocean-continent subduction


environment being the most likely setting (Mueller and Porch, 1983).
The St. Lucie Metamorphic Complex has been correlated to
units of the central Rockelide Orogen in Guinea, proximal to the Bove


Basin (Chowns and Williams


,1983; Dallmeyer and Villeneuve, 198


In addition to similarities in petrology and degree of metamorphism,
both of these metamorphic units appear to have been affected by an
extensive thermotectonic event, Pan-African II, that occurred along
the northwestern margin of Gondwana about 550 Ma ago (Dallmeyer


and Villeneuve, 1987)


In addition, there is no evidence that the St.


Lucie Metamorphic Complex was thermally affected by the


subsequent Caledonian or Hercynian orogenies.


This is consistent


with a central Rockelide origin for the complex, but not with a more
northerly origin, where the adjacent Mauritanides were strongly
deformed during the later orogenies (Roussel et al., 1984), nor with a





67

There are a number of similarities that support the correlation
of the Osceola Granitoid Complex of east central Florida with the Coya


Granite of Senegal, which is exposed in the northern Rockelides.


Both


are post-tectonic, Early Paleozoic plutons with similar petrographic


and petrologic characteristics (Dallmeyer et al.,


1987)


. These


similarities


, as well as numerous continental reconstructions, suggest


that during the Early Paleozoic, the Osceola and Coya Granites were
likely to have formed as part of the series of granitoids emplaced
along Gondwana's northwestern margin during Pan-African II
orogenies (Dillon and Sougy, 1974; Dallmeyer et al., 1987).
The Suwannee Basin Complex of north-central Florida was


deposited in an Early Paleozoic sedimentary basin.


The lower portion


of the sampled sequence of this complex consists of quartz
sandstones overlain by graptolite-bearing black shales, which are


indicative of a restricted marine environment.
stratigraphic and paleontological similarities, t


Based largely on
he Suwannee Basin


was correlated by a number of workers to the Bovy Basin of western


Guinea (Cramer, 1971, 1973


and Williams


Pojeta et al., 1976; Smith, 1982; Chowns


,1983; Venkatakrishnan and Culver, 1988).


Subsequently, a paleomagnetically determined Lower Ordovician
paleolatitude of 49 degrees south (Opdyke et al., 1987) and a
comparison of plateau ages for muscovite (ca. 505 Ma) from both
basins (Dallmeyer, 1987) have provided further support for this


correlation.


These deposits probably represent the disjointed





68

Although the present day basal contact of the Suwannee Basin
terrane is likely to be a simple unconformity, there is some evidence
to suggest that it may instead be a northward-dipping thrust fault.
Although the Suwannee Basin has been only mildly metamorphosed,
seismic reflection profiles suggest the presence of low angle faulting


within the terrane (Arden, 1992,


personal communication) and the


basal contact has the appearance of a ramped thrust fault (Arden,
1974). In addition, thrust faults are common in the central
Rockelides, which are in close proximity to the Bove Basin (e.g.,
Williams and Culver, 1982).
Smith (1993) has postulated that the pre-Mesozoic terranes of
the plateau basement, including the Suwannee Basin, the North
Florida Rhyolite, the Osceola Granite, and the St. Lucie Metamorphic
Complex, were fragmented and repositioned to their present day


configurations during the late stages of AUeghenian closure.


In this


scenario, fragments of these Gondwanan terranes were moved


laterally into their relative positions,


with separate fragments being


translocated into eastern Florida, southeastern Georgia, and
northwestern Florida, along a series of left and right lateral strike-


slip faults.


These strike-slip faults,


including the precursor to the Jay


Fault, formed in response to the differential stresses in Gondwana


resulting from a deformational collision with Laurentia.


The process


is demonstrated in Figure 19, which illustrates the formation of
strike-slip faulting in the Florida basement and the relative















Ala. Ga. S.C.


Laurentia






Iipe p4tu$ 0cen






rGondwana .
I s s
* A *L* i *^ ****** *
a aa an a ,-a .1

a - Fea si I t r Usi v"** a- ^a a


r/-- r


SA .a G. S.C.



SLaurentia








a""a""a"a" a"a "". '





*aaa a aaa 5 *a aa a ** A**b*b4
aA A aae aaaa aaa aaaa baa.6
***i.** a*... aa l a'ia a*a
^sM ats 'iTe-isi* WW:iihh *mlitA




A + A A A A A A A ,4k i!'
V*+ A A A A' Al i :0sA l :!i A
e s s ne s s aa'fl'M 1' l'l*!"!-i4!l
A A A A A A A A A A A A A A A A ^*-*~itlp rlt~ll! 'tiEaf
V V ^*"** 4 4 **tl !liIt*!*P'l!* ll *Ealtll i!
M A A A A A A A 4 4ss ei A AiA A AUSA;^ ^ E


Alaurel


Laurel


nt


nt


i S.a C.

ia \


a....aa aaaaa..ab.aa L aL.:
&* *. *ataana aaaaa ara ab *
... **aa ..7... a...a .

:::::::::::::::::::::::::::::::::::::::::::::B:::::


H _______________________________________________________________________


A-.Ga. S-


Laurentia






if
:. *: *. *"JEliM lltfr ilsfS

'e'.t -L A-TAlpA-A

L.. ........ .
t 4 A A A J t i ?i4 lii44ii&-4&il44Ai 44
Laonwanaaa a a Mea ..
A *&* ei Aa*A aa at.
& AAAA A AAA A;- '* '-< fAAAAA *


b a A h* i
MAA~eM AAAAA AAAA A 4 4 M *
AAS6AAAAA AAAA AAA AA A AA
GoAAmAnaA < .
MeA ~ aeg AAMM AA eS a aa a a
Asses.As M~mse ...
AAA~ae AA AAA*A b*A
WwWW AAM4A ..A
AAAAAaA


. I






s, irl a 1



. Plateau ,
..
"i4~ *


__


I





r 11


4 l o





70

There are other lines of evidence that support this scenario. It
is unlikely that the numerous Triassic and Jurassic features of the
plateau basement would share a common linear boundary such as
the Jay Fault zone (Chapter 2), unless this boundary existed as a pre-
existing zone of weakness. The continuity and linearity of this zone
of weakness is strongly suggestive of a long strike-slip fault zone,


which is consistent with Smith's model.


In addition, the


metamorphic Catoche terrane encountered in DSDP holes 150 kms.
west of the Florida Escarpment in the Gulf of Mexico is petrologically
similar to the St. Lucie Metamorphic Complex and displays similar


40Ar/39Ar age spectra (Dallmeyer, 1984)


. Consequently


, it appears


to be another detached fragment of the same Gondwanan source


terrane (Schlager et al.,


1984).


If one accepts a Gondwanan origin for


the Catoche terrane, this necessitates northward to westward net
transport relative to the present orientation of North America for
emplacement of the terrane fragment to the west of the Florida


Plateau.


Because Mesozoic extensional trends suggest only


southeastward transport, then northward to westward movement of
the Catoche terrane must have occurred during or after closure, but
prior to Mesozoic extension in some manner similar to that described
by Smith.
Although significant northward and westward movement along
a series of strike-slip faults is suggested, the amount of lateral offset


along the Jay Fault zone may have been rather limited.


Recent drill





71

Granite and is probably another fragment of the same terrane


(Figure 17)


In addition, Heatherington and Mueller (1991) have


found little evidence for differences in Nd model ages for the
lithospheric components from which the volcanic suite of the South
Florida Basin and the north Florida volcanic rocks were derived.
Each of these suggests that there was not enough net strike-slip
movement along this section of the Jay Fault during the Paleozoic or
Mesozoic to bring separate lithospheric units into juxtaposition.


Mesozoic


Most Late Paleozoic-Early Mesozoic reconstructions of Pangaea
place the continental basements of the Floridan Plateau, the Yucatan,
and the Bahamas Platform in the reentrant at the junctions between


the juxtaposed North American


, South American, and African plates.


With the formation of a Late Triassic interior rift system, North
America began to rift away from Gondwana along the traces of the
Paleozoic continental margins (Van der Voo et al., 1976; Mullins and


Lynts, 1977; Pindell and Dewey
et al., 1984; Van Siclen, 1984).


,1982; Klitgord et al., 1983; Klitgord
The hypothesized sequence of events


on the Floridan Plateau during this period are shown in Figure 20.
A large graben system, the South Georgia Rift, formed across
southern Georgia nearly along the trace of the Alleghenian suture


between Florida and North America (Nelson et al., 1985B).


The


configuration of the rift was largely controlled by the presence of





























































Jurassic


Basin


Jurassic

Triassic


Normal Fault

Normal Fault


Triasic


Continental Rift Basin


Fault Zone (accommodates normal
and lateral motion)


aT Yyi a &lah a _a -v aL a a_ C a. a a 4 a a tie a Ca


T^---- /





73

across the Triassic northern margin of the Yucatan block, as
suggested by the presence of red beds in the Todos Santos Formation
of southern Mexico (Buffler and Sawyer, 1985; Molina-Garza et al.,


1992


geometry


. With a Late Triassic-Early Jurassic shift in the drifting


the South Georgia Rift became an aulacogen and the


Floridan Plateau


, Yucatan, and western Bahamas basements were left


appended to the southeastern margin of North America (Mullins and


Lynts,


1977


Pindell and Dewey, 1982


Smith


1982


Burke et al.,


1984; Klitgord et al.,


1984; Van Siclen, 1984).


Smith (1982) has suggested that this shift in the drifting
geometry was caused by the formation of a Triassic hot spot and
triple junction associated with the opening of the Atlantic Ocean near


the southern tip of Florida.


This model is supported by the ages,


lithologies, and isotopic signatures of the bimodal rhyolitic and


basaltic suite in the South Florida Basin region.


The basalts of this


terrane have yielded 40Ar/39Ar ages of about 190 Ma and trace
elemental analyses have been found to be consistent with a mantle
source, mixing with Precambrian continental lithosphere, and
deposition in an extensional environment (Mueller and Porch, 1983;
Heatherington and Mueller, 1991).
Both the configuration of the Floridan Plateau basement and
recent continental reconstructions suggest that the South Florida
Basin suite is more likely to be associated with the opening of the


Gulf of Mexico, rather than the Atlantic Ocean.


The alternating basins







of the South Florida Basin terrane to the northwest.


In addition, as


suggested by the position of the salient between Africa, North
America, and South America, it is likely that during the Triassic and


Jurassic,


the continental crust of the western Bahamas block was


located immediately to the east of the South Florida Basin terrane


(e.g., Mullins and Lynts,


1977), where it would have effectively


isolated the terrane from the Atlantic oceanic rift zone (Figure 20).
Although these reconstructions suggest that the South Florida Basin
was geographically removed from the Atlantic Ocean and that the
bimodal volcanics were erupted as a consequence of the opening of
the Gulf of Mexico basin, this does not negate a Late Triassic-Early
Jurassic shift in oceanic rifting geometry as a causal mechanism for
the failure of the South Georgia Rift.
A number of Late Triassic reconstructions of the Gulf of Mexico
region indicate that the north-central Gulf region was occupied by
the Yucatan (or Maya) Peninsula block which, as supported by recent
paleomagnetic data (Molina-Garza et al., 1992), later rotated
counterclockwise away from North America during its separation
with South America (e.g., Pindell, 1985; Ross and Scotese, 1988;
Rowley and Pindell, 1989). These reconstructions suffer from a gap
in the northeastern Gulf of Mexico. There have been two general
hypotheses that account for this gap.
One hypothesis is that the gap was occupied by the crustal
blocks now occupying the Florida Straits and the South Florida Basin







Florida during the Late Jurassic.


This displacement would have


occurred along a left-lateral strike-slip fault (the Florida Elbow Fault)
which would necessarily extend across the southwestern quarter of


the Floridan Plateau (Figure 16B)


. The second hypothesis is that any


apparent gaps are accountable entirely to attenuation and subsidence
of continental crust in the eastern gulf during the Jurassic (e.g.,


Buffler and Sawyer, 1985; Dunbar and Sawyer, 1987


Salvador, 1987


Winker and Buffler, 1988).
In contrast to the obvious potential-field signature attributable
to the Jay Fault zone, there is no suggestion of a Florida Elbow Fault
extending across southwestern Florida in either Bouguer anomaly


maps or seismic reflection surveys.


In addition, the Middle Jurassic


basins and arches of the southwestern Floridan Plateau appear to
extend undisturbed across the proposed trace of the Florida Elbow


Fault.


Hence, there is little supporting evidence for this fault or any


similar structure across the southwestern plateau, and the
translocation of the Florida Straits block from the eastern gulf to its
present location by some other path seems unlikely.
Another consideration with respect to the apparent gap is that
the southwestern half of the Floridan Plateau is comprised of
attenuated (i.e., stretched and thinned) continental crust, as
suggested by the prominent extensional basins and the drill holes


intersecting granitic basement (Figure 1).


The depth-to-basement in


the Tampa, Apalachicola, and South Florida Basins demonstrates that





76

and 1.8 (Pindell, 1985). Consequently, the model that attributes an
apparent gap in the northeastern Gulf of Mexico exclusively to the
attenuation and subsidence of continental crust is preferred.
As previously mentioned, restoration of the Jurassic crustal
attenuation of the southwest Floridan Plateau results in the
relocation of the South Florida Basin region into the proximity of the
eastern gulf. Potential-field maps indicate that the South Florida
Basin region is comprised of several separate basins which trend
northwest-southeast, perpendicular to the trend of the other
Mesozoic features of the southern and western Floridan Plateau
(Figure 18). The northwest-southeast orientations of these basins
and their proximity to the Yucatan during the Late Triassic-Early
Jurassic suggests that they may have formed in response to
southwestern extensional stress as the Yucatan block rotated
counterclockwise away from the Floridan Plateau (Figure 20).
Subsequently, as North America and South America continued
to separate, a series of basement horsts and grabens formed around
the periphery of the Gulf of Mexico basin (Pindell and Dewey, 1982;
Burke et al., 1984; Klitgord et al., 1984; Buffler and Sawyer, 1985;
Pindell, 1985; Winker and Buffler, 1988). A few of the horsts in the
southeastern United States, such the Wiggins Arch underlying
southern Alabama and Mississippi, have been proposed to be
stranded blocks left behind by the departing Yucatan block or South
American continent (Smith et al., 1981; Pindell and Dewey, 1982;





77

North American crust (e.g., Pindell, 1985; Ball et al., 1988; Winker
and Buffler, 1988; Dobson and Buffler, 1990; Reitz, 1991).
The Gulf of Mexico existed as a restricted basin from about 165
Ma until about 150 Ma, resulting in the deposition of extensive salt
deposits (Pindell and Dewey, 1982; Salvador, 1987). During this
period, stress in the western Floridan Plateau was reoriented to a
north-south or northwest-southeast extensional regime, as
delineated by the northeast-southwest orientations of the horsts and
salt-containing basins of the southwestern Floridan Plateau. This
shift in the stress regime is likely to have been associated with the
continuing separation of the North American and South American
landmasses.
The mechanisms of relative motion between North and South
America during the Jurassic are not well constrained, partly because
much of the intervening crustal structure was subsequently
destroyed during the eastward migration of the Caribbean plate. It
is probable that the relative motion between the two continents was
accommodated by strike-slip movement along a series of northwest-
southeast or north-south oriented transform faults (Pindell and
Dewey, 1982; Klitgord et al., 1984; Van Siclen, 1984; Pindell, 1985;
Salvador, 1987).
It has been suggested that the Bahamas Fracture Zone/Jay
Fault system acted as an important transform fault in the eastern
Gulf of Mexico during the Jurassic, not only by accommodating







1984)


However, as mentioned, the plateau basement southwest of


the Jay Fault zone appears to be composed entirely of attenuated
continental crust, rather than of stranded individual blocks, as had


been hypothesized.


In addition


, similarities between Nd isotope


signatures of basement samples from north and south Florida imply
little net strike-slip motion along the Jay Fault zone (Heatherington


and Mueller, 1991).


Consequently, while some sinistral motion along


the Jay Fault zone during the Jurassic resulting from crustal
attenuation along its southwestern margin is likely, its proposed
roles as the southern boundary of North American continental crust
and as a major transform boundary seem implausible.
Tensional deformation in the Gulf of Mexico region ceased by


the Late Jurassic (Salvador, 1987)


indicating that the lithotectonic


blocks of the Floridan Plateau had reached their present relative


positions by that time.


A review of existing seismic reflection


profiles reveals no evidence of continued faulting or other
tectonically-induced deformation in the sedimentary column
overlying the Floridan Plateau after the Early Cretaceous, despite the
presence of numerous pre-existing potential zones of weakness.
Thus, as soon as the region became established as a passive margin,
the level of crustal stress on the plateau appears to have diminished
dramatically.





79

of Mexico associated with the collision of Cuba and the Bahamas
during the Early Paleogene has been documented; however, there is
no evidence for any associated tectonic activity or significant
horizontal lithospheric stress on the Floridan Plateau (Pindell, 1985).
Rather, the undisturbed Upper Cretaceous and Tertiary strata on the
plateau document an extended, nearly continuous period of shallow
marine carbonate and plastic deposition which occurred in response
to regional subsidence.












CHAPTER 4
SEISMICITY AND STRESS IN THE
SOUTHEASTERN UNITED STATES

Introduction
Several hypotheses have been developed to explain seismic
activity in the intraplate setting of the eastern United States;
however, specific causal mechanisms and conditions remain
enigmatic. The relative paucity of earthquakes and the tectonic
complexity of the region result in seismic hazard assessments that
are necessarily rather subjective and often the source of debate. An
improved understanding of seismicity and seismic hazard in the
eastern United States is necessary and can best be acquired from the
integration of a variety of observations.
The intention in this chapter is to characterize epicentral
density patterns in the southeastern United States and to provide a
summary of the disparate hypotheses and observations relating to
the seismicity of the region. A further objective is to evaluate
possible causes for the nonuniform distribution of seismicity, with
particular emphasis on the seismotectonic differences between the
active South Carolina-north Georgia seismic zone and the juxtaposed,
yet seismically quiescent south Georgia-Floridan Plateau region.





81

Methods


The characterization of epicentral distribution in the
southeastern United States was accomplished through the integration
of a literature review of historical and instrumental records and the


establishment of a network of digital seismographs in Florida.


seismograph network has been in operation since 1989.


This


In addition,


a review of previously published seismic reflection profiles and an


overview of the tectonic setting,


as discussed in Chapters


2 and 3,


respectively, are used as general indicators of prehistoric seismic
activity for the Floridan Plateau.


Results


Epicentral Distribution in the Southeastern United States
Compilations of historic reports of seismic activity in the
southeastern United States, except for Florida, have been provided


by Bollinger (1973A, 1975)


. In these compilations,


about 800


reported events during the period between 1754 and 1974 are


identified. Of these,


76 reports are from eastern Tennessee, 73 are


from North Carolina, and 451 are from South Carolina, while only 34


are from Georgia and 18 are from Alabama (Figure 21).


Nearly all of


the Georgia and Alabama events occurred in the northern portions of


the states.


The remainder of the reports are from Virginia,


West


Virginia, Maryland, and Kentucky.


Based on this compilation,


Bollinger identified several different seismically active zones in the


















n Florida Historical Events

, Instrumentally Located
Events

+ Historical or Poorly
Located Events


fl* ... n ) I t.ia a a_ 1 41 .a Ia. ai a -


* t 4 4f!L -L *


II II





83

350 kilometers. It covers an area of more than 145,000 square
kilometers and trends perpendicular to the structural grain of the


Appalachians (Bollinger, 1973A,B)


Although epicentral distribution


within this zone is diffuse, it includes Charleston, S.C., where an 1886
event had an estimated body-wave magnitude of 6.8 (Bollinger,


1977


, 1983) and a lengthy aftershock sequence, including more than


300 events during the subsequent 35 year period (Tarr, 1977).
Another zone, the Southern Appalachian seismic zone, extends
from central Alabama northeastward to western Virgina along the
Blue Ridge and Valley and Ridge provinces. It encompasses 162,000
square kilometers and, with respect to epicentral density, is more


active than the South Carolina-Georgia seismic zone.


Eight events


with Modified Mercalli (MM) intensities greater than VI have
occurred in this zone since 1874, including the intensity VIII event
of Giles County, Virginia in 1897.
In addition to the historic reports of seismic activity, epicentral
distribution patterns have also been determined from instrumentally


located events.


During the period from 1977-199 1


the Southeastern


United States Seismological Network (SEUSSN) located 1162 events,
which are also shown in Figure 21 (SEUSSN Contributors, 1992).
These instrumental records generally corroborate the previously
determined epicentral distribution patterns of Bollinger, implying
that the effects of population distribution, local ground response, and
reporting error have not significantly biased the historical record.





84

(1983) identified only 33 reports of earthquakes in Florida; these
reports were later used to construct a seismicity map of the state


(Reagor et al., 1987).


Most of these reports do not represent tectonic


activity in Florida (Smith and Randazzo,


1989; Lord and Smith, 1991).


Rather, many are attributable to events that occurred elsewhere and
others, based on their descriptions, are more likely to have been


blasting, military activities,


or atmospheric phenomena.


The


conclusion of these reviews was that Florida has experienced only 6


possible low-intensity historic events.


A subsequent review of


newspaper reports has revealed that 1 of these 6 possible events (on
21 June 1893) was felt in the Charleston area before being felt in
Florida and, therefore, is not likely to have originated on the Floridan


Plateau.


The remaining 5 possible events are shown on Figure 21.


Instrumental records also support the premise that the
Floridan Plateau, which includes south Georgia, is characterized by an
unusually low level of seismic activity. In fifteen years of operation,
the SEUSSN has never detected an event on the Floridan Plateau. In


addition, during the three years that the state-wide network of

digital seismographs has been in operation, there were no events
located on the plateau and only a single local event was detected.
This was a mb 3.8 event located in the Gulf of Mexico, 340 kilometers


west-southwest of Sarasota, Florida on March 31


1992.


In addition to historical and instrumental records, a review of
existing seismic reflection surveys from around the plateau (Chapter





85

plateau basement appears to have been in a low-seismicity state for
a relatively long period of time.
Consequently, historical records, instrumental records, and
geophysical data each demonstrate a pronounced difference in levels


of seismic activity between proximal regions.


To the north, in the


southern Appalachians and southern Coastal Plain, there is a
moderate to high level of seismic activity, including several historic


large magnitude events.


Immediately to the south, however, there is


an unusually low level of seismic activity, with only occasional low
intensity events. As inferred from Figure 21, the boundary between
these two regions extends approximately east-west across south-
central Georgia and Alabama and is a well-defined example of the
nonuniform nature of seismicity in the southeastern United States.
Several plausible explanations for this nonuniform distribution
of seismicity can be presented, although they are not mutually


exclusive.


The first is that this difference is caused by variability in


the lithospheric stress field, as suggested by Hatcher et al. (1987).
Another possible reason is that there are mechanical differences (i.e.,
lithology or structure) between adjacent zones (Wheeler and


Bollinger, 1984).


This would imply that local seismogenic features


exert primary control over the distribution of seismic activity in the


southeastern United States.


A final possibility is that there is a


systematic variability in the propensities of pre-existing fault zones


to slippage.


This would be attributable to variations in either the





86

Lithospheric Stress in the Southeastern United States
While the stress regime in central and eastern North America
has been established to be primarily northeast-southwest
compressive (Zoback and Zoback, 1980; Zoback, 1992), there have


been a number of contradictory observations.


This suggests the


possibilities of local perturbations in the regional field and/or the
influence of a local stress field. As a result, the sources) of
seismogenic stress in the southeast is still rather speculative.
Stress orientations in the Coastal Plain and southern
Appalachian provinces have been determined from a variety of
sources, such as the orientation of recent faulting (Schafer, 1979;
Zoback and Zoback, 1980; Wentworth and Mergner-Keefer, 1981,
1983), hydrofracturing in deep wells (Zoback and Zoback, 1980), in
situ measurements (Hooker and Johnson, 1969; Zoback et al., 1978),
fault plane solutions (Tarr, 1977; Tarr and Rhea, 1983; Johnston et
al., 1985; Zoback, 1992), borehole elongation (Plumb and Cox, 1987)
and crustal modeling (Kuang et al., 1989; Richardson and Reding,


1991).


Within these provinces, there is general agreement that the


maximum stress is compressional and horizontal.


While most


indicators suggest that the maximum compressional stress direction
is oriented northeast-southwest, many are also indicative of a
northwest-southeast oriented field.
A number of indicators have been used to infer northeast-


southwest compression.


One of these is shallow focal plane solutions





87

addition, Plumb and Cox (1987) measured borehole elongations from
eastern Tennessee to Canada, including a number from west of the
Appalachians, and found nearly their entire study area to be under


northeast-southwest compression.


While Plumb and Cox had no


measurements from the southeastern Coastal Plain, Hooker and
Johnson (1969) measured in situ stresses in 5 boreholes in Georgia,


Virgina, and North Carolina.


Their measurements were inconclusive,


as their orientations ranged from nearly north-south to nearly east-


west.


Focal mechanisms from the southern Appalachians


demonstrate that at depths averaging between 9 and 15 kms, the
basement under the Appalachian detachment is under northeast-


southwest compression (Johnston et al.,


1985).


Using a variety of


indicators, Zoback and Zoback (1989) found the eastern and central
United States to be under a relatively uniform northeast-southwest


compressional field.


More recently, M.L Zoback (1992) found that of


the 32 central and eastern United States earthquakes used in her
investigation, 25 indicated slip compatible with northeast-southwest
compression.
Alternatively, some of the stress field indicators suggest a
component of horizontal compressive stress oriented approximately


northwest-southeast.


Schafer (1979) came to this conclusion while


investigating recent surficial fault motion in the western Valley and
Ridge of eastern Tennessee, as did Wentworth and Mergner-Keefer


(1981


1983) and Prowell (1989) in studies of Cenozoic faulting from





88

characterized the Appalachian and Coastal Plain provinces as being


under northwest-southeast compression.


Finally, Bollinger (1983)


and Wentworth and Mergner-Keefer (1983) both concluded that the
1886 Charleston earthquake was probably a thrust event along a
northeast-trending structure--another indication of northwest-
southeast compression.
Within Florida, stress orientations have been deduced only
from a single report of recent tectonic faulting (Vernon, 1951). In


this account,


Vernon proposes northwest-striking normal faulting in


the west-central portion of the peninsula as a source of aerial photo


lineations and inconsistencies in drill logs.


The report has


subsequently been cited as evidence for northeast-southwest
extensional stress on the Floridan Plateau (York and Oliver, 1976;


Zoback and Zoback, 1980; Prowell, 1983)


No other reports of these


features exist, and no physical recognition of them has been
achieved, although northwest-trending jointing is common in that


area (Beatty


1977)


Accordingly, the orientation of the stress field


under the Floridan Plateau is unknown, although Zoback and Zoback
(1989) indicate that southeastern Georgia and the Atlantic
continental margin off Florida's northeast coast are under north-
northeast to south-southwest compression.
In an attempt to explain the contradictory indicators of stress


direction in the southern Appalachians and Coastal Plain,


York and


Oliver (1976) suggested that seismogenic stresses in the eastern





89

perturbed by the strain release during the 1886 event and has not


yet recovered.


Bollinger (1983) hypothesized that there were two


sets of perpendicular seismogenic features around Charleston, one
northwest-southeast and one northeast-southwest, along which
earthquakes occur without regard to the orientation of the stress


field.


McCartan and Gettings (1991) proposed that young mid-crustal


plutons locally disrupt the stress field by concentrating stress.
Hatcher et al. (1987) suggested that the eastern and central United
States could be divided into a number of individual blocks based on
geophysical signatures and that a southern group of blocks, including
the Floridan Plateau, are decoupled from the regional stress regime.
Although each of these may explain certain observations or
particular local phenomena, none resolves all of the contradictory
observations throughout the eastern United States.
The total stress field at a given site is the sum of regional and


local stresses.


Regional stress, as the name suggests, is that stress


regime which prevails over a large area and is produced by tectonic


effects (e.g., Kuang et al.,


1989)


Local stress is that induced by


surficial topographic relief or shallow crustal features or phenomena.
Remnant stress is that stress remaining in a rock after an applied
stress has been removed (Sbar and Sykes, 1973).
The primary source of regional stress in eastern North America
appears to be ridge-push from the mid-Atlantic ridge (Yang and


Aggarwal, 1979: Zoback and Zoback, 1980; Kuang et al.,


1989; Zoback,





90

resistance to plate motion (Zoback and Zoback, 1980) and lithospheric
stress induced by asthenospheric convective flow (Voight, 1969; Sbar


and Sykes,


1973).


Either of these would also produce a northeast-


southwest maximum compressional stress direction.
The contradictory indicators of stress direction suggest the
possibility of a local stress stress field in the eastern United States.
Several possible sources of local stress have been proposed.
Bollinger (1973B) and Barosh (1981) have suggested isostatic uplift
and flexure as a causal mechanism for compression; however, Brown
(1978) was unable to find a correlation between vertical crustal


movement and seismicity along the eastern seaboard.


In addition


the rebounding Appalachians are the source of the uplift in this
model, then the predicted extensional stress along the crest of the


mountains is not observed (Zoback and Zoback, 1980).


Another


proposed source of local northwest-southeast compressional stress is
gravity-induced backsliding southeastward along the Appalachian


decollement surface (Seeber and Armbruster, 1981).


This model also


predicts extension along the apex of the Appalachians, which is not


observed (Zoback and Zoback, 1980).


Finally, Kuang et al. (1989)


considered the primary sources of local stress to be topographic


relief and density contrasts within the lithosphere.


According to


their model, these would also be likely to induce northwest-
southeast compression.
Seismotectonic Features of the Southeastern United States