This paper was published in the Journal of Geophysical Research, Solid Earth, Vol. 103, no. B12, p. 30,505-30,530, december 10th, 1998, Copyright 1998 by American Geophysical Union.  This paper is also available in a pdf-version.

Structure and tectonics of the upper Cenozoic Puerto Rico-Virgin Islands carbonate platform as determined from seismic reflection studies.

Jean-Paul van Gestel Department of Geological Sciences University of Texas at Austin Austin, TX 78759 jpges@utig.ig.utexas.edu phone: 512-232-3207 fax: 512-471-8844, Paul Mann Institute for Geophysics University of Texas at Austin 4412 Spicewood Springs Road, Bldg. 600 Austin, TX 78759-8500 paulm@utig.ig.utexas.edu phone: 512-471-0452 fax: 512-471-8844, James F. Dolan Department of Earth Sciences University of Southern California Los Angeles, CA 90089-0740 dolan@coda.usc.com phone: 213-740-8599 fax: 213-740-8801, Nancy R. Grindlay Department of Earth Sciences University of North Carolina at Wilmington Wilmington, NC 28403-3297 grindlayn@uncwil.edu phone: 910-962-7421 fax: 910-962-7077.

Abstract.

The Puerto Rico-Virgin Islands carbonate platform was deposited over an area of 18,000 km2 from early Oligocene to Holocene on top of an inactive and subsiding Cretaceous-earliest Oligocene island arc. Regional single-channel and multichannel seismic reflection lines presented in this study provide the first information on the regional stratigraphy and structure of this platform that has previously been known mainly from onshore stratigraphic sections of a relatively small (2250 km2) portion of the platform exposed by late Neogene tectonic uplift along the north coast of Puerto Rico. Seismic reflection lines are used to map the thickness of the carbonate platform strata and to correlate this thickness with onshore outcrop and well data from northern and southern Puerto Rico, St. Croix (U.S. Virgin Islands), and the Saba Bank. Limestone thickness variations from a little over 2 km to almost zero are used to subdivide the Puerto Rico-Virgin Islands platform into five distinct carbonate provinces: (1) north Puerto Rico area including the onshore exposures; (2) Virgin Islands area; (3) St. Croix and Saba Bank area; (4) south Puerto Rico area; and (5) Mona Passage area. Carbonate thickness and structural information from each area are used to test five previously proposed models for the deformation and vertical movements of the platform. The most prominent feature of the platform in the Puerto Rico-Virgin Islands area is a large, east-west trending arch. The northern limb of this arch exhibits a smoother, more uniform dip than the steeper, more abruptly faulted, southern limb. The core of the arch is responsible for the exposure of arc basement rocks on Puerto Rico. The origin of this arch, which occurs over a 300 km wide area, is best explained by north-south shortening and arching, caused by interaction at depth of subducted slabs of the North America and Caribbean plates. Other important evidence for this model can be found in the Benioff zones observed in the earthquake profiles. Loading of the Caribbean plate results in downward flexing of the North America plate and causes the 4 km subsidence of the carbonate platform north of Puerto Rico.

1. Introduction

Carbonate platforms provide useful records of stratigraphic and tectonic events because they are (1) sites of rapid and uniform deposition related to the productive "carbonate factory"; (2) sediments containing numerous organisms that can be dated biostratigraphically; and (3) well stratified, easily imaged using single-channel and multichannel seismic reflection systems and can be used to map tectonically produced faults, folds, and unconformities.

The purpose of this paper is to use seismic reflection data to study the tectonic history of a large Puerto Rico-Virgin Islands carbonate platform in the northeastern Caribbean (Figure 1). Thereby we especially want to focus on the 4_ tilt of the carbonate platform north of Puerto Rico, which has resulted in a huge, 4 km subsidence of the most northern edge of this platform (Figure 2).


(181K)

Figure 1. (a) Present-day plate boundary faults of the Caribbean plate modified from Gordon et al. [1997]. Directions and rates of plate motion relative to the Caribbean plate from DeMets et al. [1996]. EPGFZ is Enriquillo Plantain Garden fault zone. Box shows location of study area (Figure 2). (b) Main tectonic arcs in the Caribbean area, modified from Gordon et al. [1997]. Arrows indicate inferred direction of opening in the Yucatan back arc basin [Rosencrantz, 1990] and the Granada back arc basin [Bird et al., 1993].


(143K)

Figure 2. Map of the study area, showing the bathymetry, contours every kilometer, the extent of the carbonate platform in the different areas, the major geological features, and the locations of our data sources. Data source locations include the track lines of the surveys, the locations of the wells, and the locations of the earthquake profiles. Bathymetry is an integration of EW 96-05 Hydrosweep, ETOPO-5 digital terrain map, and National Ocean Survey hydrographic data sets [Mercado, 1994]. Solid lines are track lines of different surveys. North America-Caribbean plate motion vector according to DeMets et al. [1996]. The light gray area shows the extent of the carbonate platform as observed in seismic reflection profiles and side scan images. The rifts are shown is a darker gray, bounded by normal faults and the darkest gray areas are the islands. PRT, Puerto Rico trench; SPRSFZ, South Puerto Rico Slope fault zone; NPRSFZ, North Puerto Rico Slope fault zone; GNFZ, Great Northern fault zone; GSFZ, Great Southern fault zone.

For this study we make extensive use of single-channel seismic reflection, side scan, bathymetry, gravity, and magnetic data, collected during cruise EW 96-05 on board of R/V Maurice Ewing in which all authors participated (Figure 2). These newly acquired data have been integrated with older available data and correlated to onland outcrops and well information. The advantage of this method is that it provides a regional set of observations on the largely submerged platform sections.
Five models have been previously proposed for the deformation and vertical movements of this carbonate platform (Figure 2). We test these tectonic models and evaluate each model in light of the new data presented in this study.
The following models are reviewed: (1) the transtension hypothesis by Speed and Larue [1991], who proposed that the dominant fault style in the Puerto Rico area was low-angle normal faulting (Figure 3a), (2) the rotating microplate hypothesis by Masson and Scanlon [1991], who proposed a counterclockwise rotation of the Puerto Rico tectonic block within a broad zone of east-west strike-slip motion, which results in extension in the northwest corner of the Puerto Rico-Virgin Islands block (Figure 3b), (3) the pinning and localized extension model by Vogt et al. [1976], who proposed that divergent features might form between the "pinned" (Hispaniola) and "unpinned" (Puerto Rico-Virgin Islands platform) areas in the Mona Passage, caused by oblique collision of the Greater Antilles with the Bahama platform (Figure 3c), (4) north-south shortening and arching hypothesis by Dillon et al. [1994], who proposed interaction of Caribbean and North America plates at depth, which occurred due to north-south crustal convergence (Figure 3d), and (5) large-scale tectonic erosion by McCann and Sykes [1984], who proposed tectonic erosion related to oblique subduction of the southeastern Bahama platform extension and Atlantic fracture zone highs (Figure 4).


(138K)

Figure 3. Simplified drawings of four of the five previously proposed models explaining the subsidence of the carbonate platform: (a) Transtension model [Speed and Larue, 1991]; (b) Rotating microplate model [Masson and Scanlon, 1991]; (c) Pinning and localized extension model [Vogt et al., 1976]; (d) North-south shortening and arching model [Dillon et al., 1994].


(170K)

Figure 4. Simplified drawing of oblique subduction model, where platform is affected by tectonic erosion related to oblique subduction of the southeastern Bahama platform and Atlantic fracture zone highs [McCann and Sykes, 1984]. The movement of the North America plate opposite to the Caribbean plate is shown for three different time periods: (a) middle Miocene (15 Ma), (b) late Miocene (7 Ma), and (c) Holocene. The plate motion vector of DeMets et al. [1996] has been used for calculation.

2. Tectonic and Geologic Setting of the Puerto Rico Area

2.1. Cretaceous-Oligocene Geologic History

Puerto Rico occupies the northeastern segment of a lower Cretaceous to Holocene island arc chain that extends from western Cuba to the north coast of South America (Figure 1b). The volcanics in the northern, or Greater Antilles, segment of the arc that includes the islands of Cuba, Hispaniola, Puerto Rico, and the Virgin Islands (Figure 1a) have been extinct since the collision with the Bahamas carbonate platform in late Paleocene to early Oligocene time. Major pulses of collision were broadly diachronous and occurred in late Paleocene/earliest Eocene in western Cuba [Gordon et al., 1997], early to middle Eocene time in central Cuba [Hempton and Barros, 1993] middle Eocene to Holocene in Hispaniola [Mann et al., 1991] and late Eocene to early Oligocene in Puerto Rico and the Virgin Islands [Dolan et al., 1991].

Because all segments of the Caribbean arc were initiated during the early Cretaceous and because they exhibit lithologic and geochemical similarities, several groups of workers have interpreted the presence of circum-Caribbean island arc rocks. These rocks are considered to be the dismembered remnants of a once continuous volcanic arc chain that ringed a Pacific oceanic plateau, which formed during early Cretaceous time and swept westward into the present-day Caribbean region during latest Cretaceous and early Cenozoic time [Pindell and Barrett, 1990].
The pre-Oligocene geology of Puerto Rico consists of a deeply eroded volcanic arc basement cut by the northwest striking Great Northern and Great Southern fault zones (Figure 2). The eroded arc rocks reach elevations of 2 km above sea level in the central range, or Cordillera Central, of Puerto Rico. The Great Southern fault zone preserves a younger, 110-km-wide strip of middle Eocene sedimentary and volcanic rocks known as the Cerrillos belt [Dolan et al., 1991; Glover, 1971] (Figure 2). Rocks of the Cerrillos belt and adjacent arc rocks exhibit large-scale folds and thrust faults, most of which verge northward.
Analogous folds and faults do not cut Oligocene shallow marine and nearshore rocks of the San Sebastian Formation and Juana Diaz Formation of early Oligocene age that overlie the middle Eocene Cerrillos belt rocks above an angular unconformity [Dolan et al., 1991]. This angular unconformity implies a major late Eocene tectonic event that resulted in uplift of basinal rocks to shallow depths, coincident with cessation of arc activity. This late Eocene uplift event is probably related to the northeast verging folds and thrust faults that cut the Cerrillos belt rocks but not the overlying Oligocene rocks [Dolan et al., 1991].
Our study of Oligocene-Holocene strata allows a simplified and more dependable analysis of the most recent tectonic responses. Such responses are more difficult to decipher when considering more complex regions that have undergone a longer period of activity. Examples of these more complex regions are the Great Southern fault zone [Erickson et al., 1991] or the middle Eocene volcanic rocks (Figure 2).

2.2. Oligocene-Holocene Geologic History and Active Tectonic Features

In northern and western Puerto Rico, the siliciclastic San Sebastian Formation forms the base for the lower Oligocene to lower Pliocene Puerto Rico-Virgin Islands carbonate platform that is the object of this study (Figure 2). The platform covers an area of 18,000 km2 and extends from the eastern Dominican Republic on the island of Hispaniola, west of Puerto Rico, to the Virgin Islands, east of Puerto Rico (Figure 2). The continuity and similarity of facies across the platform indicate a remarkable stability over this area for a period of almost 35 million years. Where onshore platform rocks have been studied in detail in northern Puerto Rico [Monroe, 1980] and southern Puerto Rico [Frost et al., 1983], they indicate deposition at sea level with minor periods of subsidence in the early Pliocene.

The platform overlies the Puerto Rico-Hispaniola microplate defined originally by Byrne et al. [1985] on the basis of earthquake data and refined by Masson and Scanlon [1991] using marine geophysical side scan surveys of offshore areas. On the regional bathymetric map the carbonate cap of platform rocks forms a smooth, gently dipping surface that contrasts with the rougher surfaces formed by mainly deformed siliciclastic rocks in deeper water areas (Plate 1).
To the north, the microplate is bounded by the Puerto Rico trench, which accommodates both strike-slip and thrust motion between the Caribbean and North America plates [Masson and Scanlon, 1991; Larue and Ryan, 1998; Grindlay et al., 1997] (Figure 2 and Plate 1). The direction of plate motion is approximately east-west with a small component of southwest convergence [DeMets et al., 1996] (Figure 2). The Puerto Rico-Virgin Islands platform does not extend to the trench but forms a north dipping surface north of its outcrop on Puerto Rico to a latitude of about 19¡N [Moussa et al., 1987].


(853K)

Plate 1. Bathymetric map with 500 m contour interval, depth in kilometers, based on the same compilation of bathymetric data described in the caption of Figure 2. Illumination is from the NNE. The boundaries of the Puerto Rico-Virgin Islands block, the Puerto Rico trench, the Muertos Trough, the Mona Passage, and the Anegada Passage all have a clear bathymetric expression.

To the west of Puerto Rico, the carbonate platform extends in a westward direction across the Mona Passage and onto the island of Hispaniola, where it has not been studied in detail. Bathymetric deeps in the Mona Passage correspond to rifts that locally extend and fragment the platform between Puerto Rico and Hispaniola and probably account for the subsidence that formed the marine pass in this area (Figure 2).
To the east of Puerto Rico, the platform trends to the east-northeast and covers most of the area of the American, British, and Dutch Virgin Islands. A small remnant of the platform, or its deeper water facies, exists in St. Croix (U.S. Virgin Islands) to the southeast of Puerto Rico [Gill, 1989]. This shelf area on St. Croix is in turn separated from the Miocene to Holocene Saba Bank (Figure 2) formed above the Aves ridge, or remnant arc of the Lesser Antilles (Figure 1b). The eastern and southeastern edge of the platform and the Puerto Rico-Hispaniola microplate are sharply bounded by the Anegada fault zone. This fault zone runs trough the deepwater Anegada Passage between Puerto Rico and St. Croix [Jany et al., 1990] and connects the Sombrero and Virgin Islands basins (Figure 2). These two basins are fault-bounded deeps that Jany et al. [1990] interpreted as pull-apart basins along the right-lateral Anegada fault zone. The southern edge of the Puerto Rico-Hispaniola microplate is defined by the Muertos Trough where the Caribbean plate is subducted beneath the microplate [Byrne et al., 1985; Ladd et al., 1990].
The continuation of the Anegada fault zone west of the Virgin Islands basin is controversial. Jany et al. [1990] connected the Anegada and Great Southern fault zones of Puerto Rico. In contrast, Masson and Scanlon [1991] continued the Anegada fault zone southwest to the Muertos Trough, which is the interpretation preferred in this paper. The projection of the Anegada fault zone encounters the Muertos Trough in the area where it appears to accommodate less thrust motion. Profiles of earthquake hypocenters through Puerto Rico show the presence of two Benioff zones descending from the surface traces of the Puerto Rico trench to the north and the Muertos Trough to the south of the island [McCann and Sykes, 1984; Dillon et al., 1994] (Figure 5). The subducting slab associated with the Puerto Rico trench has a steeper dip than the one associated with the Muertos Trough. The northern limb of the Puerto Rico-Virgin Islands platform above the northern slab has a gentle trenchward dip (about 5¡) and exhibits less complex deformation than the southern limb of the platform in southern Puerto Rico, which exhibits a steeper troughward dip (about 10¡) and more complex deformation that is perhaps related to movement along the Anegada and Great Southern fault zones [Jany et al., 1990] (Figure 2).


(109K)

Figure 5. Four north-south profiles at different longitudes with bathymetry and earthquake distribution. Locations of sections are shown in Figure 2. Note the asymmetric arching on both sides of the island arc and the Benioff zones from the North America and the Caribbean subducting plates. Darker gray area indicates the Puerto Rico-Virgin Islands carbonate platform, which is short and more faulted on the south side, and more extensive, but less faulted on the north side.

3. Methods

The main data set used in this study was collected by the authors on the EW 96-05 cruise of the R/V Maurice Ewing in May and June 1996 (Figure 2). Data types included Hawaii MR1 side scan sonar and Hydrosweep bathymetry [Grindlay et al., 1997], magnetics [Muszala et al., 1997], gravity data, and single-channel seismic reflection data (Table 1). The 36 mostly NNE-SSW oriented ship tracks provided ~5600 km of single-channel seismic reflection data, which covered most of the north coast area of Puerto Rico (Figure 2). Data were also collected in the Mona Passage area, but the seismic reflection coverage here was less dense than in the north coast area of Puerto Rico. To improve coverage of the Mona Passage, Virgin Islands, south coast of Puerto Rico, and St. Croix areas of the Puerto Rico-Virgin Islands platform, additional, older multichannel lines were used (Figure 2 and Table 1). These lines were collected by University of Texas Institute for Geophysics (UTIG) and Gulf Oil Company (Gulf) and were archived at UTIG. Because of a lack of information on data acquisition parameters, no further processing was attempted on the data of these older cruises.

The Gulf seismic reflection data set was collected by Gulf Research and Development Company in 1975 (Table 1). The Gulf lines were important for adding coverage to the complexly deformed southern margin of Puerto Rico as well as in the St. Croix, Virgin Islands, and Saba Bank areas (Figure 2).

Data parameters

R/V Maurice Ewing 96-05

UTIG R/V Fred H. Moore

Gulf Oil Company

Recording year

1996

1980

1975

Source

six air gun array

four air gun array

na

Sampling interval

2 ms

4 ms

4 ms

Shotpoint interval

43 m

72 m

na

Number of channels

single-channel

24

multichannel

Near trace

207 m

315 m

na

Far trace

207 m

3535 m

na

Processing steps

resample

demultiplex

further processing na

band pass filter

sort

gain modification

velocity analysis

deconvolution

filter / mute / scale

band pass filter

translate to reel

water velocity migration

further processing na

amplitude gain control

final result after processing

migrated data

stacked data

stacked data

Table 1. Data parameters for seismic reflection data sets as used in this study.

The UTIG data were acquired by the R/V Fred H. Moore in 1980 as part of the industry-funded "Caribbean Tectonics" project and include a total of about 1000 km of multichannel data (Table 1). While parts of these lines have been used in previous publications by Larue and Ryan [1991, 1998], most of these lines remain unpublished. Of particular interest for this study was UTIG line VB, which crosses the arch of the platform completely in the Mona Passage area (Figure 2).
The published multichannel seismic reflection line of Meyerhoff et al. [1983] has also been incorporated in this study. Interpretations from this line are important as this line provides constraints on the nature of faults and the thickness of the upper Eocene unit between the Eocene arc basement and the overlying carbonate platform in the north coast area.
Two sources of velocity data for the study area were considered in the estimate of an average velocity value for the rocks of the Puerto Rico-Virgin Islands platform: (1) Western Geophysical Company [1974] calculated an average velocity of 2.7 km/s for the carbonate platform in relation to their study of the multichannel seismic data set from which line T-111D [Meyerhoff et al., 1983] was derived and (2) Anderson [1991] calculated a velocity of 2.93 km/s for the carbonate platform using seismic reflection data collected over the Toa Baja well. For this study, we averaged these two values and use a value of 2.75 km/s to convert two-way travel time data to thickness data from the platform carbonates.

4. Description of Seismic Reflection Data From the Puerto Rico-Virgin Islands Carbonate Platform

4.1. Five Areas of the Platform

As shown on Figure 6, the platform is divided into five areas based mainly on thickness variations of the platform strata in these areas. Figure 6 is an isopach map of the thickness of the platform limestones based on lines with seismic reflection data shown in Figure 2. In these seismic reflection profiles, the change from fine layered reflections to chaotic seismic characteristics was interpreted as the basement of the early Oligocene-early Pliocene carbonate platform. This basement has been identified in all seismic lines, and the thickness of the overlying carbonate strata has been calculated using the velocity of 2.75 km/s. In the areas with well control, this correlation has been justified. In the Puerto Rico north coast basin the same uniform 4_ dip can be found both in onshore outcrop information and offshore seismic reflection profiles. For each of the five areas, the structure, stratigraphy and depocenters will be described. In addition, the stratigraphy of the Saba Bank to the southeast of the Puerto Rico-Virgin Islands platform (Figure 2) is described. Key seismic reflection lines with the main features of each of these areas are also shown. The results of these descriptions form the basis for testing the five tectonic models shown in Figures 3a-3d and 4.


 (217K)

Figure 6. Isopach map of the Puerto Rico-Virgin Islands carbonate platform with 100 m contour interval, thickness in meters. The change from a highly stratified sequence to a chaotic sequence has been interpreted in all seismic reflection profiles shown in Figure 2 as the basement of the carbonate platform. From the interpretation of this basement, the thickness of the carbonate platform has been calculated using a general velocity of 2750 km/s. Boundaries of the carbonate platform are based on interpretation done in side scan and seismic reflection data. Major depocenters include the North Coast basin and the area west of the Mona rift.

4.2. Puerto Rico North Coast Area

4.2.1. Boundaries.

This area is truncated on its northern area by the shelf break scarp that is located about 50 km off the north shore of Puerto Rico [Schwab et al., 1991; Masson and Scanlon, 1991] (Figure 7). On the east side it is bounded by the San Juan arch, which deforms both the basement and the overlying lower part of the carbonate platform. The San Juan arch extends onto the Puerto Rico margin and is named and described by Larue et al. [1998] using the Western Geophysical data set in the onshore and coastal area. This structural arch is responsible for the thin stratigraphic section seen in the Toa Baja well [Larue, 1991] and may account for the more narrow outcrop pattern of the onshore platform in the area of the city of San Juan. East of the San Juan arch, the platform carbonates decreases considerably in the Virgin Islands area, and its character differs enough to make correlations to the onshore formations of Puerto Rico difficult to impossible (Figure 6). The southern boundary of the depositional north coast basin is located on the island of Puerto Rico, where it forms an unconformity with the underlying Eocene arc rocks as seen on the compilation of onshore data by Monroe [1980] (Figure 6).


 (368K)

Figure 7. (a) EW 96-05 line 23, in the area of the carbonate platform, showing small normal faults formed by uplift in the Guajataca arch. (b) Interpretation of EW 96-05 line 23. Inset shows location of EW 96-05 line 23.

The Puerto Rico-Virgin Islands platform in the Puerto Rico north coast area is bounded on its western side by the Mona rift and the Guajataca arch, which is observed in both arc basement and the lower and upper parts of the carbonate platform (Figure 7). The Guajataca arch extends onto the Puerto Rico margin and is named and described by Larue et al. [1998] using the Western Geophysical data set in the onshore and coastal area. On EW96-05 line 23 (Figure 7), which approaches the north-south Mona rift at a slight angle, the lower, Oligocene-Miocene part of the platform deposits can be seen thinning and onlapping onto the Guajataca arch (Figure 7). The dip of the platform strata off the flank of the Guajataca arch is steeper (6-7¡) than the average amount of dip of other areas in the north Puerto Rico area of the platform (4¡) (Figure 8).

4.2.2. Stratigraphy.

Detailed lithologic and age correlations of the seismic reflection sequences seen in the platform section of EW96-05 line 4 (Figure 8) with onshore wells and outcrops are given by Larue et al. [1998] and Moussa et al. [1987] (Figure 9). They analyzed the two wells drilled in this province, the CPR-4 well and the Toa Baja well (Figure 2), and have shown good correlation with the available seismic reflection data and the outcrop information. The outcrop data show a 4¡ dipping package of lateral homogeneous carbonate layers, unconformably overlying Cretaceous-Eocene arc basement rocks (Figure 8).


 (241K)

Figure 8. (a) To the south, a cross section of the carbonate platform onshore of Puerto Rico based on interpretation of outcrops from Monroe [1980], to the north, offshore EW 96-05 line 4, in the area of the carbonate platform. (b) Interpretation of EW 96-05 line 4. An average velocity of 2750 m/s was used for the carbonate platform. The platform limestones are very consistent in thickness and 4û northward dip. Inset shows location of EW 96-05 line 4.


(75K)

Figure 9. Stratigraphic information of the Oligocene to Pleistocene carbonate strata in the different areas. Data compiled from Monroe [1980], Nemec [1980], and Gill [1989]. Eustatic sea level curve from Haq et al. [1987].

This same dip can be found in the seismic reflections profiles offshore north of Puerto Rico. In these lines a strong reflection can be recognized which forms the boundary between a sequence with high frequency, and laterally very continuous reflectors and a sequence with a much lower frequency content (Figure 8). This reflection can be extrapolated to the onshore area and shows good correlation with the unconformity, which forms the base of the carbonate platform formations. This is shown in Figure 8, where a cross section based on onland outcrops to the left and our seismic line 4 to the right has been displayed. This reflector is easily identifiable in the seismic reflection profiles in this province and is interpreted as the base of the carbonate platform.

4.2.3. Depocenter.

The major depocenter of the north coast area is called the North Coast basin and is localized in a semicircular area bounded by the Guajataca and San Juan arches (Figure 6). The platform strata thicken to the north into the North Coast basin, where the maximum thickness reaches about 1500 m near the northern edge of the platform. The North Coast basin depocenter is roughly 155 by 60 km in size and overlies the large amphitheater-shaped erosional scarp in the carbonate section described by Schwab et al. [1991] in the shelf break area. They speculated that basement faults mapped by Meyerhoff et al. [1983] using the Western Geophysical lines to the west may be responsible for localizing the present position of the carbonate margin of the shelf break. However, in the carbonate margin section no faults parallel to the shelf margin break are present.

4.2.4. Structural features.

Except for the northern scarp margin, which is characterized by mass movement [Schwab et al., 1991], and the two arches, the northern area of the platform is remarkably free of any faults or folds. The only group of faults is seen in EW96-05 line 23 crossing the eastern part of the Guajataca arch (Figure 7). Five small normal faults with displacements of 25-33 m are present on the eastern slope of the arch. These faults can be traced onto the adjacent EW96-05 line 22, and the true dip direction of the faults appears to be to the southwest. The faults occur in all the formations of the carbonate platform, which indicates that they are younger than the uppermost unit of the carbonate platform, which is early Pliocene according to Moussa et al. [1987]. Lineaments seen in the side scan images reveal that these faults are part of a group of seafloor faults that are associated with and subparallel to the northwest trending Guajataca arch. Their expression on the seafloor suggests that these faults may be active.

North of Puerto Rico, several channels are cut in the carbonate platform, along which sediment is transported from onshore into the Puerto Rico trench [Masson and Scanlon, 1991; Grindlay et al., 1997]. Some of these channels may be related to north-south trending faults in the carbonate platform. However, in the few parts of east-west trending seismic reflection lines we collected, the channels show little subsurface displacement. Furthermore, the small displacements that are observed may be related to a pull-down velocity effect between the water-filled channel and the adjacent high-velocity carbonate rocks.

4.3. Virgin Islands Area

4.3.1. Boundaries.

The Puerto Rico-Virgin Islands platform in the Virgin Islands area is bounded to the west by the San Juan arch and to the north by the abrupt erosional shelf margin about 50 km offshore (Figure 6). In the east, the platform continues as far as the island of Anegada to the northeast of the Virgin Islands (Figure 2). To the south the platform is bounded by the Anegada fault zone, which runs from the Sombrero basin through the Anegada Passage to the Virgin Islands basin (Figures 2 and 6). There are no outcrops of the carbonate platform rocks in the Virgin Islands, where only older arc basement rocks are exposed.

4.3.2. Stratigraphy.

In this province the outcrops of the Virgin Islands mainly consist of arc basement rocks. The seismic reflection profiles in this area show an upper sequence, similar to the upper sequence in the Puerto Rico north coast area, and a similar transition from high-frequency layered reflections to more chaotic basement reflectors. We have correlated the upper sequence in this area to the middle Oligocene-early Pliocene carbonate platform, found in the Puerto Rico north coast area, as seen on EW96-05 line 12 (Figure 10).


 (642K)

Figure 10. (a) EW 96-05 line 12 showing a representative part of the Puerto Rico-Virgin Islands carbonate platform in the north Virgin Islands area, where carbonate platform strata are thinner. This thinner limestone is immediately underlain by basement and shows small normal faults. (b) Interpretation of EW 96-05 line 12. Inset shows location of EW 96-05 line 12.

4.3.3. Platform in the northern Virgin Islands.

The 1100 m maximum thickness of strata in the northern area of the Virgin Islands platform is significantly less than the 1500 m thickness of strata seen in the main depocenter of the north Puerto Rico area (Figure 6).

Both the north coast and Virgin Islands areas exhibit an average dip of 4¡ for the carbonate section. Reflectors in the northern end of EW96-05 line 12 show a strong onlap onto basement (Figure 10). Moreover, the upper part of the section appears eroded, possibly by bottom currents in this deeper water area. Both factors appear to contribute to the variations in thickness of the platform strata in the Virgin Islands area.

4.3.4. Platform in the southern Virgin Islands.

In the area between the Virgin Islands and Puerto Rico, all dipping strata of the carbonate platform appear to have been eroded. A younger, horizontal platform of presumably latest Neogene age has been deposited on top of it and forms the present-day seafloor with an average depth of less than 200 m (Gulf lines LS50 to LS52 in Figure 11). Oligocene and older arc-related rocks of the Virgin Islands basement protrude through both the older gently dipping carbonate platform and the younger horizontal platform to form the island areas. On Figure 11, the northern edge of the carbonate platform shows the characteristic smooth 4¡ dip to the north.


 (411K)

Figure 11. Gulf lines LS50 to LS52, showing the arching of the Puerto Rico-Virgin Islands carbonate platform seen in the Virgin Islands area. There is a smooth northward dipping slope in the north, a flat eroded area in the middle, and a steeper dipping slope in the south. Notice the two unconformities in the north of the area. Both the middle Eocene-early Oligocene sediments and the early Oligocene-early Pliocene platform are truncated by angular unconformities related to arching. (b) Interpretation of Gulf lines LS50 to LS52. Inset shows location of Gulf lines LS50 to LS52.

On the south side of the Virgin Islands area adjacent to the Anegada Passage, the shelf-slope transition is more abrupt than in the north and is probably controlled by faults parallel to the Anegada fault zone, which runs from the Sombrero basin through the Anegada Passage to the Virgin Islands basin (Figure 2, and Gulf line LS50 in Figure 11). The ~400 m total thickness of the carbonate platform limestones is thinner than in the northern area, where the total thickness is up to 1000 m.

4.3.5. Arching.

In summary, the overall structure of the platform in the Virgin Islands area is a large, east-west trending arch, which we name the Puerto Rico-Virgin Islands arch and is shown in map view on Figure 3d. The northern limb of the arch exhibits a smoother, more uniform dip than the steeper, more abruptly faulted, southern limb. The core of the arch is responsible for the exposure of arc basement rocks in the Virgin Islands and Puerto Rico. The horizontal nature of the uppermost carbonate unit forming the shallow modern platform suggests that arching became inactive in late Neogene time.

4.3.6. Structure.

There are several normal faults affecting both the northern and southern sides of the Virgin Islands part of the Puerto Rico-Virgin Islands platform. On the north side, the faults could not be followed between lines and no fault exhibited a throw greater than 50 m (Figure 11). On the south side of the area, the throws are larger (Gulf line LS50 in Figure 11), but the coverage and quality of the seismic reflection profiles are not good enough to interpret their lateral continuity. However, on the basis of the bathymetry data shown in Plate 1, the faults probably trend east-northeast, parallel to the steep slope of the Anegada Passage.

4.4. St. Croix Area

4.4.1. Boundaries.

The boundaries of this province are the Virgin Island basin in the Anegada Passage to the north and west of the island and the Muertos Trough to the south (Figure 6). Gulf line LS-49 in Figure 12 shows the regional structure of carbonate rocks in the vicinity of St. Croix and the Saba Bank. The line shows a southeast dipping, deformed carbonate cap on the St. Croix basement block known from field studies to consist of the same island arc basement as exposed in Puerto Rico and the Virgin Islands. A fault-bounded basin filled with deeper water, and probably siliciclastic sediments separates the St. Croix area from Saba Bank.

4.4.2. Stratigraphy.

On Saba Bank, drilling at the Saba-1 well showed an Oligocene-lower Miocene volcaniclastic section conformably overlain by a 1480-m-thick, horizontal carbonate platform of middle Miocene and younger age (Figure 9). Field studies in St. Croix, summarized by Gill [1989], document the history of carbonate rocks in that area. A northeast striking rift, the Kingshill basin, formed in basement arc rocks of St. Croix in middle Miocene time (Figure 9). Deep-water carbonate facies were deposited in water depths of 600 m. Carbonate deposition continued to form a total carbonate thickness of 180 m in the Kingshill basin through the early Pliocene and included shallow water reef and carbonate bank material suggestive of a nearby platform margin. Gill [1989] speculates that this platform area might have been formed locally near the present St. Croix block or was derived from the erosion of the Puerto Rico carbonate platform. They speculated that the latter scenario would be possible if the deep-water areas of the Anegada Passage were previously closed and later horizontally translated in a right-lateral manner with movement along the Anegada fault zone, during which the Virgin Islands basin was formed.

4.4.3. Structure.

The Saba Bank shows no evidence of Miocene rift formation and appears to mark a stable area of the Caribbean plate removed from the tectonic events affecting St. Croix and Puerto Rico. On the Saba Bank, volcaniclastic deposition continued from the Oligocene through early Miocene and is presumably related to the proximity of the Lesser Antilles volcanic arc [Nemec, 1980]. By middle Miocene time, a carbonate bank was established and continues to the present day. It is possible that the St. Croix area formed a deeper water margin of the Saba Bank prior to the faulting event that formed the deep-water area that now separates them (Figure 12).


 (330K)

Figure 12. (a) Gulf line LS49, showing the carbonate platform, overlying the island arc basement on top of the St. Croix ridge and the edge of the Saba Bank area. (b) Interpretation of Gulf line LS49. Inset shows location of Gulf line LS49.

4.5. Puerto Rico South Coast Area

4.5.1. Boundaries.

On the north, this area is bounded by the edge of the onshore carbonate outcrops described in detail by Frost et al. [1983]. They correlate this carbonate platform section to the one described by Monroe [1980] and Moussa et al. [1987] in northern Puerto Rico. The boundaries are chosen at the eastern and western edges of the island of Puerto Rico. However, the eastern side of the south coast area is similar to the southern part of the Virgin Island province, where the carbonates have the same thickness and structure, and consequently, the boundary between the two provinces is not well defined. The southern boundary is defined by the abrupt southern margin of Puerto Rico (Figure 6).

4.5.2. Stratigraphy.

The carbonate outcrops in southern Puerto Rico are about the same age as the carbonate units in northern Puerto Rico. Two formations have been recognized: the Ponce and the Juana Diaz Formations, which are early Oligocene to early Miocene age and middle to late Miocene in age respectively [Monroe, 1980] (Figure 9).

4.5.3. Structure.

On Gulf lines LS24 to LS29 (Figure 13), the thickness of the southern area platform strata is variable. Major faults are indicated by the large amount of relief in the top of the basement surface. Relief in the overlying carbonate section also may be in part related to erosion along several major canyons draining the southern coast of Puerto Rico. Abundant faulting was also found in the outcrops on southern Puerto Rico [Monroe, 1980], where faults with a displacement of as much as 200 m can be found. Correlation between various outcrops is difficult due to this faulting.

The area consists of a carbonate section with a maximum thickness of 500 m (Figure 6). The general dip of the section is southward toward the Muertos Trough. The slope is much steeper (10û) than the north coast area and shows more faulting and folding both in outcrops [Frost et al., 1983] and in Gulf lines LS24 to LS29 shown in Figure 13. The thickest part of the platform is in the middle of the south coast area about at longitude 66û45'. This area coincides with the location of the onshore outcrops (Figure 6). Three exploration wells were drilled through the south coast section and show 395 to 723 m of lower Oligocene to lower Miocene shallow water carbonate facies. Correlation between the wells was impossible because of intense faulting between the closely drilled wells.


(365K)

Figure 13. (a) Gulf lines LS24 to LS29 showing the irregularity of the thickness of the carbonate strata and the faulting in the basement of the Puerto Rico-Virgin Islands carbonate platform in the south Puerto Rico area. (b) Interpretation of Gulf lines LS24 to LS29. Inset shows location of Gulf lines LS24 to LS29.

4.6. Mona Passage Area

4.6.1. Boundaries.

This area is bounded by Hispaniola to the west and Puerto Rico to the east (Figure 6). On the north the carbonate platform extends to a water depth of 2600 m. The southern edge is poorly constrained due to lack of data and structural complexity in the form of two north-south trending rift basins, the Yuma and Cabo Rojo rifts (Figure 6). These basins occur in deep water beyond the shelf break and are the sites of rapid terrigenous sedimentation from river systems both in western Puerto Rico and eastern Hispaniola.

4.6.2. Stratigraphy.

In this province, no outcrops are present, and no wells have been drilled. The information about the age of the carbonate platform is consequently limited. We assume that the age differences on both sides of the Mona rift are minimal and the platform has the same age as in the Puerto Rico north coast basin.

4.6.3. Structure.

The overall structure of this area is a large arch in the carbonate platform with gently dipping north and south flanks superimposed by mainly north striking, but also northwest-southeast oriented normal faults (Figures 14 and 15). The 120 km wavelength of the arch (Figure 15) is similar to the 140 km wide arch seen in the Virgin Islands area (Figure 11). Here the arch is more symmetrical, rather than being faulted on the south side and structurally undeformed on the north side as in the Virgin Islands area (Figure 11). The least faulted part is in the center of the arch, and more faults are present on the steeper-dipping flanks (Figures 14 and 15). All lines show that the steeper-dipping and deeper-water parts of the platform are breaking off in large blocks, ~5 km in width (EW96-05 lines 29 and 30 on Figure 16). This suggests that gravity may play some role in the normal faulting observed in the area. However, other observed faults trend at a high angle to the slope and would not be favorably oriented for gravitationally induced sliding.


 (417K)

Figure 14. (a) EW 96-05 line 35, which extends from the Yuma rift in the southwest to the Mona rift to the northeast, across an area of regional divergence. The roughly north-south normal faults are well displayed in this line, with an approximate east-west trend. (b) Interpretation of EW 96-05 line 35. Inset shows location of EW 96-05 line 35. Dashed line shows seafloor multiple.


(388K)

Figure 15. (a) UTIG north-south line VB in the Mona Passage area, where a regional arch is present and normal faults modify the arch along its northern and southern flanks. Similar normal faults are present on the north flank of the arch as seen in lines 29 and 30 in Figure 16. Interpretation of UTIG line VB. Inset shows location of UTIG line VB.


(297K)

Figure 16. (a) EW 96-05 line 29 and (b) EW 96-05 line 30 showing the faulting of the Puerto Rico -Virgin Islands carbonate platform in the north of the Mona Passage area. Normal faulting is more intense in this area than any other areas of the Puerto Rico-Virgin Islands platform. (c) Interpretation of EW 96-05 line 29 and (d) of EW 96-05 line 30. Inset shows location of EW 96-05 lines 29 and 30.

4.6.4. History of arching.

As in the case of the Virgin Islands arch, there is evidence for two periods of platform growth, with an older, steeply dipping platform, separated by angular unconformity with a younger, more horizontal platform (Figures 14 and 15). The thickness of the platform limestones in the Mona Passage area averages about 1500 m (Figure 6). In the northern part of the platform eastward thickening can be observed (Figures 14 and 15). Because of the intervening Mona rift, the Mona Passage area platform cannot be correlated to the north coast of Puerto Rico. However, it appears that both platform accumulations rest on the middle Oligocene San Sebastian Formation. Because all faults are observed cutting through the entire section, and faults control topographically depressed rifts such as the Yuma, Mona, and Cabo Rojo rifts, faulting is assumed to be young, perhaps as young as post-early Pliocene.

5. Discussion

5.1. Summary of major results of this study

The previous description of data leads to the following main conclusions, which can be compared to the models shown in Figures 3a-3d and Figure 4:

1. The Puerto Rico-Virgin Islands platform limestones accumulated during middle Oligocene to early Pliocene in the north coast area and possibly are of that age also in the Virgin Islands, St. Croix, and Mona Passage areas. The Saba Bank is a middle Miocene to Holocene platform and marks a stable area of the Caribbean plate that is unaffected by the plate boundary events seen in the areas to the north near Puerto Rico.
2. The north coast of Puerto Rico and north Virgin Islands areas are remarkably free of folds and faults. Two arches are present in the north coast area: (1) the northwest trending Guajataca arch forms the western margin of the basin and (2) the northeast trending San Juan arch divides the distinctive carbonate stratigraphy of the north coast and Virgin Islands areas. Onlap relations on the Guajataca arch suggest that it has shown activity possibly extending from the middle Oligocene through middle Miocene. The origin of this activity might be related to the rifting, which has occurred in the nearby located Mona rift. The tectonic origin of the San Juan arch is not clear. The major depocenter of the middle Oligocene-lower Pliocene carbonate platform is localized between the two arches (Figure 6).
3. The dominant feature of the Puerto Rico-Virgin Islands platform is the large east-west trending arch in the platform that extends from the Mona Passage across the Cordillera Central of Puerto Rico and through the Virgin Islands (Figures 2 and 3d). This arch accounts for the regional dips to the north and south of the platform and the elevation of the axis in the Mona Passage, Puerto Rico and the Virgin Islands (Figure 6). The age of arching appears to be long-lived, as two unconformable carbonate sections are observed both in the Mona Passage (Figures 14 and 15) and the Virgin Islands (Figure 11). The older carbonate section exhibits steeper dips, and is overlain by a younger section exhibiting lower or horizontal dips. On the basis of the known age of the platform in the north coast area, we can speculate that the arching continued from post-Eocene time to some time in the late Neogene. In the Virgin Islands the arching appears to have ceased because the upper carbonate unit is horizontal. The southern limb of the arch in the south coast area and adjacent to the Anegada Passage is steeply dipping, and may reflect movement along the Anegada fault zone.

5.2. Comparison of Observations with Previously Proposed Tectonic Models

The main predictions of five previously proposed models on the tectonics and geologic history of the Puerto Rico-Hispaniola microplate are reviewed below (Figures 3a-3d and 4). Each model will be evaluated in light of the new data presented in this study, which covers most of the Puerto Rico-Virgin Islands carbonate platform (Figure 2).

5.2.1. Transtension Hypothesis: Speed and Larue [1991]

Speed and Larue [1991] proposed that the dominant fault style in the Puerto Rico area was low-angle normal faulting that accommodated divergence in a west-northwest direction (Figure 3A). The major normal faults included the 19¡ fault of Larue and Ryan [1991] along the shelf edge of northern Puerto Rico, the Puerto Rico trench, and normal faults bounding the Anegada Passage. Low-angle normal fault surfaces were proposed to underlie most of the Puerto Rico-Virgin Islands platform. The north dipping surface of the platform in the north Puerto Rico area was attributed to a rollover anticline related to down-to-the-south movement on the 19¡ fault (Figure 3A). In our data we find no evidence for a zone of recent west-northwest directed divergence affecting the carbonate platform as predicted by this model (Figure 3A). The only divergent features observed are normal faults in the downgoing North America plate that are probably caused by bending of the plate and minor normal faults associated with the large east-west arch in the Mona Passage and the Guajataca arch (Figures 7 and 15). We find that the dip on the north coast platform can be better explained by regional arching rather than by a rollover anticline associated with normal faulting along the 19¡ fault. We do not find any evidence for this fault in the seismic reflection profiles. The steep edge of the carbonate margin in this area is attributed to gravitationally induced mass wasting [Schwab et al., 1991].

5.2.2. Rotating Microplate Hypothesis: Masson and Scanlon [1991]

Masson and Scanlon [1991] used long-range side scan sonar images and seismic reflection data to map the major tectonic features both north and south of Puerto Rico. To the north in the Puerto Rico trench, they proposed that almost pure strike-slip faulting occurs in the Puerto Rico trench between 65¡30' and 68¡ (Figure 3b), although a significant component of underthrusting occurs to the east.

To the south, they observed underthrusting of the Caribbean oceanic plateau crust beneath Puerto Rico in the central part of the Muertos Trough, but this motion becomes divergent in the Virgin Islands basin. To explain these observations, they proposed a simple model involving the rotation of the Puerto Rico tectonic block within a broad zone of east-west strike-slip motion. Counterclockwise rotation of this block induces localized divergence in the southeast (Virgin Islands basin) and northwest corners of the block (western Puerto Rico trench) as well as convergence in the northeast (eastern Puerto Rico trench) and southwest corners of the block (western Muertos Trough) (Figure 3b). A study of the paleomagnetism of the Neogene platform exposed on the north coast of Puerto Rico established 25¡ counterclockwise rotation during the period between 10 and 4 Ma [Reid et al., 1991].
Our data do not directly address this hypothesis. Many of the predicted areas affected by the rotation of the Puerto Rico-Virgin Islands platform lie in deep water and beyond the area of the carbonate platform described in this paper (Figure 3b). In the north coast and Virgin Islands area the period of major arching in the platform occurs prior to the 10-4 Ma 25¡ counterclockwise rotation of the north coast. Also the middle Miocene subsidence of the Kingshill basin on St. Croix as determined by Gill [1989] occurs prior to the 10-4 Ma period of rotation.
The subsidence and tilting of the platform appear to be a regionally coeval event all along the Puerto Rico-Virgin Islands microplate boundary and do not show the change from convergence to divergence. In conclusion, we propose that the counterclockwise rotation of Puerto Rico is probably not related to the tilting and arching of the platform. Moreover, it is not clear from the data presented on the platform if the rotation and the tilting are coeval.

5.2.3. Pinning and Localized Divergence Hypothesis: Vogt et al. [1976]

Vogt et al. [1976] proposed that the oblique collision of the Greater Antilles arc with the Bahamas platform would impede the eastward motion of the Hispaniola area (Figure 3c). Because the Puerto Rico-Virgin Islands platform has passed the eastern limit of the collisional zone marked by the Navidad Bank at 68¡30' [Mullins et al., 1992; Dolan et al., 1998], it would be unimpeded, and divergent features might form between the "pinned" (Hispaniola) and "unpinned" (Puerto Rico-Virgin Islands platform) areas in the Mona Passage. The marine geology and tectonic setting of the basins in the Mona Passage (Yuma, Cabo Rojo, Mona rifts) are described in detail by Grindlay et al. [1997] and Dolan et al. [1998].

This model predicts the localized divergence in the Mona Passage area that appears to be late Neogene in age. However, this model does not explain the east-west arching in this and the other platform areas. If the pinning process is deforming the Mona Passage in the present day, it is reasonable to assume that the same process deformed areas to the east in Puerto Rico and the Virgin Islands when the Bahama platform was located farther to the east (compare reconstructions in Figure 4). We see no evidence for north-south normal faults or any other divergent features caused by the collision of the Bahama platform with the Greater Antilles arc in these areas.

5.2.4. North-South Shortening and Arching Hypothesis: Dillon et al. [1996]

A prominent feature of the Puerto Rico-Virgin Islands platform as seen on the bathymetric (Plate 1) compilation maps is a roughly east-west trending arch in the surface of the platform that extends from the eastern coast of the Dominican Republic to the eastern edge of the Virgin Islands platform (Figure 3d). Because this arch affects rocks of the platform, the tectonic shortening event that created it is younger than the youngest folded unit of the platform, the Quebradillas Limestone of early Pliocene age [Moussa et al., 1987]. The northern flank of this arch is defined by the shallower dipping and less deformed carbonate platform compared with the steeper and more complex deformed southern margin (Figures 5a-5d). The island of Puerto Rico is the most eroded part of the arch where arc rocks are exposed up to elevations of 2 km (Figures 5b and 5c). The carbonate cap is least eroded in the Mona Passage and in the Virgin Islands, where, to our knowledge, it is an unnamed feature (Figure 5d). Faults in the Anegada Passage may also postdate the formation of the arch and separate carbonate outcrops on St. Croix and southern Puerto Rico [Gill, 1989] (Figures 5c and 5d).

One possible mechanism for the post-early Pliocene formation of the arch over the past 2 Ma is north-south crustal convergence across the Puerto Rico-Virgin Islands platform related to subduction at the Puerto Rico trench and the Muertos Trough [Dillon et al., 1996; Dolan et al., 1998] (Figure 5). Much larger amounts of northeast-southwest crustal convergence are observed to the west in Hispaniola and may be related to the oblique collision of the Bahama platform with the island [Mann et al., 1991; Mullins et al., 1992]. Smaller amounts of crustal convergence, the presence of subducted slabs, and the formation of the arch in the Puerto Rico-Virgin Islands area may be related to the subduction of Atlantic and Caribbean oceanic crust in the area east of the main Hispaniola-Bahama platform collision zone. This model best explains the major feature that we observe in these data: the regional east-west arch extending the length of the study area (Figure 3d). The thrust loading of North America lithosphere by the overriding Caribbean lithosphere bends the North America plate, disturbing it from isostatic equilibrium and causing the huge amount of subsidence of the platform. This subsidence would not only happen in the north but also in the south of Puerto Rico, which results in the east-west arch we observe in our data. We expect this arch to be nonsymmetrical on the north and south sides, because of the opposed subduction angles, and to be active over an area of hundreds of kilometers. Both subducting slabs interact in a broad zone, underlying the seismic reflection profiles we show from the Puerto Rico-Virgin Islands platform. This model would also explain the huge gravity low found off northern Puerto Rico. However, if arching is related to the north-south convergence between the two subduction zones, it would imply that subduction convergence is as long-lived as the arching, which appears to have been active from the end of the Eocene.

5.2.5. Platform Affected by Tectonic Erosion Related to Oblique Subduction of Atlantic Fracture Zone Highs: McCann and Sykes [1984]

McCann and Sykes [1984] proposed that the southeastern Bahama platform and Atlantic fracture zone highs are being obliquely subducted beneath the Puerto Rico margin. Highs on the downgoing plate like the Main ridge (Figure 2) and the Navidad Bank (Figure 2) were proposed to produce northwest trending arches in the overriding Puerto Rico-Hispaniola microplate and its overlying carbonate platform. The shallow subduction of these highs and their eastward migration would leave areas of the margin unsupported and subject to collapse as the highs moved from beneath arched areas on the overriding plate. In Figures 4a-4c, three reconstructions of the projections of the Navidad Bank and Main ridge highs beneath the overriding Puerto Rico area are presented. Areas east of the two arches shown might be expected to collapse in the wake of the two moving ridges.

Moussa et al. [1987] and Birch [1986] compiled data on the drowned platform of northern Puerto Rico and concluded that the 350 km by 50 km area of the offshore platform had undergone a 4¡ tilt to the north and had sunk to a depth of 4 km below sea level at its northern limit. The post-Pliocene age of this event yields a subsidence rate of 1 mm/yr at this limit of the carbonate platform. Birch [1986] explained the subsidence by tectonic erosion of thick island arc crust that previously extended at least as far as the northern limit of the platform (Figure 2 and Plate 1). This tectonic erosion event replaced island arc crust by oceanic crust, increased the weight of the island arc, and caused the observed subsidence.
This model predicts deformational effects and arching of the platform in directions almost perpendicular to the observed trends of the Guajataca and San Juan arches. Moreover, the latter two arches were formed in the Oligocene and Miocene well before the predicted passage of the Bahama platform according to the reconstructions shown in Figure 4. This model would require time-transgressive deformation of the platform, which would have begun in the east and progressively moved toward the west. This is not what we observe in the data, where we see that the carbonate platform is still intact and has been tilted as a coherent block.
It also predicts substantially more deformation of the platform than we observe.

6. Conclusions

In this study, newly acquired and older available seismic reflection data were interpreted, and a regional map of the carbonate platform limestones has been generated. With the use of this map, five different carbonate provinces are defined, of which the structure and stratigraphy have been described in detail. The major results of these observations are as follows:

1. The Puerto Rico-Virgin Islands platform limestones are deposited from middle Oligocene to early Pliocene in a stable area that was unaffected by the nearby plate boundaries.
2. In the Puerto Rico north coast basin, two arches are present: (1) the northwest trending Guajataca arch and (2) the northeast trending San Juan arch.
3. The dominant feature of the Puerto Rico-Virgin Islands platform is the large east-west trending arch in the platform that extends from the Mona Passage across the Cordillera Central of Puerto Rico and through the Virgin Islands (Figure 3d and Plate 1).
Five previously proposed models have been studied and compared to the results of our data. The model that best explains these results is the north-south shortening and arching model as proposed by Dillon et al. [1994] (Figure 17). This is the only model that explains the regional tilting and arching, which appears both in the Virgin Islands and in the Mona Passage area east and west, respectively, of Puerto Rico. Since the strikes of both the Puerto Rico trench and the Muertos Trough are parallel, the intersection of the two slabs at depth occurs over a several hundred kilometer wide area. Benioff zones observed in the earthquake profiles confirm this interaction at depth.


(125K)

Figure 17. Simplified drawing of the proposed model. A three-dimensional display of the bathymetry is shown, based on the same compilation of bathymetric data described in the caption of Figure 2. The structure is viewed at an angle from the ESE. The horizontal part in the middle of the figure is the carbonate platform. The asymmetric arching is clearly observed with the smooth dip to the north and steeper dip to the south. Underneath the bathymetry we have drawn the assumed continuations of the subducted slabs, which are at a different scale than the bathymetry.

Other models result in a lateral, time-progressive change of the subsidence of the carbonate platform that is not observed in the data. Both tectonic erosion, the other model that can explain the 4 km subsidence, and pinning by the Bahama platform would result in an east to west varying deformation.
The late Miocene counterclockwise rotation of Puerto Rico [Reid et al., 1991] is not explained by the north-south shortening model. The origin of this late Miocene event, and how it relates to the early Pliocene to Holocene tilting of the Puerto Rico-Virgin Islands carbonate platform, remains uncertain.

Acknowledgments.

This work has been funded by National Science Foundation grants NSF-OCE-9504118 and NSF-OCE-9796189 to Grindlay, Mann, and Dolan. Van Gestel thanks the UTIG student cruise fund for the opportunity to participate in EW 96-05. Special thanks to the captain and crew of the R/V Maurice Ewing and the University of Hawaii MR1 support team. UTIG contribution 1387. CMSR contribution 197.

References

Anderson, R. N., Geophysical logs from the Toa Baja scientific drillhole, Puerto Rico, Geophys. Res. Lett., 18, 497-500, 1991. Birch, F. S., Isostatic, thermal and flexural models of the subsidence of the north coast of Puerto Rico, Geology, 14, 427-429, 1986.

Bird, D. E., S. A. Hall, J. F. Casey, and P. S. Millegan, Interpretation of magnetic anomalies over the Grenada Basin, Tectonics, 12, 1267-1279, 1993.

Byrne, D. B., G. Suarez, and W. R. McCann, Muertos Trough subduction-Microplate tectonics in the northern Caribbean, Nature, 317, 420-421, 1985.

DeMets, C., T. Dixon, F. Farina, E. Calais, P. Jansma, and P. Mann, GPS-derived velocities in the northeastern Caribbean: Implications for Caribbean-North America motion and the NUVEL-1A model, Eos Trans. AGU, Fall Meet. Suppl., 77 (46), F143, 1996.

Dillon, W. P., N. Terence Edgar, K. M. Scanlon, and D. F. Coleman, A review of the tectonic problems of the strike-slip northern boundary of the Caribbean plate and examination by GLORIA, in Geology of the United States' Seafloor: The View From GLORIA, edited by J. V. Gardner, M. E. Field, and D. C. Twichell, pp. 135-164, Cambridge University Press, Cambridge, U.K., 1994.

Dolan, J., P. Mann, R. de Zoeten, C. Heubeck, J. Shiroma, and S. Monechi, Sedimentologic, stratigraphic and tectonic synthesis of Eocene-Miocene sedimentary basins, Hispaniola and Puerto Rico, in Geologic and Tectonic Development of Hispaniola, edited by P. Mann, G. Draper, and J. F. Lewis, Spec. Pap. Geol. Soc. of Am., 262, 217-264, 1991.

Dolan, J. F., H. T. Mullins, and D. J. Wald, Active tectonics of the north-central Caribbean: Oblique collision, strain partitioning, and opposing subducted slabs, in Active Strike-Slip and Collisional Tectonics of the Northern Caribbean Plate Boundary Zone, edited by J. F. Dolan, and P. Mann, Spec. Pap. Geol. Soc. of Am. 326, in press, 1998.

Erickson, J. P., J. L. Pindell, and D. K. Larue, Fault zone deformational constraints on Paleogene tectonic evolution in southern Puerto Rico, Geophys. Res. Lett., 18, 569-572, 1991.

Frost, S. H., J. L. Harbour, M. J. Realini, and P. M. Harris, Oligocene Reef Tract development southwestern Puerto Rico, Part 1, report, 144 pp. Univ. of Miami, Miami, Fl., 1983.

Gill, I. P., The evolution of Tertiary St. Croix, Ph. D. thesis, 287 pp., Louisiana State University, Baton Rouge, Louisiana, 1989.

Glover, L., III, Geology of the Coamo area, Puerto Rico, and its relation to the volcanic arc-trench association, U.S. Geol. Surv. Prof. Pap., 636, 102 pp., 1971.

Gordon, M. B., P. Mann, D. Caceres, and R. Flores, Cenozoic tectonic history of the North America-Caribbean plate boundary zone in western Cuba, J. of Geophys Res., 105, 10,055-10,082, 1997.

Grindlay, N. R., P. Mann, and J. Dolan, Researchers investigate submarine faults north of Puerto Rico, Eos Transactions AGU, 78 (38), 404, 1997.

Haq, B.U., J. Hardenbol, and P. R. Vail, Chronology of fluctuating sea levels since the Triassic, Science, 235, 1156-1167, 1987.

Hempton, M. R., and J. A. Barros, Mesozoic stratigraphy of Cuba: Depositional architecture of a southeast facing continental margin, in Mesozoic and Early Cenozoic Development of the Gulf of Mexico and Caribbean Region: A Context for Hydrocarbon Exploration, edited by J. L. Pindell and R. F. Perkins, pp. 193-209, Gulf Coast Section Soc. Econ. Paleont. and Mineral. Found., Houston, Tex., 1993.

Jany, I., K. M. Scanlon, and A. Mauffret, Geological Interpretation of combined Seabeam, Gloria and Seismic data from Anegada Passage (Virgin Islands, North Caribbean), Mar. Geophys. Res., 12, 173-196, 1990.

Ladd, J. W., T. L. Holcombe, G. K. Westbrook, and N. T. Edgar, Caribbean marine geology; Active margins of the plate boundary, in The Geology of North America. vol. H, The Caribbean Region, edited by G. Dengo and E. J. Case, pp. 261-285, Geol. Soc. of Am., Boulder, Colo., 1990.

Larue, D. K., The Toa Baja drilling project, Puerto Rico: Scientific drilling into a non-volcanic island-arc massif, Geophys. Res. Lett., 18, 489-492, 1991.

Larue, D. K., and H. F. Ryan, Extensional tectonism in the Mona Passage, Puerto Rico and Hispaniola: A preliminary study, in Transactions of the 12th Caribbean Geological Conference in St. Croix, edited by D. K. Larue and G. Draper, pp. 223-230, Miami Geological Society, Coral Gables, Fl., 1991.

Larue, D. K., and H. F. Ryan, Seismic reflection profiles of the Puerto Rico trench: Shortening between the North American and Caribbean plates, Spec. Pap. Geol. Soc. of Am. 322, 193-210, 1998.

Larue, D. K., R. Torrini Jr., A. L. Smith, and J. Joyce, North Coast Tertiary basin of Puerto Rico: From arc basin to carbonate platform to arc-massif slope, Spec. Pap. Geol. Soc. of Am. 322, 155-176, 1998.

Mann, P., G. Draper, and J. F. Lewis, Overview of the geological and tectonic development of Hispaniola, in Geologic and Tectonic Development of Hispaniola, edited by P. Mann, G. Draper, and J. F. Lewis, Spec. Pap. Geol. Soc. of Am. 262, 1-28, 1991.

Masson, D. G., and K. M. Scanlon, The neotectonic setting of Puerto Rico, Geol. Soc. of Am. Bull., 103, 144-154, 1991.

McCann, W. R., and L. Sykes, Subduction of aseismic ridges beneath the Caribbean plate: implications for the tectonics and seismic potential of the northeastern Caribbean, J. Geophys. Res., 89, 4493-4519, 1984.

Mercado, A., Digitization of National Ocean Survey hydrographic survey "smooth" sheets for Puerto Rico and the U.S. Virgin Islands, Sea Grant College Program, Mayaguez, Puerto Rico, 116 pp., 1994.

Meyerhoff, A. A., E. A. Krieg, J. D. Cloos, and I. Taner, Petroleum potential of Puerto Rico, Oil Gas J., 81, 113-120, 1983. Monroe, W. H., Geology of the middle Tertiary formations of Puerto Rico, U.S. Geol. Surv. Prof. Pap., 953, 93 pp., 1980.

Moussa, M. T., G. A. Seiglie, A. A. Meyerhoff, and I. Taner, The Quebradillas Limestone (Miocene-Pliocene), northern Puerto Rico, and tectonics of the northeastern Caribbean margin, Geol. Soc. Am. Bull., 99, 427-439, 1987.

Mullins, H. T., N. Breen, J. Dolan, R. Wellner, J. Petruccione, M. Gaylord, B. Andersen, A. Mellillo, A. Jurgens, and D. Orange, Carbonate platforms along the southeast Bahamas-Hispaniola collision zone, Mar. Geol., 105, 169-209, 1992.

Muszala, S. P., N. R. Grindlay, P. Mann, and J. F. Dolan, Euler deconvolution of marine magnetic anomaly data north of Puerto Rico: Constraints on the neotectonic deformation of the northeastern North America-Caribbean plate boundary, Eos Trans. AGU, 78 (17), Spring Meet. Suppl., S115, 1997.

Nemec, M. C., A two-phase model for the tectonic evolution of the Caribbean, in Transactions of the 9th Caribbean Geological Conference, pp. 23-24, 1980.

Pindell, J. L., and S. F. Barrett, Geological evolution of the Caribbean region: A plate-tectonic perspective, in The Geology of North America, vol. H, The Caribbean Region, edited by G. Dengo and J. E. Case, pp. 405-432, Geol. Soc. of Am., Boulder, Colo., 1990.

Reid, J. A., P. W. Plumley, and J. H. Schellekens, Paleomagnetic evidence for Late Miocene counterclockwise rotation of north coast carbonate sequence, Puerto Rico, Geophys. Res. Lett., 18, 565-568, 1991.

Rosencrantz, E., Structure and tectonics of the Yucatan basin, Caribbean Sea, as determined from seismic reflection studies, Tectonics, 9, 1037-1059, 1990.

Schwab, W. C., W. W. Danforth, K. M. Scanlon, and D. G. Masson, D. G., A giant submarine slope failure on the northern insular slope of Puerto Rico, Mar. Geol., 96, 237-246, 1991.

Speed, R., and D. K. Larue, Extension and transtension in the plate boundary zone of the northeastern Caribbean, Geophys. Res. Lett., 18, 573-576, 1991.

Vogt, P. R., A. Lowrie, D. R. Bracey, and R. N. Hey, Subduction of Aseismic ridges: effects on shape, seismicity, and other characteristics of consuming plate boundaries, Spec. Pap. Geol. Soc. Am. 172, pp. 1-59, 1976.

Western Geophysical Company, Offshore geophysical investigations for siting of a nuclear power station on Puerto Rico, for Puerto Rico Water resources Authority, Preliminary Safety Analysis Report, U.S. Atomic Energy commission, Docket No. 50-376, North Coast Nuclear Plant No. 1, vol. III, 101 pp., 1974.

upLast changes:
For any comments on this page you can always mail me at comments@vangestel.com