Variation of palaeostress patterns along the Oriente...

20
Variation of palaeostress patterns along the Oriente transform wrench corridor, Cuba: significance for Neogene–Quaternary tectonics of the Caribbean realm Y. Rojas-Agramonte a,b, * , F. Neubauer a , R. Handler a , D.E. Garcia-Delgado b , G. Friedl a , R. Delgado-Damas b a Fachbereich fu ¨r Geographie, Geologie und Mineralogie, Universita ¨t Salzburg, Hellbrunner Strasse 34, A-5020 Salzburg, Austria b Instituto de Geologı ´a y Paleontologı ´a, Vı ´a Blanca y Lı ´nea del Ferrocarril s/n, San Miguel del Padro ´n 11000, Havana, Cuba Received 16 October 2003; accepted 19 November 2004 Available online 20 January 2005 Abstract In this study, we address the late Miocene to Recent tectonic evolution of the North Caribbean (Oriente) Transform Wrench Corridor in the southern Sierra Maestra mountain range, SE Cuba. The region has been affected by historical earthquakes and shows many features of brittle deformation in late Miocene to Pleistocene reef and other shallow water deposits as well as in pre-Neogene, late Cretaceous to Eocene basement rocks. These late Miocene to Quaternary rocks are faulted, fractured, and contain calcite- and karst-filled extension gashes. Type and orientation of the principal normal palaeostress vary along strike in accordance with observations of large-scale submarine structures at the south-eastern Cuban margin. Initial N–S extension is correlated with a transtensional regime associated with the fault, later reactivated by sinistral and/or dextral shear, mainly along E–W-oriented strike-slip faults. Sinistral shear predominated and recorded similar kinematics as historical earthquakes in the Santiago region. We correlate palaeostress changes with the kinematic evolution along the boundary between the North American and Caribbean plates. Three different tectonic regimes were distinguished for the Oriente transform wrench corridor (OTWC): compression from late Eocene–Oligocene, transtension from late Oligocene to Miocene (?) (D 1 ), and transpression from Pliocene to Present (D 2 –D 4 ), when this fault became a transform system. Furthermore, present-day structures vary along strike of the Oriente transform wrench corridor (OTWC) on the south-eastern Cuban coast, with dominantly transpressional/ compressional and strike-slip structures in the east and transtension in the west. The focal mechanisms of historical earthquakes are in agreement with the dominant ENE–WSW transpressional structures found on land. D 2004 Elsevier B.V. All rights reserved. Keywords: Transform fault; Wrench corridor; Sierra Maestra; Santiago basin; Santiago deformed belt; Palaeostress; Cuba 0040-1951/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.tecto.2004.11.006 * Corresponding author. Present address: Institut fu ¨r Geowissenschaften, Universita ¨t Mainz, D-55099 Mainz, Germany. E-mail addresses: [email protected] (Y. Rojas-Agramonte)8 [email protected] (F. Neubauer). Tectonophysics 396 (2005) 161– 180 www.elsevier.com/locate/tecto

Transcript of Variation of palaeostress patterns along the Oriente...

www.elsevier.com/locate/tecto

Tectonophysics 396 (

Variation of palaeostress patterns along the Oriente transform

wrench corridor, Cuba: significance for Neogene–Quaternary

tectonics of the Caribbean realm

Y. Rojas-Agramontea,b,*, F. Neubauera, R. Handlera, D.E. Garcia-Delgadob,

G. Friedla, R. Delgado-Damasb

aFachbereich fur Geographie, Geologie und Mineralogie, Universitat Salzburg, Hellbrunner Strasse 34, A-5020 Salzburg, AustriabInstituto de Geologıa y Paleontologıa, Vıa Blanca y Lınea del Ferrocarril s/n, San Miguel del Padron 11000, Havana, Cuba

Received 16 October 2003; accepted 19 November 2004

Available online 20 January 2005

Abstract

In this study, we address the late Miocene to Recent tectonic evolution of the North Caribbean (Oriente) Transform Wrench

Corridor in the southern Sierra Maestra mountain range, SE Cuba. The region has been affected by historical earthquakes and

shows many features of brittle deformation in late Miocene to Pleistocene reef and other shallow water deposits as well as in

pre-Neogene, late Cretaceous to Eocene basement rocks. These late Miocene to Quaternary rocks are faulted, fractured, and

contain calcite- and karst-filled extension gashes. Type and orientation of the principal normal palaeostress vary along strike in

accordance with observations of large-scale submarine structures at the south-eastern Cuban margin. Initial N–S extension is

correlated with a transtensional regime associated with the fault, later reactivated by sinistral and/or dextral shear, mainly along

E–W-oriented strike-slip faults. Sinistral shear predominated and recorded similar kinematics as historical earthquakes in the

Santiago region. We correlate palaeostress changes with the kinematic evolution along the boundary between the North

American and Caribbean plates. Three different tectonic regimes were distinguished for the Oriente transform wrench corridor

(OTWC): compression from late Eocene–Oligocene, transtension from late Oligocene to Miocene (?) (D1), and transpression

from Pliocene to Present (D2–D4), when this fault became a transform system. Furthermore, present-day structures vary along

strike of the Oriente transform wrench corridor (OTWC) on the south-eastern Cuban coast, with dominantly transpressional/

compressional and strike-slip structures in the east and transtension in the west. The focal mechanisms of historical earthquakes

are in agreement with the dominant ENE–WSW transpressional structures found on land.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Transform fault; Wrench corridor; Sierra Maestra; Santiago basin; Santiago deformed belt; Palaeostress; Cuba

0040-1951/$ - s

doi:10.1016/j.tec

* Correspon

E-mail addr

2005) 161–180

ee front matter D 2004 Elsevier B.V. All rights reserved.

to.2004.11.006

ding author. Present address: Institut fur Geowissenschaften, Universitat Mainz, D-55099 Mainz, Germany.

esses: [email protected] (Y. Rojas-Agramonte)8 [email protected] (F. Neubauer).

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180162

1. Introduction

Transform faults play a key role in plate kine-

matics as they link divergent and convergent plate

boundaries with each other. In theory, they represent

strike-slip boundaries along which there are no

major transtensional or transpressional forces as

plate-driving forces, e.g., slab pull or ridge push

acting parallel to it. However, many major transform

faults include restraining and releasing bends

imposed by strength inhomogenities of the involved

lithospheric plates and oblique plate motion. Such

situations create wide transform wrench corridors

rather than narrow transform faults. In nature, strike-

slip faults often represent wide wrench corridors,

specifically when rheologically weak cover sedi-

Gran

N

Fig. 8Fig. 11

Cuba

CaJamaica

Gulf of Mexico

North American Plate

Trough

Cayman

Oriente FaultSDB

YucatanBlock

ChortisBlock

NicaraguanRise Colombia

Basin

90°W 80°W

Yucatan Basin

North

n

Caribbea

TransformSystem

SM

Sierra Maestra

Paleogene volcanic units Cretaceous volcanic u

Eocenic granitoids

Swan fault

Neogene-Quaternary deposit Middle Eocene-Oligoc

Las TunasHolguin

Sea

SubmarinBahamasPlatform

Pil nó

Cauto basin

Fig. 1. (a) Simplified map of the Caribbean realm showing the location of

Cuba and is shown in detail in panel (b). SDB—Santiago Deformed Belt. S

Iturralde-Vinent, 1996).

mentary rocks are involved (Mandl, 1988). Such

wrench corridors comprise structures such as releas-

ing and restraining bends, strike-slip duplexes, and

negative and positive flower structures along fault

oversteps (e.g., Sylvester, 1988; Zolnai, 1988).

Classical experiments (Riedel, 1929; Wilcox et al.,

1973; Mandl, 1988) show that, along seemingly

straight segments of major strike-slip faults, no

through-going straight segments evolve during for-

mation of such faults. An array of en echelon

arranged Riedel shears, P-shears, and antithetic

Riedel shears (cross-faults) evolve along such

corridors, with nearly undeformed shear lenses in

between anastomosing faults. The anastomosing

zones of slip are termed the bfault coreQ (Caine et

al., 1996). A wide zone of rocks, which can be

Piedra

Fig. 9

Nipe-Guacanayabo fault

Puerto Rico

ribbean Plate

Hispaniola

Lesser Antilles

n BeataRidge

VenezuelanBasin

Aves

Ridge

20°N

70°W

Atlantic Ocean

nits

Northern ophiolite

a

Cretaceous metavolcanicsene sediments

N

Guantanamobasin

100 km

e

b

the Oriente Fault. Dashed rectangle locates the eastern province of

M—Sierra Maestra. (b) Geological sketch map of eastern Cuba (after

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 163

distant from the fault zone itself, is influenced by

fault motion along such and is known as bdamage

zoneQ (Caine et al., 1996). Furthermore, rigidity

contrasts between continental and oceanic litho-

spheres along transform systems as well as the

presence of releasing and restraining bends may

strongly influence the structural evolution of a

transform fault system. This is the case along the

Oriente transform wrench corridor (OTWC), also

described as Oriente Transform Fault, of the north-

ern Caribbean realm and its expression in the coastal

ranges of the Sierra Maestra, southeastern Cuba,

(Fig. 1). We investigated the on-land structures and

expressions of the OTWC in late Miocene–Quater-

nary rocks on the southern margin of this mountain

range. These observations document the structural

Fig. 2. (a) Simplified map of the late Miocene-Quaternary deposits in the

margin showing the distribution of main tectonic elements along the Orient

Compiled using unpublished maps available from the Institute of Geolog

shown (Fig. 4). (b) Inset map showing pattern of main active faults defined

Rueda Perez et al., 1994 and Kuzovkov et al., 1988). The most important f

fault zone system with a NW-SE orientation. (c) Morphotectonic different

Western Sierra Maestra; SMC—Central Sierra Maestra; SME—Eastern Si

(after Hernandez Santana et al., 1991, Iturralde-Vinent, 1991).

arrangement and permit interpolations on plate

motion and changes in external palaeostress con-

ditions along a major strike-slip wrench corridor also

from places within the damage zone and distant

from places of principal slip.

A better understanding of the geology of Cuba,

especially of the kinematic processes related to the

OTWC, is important to explain the tectonic evolution

of the Caribbean region and to relate Neogene to

Recent deformation to the kinematics of the Carib-

bean plate. We present an analysis of deformational

features in late Miocene and Quaternary sediments in

order to document and evaluate the type and

orientation of palaeostress patterns along strike within

the OTWC. Our study is also aimed at helping to

understand the palaeostress and kinematic conditions

Sierra Maestra and the submarine sectors of the Cuban continental

e Fault. The map shows the main fault system in the Sierra Maestra,

y and Paleontology, Havana. Location of cross-section (A–AV) isby geodetic–altimetric measurements in the period 1983–1994 (after

ault system is that with NE-SWorientation followed by the Baconao

iation in the Sierra Maestra macroblock: CC—Cabo Cruz; SMW—

erra Maestra; B-SC—Baconao-Santiago de Cuba; GP—Gran Piedra

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180164

of transform faults, for which data are largely missing

from situations deviating from pure strike-slip con-

ditions (e.g., Sylvester, 1988). Another issue is the

documentation of structures whose orientations and

kinematics fit in with historical earthquakes and recent

GPS surveys.

The Sierra Maestra is an E–W trending mountain

range (Figs. 1 and 2) that occurs immediately north of

the North Caribbean (Oriente) transform system

(Hernandez Santana et al., 1991; Figs. 1a and 2),

extending from several kilometers west of Pilon to

Guantanamo basin (Fig. 1b). The OTWC separates the

Caribbean plate from the North American plate and is

connected with the Swan fault to the west through the

Mid-Cayman spreading centre (Fig. 1a), which has

generated the oceanic crust of the Cayman trough

since the latest middle Eocene (Rosencrantz et al.,

1988; Fig. 1a).

Our study region extends for some 140 km E–W,

from near Pilon (La Mula) to near Playa Colorada

(east of Santiago de Cuba; Fig. 2a), and is parallel to

the submarine wall of the OTWC. A steep topo-

graphic gradient is typical for the region, ranging from

Fig. 3. Earthquake focal mechanisms along the Swan and Oriente fault

Spreading Center; EPGF—Enriquillo Plantain Garden fault; WF—Walton

for different styles of fault movement (from Stewart and Hancock, 1994)

1974 m (Pico Turquino) to �6642 m in the southern

Oriente deep (Magaz Garcıa, 1989; Fig. 2a).

2. The Oriente Transform Wrench Corridor

(OTWC)

The OTWC from the southeast Cuban margin to

north-western Hispaniola is an area of strong present-

day seismicity, reflecting ongoing motion along the

fault. The seismic energy released during earthquakes

provides an estimate of the coseismic displacement on

a strike-slip fault (Brune, 1969; Anderson, 1979).

Much of this deformation is concentrated along the

releasing and restraining bends of the OTWC such as

the Cabo Cruz basin and the Santiago deformed belt

(SDB; Calais et al., 1998; Moreno et al., 2002; Fig.

2a). Most solutions of earthquake focal mechanisms

indicate strike-slip motion in the Santiago sector and

help to define the trace of the fault, which is

dominantly below sea level (Mann and Burke, 1984;

Calais and Mercier de Lepinay, 1991; Fig. 3). Based

on teleseismic information, Rosencrantz and Mann

s, southern Cuba (after Calais et al., 1990, 1998). CSC—Cayman

fault. Inset rectangle in lower right shows idealized focal mechanism

.

Table 1

Historical earthquakes reported in Cuba for the last century (Chuy

Rodrıguez, 1999, and references therein)

Year Lat. N. Long. W. Ms H I Locality

1903 (19.90) (76.00) (5.7) (30) 7.0 Santiago de Cuba

1906 (19.65) (76.25) (6.2) (30) 7.0 Santiago de Cuba

1914 (21.22) (76.17) (6.2) (32) 7.0 Gibara

1914 (19.45) (76.30) (6.7) (30) 7.0 Santiago de Cuba

1926 (20.30) (77.10) (5.4) (15) 7.0 Manzanillo

1930 (19.90) (76.00) (5.8) (25) 7.0 Santiago de Cuba

1932 (19.80) (75.80) (6.75) – 8.0 Santiago de Cuba

1939 (22.50) (79.25) (5.6) – 7.0 Remedios–

Caibarien

1947 (19.90) (75.30) (6.75) (50) 7.0 Santiago de Cuba

1976 (19.87) (76.87) (5.7) (15) 8.0 Pilon

1992 (19.62) (77.70) (7.0) (30) 7.0 Cabo Cruz

Ms—magnitude of superficial waves; H—depth of hypocentre; I—

intensity.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 165

(1991) suggested the existence of the Gonave micro-

plate to the south of Cuba (Fig. 3), which is bounded

to the north by the OTWC and to the south by the

Walton and Enriquillo–Plantain Garden faults, as part

of the Jamaica transform fault. The existence of this

microplate suggests that the rate of opening of the

Cayman trough (Figs. 1a and 3) is a measure of plate

motion between this microplate and the North

American plate, which implies a Caribbean–North

America relative offset rate of 20 mm/year or more

(Rosencrantz and Mann, 1991). This is in good

agreement with GPS results of DeMets et al. (2000)

and Mann et al. (2002), which suggest an average

movement rate for the Caribbean plate of 18–20F3

mm/year with 18F2 mm/year of boundary-parallel

slip.

The southern Cuban margin contains several first

order tectonic structures on the sea floor (Fig. 2a). The

Cabo Cruz basin is a narrow E–W trending depres-

sion, interpreted by Calais et al. (1998) as a pull-apart

basin created by left-lateral shear along the two

dextrally offset faults that bound it to the north and

to the south. Left-lateral displacement along these

faults imposed tensional strain in the relay area and

caused extension and subsidence (Calais et al., 1998).

This basin is internally divided into a series of oblique

horst and grabens delimited by normal faults.

The Oriente deep is an east–west trending depres-

sion bounded by a prominent scarp, probably gen-

erated by the OTWC. The Oriente deep provides clear

evidence of active transpressional tectonics (en

echelon folds and reverse faults) occurring along a

major strike-slip fault that was probably initiated

during the late Pliocene (Calais and Mercier de

Lepinay, 1990).

The Chivirico basin is a small depression, perched

upon the Oriente wall and interpreted as a pull-apart

basin formed in a dextral offset, ben echelonQ segment

(Calais and Mercier de Lepinay, 1991). It is bordered

to the north and south by two significant escarpments.

Based on seismic reflection data, Calais et al.

(1989) documented the existence of the Santiago

deformed belt to the south of Santiago the Cuba (Fig.

2a). It is a narrow submarine mountain range

extending over 300 km long and 10–30 km wide

along the OTWC with faults and thrust faults showing

clear evidence of transpressional deformation; a

positive flower structure was interpreted for the

western part of this belt (Calais and Mercier de

Lepinay, 1990; Calais et al., 1998). This belt

comprises several thrust sheets which indicate active

convergence along this sector of the OTWC since the

late Pliocene (Calais et al., 1998).

Calais et al. (1990) defined the stress field along

northern Hispaniola, where the trace of the OTWC

can be followed onland (Fig. 1a). It is characterized by

horizontal principal axes r1 and r3, compatible with

strike-slip kinematics; the maximum stress (r1) is

always highly oblique to the 1108N direction of the

main strike-slip fault zone. Calculations show that the

relative motion vector strikes around 808N from Cuba

to eastern Hispaniola (Calais et al., 1990). Therefore,

these authors concluded that the NNE–SSW direction

of r1 in the vicinity of the transcurrent fault zone is

very different from the direction of the relative motion

vector, and the kinematics of the Caribbean plate can

therefore not be directly deduced from the determi-

nation of the r1 direction.

Santiago de Cuba (Fig. 2b) is the second largest

city in Cuba and is considered to have the greatest

seismic hazard of any region on the island. Santiago is

crossed by two active normal faults (Quintero and

Sardinero faults; Fig. 2b) that cause tectonic insta-

bility in the region (Rueda Perez et al., 1994). Reports

of earthquakes in Cuba date back as far as the 16th

century (Alvarez et al., 1973; Calais et al., 1998), but

it was not until 1855 that the first seismic catalogue

was published in a paper by Andres Poey (1855).

Table 1 shows a list of reported earthquakes in

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180166

southeastern Cuba with an intensity of 7 or more

MSK (Medvedev–Sponheuer–Karnik) for the last

century (Chuy Rodrıguez, 1999).

3. Geomorphology and stratigraphy of Miocene–

Quaternary deposits

The southern margin of the Sierra Maestra exposes

Cretaceous to Eocene rocks and late Miocene–

Quaternary deposits in several depressions (graben)

facing the OTWC. The pre-Neogene rocks were

deformed by nearly east–west trending folds and

north-vergent thrust faults (Fig. 4) while the late

Miocene–Quaternary units were deposited uncon-

formably on top of the pre-Neogene rocks and appears

less deformed. The latter can be used to monitor

vertical motion of the region as well as distant effects

of wrenching along the OTWC.

The sinistral movement of the OTWC is respon-

sible for the present configuration of the Sierra

Maestra macroblock (Hernandez et al., 1989). NE-

trending faults and subordinate ENE–WSW-trending

faults (Fig. 2a,b) formed in response to this movement

and are important features in the region determining

some geomorphologic and dynamic features in this

area (Hernandez Santana et al., 1991; Rueda Perez et

al., 1994). Rueda Perez et al. (1994) describe

horizontal displacements with an E–W orientation

and formation of structures due to N–S extension for

Late Miocene–Quaternary rocks in the Santiago

Fig. 4. Cross-section across a portion of Central Sierra Maestra (see Fig. 2 f

close to the coast (after Rojas-Agramonte, 2003). The compressive ev

Compressive structures were overprinted by widespread extensional stru

Miocene (Rojas-Agramonte, 2003). We correlate the extensional event wi

rocks in southern Sierra Maestra.

region. Normal faults in this region experienced

downward displacement up to 40 mm during the

period 1983 to 1990 (Rueda Perez et al., 1994). These

authors proposed a geodynamic scenario whereby the

zone of main tectonic weakness has a NE–SW

orientation. The Baconao fault zone was described

by these authors as the second most important fault

system in this region (Fig. 2b).

These neotectonic faults system is mainly respon-

sible for the present coastal configuration and for

emplacement and formation of different morphostruc-

tural blocks in the Sierra Maestra (Fig. 2c). These

blocks have a strong fracturing in a way of horst

(Hernandez Santana et al., 1991) and grabens located

in the southern sectors close to the coast (Iturralde-

Vinent, 1991). The Central Sierra Maestra macroblock

(SMC; Fig. 2c) is the most elevated in the eastern

Cuban region with the most intense neotectonic

movements whereas the Boniato-Santiago de Cuba

block (B-SC; Fig. 2c) constitutes the link between

western Sierra Maestra and the Gran Piedra Moun-

tains, the Santiagio basin formed in this block and is

the most depressed graben in the region.

The stratigraphy of the late Miocene–Quaternary

deposits is shown in Fig. 5 and has been described in

detail by Cabrera-Castellanos et al. (2003). In general

terms, the basal La Cruz Fm. (Vaughan, 1919) is of

late Miocene to early Pliocene age and is charac-

terized by an alternation of terrigenous and carbonatic

sediments, suggesting sea level changes due to

episodes of subsidence and uplift. The overlying late

or location) showing normal faults overprinting compressive features

ent occurred between middle Eocene to early Oligocene times.

ctures, mainly by S-directed normal faults in the late Oligocene to

th the first deformation (D1) described for late Miocene–Quaternary

Fig. 5. Stratigraphic scheme of late Miocene–Quaternary formations of the southern Sierra Maestra. *Geological time scale, calibration after

Remane et al. (2002).

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 167

Pliocene to Pleistocene Rıo Maya Fm. (Franco, 1976)

consists of biohermic limestones with well-preserved

corals. Sedimentation occurred during fast tectonic

uplift with some periodicity, leading to the formation

of up to four terrace levels of eustatic origin (Cabrera-

Castellanos et al., 2003). The late Pleistocene

Jaimanitas Fm. (Brfdermann, 1940) occupies the

first Pleistocene terrace in the area with a height of

approximately 3–8 m (Cabrera-Castellanos et al.,

2003). Fossiliferous limestones with well-preserved

shells and corals characterize this formation as well as

karstic cavities filled by reddish calcareous, loamy,

ferruginous clays. The biocalcirudites of the intertidal

La Cabana Fm. (Cabrera-Castellanos et al., 2003),

together with the calcarenites and breccia-conglom-

erates of the Salado Fm. (Cabrera-Castellanos et al.,

2003), represent transgressive cycles within the great

Wisconsin regression. Locally, the above lithology

also occurs as fillings within the formations described

before. The Salado Fm. (Kartashov et al., 1981)

contains well-preserved Strombus gigas shells. The

youngest deposits in the area (undifferentiated units)

are Holocene in age and are divided into terrigenous

and marine units. The latter are represented by

conglomerate and coarse-grained sandstone.

Neotectonic movements related to the OTWC and

eustatic sea level changes controlled sedimentation of

the above formations. Coarse clastic material of the

Salado Fm. indicates surface uplift and documents

rapid, tectonically controlled sedimentation together

with sea level changes.

The late Miocene to Quaternary deposits (Fig. 5)

generally increases in elevation (including several

terraces) from la Mula to Playa Colorada (Fig. 6).

Terraces in Jaimanitas Fm. can be seen at several

localities along the southern Cuban coast. Shantzer et

al. (1975) considered the formation of these three

terraces to be exclusively due to glacio-eustatic and

not tectonic control, related to a sea level rise during

the Sangamon Interglacial in North America. Up to

four Plio-Pleistocene terrace levels are found in the

Rıo Maya Fm. (Fig. 6). These terraces reach altitudes

Caribbean Sea

Marine terraces

Jaimanitas Fm.

Rio Maya Fm.

E

N

uplifted

Fig. 6. View looking to the east of Playa Colorada. Terraces exposed in Rıo Maya Fm. People for scale.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180168

of up to 200 m in the eastern part of the area, whereas

the rocks from the western and central parts (from La

Mula to the west of the Santiago basin) appear less

elevated. Consequently, these terraces must be of

tectonic origin.

Those deposits in the southern part of the Sierra

Maestra unconformably overly Cretaceous and Palae-

ogene volcanic arc and sedimentary sequences (Fig.

1b). Eocene arc-related I-type granitoids of tonalitic–

trondhjemitic composition are exposed in several

massifs along the southern flank of the Sierra Maestra

mountain range (Rojas-Agramonte et al., 2004). The

still active sinistral transcurrent El Cristo fault (Perez

Perez and Garcıa Delgado, 1997; Fig. 2b) branches off

to the NW from the late Miocene to Quaternary

deposits in the Santiago basin. This fault is part of the

OTWC and is defined as a synthetic Riedel fault

(Perez Perez and Garcıa Delgado, 1997). The sub-

sidence and depositional history of the Santiago basin

was controlled by strike-slip deformation, related to

Miocene activity of the El Cristo fault, and synsedi-

mentary tectonics (Perez Perez and Garcıa Delgado,

1997). Most late Miocene–Quaternary formations that

are found along the southern Cuban coast are exposed

in the Santiago basin (minimum thickness ca. 200 m),

which has a regular geometry and is bounded by faults

(Fig. 2b). The El Cristo fault currently has a sinistral–

transcurrent character and constitutes the western

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 169

border of the Santiago basin, whereas the Baconao

fault zone forms the N–NE border (Fig. 2b) with the

Gran Piedra elevated (horst) block (Fig. 2c). The

southern boundary on-land is defined by the fault-

related straight coastline.

4. Methodology

Data from slickensides, striations, joints and

tension gashes were collected at ca. 60 stations within

late Miocene to Quaternary formations between Playa

Colorada and La Mula (Fig. 2a) in order to evaluate

the kinematics and stress history of the region. Each

station is a major outcrop along the coast with almost

uniform lithology. Depending on outcrop conditions,

we tried to measure and separate differently oriented

extensional veins filled with red soil in karst or calcite,

as well as tension gashes and fault planes with

striations. The sense of movement along the faults

was deduced from kinematic indicators, e.g., dis-

placed markers. Structures like left- or right-stepping,

releasing bends, shear lenses, Riedel shears and

conjugate Mohr shears were also observed on a small

scale in the study area; they also helped in defining

the sense of movement.

Fig. 7. (a) Unnamed Holocene deposits from locality 85, showing a broad

vein reactivated as sinistral strike-slip fault due to NE–SW compression.

(locality 31) showing normal fault. (d) Strike-slip fault within the Jaimanita

stress field.

The sequence of faulting and displacement was

determined according to criteria proposed by Petit

(1987) and Gamond (1983, 1987). Palaeostress

orientation patterns were evaluated from these faults

and slickenside data, using numerical and graphical

inversion methods as proposed by Angelier and

Mechler (1977), Angelier (1979, 1989), Armijo et

al. (1982), and Marret and Almendinger (1990). These

inversion methods indicate a strain rate rather than

palaeostress patterns with relative magnitudes of

principal stress axes (Twiss and Unruh, 1998). We

used the computer program package TectonicsFP

(Reiter and Acs, 1996) for fault-slip analysis and

determination of maximum (r1) and minimum (r3)

palaeostress axes. The stress regime is determined by

the nature of the vertical stress axes: extensional when

r1 is vertical, strike-slip when r2 is vertical, and

compressional when r3 is vertical.

A difficult point in defining successive palaeostress

tensor groups in the study area was to determine the

relative timing of formation of the different fault and

fracture systems. Relative timing of successive fault

and fracture development is indicated by overprinting

relationships such as consistent fault superposition.

Moreover, the age of deposition of the rocks is well

constrained.

NS

248/85

280/87

218/85

β α

σ 1

α β<<

open antiform formed due to ca. E–W compression. (b) Extensional

Small-scale releasing bend can be also observed. (c) La Cruz Fm.

s Fm. (locality 94). Sketch to the right shows movement pattern and

σ1 σ 2 σ3

83

819333

31

Santiago de Cuba

Caribbean Sea

9592

94

89

86

85

84

8790

20° 00´

75° 55´

Jaimanitas Fm.

La Cruz Fm.

El Cobre GroupLower Paleocene-Middle Eocene

Río Maya Fm.Río Seco

Río

ElS

ardi

nero

Río

San

Juan

Río

Just

ici

Salado Fm.

N

82

8987

90 (D )1

Locality 81

D : 172/881 D : 70/882 213/90

σ1N N

Conjugate Mohrshear (260/84)

D3

33 92 95

81 (D )384

94

85-86 (D )4

85, 90(karst-sinter infilling)

75° 50´ 75° 45´

92, 95, 81, 84,85, 87, 89, 90

92, 95, 81, 85,86, 87, 89, 90

92, 94, 95, 81,85, 86, 89, 90

33, 92, 94, 95,84, 86, 87, 89

Santiago basin

(D )2

D1 D2 D3D2

σ1

(D )2

b

a

2.5 km

Fig. 8. (a) Geological map of Santiago basin showing key outcrops with fault slip data, extensional veins and joints. (b) Relative sequence of

structures observed at locality 81.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180170

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 171

5. Structural results

The late Miocene–Quaternary deposits are folded,

faulted, and fractured and contain subvertical calcite-

and karst-filled tension gashes; for example, at

locality 85 (Fig. 7a), we found a gently S-dipping

fold, whereas at other localities (e.g., locality 94, Fig.

8), several strike-slip and subordinate normal faults

were measured (Fig. 7b–d) with offsets in the order of

0.1 to several metres in late Miocene–Quaternary

rocks.

The relative chronology of structures and, there-

fore, palaeostress conditions was deduced from over-

printing relationships in key outcrops, indicating the

existence of several events. This relationship led to

separation of different phases of deformation that

were computed from 60 outcrops. Palaeostress pat-

terns vary along strike from E to W: Four phases of

deformation in the eastern part (D1–D4) and three in

the western part (D1, D3, D4) were distinguished

(Table 2).

Table 2

Phases of deformation in Neogene–Quaternary rocks in the southern part

Western block G

Structures Interpretation

D1 Karst-filled E-trending

extensional veins

ENE–WSW trending tension

gashes caused ca. N–S-

to NNW–SSE-directed

extension

D

D2 – – D

D3 N-trending sinistral strike-slip

faults and E- trending,

dextral strike-slip faults.

NNW–SSE-directed

compression

D

D4 Dextral strike-slip faults;

normal faults were

reactivated. Extensional veins

and fractures also occur.

E–W- to ENE–WSW-directed

compression and N–S

extension.

D

The earliest deformation recorded in the eastern

part of the Sierra Maestra (Gran Piedra block) is

represented by N–S-oriented calcite- and karst-filled

extensional fractures seen at several localities (81, 82,

84, 87, 89, 92, 95, and 97; Fig. 8); these fractures

appear curved. These fractures were overprinted by N-

trending dextral and ca. E- and NE-trending sinistral

strike-slip faults as well as by extensional veins (ca.

10–20 cm wide) filled by calcite and laminated red,

karstic material (locality 95; Fig. 8) showing exten-

sion in a NW–SE direction. Superimposed onto these

are conjugate Mohr shears recording a NW–SE

maximum orientation of principal stress (Fig. 8b).

We also observed S-dipping normal faults displaying

N–S extension at locality 90 (Fig. 8), as well as

subvertical veins filled by karst and travertine. Small-

scale structures such as releasing bends in dextral and

sinistral strike-slip faults and shear lenses were also

observed at the above localities (e.g., Fig. 7b).

At locality 33 (Fig. 8), we measured veins filled by

red soil in karst that also formed due to N-S extension.

of the Sierra Maestra (D1–D4)

ran Piedra block

Structures Interpretation

1 E-trending extensional

fractures.

N–S-directed extension

2 Sinistral and N-trending

dextral strike-slip faults.

Extensional veins also

formed during this event,

filled with calcite.

NE–SW- to nearly

N–S-directed compression.

3 Dextral strike-slip faults as

well as conjugate Mohr shear,

reverse faults also appear.

NE–SW-oriented extension

along NW-trending

extensional veins and

fractures were also formed as

a consequence of this

compression.

NW–SE-directed

compression

4 Reactivation of ca. E-trending

sinistral strike-slip faults. Ca.

E-trending extensional veins

and karst-filled extensional

fractures are associated with

this stage.

ENE–WSW- to E–W-directed

compression and N-S

extension.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180172

These veins were reactivated as dextral faults due to

N–S compression and frequently show an anasto-

mosing or braided pattern on a small scale. NW-

trending sinistral strike-slip faults were also meas-

ured at this locality. Normal faults were recorded at

locality 31 (Fig. 7c) and formed due to E–W

extension. They were overprinted by reverse faults

resulting from NW–SE compression. Veins filled by

karst were documented at locality 94 and are due to

E–W extension. They were reactivated as sinistral

and dextral strike-slip faults formed in response to

NW–SE and NE–SW maximum principal stress (Fig.

7d). Locality 98 (Fig. 9) displays a conjugate vein

system, recording the same pattern as described

above, but we also measured veins formed in

response to N–S extension (Fig. 10a) and disrupted

96

9798

Caribbean Sea

99 Mag

dale

na

97 96

96-99, 104 96-96-99, 104

75° 40´ 75° 35´

Jur a

g ua

2 km

D1 D2

(D )2

Fig. 9. Map of the eastern part of the study area showing outcrops where da

8 for legend.

by dextral faults due to E–W compression. Fractures

formed due to NW–SE extension at locality 104

were reactivated as dextral faults that formed due to

E–W compression.

At locality 85 (Fig. 7a), a broad open fold with

gently S-plunging fold axis was measured as well as

extensional veins filled by sinter and karst. An en

echelon arrangement of oblique to dextral faults was

also recorded at this locality.

The western part of the study area (La Mula)

displays similar structural patterns, but these belong

to fewer structural events (Table 2). At localities 79

and 80 (Fig. 11a), we observed subvertical, ENE-

striking extensional fractures which were subse-

quently reactivated as dextral (up to ca. 40 cm of

lateral displacement) and/or sinistral strike-slip faults,

104Playa Coloradas

98 (D )497 (D )3

99, 104 96-99, 104

19° 55´

75° 30´

Agu

ada

delo

sB

ueye

s

N

D3

ta for fault slip, extensional veins, and joints were measured. See Fig.

Fig. 10. (a) Extensional vein, filled by calcite due to N–S extension in Jaimanitas Formation. (b) Dextral strike-slip fault displacing preexisting

fracture. (c) Transtensional structure, showing displaced coral in Jaimanitas Fm. due to a sinistral-oblique normal fault. See pen for scale (length:

16 cm).

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 173

recording ENE–WSW and NNW–SSE compressions,

respectively. Sinistral and dextral strike-slip faults

were also recorded at this locality; they formed in

response to NE–SW and NW–SE compressions,

respectively.

In the central part of the area, similar observations

at localities 26–28 and 34 (Fig. 11a, b) show that the

earliest deformation is documented by NE-SW exten-

sion, resulting in calcite-filled extensional veins (up to

15 cm wide). The next younger karst-filled fractures

resulted from NW–SE extension and were later

reactivated as dextral faults recording an E–W

maximum principal stress and a slight shift to ENE–

WNW compression. Common features in the area are

E–W trending faults, which often appear curved in

plan view.

Based on these observations, we propose a

succession of deformation phases as summarized in

Table 2. The palaeostress orientations in the western

and central parts of the Sierra Maestra are considered

to represent the same events as in the Gran Piedra

block. Our field observations and the above data set

show the palaeostress evolution along the southern

coast of the Sierra Maestra, which corresponds to the

systematic formation of extensional features and their

subsequent reactivation as strike-slip faults. During

each stress regime, the generation of new faults was

accompanied by reactivation and slip on preexisting

surfaces.

6. Discussion

We now discuss the tectonic evolution of the late

Miocene–Quaternary deposits along the southern

margin of the Sierra Maestra during the evolution of

the OTWC. The present southern slopes of the Sierra

Maestra expose the deep sectors of an early Tertiary

volcanic arc and sedimentary rocks, including a belt

of Eocene granitoid bodies in its centre. Changes in

palaeostress correspond to distinct episodes in the

kinematic evolution of the northern Caribbean

domain, and the deformation recorded indicates a

predominance of strike-slip movement. The palaeo-

stress tensor groups portray distinct changes over

time. Based on the relatively well-established tectonic

history of the OTWC along the south-eastern Cuban

margin (e.g., Calais et al., 1998 and references

therein), we correlate differently oriented palaeostress

groups onland with the evolution of the northeastern

Caribbean during Neogene–Quaternary times. A

number of models have been established for the

Neogene to Recent tectonic evolution of the north-

eastern Caribbean realm (Leroy et al., 2000; Pubellier

N

79, 80

8079 (D )3 2634

Caribbean Sea

20° 20´76° 00´76° 30´

La Mula

80

20 km

7926 27

28 34

27 34

Locality 27D3

D : 256/851

D : 318/904

N

79, 80

b

a

80

D1

(D )4

D1D1D3

D3

D3

Fig. 11. (a) Fault slip data and extensional veins of the western part of the study area up to La Mula localities (79, 80). (b) Sequence of

deformation observed at locality 27. For legend, see Fig. 8.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180174

et al., 2000; Mann et al., 2002). Our own observations

support and also modify these models (Fig. 12).

The initial event recorded in late Miocene–Quater-

nary rocks was extensional (D1). It is associated with

activity along the OTWC, and is the best documented

for this period in the Caribbean region (starting in

early Miocene according to Iturralde-Vinent and

Macphee, 1999). Iturralde-Vinent and Macphee

(1999) stated that in the early middle Miocene, the

Caribbean region shows the effect of disruption of

continuous landmasses (e.g., Cuba and Hispaniola).

Localized extensional features began to form as

grabens, trenches, and pull-apart basins.

We correlate our first deformation (D1, N–S

extension) with the extensional regime described

above and subsidence to the south of the area, with

formation of basins due to a transtensional regime

associated with the OTWC (Calais and Mercier de

Lepinay, 1991). This regime is well documented by

normal faults and infilled extensional fractures in

the working area (Fig. 10c). The disruption of Cuba

and Hispaniola was already taking place. Apatite

fission-track ages suggest that this event occurred

between middle and late Miocene (Rojas-Agra-

monte, 2003).

We correlate the first phase of deformation in the

western part of the Sierra Maestra (D1, NNW–SSE to

N–S extension) with a transtensional regime associ-

ated with the OTWC and with large subsidence of the

Oriente deep (Fig. 12a). As stated before, the process

of disruption between Cuba and Hispaniola was

already taking place and also affected the stress

regime in this area (NW–SE-directed extension).

The second phase of deformation (D2) corre-

sponds to NE–SW and nearly N–S-directed com-

pression, generating sinistral and dextral strike-slip

D1

D2

D3

D4

Hispaniola

Yucatanblock

spreadingat Cayman

ridge

subduction

Beata ridge

transtension

Beata ridge

sinistraltransform motion

local dextral inversion oftransform motion

Cayman ridge

Cuba

SMSM Sierra Maestra

Bahamas

carbonate platform

a

b

c

d

OTWC

OTWC

OTWC Oriente transformwrench corridor

indentationof the

Beata ridge

SDB

Caribbean Plate

Yucatan

basin

ChortisBlock

Jamaica

GMP

GMP

GMP Gonave microplate

SDB

SDB Santiago Deformed Belt

North American Plate

Atlantic Ocean

Swan Fault

Lesser Antilles

Bahamas

carbonate platform

North American Plate

Atlantic Ocean

Caribbean Plate

Swan Fault

Fig. 12. Models for the Neogene to Recent tectonic evolution of the northeastern Caribbean realm (according to Leroy et al., 2000; Pubellier et

al., 2000; Mann et al., 2002 and own observations). (a) Early Miocene N–S extension along the southern edge of the Sierra Maestra Mountains.

(b) Middle Miocene to early Pliocene sinistral transform motion along the Oriente transform wrench corridor. (c) Late Pliocene–Quaternary

dextral inversion at eastern sectors of the Oriente transform due to lateral escape from the frontal tip of the Beata ridge indenter. (d) Recent E–W

contraction due to lateral escape from the frontal tip of the Beata ridge indenter.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 175

faults and correlating with sinistral movements along

the main fault to the south. Because of nearly N–S

compression, extensional veins filled with calcite

were also formed. These reflect an extensional event

related to stress reorganization along strike of the

OTWC and coincide with movement of the Car-

ibbean plate to the ENE. At this phase, the Oriente

fault became a transform system (Fig. 12b).

Probably related to deformation processes in the

SDB is NW–SE-oriented compression (D3) defined

by E–W-oriented dextral strike-slip faults as well as

conjugate Mohr shears and reverse faults. The second

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180176

deformation in the western block is correlated with the

third phase in the Gran Piedra block (see Table 2), this

is why we label it D3. We associate this phase (D3,

NW–SE-directed compression) with deformation pro-

cesses along the Oriente deep when the sedimentary

fill of this basin began to undergo compression (Calais

and Mercier de Lepinay, 1991). This deformation

probably led to the formation of dextral strike-slip

faults on the south-eastern Cuban coast. An explan-

ation for the short-living dextral inversion of the

Oriente transform wrench corridor could be the

northeastward indentation of the Beata ridge (Fig.

12c), which is part of the Mesozoic Caribbean large

igneous province (Hoernle et al., 2004), into Hispa-

niola (e.g., Leroy et al., 2000; Pubellier et al., 2000),

creating there blocks, which lateral escaped to the east

and west. Such effects are well known from the frontal

part of indenters according to the slip-line theory (e.g.,

Molnar and Tapponnier, 1978).

The final deformation (D4) corresponds to

approximately E–W-directed compression, shifting

the palaeostress regime of the area in this direction

and giving rise to sinistral strike-slip faults and

Riedel shears. This is in agreement with observa-

tions at locality 85 (Fig. 7a) where Holocene gravels

and sands were folded and fractured recently in

response to E–W compression. Compressional defor-

mation even affected the Holocene sediments along

the south-eastern Cuban margin and was active and

synchronous with activity along the Oriente trans-

form fault (Calais et al., 1989), thus affecting the

morphology of the coast. Cabrera-Castellano et al.

(2003) interpreted these Holocene deposits as bbeachrocksQ and explained the existence of broad flexures

(Fig. 7a) by the specific morphology of the coast

and because of lateral currents resulting from

refraction of waves perpendicular to the coastline.

However, our field observations and measurements

show that these structures represent open folds

because (1) the fold is exposed in a bay and not

at a cape and because (2) similar gently dipping

rocks were observed at other exposures (e.g., local-

ity 86; Fig. 8). In western Sierra Maestra, the final

phase of deformation (D4, E–W- to ENE–WSW-

directed compression) generated dextral strike-slip,

and normal faults were reactivated. Extensional

veins and fractures due to N–S extension were also

formed.

This plate boundary has experienced a complex

Quaternary reorganization and became redefined and

modified during each phase of deformation as was

already stated by Pubellier et al. (2000) for the

western part of Hispaniola. GPS measurements show

a nearly eastward motion (ca. 708N–808E) of the

Caribbean plate in respect to North America. This

motion is nearly subparallel to the OTWC in northern

Hispaniola, but slightly oblique in southern parts of

Hispaniola (Mann et al., 2002), creating some trans-

pressive strain along the OTWC. This transpressive

motion could trigger some strain due to lateral

westward escaping blocks, which are in opposite

motion as motion created at the Cayman Ridge (Fig.

12d). These concurrent movements are likely respon-

sible for the ca. E–W to ENE–WSW D4 shortening

along of the eastern part of the OTWC.

According to Iturralde-Vinent (1998) the OTWC

had two main stages of development, the first one was

a compressive event in the late Eocene–Oligocene

(Fig. 4) and the second one was an extensive event

starting in the Miocene when the Cayman trough

opens. Our on-land observations are supported by

marine geophysical studies carried out off the south-

ern Cuban coast (Calais and Mercier de Lepinay,

1991). These authors described two major periods

related to the OTWC for the northern Caribbean

region, each one characterized by a distinct tectonic

regime starting in the Pliocene. The first one was a

general transtensional regime, accompanied by a large

subsidence of the Oriente deep (Fig. 2), which would

correspond with the first deformation described on-

land (D1); they explained this extensional tectonic

regime by the right stepping en echelon geometry of

the OTWC at sea. The transtensional regime was

followed by compressional tectonics during a trans-

pressional regime, when the sedimentary fill of the

Oriente deep began to undergo compression. This

event is still active and can be correlated with our D2,

D3 and probably D4 deformation. In summary, with

the above data, we can conclude that in general the

OTWC had three different tectonic events: compres-

sion from late Eocene–Oligocene, transtension from

late Oligocene to Miocene and transpression from

Pliocene to Present.

Moreno et al. (2002) suggested a thrust fault

regime along the southern Cuban margin from the

stress inversion of earthquake focal mechanisms.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 177

They obtained a nearly horizontal r1 (oriented ENE–

WSW) and nearly vertical r3, in agreement with the

dominant structural trend associated with the SDB.

The final deformation recorded on-land (E–W to

ENE–WSW compression) is also in good agreement

with their result. The earthquake epicentres are

located mainly to the south of the Santiago area

due to its location along the northern margin of the

SDB. The most recent earthquakes (Table 1) show a

westward propagation from Santiago (1947) to Cabo

Cruz (1992). The velocity of vertical movements in

Cuba, determined from geodetic methods, shows that

eastern Cuba displays extreme values of up to 12–15

mm/year, whereas central and eastern Cuba only

records a maximum velocity of 4 mm/year, suggest-

ing that eastern Cuba experiences greater vertical

neotectonic activity than the rest of the island

(Iturralde-Vinent, 2003).

The late Miocene to Quaternary reef and detrital

limestones generally were uplifted as part of a

system of marine terraces from la Mula to Playa

Colorada. The Jaimanitas Fm. occupies the first

Pleistocene terrace and can be seen at several

localities along the Cuban coast, with up to 8 m

high in western and central Cuba and probably up to

20 m in southeastern Cuba (Bresznyanszky et al.,

1983; Iturralde-Vinent, 2003). Shantzer et al. (1975)

considered the formation of this terrace to be

exclusively due to glacial and not tectonic control,

related to a sea-level rise during the Sangamon

Interglacial in North America. However, since

subsidence of the south-eastern Cuban coast is

fault-controlled and not only due to eustasy (Calais

and Mercier de Lepinay, 1990), we consider the

formation of this terrace to be related to neotectonic

movement along the OTWC and the compressive

action of the SDB. The terraces here are also more

elevated than others described along the northern

Cuban coast as well as in the central-western sectors

of the southern Cuban coast, supporting strong and

fast uplift due to this local tectonic movement

(Iturralde-Vinent, 2003). Like at other sites of the

Cuban coast, up to four Plio-Pleistocene terrace

levels are found in the Rıo Maya Fm. (Fig. 6),

obeying tectonic control. These terraces reach

altitudes of up to 200 m in the eastern part of the

area and seem to be closely related to compressional

activity in the SDB to the south, whereas in the

western and central parts (from La Mula to the west

of the Santiago basin), they appear less elevated.

Offset and bifurcations along strike-slip faults

create either transtensional or transpressional areas

(Davison, 1994). Offset faults can remain unlinked

until larger displacements have built up, resulting in

pull-apart basins or push-up mountain ranges. The

reasons for offset in strike-slip faults may be due to

propagation of separate faults which subsequently

become linked (Davison, 1994). We interpret the El

Cristo fault (Fig. 2), with a normal-slip component, as

an off splay of the OTWC, leading to the formation of

the Santiago basin in Neogene–Quaternary times (Fig.

2). Due to the proximity of the Santiago basin to the

main strike-slip fault, we interpret its formation as an

internal basin to the strike-slip zone, formed in a local

transtensional area along the active fault, as a possible

pull-apart basin. However, the overstepping fault on

the southeastern side of the Santiago basin is not

exposed. This makes the pull-apart basin origin

somehow hypothetical. The Neogene–Quaternary

structural pattern and formation of the Santiago basin

were strongly influenced by changes of the palaeo-

stress field along the OTWC, changing from strike-

slip to an extensional regime. We suggest that the

partial uplift of the Santiago basin occurred during our

D3 and D4 phases of deformation. The Chivirico pull-

apart basin (Fig. 2) to the south of the area has not

been as well studied by marine geophysical methods

as other structures related to the OTWC along the

southeastern Cuban margin. The geometry of this

basin was already traced by Calais and Mercier de

Lepinay (1991), and these authors conclude that the

basin developed on tensional relays along the main

left-lateral strike-slip fault. Due to the proximity of

these two basins and their position, we conclude that

they probably formed under the same regime with

similar kinematics probably starting in the late

Miocene.

Although E–W compression was responsible for

the last deformation recorded in the late Miocene–

Quaternary deposits of the Sierra Maestra, a trans-

pressional regime now dominates the area as indicated

by earthquake focal mechanisms (Calais et al., 1998;

Moreno et al., 2002).

Our new data from the Oriente transform wrench

corridor indicate a complicated evolution of this

transform system. Structures range from extension/

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180178

transtension during initiation of the transform fault

system through pure strike-slip transform motion to

late-stage complexities with transpressional structures

including local short-living shear reversal. These

transpressive motions seem to represent accommo-

dation effects of both indentation of an oceanic

plateau and individualisation of a microplate

(Gonave microplate). Such complicated evolutionary

paths at the interface between continental to oceanic

plates may represent rather the rule than exception

due to rigidity contrasts between involved litho-

spheric plates.

7. Conclusions

Palaeostress analysis of small-scale faults, joints,

and tension gashes of the OTWC along the south-

eastern Cuban coast enable us to establish the

timing and orientation of the stress regime that

resulted in Neogene separation of Cuba and

Hispaniola. The formation of the Santiago basin

in an offset of the OTWC is due to major sinistral

displacement along the transform fault. Consistent

datasets were obtained for all analyzed sites. Our

results indicate that the OTWC exhibits a fault and

fracture pattern which agrees with first motion

solutions derived from earthquake focal mecha-

nisms. We distinguish two different groups of

tectonic events for the OTWC: transtension during

late Miocene (?) (D1) and transpression from

Pliocene to Present (D2–D4).

Palaeostress analysis reveals four phases of defor-

mation for the easternmost part of Sierra Maestra

(Santiago basin up to Playa Colorada) and three

phases for the central and western parts up to La

Mula, with an age range from late Miocene to

Quaternary. These phases are:

D1: N–S-directed extension is mainly associated

with karst-filled extensional veins and normal faults.

We correlate this event with the regional kinematics in

the northern Caribbean imposed by the opening of the

Cayman trough. During the early middle Miocene, the

Caribbean region experienced the effects of the

separation of Cuba from Hispaniola. Localized

extension occurred, and grabens, pull-apart basins

and troughs began to form as well as subsidence in the

Oriente deep.

D2 and D3: NE–SW to nearly N–S and subsequent

NW–SE-directed compression generated sinistral and

dextral strike-slip faults and conjugate Mohr shears

that correlate with strike-slip movement along the

main ENE-trending fault to the south. These phases are

also associated with a transpressional regime in the

SDB and with deformation processes along the Oriente

deep when the sedimentary fill of this basin began to

undergo compression. We suggest that the Santiago

pull-apart basin formed during this event. Transten-

sional features due to E–Wand NE–SWextension also

formed during this deformation, displaying an event of

extension related to a stress reorganization along strike

of the OTWC and coinciding with the relative move-

ment of the Caribbean plate to the ENE. At this phase,

the OTWC became a transform fault.

D4: ENE–WSW- to E–W-directed compression

corresponds to a shift of the palaeostress regime in

this direction and gives rise to reactivation of sinistral

strike-slip faults and formation of Riedel shears.

Extensional veins and karst-filled extensional frac-

tures are associated with this phase and record N–S-

directed extension. This process is consistent with

stress orientations deduced from earthquake focal

mechanisms.

The above phases of deformation describe the

dynamics of the south-eastern Cuban coast at the

leading edge of the North American plate during

formation and development of the northern Caribbean

Oriente transform wrench corridor. We compare late

Miocene to Quaternary kinematic and stress directions

with published results from earthquake focal mecha-

nisms for the southern Cuban margin. These are in

agreement with active deformation resulting from ca.

E–W compression.

Acknowledgements

Y. R-A. acknowledges support through a fellow-

ship of the Austrian Academic Exchange Service. We

acknowledge useful discussions and comments on the

manuscript by Manuel Iturralde-Vinent, Guillermo

Millan Trujillo, Miguel Cabrera Castellanos, and for

thorough remarks by Dickson Cunningham which

helped to clarify data presentation and interpretation.

We also acknowledge support through grants of the

Stiftungs- und Ffrderungsgesellschaft of Salzburg

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180 179

University and the Institute of Geology and Palae-

ontology in Havana, Cuba which supported field work

in Sierra Maestra. This is a contribution to IGCP-

Project 433 (Caribbean plate tectonics).

References

Alvarez, J., Blanco, P., Medvedev, S.V., Menende, L., Shteynberg,

V.V., 1973. The seismic conditions of Santiago de Cuba. Bull.

Acad. Sci. USSR, Earth Phys. 5, 320–324.

Anderson, J.G., 1979. Estimating the seismicity from geological

structure for seismic-risk studies. Bull. Seismol. Soc. Am. 69,

135–158.

Angelier, J., 1979. Determination of the mean principal direction of

stresses for a given fault population. Tectonophysics 56, 17–26.

Angelier, J., 1989. From orientation to magnitudes in palaeostress

determination using fault slip data. J. Struct. Geol. 11, 37–50.

Angelier, J., Mechler, P., 1977. Sur une methode de recherche des

contraintes principales egalement utilisable en tectonique et en

seismologie: la methode de diedres droits. Bull. Soc. Geol. Fr.

19/6 (7), 1309–1318.

Armijo, R., Carey, E., Cisternas, A., 1982. The inverse problem in

microtectonics and the separation of tectonic phases. Tectono-

physics 82, 145–160.

Bresznyanszky, K., Franco, G., Radocz, G., 1983. Perfiles

comparativos de las areas de Cabo Cruz y Maisı. In: Nagy, E.

(Ed.), Contribucion a la geologıa de Cuba oriental. Editorial

Cientıfico-Tecnica, Havana, pp. 169–172.

Brfdermann, J., 1940. Determinacion geologica de la Cuenca

Vento. Rev. Soc. Cub. Ing. 34, 272–315.

Brune, J.N., 1969. Seismic moment, seismicity, and rate of slip

along major fault zones. J. Geophys. Res. 73, 777–784.

Cabrera-Castellanos, M., Garcıa-Delgado, D., Rojas-Agramonte, Y.,

Reyes-Perez, C., Rivera-Alvarez, Z., 2003. Depositos Cuater-

narios al sur de la Sierra Maestra: V. Congreso de Geologıa y

Minerıa, Havana, Cuba, CD-ROM. ISBN: 959-7117-11-8.

Caine, J.S., Evans, J.P., Forster, C.B., 1996. Fault zone architecture

and permeability structure. Geology 24, 1025–1028.

Calais, E., Mercier de Lepinay, B., 1990. A natural model of active

transpressional tectonics, the en echelon structures of the Oriente

Deep along the northern Caribbean transcurrent plate boundary

(southern Cuban margin). Rev. Inst. Fr. Pet. 45, 147–160.

Calais, E., Mercier de Lepinay, B., 1991. From transtension to

transpression along the northern Caribbean plate boundary off

Cuba: implications for the Recent motion of the Caribbean plate.

Tectonophysics 186, 329–350.

Calais, E., Mercier de Lepinay, B., Renard, V., Tardy, M., 1989.

Geometry and tectonic regime along a transcurrent plate

boundary: the northern Caribbean border from Cuba to Hispa-

niola, Greater Antilles. C. R. Acad. Sci. Paris 308, 131–135.

Calais, E., Mercier de Lepinay, B., Bethoux, N., 1990. Stress/

kinematics relations along a lithospheric strike-slip fault: the

northern Caribbean plate boundary from Cuba to Hispaniola.

C. R. Acad. Sci. Paris 311, 1259–1266.

Calais, E., Perrot, J., Mercier de Lepinay, B., 1998. Strike-slip

tectonics and seismicity along the northern Caribbean plate

boundary from Cuba to Hispaniola. In: Dolan, J.F., Mann, P.

(Eds.), Active strike-slip and collisional tectonics of the northern

Caribbean plate boundary zone, Geol. Soc. Amer. Spec. Pap.,

vol. 326, pp. 125–141.

Chuy Rodrıguez, T., 1999. Macrosısmica de Cuba y su aplicacion

en los estimados de peligrosidad y microzonacion sısmica,

Unpubl. dissertation, Univ. Santiago de Cuba.

Davison, I., 1994. Linked fault systems; extensional, strike-slip and

contractional. In: Hancock, P.L. (Ed.), Continental deformation.

Pergamon Press, Oxford, pp. 121–142.

DeMets, C., Jansma, P.E., Mattioli, G.S., Dixon, T., Farina, P.,

Bilham, R., Calais, E., Mann, P., 2000. GPS geodetic constraints

on Caribbean–North American plate motion. Geophys. Res.

Lett. 27, 437–440.

Franco, G.L., 1976. In: Nagy, E., et al., Texto explicativo del mapa

gelogico de la provincia de Oriente a escala 1:250 000. Inst.

Geol. Palaeont., Ministry of Industry and Basic Research,

Havana, Cuba (unpubl.).

Gamond, J.F., 1983. Displacement features associated with fault

zones: a comparison between observed and experimental

models. J. Struct. Geol. 5, 33–45.

Gamond, J.F., 1987. Bridge structures as sense of displacement in

brittle fault zones. J. Struct. Geol. 9, 609–620.

Hernandez, J.R., Gonzalez, R., Arteaga, F., 1989. Diferenciacion

estructuro-geomorfologica de la zona de sutura de la micro-

placa cubana con la Placa Caribe. Editorial Academia, La

Habana, pp. 48.

Hernandez Santana, J.R., Dıaz Dıaz, J.L., Magaz Garcıa, A.,

Lilienberg, D.A., 1991. Evidencias morfoestructuro-geodinami-

cas del desplazamiento lateral siniestro de la zona de sutura

interplacas de Bartlett. Morfotectonica de Cuba Oriental,

Havana, pp. 5–9.

Hoernle, K., Hauff, V., van den Bogaard, P., 2004. 70 m.y. history

(139–69 Ma) for the Caribbean large igneous province. Geology

32, 697–700.

Iturralde-Vinent,M.A., 1991. Deslizamientos y descensos del terreno

en el flanco meridional de la Sierra Maestra, Cuba sudoriental.

Morfotectonica de Cuba Oriental, Havana, pp. 24–27.

Iturralde-Vinent, M.A., 1996. Introduction to Cuban geology and

geophysics. In: Iturralde-Vinent, M.A. (Ed.), Cuban Ophiolites

and Volcanic Arcs. International Geological Correlation Pro-

gramme 364, Spec. Contrib., vol. 1. Museo Nacional de Historia

Natural, Havana, Cuba, pp. 3–35.

Iturralde-Vinent, M.A., 1998. Synopsis de la constitucion geologica

de Cuba. Acta Geol. Hisp. 33, 9–56.

Iturralde-Vinent, M.A., 2003. Ensayo sobre la paleogeografıa del

Cuaternario de Cuba. V Congreso de Geologıa y Minerıa,

Havana, Cuba, CD-ROM: ISBN-959-7117-11-8 (www.ig.

iutexas.edu/caribplate/caribplate.html).

Iturralde-Vinent, M.A., Macphee, R.D.E., 1999. Paleogeography of

the Caribbean region implications for Cenozoic biogeography.

Bull. Am. Mus. Nat. Hist. 238, 0–95.

Kartashov, I.P., Cherniajowski, A., Penalver, L.L., 1981. Estrati-

grafıa de los depositos Plioceno-Cuaternarios de Cuba. Nauka

Publ. House, Moscow. 145 pp.

Y. Rojas-Agramonte et al. / Tectonophysics 396 (2005) 161–180180

Kuzovkov, G., Zinchenko, V., Alcaide, J., Duranona, D., Cobian, T.,

Mendez, I., Rodrıguez, R., Sanchez, R., Guerra, M., Genos, B.,

Lay, M., Mondelo, F., Rodrıguez, M., 1988. Informe sobre el

levantamiento geologico a escala 1:50 000 y busqueda

acompanante ejecutada en el area de la Gran Piedra al este de

Santiago de Cuba en los anos 1983–1987. CNFG. La Habana

(Inedito).

Leroy, S., Mauffret, A., Patriat, P., Mercier de Lepinay, B., 2000. An

alternative interpretation of the Cayman trough evolution from a

reindenification of magnetic anomalies. Geophys. J. Int. 141,

539–557.

Magaz Garcıa, A.R., 1989. Mapa hipsometrico de Cuba a escala 1:1

000 000. In: Atlas Nacional de Cuba. IV. 1.2.3.

Mandl, G., 1988. Mechanics of tectonic faulting. Elsevier,

Amsterdam, p. 407.

Mann, P., Burke, K., 1984. Neotectonics of the Caribbean. Rev.

Geophys. Space Phys. 22 (4), 309–362.

Mann, P., Calais, E., Ruegg, J.C., DeMets, Ch., Jansma, P.E.,

Mattioli, G.S., 2002. Oblique collision in the northeastern

Caribbean from GPS measurements and geological observa-

tions. Tectonics 21, 1057.

Marret, R., Almendinger, R.W., 1990. Kinematic analysis of fault

slip data. J. Struct. Geol. 12, 596–612.

Molnar, P., Tapponnier, P., 1978. Active tectonics of Tibet.

J. Geophys. Res. 83, 5361–5375.

Moreno, B., Grandison, M., Atakan, K., 2002. Crustal velocity

model along the southern Cuban margin: implications for the

tectonic regime at an active plate boundary. Geophys. J. Int.

151, 632–645.

Perez Perez, C.M., Garcıa Delgado, D., 1997. Tectonica de la

Sierra Maestra (Sureste de Cuba). In: Furrazola-Bermudez, G.,

Nunez Cambra, K. (Eds.), Estudios sobre Geologıa de Cuba.

Centro Nacional de Informacion Geologica, Havana, Cuba,

pp. 464–476.

Petit, J.P., 1987. Criteria for the sense of movement on fault surfaces

in brittle rocks. J. Struct. Geol. 9, 597–608.

Poey, A., 1855. Tableau chronologique des tremblements de terre

ressentis a l’ile de Cuba, 1551–1855. Paris, France.

Pubellier, M., Mauffret, A., Leroy, S., Marie Vila, J., Amilcar, H.,

2000. Plate boundary readjustment in oblique convergence:

example of the neogene of Hispaniola, greater Antilles.

Tectonics 19, 630–648.

Reiter, F., Acs, P., 1996. TectonicsFP. Software for structural

geology. Innsbruck University, Austria. http://go.to/TectonicsFP.

Remane, J., Cita, M.B., Dercourt, J., Bouysse, P., Repetto, F., Faure-

Muret, A. (Eds.), 2002. International Stratigraphic Chart.

Riedel, W., 1929. Zur Mechanik geologischer Brucherscheinungen.

Zentralbl. Mineral., Geol. Palaontol. 1929B, 354–368.

Rojas-Agramonte, Y., 2003. Tectonic evolution of the Sierra

Maestra mountain range, Cuba: from subduction to arc-

continent collision and transform motion. PhD thesis, Faculty

of Natural Sciences, University of Salzburg, p. 146.

Rojas-Agramonte, Y., Neubauer, F., Krfner, A., Wan, Y.S., Liu,

D.Y., Garcia-Delgado, D.E., Handler, R., 2004. Geochemistry

and age of late orogenic island arc granitoids in the Sierra

Maestra, Cuba: evidence for subduction magmatism in the early

Palaeogene. Chem. Geol. 213, 307–324.

Rosencrantz, E., Mann, P., 1991. SeaMARC II mapping of

transform faults in the Cayman trough, Caribbean sea. Geology

19, 690–693.

Rosencrantz, E., Ross, I.R., Sclater, J.G., 1988. Age and

spreading history of the Cayman trough as determined from

depth, heat flow and magnetic anomalies. J. Geophys. Res.

93, 2141–2157.

Rueda Perez, J.S., Arango Arias, E.D., Lobaina Teruel, A., 1994.

Algunos resultados del estudio de los movimientos recientes de

la corteza terrestre en el poligono geodinamico Santiago de

Cuba. Edition ORISOL, Instituto Cubano de Geodesia y

Cartografıa, Holguın, Cuba, 20 pp.

Shantzer, E.V., Petrov, O.M., Franco, G.L., 1975. Sobre las

formaciones costeras del Holoceno en Cuba. Las terrazas

pleistocenicos de la region Habana- Matanzas y los sedimentos

vinculados a ellas. Academia de Ciencias de Cuba. Ser. Geol.

21, 1–26.

Stewart, I.S., Hancock, P.L., 1994. Neotectonics. In: Hancock,

P.L. (Ed.), Continental Deformation. Pergamon Press, Oxford,

pp. 370–409.

Sylvester, A., 1988. Strike-slip faults. Geol. Soc. Amer. Bull. 100,

1666–1703.

Twiss, R.J., Unruh, J.R., 1998. Analysis of fault slip inversions: do

they constrain stress or strain rate? J. Geophys. Res. 103,

12205–12222.

Vaughan, T.W., 1919. Fossil corals from Central America, Cuba and

Puerto Rico with an account of the American tertiary,

Pleistocene and recent corals reefs. U.S. Nat. Mus. Bull. 103,

189–525.

Wilcox, R.E., Harding, T.P., Seely, D.R., 1973. Basic wrench

tectonics. Am. Assoc. Pet. Geol. Bull. 57, 74–96.

Zolnai, G., 1988. Continental wrench-tectonics and hydrocarbon

habitat. Amer. Assoc. Petrol. Geol., Mediterranean Basins

Conference and Exhibition, Nice, France. 37 pp.