Crustal structure of eastern Cuba, derived by constrained 3D gravity inversion

Eduardo Diego Arango-Arias*^,**, Marco Antonio Pérez-Flores** and José Alberto Batista-Rodríguez***

* Centro Nacional de Investigaciones Sismológicas Calle 17 No. 61, Rpto. Vista Alegre Santiago de Cuba, Cuba

** Centro de Investigación Científica y de Educación Superior de Ensenada Applied Geophysics Department Carr. Ensenada-Tijuana, 3918 Zona Playitas, Ensenada Baja California, México. *Corresponding author: elgato601115@gmail.com

*** Universidad Autónoma de Coahuila, Escuela Superior de Ingeniería. Blvd. Adolfo López Mateos 26800 Nueva Rosita Coahuila, México

Received: February 21, 2013.
Accepted: October 08, 2013.
Published on line: July 01, 2014.

Resumen

Se obtuvo un modelo tridimensional de la corteza mediante un proceso de inversión de datos gravimétricos para la región oriental de Cuba. Los datos y el modelo cubren un área rectangular de 64 600 km². El modelo inicial fue constreñido con la geología de superficie, la información sísmica y de perforación. Se aplicó un algoritmo de inversión que utiliza los datos de gravedad para estimar las topografías 3D a partir de las unidades geológicas principales. El modelo nos proporciona información cuantitativa sobre las profundidades y espesores de las formaciones geológicas más importantes. En el mismo se observan las secuencias alóctonas de diferente composición y origen sobre el basamento carbonatado de la Plataforma de Bahamas. La mayoría de los máximos en la anomalía de la gravedad se deben a la presencia de mantos más densos de ofiolitas poco profundas. Se destaca al suroeste el máximo gravimétrico provocado por la presencia de la corteza oceánica más densa generada en el Centro de Dispersión de Caimán.

Palabras clave: corteza, inversión, gravedad, ofiolitas, acreción.

Abstract

Key words: Eastern Cuba, crustal structure, constrained inversion, gravity, ophiolites, accretion.

Introduction

Models of genesis and evolution for the Caribbean-Cuba region show little agreement; in-situ (Giunta et al., 1997,James, K.H; 2003) and allochthonous (Pindell et al., 1990, 2009; Iturrade-Vinent, 1998, 2002; Cobiella, 2005; García-Casco et al., 2008 and Sommer et al., 2011). Allochthonous models indicate the area is formed by fragments of the ancient Caribbean plate that overrode the continental Bahamas margin. Over this margin are the Paleocene Volcanic Arc (PVA) and Neogene-Quaternary sediments forming basins. Other studies took into consideration seismic refraction, gravity, deep drilling boreholes and satellite images. Studies from 30 years ago showed that the island is on transitional crust, with 17 to 30 km thick and characterized by three layers: an upper volcanic-sedimentary layer with P velocity of 4.0-4.8 km/s, a lower layer with P of 5.8-6.4 km/s and the deepest layer with P of 6.3-6.7 km/s (Bovenko et al., 1982; Otero et al., 1998; Bush and Shcherbakova, 1986). Alternatively, Eastern Cuba is constituted to the south by an oceanic crust and to the north by continental crust (Tenreyro et al., 1994). Otero et al. (1998) based on re-interpreted seismic and gravity data proposed that the crust southward of Cauto-Nipe (Figure 1) is oceanic, ~20 km thickness, and below the basin there is a fine-transitional crust with 20 to 30 km thick and northward continental crust. Through teleseismic data, Palau et al. (2006) determined the presence of a 1 km thick shallow layer with P velocity of 3.6 km/s, underneath a 6 km thick layer with P of 5.8 km/s and a deeper 13 km thick layer with 6.9 km/s for P. Recently, Gonzalez et al. (2011) defined a 16 to 30 km earth crust thickness for eastern Cuba, through joint inversion of Rayleigh waves dispersion and receptor functions. All these models are assumed one-dimensional or horizontal stratified.

Such varied results and explanations indicate that a further research is required. The main purpose of our research is to obtain the structure of the earth crust for Eastern Cuba using a dense gravity data. We extended a method that was previously used in a reduced area of Eastern Cuba with mining purposes and using only magnetic data (Batista et al., 2007). We used the Gallardo et al. (2003) algorithm, which minimizes the quadratic norm of differences between gravity data and the model response, constraining the solution or model with the surface geology, boreholes (~3.5 km as maximum) and seismic reflection profiles. Every unit is simulated with a regular mesh of 10 km x 10 km prisms. Along the iterative inverse process, the depths to every prism is moved automatically in order to fit the gravity data and restricted to obey the constraints imposed.

The study area is a rectangle of 190 km by 360 km that covers the southeast of Cuba. A large land area and two pieces of oceanic crust are involved; the Atlantic Ocean and the Caribbean Sea where the Oriente Fault Zone (OFZ) is located (Figure 1). The area limits in "Southern Cuba coordinates system" are (100,000; 290,000) m North and (420,000; 760,000) m East. The rectangular region was divided into a grid of prisms with 10 km x 10 km on surface as shown in Figure 1.

Geologic context

Iturralde (1998) recognizes two levels of the geological structure of Cuba: the substrate folding and the neo-autochthonous. Each consisting of different geological units.

In our study area (Figure 2), the folded substrate is composed by: ophiolites, Cretaceous and Paleogene Volcanic Arcs. Ophiolites (West-East strips) located at the north, are over the Bahamas platform and under the Cretaceous Volcanic Arc (CVA). Sometimes, ophiolites and CVA are mixed forming an ophiolitic mélange. It is assumed that this ophilites were emplaced when the collision between the extinct CVA and the Bahamas Platform occurred.

Ophiolites eastward the area (Figure 2) are located in the Mayarí-Sagua-Baracoa massif. These ophiolites sheets are over olistostromes and over the CVA.

The Albian-Campanian volcanic arc is Cretaceous in age. It lies in tectonic contact with the northern ophiolites. Near the contact, the arc rocks are even more deformed, with fissured and foliated areas, and with chaotic masses that contain a mixture of ophiolites blocks, vulcanite and plutons.

For simplification, we refer to this volcanic arc as CVA. This is constituted by volcanogenic-sedimentary complexes (calcalkaline and alkaline composition), plutons and the metamorphic complex.

The Paleocene volcanic arc (VPA) is a characteristic of the oriental south portion of the island (our area). Its age goes from the upper Daniense to the lower Eocene in the western part of Cuba, Jamaica, Hispaniola, Puerto Rico and Virgin Islands. The Paleocene arc was formed over the deformed remains of the CVA-Ophiolites units. VPA is constituted by volcano-sedimentary and plutonic rocks with different composition.

Neo-autochthonous units are represented by sedimentary rocks originated from the upper Eocene to recent. Three sedimentary cycles can be recognized; first, a stadium of the upper Eocene to the Oligocene, second, the lower to upper Miocene, and the Pliocene to the recent.

Method

Gravity data consist of a rectangular mesh of 340 km East direction and 190 km North direction, with interpolated data every 3 km. This design is optimal because we are looking for low spatial frequency structures. This produces at least nine observations over every 10 km x 10 km prism. Data was collected and processed by Instituto de Geología y Paleontología de Cuba (IGP) and represents the complete Bouguer anomaly (Blakely, 1996), using 2.3 gr/cm³ for the earth crust density (Figure 3).

The inversion process moves the top and bottom depth for every single prism at every geological unit. Restrictions are imposed to not allow overlap or spaces between prisms. The quadratic norm of the differences is minimized between data (g_o) and model response (g_r) plus a smoothing term (equation 1).

subject to

Where m is the unknown vector containing the depths from every prism. Matrix D is the horizontal (x, y) first derivatives of the depths. This term minimizes the top depths differences between adjacent prisms. Term β magnifies or dismisses this term. When it is zero the model shows very rough top topography for every unit; when large, every topographic unit looks very smooth, except where the data (first term in equation 1) requires larger jumps. This can happen where geological faults are located.

Depth determination is quoted by means of quadratic programming (Gill et al., 1986), using inequalities or equalities. This allows introduction of surface geology, wells and seismic data as constraints.

Surface geology is introduced as a priori information. Figure 4 shows that unit-1 prisms are displaced to allow units-2 and 3 prisms outcropping. This is performed in the algorithm by collapsing the prisms' thickness to zero.

For the modeling, the horizontal prism area was fixed in 10 km x 10 km as shown in Figure 1, giving a set of 646 prisms for every geological unit.

In order to reduce the non-uniqueness, density contrasts are considered as known. Densities were obtained by direct sampling on the surface. Those densities have a variance range due to heterogeneities inside the geological unit. In the inversion process we adjusted the density contrasts along those ranges. The inversion is not completely automatic because we had to try different density contrasts as fine adjustments.

Results and discussions

The most recent hypothesis about the Southeastern Cuba states the crustal structure consists of folded basement overriding Bahamas platform. The folded basement is constituted of ophiolites thin sheets intercalated over and under the Cretaceous Volcanic Arc (CVA). At the Southeast of our study area, the Paleogene Volcanic Arc (PVA) rocks are predominant and lay over the ophiolites flakes and CVA package (Iturralde-Vinent 1998; Iturralde-Vinent et al. 2002; Cobiella et al. 2011). These geological models were constructed from petrological data taken from surface rocks. The geological unit at depth is inferred with a large probability error that increases with depth. In contrast, our 3D model obtained from gravity data has a quantitative character. We can estimate depths, thickness and dip angles with some probability error that also increases with depth, but these errors are smaller than the geological ones. Thus the algorithm takes the surface geology and boreholes information as geologists would and then guides the 3D structures at depth obeying the physics behind the gravity data. Our procedure reduces the uncertainly at depth considerably.

Before presenting the 3D model obtained, it is important to show the model response (Figure 5B) (5A) and the differences between data and response (Figure 5C). Model response has a 6% data misfit. We reproduced the gravity highs (from 140 to 214 mGal), and also the Cauto-Nipe basin gravity low at the NW with values from 0 to 10 mGal. The differences map has a minimum and maximum of -4 to 4 mGal and shows almost a random behavior. The main differences concentrate at the south border of the island because the 10 km x10 km prisms do not fit exactly the steep coastline (Fig 1). There is a surface excess or deficiency of mass producing a misfit. Excluding that border misfit, general misfit must be lower than 6%.

In Figure 6 is shown the 3D density model obtained. We show the bottom topography for every geological unit. Figure 6A (6B, C, E, F, G, H) shows the surface topography and bathymetry. This surface is known and therefore constrained. Sea effect was taken in consideration. According with the units shown in Table 1, sea bottom topography corresponds with sediments top topography, sediments bottom corresponds with PVA top and so on. Top mantle topography is not shown because it corresponds with the bottom oceanic crust depth. In the bottom depth maps, positive value means above and negative value below sea level are shown. For example: PVA outcrops at Sierra Maestra, therefore showing positive elevation levels at sediments bottom topography (Figure 6A). While these contour plots are not visually informative, they can be digitized and used for future research.

For Central and Eastern Cuba (see oval in Figure 7), Otero et al. (1998) argued that there is an oceanic crust transitioning to continental northward. In order to test that hypothesis, we obliged the 3D model to put oceanic crust at the very bottom of the crust (denser than continental). The gravity response was so high that the algorithm never got a good convergence as we can see in Figure 7B. If we put a continental crust instead (Bahamas platform), the fitness is optimal as we can see in Figure 7A. We therefore postulate that Bahamas platform must be there because it is less dense than oceanic crust.

Figure 8A shows the surface geology declared at every surface prism. It also shows the location of four cross-section of the 3D model. Figure 8B show those cross-sections. At cross-section AA' and BB', we can see that the Bahamas platform dips inside the mantle with a slope close to 45_o. However at the East (cross-section CC'), the same platform arises forming the Mayarí-Moa-Baracoa massif. Cuevas (1998) believes that this massif was raised by isostatic compensation movements. Chang (2003) argues that the Mohorovicic boundary is moving upward. Based on our model, the latter hypothesis is more probable, because we do not see isostatic roots in cross-section DD'.

The Bahamas platform belongs to the North-America continental plate (Figure 1). Cross-section AA' (Figure 8) shows clearly how this platform penetrates the mantle in a kind of slab with an approximate thickness of 10 to 12 km, dipping southward. It begins with a low angle at North, where it almost outcrops, increasing the angle southward to 45_o below the Cauto-Nipe basin. Cross-section BB' is similar but the PVA is less overlapped. Cross-section CC' shows a rebound or vertical uplift of the Bahamas platform. It seems that the PVA is distributed more at the West of this part of the island. Iturralde-Vinent (1998), Cobiella (2005), García-Casco et al. (2008) and Sommer et al. (2011) have suggested this slab before, but this is the first geophysical evidence. The Bahamas platform density is very close to 2.3 g/cm³, meaning that density contrast is almost zero. The shallow presence of Bahamas platform at NW (~2 km) justifies the low gravity values.

At cross-section AA', the Cauto-Nipe fault is signed as a high depth gradient. This fault cuts the northern side of the Sierra Maestra massif, dipping northward.

Southward in the Cuba island, the Oriente fault is present (Figure 1). The fault signals the change from continental to oceanic crust. Even southward of cross-section AA' (Figure 8B), the 3D model cuts this fault. The high gravity anomaly requires a denser body which could correspond to oceanic crust. This new oceanic crust has been shifted from West to East by the Caiman trench. The gravity anomaly decreases southward in cross-sections BB' and CC' (Figure 8B). The 3D model justifies this with the less dense Gonave continental micro plate (Calais et al. 2002, 2006).

Cross-section BB', at the center cuts an ophiolite body of 4 km thick. This high density body causes the Levingstone high gravity anomaly (number 1 at Figure 3). Levingstone had been interpreted as caused by an ultrabasic mantle intrusion (Otero et al. 1998). We sought to understand the origin of such gravity anomaly highs (Figure 3). We made two additional cross-sections over the 3D density model. Cross-section EE' (Figure 8B) crossed Levingstone, La Guira and Piloto gravity highs (number 1, 2 and 3 at Figure 3). Figure 8A shows a great correspondence between the ophiolites bodies and the gravity anomaly highs. This is explained because the density contrast is +0.75 gr/cm³ (Table 1) and also because those bodies are shallower producing three high-frequency features over the gravity anomaly. Cross-section CC' passes over the El Salvador gravity high (number 4 at Figure 3). Despite the smoothing (Figure 8), it is clear that ophiolites at km 80 are producing the gravity highs.

Cross-section FF' (Figure 9) passes over La Perrera gravity high (number 5 at Figure 3). Figure 9B shows that in this case the ancient ocean crust keeps strong correlation with this anomaly.

The Eastern Island has a more complex geology, including the Mayarí-Moa-Baracoa ophiolitic massif. Iturralde-Vinent (1998), Cobiella (2005) and Sommer et al. (2011) suggest that the evolution of this area is different. Cross-section CC' (Figure 8B) shows that the Bahamas carbonate basement raised up, forming a horst bellow the Mayarí-Moa-Baracoa massif. Here, crustal thickness is 15 km.

The southern corner of cross-section CC' (Figure 8B) reached the deformed Santiago de Cuba belt, formed by the rising of the ancient ocean crust. Tectonically this was due to a transpressive process produced by the oblique contact between the Caribbean plate and Gonave microplate. It is important to emphasize that here the Gonave microplate is constituted by the CVA and ophiolites over the ancient oceanic crust that have been migrating from the Yucatan basin and is displaced by the Oriente fault to the present position. Similar crust composition is observed at La Española Island (Case et al. 1990) where ophiolites and CVA outcrop.

Current discussion seeks to explain whether the Bahamas platform and Caribbean plate collisioned or subducted each other. Iturrade-Vinent (1998), Cobiella (2005) and Sommer et al. (2011) argue that the Bahamas platform subducted under the Caribbean plate (see cross-section AA' at Figure 8B). García-Casco et al. (2008) argues for a collision. From our results, we propose a tectonic emplacement. At Manaibon elevation, the ophiolites can be seen over the Bahamas platform. At Manaibon and Cupey Sierra, olistolites appear inside the ophiolites, indicating that the ophiolites formation dragged over the top of the Bahamas platform when the overriding process occurred. Thus a low density Bahamas platform was overridden by a denser Caribbean plate. Since normal subduction phenomena occurs when a denser layer is overridden by a less dense layer, in this case, collision occurred rather than subduction.

When we conceived the conceptual model, we assumed that might be in the bottom, remains of an ancient crust and an oceanic crust younger (CSC) from the area of generation (Cayman Trench). Both were modeled as a single unit because they have approximately the same density, but in the same cross-section AA' (figure 8) they are shown as two types oceanic crusts.

A limitation of this research is that we used 10 km x 10 km prisms area, therefore, we cannot resolve high spatial frequencies. Nevertheless, the gravity anomaly is very smooth (Figure 3). There is no high frequency information in such a gravity anomaly. Another limitation is that we could not differentiate the Cuban metamorphic complex from the Cretaceous volcanic arc because densities are similar, and also between sedimentary volcanogenic sequences (Paleogene age).

Three-dimensional gravity inversion can be done using the same size grid of prisms and assuming densities as unknowns implying a lineal process with a single iteration solution, but the non-uniqueness is quite high by the huge number of prisms. In this research, our inversion procedure considers depths as unknown and densities as known, bringing a non-lineal problem that must be linearized by an iterative process. In this inversion, the non-uniqueness diminishes considerably because the number of prisms is much smaller. However, we also applied geological and geophysical constraints to diminish such non-uniqueness even further. We used the Occam's razor criteria that warrants a most probable 3D model.

The uncertainty inside the model is not constant. It increases with depth but also at the 3D model edges because of the lack of data. High frequency details may also have high uncertainty because the gravity anomaly is very smooth. Consequently, we looked for the simplest but the most probable 3D model. For scale purposes, 10 km x 10 km area prisms were optimal.

The smoothing parameter or roughness penalization avoids large depth steps between contiguous prisms. However, if the resolution of gravity data is optimal, the fitness term becomes more important than the smoothing term. In our 3D density model, large depth steps appear when data require them, because we are using an intermediate value for the smoothing parameter.

Parallel 3D topographies mean linear correlations between the unknowns. Problem increases when increasing the prism number and also with deeper formations. However, our model does not exhibit high linear correlation at the deepest part, because we did not use a large number of prisms.

The gravity response from the 3D density model obtained reproduces very well the general shape of the data, particularly the gravity lows at Cauto basin (NW of the map), with values between 0 and 10 mGal. It also fits the named gravity maximums. We determined that those maximums are caused by the presence of shallow ophiolite sheets mainly.

Crustal thickness obtained from the 3D model is 10 to 15 km southward and thicker northward. This result is agree with previous qualitative geological models.

The density model further shows the complex 3D topographies of the Bahamas platform, the Cretaceous volcanic arc, the Paleogene volcanic arc, and the ophiolites sheets, and how they intrude each other.

The 3D density model shows the new oceanic crust from the Gonave microplate (SE of the area) caused by the pull of the Cayman spreading center.

The response of the ophiolites sheets and Cretaceous volcanic arc overriding the Bahamas platform provides evidence for collision rather than subduction between the ancient Caribbean plate and the Bahamas platform (present North-American plate).

Acknowledgements

We thank the Institute of Geology and Paleontology of Cuba and especially Eric Escobar and Fernando Mondelo for giving us the data of the gravity anomaly used in this study.

References

Arango E.D., 1996. Geodynamics in the region of Santiago de Cuba, on the plate boundary between the North America and the Caribbean), MSc thesis, National Polytechnic Institute of Mexico, Internal Report of the National Center for Seismological Research, Santiago de Cuba, 110pp.         [ Links ]

Arnaiz-Rodríguez M., Garzon Y., 2012. Nuevos mapas de anomalías gravimétricas del Caribe. Rev. Interciencia, 37, 3.         [ Links ]

Batista J., Pérez M., Quiroga G., Gallardo l., 2007, Case History. Geometry of ophiolites in eastern Cuba from 3D inversion of aeromagnetic data, constrained by surface geology. Geophysics, 72, 3, B81–B91.         [ Links ]

Blakely R.J., 1996, Potential Theory in Gravity and Magnetic Applications: Cambridge, UK, Cambridge University, Press.         [ Links ]

Bovenko V.G., Shcherbakova B.E., Hernández H., 1982, Novyye geofizicheskiye dannyye o glubinnour stroyenii vostochnoy kuby Sovetskaya Geologiya, 9, 101–109. Translation in International Geology Review, 24, 1155–1162.         [ Links ]

Bush V.A., Shcherbakova I.N.,1986. New data on the deep tectonics of Cuba, Geotectonics, 20, 192–203.         [ Links ]

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 Active Strike-Slip and Collisional Tectonics of the Northern Caribbean Plate Boundary Zone. Geol. Soc. Am., Special Paper 326.         [ Links ]

Calais E., Han J.Y., Demets C., Nocquet J.M., 2006, Deformation of the North American plate interior from a decade of continuous GPS measurements. Journal of Geophysical Research, 111, B06402.         [ Links ]

Calais E., Mazabraud Y., Lépinay B.M., Mann P., Mattioli G., Jansma P., 2002, Strain partitioning and fault slip rates in the northeastern Caribbean from GPS measurements. Geophysical Research Letters, 29, 18, 1856.         [ Links ]

Case J.E., Macdonald W., Fox P, 1990, Caribbean crustal provincies; Seismic and gravity evidence, in The Caribbean region. The Geology of North America, Edited by Dengo, G and Case, J, E. Geol. H, 328, pp. 15-36, The Geological Society of America.         [ Links ]

Chang M., 2003, Respuestas de los campos físicos en el territorio de Oriente. Compilación monográfica. Instituto de Geología y Paleontología. La Habana. Cuba. Unpublished.         [ Links ]

Cobiella-Reguera J.L., 2005, Emplacement of Cuban Ophiolites. Geologica Acta, 3, 247–268.         [ Links ]

Cuevas J.L.,1998, Estudios sobre Isostasia en Cuba: Una Caracterización y Delimitación de Zonas Potencialmente Sísmicas. Rev. Boletín Geológico y Minero, Inst. Tecnológico GeoMinero de España, 109, 3, 265-278.         [ Links ]

Cuevas J.L, Diaz L.A., Polo B., 2001, Regionalización gravimétrica en el Caribe Centro Occidental (I): Nuevos mapas de anomalías de Bouguer total y aire libre de Cuba a escala 1: 500 000 (Gravimetric regionalization in West-central Caribbean (I): New maps of total Bouguer anomalies and free-air of Cuba at 1: 500 000), Mem. GEOMIN, 93–104.         [ Links ]

Enman S.V., Belousov T.P., Marquez M.E., Rueda J.S., Jorge G.D., 1997, Recent crustal movements and morpho-structural pattern of Southeastern Cuba: Santiago de Cuba Geodynamic Research Site. Izv. Phys. Solid Earth, 1, 55–69.         [ Links ]

Gallardo L.A., Pérez-Flores M.A., Gómez-Treviño E., 2003, A versatile algorithm for joint 3D inversion of gravity and magnetic data. Geophysics, 68, 949–959.         [ Links ]

Gallardo L.A., Pérez-Flores M.A., Gómez-Treviño E., 2005, Refinement of three-dimensional multilayer models of basins and crustal environments by inversion of gravity and magnetic data. Tectonophysics, 397, 37–54.         [ Links ]

García-Casco A., Iturralde-Vinent M., Pindell M., 2008, Latest Cretaceous Collision/Accretion between the Caribbean Plate and Caribeana: Origin of Metamorphic Terranes in the Greater Antilles, International Geology Review, 50, 9, 781-809.         [ Links ]

Gill P.E., Hammarling S.J., Murray W., Saunders M.A., Wright M.H., 1986, User's guide for lssol (version 1.0): A Fortran package for constrained linear least squares and convex quadratic programming: Department of Operations Research, Stanford University technical report SOL 86-1.         [ Links ]

Gonzalez O., Álvarez L., Guidarelli M., Panza G.F., 2007. Crust and upper mantle structure in the Caribbean region by group velocity tomography and regionalization, Pure appl. Geophys., 164, 1985–2007.         [ Links ]

González O., Álvarez L., Moreno B., Panza G.F., 2011, S-Wave Velocities of the lithosphere–asthenosphere system in the Caribbean region. Pure and applied Geophysic, 169, 1-2, 101-122.         [ Links ]

Giunta G., Beccaluva L., Coltorti M., Siena F., 1997, Ophiolitic units of the southern margin of Caribbean plate in Venezuela: a reappraisal of their petrogénesis and original tectonic setting. Memorias del VIII Congreso Geológico Venezolano, tomo 1, 331-337.         [ Links ]

Iturralde-Vinent M.A., 1996, Geología de las ofiolitas de Cuba, in Ofiolitas y Arcos Volcánicos de Cuba IGCP Project 364, Special Contribution 1, 83–120.         [ Links ]

Iturralde-Vinent M.A., 1998, Sinopsis de la Constitución Geológica de Cuba. Acta Geológica Hispánica, 33, 9–56.         [ Links ]

Iturralde-Vinent M., Gahagan, L., 2002, Late Eocene to Middle Miocene Tectonic Evolution of the Caribbean: Some principles and their Implications for Plate Tectonic Modeling. In T. A. Jackson, ed., Caribbean Geology Into the Third Millennium. Transactions of the Fifteenth Caribbean Geological Conference. 47-62. Ed. Pear Tree Press Ltd.         [ Links ]

James K.H., 2003, Caribbean Plate Origin: Discussion of Arguments Claiming to Support a Pacific Origin; Arguments for an In-Situ Origin. American Association Petroleum Geologist, International. Conference, 8-9.         [ Links ]

Mondelo F., Sánchez R., Pérez P., Pardo M., 2011, Catálogo de mapas gravimétricos de la República de Cuba. Fourth Cuban convention in Earth Sciences. Congress abstracts in digital version., ISBN 978-959-7117-30-B. 1-34.         [ Links ]

Moreno B., Grandison M., Atakan K., 2002, Crustal velocity model along the southern Cuba margin. Implications for the tectonic regime at an active plate boundary. Geophys. J. Int., 151, 632–645.         [ Links ]

Otero R., Prol J.L., Tenreyro R., Arriaza G.L., 1998, Características de la corteza terrestre de Cuba y su plataforma marina. Rev. Min. Geol., 15, 31–35.         [ Links ]

Palau R., Moreno B., Blanco M., 2006, Modelo de velocidades de Cuba Oriental. Revista Geológica de América Central, 34-35, 109-119.         [ Links ]

Pérez M., Suárez F., Gallardo L., González A., Rogelio F., 2004, Structural pattern of the Todos Santos, Coastal Plain, based on geophysical data. Ciencias Marinas, 30, 2. 349–364.         [ Links ]

Pindell J.L., Barrett S.F., 1990, Geologic evolution of the Caribbean region: A plate tectonic perspective. In Dengo, G. and Case, J.E. (eds.): The Caribbean Region, The Geology of North America, 405–432, Vol. H, 40:435        [ Links ]

Pindell J., Kennan L., 2009, Tectonic evolution of the Gulf of Mexico, Caribbean and northern South America in the mantle reference frame: an update, in The Origin and Evolution of the Caribbean Plate. Geol. Soc. Lond. Spec. Publ. 328, 1–55.         [ Links ]

Rojas-Agramonte Y., Neubauer F., Garcia-Delgado D.E., Handler R., Friedl G., Delgado-Damas R., 2008, Tectonic evolution of the Sierra Maestra mountains, SE Cuba, during tertiary times: from arc-continent collision to transform motion. J. South Am. Earth Sci., 26, 125–151.         [ Links ]

Shcherbakova B.E., Bovenko V.G., Hernández H., 1978, Stroyeniye zemnoy kory Zapadnoy Kuby (Crustal structure in West Cuba), Sovetskaya Geologya, 8, 138–143. Translation in International Geology Review, 20, 1125–1130.         [ Links ]

Sommer M., Hüneke H., Meschede M., Cobiella-Reguera J., 2011, Geodynamic model of the northwestern Caribbean: scaled reconstruction of Late Cretaceous to Late Eocene plate boundary relocation in Cuba. Neues Jahrbuch für Geologie und Paläontologie – Abhandlunge (Band 259, Heft 2), 259(3), 299-312.         [ Links ]

Tenreyro R., López J.G., Echevarría G., Álvarez J., Sánchez J.R., 1994, Geologic evolution and structural geology of Cuba, AAPG Annual Convention.         [ Links ]