Late Holocene marine terraces of the Cartagena region, southern Caribbean: The product of neotectonism or a former high stand in sea-level?

https://doi.org/10.1016/j.jsames.2009.08.010Get rights and content

Abstract

The detailed stratigraphic survey and paleontological study (mollusks, corals, foraminifera and ostracods) of four low-level, ∼3 m, marine terrace sections: Punta Canoas, Manzanillo del Mar, Playa de Oro, and Tierra Bomba Island, from the Cartagena region, southern Caribbean, supplemented with 22 radiocarbon dates, reveals that the northern terraces were deposited as parasequences in a clastic depositional system compared to the Tierra Bomba Island succession that was deposited in a carbonate depositional system between ∼3600 and ∼1700 cal yrs BP. Drier conditions and the southern location of the ITCZ at about 3 ka triggered stronger easterly Trades and more dynamic southwestward sediment drift fed by the Magdalena River mouth, thus promoting the formation of sand spits that ultimately isolated the Cienaga de Tesca coastal lagoon from the Caribbean Sea. Our estimates support the hypothesis that the present position of the terraces is the product of neotectonism rather than a higher 3 ka, sea-level. Upheaval of the terraces varies between ∼3.8 mmyr−1 at Punta Canoas and ∼2.2 mmyr−1 at Tierra Bomba to ∼1.5 mmyr−1 at Manzanillo del Mar and Playa de Oro terraces. Our study corroborates previous contentions on the role of mud diapirism and the dynamics of the Dique Fault as late Holocene upheaval mechanisms.

Resumen

La descripción estratigráfica detallada y el estudio paleontológico (moluscos, corales, foraminíferos y ostrácodos) de cuatro niveles bajos, ∼3 m, de las secciones de las terrazas marinas de: Punta Canoas, Manzanillo del Mar, Playa de Oro y Tierra Bomba en la región de Cartagena, sur del Caribe, suplementado con 22 dataciones radiocarbono, revela que las terrazas del norte fueron depositadas como parasecuencias en un sistema clástico comparadas con la sucesión de la isla de Tierra Bomba que fue depositada en un sistema carbonatado entre ∼3600 y ∼1700 años cal AP. Condiciones mas secas y la localización mas al sur de la ZCIT hace 3 ka resulto en vientos alisios del este mas fuertes y una deriva litoral hacia el suroeste mas dinámica alimentada por la desembocadura del Río Magdalena promoviendo así la formación de espigas de arena las cuales finalmente aislaron la Cienaga de Tesca del Mar Caribe. Nuestros estimativos apoyan la hipótesis que la posición presente de las terrazas es el producto de neotectonismo y no el de una posición del nivel del mar mas alta hace 3 ka. El levantamiento de las terrazas varia entre ∼3.8 mmyr-1 en Punta Canoas y ∼2.2 mmyr-1 en Tierra Bomba a ∼1.5 mmyr-1 en Manzanillo del Mar y Playa de Oro. Nuestro estudio corrobora propuestas previas con respecto al papel del diapirismo de lodo y la dinámica de la Falla del Dique como mecanismos responsables del levantamiento de las terrazas bajas durante el Holoceno tardío.

Introduction

Whether or not the marine terraces of the Cartagena region are the result of a former 3 m high stand in sea-level at ∼3 ka or recent tectonic upheaval, their depositional and paleo-bathymetric history should be interpreted first. This paper provides such a reconstruction and discuses neotectonic activity in a location that lies in the southern extreme of the Luruaco anticlinorium in the Sinu belt (Duque-Caro et al., 1983) where the Caribbean – South American plates converge (e.g. Taboada et al., 2000, Trenkamp et al., 2002, Flinch, 2003).

Sea-level changes in the Holocene respond to a complex balance between: (1) glacio-isostatic and hydro-isostátic effects, (2) subsidence and/or upheaval, and (3) sediment supply or erosion (e.g. Emery and Aubrey, 1991, Pirazzoli, 1996, Emery and Myers, 1996). Following deglaciation, the hydro-isostatic response of the continental margins was diverse and related to the distance to the polar ice caps of the northern hemisphere (e.g. Lambeck, 1993). Locations that where far away from former ice sheet, i.e. far-field regions such as Australia and Japan, have been studied for Holocene sea-level changes (Nakada and Lambeck, 1989, Chappell, 1987, Yokoyama et al., 1996, Lambeck, 2002). These studies reveal that most of the global ice sheets melting ceased ca. 7 ka when higher than present day sea-level indicators are observed at the far-field regions. The southern Caribbean is classified as intermediate-field, from the Laurentide ice sheets, where the ice loading component from the ice sheet is still considerable (Clark and Linge, 1979). Therefore, glacio-hydro-isostatic models predict continuous rise in sea-level throughout the Holocene including the present day highest sea-level (Lambeck et al., 2002).

Attempts to build a southern Caribbean Holocene sea-level curve include: (1) a palynological study, supported by radiocarbon dates, in eastern Venezuela (Rull et al., 1999), that in the absence of tectonic and glacio-hydro-isostatic considerations should be regarded as preliminary, (2) the dynamics of Rhizophora mangle and the precise radiocarbon dating of peat levels in Trinidad that suggest that sea-level rose from −9 m to −2 m between 7 and 2 ka (Ramcharan, 2004), and (3) a compilation of relative sea-level data and glacio-hydro-isostatic model (Milne et al., 2005) that allowed them to predict a relative sea-level curve for Curacao, i.e. a relative sea-level rise from the last glacial maximum (LGM), level to ca. 3 m at about 7 ka and the steady rise of ca. 1 m at about 5 ka.

The origin of the lower, i.e. ca. 3 m high, coastal terraces of the Colombian Caribbean remains controversial with regard to either a former high sea-level position at ca. 3 ka or recent tectonic upheaval. Historically, research on the low-level terraces (see Fig. 1 for location) has gone through: (1) stratigraphic and paleontological surveys of the ca. 3 m high Tierra Bomba Island terraces which were dated as 2850 ± 150 14C yrs BP (De Porta and De Porta, 1960, Richards and Broecker, 1963, De Porta et al., 1963), and 2800 and 369014C yrs BP, and interpreted as the result of a higher sea-level position (Burel et al., 1981, Vernette, 1989a), (2) ∼2500 and 8900 14C yrs BP dates on corals and mollusks from south of the Morrosquillo Gulf (Fig. 1c), that were compared with the hydro-isostatic curve for Brazil as a reference to estimate upheaval rates of ∼2–5 mmyr−1 and subsidence rates of ∼0.7 mmyr−1 along the coast (Page, 1983), (3) radiocarbon dates on peats, calcareous nodules, mollusks and corals collected from 11 samples from the continental shelf offshore Cartagena, that were used to construct a sea-level change curve, whose pattern is more alike type V than IV; therefore appearing to be more of a far-field type than an intermediate-field type (cf. Clark and Linge, 1979), where the effect of neotectonísm apparently was minimum (Javelaud, 1987) and, (4) a micropaleontological (foraminifera) study of 6 cores from the Barú Island (Fig. 1c) coastal lagoons and 18 cores from the continental shelf to reconstruct paleobathymetry (Parada, 1996).

Major pitfalls on these studies are the absence of stratigraphic columns, the dating of only few fossils without any control on their position in the terraces, taphonomy, and transport, uncalibrated ages or the complete absence of radiocarbon dates as is the case of Parada’s (1996) study. Furthermore, interpretations of either a +3 m high sea-level stand at 3 ka or neotectonic upheaval, in those studies, were unsupported by a geophysical or tectonic study.

This paper documents the detailed survey of ten stratigraphic columns from four terrace sections with a sequence biostratigraphy and taphonomy approach plus a paleo-bathymetric reconstruction, supported by 22 radiocarbon dates. From this information we estimate sediment accumulation and upheaval rates, and explore the role of the southward migration of the intertropical convergence zone (ITCZ; Haug et al., 2001) in the construction of the depositional sequences, and evaluate the role of local faults. Herein we present data from four stratigraphic columns only. For full details of the other columns and cores see Gomez (2005) and Delgado (2004). Details of the taphonomic study will be published elsewhere.

The Cartagena region is located in an active tectonic zone where the Caribbean and South American plates interact. The Caribbean plate moves southeastward, colliding with the South American plate at a speed of 1–2 cmyr−1 (e.g. Taboada et al., 2000; Fig. 1a). Oblique convergence initiated in the Tertiary and shaped the poly-history Lower Magdalena Basin, where two provinces are recognized, the western Sinú belt and the eastern San Jacinto belt, separated by the Sinú lineament (Duque-Caro, 1980). Oblique plate convergence, together with high accumulation of terrigenous sediments on the continental margin results in the extensive formation of mud diapirs and volcanoes, which are characteristic of the Sinú belt and the present continental margin (e.g. Vernette, 1989b, Ordoñez, 2008). Mud diapirism in the Sinú belt apparently is responsible of most of the upheaval of the coastal terraces and subsidence in the Morrosquillo Gulf area (e.g. Page, 1983, Duque-Caro, 1984, Vernette, 1989b). The Sinú belt is composed of folded Miocene to Pliocene marine rocks affected by east-dipping thrusts, mud diapirism and NW-SE faults and lineaments, e.g. the Dique Fault (Fig. 1; Duque-Caro, 1980, Vernette, 1989b) or Canoas Fault of others (e.g. Ruiz et al., 2000). The La Popa Formation, a Pleistocene reef, caps most of the hills on the Tierra Bomba and Baru islands and north of Cartagena (La Popa hill) and constitutes the ∼20 m high level terraces. By contrast, there is a low Holocene marine terrace level, which occurs at ∼3 m all over the region, and is the object of this study.

Sediment yield over the Caribbean coast is controlled by the seasonal (latitudinal) migration of the ITCZ which controls precipitation in northern South America (Fig. 1b), i.e. a dry season (November to March), a transition season (April to August), and a rainy season (August to November). Similarly, northeast Trade Winds are stronger during the dry season, when the southwesterly Caribbean Current is stronger and the wave front hit the shore obliquely thus resulting in a southwestward longshore current and sediment drift (Fig. 1c). During the rainy season, when the northeasterly Darien Counter-Current is stronger, drifting of sediments is reduced (e.g. Pujos et al., 1984). However, during this season, hurricanes that originate as tropical depressions, occasionally hit the area thus transporting sediments from the northeast, e.g. hurricane Joan that reached the Cartagena region on October 1988 (Lawrence and Gross, 1989).

Therefore, the interaction between tectonism, climate and oceanographic dynamics has resulted in a coastal setting where a number of geomorphologic units can be recognized. Among them: (1) coastal plains associated to fluvio-marine sedimentary processes, (2) coastal lagoons partially or completely isolated from the sea by coastal bars (Gayet and Vernette, 1989), (3) beaches and coastal barrier islands, (4) coral reefs and, (5) marine terraces. The predominant supply of terrigenous sediments from the northeast, i.e. the Magdalena River, has resulted in two ecosystem and depositional settings, clastic and carbonate, north and south of the Cartagena Bay (Fig. 1c; Diaz and Puyana, 1994), respectively. These different depositional systems are reflected in the composition and spatial heterogeneity of the sedimentary marine terraces, i.e. the northern Punta Canoas, Manzanillo del Mar, and Playa de Oro terraces are mostly clastic, whereas the southern Tierra Bomba Island terraces are mostly calcareous (Burel and Vernette, 1981). In this paper we reconstruct the evolution of these depositional systems for the late Holocene.

Section snippets

Methods

To capture lateral variations we surveyed ten stratigraphic successions on four terrace sections (Fig. 1c), one in Punta Canoas (10°33′23.3″N, 75°30′20″W), three in Manzanillo del Mar (10°30′44.6″N, 75°30′2.8″W), three in Playa de Oro (10°30′19.1″N, 75°30′3″W), and three in Tierra Bomba Island (10°22.39′N, 75°34. 20′W). The altitude of the terraces was measured by drawing beach profiles, connecting the base of each stratigraphic column with the highest swash mark, with the aid of a Look level

Results

Our topographic survey shows the varying altitudes of the Cartagena terraces (Fig. 2). The top of Punta Canoas, the highest terrace, lies at ∼680 cm above sea-level (cmasl), whereas the top of Playa de Oro, the lowest terrace, lies at ∼212 cmasl. The terraces were deposited between ∼3400 and ∼1600 yrs BP (Table 1). Two dates, codes 7 and 22, were excluded from our upheaval estimates. Sample code 7, collected in the Tierra Bomba terrace (TB-1 section), is considered as a reworked specimen because

Depositional environments

The above-mentioned faunal, taphonomic, and radiocarbon features for the Punta Canoas section suggest very low sedimentation rates and deposition in a shallow water environment, except for layer 4. Facial variations on the Punta Canoas stratigraphic section suggest a change from a fluvial dominated deltaic system to an estuarine and/or lagoon system (cf. Walker, 1984, Walker and James, 1992); interpretation that is supported by the molluscan assemblages, e.g. the presence of C. cancellata, C.

Conclusions

Evidence presented herein suggests that:

  • (1)

    The Punta Canoas, Manzanillo del Mar, and Playa de Oro successions (parasequences) were deposited in a clastic depositional system compared to the Tierra Bomba Island succession that was deposited in a carbonate depositional system during the late Holocene, i.e. between ∼3.6 and ∼1.7 ka.

  • (2)

    Drier conditions and the southern location of the ITCZ ∼3 ka triggered stronger easterly Trades and more dynamic sediment southwest drift fed by sediment yield from the

Acknowledgements

We specially thank the Foundation for the Promotion of Science and Technology and Universidad EAFIT for their support. YY is partly supported from JSPS grand-in-aid for scientific research and COE program of the University of Tokyo. The collaboration of Erika Siegert on our field campaign and lab preparation is greatly appreciated. We thank the Tierra Bomba Island high school (Grade 11) students and teacher who made possible our field work. We thank Camilo Ordoñez for kindly providing us his

References (63)

  • Bosscher, H., 1992. Growth potential of coral reefs and carbonate platforms. Ph.D. Thesis, Vrije University, Amsterdam,...
  • T. Burel et al.

    Evidencias de cambios del nivel del mar en el Cuaternario de la region de Cartagena. 1er Seminario sobre el Cuaternario en Colombia

    Revista CIAF

    (1981)
  • T. Burel et al.

    Indice d’une ligne de rivage émergée dans la région de Cartagena (Colombie) Les lagunes cotieres

    Bulletin de Institute Géologie du Bassin d’Aquitainey Costeras, Bordeaux

    (1981)
  • J. Chappell

    Late Quaternary sea-level changes in the Australian region

  • Correa, I.D., 1990. Inventario de acreción y erosión litoral (1793–1990) entre Los Morros y Galerazamba, Departamento...
  • J. De Porta et al.

    El Cuaternario marino de la Isla de Tierrabomba (Bolivar)

    Boletín de Geología UIS

    (1960)
  • J. De Porta et al.

    Nuevas aportaciones al Holoceno de Tierrabomba

    Boletín de Geología UIS

    (1963)
  • Delgado, A., 2004. Reconstrucción paleobatimétrica de las terrazas marinas holocenicas de la region de Cartagena...
  • Diaz, J.M., (Ed.), 2000. Areas Coralinas de Colombia. INVEMAR Spec. Publ. 5, pp....
  • Diaz, J.M., Puyana, M., 1994. Moluscos del Caribe Colombiano, un catalogo ilustrado. Colciencias – Fundacion Natura –...
  • J.M. Diaz et al.

    Reef associated molluscan fauna of the Santa Marta area, Caribbean coast of Colombia

    Boletín Investigaciones Marinas y Costeras, Invemar

    (1990)
  • H. Duque-Caro

    Geotectónica evolutiva de la región Noroccidental colombiana

    Boletín Geológico

    (1980)
  • H. Duque-Caro

    Structural style, diapirism and accretionary episodes of the Sinu-San Jacinto terrane, southwestern Caribbean borderland

    Geological Society of America Memoir

    (1984)
  • Duque-Caro, H., Page, W.D., Cuellar, J., 1983. General Geology, Geomorphology and Neotectonics of Northwestern Colombia...
  • K.O. Emery et al.

    Sea Levels, Land Levels and Tide Gauges

    (1991)
  • Emery, D., Myers, K.J., 1996. Sequence Stratigraphy. Blackwell Science, 297...
  • J.F. Flinch

    Structural evolution of the Sinu-Lower Magdalena area (northern Colombia)

    AAPG Memoir

    (2003)
  • J. Gayet et al.

    Les lagunes cotieres

    Bull. Inst. Geol. Bassin d’Aquitaine, Bordeaux

    (1989)
  • Gomez, A., 2005. Reconstruccion de cambios paleobatimétricos holocenicos de las terrazas marinas de la region de...
  • H.M. Granja et al.

    Sea-level rise and neotectonism in a Holocene coastal environment at Cortegaca Beach (NW Portugal): a case study

    Journal of Coastal Research

    (1996)
  • G. Haug et al.

    Southward migration of the intertropical convergence zone through the Holocene

    Science

    (2001)
  • Cited by (33)

    • Sedimentary records of extreme wave events in the southwestern Caribbean

      2018, Geomorphology
      Citation Excerpt :

      Such terraces are found throughout the Colombian coast. Martínez et al. (2010) reported that terraces on Tierra Bomba Island date to between 2260 and 2020 BP. Although coralline terraces of the ERIA have not been directly 14C-dated, it is thought that the similar geomorphic structures on Tierra Bomba Island provide a comparable age, given their geographic proximity and similar geodynamic environment.

    View all citing articles on Scopus
    View full text