ALONG STRIKE CONTINUITY OF THE LEEWARD ANTILLES ARC IN NORTHWESTERN SOUTH AMERICA AND EVIDENCE OF STRAIN PARTITIONING

Eleine Melisa Vence

The university of Texas at Austin

INTRODUCTION

The interaction of the South Caribbean Plate and the South America Plate has created distinctive basement provinces that extend westward from Venezuela into the coastal and offshore area along the Caribbean margin of Colombia, where they has been poorly documented. These provinces form basement highs which are mappable at the basement level (Fig. 1) and they can be defined as well, on the basis of their gravity signature, radiometric age dating of their various rock assemblages, depth to basement as determining from mapping of seismic reflection data, well data and the type and style of sedimentary basins overlaying them.

Fig. 1. Continuity of the Caribbean Basement Provinces along the northern margin of Colombia

Key to abbreviations: VB = Venezuelan basin; SCDB = South Caribbean deformed belt; LAA = Leeward Antilles arc; AR = Aves Ridge; BB = Bonaire basin; GB = Grenada basin; TB = Tobago basin; MB = Maracaibo basin

The aim of this work is to strength the theory of the continuation of the Leeward Antilles Arc in the Northeastern margin of Colombia as it is suggested by the TWT structural regional map at the top of Acoustic Basement (fig. 1) by the integration and interpretation of Gravity, Raster and Well data, and to propose a model that explain its current geometry and arrangement based on the interpretation of Plate Velocity Vectors from GPS data.

HYPOTHESIS

If the evidences of the continuity of the LAA in northwestern South America are true, Gravity data along well and radiometric data should support that interpretation and a better constraint of an curved plate boundary can be elucidated. If this is true this curved boundary is reflected in strain partitioning in the area? And how much is the arc parallel extension?

OBJECTIVE

The objectives of this project are:

- Built an integrated database in ArcGis that will allow the identification of the Leeward Antilles arc in northwestern South America.

- To standardized the projection of the project in WSG 1984

- To Create a map of Basement terrains with the results of the Gravity and well data interpretation

- Evaluate the evidences of the Strain Partitioning in the area, based on GPS data and in the case of a positive evidence, measure the Arc parallel extension in the Leeward Antilles Arc

METHODOLOGY

1. Data compilation and evaluation of the data readily to use and the one that need further reprocessing.

2. Integration and reprocessing of Gravity data, well data and topography data that leads to the interpretation of the continuity of the Leeward Antilles Arc in northwestern South America and others basement terrains.

3. Built a personal geodatabase for storing the basement terrains interpretation and the digitization products and topology of this work.

4. Integration of the Plate velocity Vectors according to GPS data and relative to a stable South America Plate with the results of the previous map, in order to evaluate the evidences of the Strain Partitioning in the area.

5. Creation of a map that reflects the analysis of the strain partitioning in the area and the quantification of the arc parallel extension.

AVAILABLE DATA

1. Gravity data from GEOSAT (Raster and Tiff) and topography raster data were accessible from the CBTH project at the Geophysical Institute of The University of Texas at Austin, (Fig 2, 3a & 3b).
2. Regional Fault data was downloaded from the USGS web page, where there is free geological metadata for South America.
3. Radiometric ages from wells and a portion of the basement terrains interpretation in east Venezuela was taken from Gorney, 2005, (Fig. 4)

Fig. 2. Topography raster data from CBTH project.

Fig. 3a. Gravity raster data from GEOSAT. Took from CBTH project.

Fig. 3b. Gravity TIFF data from GEOSAT. Took from CBTH project. Due to this image has an additional process in other program. The interpretation of the LAA was made above this image rather than in the Raster data.

4. Wells ages and basement lithology was compiled in part from my thesis in the Guajira area, and the rest of the wells (Central Colombia) were compiled from published data (Flinch, 2003) combined with free Colombia database information in www.epis.com.co, (Fig. 5)
5. Plates velocity vectors from GPS data were consulted in different papers, however at the end, the only sourced used was Corredor, 2003. Where there is a previous compilation of the main GPS data available for the study area.

Fig. 4. Basement Provinces of Western Venezuela and Dutch Antilles, based on Gravity data interpretation and well age constraint. Gorney, 2005

DATA REPROCESSING

The data from 26 wells was compiled and an excel spreadsheet was created with that information (Fig. 6). This process implies a lot time and effort since the well data was compiled from different sources and just the wells that effectively drilled the basement were selected. This data can be imported to ArcGis as new shapefile (fig. 6)

Fig. 6. Excel Spreadsheet with the compilation of ages basement types for central Colombia wells. Below the importation product of this information reflected in Wells_Controlages shapefile.

DATA PROCESSING

1. The first step in the data processing was to define the extension of the study area. Then a Personal Geodatabase (Covers_Tesis.mdb) was created with a Feature Dataset called Basement_Provinces that includes a feature class called Map_Area, where a polygon was defined with the extension of the area of interest (fig. 7a & b).

	Fig. 7a. The figures illustrate the process of how a new feature dataset and a feature class can be created and the final structure of Cover_Tesis.mdb

Fig. 7b. View in ArcCatalog of the structure of the Cover_Tesis.mdb (personal geodatabase)

After that the data available was clipped to the Map_Area extension, using either Clip Analysis Tool for features datasets and shapefiles and Clip Data Management Tool for raster data. (fig. 8 & 9)

Fig. 8. This figure shows the Clip tool, which is found in the Arctoolbox. At the right is an example in how to clip raster with Clip Data Management tool and the feature dataset “Map_Area” defined in the step above.

				Faults clipped with Clip Analysis Tool

Fig. 9. Some examples of how the data was clipped with Clipped Analysis Tool in the case of the Faults shapefile from USGS, and the clip made to the Topography raster with Clip Data Management tool (right). The same process was applied to the Gravity data.

2. The second step was to create a map with the Gravity, topography and the well data previously imported as a shapefile. Over this map the Gorney, 2005 figure was digitized, following these steps:

- Add the Figures

- Right click in tool→ Georeferencing → Fit to Display

- Look for good markers and create a first order polynomial transformation, with at least three points

- Finally, the Rectification for the figure (fig. 10).

Fig. 10. Gorney, 2005 figure georeferenced.

3. The third step was to digitize the wells compiled by Gorney in a feature class stored in the Geodatabase. The feature class has a subtype where the basement type characteristics were stored, (fig. 11).

3. This step was to label all the wells with basement type and interpreted the stinking character and the well age. For this area the Oceanic basement will be Cretaceous and younger, and the Arc related basement is generally metamorphous. This point differentiated Arc related basement with Back Arc basement. Additionally, the basement related to the South American Plate is Jurassic and older, mostly Paleozoic. (Fig. 12)

Fig. 11. Gorney Wells digitization and creation of Gorney Wells feature class.

Fig. 12. Basement Terrains boundaries interpretation in the study area. Additionally it shows the digitization of the Trench in Faults feature class

Step 5. In this step the topology for the Terrains feature class was created and validated (Fig. 13 a). The topology is useful for identifying and corrects the mistakes in the digitization that can bring problems in the creation of a new polygon features class from lines. The topology was created in the Basement_Provinces (Dataset)→ New→Topology. This step requires that the elements of this feature dataset are not open in any ArcMap window. After that, the topology is loaded to a new map in ArcMap.

Furthermore a Domain called Basement Type was created in the Personal Geodatabase and added later to the Polygon feature class produced from the topology (fig. 13 d, e). The polygon feature class was called Bst_Terrains-poly. Based on this polygon a layer with different colors for each basement terrain was created in order to keep the color format for future maps (fig. 13c).

Fig. 13e illustrates how the Domain is incorporated in the Polygon feature class

	Fig. 13 d. The creation of a Domain in the Geodatabase

		Fig. 13a. Validation of topology in ArcGis

		Fig. 13c. The polygons were colored according to the Domain assigned.

Fig. 14. Basement Terrains map from Norwestern South America

6. For this step a new map was created in order to manage the Plate Velocity vectors separately. The Corredor, 2005 picture was georeferenced and rectify following the steps described in step 2 and the vectors were digitized following the methodology of step 3. For this purpose a new feature class called Vectors_cor was created. And the digitized vectors were under Shape_length analysis (Fig. 15). With this tool, a correlation between length and velocity magnitude was found. And it was possible to know the velocity magnitude of the vector despite that it was not stated in the Corredor, 2003 figure. (fig. 16)

Fig. 15. Shape Length Analysis to Vectors_cor Feature Class (Plate Velocity vectors).

Fig. 16. Georeferenced and rectify Corredor, 2003 figure. Please note the correlation between the color of the vector derived from Shape_Length Analysis and its magnitude.

Step 7. This in an intermediate step in the process of the construction of the vectors map is visualized in fig and consisted in overlay the feature class created with the vectors over the gravity and topography map. (see, Fig. 17)

.

Fig. 17. Plate Velocity Vectors over topography of northwestern South America and GEOSAT Gravity. Please note, that the velocity magnitudes are higher parallel to the East – West direction and are aligned with the East-West strike-slip movement in the study area.

Step 8. Evidences of Arc Parallel Extension: Having identified in the previous work (Basement Terrains map from Northwestern South America) the continuation of the Leeward Antilles Arc in the study area, it can be stated with confidence that in the western margin of South America, the interaction between the South Caribbean Plate (SCP) and South America Plate (SAC) occur through a close arched (curved) boundary. Additionally, is well know for previous studies that there has been an oblique collision of the SCP against SAC since Paleogene times when the former collided against the Bahamas platform at the north and changed its movement direction from NW to EW. These characteristics resemble the geometry found in Aleutian Island chain, described by Llalemant, 2000 (fig. 18)

Fig. 18 Description of the Arc-Parallel Extension in Aleutian Island Chain. From Lallemant, 2000

Hypothetical model for displacement partitioning and trench-parallel migration and extension of arc-forearc terrane due to oblique plate convergence along arcuate convergent plate boundary. Obliquity angle (α) of convergence-rate vector V between oceanic (Pacific) and volcanic arc (on North American plate) increases from A to C, from about 20° to 90°.Vector component (sub)normal to plate margin (Vn) decreases and component parallel to margin (Vt) increases from A to C. Vn is expressed by arc-normal shortening (arc-parallel folds and thrust faults) and Vt causes arc-forearc terrane to migrate westward along one or more strikeslip faults. As result of increase of Vt, arc-forearc terrane undergoes stretching (arc-perpendicular normal faults and fractures). Diagrams at p, q, and r are hypothetical earthquake focal mechanism plots related to thrusting, strike-slip faulting, and normal faulting, respectively.

In the case of my study area where there is oblique convergence and a curved plate boundary, the displacement is partitioned into two components. One component is essentially perpendicular to the trench and is responsible for contractional structures oriented parallel to the Plate boundary, another is oriented parallel to the plate boundary, (Fig 19). The evidences of this are supported by the orientations of the Plate Velocity Vectors that can be compared with those described by Lallemant, 2000. The normal faults products of the extension were digitized in a new shapefile and are appreciated in the figure below.

Fig. 19. Evidences of Arc- Parallel Extension in the Leeward Antilles Arc, along the northwestern margin of South America (Eastern Colombia – Western Venezuela).

Moreover, the figure above shows that the analogues to Vt vectors (E-W Vectors) are larger in magnitude (21 – 30 mm/yr) than the analogues to Vn Vectors (Perpendicular to the Trench), 25 – 30 mm/yr. This situation leads to stretching conditions along the Arc.

Step 9. One of the objectives of this work is to measure the magnitude of the extension between the Leeward Antilles Arc in the area. For that reasons lines with double arrows were digitized perpendicular to the direction of the normal fault, in a new shapefile called extension. In order to know the magnitude in kilometers of these lines, the data frame properties need to be modified, since the coordinates are set for GCS_WGS_1984 and the measurements given by this systems are in decimal degrees (Fig. 20).

Fig. 20. Coordinate system of the project, CGS_WGS_1984

Therefore, the coordinates were change to UTM zone 17 that is the UTM that covers the study area: in Data Frame properties → General, and the display was changed to kilometers. As a result, the lines with double arrows perpendicular to the faults were labeled by shape length, as it can be visualize in fig. 21.

RESULTS AND CONCLUSIONS

The main results of this work are summarized in two maps, one that shows the identification of the Leeward Antilles Arc in northwestern South America and support its continuity from western Venezuela to eastern Colombia based on its gravity characteristics and well age constraint (Fig. 14) and the second one which summarized the strain partitioning and state the magnitude of the arc parallel extension in the study area in kilometers (fig. 22).

The main conclusions elucidated from this work are:

- An integrated database in ArcGis, including gravity data, topography, well data and a series of published works, lead to a better analysis of the different geological processes and their outcomes in my study area.

- The topology is a very useful tool in order to create polygons from lines, and those polygons allow the superposition of the new information created in an easy way over previous or future maps.

- Layer→Properties→Symbology, is an useful tool for enhancing the main properties of a raster, shapefile and feature classes. Also it helps to highlight additional properties, i.e, in the case of the digitized vectors, the Shape Length help to find the velocity magnitude in the vectors that were not labeled in the original figure.

- The gravity anomalies in the northeast part of Colombia are striking characteristics of the Leeward Antilles arc and correlate with the basement highs mapped along the northern margin of South America. Additionally, this assumption is supported by the age and basement type present in the Gravity anomalies: Cretaceous and Metamorphous basement with oceanic affinity.

- The well integration, allowed the identification of others basement terrains, and the South Caribbean plate boundary. This terrain is characterized by ages from Jurassic and older, mainly Paleozoic.

	Fig. 22. Magnitude of the Arc Parallel extension in the Leeward Antilles Arc, northwestern Venezuela – northeastern Colombia

- The main implication of the identification of the continuity of the Leeward Antilles arc in Western Colombia, is the improvement of the plate Boundary constraint which shows a more arched or curved boundary.

- The Gravity anomaly found in western Colombia does not match in age with the Leeward Antilles Arc, since its correlative onland portion is younger in age. This anomalies can be interpreted an accretionary prism related to the South Caribbean deformed belt, supported with seismic information.

- The integration of the Plate Velocity vectors in the ArcGis model, lead to the identification of the additional processes underwent by the Leeward Antilles Arc in the study area and help to elucidated and analogy between this Arc and the Aleutian Island Arc, described by Llalemant, 2000.

- The direction of the vectors in relation with the South Caribbean Trench reflects a strain partitioning in the area, divided in two components. One perpendicular (shortening) and another parallel, responsible for the right lateral strike slip movement in the area and the east west migration of the arc, due to the arc parallel extension is produced by the higher magnitudes of this vector in the area compared with the perpendicular vector.

- The magnitude of the extension produced by the high magnitude in the East-West velocity vectors seems to be affected by the increasing in magnitude of the perpendicular vectors as well. Due to, there is an increase in the magnitude of the extension from 35.28 to 37.68 km, when this vector increase from 17-18 to 23-24 mm/yr

BIBLIOGRAPHY

Calais, E., et al., 2002. Strain partitioning and fault slip rates in the northeastern Caribbean from GPS measurements. J. Geophys. Res., 29, 1856-1859.

Corredor, F., 2003. Seismic strain rates and distributed continental deformation in the northern Andes and three-dimensional seismotectonics of northwestern South America. Tectonophysics 372, 147-166.

Colmenares, L., Zoback, M., 2003. Stress field and seismotectonics of northern South America. Geology 31, 721-724.

Flinch. J.F., 2003. Structural evolution of the Sinu-Lower Magdalena area (Northern Colombia), in C. Bartolini, R.T. Buffler, and J. Blickwedge, eds., The Circum-Gulf of Mexico and the Caribbean: Hydrocarbon habitats, basin formation, and plate tectonics: AAPG Memoir 79, p. 776-796

Lallemant, H. G., Oldow, J.S., 2000. Active displacement partitioning and arc-parallel extension of the Aleutian volcanic arc based on Global Positioning System geodesy and kinematics analysis. Geology 28, 739-742.

Mann, P., et al., 2006. Regional geologic and tectonic setting of the maracibo Supergiant basin, western Venezuela. AAPG Bulletin 90, 445-477.