GISdevelopment.net ---> GIS for Oil & Gas Proceedings 2001

Photogeology and GIS in oil exploration in ecuador

Marcelo Echeverría
National Polytechnic University
Department of Oil studies and Natural Gas & CEPEPOL
P.O. Box 17-01-27-59, Quito, Ecuador
PHONE: 011-593-2- 507127 FAX: 011-593-2-449476
Email: mechever@mail.epn.edu.ec, jmecheve@ecnet.ec

Wibold Jongsma
Department of Informatics and Computer Sciences
Unit for Artificial Intelligence and Geographical Information Systems (UNISIG)
National Polytechnic University, Quito, Ecuador.
Veronica Echeverría CEPEPOL National Polytechnic University
Quito, Ecuador


Abstract:
Oil exploration in the Amazon basin of Ecuador helped us to generate geological, structural maps using GIS, air photos, satellite and radar images. This tool and materials not only allowed a faster way of working but also permitted the development of new experimental Geological lineaments Intersection Density Methodology (GIDM). The GIDM permits to establish a correlation between areas of high geological lineaments intersection density and the possible presence of petroleum fractured fields associated with these fractured zones.

Introduction
The petroleum exploration is considered an art in which many geological variables are gathered, using expensive and sophisticated methods. Afterwards, data is carefully interpreted, to finally make predictions which are very important, since the investments that have to be made for an exploitation are really high, and often large parts of the population depend on them.

The present study is motivated by the idea to develop a method in which remote sensors are used in combination with field data and geophysical measurements in order to optimize future investments. In the petroleum exploration, especially in remote areas, aerial photos, satellite and radar images have played a mayor role to establish the geological mapping foundation. However, the proposed methodology has to be reviewed for its applicability before the petroleum exploration can start and demands are translated in petroleum barrels. In order to fulfill this requirement, the method has been experienced on a pilot area which will be described later on in this article. Within this framework, the help of different petroleum companies has been very useful, because in addition to the contracted aerial photo interpretations, they contributed with the experimentation and refinement of the proposed method.

The main philosophy behind the methodology is that when there are concentrations of geological lineament intersections, then the geological structure is highly fractured, and can possible serve as a oil reservoir.

Therefore the idea of the methodology is to make better use of geological fractures, interpreted on remote sensing imagines and translated into (photo) lineament maps. Based on the geological lineament intersections (GDIM) isolines are extracted with a certain density value, converting it into a density map. According to this density map, recommendations can be formulated to indicate which geological structures are favorable for petroleum exploration. The use of remote sensors within GDIM is also very useful in areas with buried structures under sediments because they let us visualize these structures by its reflection due to the relative transparency of the sediments.

Methodology
The methodology starts with the interpretation of the linear geological structures on the remote sensing images. Logically, the type of images used depends on its availability and the available budget. In case radar or optical satellite images are used, they have to be imported into a GIS, georeferenced and, an optimal combination of bands has to be established. In case of Landsat TM images at least a combination of band 4 and 7 are recommended (Lillesand & Kiefer, 2000). Due to its scale, aerial photos are often preferred for photo lineament studies, but for example in case of the equatorian Amazon, they are not always available. On the other hand, with optical satellite images it is very hard to find images with very low cloud cover which allows a successful interpretation. Finally, with radar images there can be a problem with shadows formed by relief which often interfere with geological structure reflection patterns.

This problem occurs in the higher parts of the Amazon basin. However, more towards the east, further down in the Amazon basin, this problem is almost solved because there is very little relief. Therefore, often a combination of different types of images is preferred to complement one another.

For the implementation of the GDIM the following steps are proposed:

1. Preprocessing the remote sensing images in GIS software package if necessary, a) In case of satellite images, importing the different bands separately, georeferencing the images, select the study area, eventually apply some enhancement procedure to improve the quality of the image. b) In case of aerial photos, scanning of the photos, import the different photos, digitizing or importing contour lines, interpolation of contour lines into a Digital Elevation Model (DEM), generation of a digital ortho photo using Ground Control Points (GCP´s) and photo camera parameters like focal length, generating a digital ortho photo mosaic by joining the individual digital ortho photos together. This way a digital non-classified raster base map is created.

2. Identify and outline geological lineament structures through optical interpretation on the computer screen or in case of aerial photos, direct on the photos and afterwards, through onscreen digitizing, on the digital ortho photo mosaic. This way a vectorial segment map is created within the GIS.

3. Location of geological lineament structures intersection points on the screen. With this step a third vectorial point map layer is aggregated within the GIS.

4. Overlaying a vectorial matrix with a certain grid size (each grid cell forms a polygon) for which the density of intersections points is calculated. (for example number of intersections points per ha or km2). To each grid cell the density value is assigned. For small areas this procedure can be easily done manually, however, for larger areas, with some simple programming this procedure can be automated.

5. Assignment of the density value of each grid cell in its center, this procedure generates a new point map in which the points are equally distributed but several points may have the value zero. Afterwards the vector point layer is rasterized, using the same grid cell as was chosen for the above-mentioned vector matrix.

6. Interpolation of the raster image created in step 5, which generates a Digital Terrain Model (DTM) with a density value in each pixel for the whole study area.

7. If required a vectorial segment map can generated through a raster-to-vector operation based on the DTM in order to create an isoline density map. The advantage of creating an isoline density segment map is that it facilitates an easier visual interpretation. Besides it can be overlayed with other map layer like for example the raster base map.

Result of these steps are visualized in maps 2-4 at the end of this article.

Logically, the GIS permit us to incorporate other thematic layers or additional information within the non-spatial database. Therefore this tool not only helps us to automate the GDIMethodology but also assists in establishing a (spatial) information base.

Applications of the GDIM within the amazonic region
In order to demonstrate the use of the proposed methodology, we describe here its application within a pilot area in the Equatorian Amazon region. The pilot area is the oil exploration block 28 of the Tripetrol Exploration Company, which is located west of the Andes, at the beginning of the Amazon Basin and between the main rivers Pastaza, Puyo and Napo (see map 1). The exploration block and its direct surroundings cover an area of about 8,000 km2.

Within the area there are sedimentary rocks present of marine and continental origin, expressed mainly as Cretacic, Tertiary and Quaternary terrains. The resulting lithologies generate favorable environments for the formation of petroleum generating rocks.

The Tectonic environment of block 28 is a compressive "back arc, in which the Jurassic/Cretaceous terrains form tight folds at one side and Abitagua intrusive at the other side. The Tertiary terrains on the other hand, form very wide foldings and softer thrusts towards the west plain.

The geological formations are oriented in a North-Northwest direction, towards the Pastaza faults, where it turns east in a North Northeast direction.

The intrusive body of Abitagua is highly fractured because of its rock nature and limits with the West basin with faults contact.

The remote sensing products which were used are the following: 1) Landsat TM images, of several years, joined together in a mosaic. 2) SAR Radar imagines 3) Aerial photographs scale 1:60.000. These products were obtained by the company INFOESPACE.

For the interpretation of the Landsat TM satellite images a 2,7,4 band combination was used, this combination permits to study the geological structures but furthermore, is of use in a second phase when access roads to the exploration have to be planned, to often very in hostile areas.

Geological lineaments
In spite of the thick and continuous vegetation layer of the humid tropical forests, the cartographic alignments by photo geology are various and well recognized. An old system of Geological Lineaments, with a North-Northwest direction is located at one side of of the study area, and at the other side the great alluvial plain known as the Pastaza Depression is found. A young or reactivated system of an old system of Geological Lineaments with a North-Northwest direction can be recognized witin the remote sensing due to the transperancy of the younger sedimental Meza formation and is located in the surroundings of the two comunities “El Puyo” y “10 de August”. This system limits the Mirador anticlinal and controls the course of the great Pastaza River. The young Geological Lineaments with the North-Northeast direction are associated to Pastaza River and cuts the proceding of the older or reactivated Geological Lineaments. The youngest Geological Lineaments with a North-South direction cut the recent alluvial deposits.

Faults
The Geological Lineaments that prove some kind of displacement have been considered as line faults. While on the other hand, Geological Lineaments, varying between North-Northwest and North-Northeast are interpreted as thurst faults. The most remarkable thrust faults are the following:
  1. The faults associated to the intrusive rocks,
  2. The Fatima Fault that is bordered at East by the anticlinal Mirador, and crosses the beginning of the Bobonaza river, and,
  3. The Cosano Villano Bobonasa fault that borders in the East the Oglan-Autapi anticlinal.
Geological Lineaments with an East-West direction coincide with the normal faults expected at the bottom and at the beginning of the Mirador folding system. This way the Geological Lineaments that can be identified in the terrain through remote sensing images are limited to the North and the South by the anticlinal Mirador. The Geological Lineament that forms the main course of the Pastaza River is interpreted as a right or dextral fault that displaces former structures and is generrally known as the Pastaza River fault.

Foldings
Within the study area there are tectonic forzes with Eastwest and Southwest characteristics of the back arc that result in different folding, faults and fractures systems with clear defined directions.

Therefore any structural element can be predicted by guiding us by the Geological Lineaments. Based on this analysis, it can be concluded that the combination of an anticlinal structure and a fault, forms a double structure which is favorable for the presence of high structures, or with other words the possibility of oil reservoirs. When Geological Lineaments are identified that correspond to faults, it is also probable that high structures are found.

The axis of the foldings in the remote sensing images are evidenced of the morphologic oval characteristics of the formation. Meanwhile, the curved form of the drainage patterns and the riverbed, can be of great help to find the axis of the foldings, because these suggest obstacles of a structural nature. The main directions of the foldings axis South of the Pastaza River fault are North-Northwest – South-Southeast. North of the same fault the directions are North-South and NorthNortheast – South-Southwest. The most pronounced anticlinal within the study area is the Mirador anticlinal.

Construction and use of GDIM
With the use of the different remote sensing images as were indicated before, the PDIM was applied, generating density isoline every 5 units of intersection density. Based on the intersection density units, and considering a map scale of 1:100,000 a classification was made in which maximum values are considered to be superior to the 40, high values start at 30 and go up to 40, medium values are in a range between 15 and 30, and finally, low values are the minor to 15.

Taken into account this classification it was found that the PDIMethodology gives high values at the Westside of block 28 following a North-South direction, which corresponds with the instrusive rocks belt and the instrusive fault zone. From a prospection point of few this indicates that it is possible to find fractured deposits at the high structure near the Pastaza River and Illocullin River. Areas with high values of intersection density are found in the Arajuno River, which starts in the center of block 28 and heads towards its Northeast and Southeast corners.

Clear trends that can be distinguished with the PDIM in this study area, are the North-Northeast and North-West directions of the geological lineaments, the first one coincides with the failure transcurrent dextral system type strike-slip while the second one coincides with the transcurrent sinestral antithetical systems that aligns following the Pastaza River.

Conclusions
It can be concluded that with the help and interpretation of remote sensing images, like satellite imagines, radar imagines and aerial photographs, it is possible to identify certain useful geological information, part of a petroleum prospecting, that let us formulate hypothesis on the presence of petroleum reservoirs.

Considering the different remote sensing products applied for this study, the radar images are the ones that best express the structural characteristics. The Landsat TM images also show the structures, but cloudcover are often a problem to obtain good quality images of the Amazon region. On the other hand, the aerial photographs show more detail due to its scale but have the same problem with clouds. Also the processing phase of the photos into a ortho photo mosaic is a more time consuming job.

The identification of geological lineaments on remote sensing images allows us to make a density map of the geological lineaments intersections. This is considered to be a great tool to help to find the location and distribution of the structures, which consists mainly of failures and litological differences.

In block 28 there is a coincidence between the distribution of the density isolines generated with the PDIM and the fault transcurrent dextral strike-slip systems in a North-Northeast direaction transcurrent to the sinestral antiethic fault in a North - West direction.

When applying the PDIM on the study area, block 28, it gives high value at the Westside for the high structure near the Pastaza and Illocullin Rivers. This result indicates a great possiblity that oil concentrations can be found in these fractured zones through prospections. Areas with high values of intersection density are found in the Arajuno River, which starts in the center of block 28 and heads towards its Northeast and Southeast corners.

Finally, this proposed methodology is considered an important improvement to predict the possible location of high structures, and in relation possible oil reservoirs, by means of geological lineament identification using remote sensing images and GIS.

References
  1. Bally, A. W., 1983, Seismic expression of structural styles.
  2. Costa-Cabral, M.C. and Burges, S.J., 1994. Digital Elevation model Networks (DEMON), A model of flow over hillslopes for computation of contributing and dispersal areas, Water Resour. Res., 30 (6), pp. 1681-1692.
  3. Eastman, R., 1995. Idrisi for Windows, Student Manual.
  4. Faifield J. and Leymarie, P., 1991. Drainage Networks from grid digital elevation models, Water Resour. Res.,27 (5), pp. 709-717.
  5. Floyd F. Sabins, JR., 1987 Remote Sensing, Principles and Interpretation.
  6. Jenson, S.K. and Dominigue, J.O., 1988. Extracting topographic structure from digital elevation data for geographic information analysis, Photogramm. Eng. Remote Sens., 54(11), pp. 1593-1600.
  7. Lillesand, T.M. and Kiefer, R.W, Remote Sensing and Image interpretation, 4 th ed, John Wiley &Sons, In, New York, USA.
  8. Richards, J.A., 1994. Remote Sensing Digital Image Analysis: An Introduction, springer-Verlag, Berlín, Germany, p.340.
  9. Robinson Villanueva N. Técnicas Estadísticas y Análisis Factorial aplicados a las Ciencias de la Tierra.Lima, Perú, pp. 32-61 Photogrammetric Engineering & Remote Sensing. Vol. 57, Nº9, pp.1163-1169. U.K. Nationale Remote Sensing Centre, Farnborough, Hants, United Kingdom.
  10. Zhang, W. and Montgomery, 1994. D.R., Digital elevation model grid size, landscape representation, and hydrologic simulations, Water Resour. Res., 30 (4), pp. 1019-1028.

Map 1: Location of oil explotation blocks in Ecuador Map



2: SAR Radar image of the study area Map



3: Geological lineamentas and intersections



Map 4: Density isolines


© GISdevelopment.net. All rights reserved.