Geological interpretation and reservoir maps generated by oil and gas exploration and production companies are mostly picture files but with various attribute elements also encoded in them. This article examines a procedure for decoding the graphics and attribute elements and creating a GIS-based layered database
Achintya Pal, Consultant/Domain Expert to ONGC
Oil and gas Exploration and Production (E&P) companies the world over generate huge volumes of geological interpretation and reservoir estimate maps in their quest to acquire hydrocarbon reserves. The interpretation maps indicate possible existence of oil/gas (Fig.1) while the reservoir estimate maps depict the extent of reserves of various thicknesses in a proven field (Fig.2). These maps, as final products, are normally preserved as .CGM files that are essentially picture files but with various attribute elements also encoded in them. This article discusses a procedure for decoding all the graphics and attribute elements (e.g., line width, line color, polygon fill color) and using the latter to build a layered database of cultural, geological and other features in actual geographical coordinates that is central to any GIS.
Fig.1 Typical geological and geophysical interpretation map prepared in oil and gas E&P industry
Fig.2 Typical sandstone (oil or gas reservoir) thickness mapgenerated in oil and gas E&P industry
The procedure – breathing life into a picture (conversion to live data)
This open third party software is essential for the specific case of E&P industry because CGM files cannot be directly imported into ArcGIS. For this reason, a customized CGM file decoding software has been developed though several CGM viewers are freely available on internet. The software, which works in both windows and UNIX environments, first extracts attribute elements along with the “device” coordinates encoded in the CGM files and then converts the latter into actual “user” coordinates by geo-referencing, consistent with a particular Cartographic Reference System (CRS).
Fig.3 Flowchart of the procedure for decoding CGM file and import to ArcGIS
While interpretation maps at various geological levels are encoded mostly as poly-lines and polygons, it has been found that some oil and gas reservoir extent maps are encoded as the “control points” out of which the actual shapes are to be reconstructed as “Bezier Curves” that are weighted interpolation curves akin to spline curves. Any composite map is then segregated into various “layers” in ArcGIS using the “select by attribute” feature.
Fig.4 A zoomed lower left part of the sandstone thickness map shown in Fig. 2. The cubic Bezier curve interpolation, with four control points, in the left is clear with the fourth point of one segment (points 1 to 4) being the first point of the next (points 4 to 7). The two segments are joined smoothly at point number 4.
The layered structure in ArcGIS
Fig.5Different geological (contours, faults) and cultural features can be segregated and stored as layers in ArcGIS
The procedure can handle both vector as well as raster maps
In addition to the “vector” maps, any “raster”ones like old legacy maps available only as paper prints can also be digitized using the decoding software. Raster files like .JPG or .TIFF are converted to .CGM files using graphics packages and then the same procedure is employed to extract the cell or pixel coordinates. Many maps which contain both vector and raster components are read by the same module so that it is transparent to the user.
Possibly the biggest advantage of ArcGIS based database from the point of view of the oil industry is that since the modern generation geo-scientific interpretation packages like R5000, Petrel, Kingdom are equipped to directly accept .shpand .prjfiles (containing map projection parameters) output by ArcGIS, any archived map can be retrieved very quickly into workstations for evaluation and comparison with other outputs.
The above comprehensive procedure for creating a corporate database of interpretation and reserve estimate maps are being used in ONGC India. The procedure has the potential to cater to the need of oil E&P companies worldwide.
Examples
The following figures illustrate the capability of converting various scanned or ordinary picture files into usable data in actual geographical coordinates and displaying multiple maps in the same area, obtained from different sources, in one frame.
Fig.6Original remote sensing image of Saurashtra as .JPG file
Fig.8 Composite Kutch-Saurashtra area with geological and remote sensing maps along with coastlinedata from independent source displayed in ArcGIS
Summary of capabilities of the procedure for creating a GIS based database from CGM files
1. Vector or raster maps from diverse sources may be digitally integrated and displayed simultaneously or layer-wise in any scale. Maps may be
(a) geological
(b) remote sensing
(c) topo-sheets
(d) gravity-magnetic
(e) geologicalinterpretation
(f) reservoir extent
2. Different layers of a map can be segregated
(a) on the basis of line width/line colour/fill colour etc. in case of vector maps (with polylines or polygons) such as interpretation maps or relatively recent REC maps
(b) on the basis of pixel colours in case of raster maps (pictures as .jpg/.tiff files) such as old legacy interpretation or hydrocarbon reserve estimate maps
3. Individual layers or composite layers of a digital map can be displayed on ArcGIS or other popular mapping packages in oil industry such as ZMap with their original colours in one go (requiring minimal user intervention)by importing a comprehensive colour table generated by the CGM decoding and geo-referencing modules.
4. Geological and geophysical interpretation carried out in one software package can be transplanted into another package. For example, cultural data already available on Landmark may be transported to Petrel and superposed on other interpretation features. This is done by
(a) Exporting contents of the map on Landmark to a CGM file as a normal procedure.
(b) Converting the CGM picture file to live data (coordinates in easting/northing) and storing them layer-wise on ArcGIS as shapefiles (.shp)
(c) Directly importing these .shp files to Petrel. (Modern day interpretation packages such as R5000, Petrel, Kingdom are equipped to accept the .shp files directly)
Conclusion
The above procedure can be effectively utilized for other digitization jobs such as that of a hardcopy time-depth curve. It is equally applicable to other exploration activities like minerals, groundwater etc.
