GIS data modeling for well spot planning

GIS data modeling for well spot planning

Dr. Robert C. Maggio
Manager, GIS Operations
Petris Technology, Inc.
1900 St. James Place, Suite 700
Houston, TX 77056
713-403-8451
Emailt: maggio@petris.com

Introduction
Oil and gas exploration uses considerable data regarding underground geologic structure to determine the placement of wells. There are sophisticated software systems that aid the geologist in the placement of oil wells. These software systems depend on the geologist's interpretation of the geologic strata, production figures from existing wells, as well as a vast amount of experience. The result is the placement of the bottom of a new well. The exploration software systems locate new wells by assigning them a world coordinate; most often, this coordinate reference is in degrees of latitude and longitude. This method effectively locates the well. However, this is only a coordinate. Most exploration software systems do not reference this new coordinate to landcover features at the surface.

The geographic information system (GIS) is an integrated system of software tools that capture, store, analyze, portray, and output spatially referenced data. The GIS is based on coordinate systems; and much like the software system used by the exploration geologist, it represents data in two or three dimensional space. Each system has its strengths and weaknesses. The exploration system is designed to aid the geologist in finding oil. The GIS is an analytical tool whose forte is the representation of many data types in a single view. An integration of these two capabilities could equip the exploration geologist with a very powerful decision making tool.

One of the challenges to a geologist when a new well has been spotted, is the placement of the surface location of the well. The decision to drill the well vertically or horizontally is based both on geologic evidence and surficial landcover type. Placing oil wells at the surface in residential areas, on highways, in rivers, or on highly rugged terrain is not practical. For this reason, it is prudent to integrate the well spots with detailed landcover data. This landcover Therefore, the final decision as to where to place the well may depend on the landcover at the surface. data is often depicted on aerial photographs or satellite imagery.

Another key factor in the placement of the surface location of oil wells is the density of the wells in a production unit or field. The GIS has a sophisticated set of spatial analysis tools that will aid the geologist is the grouping of wells according to criteria set forth by regulatory agencies. Grouping of wells into units and fields is a service the GIS can provide. Additionally, since the GIS integrates both mapped data and remotely sensed imagery, it will allow the geologist to plan well densities, while at the same time, use good judgement in their placement with reference to landscape elements.

The data

Well Location Data

A portion of the data used in this analysis is extracted from the geologic exploration software system Geographix. The system yields coordinate pairs that reference the location most likely to produce oil. No effort will be made to explain the operation of the system or its use. For this presentation, it merely produces a location to drill a well. The coordinates from Geographix are integrated into the GIS using a process known as geocoding. This function takes an X,Y coordinate pair listing and generates a map. The map may contain one to any number of positional locations. In addition to the locational information for each well, data describing the well such as depth, operator, owner, etc. is also stored with its location.

Aerial Photography

The landcover data stored in the GIS is digital, color aerial photography. The photography was imaged using a conventional aerial camera that produced a 9" X 9", true color photograph. The negative scale was 1:24,000 or 1" = 2000'. This is a common scale used for landuse/landcover mapping. The photographs were then scanned at a resolution of 400 dots per inch to produce 1- meter pixel resolution digital images. Due to time constraints and equipment available to image the photographs, there was no attempt to geo-reference the images using aerial Global Positioning Systems (GPS) equipment.

The geo-referencing of the aerial photographs was developed manually. Ground control points that appear both on the photographs and on the ground were visited in the field using GPS equipment. Features visited in the field included road intersections, stock tank dams, fence corners, oil tank batteries, and other man made features. Using the GPS equipment, coordinates were collected for four to eight points per photograph. A digital image processing system was used to geo-reference each photograph. The result of this process was a coordinate location for each pixel on each aerial photograph.

Each individual, geo-referenced aerial photograph was of great value and could be loaded directly into the GIS. However, an additional step of mosaicing all of the geo-referenced aerial photographs into a single, seamless image greatly enhanced its utility. The digital image processing system, ER Mapper was used to mosaic each of the individual geo-referenced images into a single image. The process of mosaicing multiple, large-scale images into a single, large- scale image produced an image that was seamless, and yet preserved the 1-meter pixel resolution of the original photography. The ER Mapper image mosaicing process also conducted edge matching, color balancing, edge feathering among images, and image compression. The original, seamless image exceeded 108 gigabytes, but was compressed to 482 megabytes. This compression is very important when displaying large images in the GIS. The ER Mapper software provides a "plugin" to most GIS packages that display imagery.

Unit and Field Boundaries

A grouping of wells makes up a unit and a group of units makes up a field; the geologist develops these units and fields. The grouping of the units and fields is typically done using a computer aided drafting and design (CADD) system. The units and fields are typically digitized from a map containing roads and other cultural features. Sometimes, they are digitized from georeferenced aerial photographs. The units and fields are then imported into the GIS.

Figure 1: Well spots are represented as dots while the
units are contained within the Field. The Units have been buffered.

The GIS database and anlaysis

The GIS database developed for this well spotting application contained the three base map layers that have been previously described. The wells database contained fifteen attributes:

API Number	Field Name
Operator	Well Name
Well Number	State
County	Date
Latitude	Longitude
Depths

The units database contained the following attributes: unit name, field name, and acreage. The field layer merely contained the units. Figure 2 shows a production field with wells represented as points. The units are drawn as polygons with wells contained therein. The grouped units are stored as a field. Figure 3 shows a group of wells with 1320' buffers. This buffer distance reflects regulatory constraints on the number of wells that may be contained in a unit and the density of the proximity of wells to one another. This analysis was conducted in the GIS using attributes from the databases described above.

Figure 2: Buffered wells (circles) are superimposed over
units to depict well placement densities with reference to units.

Figure 3: Geo-referenced image with well spots and
well buffers superimposed.

The units represented in figure 3 are buffered toward the inside to comply with further regulatory constraints. Each unit is buffered 1320 feet to the inside of the unit. Buffering the units and superimposing the well buffers on the units may also visually depict compliance with this regulation. Well spots may be adjusted by using the GIS to select appropriate placement to comply with regulations.

Output from the Geographix software has driven the GIS display and analysis to this point. Buffering has taken place on the existing data. However, care must now be taken to determine whether the placement of wells is appropriate with respect to the existing landcover. Traditional methods of locating wells on the landscape have used field surveying with transits or GPS equipment. This method requires considerable, expensive fieldwork. Geo-referenced aerial imagery may reduce the need for this expensive fieldwork; however, it will not replace it. Since all spatial data contained in the GIS is geo-referenced, the well spots, well buffers, and unit buffers are readily superimposed onto geo-referenced imagery. Figure 3 depicts a digital image with well spots and well buffers superimposed.

Inspecting the imagery with the superimposed well locations will identify areas where current landuse and potential well spots might conflict. In the imagery, it is obvious that some of the wells were already in place (white spots around the well spot symbols) while others were merely in the planning phases (well spots with no surrounding white spots). The white scar on the landscape surrounding a well spot is typically made by the placement of gravel around the well head. Inspection of the imagery with superimposed well spots will provide the engineer the opportunity to select alternate surface locations for wells that conflict with existing landuse. Examples of this would include wells placed in roadways, ponds, and on urbanized structures.

Conclusions
Geo-referenced, digital imagery and GIS analysis may prove to be and invaluable addition to the tool kit of the exploration engineer. The analytical processing provided in the GIS will augment the analysis provided by conventional exploration software systems. In concert, the two tools will allow the engineer the opportunity to make better-informed decisions and assist with the compliance to regulatory agency criteria for well spotting.