Geological problem, geophysical tool: An alternate workflow

March 5, 2001
Cross Timbers Oil Co. obtained controlling interest in Middle Ground Shoal field from Shell Oil Co. in 1998.

Cross Timbers Oil Co. obtained controlling interest in Middle Ground Shoal field from Shell Oil Co. in 1998. Middle Ground Shoal field in Cook Inlet, Alaska, is a north-south trending anticline productive from multiple zones and structurally complicated with portions of its west limb overturned and cut by thrust faults. The extreme dip of these beds has adversely affected the quality of the seismic in this area with the result that the wells comprise the most reliable available data set.

Cross Timbers wants to boost production from the field and has planned a number of wells to be drilled during the next few years. The field's structural complexity is well documented and was one of the reasons Cross Timbers purchased the properties. However, visualization of the west flank structure proved to be a challenge using conventional mapping techniques.

Because this was an essential element for maximizing the benefits from these wells, Cross Timbers requested the assistance of the E&P Workflow Consulting team from GeoQuest, an operating unit of Schlumberger, in the construction and visualization of a structural framework of the field's productive horizons.

Vertical and overturned beds on the field's western side precluded the use of many software tools typically used in the development of structural frameworks from geological data. A workflow was developed for this process that employed interpretation programs normally used for seismic interpretation. This combination of seismic interpretation and visualization tools, populated with solely geological data (well locations, deviation surveys, and marker tops), ultimately provided Cross Timbers with the collaborative means to reach the consensus needed among its technical staff and vendors for confident and timely decisionmaking.

This article describes this workflow and the tools used to bring about this consensus.

Introduction

Cross Timbers is an exploration, exploitation, and development company primarily interested in the acquisition of long-term hydrocarbon producing properties of high quality. The company has historically concentrated its activities in Oklahoma, Arkansas, Kansas, Wyoming, New Mexico, and Texas and recently expanded into Alaska.

Click here to enlarge image

In October 1998 Cross Timbers acquired controlling interest of two platforms in Middle Ground Shoal field from Shell (Fig. 1). The field, discovered in 1963, is highly structured with production from multiple zones within the Miocene-Oligocene aged Tyonek formation. The primary productive zones occur between the subsea depths of 7,300 ft and 10,000 ft.

Production began in 1966, and the field has produced 185 million bbl of oil to date from four platforms. Production averages 5,000 b/d from multiple zones in the Hemlock formation. Cross Timbers produces 3,600 b/d from 27 wells drilled from two platforms.

The anticline's eastern limb dips to the east with dips approaching 55 degrees in places. Its western limb has high-angle dips and in some locations is overturned with dip 85 degrees to the east. A highly successful pressure maintenance program began in 1968 for the field's eastern flank. Injection into the field's west flank began in 1997. Eleven injection wells are operating.

In addition to the folding that has taken place, Middle Ground Shoal field has undergone deformation by thrust faulting. Three high-angle thrust faults are known to have formed in the western limb of the anticline. The deepest and most westerly of these has little well control but appears to have a maximum of 320 ft of displacement with a fault plane that dips to the east at 50 degrees. The shallower, crestal thrust has a maximum displacement of 200 ft and dips approaching 70 degrees in some places.

The problem

With purchase of the field, Cross Timbers began working on a number of methods to increase production.

One possible method investigated is the expansion of the pressure maintenance program under way on the field's western flank. To accomplish this properly a thorough understanding of the field's complex geologic structure had to be obtained and accurately communicated to all professionals working the field. This would not only facilitate the pressure maintenance program but would also aid in the determination of potential drilling opportunities.

During review of the field's geology it became apparent that the structural complexity of the west flank, combined with the highly deviated wellbores, would require some unconventional mapping techniques.

The Cross Timbers geoscience group constructed six Hemlock formation net pay isopach maps using wellbore data from the 31 west flank penetrations. This mapping led to the identification of several "locations" where additional deviated wellbores could be placed.

During meetings with management in late 1999 to discuss these drilling opportunities, it became apparent that the newly constructed two-dimensional maps could not completely convey the three-dimensional (3D) nature of the structure. The geoscience group was asked to pursue some type of 3D visualization program that would help confirm the opportunities identified by the new mapping.

The GeoQuest E&P Workflow Consulting Group in Houston was contacted and requested to develop a solution. The greatest problem facing the workflow team was the fact that geologic horizons are handled by most software as gridded surfaces. In addition, the vast majority of software commercially available at this writing does not permit multiple depth values for the same X and Y coordinates. The overturning of the formations along the western flank of the field prevented the available software tools from being used in the normal manner.

The solution

After some discussion the workflow team decided to attempt the construction of a 3D structural framework using a number of GeoQuest seismic interpretation tools in an unconventional approach. All available information concerning well locations, deviation surveys, and geologic marker tops was obtained and loaded into a project data base. Cross-sections and structural maps of the field also were made available by Cross Timbers to the workflow team.

An artificial 3D seismic survey composed entirely of dead traces was also loaded into the data base. This survey was in depth, and 50-ft by 50-ft bin spacing was used for the survey spacing. This artificial survey of dead traces was used as a "3D canvas" on which the field's structural framework was developed.

The reservoir intervals on the eastern flank of the structure were mapped by conventional means. The markers for the wells were gridded and smoothed using an appropriate mapping application. The interpreted position of the main crestal thrust fault, which acts as a reservoir flow barrier separating the east and west flanks of the field, was used as a boundary for the grids. The resultant grids were checked for accuracy in the 3D visualization software. Some grids were found to cross, due to isolated incorrect marker picks, which were adjusted before proceeding.

To overcome the limitation of the inability of horizons to contain multiple depth values at a point, it was decided that these would be interpreted by classifying them in the project as faults instead of horizons. A GeoQuest seismic interpretation package was selected to use in the development of the interpretation. This product allows for the use of multiple Z values for faults and is directly tied to GeoViz 3D visualization and interpretation software.

The borehole paths of the wells, along with the marker tops interpreted by Cross Timbers' geologist, were posted within the interpretation and visualization programs. Horizons were interpreted as faults on the survey in-lines where these artificial seismic lines intersected boreholes using marker tops posted on the borehole paths as guides. Figs. 2 and 3 illustrate how this was achieved.

Horizons as faults

Click here to enlarge image

Fig. 2A depicts an area of the artificial 3D where a well's borehole (well C12-23) parallels an in-line (line 483). - A--Horizon intersections along the C12-23 borehole parallel to in-line 483. B--Due to the subparallel relationship of the borehole to the in-lines, horizons had to be initially interpreted on multiple in-lines in order to establish their true structural configuration.

Click here to enlarge image

In this instance all of the zones of interest, HC through HZ, could be interpreted on the line because the borehole trajectory nearly parallels it. This simplified the interpretive process since the spatial orientation of the borehole intersections of the zones of interest could be observed in a near correct manner along the in-line.

Click here to enlarge image

Fig. 2B shows the map and cross-sectional views of in-lines 422 and 427 where they intersect the borehole of well C24A-23 near the HC zone pick. Because the borehole is not parallel to the in-lines, the true structural relationship between the HC zone pick and the other zones cannot be observed or interpreted along one line. Therefore only the HC horizon was initially picked on in-line 422. The other horizons were added to this line as gaps in the interpretation were filled in following the initial interpretation phase.

Click here to enlarge image

Fig. 2B also shows in-line 427 where it intersects the same well C24A-23 at the near structurally correct location for the HR zone penetration. The units between HC and HR were interpreted on the in-lines between 422 and 427 where their closest approach occurred to minimize structural inaccuracies associated with projecting them from out of the plane of section.

Using this method, a structural interpretation for the various zones was developed that best honored the well control and allowed multiple depth values for the zones. Once this was completed all of the other zones were added to the in-lines that contained only one or two of the zones due to the angle at which the borehole intersected them.

As the interpretation was being built up it was also displayed within GeoViz. Using the 3D visualization proved to be an important step, allowing the interpretation to be quality-controlled as it was developed. Variations within zone thicknesses and differences in bed shape were readily identified. Adjustments were then made within the seismic interpretation software.

Closing other gaps

Once these adjustments were completed, a number of the survey's in-lines were still without interpretation, these being the ones between the wells providing the control.

These gaps constituted a considerable portion of the survey. To infill these areas the fault interpolation tool within the seismic interpretation program was utilized. This interpolation tool was used extensively during this stage of the project and resulted in a significant time savings.

A--Shows how HR horizon was interpolated between lines of control for the footwall in a structurally uncomplicated area.
Click here to enlarge image

Fig. 3 shows a structurally uncomplicated portion of the study area where overturning has not occurred and where the heave of the thrust has positioned the hanging wall over the footwall by some 150 ft. The ribbon posting for two lines on the basemap shows where the HR horizon has been interpreted in this portion of the study area. The 15 in-lines in between have not had the HR horizon interpreted on them. Fig. 3A shows how the interpretation of the footwall was first interpolated between the lines of control. Following this the hanging wall was interpolated (not shown). Note that where ribbon posting for the hanging wall overlays that of the footwall, it completely conceals the interpretation for the lower plate.

Click here to enlarge image

This method allowed the interpretation to be quickly propagated through many areas of the survey. However, in areas more complicated structurally, where overturning of the zones had occurred, and in particular where multiple thrust faults cut the section, it was necessary to limit the ribbon posting to a particular depth range.

B--Middle thrust plate of HR zone in a more complex area could only be interpolated by limiting the Z-range display to a specific depth range.
Click here to enlarge image

Fig. 3B illustrates why this was necessary. Fig. 3B contains an area of the field where the overturning is significant.

Click here to enlarge image

The upper map view shows the ribbon posting for the entire section. Note that the portion of the zone underneath the overhang, the middle thrust plate, is completely obscured in this map view.

The lower map views show the same area where the ribbon posting has been limited to a depth range between 9,550-10,250 ft.

With the middle plate now being the only data displayed on the ribbon posting map the interpolation tools can be used to interpolate this section of the HR horizon between the two control lines.

Using this methodology all productive zones were interpreted at the well control and then interpolated through the interwell areas for the entire study area. Once this was accomplished three proposed boreholes were added to the GeoFrame project. These proposed locations were then viewed along with the interpretation of the zones and thrust faults in 3D space using GeoViz.

Using this 3D model, Cross Timbers' geological and engineering professionals were able, for the first time, to visualize a common structural framework. This proved to be a valuable collaboration method from which it was determined that two of the proposed boreholes needed adjustments in order to hit their intended objectives.

The presentations

Upon the completion of the technical portion of the project, three presentations were given to Cross Timbers management and various vendors associated with the project.

The first, at Schlumberger's office in Oklahoma City, was a working session with the geologist and drilling engineers. The second, also in Oklahoma City, was given to local management and used GeoViz's stereoscopic capabilities.

The structural frameworks, along with the boreholes of the existing and proposed wells, were displayed on a large 52-in. Mitsubishi monitor connected to a Silicon Graphics workstation.

A stereoscopic emitter, synchronized with blinker glasses worn by the observers, was also attached to the workstation. This allowed the participants to visualize the interpretation in true 3D space during the presentation.

The third presentation was given in Fort Worth to Cross Timbers upper management and was made using GeoQuest's VisionDome, a portable virtual reality system environment. This system provides an immersive environment for collaboration and also allows for the use of the stereoscopic viewing technology.

The VisionDome was set up in the lobby of Cross Timbers headquarters, the historic W.T. Waggoner Building in Fort Worth. The original banking lobby of the building's first floor was an ideal setting for the VisionDome's assembly and use. During one day a number of presentations of the newly developed model of Middle Ground Shoal field were made to Cross Timbers upper management and staff.

The ability to view the vertical west flank in a horizontal planar orientation proved invaluable and led to a consensus of where additional wells needed to be located to effectively develop this part of the field. The Alaska team was given permission to proceed with the proposed drilling/workover program and charged with keeping the model updated for future presentations.

The outcome

The situation that Cross Timbers personnel encountered when Middle Ground Shoal field was acquired was not unique to the company.

Quite often a disconnect can occur between various members of an asset team, particularly in areas of complicated geology, due to structural and/or stratigraphic considerations.

Full 3D visualization tools are an essential aid when trying to resolve these disconnects and technical misunderstandings. Although the majority of the "hard" data available for the field was geologic in nature, a unique workflow that utilized traditional geophysically oriented tools was developed to examine it.

This combination of imaginative workflow processes and powerful geophysical visualization tools available within GeoViz resulted in the development of the structural model needed to bring the technical team to a common understanding of the structure and the problem.

Additionally, the immersive setting of the VisionDome provided the enhanced collaborative environment needed to quickly convey key aspects of the field's reservoir geometry. This in turn allowed rapid consensus to be reached and prompt decisionmaking to occur.

The authors

Click here to enlarge image

Brian Toelle is a senior geoscientist with GeoQuest in Houston. He was employed as a geologist with Texaco in 1981-89, evaluating basins and generating exploration plays and prospects in West Texas, the Rocky Mountains, and offshore California. He worked as a geophysicist at GeoQuest in 1989-92 and in 1992-97 evaluated 3D seismic surveys with Saudi Aramco in Dhahran. He has an MS degree in natural science, emphasis on structural geology, from Stephen F. Austin State University. E-mail: [email protected]

Click here to enlarge image

Karen Glaser is manager of the technology solutions group for GeoQuest North America. She was employed as an exploration and development geologist for West Texas by Amoco Production Co. in 1983-86. She then worked at Exxon Production Research in 1991-94 as a sequence stratigrapher. She joined GeoQuest in 1995 after working as a consultant. She has a PhD degree from Rice University and an MS degree in petroleum geochemistry from the University of Oklahoma.

Click here to enlarge image

Michael Langeler joined Cross Timbers in 1998 as a senior geologist responsible for the Alaska properties. He started as a mud logger for Imco Services in 1980-82 in the Gulf of Mexico, then worked as a geologist for Sun/Oryx Energy in 1982-98 in Central and South Texas, New Mexico, Oklahoma, and the Rocky Mountains. He has a BS degree in geology from Eastern Illinois University.