Applied Digital Subsurface Mapping

Based on a workshop sponsored by PTTC's Midwest Region on April 18-19, 2002 at Wabash Valley College, Mt. Carmel, Illinois

BOTTOM LINE

The University of Illinois has concentrated on improved methods of digital mapping, including Trend Analysis and 3-D Modeling. Trend Analysis is based on a method for differentiating regional and local components in a map area. Three-dimensional modeling provides a tool for visualization of the spatial distribution of data collected by geologic, geophysical, petrophysical and engineering means. An example of trend analysis would be to describe an anticline as a local anomaly occurring along a regional dipping surface. 

PROBLEM ADDRESSED

Petroleum exploration depends on identification of anomalies between regional and local components in a map area. Trend analysis provides a tool for efficient discrimination and 3-D modeling is the best means for visualization of the data. 

KEY WORDS:

3-D Modeling, Digital Mapping, Trend Analysis, Visualization Tools

SPEAKERS

Hannes E. Leetaru, Illinois State Geological Survey, Champaign, Illinois

TECHNOLOGY OVERVIEW

Trend analysis was first used by Levorson in 1927 as a method to remove the regional structural component in the mid-continent. Manual trend analysis was time consuming and not always accurate. Advanced computer software packages eliminate the limitations of manual methods and speed up trend analysis. The most used trend analysis techniques are polynomial curve fitting and Fourier analysis.

Classification of Trend-Surface Analysis
Trend-surface analysis is a form of multiple regression. A surface is mapped using polynomials to create the best fit with observed data. Using geographical coordinates, a first-order trend surface is represented by a plane. In structural mapping, the first-degree trend will show the regional dip of the surface. Second-degree trend surface introduces curvature. Third-degree surfaces approximate a saddle shape and have 10 coefficients. 

Steps to Trend Surface Analysis
Evaluate the general trend of the data using first-order data and use higher degree trends to interpret more complex or detailed patterns. Most computer packages use relative variance or standard deviation instead of absolute values to map surfaces, and show the map in grid form. The change in the variance between the third- and fourth-degree surfaces is normally much smaller than that between first- and second-degree surfaces. 

Problems with Trend Surface Analysis
Trend-surface maps are mathematical models that do not necessarily reflect reality. Computer-generated maps may have serious edge effects with extremely high or low values. One can eliminate this problem by creating a buffer zone around the desired map area. Data distribution within the map area may also create problems, such as elongation of lenticular trends. Trend analysis may also overemphasize well clusters based on the well control data, or misinterpret surfaces with wide dip variations. 

When to Use Trend Surface Analysis
Trend surface analysis is best used when one is looking for subtle changes in structure and to find structures not found using conventional mapping tools. Examples of areas where trend analysis has been successfully used are the Hugoton Embayment of Kansas and the Ste. Genevieve Limestone in Illinois. Oil fields in the Ste. Genevieve appear to be related to subtle residual high structures that are difficult to identify. The search for pinnacle reefs is also enhanced by trend surface analysis, using well control data to look for structural noses hidden using shale drapes. In addition to its use in structural mapping, trend surface analysis can be used for isopach and isolith maps. As an example, using trend surface analysis in deltaic systems with overlapping channels to isolate individual channels and identify potential new plays.

Evaluation of Trends and Residuals
Examples from Illinois use contour intervals in the Benoist sandstone in oil fields in the deepest parts of the Illinois Basin to interpret subtle structural noses. Previous structural maps of the area did not show any anticline features, but first-degree residual mapping revealed anomalous high areas in these oil fields. At Storms Consolidated oil field, producing from the Aux Vases formation, new reserves were mapped from deeper pay zones using first- through fourth-degree trend surface analysis. Almost every Aux Vases reservoir compartment is located on residual highs. Trend analysis has proved to be an important technique for enhancing map data. 

3-D Modeling
Three-dimensional modeling is a visualization tool for spatial data. It can be used to develop final maps, or be used for additional modeling in reservoir simulator software packages. Companies use 3-D modeling as a database to store data on perforations, determine lateral continuity between strata, and to locate individual wellbores within a reservoir. Petroleum companies and environmental concerns use 2-D maps to characterize reservoirs, aquifers and hazardous waste sites, but 2-D characterization does not produce a detailed map. In recent years there have been tremendous advances in 3-D technology. Geological and engineering data are used to interpret both outcrops and subsurface surfaces. 3-D modeling can be used to correct cross sections and maps of the same area, and to develop a series of cross sections into fence diagrams.

Variations in porosity and permeability can be best visualized with 3-D modeling. Such modeling of permeability intervals is very important in planning water floods and secondary recovery projects. 3-D models can illustrate both lateral and vertical changes in permeability. 

Methodology
Adequate well control data for all horizons is important in constructing subsurface maps. Field examples from Illinois were used to illustrate the advantages of 3-D modeling over surface 2-D mapping, illustrated by more accurate paleotopographic maps with 3-D modeling. Unconformities should always be used as key mapping surfaces. Local unconformities need to be correlated with the regional geologic framework. Wireline data from gamma ray and spontaneous potential curves can be used to differentiate sandstone from shale. Neutron, density and sonic logs measure the amount of porosity in the rock, but data need to be normalized prior to modeling. Older logs, in particular, are often inaccurate because of improper calibration, and need to be correlated. 

3-D Modeling Methods
The first stage is the construction of 2-D grids, including key mapping horizons. Tighter grid intervals yield higher resolution models. Deterministic and stochastic approaches are used to attach 3-D attributes to the grid. Weighting factors based on geologic and engineering data need to be added to create smoother and more accurate models. 3-D modeling is very graphic and requires large amounts of computer memory. As the 3-D model is moved into reservoir simulators, more dedicated workstations are required to manage the data. 

Pitfalls in 3-D Modeling
To create a realistic model, the initial stratigraphic framework may need to be changed when viewed in three dimensions. Subtle features on a cross section, such as an unconformable surface, should be more apparent in a 3-D visualization. Poor data from wireline logs will be more obvious with 3-D visualization, as will be errors in correlation. It is important to realize that anomalies that are smaller than well spacing will not be visualized in a 3-D model. Procedures that produce smooth grids and average the data may be correct for laterally continuous beds, but the models will be incorrect for more heterogeneous reservoirs. 

Using 3-D Modeling as a Decision Making Tool
The data used to construct 3-D models is qualitative. Geological expertise is still required to evaluate facies changes and other inter-well spacing problems. The value of the 3-D model is that it demonstrates the relationships of input parameters to increase the speed and accuracy of the geologist's interpretations, and allows the geologist to look at different scenarios and select the best fit. The ultimate goal of 3-D modeling is to improve reservoir management. 

CONNECTIONS:

Hannes Leetaru
Oil and Gas Section
Natural Resources Building
615 E. Peabody Drive
Champaign, IL 61820
Phone: 217-333-5058 Fax: 218-333-2830
Email: leetaru@geoserv.isgs.uiuc.edu

For information on PTTC's Midwest Region and it's activities contact:
Steve Gustison 
Illinois State Geological Survey
615 East Peabody Drive, Champaign, IL 61820
Phone 217-244-9337
Email: gustison@isgs.uiuc.edu

Disclaimer: No specific application of products or services is endorsed by PTTC. Reasonable steps are taken to ensure the reliability of sources for information that PTTC disseminates; individuals and institutions are solely responsible for the consequences of its use.

The not-for-profit Petroleum Technology Transfer Council is funded primarily by the US Department of Energy’s Office of Fossil Energy, with additional funding from universities, state geological surveys, several state governments, and industry donations.

Petroleum Technology Transfer Council, 16010 Barkers Point Lane, Ste 220, Houston, TX 77079
toll-free 1-888-THE-PTTC; fax 281-921-1723; Email hq@pttc.org; web www.pttc.org