QUICK FACTS
The LINK Collaborative Research Program is using PV-WAVE to analyze data by using combinations of dynamically linked scatterplots, contour plots and histograms. This analysis is done to get a better understanding of the accuracy of seismically derived velocities and for estimating Earth velocity. The end results offer insight on subsurface structures and hydrocarbon exploration.

THE PROBLEM
In hydrocarbon exploration and field appraisal, seismic surveys are the primary source of information about subsurface structures. Because the seismic record is referenced to time, assumptions about the velocity of the signal through the Earth are made in order to connect the time-based interpretation to a depth model. However, current methods for estimating Earth velocity have limited accuracy.

As part of the LINK collaborative research program, funded by the Oil and Gas projects and Supplies Office (OSO) of the UK Department of Trade and Industry and the National Environment Research Council (NERC), Scott Pickford has teamed up with Imperial College, London, to address this problem.

The project has been using PV-WAVE as its development environment because of the system's high-level graphic analysis capabilities. The development team includes petrophysicists and programmers, all of whom use PV-WAVE to differing degrees. But all involved agree that without the use of a 4GL such as PV-WAVE, much of the system development would be impossible. Two systems have been completed: IC2 and Gather Snapper. A further system, Minerva, is under development.

When the "IC2" and "Gather Snapper" projects were completed at the end of 1995, the industry sponsors (British Gas, Elf, Enterprise, OMV and Geco-Prakla) received fully engineered software packages based on methodology developed during the project, and a detailed structural interpretation of the North Sea's southern gas basin blocks, which make up the data set. In addition, the software will be sold commercially.

THE SOLUTION
During the project's three-year duration, several significant milestones have been passed.

A two-year research program by Imperial College successfully completed a fully functional prototype Interactive Clustering and Multivariate Attribute Analysis (IC) software package. Although primarily designed for seismic data, the software can handle any multivariate data set.

Data are analyzed in PV-WAVE using combinations of dynamically linked scatterplots, contour plots, histograms, etc. Regions can be highlighted in one window and the corresponding points displayed in other windows; all plots are linked, giving a powerful tool to explore structure in a multivariate data set.

The user can subdivide the seismic data set into time windows corresponding to geological horizons and compute a wide range of attributes ranging from widely used parameters such as trace energy.

A regional well log and seismic database including raw traces, stacks, maximum coherence stacking velocities and migrated data has been established. Routines for handling log data have been completed and will be incorporated into the Log Data Analysis Environment. A link between the well-based and seismic domains via the Discrete Wavelet Transform will be developed.

An understanding of the accuracy and reliability of seismically derived velocities and interval velocities derived from them is crucial to the method. As such, the Gather Snapper has been designed to allow the user to examine the geometry of events observed on seismic gathers and extract velocities from them in various ways. A study has been made of the magnitudes of error and the mathematics of error propagation during the process of extracting interval velocities from the RMS velocities obtained from seismic gathers. Ray tracing routines that enable the user to transfer data between the unmigrated time, migrated time and depth domains supports this.

The final software effort concentrated on linking the various software packages developed during the project to one another and other industry standard mapping packages, and establishing data connectivity with the oil company sponsors' databases.

Project leader Christopher Skelt of Scott Pickford, is not a programmer. He is a leading petrophysicist fascinated by using computers to solve problems. Skelt himself has developed a prototype "Minerva" system in PV-WAVE, that analyzes a set of wireline log and core data into component rock and fluid volumes using a variety of stochastic and deterministic mathematical techniques. With the backing of the DTI, Minerva is set to move from prototype to commercial system. The delivered software will provide other petrophysicists with a useful analysis tool.

Minerva has been designed to allow the user to investigate the validity of each measurement and its influence on the result. Typically, a series of models is applied sequentially, yielding a series of results, each of which is appropriate to one or more of the petrophysical environments encountered in the well. Other applications allow the user to refine parameter choice, synthesize combined results from the individual models, compare results with benchmark data, etc.

Minerva is already used in-house at Scott Pickford to good effect, and Skelt has handed over the system to his programming team for further development. "Minerva is unique in that it has been designed by petrophysicists for petrophysicists," said Skelt. "Using PV-WAVE allowed me to interactively develop a working system on an ad-hoc basis by turning my own knowledge into an analysis tool. I have been able to turn this working core over to the in-house developers to polish."

Scott Pickford's team will add a Motif® GUI that follows the logical flow of the petrophysical task. They will also develop links with industry standard databases and common petrophysical software packages.

RETURN ON INVESTMENT
According to Skelt, PV-WAVE allowed him to produce output that is familiar to petrophysicists. The color output mimics traditional plots but is interactive, allowing the user to experiment with the analysis and to filter out erroneous data. "Traditional, one-color line charts were static," explained Skelt. "By using PV-WAVE we can bring the graphs alive, adding color coding and allowing the user to interactively screen out erroneous data, or we can simply let the system start the process for him."

The functionality of PV-WAVE is such that the core algorithms are very short. The basic problem relates to an overdetermined linear set of equations. PV-WAVE will calculate and display the solutions, and Skelt's system allows users to develop a most probable single solution, using their own knowledge of petrophysics.

WORLD CLASS PRODUCTS, SERVICES, AND SUPPORT
Visual Numerics has provided technical software solutions for numerical analysis and visualization for over 30 years. The company's software products help users understand complex data from a variety of sources and build business-critical applications. Visual Numerics offers two product lines: the IMSL® Numerical Libraries for powerful mathematical and statistical analysis and the PV-WAVE® visual data analysis development environment. Visual Numerics also offers customized consulting services for applications that involve mathematical, statistical, or visual data analysis to meet today’s business analytical needs.