Raster images offer low-cost well log preservation

Susan L. Cisco
Cisco & Associates
Austin

Raster and vector logs
Both raster and vector well logs are digital representations of well logs. Raster digital well logs represent bitmaps of the log image in a rectangular array of black and white dots called pixels. The white pixels are represented by zeros, and the black pixels are represented by ones.
Raster images are generated using a relatively simple and inexpensive optical scanner.
Vector digital well logs contain the numeric values of well log curve. Vector data are more expensive to digitize, requiring considerable time for hand digitizing and editing.

Well log data saved as depth-calibrated raster images provide an economic alternative to digital formats for preserving this valuable information into the future.

Although often thrown away after vectorization, raster imaged well logs may be the key for a global computer-readable format for legacy hardcopy data (see box for definition of raster and vector logs).

A legacy log is one obtained in the past that does not have the same format as current logs.

Over the past several decades, oil and gas companies have acquired an enormous volume of well logs, maps, and seismic and core data. Preservation of, and access to, valuable data such as well logs is critical to the petroleum industry.

These legacy data are stored on multiple media and contain information for a variety of applications in addition to resource exploration and development, such as environmental protection, water management, global change studies, and basic and applied research.

Well logs

Well logs provide fundamental information to geologists and petroleum engineers who support exploration and development activities. Decades of legacy well log data are recorded on different physical media (paper, microfilm, optical, and magnetic) in a variety of so-called standard formats.

Because of an enduring value, most well log data are retained indefinitely, so that the volume increases over time. In some cases, well log data are unique and cannot be replaced because of urban development and new restrictions that place land off limits to exploration and development activities.

The marketplace offers a variety of solutions for the management of well logs throughout their entire life.

The comments in the box on storage reflect some of the problems related to log storage.

Three criteria for evaluating data management are: acquisition cost, storage requirements, and ease of use by geoscientists.

Acquisition cost

For comparison purposes, several well known vendors of well log data were asked to quote prices for well logs in paper, microfiche, digital raster, and vector data formats. They were to assume a typical suite of two logs (such as resistivity and neutron density), four curves per log, for a 12,000 ft well.

Table 1 [97948 bytes] summarizes the costs. One should understand that prices are only illustrations and can vary depending on the volume and condition of the logs and the required turnaround time.

The cheapest storage format for well logs is microfiche at a cost of $2.25/two well logs on one microfiche. Microfiche readers with continuous form printers, no longer manufactured, are required to print well logs on paper.

Thus, the future of well logs on microfiche is likely to be limited to its use as a preservation medium for secure storage for up to 500 years.

The oldest and most familiar storage format for well logs is paper format. A well log duplicated at half scale (1 in. of well log equals 100 ft of well bore) costs between $5 and $20.80/log. The $5 price is for Cambe Geological Services Inc. log library members who pay at least $300/month in member fees and the $20.80 price is at Riley's Electric Log Inc.

The most expensive storage format for well logs is digitized vector data, ranging from $100/well to Cambe log library members (if the log already has been digitized), to $1,000/well for a 5-in., full-scale log from A2D Technologies Inc.

A newcomer to the well log data marketplace is depth-calibrated raster images of well logs. First proposed by Interpretive Imaging LLC., smart rasters are raster images of well logs enhanced with depth calibration information. Priced at $15/log, the data set includes well header/well spot data that are integrated into a data base management system.

Additional software can be purchased to vectorize portions of the raster images on demand.

Storage requirements

Paper well logs take up a large volume of physical storage space and quickly create mountains of information (Table 2 [92018 bytes]). Storing paper well logs on-site in active file areas costs $0.127/log/year at Amoco Production Co., not including personnel costs or filing supplies. Logs can be stored for $0.43/log/year in an off-site storage area, not including labor or cost of filing supplies.

Microfiche takes 98% less space than paper. Silver halide original microfiche must be stored in a climate-controlled environment to ensure media stability.

A raster image of a well log requires more computer storage than vector data for the same well log, depending on the complexity of the source well logs. Both raster and vector formats require investments in data refreshing and migration to new platforms over time.

Ease of use

For most geoscientists, paper presents the most familiar storage format for well logs. To construct cross sections and correlations for geologic evaluations, paper well logs must be manually cut and pasted up on a drafting table or empty wall, a time-consuming operation. Also, well log users have to be located where the logs are stored physically.

With microfiche, the same constraints on correlation and cross section construction apply; however, paper duplicates must be printed prior to the time-consuming cut-and-paste process.

Raster well logs can be manipulated side-by-side on the screen of a geologist's work station for correlation, tops picking, and structural and stratigraphic cross section construction. The depth-calibrated (intelligent) raster logs from Interpretive Imaging allow geologists to automatically hang logs on any datum in the correct (true) scale and print them.

Vector well logs provide all the functions that raster images do plus more, such as water saturation and mineralogy studies. Simultaneous access to well logs by multiple users, including access by individuals in remote locations is possible with networked raster and vector data.

The future

According to The Information Store, Houston, there are 32 known national geological and geophysical data repository efforts under way as of September 1996. As more resource exploration ventures are partnered to share the cost and risk, oil and gas companies need to share geological and geophysical data from exploration fields of mutual interest.

Within companies, interdisciplinary data sharing is required to seamlessly integrate data from field operations with data from corporate offices. Oil and gas companies are outsourcing more data management functions to oil field service providers.

The motivation is more compelling than ever to integrate the various formats of geological and geophysical data within a company (or among several business entities) into one standardized system so that individuals, companies, and countries can share information throughout the entire life cycle of an oil or gas field.

Legacy hardcopies

Although they lack "computer intelligence," the value of legacy hardcopy analog well logs will not disappear anytime soon. In the future, legacy paper and reproducible logs will be stored in inexpensive, secure off-site storage and digitized on demand.

Although the cheapest storage format for well logs is microfiche, it lacks computer intelligence. Thus, the future of microfiche will be limited to its use as a preservation medium for long-term storage.

Digital data

Emphasis in well log data management over the last 10 years has been on "intelligent" well logs in the digital vector format. The high cost of vector data acquisition (from $100 to $1,000/well) means that companies can afford to vectorize only selected wells, logs, and log intervals.

At $15/well log, depth-calibrated (intelligent) raster well logs offer an economical alternative digital format and make it feasible to acquire complete digital well log data sets for geologic studies. With raster data, selected portions of well logs can be converted to vector data on demand.

The feasibility of using raster data as the global computer-readable format for well log data management is questioned by many petroleum industry watchers. In practice, a raster image of a paper or microfiche well log must be made prior to digitizing it as vector data. Some companies "throw away" the raster images when vectorization is complete.

Dave Kotowych at A2D Technologies explained that, "most of our customers don't want the raster file because it is a nonintelligent data set. With a digital file, you can reproduce a hard copy log if necessary." Also, raster files are larger than vector files and tie-up printer and Intranet/Internet resources when networks are not configured correctly.

Costs, benefits

Clearly, the global petroleum industry needs an economical information environment that integrates all geological and geophysical data associated with the oil or gas field life cycle. Although the inherent intelligence of vector data is invaluable, high costs prohibit their use as a global computer-readable format for legacy hardcopy data.

Maybe it is time to question conventional thinking about digital formats and consider the more economical approach of raster well log data as the bedrock of the digital information environment of the future.

Problems with well log storage

VARIOUS PERSONS HAD THESE comments and observations on well logs and storage:

"Standards are absolutely necessary for effectively managing well logs," said Doris Hughes, manager of information resources and services at Crimson Corp. Services, to whom BP Exploration Inc., in Houston, outsources the management of its hardcopy exploration, production, and administrative data.

Hughes adds: "API (American Petroleum Institute) well identifier numbers need to be readily available on all well logs. We also need standard naming conventions for well names that can be linked to the API numbers and standard nomenclature for identifying different types of well logs."

Well names cause Amoco Production Co. problems too, according to Mary L. Cytrynbaum, senior information specialist. "It is difficult to enter well information correctly into data bases because well log header information may be incomplete or contain incorrect information."

One example, she said, was a well whose name had been entered into Amoco's data base three different ways: San Jan Gravel A 1E, San Juan Gravel A-1-E, and San Juan Gravel A-1E.

Additional variations on the well name are possible. Also, well naming can cause problems when a well is sold, and the new owner changes the well's name. Variations in well names can cause problems when retrieving well logs from data bases and filing systems.

Ed Garner's principal problems with paper well logs are their sheer volume and the personnel required to process them into the well log data management system. Garner is manager of the University of Texas at Austin, Bureau of Economic Geology Well Log Library.

Garner said, "When you file well logs geographically by state, then by township-range section and well name, you end up constantly shifting well logs from shelf-to-shelf, drawer-to-drawer, or moving them to off-site storage because most of the time, you don't know or can't predict what sections of the log library will expand."

Garner resolved the problem by assigning a unique identifying number to each well log and arranging them numerically. Thus, the filing system always grows at one end of the system where growth can be effectively managed. Refiling is easier because well logs going back into the filing system can be sorted into numerical order. Misfiles are reduced because it is easier to file numerically then alphabetically. There are 26 letters to remember and only 10 numbers.

Neil McNaughton (editor, Petroleum Data Manager) said that petroleum data management, "...has reached the state where something just has to be done: old tapes need remastering, interpretations are rarely correctly archived, and paper data, which may still contain mission-critical information, is often effectively lost to posterity."

The Author

Susan L. Cisco is president and principal consultant of Cisco & Associates, a consulting and research practice that advises companies on records retention issues and implementation of document management systems. She has more than 20 years' experience in coordinating records and information management activities in numerous industries. In addition, she is a lecturer at the University of Texas, Graduate School of Library & Information Science.

Cisco holds a BA from Northwest Missouri State University, and a Masters in Library Science and PhD from the University of Texas at Austin. She is a certified records manager, a member of the Association for Records Managers & Administrators (ARMA), and a member of the Association for Information & Image Management. Currently, she is manager of ARMA's industry-specific program for the petroleum, utilities, and environmental industries.