Different arrangements of words make different meanings, and different arrangements of meanings produce different effects.
–Pascal, Pensees, fragment 784
CD-ROM technology is poised to become the worldwide medium of choice for information distribution and storage. CD-ROMs will become the floppy disks of the future. CD recorders (CD-Rs) will be as common place as an office photocopier. The fact that the SEG has made all the technical articles and information from THE LEADING EDGE and from the GEOPHYSICS journals (1936-present) available on CD-ROM (GEOROM II), speaks strongly about the present utility and the future longevity of the medium. Besides GEOROM II, there are various other geophysical and geological auxiliary publications on CD-ROM, such as the American Geological Institute's GEOREF available from SilverPlatter Information Inc.
The success and popularity of a product is related to the guidelines established to facilitate the usage of the product. Historically, fundamental to the success of computer technology was the establishment of 9-track standards in the early 1950' s. Likewise, the SEG committees established digital recording standards in the early years of digital seismic data recording. Similarly, the establishment of international standards for the physical and logical format of compact discs played a major role in the worldwide acceptance of CD-ROM by the computing community.
At these early stages of applying CDROM technology for data storage by the seismic community, it is a categorical imperative that "arrangements different" from SEG conventions be avoided. The sensible approach for storing digital seismic field data on CD-ROM is to continue utilizing the recommended SEG standards.
"A proposal for the logical file structure of a CD-ROM was developed by a group of industry representatives meeting at the High Sierra Hotel and Casino in Lake Tahoe, Nevada, in 1985. The original participants in what became known as the High Sierra Group (HSG) were Apple Computer, Digital Equipment Corporation (DEC), Hitachi, Microwave Systems, Microsoft, Phillips, Reference Technology, Sony, 3M, TMS, Video-tools, Xebec, and Yelick.
The intent of the group was to define a file system that would be tailored to the characteristics of the optical read-only medium, yet would still be acceptable to a wide variety of operating systems running on a variety of computer hardware. Their work, which became known as the HSG proposal, was then submitted to the International Standards Organization (ISO). After some minor modifications and enhancements were made, the HSG proposal was adopted in late 1987 as ISO standard 9660."1
ISO 9660 interchange level 1 compliant filenames and directory names must satisfy the following conditions:
- Eligible characters are upper case "A" to "Z", "0" to "9" and"- "
- Filenames maximum of 8 characters
- Filenames extensions maximum of 3 characters
- Directory names maximum of 8 characters with no extension period separator "." permitted.
Directory depth is limited to eight levels. Furthermore, each file must be contiguous.2
SEG Field Data on CD-ROM
- Seismic field data stored on CDROM should comply with ISO 9660 interchange level 1 in addition to:
- maintaining the original form of the field data on tape as much as possible
- avoiding any possible confusion for the end user as to the format or composition of the original field data from tape.
- Directory structure may be determined by the particular client.
- Directories for area and line may be present, although mixing areas on a single CD-ROM is not recommended.
- Individual files consist of each field file with an associated header file if that file header is not attached to the body of the field file. The byte ordering or data format should not be changed.
- For large block seismic data, each block on tape should become a file on disc.
For trace data, such as SEG D demultiplexed format, the file header should form a separate header file, and all the traces for that file should be concatenated into an associated file. Individual trace positions should be able to be determined from the header file. A separate header is necessary for clarity and compatibility with large block. The SEG D format number in the header will indicate large block or trace data. A separate file header should always be used since some earlier "SEG D" data was recorded with a header length that did not agree with the length outlined in the header itself.
A separate log file containing the records read, their lengths and any eofs encountered could be saved in each tape directory. The log file should be in text format.
- The filename should be the file number itself, prefixed by zeros to fill out eight character positions. If the user so chooses, the first four characters may be "FILE" if the values of the file numbers do not exceed 9999.
- If a separate header block is present it is given an extension starting with the letter "H" and the seismic data body block is given an extension starting with the letter "S".
This makes the filenames sortable alphabetically with the header filename being ordered first.
There are three noteworthy consequences resulting from adopting the above approach.
Firstly, there are no extra bytes or extraneous information. The data is in its original untampered form.
Secondly, outsourcing services are not necessary when reproducing the seismic data from one CD-ROM onto another CDROM. No extra or special software is required and therefore no extra education or training process is required for staff to create a partial or complete copy of the seismic data. Thirdly, any user of the CD-ROM can readily identify the contents of the disc since each file from tape has a disc filename.
An obvious question is, "Is CD-ROM technology suitable as an archive technology for seismic data?" There are three fundamental objectives of any archive project.
- to preserve, to consolidate and to organize the database for future utilization
- to provide a high degree of accessibility of the database
- to provide an unified means of administration of the database
In recent years, archive projects of seismic field data have primarily focused only on the first of the three archive objectives. Only by fulfilling all three archive criteria, can a project be seen as successful. The net effect of a successful archive project is that the control of an invaluable database is regained and retained by the owners of the database.
The merits of CD-ROM technology sets CD-ROM technology apart as the only technology providing an economical means of regaining and retaining control of your archives.
|Standard Seismic Data3|
|FILE0001.S1||SEG 1 data|
|FILE0001.S2||SEG 2 data|
|FILE0001.HA||SEG A header|
|FILE0001.SA||SEG A data|
|FILE0001.HB||SEG B header|
|FILE0001.SB||SEG B data|
|FILE0001.HC||SEG C header|
|FILE0001.SC||SEG C data|
|FILE0001.HD||SEG D header|
|FILE0001.SD||SEG D data|
|L1NE0001.SP1||SEG P1 Positional Postplot location data|
|L1NE0001.SY||SEG Y data includes the EBCDIC and binary file headers concatenated with the data to make a single file. This has been a common practice in the industry for SEG Y disk files and should not be a problem since the headers are always a fixed length of 3600 bytes.
The tape image format should be maintained with EBCDIC characters and IBM mainframe byte order as recommended by the SEG standard.
Non-Standard Seismic Data
Some existing seismic data does not adhere to the SEG standards, but may be stored on CDROM in a similar manner, maintaining as closely as possible the recorded image and blocking on tape. This, for example, may be data that has an attached header, and the absence of a separate header file on disc would serve as one indication of the non-standard formats.
|FILE0001.SA||EPR format A (attached header)|
|FILE0001.SC||DDS or LRS format C (attached header) format unknown or unsure:|
|FILE0001.H||header (if it exists)|
|L1NE0001.SGY||Non-standard SEG Y with ASCII header and modified format such as IEEE data format and/or PC byte order may exist for some stacked data, but is NOT a recommended format for field data.|
1 Mirecki, T., "A Tree Grows in the High Sierra": PC TECH JOURNAL, October 1988: p. 58
2 Compact Disc Terminology 2nd Edition 1993: Disc Manufacturing, Inc.
"Recommended standards for digital tape formats": Geophysics,v. 32, p. 1073-1084; v. 37, p. 36-44; v. 40, p.344-353 "SEG standard exchange formats for positional data": Geophys~s,v.48, p.488-503 "Recommended standard for seismic (/radar) files in the personal computer environment": Geophys~s,v. 55, p. 1260-1271