load

Group:

Authors:

Igor B. Morozov, 01/97, dataset access added 10/00

Dataset data load

Loads seismic traces from a dataset given in field NAME. This name can contain:

1. The name of a dataset created by tool DATASET. This dataset can contain multiple files of various types, data retrioeved from SQL databases, Unix pipes, or internet connections. See the documentation for tool DATASET.
2. File name of Unix file name mask for ordinary files. The interpretation of file formats is based on file-name suffixes ("extensions") listed below.
3. Named buffers of traces created in the same job by tool STORE (see list STORE)

When using trace data files without explicit specification of their formats (even when included in datasets presented by tool DATASET), the following automatic interpretations of file-name suffixes is used:

• ".sia" or ".igeos" - native SIA (IGeoS file format (same as input by tool DSKRD
• ".segy" or ".sgy" - SEGY format (revision 1 is currently default)
• ".sgyi1" - SEGY format with data encoding in 1-byte integers
• ".sgyi2" - SEGY format with data encoding in 2-byte integers
• ".sgyi4" - SEGY format with data encoding in 4-byte integers
• ".sgyf4" - SEGY format with data encoding in 4-byte floats
• ".sgyp" or ".sgp" - Passcal SEGY extension
• ".seg2" or ".dat" - SEG 2 format
• ".su" - Seismic Unix
• ".sui2" - Seismic Unix with data as 2-byte integers
• ".sui4" - Seismic Unix with data as 4-byte integers
• ".seed" - IRIS SEED format
• ".mseed" or ".miniseed" - IRIS MiniSEED format
• ".dlseed" - IRIS datalsess SEED format
• ".sac" or ".sc" - SAC format (binary)
• ".sac-be" - binary SAC format with big-endian encoding
• ".sac-le" - binary SAC format with little-endian encoding
• ".mat", ".oct", or ".octave" - text workspace files produced by "save" command in Octave (and maybe Matlab with proper options selected)

NOTE: not sure whether all of these formats are currently completely implemented.

Notes about some input formats

• SEGY format extensions can be further adjusted by lisy SEGYFMT. Trace header remapping for SEGY and other binary formats can be done by lists THEADER and SHEADER.
• From Octave (".mat") files, matrices named "data" are loaded by columns as data traces. Each column is loaded as a separate single-component trace. Variables "timing" are used as trace timing information and should contain (at least) two elements. The value timing(1) is interpreted as the time start of the trace, and timing(2) is interpreted as sample interval. Any other variables are loaded into trace headers provided the headers with the same names are defined in the job (in tools preceding LOAD or in list HEADERS). If the Octave variable (for example, "ffid") contains only one element, its value is assigned to teh header "ffid" in all traces loaded from the workspace file. If the variable is a matrix with more than one element (for example, "chan"), then each of its values is assigned to the corresponding output trace.

Other miscellaneous notes

• All HDRMATH operations can be performed with headers of input traces.
• Input traces can be selected by using the same selectors as used by tool IF.
• Specified time ranges can be extracted from input traces by list TRANGE.
• List INSERT allows several ways for merging the data input from current dataset with the traces entering the tool from preceding processing. List ENSEMBL can be used to ensure that the data are loaded by complete trace ensembles.
• For interactive processing, LOAD offers entry points (list NAME) for receiving commands from interactive tools.

See also:

DATASET, STORE, DSKRD, RDSEG2, READBF, WSERV

Module call format: