File formats for Pasteup

Pick file

Pasteup can handle several type of pick file formats.

Pasteup format

Pasteup pick format is a text file format based on the MacRay pick file format, but slightly extended.

The first line has a special meaning. It consists of 4 columns.

column contents
1 receiver (or shot) x-position (km)
2 receiver (or shot) z-position (km)
3 must be 0 (obsolete parameter)
4 must be 1 (obsolete parameter)
Meaning of each column in the first line.

The rest rows of the pick file contains each pick information. Each row must have 5 columns as shown in below.

column contents
1 shot number (integer)
2 offset distance between source and receiver (km)
3 traveltime (sec), not reduced
4 uncertainty, or accuracy of traveltime data (sec)
5 pick type (integer between 1 and 200)
Meaning of each column in the "pasteup" pick format.

Here is an example.

50.000 1.500 0.0 1.0
522 6.386 4.275 0.02 1
523 6.282 4.260 0.02 10
524 6.178 4.246 0.05 1
525 6.075 4.232 0.05 4

RAYINVR format

RAYINVR is one of the most famous program package of raytracing and inversion developed by Colin Zelt. See the documents of RAYINVR for details.

GeoCT-II format

GeoCT-II is a commercial packages for inversion analysis. See its manual for details.

Mochi format

Mochi format is originally used by Dr. K. Mochizuki. This format is very simple and can be easily handled. The first line represents the number of picks in the file and remaining lines represent x(km) and t(sec).

Here is an example.

3
1.0   3.123
2.5   4.224
3.2   5.197

Synthetci traveltime files

Pasteup supports several formats for synthetic traveltimes, including MacRay, RAYINVR, GeoctII. Pasteup identifies these formats automatically.

However, Pasteup displays only a record section of a common shot or a common receiver data, therefore, synthetic pick file must contain only one shot point or only one receiver.

Parameter file (resource file)

Pasteup can save all the parameters to draw a record section in a parameter file (resource file). The file contains data file names, drawing conditions and so on.

The parameter file is just a tclsh script composed of only "set" assignment statements. Very easy to read even if you don't know the grammar of the tclsh. Usually, you don't need to see and edit the parameter file.

Seismic data file

Pasteup supports several data formats such as Seismic Unix, SEG-Y and so on. In either case, the data must be a receiver gather or a shot gather.

The Developer uses SU format. Therefore, recommended file format is SU.

Seismic Unix (SU) format

Seismic Unix (SU) format is based on the SEG-Y. See the SU manual for detail. The supported SU format is the usual float format (depending on the CPU architecture). XDR format is not supported.

Pasteup reads the following header words. All of these are the standard SEG-Y headers.

Mandatory header words

Header word Description
dt Time sample interval
ns Number of samples
trid Trace identification number
If the SU file is a migrated image, trid must be 200, otherwise trid must be 0. In the migratated image, dt means depth interval.

Optional Header words

Header word Description
derlrt Delay recording time
tstat Total static applied in millisecond
laga Lag time A (millisecond)
lagb Lag time B (millisecond)
offset Offset distance in meter
tracl Trace sequence number within line
tracr Trace sequence number within SU file
tracf Trace sequence number within field record
fldr Field record number
ep Energy source point number
cdp Ensemble number

Starting time of the trace is calclated by

delrt + tstat + lagb - laga

In the receiver gathers, shot number is read from one of tracl, tracf, tracr, fldr; and the receiver number is read from one of fldr or ep.

In the shot gathers, receiver number (site number) is read from one of tracl, tracf, tracr, fldr; and the shot number is read from one of fldr or ep.

SEG-Y

SEG-Y is one of the standard file formats for seismic data defined by the Society of Exploration Geophysicists (SEG). Each trace in SEG-Y format has 240 bytes header, but the standard SEG-Y defines only the first 180 bytes of them. The rest of the header (181-240 byte) are left for optional use. Therefore, several extended SEG-Y formats have been developed using this optional header.

As you know, SU is one of extensions of SEG-Y format, pasteup reads the same header words in the SEG-Y format as in the SU format.

Hopefully, The following SEG-Y extension is supported.

Passcal SEG-Y

This format was defined by IRIS PASSCAL Instrument Center. For details, see the web page.

The sampling interval is determined by the Passcal extended header, located between 201-204 byte, not by the standard SEG-Y header. Delay of recording time is computed using the extended header are, located between 207-220 byte.

IASPEI SEG-Y

Pasteup supports IASPEI SEG-Y format of version 3.00 (January 25, 1993). See their documents for details.

Delay of recording time is computed using the data information in the extended trace header, located between 181-200 byte.

Kasahara-labo format

Historically speaking, pasteup was developed for the data format used in Kasahara-lab, Earthquake Research Institute (ERI), University of Tokyo. This format is based on the Urabe-Hirata format, but slightly different from it. In this format, seismic data consists of two files; a waveform file and a corresponding record header file.

Although this file format is not sophisticated, it has some advantages.

  1. Waveform can be 2-byte integer, 3byte integer and 4-byte float.
  2. Header file is a text file and easily readable, editable.

In some cases, the storage area is much saved if you use this format in place of SU or SEG-Y.

Waveform data are stored in block unit, where 1 block consists of 1024 samples. Record header file have information about the shot time and recording delay, the corresponding block position and so on.

Record header format

The record header file is a text file; a shot per a line. Each line has 83 letters. Each columns has the following meanings.

Columns Format Description
0-3 4d Line number in the file, arbitrary number, not the exact line number
5-8 4d Shot number
10-11 2d Shot time (Month)
12-13 2d Shot time (Day)
14-15 2d Shot time (Hour, in 24H)
16-17 2d Shot time (Minuite)
18-23 6.3f Shot time (Second)
25-32 7.3f Difference between True-shot-time and time of first-sample, in second
34-41 8.3f Offset distance in km
42-47 6.3f The error of offset in km
49-55 7.2f Sampling rate in Hz
56-60 5d Accumulated block number
61-63 3d Block number for this trace
65-68 4d Sample number in the last block of this trace
69-72 4d The weight of explosion
73-78 6.3f The time correction for the depth of explosion, in second
79-83 5d Water depth in meter

Here is a sample record header file.

  15   15  831 11920.554   -0.004   32.936 0.000  100.00    0  4  928   0 0.020    0
  16   16  831 11940.554   -0.004   32.870 0.000  100.00    4  4  928   0 0.020    0
  17   17  831 120 0.554   -0.004   32.801 0.000  100.00    8  4  928   0 0.020    0
  18   18  831 12020.554   -0.004   32.735 0.000  100.00   12  4  928   0 0.020    0

Tohoku-Univ. format

The Tohoku Univ. format was, I think, developed by Tohoku University. This format is multiplex; 4-component data are in one file. Each trace has trace header information such as shot position (latitude and longitude), shot time, trace length, sampling interval and amplitude gain. However, there is no information about offset distance and receiver position(or shot position in a shot gather).

Thus, to make a record section, you need to provide the receiver position (or shot position in a shot gather) by latitude and longitude, then the pasteup compute offset distance between shots and the receiver on the fly based on the WGS84 ellipsoid.