Overview

This section discusses the contents of the SAC data file, describes the binary and alphanumeric formats of this file, and documents the SAC header in detail.

Since version 100.0, SAC can handle binary data files in either endian (byte order), so big-endian systems (Sun Solaris, Mac PPC) can read SAC data files written on little-endian systems (Linux, MAC i686, Cygwin) and vice versa.

Each signal is stored on disk in a separate SAC data file. Through SAC version 101.6a, these files contain a fixed length header section followed by one or two data sections. The header contains floating point, integer, logical, and character fields. Evenly spaced data files have only one data section which contains the dependent variable. Unevenly spaced data and spectral data files contain two data sections. For unevenly spaced data, the first data section contains the dependent variable and the second contains the independent variable. For spectral files the first component is either the amplitude or the real component and the second component is either the phase or imaginary component.

All reals in the SAC header and data sections are single-precision (32 bits, 4 bytes). As discussed in the section on precision in TUTORIAL, although having single-precision for the data is sufficient, having only single-precision for time and distance variables in the header is no longer good enough to handle many modern data sets. SAC version 102.0 solves this problem. The header is unchanged, but a "footer" is added with 22 header variables in double-precision. Version 102.0 can handle "old" SAC data files with no footer or files with a footer, so there is complete compatibility. SAC v102.0 distinguishes between "old" and "new" files by the header variable NVHDR: if NVHDR = 6, the SAC file is in "old" format, if NVHDR = 7, it is in "new" format with a footer. See see the section on Floating-Point Precision in SAC in TUTORIAL for a more detailed discussion.

SAC Binary Format for NVHDR = 6

The binary format is the most common one use. It is is the default format in the READ and WRITE commands and in the subroutine library (RSAC1, RSAC2, WSAC1, WSAC2, WSAC0.) These are binary (unformatted) files.

Structure of SAC Binary Data File for NVHDR = 6

SAC Binary Header

The following table shows the contents and layout of the SAC binary data file header. The W and T columns give the beginning word and header data type for the header variables named on that line. These header variables and data types are described later in this section. If the name is INTERNAL then that variable is internal to SAC and not normally of interest to the user. If the name is UNUSED then that variable is not currently being used. For any given file, some of these variables will not have meaningful values. These are refered to as "undefined variables" for that file. For each data type, a special value signifies this undefined state. They are listed in a table at the end of this section.

Header Data Types

This table lists the header types and their definitions. The third column lists the special value used to signify that a particular header variable is undefined in a particular file.

Header Variables

KEVNM is 16 characters (4 words, 16 bytes) long.

All other K fields are 8 characters (2 words, 8 bytes) long.

Structure of SAC Binary Data File for NVHD = 7

The only difference between a SAC binary data file with NVHDR = 6 and one with NVHDR = 7 is the addition of a "footer" following the data section(s). The footer is comprised of 22 double-precision (64-bit) floating point variables: DELTA, B, E, O, A, T0 ... T9, F, EVLO, EVLA, STLO, STLA, SB, and SDELTA. The names are the same as in the header; SAC distinguishes between the single- and double-precision variables in internal calculations by their location in the file.

SAC Alphanumeric Format

This file is essentially the alphanumeric equivalent of the SAC binary data file. The header section is stored on the first 30 80-character lines. For both NVHDR = 6 and NVHDR = 7, this is followed by one or two data sections. The data are in 5G15.7 format. The following table shows the line number, formats and names of the variables on the header section lines.

SAC Alphanumeric Data File Example

The header section and first five lines of the data section of a sample SAC alphanumeric data file is shown below. You can reproduce this file (with the entire data section) on your system by executing the following commands:

SAC> FUNCGEN SEISMOGRAM
SAC> LH
SAC> CH NVDHR 7
SAC> WRITE ALPHA TEMP1

You can read this alphanumeric file directly into SAC with the following commands:

SAC> READ TEMP1
SAC> LH

SAC knows the difference between an alphanumeric and binary SAC files and should be able to read them directly. You can specify an ALPHA flag to the READ command for backwards compatibility:

SAC> READ ALPHA TEMP1
SAC> LH

Listed next are the first lines of TEMP1 -- header plus first 8 lines of data and ending with the 22 lines of the footer (one line per variable).:

    0.01000000      -1.569280       1.520640      -12345.00      -12345.00
      9.459999       19.45000      -41.43000       10.46400      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
     -12345.00       48.00000      -120.0000      -12345.00      -12345.00
      48.00000      -125.0000      -12345.00       15.00000      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
      373.0627       88.14721       271.8528       3.357465      -12345.00
     -12345.00    -0.09854718       0.000000       0.000000      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
     -12345.00      -12345.00      -12345.00      -12345.00      -12345.00
     1981        88        10        38        14
        0         7         0         0      1000
   -12345    -12345    -12345    -12345    -12345
        1        50         9    -12345    -12345
   -12345    -12345        42    -12345    -12345
   -12345    -12345    -12345    -12345    -12345
   -12345    -12345    -12345    -12345    -12345
        1         1         1         1         0
CDV     K8108838
-12345  -12345  -12345
-12345  -12345  -12345
-12345  -12345  -12345
-12345  -12345  -12345
-12345  -12345  -12345
-12345  -12345  -12345
-12345  -12345  -12345
   -0.09728001    -0.09728001    -0.09856002    -0.09856002    -0.09728001
   -0.09600000    -0.09472002    -0.09344001    -0.09344001    -0.09344001
   -0.09344001    -0.09344001    -0.09472002    -0.09472002    -0.09344001
   -0.09344001    -0.09216000    -0.09216000    -0.09216000    -0.09216000
   -0.09088002    -0.09088002    -0.09216000    -0.09344001    -0.09472002
   -0.09472002    -0.09472002    -0.09472002    -0.09472002    -0.09472002
   -0.09344001    -0.09344001    -0.09216000    -0.09088002    -0.09088002
   -0.09216000    -0.09216000    -0.09216000    -0.09344001    -0.09472002
   ....
0.0099999997764825821
9.4599990844726562
19.449998861178756
-41.430000305175781
10.46399974822998
-12345
-12345
-12345
-12345
-12345
-12345
-12345
-12345
-12345
-12345
-12345
-125
48
-120
48
-12345
-1234

SAC Header Variables

This table lists the header variables, their types, and descriptions. They are grouped by category: required fields, time fields, phase picks, instrument parameters, station parameters, event parameters, misc. The header types are defined in the second table.

Phase Picks

Instrument Fields

Station Fields

Event Fields

Miscellaneous Fields

Enumerated Header Field Values

The enumerated header field values are stored in the header as integers. Their names and values are given in the table below.

REFERENCES

[IAU2000] https://link.springer.com/article/10.1023/A:1013939327465

Seidelmann, P., V. Abalakin, M. Bursa, M. Davies, C. De Bergh, J. Lieske, J. Oberst, J. Simon, E. Standish, P. Stooke, et al. (2002), Report of the IAU/IAG working group on cartographic coordinates and rotational elements of the planets and satellites: 2000, Celestial Mechanics and Dynamical Astronomy, 82 (1), 83–111.

[IAU-WGCCRE] https://astrogeology.usgs.gov/groups/IAU-WGCCRE

[Meftah2015] https://link.springer.com/article/10.1007%2Fs11207-015-0655-6

Meftah, M., A. Irbah, A. Hauchecorne, T. Corbard, S. Turck-Chi`eze, J.-F. Hochedez, P. Boumier, A. Cheva- lier, S. Dewitte, S. Mekaoui, et al. (2015), On the determination and constancy of the solar oblateness, Solar Physics, 290(3), 673–687.

• 8.189e-6 (Flattening)

[Rozelot2001] https://link.springer.com/article/10.1023/A:1005238718479

Rozelot, J. P., Godier, S., & Lefebvre, S. (2001). On the theory of the oblateness of the Sun. Solar Physics, 198(2), 223-240.

• 8.33e-6 (Flattening)
• 6.50e-6 - 10.20e-6 (Flattening Range)

LATEST REVISION

Version 102.0