Read and manipulate miniSEED records


scmssort reads unsorted (and possibly multiplexed) MiniSEED files and sorts the individual records by time. This is useful e.g. for simulating data acquisition and playbacks. Removing of duplicate data and trimming of time window is available.

scmssort reads single files and output to the command line. Cat many files to read them at the same time. In this way huge amount of data can be processed efficiently.

Applications to miniSEED records:

  • Sort records by time, e.g., for playbacks.

  • Remove duplicate records from files and clean waveform archives.

  • Filter data records, i.e. keep or remove them, based on

    • time windows,

    • stream lists where each line has the format NET.STA.LOC.CHA including regular expressions. Such stream lists can be generated, e.g., using scinv.


  • Combine with scart or msrtsimul to archive data or to make playbacks with real-time simulations.

  • Filter data by stream IDs using NSLC lists which can be generated using scinv.


  1. Read a single miniSEED data file. The records are sorted by endtime and duplicates are removed.

    scmssort -vuE unsorted.mseed > sorted.mseed
  2. Read all files ending with “.mseed” at the same time. The data are trimmed to a time window and duplicated or empty records are ignored.

    cat *.mseed | scmssort -vuiE -t 2020-03-28T15:48~2020-03-28T16:18 > sorted.mseed
  3. Remove streams listed by stream code and sort records by end time. Also ignore duplicated or empty records. Stream lists can be generated, e.g., by scinv.

    scmssort -vuiE --rm -l stream-list.txt test.mseed > sorted.mseed
  4. Extract streams by time and stream code and sort records by end time. Also ignore duplicated or empty records.

    echo CX.PB01..BH? | scmssort -vuE -t 2007-03-28T15:48~2007-03-28T16:18 -l - test.mseed > sorted.mseed
    scmssort -vuiE -t 2007-03-28T15:48~2007-03-28T16:18 -l stream-list.txt test.mseed > sorted.mseed

Command-Line Options

scmssort [options] input


-h, --help

Show the command-line help and exit.

-E, --sort-by-end-time

Sort by end time.

-i, --ignore

Ignore all records which have no data samples.

-l, --list

File with streams to filter the records. One stream per line. Instead of a file read the from stdin (-). Line format: NET.STA.LOC.CHA - wildcards and regular expressions are considered. Example: CX.*..BH? .

NSLC lists may be generated, e.g., using scinv.

-o, --output

Name of output file for miniSEED data (default is stdout).

-r, --rm

Remove all traces in stream list given by --list instead of keeping them.

-t, --time-window

Specify time window (as one -properly quoted- string). Times are of course UTC and separated by a tilde ~



-u, --uniqueness

Ensure uniqueness of output, i.e. skip duplicate records.

-v, --verbose

Run in verbose mode.