Submodules

Name	Purpose
ASCII	ASCII file formats (includes GeoCSV, SLIST, and variants)
FastIO	Replacement low-level I/O functions to avoid thread locking
Quake	Earthquake seismology
RandSeis	Generate SeisBase structures with quasi-random entries
SEED	Standard for the Exchange of Earthquake Data (SEED) file format
SUDS	Seismic Unified Data System (SUDS) file format
SeisHDF	Dedicated support for seismic HDF5 subformats
UW	University of Washington data format

Using Submodules

At the Julia prompt, type using SeisBase.NNNN where NNNN is the submodule name; for example, using SeisBase.Quake loads the Quake submodule.

RandSeis

This submodule is used to quickly generate SeisBase objects with quasi-random field contents. Access it with "using SeisBase.RandSeis"

The following are true of all random data objects generated by the RandSeis module:

Channels have SEED-compliant IDs, sampling frequencies, and data types.
Random junk fills :notes and :misc.
Sampling frequency (:fs) is chosen from a set of common values.
Channel data are randomly generated.
Time gaps are automatically inserted into regularly-sampled data.

SeisBase.RandSeis.randPhaseCat — Function

randPhaseCat()

Generate a random seismic phase catalog suitable for testing EventChannel, EventTraceData, and SeisEvent objects.

source

SeisBase.RandSeis.randSeisChannel — Function

randSeisChannel()

Generate a random channel of geophysical time-series data as a SeisChannel.

Keywords

KW	Default	Type	Meaning
s	false	Bool	force channel to have seismic data?
c	false	Bool	force channel to have irregular data?
nx	0	Int64	number of samples in channel [1]
fs_min	0.0	Float64	channels will have fs ≥ fs_min
fc	0.0	Float64	rolloff frequency [2]

if nx ≤ 0, the number of samples is determined randomly
specifing fc with c=false returns a geophone instrument response

See also: randSeisData, fctoresp

source

SeisBase.RandSeis.randSeisData — Function

randSeisData()

Generate 8 to 24 channels of random seismic data as a SeisData object.

randSeisData(N)

Generate N channels of random seismic data as a SeisData object.

Keywords

KW	Default	Type	Meaning
s	0.6	Float64	chance each channel is seismic data
c	0.2	Float64	chance each channel is irregular (fs=0)
nx	0	Int64	number of samples in each channel
fs_min	0.0	Float64	regular channels have fs ≥ fs_min
a0	false	Bool	recalculate :resp constants?

Behavior

If nx ≤ 0, the number of samples is determined randomly for each channel
100*s is the minimum percentage of channels with guaranteed seismic data
100*c is the maximum percentage of channels with irregularly-sampled data
s takes precedence over c; they are not renormalized. Thus, for example:
- randSeisData(12, c=1.0, s=1.0) works like randSeisData(12, c=0.0, s=1.0)
- randSeisData(c=1.0) works like randSeisData(s=0.0)
For N channels, s*N is rounded up; c*N is rounded down. Thus:
- randSeisData(10, c=0.28) and randSeisData(10, c=0.2) are equivalent
- randSeisData(10, s=0.28) and randSeisData(10, s=0.3) are equivalent
By default, :resp.a0 = 1.0 for all channels; a0=true calls resp_a0!(S)
- :resp.p and :resp.z are random; a0=true isn't guaranteed to work

SEED

Submodule for the Standard for the Exchange of Earthquake Data (SEED) file format; includes additional functionality.

Scanning SEED Volumes

SeisBase.SEED.scan_seed — Function

soh = scan_seed(fname::String[, KWs])

Scan seed file fname and report properties in human-readable string array soh.

General Keywords

quiet (Bool): true to only return compact summary strings (no stdout)
memmap (Bool): true to use memory mapping
v (Integer): v > 0 increases scan verbosity

Output Keywords

These are all Booleans; false excludes from scan.

npts: Number of samples per channel (default: true)
ngaps: Number of time gaps per channel (default: true)
nfs: Number of unique fs values per channel (default: true)
seg_times: Exact gap times (default: false)
fs_times: Exact times of fs changes (default: false)

Note that seg_times and fs_times dump verbose per-channel tabulation to stdout.

Users are encouraged to submit feature request or issues if there's a need to scan for other changes within a SEED volume.

Interaction with Online Requests

scan_seed cannot interact directly with online SEED requests. As a workaround, do $get_data(..., w=true)$ to dump the raw request directly to disk, then scan the file(s) created.

Caution

Rarely, the number of gaps reported is off-by-one from read_data.

source

UW

The UW submodule extends functionality for the University of Washington (UW) file format(s).

The UW data format was created in the 1970s by the Pacific Northwest Seismic Network (PNSN), USA, for event archival. It remained in use through the 1990s. A UW event is described by a pickfile and a corresponding data file, whose filenames were identical except for the last character. The data file is self-contained; the pick file is not required to read raw trace data. However, station locations were stored in an external text file.

Only UW-2 data files are supported by SeisBase. We have only seen UW-1 data files in Exabyte tapes from the 1980s.

SeisBase.UW.uwpf — Function

H, R = uwpf(pf[, v])

Read UW-format seismic pick file pf into SeisHdr object H, with seismic source description (focal mechanism) returned in SeisSrc object R.

uwpf!(W, pf[, v::Integer=KW.v])

Read UW-format seismic pick info from pickfile f into SeisEvent object W. Overwrites W.source and W.hdr with pickfile information. Keyword v controls verbosity.

Caution

Reader has no safety check to guarantee that pf is from the same event.

source