SAGE Facility MUSTANG metrics Web Service Documentation

This metric calculates Signal to Noise Ratio (SNR) as the ratio of the RMS variance for the 30 seconds following versus the 30 seconds prior to the predicted first-arriving P phase from an event of magnitude >= 5.5.

sample_snr is well suited to identifying events with clear first arrivals at a particular station for tasks such as verifying polarity and channel orientation and for first motion studies. SNR values greater than 10 generally have clearly visible arrivals, whereas arrivals with lower SNR values may require noise filtering.

Because sample_snr is run only on events with magnitudes >= 5.5 in the NEIC PDE catalog, it is unnecessary to cull noise “events”, as is necessary with sample_stalta. Those interested in identifying events with clear arrivals from local sources having smaller magnitudes, however, will need to use max_stalta.

Events – magnitude >= 5.5
Traces – one N.S.L.C (Network.Station.Location.Channel) per measurement
Window – 60 seconds (UTC) centered on the predicted P arrival from the event
Data Source – IRIS miniSEED archive or IRIS PH5 archive, and USGS event service

SEED Channel Types – ?H?, ?L?, ?N?, ?G?, ?P? | High Gain, Low Gain, Accelerometer, Gravimeter, Geophone | excluding very long period channels

• Request a list of events having a magnitude >= 5.5.
• For each event,
  • Request a list of N.S.L.Cs that have data starting 2 minutes before through 28 minutes after the event origin time.
  • For each N.S.L.C,
    • Request the predicted travel time of the first arriving P phase between event and station,
    • Request data starting 30 seconds before and ending 30 seconds after the predicted P arrival time,
    • Omit time series with gaps or incomplete data during this 1-minute window,
    • Calculate the RMS variance of the 30 second signal and noise windows:
      

          mean(x) as m(x) = ( sum[i=1..n](x[i]) ) / n
          RMSvariance = sqrt( ( sum[i=1..n]( (x[i]-m(x)) ^2) ) / n )
      

    • Return the ratio of signal RMS variance to noise RMS variance:
      

          SNR = RMSvariance(signal) / RMSvariance(noise)
      

value – Signal to Noise Ratio (unitless) of the first 30 seconds of an event
target – the trace analyzed, labeled as N.S.L.C.Q (Network.Station.Location.Channel.Quality)
start – the integer second (UTC), rounded down, prior to the time series window described in Data Analyzed
end – the integer second (UTC), rounded up, following the time series described in Data Analyzed
lddate – date/time the measurement was made and loaded into the MUSTANG database (UTC)

Travel times provided to this metric by web services were generated using the iasp91 (1-D) earth model. Relying on a 1-D earth model to predict travel times can result in errors when the model departs from the true 3-D earth structure. While errors in the start times of the windows will affect SNR results to some degree, these errors are expected to be small relative to window size.