The SeisSound Visualization is an audio/video-based IRIS DMS data product that illustrates the frequency and amplitude content of seismograms. Conveying the seismograms frequency content both visually and audibly produces a better understanding of their spectral content.
This data product (for a sample click image on right or to visit SeisSound data product repository, click here ) conveys the data in four different ways:
- an image of the original seismogram
- an image of a filtered version of the seismogram
- an image of the spectrogram of the filtered trace
- an audio playback of the seismic signal that is time-compressed and amplitude-normalized so that the sound is audible for human ears (roughly 20 Hz – 20 kHz)
The temporal evolution of the seismogram is depicted by the seismogram turning from a light gray color to a dark black color as time evolves. A pink vertical line in each of the sub-plots indicates the current time of the playback, as does the time counter in the upper right. The SeisSound data products do not reproduce the seismograms in true time, as this would make the movies too long and boring to listen to because the sound would be out of our hearing range. Instead the playback is sped up by the factor indicated in the x-label. For example, a speed up factor of 500 will transform the frequencies 0.5 to 40 Hz in the seismogram to 250 (0.5*500) Hz to 20,000 (40*500) Hz range of our hearing. What’s your hearing range? Test it and find out here. Both the spectrogram and audio file are created after applying a 0.5 Hz high-pass filter. To hear the full range of sounds we recommend listening to these data products with earbuds or good speakers.
It is difficult to transform seismic data in a way that both the high-frequency and low-frequency signals become audible at the same time. This is because each would require a different speed-up factor to make the signals heard. To solve this problem we apply an appropriate speed-up factor to hear the higher-frequency (>0.5 Hz) energy and then use MATLAB’s voltage controlled oscillator (vco) function to create a sound signal that oscillates around an audible center frequency in proportion to the amplitude of the seismogram’s lower-frequency energy 1 . We make the length of this modulated signal identical to the time-compressed higher-frequency channel. This allows us to save these two sound outputs as a stereo sound pair 2 . The stereo sound from these dual frequency bands, in combination with the video images, allows the correlations between the lower-frequency surface-waves and high-frequency triggered earthquakes and tremor signals (if they exist) to be assessed.
The color of the spectrogram indicates amplitude, where warm colors are higher amplitudes and cool colors are lower amplitudes. These amplitudes represent the log base 10 of the normalized amplitude, multiplied by a scale factor of 10. All SeisSound data products have the same scaling so comparisons can be made between products.
Information about the earthquake is displayed in the first line of the title, and information about the seismic recording station, and its distance and azimuth to the earthquake, is shown in the second line of the title. Listed in the top y-axis label is the type of seismogram (i.e., velocity) and the seismogram’s channel code (i.e., BHZ). There is an option to show seismic wave phase arrival times, which are indicated with short blue vertical lines and associated blue text that displays the phases (i.e., P, S, SKS, R1).
Exporting the audio channel of the SeisSound product for an event to the corresponding ground motion visualization (GMV) product produces an interesting composite movie. Click on the image below to see and hear such a composite movie created for the Sea of Okhotsk M8.2 event of May 24, 2013. In this case the audio was played only 102 times faster than true speed in order to stay in sync with the video presentation.
The following table summarizes the parameters used to generate the SeisSound data product.
|Filter||0.5 Hz high-pass|
|VCO filter||0.5 Hz low-pass|
|Approximate number of Frames requested for the movie||100|
|Image/Movie height (pixels)||720|
|P slowness (s/degree)||7.2|
|Rayleigh slowness (s/degree)||28.5|
|List of requested seismic phases||P, Pn, Pg, Pdiff, pPdiff, sPdiff, PcP, PP, PP, PS, PKP, SKS, S, Sn, Sg, Sdiff, pSdiff, sSdiff, SP, ScS, SS, R1, R2|
|Number of seconds between two consecutive phases to avoid overprinting phase markers (second)||600|
(links to information about the earthquake)
(26 December 2004; Sumatra; magnitude 9)
 To do this we modulate the amplitude (volume) of the vco tone by the sum of two envelope functions. The first is the envelop of the full data trace and the second is the envelop of the data trace after a low-pass filter (<0.5 Hz) has been applied. We find that using the sum of these two envelopes we can highlight both the arrival of the largest amplitudes in the seismogram in addition to the arrival of higher frequency energy.
 This is accomplished using MATLAB’s wavwrite function.
To cite the IRIS DMC Data Products effort:
- Trabant, C., A. R. Hutko, M. Bahavar, R. Karstens, T. Ahern, and R. Aster (2012), Data Products at the IRIS DMC: Stepping Stones for Research and Other Applications, Seismological Research Letters, 83(5), 846–854, doi:10.1785/0220120032.
To cite the source of the SeisSound data product:
- Kilb, D., Z. Peng, D. Simpson, A. Michael, M. Fisher and D. Rohrlick, (2012), Listen, Watch, Learn: SeisSound Video products, Seis. Res. Lett., doi:10.1785/gssrl.83.2.281, 2012.
To cite the IRIS DMC SeisSound or reference use of its repository:
- IRIS DMC (2012), Data Services Products: SeisSound The Audio/Video Seismic Waveform Visualization, doi:10.17611/DP/SSND.1.
- (2011) Zhigang Peng, Associate Professor, School of Earth and Atmospheric Sciences, Georgia Institute of Technology: Original concept and codes.
- (2012) Debi Kilb, Project Scientist, Scripps Institution of Oceanography/University of California, San Diego: Adapted and refined the codes and visualizations for IRIS under sub-award number 86-DMS, which was awarded under cooperative agreement No. EAR-0733069 issued by the NSF under CFDA No. 47.050.
University of California, San Diego