Research ArticleEARTHQUAKES

The evolving interaction of low-frequency earthquakes during transient slip

See allHide authors and affiliations

Science Advances  22 Apr 2016:
Vol. 2, no. 4, e1501616
DOI: 10.1126/sciadv.1501616
  • Fig. 1 Evolution of LFE clustering in the Guerrero subduction zone.

    (A) Map view of the Guerrero subduction with all 1120 LFE epicenters shown as black points. The colored boxes represent the two LFE source regions: the transient zone in purple and the sweet spot in red (5, 8). (B) Vertical profile of the subduction beneath Guerrero, Mexico. The two labeled colored boxes represent the LFE source regions in (A). The plate boundary transitions from a mostly locked interface to a mostly sliding one between the two regions of slow slip, represented by the two gray patches (7, 18). (C) LFE clustering as a function of distance from the trench, as evidenced by the measured power law exponent α (see text). (D) Inter–slow-slip coupling along the Guerrero subduction zone (17).

  • Fig. 2 Time clustering of LFEs.

    (A) Recurrence intervals of the transient zone LFEs (see Fig. 1), defined as the elapsed time between sequential events. The two dashed boxes indicate the two time periods analyzed. (B) Cumulative event count of a single transient zone LFE source, whose stacked waveforms are shown in fig. S1. The zoomed inset highlights the distinctly different timing during the two time periods. (C) Autocorrelations of the event count signals (see text) for the events shown in (B). (D) Event count spectra of the two autocorrelograms in (C). The positive linear slope in log space indicates a power law distribution of event timing and corresponds to the power law exponent α.

  • Fig. 3 Tracking slow slip in time with the power law exponent α.

    The average running 10-day estimate of α is plotted for the all LFE sources in the transient zone (purple) and the sweet spot (red). Values of α increase for each of the geodetically detected slow-slip events (6, 18), represented by the gray patches in the background. The dashed border for each gray patch corresponds to the slow-slip regions in Fig. 1B.

  • Fig. 4 Interaction across LFE sources.

    (A and B) Each cell in (A) and (B) represents the spatially smoothed correlation between different LFE event counts (see text). The inter–slow-slip period is shown in (A), whereas the co–slow-slip is shown in (B). The sweet spot exhibits high levels of interaction during both time periods, whereas the transient zone is only correlated during the Mw 7.5 2006 slow-slip event (7).

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/2/4/e1501616/DC1

    fig. S1. Stacked waveforms of a transient zone LFE source during the inter– (black) and co–slow-slip (red) time periods.

    fig. S2. Three synthetic catalogs from our numerical model.

    fig. S3. Parametric estimation of the power law exponent α with and without slow slip.

    fig. S4. Stability of event count time series autocorrelation with respect to analyzed window duration.

    fig. S5. Stability of event count time series spectrum with respect to analyzed window duration.

    fig. S6. Stability of event count time series autocorrelation with respect to analyzed bin width.

    fig. S7. Stability of event count time series spectrum with respect to analyzed bin width.

    table S1. Numerical model parameters used in figs. S2 and S3.

  • Supplementary Materials

    This PDF file includes:

    • fig. S1. Stacked waveforms of a transient zone LFE source during the inter–(black) and co–slow-slip (red) time periods.
    • fig. S2. Three synthetic catalogs from our numerical model.
    • fig. S3. Parametric estimation of the power law exponent α with and without slow slip.
    • fig. S4. Stability of event count time series autocorrelation with respect to analyzed window duration.
    • fig. S5. Stability of event count time series spectrum with respect to analyzed window duration.
      fig. S6. Stability of event count time series autocorrelation with respect to analyzed bin width.
    • fig. S7. Stability of event count time series spectrum with respect to analyzed bin width.
    • table S1. Numerical model parameters used in figs. S2 and S3.

    Download PDF

    Files in this Data Supplement:

Navigate This Article