Electronic Supplement to
Three Ingredients for Improved Global Aftershock Forecasts: Tectonic Region, Time-Dependent Catalog Incompleteness, and Intersequence Variability

by Morgan T. Page, Nicholas van der Elst, Jeanne Hardebeck, Karen Felzer, and Andrew J. Michael

This electronic supplement contains figures that show 95% confidence limits for Omori parameters within tectonic regions (Fig. S1) and alternative Omori fits that take into account b-value variation (Fig. S2), combine the smaller tectonic regions into larger regions (Fig. S3), and extend the fitting window to 100 days postmainshock (Fig. S4). We also provide results of a synthetic test to demonstrate the performance of the inversion for the intersequence a-value variability (Fig. S5).

Figures

Figure S1. Omori parameter 95% confidence limits for sequences within the García et al. (2012) tectonic regions (Fig. 3 of the main article). Maximum-likelihood solutions (see Fig. 5 of the main article for fits) are shown by the plus symbols.

Figure S2. Stacked Omori fits for aftershock sequences within the García et al. (2012) tectonic regions (Fig. 3 of the main article) in the format of Figure 5 of the main article, using maximum-likelihood b-values fit to all earthquakes in the catalog (National Earthquake Information Center [NEIC], 1990–2015; see Data and Resources) within each region. Insets: magnitude–frequency distributions for earthquakes in each region, with maximum-likelihood b-value fits and b = 1, for comparison.

Figure S3. Stacked Omori fits for aftershock sequences within the all active nonsubduction region (ANSR) and subduction zone (SZ) García et al. (2012) tectonic regions (Fig. 3 of the main article) in the format of Figure 5 of the main article.

Figure S4. 100-day fits for aftershock sequences within the all García et al. (2012) tectonic regions (Fig. 3 of the main article) in the format of Figure 5 of the main article. Light blue dotted lines show 10-day fits (Fig. 5 of the main article) for comparison.

Figure S5. Synthetic test results. We generate 10-day synthetic aftershock sequences for all mainshocks in the global dataset, assuming the mean global p-value and the distribution of a-values shown. Recovered a-value distributions for different mainshock magnitude bins show that the inversion recovers a slightly wider a-value distribution with the correct mean.

Data and Resources

The National Earthquake Information Center (NEIC) catalog, also known as ComCat, is available at http://earthquake.usgs.gov/earthquakes/search/ (last accessed September 2015).

Reference

García, D., D. J. Wald, and M. G. Hearne (2012). A global earthquake discrimination scheme to optimize ground-motion prediction equation selection, Bull. Seismol. Soc. Am. 102, 185–203, doi: 10.1785/0120110124.

[ Back ]