This supplement contains supplemental tables and figures and the gridded, long-term earthquake rate model for the central and eastern United States (CEUS).
Supplemental tables describe the smoothing parameters that optimize information gain from the adaptive and fixed smoothed-seismicity models.
Supplemental figures highlight the high-seismicity-rate regions of eastern North America (ENA) addressed in this paper and compare the final earthquake rate model to one derived from fixed smoothed-seismicity models.
Table S1. Optimized neighbor numbers from adaptive smoothed-seismicity models. Neighbor numbers are presented for all forecast and testing catalog combinations. For cases in which the intermediate neighbor distances optimized information gain, the pair of neighbor numbers used in computing the geometric mean are presented (e.g., 4/5).
Table S2. Optimized smoothing distances from fixed smoothed-seismicity models. Smoothing distances presented for all forecast and testing catalog combinations.
Figure S1. Example of the method for determining minimum magnitude of completeness (Mc) for earthquakes in the completeness zone ZC4 1980–2012. The figure presents the cumulative and incremental counts of earthquakes (left vertical axis) as a function of magnitude. The goodness-of-fit (GFT) parameter of Wiemer and Wyss (2000) is plotted on the right vertical axis.
Figure S2. Estimation of b-value for all earthquakes from the CEUS catalog for 1980–2012 follows from the maximum likelihood method (Aki, 1965; Utsu, 1965). Predicted earthquake rates use b=1.0. The predicted and observed rate of earthquakes for 3≤M≤5.5 is plotted.
Figure S3. Earthquake rate model for the CEUS from an adaptive smoothed seismicity model for the 1993 completeness level with regional minimum magnitudes of completeness listed in Table 2 of the main article, annotated with the place names of high-seismicity regions in ENA. This figure is a modified version of Figure 4 of the main article. Annual rates are plotted in a log10 scale and represent the number of M≥5.0 earthquakes. The dashed line between western North America and central and eastern North America delineates the western boundary of the CEUS for the purposes of this study. High seismicity rates (>0.002 yr-1, M≥5.0) are found for the following regions: (1) eastern part of the Intermountain seismic belt (EISB); (2) the New Madrid seismic zone (NMSZ); (3) the eastern Tennessee seismic zone (ETSZ); (4) the western Quebec seismic zone (WQSZ); (5) central Oklahoma (COK); (6) seismicity near Charleston, South Carolina (CHAR); (7) the Charlevoix seismic zone (CSZ); (8) the Norumbega fault zone (NFZ); (9) north Texas (NTX); (10) the Lancaster seismic zone (LSZ); (11) the central Virginia seismic zone (CVSZ); and (12) northeastern Ohio (NOH).
Figure S4. (a) Forecast log10 annual rates for M≥5.0 earthquakes from fixed smoothed seismicity from 50 km smoothing distance using the earthquake catalog for 1993–2012. (b) Comparison of forecast earthquake rates presented as the difference in M≥5.0 annual rates from adaptive and fixed smoothed-seismicity models.
Download: Model_S1.txt [Space-delimited Plain Text Table; ~3.8 MB]. Gridded, long-term earthquake rate model for the central and eastern United States.
Aki, K. (1965). Maximum likelihood estimate of b in the formula log N= a−bM and its confidence limits, Bull. Earthq. Res. Inst. 43, 237–239.
Utsu, T. (1965). A method for determining the value of b in a formula log n=a-bM showing the magnitude-frequency relation for earthquakes. Geophys. Bull. Hokkaido Univ. 13, no. 99, 103.
Wiemer, S., and M. Wyss (2000). Minimum magnitude of completeness in earthquake catalogs: Examples from Alaska, the western United States, and Japan, Bull. Seismol. Soc. Am. 90, no. 4, 859–869.
[ Back ]
