Electronic Supplement to
Model Validations and Comparisons of the Next Generation Attenuation of Ground Motions (NGA-West) Project

by James Kaklamanos and Laurie G. Baise

 

Additional Information on Testing Datasets

The electronic supplement to this article contains four supplemental tables (available as HTML tables and as .csv files) that contain information about the earthquakes and ground motion data used in our analyses. Table S1 (a "mini-flatfile") contains the data for the 1060 ground motion records used in our blind comparison tests, using data from seven recent earthquakes recorded in California. Tables S2 to S4 contain details on the subsets used to test the ground motion prediction equations, including the number of ground motion records in each category from each earthquake. Table S2 describes the mainshocks from the NGA flatfile, Table S3 describes the aftershocks from the NGA flatfile, and Table S4 describes the seven earthquakes used in the blind comparisons.

---

Table Descriptions

Table S1. Mini-flatfile of the 1060 ground motion records used in the blind comparison tests. As explained in the article, the data are from seven recent California-recorded earthquakes not present in any of the databases used to develop the NGA models: the (1) 2003 M 6.5 San Simeon, (2) 2004 M 6.0 Parkfield, (3) 2005 M 5.2 Anza, (4) 2007 M 5.4 Alum Rock, (5) 2008 M 5.4 Chino Hills, (6) 2010 M 7.2 Baja, and (7) 2010 M 5.7 Ocotillo earthquakes. The column headers of this table are designed to be similar to those in the NGA flatfile (Pacific Earthquake Engineering Research Center, http://peer.berkeley.edu/ngawest/index.html).

• The station network codes correspond to the following seismic networks: BK (University of California Berkeley Digital Seismic Network, http://www.ncedc.org/bdsn/); CE (California Geological Survey Strong Motion Instrumentation Program, http://www.conservation.ca.gov/cgs/smip/Pages/Index.aspx); CI (Southern California Seismic Network, http://www.scsn.org/index.html); NC (Northern California Seismic Network, http://www.ncedc.org/ncsn/); NP (National Strong-Motion Project, http://nsmp.wr.usgs.gov/); and WR (California Department of Water Resources, http://wwwdwr.water.ca.gov/).

• Finite rupture models were used for three of the larger events: Parkfield (Dreger, 2004), San Simeon (Rolandone et al.,, 2004), and Baja (Crowell and Bock, 2010). For these earthquakes, data have been populated in the columns describing the rupture plane geometry and the different source-to-site distance measures. Finite rupture models were not available for the four smaller events, but the strike, dip, and rake were determined by selecting the likely nodal plane from moment tensor solutions from the Southern California Seismic Network (SCSN, http://www.scsn.org), Berkeley Seismological Laboratory (BSL, http://seismo.berkeley.edu/), and Southern California Earthquake Center (SCEC, http://www.data.scec.org/), as described in the Data and Resources section of the article. Epicentral and hypocentral distances are listed for ground motion records for earthquakes without finite fault models, but the other distance measures are left blank.

• The geological units follow the convention of the NGA flatfile (http://peer.berkeley.edu/ngawest/index.html), as described by Table 2 of Chiou et al. (2008), which is based upon Table 1 of Wills and Clahan (2006). When Vs30 is inferred from surficial geology, the numerical values are determined from Chiou et al. (2008). If possible, geology and Vs data were obtained for stations in the NGA flatfile that recorded other earthquakes. Measured Vs data were obtained for some stations in the Parkfield vicinity from Thompson et al. (2010) and Real (1988). Data on surficial geology were obtained by the 1:250,000 scale regional geologic maps published by the California Geological Survey (Jennings, 1958, 1959, 1962, 1967; Jennings and Strand, 1958, 1959; Matthews and Burnett, 1965; Rogers, 1965, 1967; Smith, 1964; Strand, 1962; and Wagner et al., 1991), as displayed by Shakal et al. (2005) for stations in the Parkfield vicinity.

• Ground motions were obtained from the databases of (1) the Center for Engineering Strong Motion Data (CESMD, http://strongmotioncenter.org/), (2) the National Strong-Motion Project (NSMP, http://nsmp.wr.usgs.gov/), and (3) the Consortium of Organizations for Strong-Motion Observation Systems (COSMOS, http://db.cosmos-eq.org). The simple geometric mean horizontal component of ground motion is listed for PGA and Sa at spectral periods of 0.1, 0.2, 0.3, 0.5, 1.0, and 2.0 seconds.

Table S2. Information on the mainshocks in the NGA flatfile used in our testing subsets. For each earthquake, we list the numbers of records in the following subsets: (a) subdivision by site class: soil sites (180 < Vs30 < 450 m/s) and rock sites (450 < Vs30 < 1300 m/s); (b) subdivision by distance: small distance (Rrup < 10 km), medium distance (10 < Rrup < 100 km), and large distance (100 < Rrup < 200 km); (c) subdivision by magnitude: small magnitude (5 < M < 6.5) and large magnitude (6.5 < M < 8); (d) sites with measured Vs profiles versus sites with inferred Vs profiles; and (e) a special subset consisting of rock sites with small-to-medium distances in order to test the ten models simultaneously.

Table S3. Information on the aftershocks in the NGA flatfile used in our testing subsets. For each earthquake, we list the numbers of records included on soil sites (180 < Vs30 < 450 m/s) and rock sites (450 < Vs30 < 1300 m/s).

Table S4. Information on the seven earthquakes used in our blind comparison tests. For each earthquake, we list the numbers of records in the following subsets: (a) subdivision by site class: soil sites (180 < Vs30 < 450 m/s) and rock sites (450 < Vs30 < 1300 m/s); (b) subdivision by distance: small distance (Rrup < 10 km), medium distance (10 < Rrup < 100 km), and large distance (100 < Rrup < 200 km); and (c) a special subset consisting of rock sites with small-to-medium distances in order to test the ten models simultaneously. See Table S1 for the ground motion database.

---

References

Chiou, B. S.-J., R. Darragh, N. Gregor, and W. Silva (2008). NGA project strong-motion database, Earthq. Spectra 24, 23-44.

Crowell, B., and Y. Bock (2010). Fault slip model for the Mayor-Cucapah earthquake, Scripps Institution of Oceanography, University of California, San Diego, available at http://geoapp03.ucsd.edu/gridsphere/gridsphere?cid=Sierra+El+Mayor+Earthquake (last accessed Dec. 2010).

Dreger, D. (2004). Preliminary slip model, Berkeley Seismological Laboratory, University of California, Berkeley, available at Finite-Source Rupture Model Database, http://www.seismo.ethz.ch/srcmod/ (last accessed Dec. 2010).

Jennings, C. W. (1958). Geologic map of California: San Luis Obispo sheet, Calif. Div. Mines Geol., scale 1:250,000.

Jennings, C. W. (1959). Geologic map of California: Santa Maria sheet, Calif. Div. Mines Geol., scale 1:250,000.

Jennings, C. W. (1962). Geologic map of California: Long Beach sheet, Calif. Div. Mines Geol., scale 1:250,000.

Jennings, C. W. (1967). Geologic map of California: Salton Sea sheet, Calif. Div. Mines Geol., scale 1:250,000.

Jennings, C. W., and R. G. Strand (1958). Geologic map of California: Santa Cruz sheet, Calif. Div. Mines Geol., scale 1:250,000.

Jennings, C. W., and R. G. Strand (1969). Geologic map of California: Los Angeles sheet, Calif. Div. Mines Geol., scale 1:250,000.

Matthews, R. A., and J. L. Burnett (1965). Geologic map of California: Fresno sheet, Calif. Div. Mines Geol., scale 1:250,000.

Real, C. R. (1988). Turkey Flat, USA, site effects test area, Technical Report No. 88-2, California Department of Conservation, Division of Mines and Geology, Earthquake Shaking Assessment Unit.

Rogers, T. H. (1965). Geologic map of California, Santa Ana sheet, Calif. Div. Mines Geol., scale 1:250,000.

Rogers, T. H. (1967). Geologic map of California: San Bernardino sheet, Calif. Div. Mines Geol., scale 1:250,000.

Rolandone, F., R. Burgmann, D. Dreger, and M. Murray (2004). Coseismic slip distribution of the 22 December 2003 San Simeon earthquake, Annual Report 2003-2004, Berkeley Seismological Laboratory, University of California, Berkeley, http://seismo.berkeley.edu/annual_report/ar03_04/node19.html (last accessed Dec. 2010).

Shakal, A., V. Graizer, M. Huang, R. Borcherdt, H. Haddadi, K. Lin, C. Stephens, and P. Roffers (2005). Preliminary analysis of strong-motion recordings from the 28 September 2004 Parkfield, California earthquake, Seismol. Res. Lett. 76, 27-39.

Strand, R. G. (1962). Geologic map of California: San Diego-El Centro sheet, Calif. Div. Mines Geol., scale 1:250,000.

Thompson, E. M., R. E. Kayen, B. Carkin, and H. Tanaka (2010). Surface wave site characterization at 52 strong motion recording stations affected by the Parkfield M6.0 earthquake of 28 September 2004, U.S. Geol. Surv. Open-File Rept. 2010-1168, 117 p.

Wagner, D. L., E. J., Bortugno, and R. D. McJunkin (1991). Geologic map of the San Francisco-San Jose Quadrangle, Calif. Div. Mines Geol., Regional Geologic Map 5A, scale 1:250,000.

Wills, C. J., and K. B. Clahan (2006). Developing a map of geologically defined site-condition categories for California, Bull. Seismol. Soc. Am. 96, 1483-1501.