Civil Engineering Reference
In-Depth Information
the scaling procedure outlined by Baker (2011) is not expected to adversely
impact the resulting structural responses.
The procedure outlined by Baker (2011) is straight forward for sites
where only crustal earthquakes occur (e.g., California). A complicated
aspect in constructing a CMS for a site in south-western British Columbia,
in comparison with a site in California, is that three earthquake types,
having distinctly different characteristics, contribute to the overall seismic
hazard. Therefore, three CMS must be constructed for record selection
(Goda and Atkinson, 2011): 'CMS-Crustal', 'CMS-Interface' and 'CMS-
Inslab'. In order to account for these events, two sets of CMS can be con-
sidered as proposed by Goda and Atkinson (2011). The fi rst approach is the
CMS-Event-based procedure which is based on the weighted average of
the CMS that are computed by using applicable GMPEs and the corre-
sponding scenarios for the three earthquake types, in proportion to the
relative infl uences of the scenarios (i.e., the number of records from each
event type is proportional to the percentage of contribution of that event
type in seismic hazard). The second approach is the 'CMS-All-based' which
is the weighted average of 'CMS-Crustal', 'CMS-Interface' and 'CMS-
Inslab' by considering relative infl uences of the individual earthquake types
without considering different earthquake type records in the selection pro-
cedure. In fact 'CMS-All-based' is the simplifi ed version of 'CMS-Event-
based' in which different event types are considered in the construction of
the CMS, but in selecting the records the relative infl uences of the scenarios
are not considered. This results in variability of the seismic response based
on the 'CMS-All-based' approach being smaller than that based on the
'CMS-Event-based' approach. However, it must be noted that for each
earthquake type, the variability of the selected records in general tends to
be underestimated for CMS-based or UHS-based approaches, since a close
match to a target response spectrum is imposed (Baker, 2011). Nevertheless,
the 'CMS-Event-based' approach can account for the variability of the
CMS among different event types more reasonably than does the 'CMS-
All-based' approach. It has been shown that the median IDA results com-
puted using the 'CMS-Event-based' and 'CMS-All-based' are in good
agreement (Tehrani
et al.
, 2012; Goda and Atkinson, 2011).
Figure 21.3 shows the CMS at 2% probability of exceedance in 50 years
(i.e., return period of 2475 years), computed at
T
0.7 s for different event
types separately using the deaggregation results presented in Table 21.1.
The BA08 GMPE (Boore and Atkinson, 2008) and Z06 GMPE (Zhao
et al.
, 2006) were used for crustal events and subduction events, respectively.
As expected, all CMS are approximately equal to the spectral acceleration
values of the UHS at the target vibration period considered (i.e.,
T
=
0.7 s).
The CMS-Interface has rich spectral content in the long period range, while
the CMS-Inslab has rich spectral content in the short period range. The
=
Search WWH ::
Custom Search