Geoscience Reference
In-Depth Information
to the strong motion records obtained by the UPSAR dense array during the 2003 San
Simeon M6.5 and the 2004 Parkfield M6.0 earthquakes (Fletcher et al., 2006; Wang et
al., 2006). Ground conditions are quite different for the Parkway Valley and UPSAR
arrays. Referring to Ch´avez-Garc´ıa et al. (1999) and to Wang et al. (2006), respectively,
for more details, it is sufficient to recall here that Parkway Valley is a small basin, 30
to 40m deep, filled with layers of mixed mud, gravels, and sands with
V
s
varying from
∼
s, overlying a stiff conglomerate formation, while the UPSAR array is
locatedonstiffersoils,possiblycharacterizedbymoderatetopographyeffects(34mele-
vation difference between the lowest and the highest station; geological structures fairly
uniform and horizontally layered;
V
s
∼
150 to
∼
350m
/
/
s at surface).
A summary of theearthquake records considered isprovided in Table 18.1.
400m
For the UPSAR array, the spacing between adjacent stations is less favorable than for
Parkway Valley, as it ranges from few tens of meters up to few hundreds of meters.
However, considering the rule of thumb of a 90% accuracy in the evaluation of strains
if the separation distance
among adjacent stations is less than 1/4 of the dominant
wavelength(Bodinetal.,1997),anapparentwavepropagationvelocityatthesite
V
app
=
2
∆
250m, it can be deduced
thatthedisplacementwavefieldcanbereconstructedwithasufficientdegreeofaccuracy
up to
f
.
5km
/
s (Fletcher et al., 2006) and a representative value
∆ =
5Hz. A more accurate determination, not reported here for
brevity, suggests that the average error in terms of
PGV
predicted by the interpolation
procedure ranges between 18% and 34% for both arrays. A similar level of error can be
indirectly inferredalsofor
PGS
,due toitsstrong correlation with
PGV
.
=
V
app
/
4
∆ =
2
.
The spatial interpolation of the displacement wavefield at ground surface allows one to
obtain some interestingfeatures of the strainfield, such as:
- the principal strains and principal directions and their possible correlation with the
prevailing wave propagation direction;
- the variation with azimuth of the strain tensor components and of their correlation
with the representative parameters of ground motion severity, such as
PGV
.
Table 18.1. List of earthquakes considered for experimental evaluation of ground strains. Peak
values are calculated as the maximum absolute value of the horizontal components, among all
stations
Earthquake
M
w
Epicentral
No. of
PGA
PGV
PGD
(
cm
/
s
2
distance
available
)
(cm/s)
(cm)
(km)
records
Parkway 1
4.9
81
17
10.7
0.53
0.037
Parkway 2
4.2
81
17
0.8
0.04
0.003
Parkfield
6.0
11.6
11
487.1
25.40
4.27
San Simeon
6.5
55
11
238.0
16.55
5.71
