Effect of soil–structure interaction and spatial variability of ground motion on seismic risk assessment of bridges - Seismic Risk Analysis and Management of Civil Infrastructure Systems - page 608

Albansesi et al. (2007) have suggested an alternative approach based on the
use of a 'relative displacement spectrum'.
Eurocode 8 on the other hand, in the latest version of its Part 2 referring
to Bridges (CEN, 2005), also prescribes increased levels of minimum seating
(or 'overlap') lengths for the abutments, to ensure the function of the
support under extreme relative displacements that may result due to spatial
variability of seismic motion. It is also noted that the relevant seating
lengths are estimated with an additional safety margin equal to 2 com-
pared with the safety factor imposed for the stresses induced at the bridge
system. The most important development in Eurocode 8 is that it is cur-
rently the only seismic code worldwide that provides a clear and detailed
framework for considering the effect of spatial variability of ground motion
in bridge design, prescribing both a simplifi ed and an analytical approach,
the latter being included as an 'informative' annex.
In particular, Eurocode 8 clearly recognizes that since spatial variability
of seismic action is a situation wherein the ground motion at different sup-
ports of the bridge differs, the seismic action cannot be based on the char-
acterization of motion at a single point. According to EC8, spatial variability
has to be considered for bridges of continuous deck when one or both of
the following conditions apply: (a) soil properties along the bridge vary in
such a way that the soil at the various supports corresponds to more than
one category (as specifi ed in Eurocode 8 - Part 1), and (b) soil properties
along the bridge are approximately uniform, but the length of the continu-
ous deck exceeds a specifi ed limit, L lim ; the recommended value of L lim is
equal to L g /1.5, where the length L g is the distance beyond which motions
can be considered as completely uncorrelated and is given in Table 22.1 as
a function of the ground conditions (to visualize, see Sextos and Kappos,
2009).
For the general case, the potential maximum values of the considered
seismic action effect (i.e. member force or deformation) can be estimated
through an adequate (albeit simplifi ed) procedure. This method should be
followed, unless a more accurate analysis is carried out. To this end, a more
detailed procedure for the assessment of the asynchronous motion effects
in the frequency domain (Der Kiureghian et al. , 1997) is proposed in an

Table 22.1 Limiting length to consider spatial variability effects, as a
function of ground type according to Eurocode 8

Ground type A B C D E

L g (m) 600 500 400 300 500

L lim (m) 400 333 266 200 333