Geoscience Reference
In-Depth Information
at lower frequencies when the thicknesses involved are large. Site effects can be due to
the heterogeneity of the subsoil materials (impedance contrasts) or to irregular geometry
(amplification due to topography). Naturally, real cases involve a combination of these
two. The major effect on ground motion due to local geology is the amplification of
ground motion because of impedance contrasts underground. For example, in the case
of Mexico City, this amplification attains a factor of 40 (in average, individual obser-
vations reach a factor of 80) at the resonant frequency on the soft clays that cover the
lake bed zone of this valley. This amplification is in stark contrast with that due exclu-
sively to geometry of an otherwise homogeneous medium, i.e., the topography effect,
which does not exceed a factor of 2 even in the case of 3D topographies (e.g., Bouchon
etal.,1996).Giventhatitistheimpedancecontrastthatcausesthelargestamplifications,
it is the geometrical irregularities of the interfaces between sediments and rock that are
most relevant. In addition, soft soils are easily eroded and the free surface usually has
not a pronounced topography. The recent years have seen many publications where the
interaction between soft soils and an irregular subsoil structure combine to enhance the
observed local amplification. Thus, we have observed a change from the 1D paradigm
to the 2D and 3D models that commonly appear in the literature these days. With the
increase in the number of dimensions (which entails a parallel increase in the number
of details required to build the model), the loss of generality naturally follows. As the
complexity of the models increases, the possibility of extrapolating the results of par-
ticular case studies to a general application decreases significantly. This is reflected in
the current difficulties we experience to translate the knowledge we have gained on site
effects into expressions that could be generally applicable in a building code, for exam-
ple.Forthisreason,considerationofsiteeffectsinbuildingcodes isstillgenerallybased
on the1D paradigm.
This paper presents a compilation of results that reflects the current approaches to site
effects, without any attempt to present a complete review of the field. A cursory search
ininternetusinggooglefindsalmosthalfmillionreferencestositeeffectsinseismology,
evidence of the health of the research on this subject and of its importance in practical
studies. The following section deals with the estimation of site effects, i.e., their direct
measurement.Thedifferenttechniquescurrentlyusedarementioned,andtheadvantages
and disadvantages discussed. After that, the use of modelling is tackled. It is suggested
that, currently, it is not so much the modelling method as the building of a model and
its validation that are the hardest problems. A brief review of the different ways used to
estimate mechanical properties of the subsoil and its geometry is presented. Then, the
different approaches that have been proposed to include consideration of site effects in
the reduction of seismic risk are presented. Their use depends very much in the case
to case needs; either a building code, a microzonation map, or a reliable estimate for a
significant structureonitsown. Finally, some concluding remarks areoffered.
Throughoutwewillnotdealwithnon-linearbehaviour.Non-linearityis,ofcourse,avery
importantsubject,butonethatmaybeconsideredasasecondstageinsiteeffectstudies.
Indeed,areliableestimateoftheeffectsofnon-linearityongroundmotionrequiresavery
dependableestimateoflinearsiteeffectsatthesiteofinterest.Ontopofthis,thenumber
