Hydraulic Structures: Fourth Edition

WORKED EXAMPLE 687

struct the model, and a valid simulation of zoning in the prototype dam is
impossible. Even for simple homogeneous model embankments on a
homogeneous foundation difficulties arise in attempting to translate model
data to prototype scale as most problems are stress-path dependent, i.e.
they relate to short- and longer-term loading histories. The problems are
also stress-level dependent, i.e. they are a function of self-weight for the
embankment and foundation complex. Bassett (1981) and Ko (1988)
provide a comprehensive introduction to contemporary views on the use
and limitations of physical models in geotechnical design and a further
perspective is provided in Taylor (1994). The utility of such models is
effectively restricted to study of deformation modes and failure mechan-
isms for simplified profiles.

16.2.4 Modelling of seismic response

Studies of the seismic response of models of concrete gravity dams using
the ‘shaking table’ technique have been reported by Mir and Taylor
(1994). The requirement for specialist facilities and the problems with
valid physical simulation of the prototype dam constrain any application of
this technique outside the research laboratory. Comparable and valid
model studies of the behaviour of embankment dams are generally con-
sidered not to be possible, but physical model tests conducted for compari-
son with numerical analyses have been reported by Finn (1990).
The application of centrifugal modelling techniques to specific
aspects of seismic stability for embankments is discussed in Pilgrim and
Zeng (1994) and in Dewoolkar, Ko and Pak (1999).

Worked Example 16.1

A river, transporting sediment, flows into a tidal estuary. The maximum
freshwater flow into the estuary is 4000 m^3 s^1. It is required to make a pre-
liminary design of a scale model in a laboratory where the space available
dictates the horizontal scale, Ml250; the pumping capacity available for
the model is 27 l s^1 and it is desirable to use it reasonably fully to avoid
viscous effects on the model. Establish

1. a suitable vertical scale for the model,


2. the discharge rate,


3. the tidal period scale,


4. the scale of the fall velocity of suspended sediment,


5. the probable scale of the bed material, and


6. the scale for the density of the bed load.