Environmental Engineering Reference
In-Depth Information
16
Handling Uncertainty
in Coastal Modelling
DOMINIC E. REEVE, JOS ´ HORRILLO-CARABALLO
AND ADRI ´ N PEDROZO-ACU ˜ A
Introduction
For coastal applications, the main quantities
required for design are a description of the wave
conditions, the water level variations and the
movement of sediment. For waves, rather than
describing the sequence of individual waves it is
common practice to provide a statistical descrip-
tion of the conditions that define representative
wave heights, wave periods and wave directions.
For water levels, the situation is slightly different.
In many locations around the world the predom-
inant component of water level variations at the
coast is due to tides. Tides result from the gravi-
tational attraction between the Sun, Moon and
Earth, are well understood and are predictable to
a good level of accuracy. An additional, and less
predictable, component to the water level is
'surge'. This term describes changes in the water
level arising from meteorological effects. Sedi-
ment transport is not fully understood and is
currently the subject of intense research activity.
Some of the factors that make this such a difficult
area include: the range of sediment particle shapes
and sizes; the very different responses of cohesive
sediments, sands and gravels to the same wave
conditions; the three-way interaction between
the shape of the seabed, the incoming waves and
the prevailing water levels; the vast difference in
timescales between the period of an individual
waveandtheperiodoverwhichsignificantchanges
in beach morphology occur; and, similarly, the
large range of spatial scales over which sediment
transport and morphological changes occur.
Forecasting methods have evolved as our
understanding of
'Why should we bother to model coastal
processes?' is a question often asked by practi-
tioners. It is a natural question, which has a
straightforward answer: for the same reason that
we try to forecast future weather, stock market
movements, and the outcome of sporting events.
That is, to try to gainwarning of dangerous events,
to make a fortune or to gain an edge over the
competition. In coastal engineering, modelling
is used as part of the process of designing flood
defence or coastal protection works to withstand
the conditions that they are likely to experience
over the period for which they are intended to
operate. In this chapter a range of modelling tech-
niques used in coastal engineering is illustrated
via a set of case studies. Particular attention is
given to ambiguities accompanying the model-
ling, andmethods for quantifying this uncertainty.
Modelling is the process through which predic-
tions of the future are made. The modelling
process may not attempt to provide a detailed
forecast of actual future events, but might be
aimed at predicting the outcome under particular
'worst case' scenarios. Modelling can take many
forms, typically involving quantitative methods
such as theoretical analysis or computational
simulation.
the physical processes has
Search WWH ::




Custom Search