Geology Reference
In-Depth Information
correctly identi
whereby performance is
predicted in response to changes brought about by the engineering
works or in the long term from environmental impacts, including
rainfall, rises in groundwater pressures and vibration shocks. There
may be a long sequence of events contributing to the outcome, and
the geotechnical team need to predict this sequence or to recognise the
sequence when carrying out forensic studies of failures. The need for a
dynamic approach rather than just static is illustrated by the concept of
bore pile design. A pile can be designed to carry load through skin
friction, as well as through end bearing, but the designer needs to take
account of how the load is taken up sequentially. In reality, depending
on the geological conditions and geometry of piles, the skin friction in
the upper part of a long pile will come into play in carrying the load
from the structure, long before any load will reach the toe of the pile;
appreciation of this process will help the designer produce a cost-
effective solution.
The design model might be used as the basis for predictive numerical
modelling, but quite often a conceptual model can be used in its own
right to identify the hazards and the best way to proceed at a site. When
failures occur, it is often the conceptual engineering geological model
that helps to explain what has happened. The parameters are very
much secondary
ed
-
a mechanism model
-
it is rare to be able to be very certain regarding
parameters such as strength and water pressures and often a range of
possible solutions will
-
fit the facts equally well (Lerouiel & Tavernas,
1981). An example of an engineering geological model used to explain
a major landslide is given in
Chapter 7
(the Pos Selim landslide,
Malaysia) and discussed in Malone et al. (2008). Experience tells us
when a model makes sense in terms of likely strengths and mechan-
isms. A complex 3D modelling exercise would often only serve to
con
rm what we can already tell by judgement. Examples of concep-
tual models used to explain the observed facts from two landslides are
Facts to be taken into account
-
The failure occurred in August 1982, several days after heavy rain (Choot, 1983). It constituted
general distress in an elliptically shaped area in the centre of a cut slope, with kicking out at the toe by
about 750mm
(Figures B3-4.1 a
nd
B3-4.2).
-
Nomovement had been detected previously, despite evenmore severe rainfall inMay 1982. However,
one piezometer in the centre of the distressed area had begun to show positive pore pressures since the
May rainstorm.
-
The failure was in severely weathered granite with daylighting (partly clay-in
lled
discontinuities).
