Geology and ground models - Practical Engineering Geology - page 100

review is given in GEO (2007) and some principles are set out in Box 3-3.
Ground models in this context are essentially 3D models of the geolo-
gical conditions at a site, together with environmental in uences and

hazards. Numerical and physical models can be designed based on the
conceptual ground model and might be used in the development of
ground models, for example, in simulating the development of in situ
stresses.

Box 3-3 Principles of ground modelling

1. Ground models for a site should be based on adequate interpretations of geology and
hydrogeological conditions; adequate with respect to the engineering circumstances and
project requirements.
2. Models should be extensive enough to include all the ground that will be affected by the works.
For example, a building will stress the ground signi cantly to a depth of up to twice the breadth

of the foundation footprint. A new dam and reservoir may in

uence the terrain and environment
for a large area, many kilometres from the actual dam, through changes in ground water levels
and perhaps induced seismicity.
3. Engineering geological models should make sense in terms of the geological history at a site. This test
is sometimes failed in ground models produced by non-geologists.
4. Make sure that the models incorporate all the features of the ground important to physical
performance as conditions change (e.g. increased or decreased loading by the engineering works or
the application of fluid pressures).

5. The creation and testing of several simple models exploring the sensitivity of the site to various
assumptions will often be more revealing than a single complex model. This is certainly true when it
comes to using numerical and physical models.
6. Geotechnical engineers and engineering geologists should act as detectives in characterising a site,
studying the evidence, hypothesising and testing hypotheses through the collection of additional
data, including the output from numerical analyses. Several iterations may be necessary before the
models are adequate (Star

eld & Cundall, 1988).

A preliminary geological model based on desk study, together
with geomorphological interpretation, should be used for planning
ground investigation, which will allow the model to be checked and
re ned. Fookes (1997) suggests that a model can simply be a

written description or presented as cross sections or block diagrams
and plans. It might be focused on some aspect such as ground-
water, geomorphology or rock structure but should be targeted at
the engineering needs of the project. At a later stage, the geological
model may be split into units which can be characterised in terms
of engineering properties and anticipated performance. In some
locations and for particular projects, rocks that are quite different
in origin and age might be lumped together because they can be
expected to behave in a similar manner. Complex geology does not