Geology Reference
In-Depth Information
reports) that all parties buy into for contractual purposes before the
works actually begin. For larger tunnelling contracts in the UK, and
increasingly elsewhere, it is now mandatory that the hazards and risks
are assessed and managed in a consistent manner (British Tunnelling
Society, 2003). This is also the general case for some standard contracts
(FIDIC). This was introduced largely because insurance companies were
receiving an increasing number of claims due to tunnelling projects
going seriously wrong and were threatening simply to withhold insur-
ance on
such risky, poorly investigated, poorly thought-through and
mismanaged projects
'
(Muir Wood, 2000).
Unfortunately, in practice it is often not that simple to de
'
ne engi-
neering geological conditions in a distinct and unambiguous manner.
If one tries to be very speci
c (say on the rock type to be encountered)
then it would be relatively easy for the contractor to employ a specialist
at a later stage to dispute the rock description in detail and then to
allege that the slight difference in rock type caused all the dif
culties
that followed (excess wear, higher clay content etc., etc., plus delays and
general loss of productivity). Drafters of reference conditions some-
times resort instead to broad characterisation, perhaps using rock mass
classi
cations such as Qor RockMass Rating (RMR), as introduced in
Chapters 4
and
5
and Appendix C. The problem there is that such
classi
cations are made up of a range of parameters such as strength
and fracture spacing, each of which can be disputed because geology is
never that simple (or uniform). Furthermore, experienced persons
can often draw very different conclusions from the same data set.
Fookes (1997) reports an exercise where he asked two engineering
geologists familiar with rock mass classi
cations to interpret the same
sets of boreholes and exposures for a particular tunnel in terms of
RMR and Q value. One came up with an RMR
11 (extremely
poor rock and danger of immediate collapse); the other RMR
=
62
(fair rock and that no support is required). The Q value interpretations
were similarly quite different (extremely poor vs. fair rock). In this
particular case, the rock contained incipient cleavage (slate) and the
different opinions on classi
=
cations mostly hinged upon whether that
cleavage was considered a joint set or not
the standards and guidance
documents do not help very much in this regard, as discussed in
Chapters 3
and
4.
The main point is that despite reference conditions
being set out with good intentions of helping the contractor to price the
job and avoiding dispute, there is no guarantee that this will be
achieved.
It is the normal case that the extent of geological/geotechnical units
and position and nature of faults, for example, are uncertain. The
geotechnical baseline report should present the best interpretation of
the ground conditions by the designers and state any limitations and
reservations. In doing so, the rationale should not be, somehow, to
outwit the contractor contractually, but to allow the contractor to
-
