Geology Reference
In-Depth Information
Likely mechanism
1.
After cutting of slope for road construction, the slope was coated with concrete. The concrete
dammed the truncated river valley seen in
Figure B3-4.4.
2.
Probably minor movements had been occurring intermittently for many years, as evidenced by
infilled and slickensided joints.
3.
The May 1982 rainstorm probably resulted in high dammed water levels in the old river
valley, feeding water laterally into the eventual failure area. This led to dilation of the rock
mass and increased permeability, so that a piezometer that had been dry became responsive
after May 1982.
4.
The August 1982 storm led to a rise in water in the dammed river valley. It took several days for the
critical pressures to develop.
5.
Water moved laterally into the adjacent rock mass. Water pressures reduced effective stress and
initiated failure.
6.
The slope reached a metastable condition as the rock mass dilated and water pressures were
dissipated.
Coda
The distressed area was cleaned off and the slope further investigated. Despite a lack of any
'
useful
piezometer information
, the engineers responsible for the slope decided to install more than 100
deep and expensive caissons and interconnecting drains in three lines, right across the slope, rather
than concentrating on the most likely sources of shallow in
'
ltration and through
ow, such as the
old river valley. In August 1984, during re-pro
ling, major wedge failures occurred by sliding on
clay-
lled joints
(Figure B3-4.6)
. The remedial measure adopted (apart from the deep drainage system
already instigated) was to cut back the slope as far as feasible and to install more than 100 permanent
ground anchors, each stressed to more than 1,000kN through a grillage of ground beams (Buttling,
1986).
River valley colluvium nicely exposed
Wedge failure August 1984
Slopes re-graded after 1982 failure
