Underground Construction - Hydrogeology: Objectives, Methods, Applications - page 272

increasing degradation and persistent indicators that water pressure
was rising behind the roof required multiple phases of strengthening
construction, in 1856-57, 1873-74, 1882, and 1945 (re-coating the
walls, drainage, injections). On February 13th, 1960, a sudden cave-in
obstructed the tunnel over a length of 25 meters, collapsing the roof
over a length of 12 meters and a width of 6 meters. A sinkhole also
appeared at the surface, 23 meters above the roof of the tunnel, 48
hours after the accident. The disaster was attributed to the gradual
formation of a rising bell-shaped cavity above the roof of the tunnel due
to the slow washing away of fi ne clay particles by water circulations.
The tunnel was put back into circulation on June 20th, 1960, after the
caved-in mass was consolidated through injection and roof and walls
were rebuilt (Remondet & Valentin, 1961).
The prediction of water infl uxes during an underground project is not
always easy, and sizeable uncertainties can persist after the completion
of preliminary studies. In diffi cult cases, small-diameter reconnaissance
galleries are sometimes excavated in order to verify the nature and geometry
of the terrain, as well as the hydraulic characteristics of the rock mass. The
piercing of a water-bearing formation can, in addition, pre-empty the aquifer
and reduce hydrostatic pressure before the excavation of the large-diameter
passage. It is highly recommended, in all cases, but particularly when there
is a high probability of encountering water-bearing units, that exploratory
boreholes be drilled over a representative distance ahead of the excavations
in progress.
The digging of a tunnel in a moderately permeable terrain often allows
gradual drawdown of the aquifer in order to gravitationally evacuate
the collected outfl ow, if the tunnel is rising slightly. If the tunnel slopes
downwards, appropriate pumping mechanisms must be put in place in
order to evacuate the water collected by the excavation and to prevent its
immersion.
In highly permeable formations, it may be necessary to induce
drawdown in the zone being infl uenced by the tunnel, with the help of
pumping in wells dug from the surface (if the depth of the tunnel allows
it) or from drains drilled as the work progresses in front of and around
the tunnel.
The catchment and evacuation of water can sometimes prove diffi cult,
particularly during the crossing of perennial or temporary karst passages,
as is illustrated by the case of the Pouzergues railway tunnel (Brive to
Montauban line), built in 1878 to 1882 over a length of 880 meters at a
maximum depth of 40 meters. This tunnel passes through a limestone
promontory over a marly basement. During construction, it cut into
temporarily active karst, with a peak discharge of 300 L·s -1 , resulting in the