Muir Wood Lecture 2014

Some Critical Aspects

Of Subaqueous Tunnelling
Georgios Anagnostou - ETH Zurich, Switzerland

N ISBN : 978-2-9700858-6-7

20975-Muir Wood 2014-.indd 1

MAY 2014

19/06/14 10:22

Muir Wood Lecture 2014 - Some Critical Aspects Of SubaqueaousTunnelling - NISBN : 978-2-9700858-6-7 / MAY 2014
Layout : Longrine Avignon France www.longrine.fr
The International Tunnelling and Underground Space Association/Association Internationale des Tunnels et de lEspace Souterrain (ITA/AITES) publishes this report to, in accordance with its
statutes, facilitate the exchange of information, in order: to encourage planning of the subsurface for the benefit of the public, environment and sustainable development to promote advances
in planning, design, construction, maintenance and safety of tunnels and underground space, by bringing together information thereon and by studying questions related thereto. This report
has been prepared by professionals with expertise within the actual subjects. The opinions and statements are based on sources believed to be reliable and in good faith. However, ITA/AITES
accepts no responsibility or liability whatsoever with regard to the material published in this report. This material is: information of a general nature only which is not intended to address the specific
circumstances of any particular individual or entity; not necessarily comprehensive, complete, accurate or up to date; This material is not professional or legal advice (if you need specific advice,
you should always consult a suitably qualified professional).

20975-Muir Wood 2014-.indd 2

19/06/14 10:22

Muir Wood Lecture 2014

Some Critical Aspects

Of Subaqueous Tunnelling
Georgios Anagnostou - ETH Zurich, Switzerland

High potential for damage, relatively high pore pressures

and limited pre-construction accessibility are all features of
subaqueous tunnels. Potential hazards include high water
inflows or even a complete flooding of the tunnel in the case of a
connection opening up to the seabed. In subaqueous tunnels,
very high pore pressures may occur at small depths of cover,
i.e. often in combination with a low shear strength ground,
resulting in particularly adverse effects in terms of stability and
deformations of the opening. This lecture illustrates some of
the geomechanical issues relating to subaqueous tunnels (face
stability in fault zones, the limits of open mode TBM operation
in weak sedimentary rocks and the effect of advance drainage
in squeezing ground) with reference to five case studies the
Storeblt tunnel, the Melen 7 Bosphorus tunnel, the Lake
Mead Intake No 3 tunnel, the Zurich Cross Rail and the future
Gibraltar Strait tunnel project.

20975-Muir Wood 2014-.indd 3

19/06/14 10:22

1 >> Introduction

Limited pre-construction accessibility, high potential for damage and

relatively high pore pressures are all features of subaqueous tunnels.
In addition to the high cost of marine operations, the ship movements
or sea currents that often prevail in straits (Fig. 1) may cause frequent
interruptions to probing operations, thus making exploratory
campaigns very demanding. Systematic and time-consuming advance
probing during construction is often indispensable. The limited
accessibility not only increases uncertainty in the planning phase, but
also narrows the range of technical options for construction: offshore
ground improvement works, for example, share the same difficulties as
exploratory campaigns and are also extremely costly (Fig. 2).

Intermediate attacks (which might reduce the impact of geological

uncertainty and, in the case of long tunnels, also construction time) are
possible only in very exceptional cases and only for works of a limited
nature (Fig. 3).

Figure 1 : Sea currents in the Gibraltar strait and in the

Bosphorus strait

Figure 3 : Marmaray immersed tunnel project: An exceptional case

of a shaft through water [2]

Figure 2 : Storeblt tunnel. Large scale pore water pressure relief

below the seabed using pumping wells [1]

Figure 4 : Historic painting of the Thames tunnel flood of 1827

The high damage potential associated with subaqueous tunnelling

arises from the possibility of high water inflows or even a complete
flooding of the tunnel in the case of a hydraulic connection opening up
to the seabed. Tragic events of this kind have occurred many times, as
for example during the construction of Brunels Thames tunnel, the first
subaqueous tunnel in the world (Fig. 4).

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 4

19/06/14 10:22

1 >> Introduction

The risk of flooding is of course not limited to subaqueous tunnels. It

also exists in mountain tunnels where there are fault zones consisting
of so-called swimming ground [3]. The characteristic of subaqueous
tunnels, however, is that the potential volume of water recharge is
practically unlimited.
High pore pressures are not specific to subaqueous tunnels either. In
contrast to mountain tunnels, however, they may occur in combination
with a small depth of cover. As the head difference between the water
level and the tunnel has to be dissipated within a smaller distance, the
pore pressure gradients and consequently the destabilizing seepage
forces are higher (Fig. 5).
Furthermore, the small depth of cover means that the effective stress
and the shear resistance of the ground may be low relative to the pore
pressure. This effect is particularly pronounced in (but not limited to)
underconsolidated marine deposits, i.e., soils that are still consolidating
under the existing overlying sediment load and which contain an
excess pore pressure that carries part of the overburden.
The present lecture addresses some of the geomechanical questions
relating to subaqueous tunnels by means of selected projects, which,
besides their obvious differences in terms of overburden and depth
below the water table (Fig. 6), are also different with respect to the
type and extent of the geotechnical challenges. We start with the
issue of face support by earth pressure balance (EPB) shields under
suboptimum ground conditions, referring to the example of the
Storeblt railway tunnel (Section 2). This project provides an excellent
example of the interplay between geotechnical, material-technological,
mechanical- and process-engineering aspects in tunnelling. We next
investigate the limits of open mode TBM operation by considering
two hydraulic tunnels the Bosphorus-Melen 7 tunnel (Section
3) and the Lake Mead Intake No 3 tunnel (Section 4). In the first
tunnel, which mostly crosses competent hard rocks, the potential
hazard of face collapse is localized in distinct tectonic or volcanic
structures (faults, dykes), while in the Lake Mead tunnel, where weak
sedimentary rocks prevail over long portions of the alignment, this
hazard is present across the entire construction process. Section 5 is
about the conflicting criteria of settlement limitation and safety against
blowout that had to be satisfied during the slurry shield construction
of a part of the Zurich Cross Rail project due to the rare combination
of subaqueous tunnelling and tunnelling underneath a building. The
lecture closes with a future cutting-edge project, the Gibraltar Strait
tunnel (Section 6). Heavily squeezing conditions (i.e. extremely large
deformations of the ground) are expected over such a long portion of
the alignment that they represent a key technical feasibility factor in this
project.

Figure 5 : Contour lines of the hydraulic head, initial effective stress

0 and initial pore water pressure p0 for a deep tunnel (left hand
side) and, for a shallow underwater tunnel (right hand side)

Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 5

19/06/14 10:22

1 >> Introduction

Figure 6 : Longitudinal profiles

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 6

19/06/14 10:22

2 >> Storeblt railway tunnel

2.1 PrOJECT
The Storeblt fixed link project involves a twin bored railway tunnel
8 km long which connects the islands of Sprog und Seeland in
Denmark [4, 10, 11, 12]. The tunnel was constructed in 1990 - 1997
using four EPB shields (two starting from Sprog and two starting from
Seeland) of 8.75 m diameter. The ground in the project area consists
of fissured marls and glacial tills, each accounting for about 50% of the
alignment (Fig. 6a).
Tills are particularly unfavourable for tunnel construction: the till in
question is an overconsolidated soil with up to 20% clay in the shallow
part of the alignment and a very heterogeneous material in the deeper
part (containing irregular sand lenses, gravels and glacial boulders
and less than 10% clay). Typical parameter values for the till were a
cohesion c of 20 kPa, a friction angle of 35 and a coefficient of
permeability k of 10-7 - 10-5 m/s [13].
Considering the low shear strength and the high permeability of the till
as well as the high hydrostatic pressure (up to 5 bar), the main hazard
scenario was a collapse of the tunnel heading with subsequent failure
propagation up to the seabed. This became evident soon after the
start of TBM excavation. During a weekend in October 1991, a face
collapse occurred which created a hydraulic connection to the seabed.
The quantity of water flowing in increased rapidly to 4 m3/sec and led
through open bulkhead doors to flooding of the tunnel (at that time
350 m long), the TBM launch pit in Sprog and from there the parallel
tunnel as well (Fig. 7).

The remedial work took about 8 months to complete. In addition

to this spectacular incident there were several delays due to the
high abrasiveness of the tills and the frequent occurrence of granitic
boulders which necessitated time-consuming man entries into the
working chamber. Furthermore, it was often impossible to stabilize the
tunnel face by operating the EPB shield in closed mode. Towards the
end of 1993, when the two machines coming from Seeland reached
the deepest tunnel stretches in the till, these problems became so
critical that the TBMs were no longer able to advance.
2.2 Mechanics of face stabilization
The difficulties outlined above triggered thoroughgoing investigations
into the mechanics of face stabilization by EPB shields under
the specific non-ideal operational conditions that prevail in high
permeability ground [14].
Under ideal operational conditions (characterized by a fine-grained,
low-permeability soil under low hydrostatic pressure), the excavated
muck can be seen from a soil mechanics point of view as a
monophasic medium. The tunnel face is supported by the total
pressure in the muck, i.e. the pressure that is monitored by the
pressure trans-ducers in the cutterhead. A coarse-grained muck,
however, is a bi-phasic material consisting of solid grains and pore
water. In this case, a distinction must be made between the pore water
pressure and the pressure in the solids (hereafter referred to as pore
pressure and effective support pressure, respectively). Although
the sum of these two pressures is equal to the total pressure, they act
in different ways: the solids exert a stabilizing stress directly upon the
face, while the pore pressure is decisive in terms of the hydraulic head
field in the ground ahead of the tunnel face. If the pore water pressure
in the muck is equal to the in situ hydrostatic pressure (i.e. the hydraulic
head hF in the working chamber is equal to the sea level elevation hg),
then the ground is acted upon only by the force of gravity (represented
schematically by the submerged unit weight y in Fig. 8a).
In this case a low effective support pressure s or a low cohesion c are
sufficient for face stability. For the typical shear strength parameters
of glacial tills, limit equilibrium computations after [14] show that
the necessary effective support pressure s is equal to zero, i.e.
compensation of the water pressure will suffice for stability.

Figure 7 : Storeblt tunnel. Flooding of the Sprog construction

site [1]

Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 7

19/06/14 10:22

2 >> Storeblt railway tunnel

If, on the other hand, the hydraulic head hF in the chamber is lower
than the sea level elevation, then water will flow towards the tunnel
face, thus exerting seepage forces upon the ground. Figure 8 shows
the contour-lines of the piezometric head under atmospheric pore
water pressure in the working chamber. The seepage forces s are
oriented perpendicular to the contour-lines. Their magnitude increases
linearly with the head gradient and is, therefore, higher close to the
tunnel face. The seepage forces are unfavourable for stability and
necessitate a higher effective support pressure s in order for the face
to be stable. The lower the hydraulic head hF in the working chamber,
the higher the seepage forces and, consequently, the higher will be the
necessary effective support pressure s.

In conclusion, the ground response is controlled by the joint effect of

pore pressure and effective support pressure. However, these two
parameters cannot be controlled directly in practice. They depend
on the characteristics of the excavated ground, the way the ground
is mixed in the work chamber, the rotational speed of the screw
conveyor and the excavation advance rate. So, the ground response
to tunnelling by an EPB shield depends to a large degree on a complex
interplay of geotechnical and operational factors.

Figure 8 : Body forces (, s) acting upon the ground, necessary effective support pressure s and contour lines of the hydraulic head h:
(a) hydraulic head in the working chamber according to the sea level; (b) atmospheric pore water pressure in the working chamber

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 8

19/06/14 10:22

2 >> Storeblt railway tunnel

Let us consider now the conditions at chainage 12+700 of the

Storeblt tunnel, where it proved impossible to maintain water
pressure within the working chamber. The black line in Figure 9 shows
the relationship between the necessary effective support pressure
and the hydraulic head hF in the excavated muck.

the working chamber, may cause pressure fluctuations over time.

The large spatial or temporal variations in support pressure mean
that the pressure may be locally or temporarily insufficient, which
explains the difficulties observed in face stabilisation.
These operational and geotechnical problems can be avoided by
keeping the effective pressure s low. Figure 9 illustrates the basic
options in this respect: pore pressure relief in the ground ahead of
the face (blue line, Point C); ground improvement by grouting (red
line, point D); high hydraulic head in the working chamber (point B).
As can be seen from the red line in Figure 9, which applies to the
unfavourable combination of a high in situ hydraulic head (hG =
45 m) with atmospheric pressure in the chamber (hF = 0), even
rather modest ground improvement (characterized by a cohesion
increase from 20 to 75 kPa) leads to a considerable reduction in the
necessary effective support pressure s from 200 kPa to only about
70 kPa (point D).
Pore pressure relief by systematic drainage reduces the seepage
forces and thus also the necessary effective support pressure
considerably. The blue line in Figure 9 shows the relationship
between the effective support pressure s and the hydraulic head
hF in the chamber, assuming, for example, that drainage reduces
the hydraulic head hG in the ground from 45 m to 15 m. For the
given shear strength of the till, the necessary support pressure
would decrease to about 50 kPa (Fig. 9, point C). The effectiveness
of drainage depends, however, on the permeability of the ground.
As it was high in the present case, pore pressure relief was often
insufficient.

Figure 9 : Storeblt tunnel, ch. 12+700. Necessary effective

support pressure s as a function of the hydraulic head hF in the
working chamber (computation after [14])
If the pore water in the muck is atmospheric (point A), then the
necessary effective support pressure s amounts to about 200 kPa.
Such a high effective stress increases the frictional shear resistance
of the muck against the rotating cutterhead, thus leading to
excessive torque demand, which explains why it was impossible to
advance the TBMs. In addition, due to its high frictional resistance,
the muck no longer behaves as a fluid, with the consequence that
the distribution of the support pressure across the face may be
highly irregular instead of hydrostatic. Furthermore, a high effective
stress compacts the muck, thus increasing its stiffness, which, in
combination with irregularities in the mass flows into and out of

For the given shear strength of the till and the given in situ
hydrostatic head hG (Fig. 9, black line), a low effective support
pressure would be sufficient for stability only in combination with
a sufficiently high hydraulic head hF in the working chamber (e.g.
moving from point A to point B of Fig. 9). Normally, the water
pressure in the chamber can be kept high by creating a
low-permeability muck that acts as a barrier to groundwater flow.
The addition of bentonite slurry, polymers or foams to the muck
can help here to some extent. In the case in point, conditioning
attempts remained unsuccessful. In order to maintain a high water
pressure in the chamber, it was necessary to apply back-pressure
by installing a piston pump at the end of the screw conveyor.
This made it possible for one machine to continue excavation in
the glacial tills and to enter the underlying stable marls. The other
machine was put out of operation; as the TBM coming from Sprog
made good progress in the meantime, the decision was taken to
shift the joining point of the two drives to the location of the stuck
TBM [12].

Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 9

19/06/14 10:22

3 >> Melen 7 Bosphorus Tunnel

3.1 PrOJECT
The Melen 7 tunnel is located about 15 km north of central Istanbul. It
is the first bored tunnel underneath Bosphorus and also the first bored
tunnel in the world connecting two continents. The tunnel is a key
element of Istanbuls drinking water supply [18]. It has a bored diameter
of 6.11 m and is 5550 m long. A 3400 m long part, which includes the
section under the sea, was constructed using a shielded TBM (Fig. 6c).
The TBM was designed to sustain a maximum hydrostatic pressure of
13.5 bar during standstills and to operate in EPB mode at pressures of
up to 4 bar.
At the deepest section of the alignment, the TBM operated 135 m
below the sea level and 70 m below the seabed. The bedrock in
the project area consisted of alternating calcareous shales, clayey
or sandy limestones and sandstones. Due to thick granular alluvial
deposits, however, the minimum rock cover was as little as 35 m
in the marine section of the tunnel. This, in combination with the
existence of several faults as well as volcanic dykes in the project
area, implied a higher probability of encountering very high water
quantities or pressures in fault zones or in dykes communicating
with the seabed. The thickness of the dykes ranges from a few
meters up to 50 m and they consist of hard rocks with compressive
strengths of up to 140 MPa, which may nevertheless be fractured
and weathered. In the contact zones, the host sedimentary rock
is often completely crushed or disintegrated into a fine-grained
material. Due to this geological setting, together with previous
experience of tunnelling in the region, where groundwater problems
have been associated mostly with fault zones and/or dykes [6],
hazards such as a face collapse with subsequent tunnel flooding
were a major concern during the preparations for tunnelling.

Figure 10 : Melen tunnel. Fault types

The condition and behaviour of the ground in the faults depended

mainly on the dominant lithology of the competent host rock. Fault
zones in limestones with minor shale fractions appear blocky and
brecciated (Fig. 10a), while in predominately shaly rocks the fault
material is fine-grained (clayey or silty) and resembles soft ground
(Fig. 10b). Potential problems in fractured dykes and blocky fault
zones include high water inflows, which may cause difficulties in
mucking-out and in the installation of the segmental lining and
the annulus grouting, as well as rock instabilities in front of the
TBM, which may block or damage the cutter head. The risk of
high water inflows and the need for impermeabilization grouting
can be assessed on the basis of the water quantities observed
in exploratory boreholes systematically drilled ahead of the TBM
during construction. Besides sealing the rock mass, grouting also
improves its strength, thereby helping to deal with the stability
problems mentioned above at the same time.

10

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 10

19/06/14 10:22

3 >> Melen 7 Bosphorus Tunnel

Faults with fine-grained infillings were likely to be encountered in the more

shale-rich formations. The possible hazards arose from the combination of
high pore pressures with the low strength and high deformability of the fault
material and they included instabilities of the ground in front of the TBM, which
might have blocked the cutter head, as well as excessive deformations of the
ground, which can lead to shield jamming in the case of extended faults. It
should be noted that such problems cannot be identified in advance on the
basis of the water quantities in probe holes because the water inflows are very
limited in fine-grained materials of low permeability and may give a false sense
of security. Low stand-up times and face instabilities have been observed, for
example, in the Vardoe and Ellingsoy subsea tunnels in spite of less than 10
and 30 l/min, respectively, water ingress in the probe holes [15].
3.2 Stability assessment of faults with fine-grained
infillings
In the following, we discuss the parameters that govern stability in faults
with silty-clayey infillings on the simplified example of a vertical fault striking
perpendicular to the tunnel axis (Fig. 11a).
Only undrained conditions will be considered, i.e. the conditions prevailing
during continuous excavation or during short standstills. During longer
standstills, stability conditions become increasingly less favourable. The
assumption of short-term conditions is nevertheless reasonable given the
limited extent of the fault zones. Should the geological pre-exploration show
the presence of a fault, organizational measures (such as rescheduling
machine maintenance) can be undertaken in order to allow continuous rapid
excavation through the fault without interruptions.
The parameters governing short-term stability are: the undrained shear
strength Su of the ground, the depth of cover, the sea depth and the thickness
d of the fault. Their effect can be estimated by means of limit equilibrium
computations. The diagram in Figure 11 shows the necessary face support
pressure s as a function of the fault thickness d in respect of three cross
sections in the subsea section of the tunnel.

The following conclu-sions can be drawn from this diagram: For a given depth
below the sea level, the neces-sary face support pressure depends sensitively
on the depth of cover; the highest values correspond to the cross section A
in the central portion of the alignment. Tunnelling ex-perience indicates indeed
that the tunnel stretches with the highest water pressure and the smallest
depth of cover are the most critical [15, 17]. Furthermore, the narrower the
fault, the lower will be the necessary face support pressure. This is due to the
stabilizing effect of the shear stress at the interface between the fault and the
competent shale.
As can be seen from Figure 11b, face support is needed only if the fault
thickness d is greater than the critical thickness dcr . Figure 11c shows the
critical fault thickness dcr as a function of the undrained shear strength Su of
the ground for the three cross sections. Faults narrower than 2 - 3 m do not
present a stability problem even at the deepest portion of the alignment and
under extremely low shear strength values. Depending on the strength of the
fault material, thicker faults may necessitate face support (operation in closed
mode), ground improvement by grouting or pore pressure relief in the ground
ahead of the TBM by advance drainage. For shear strengths Su higher than
50 - 60 kPa, the face would remain stable without any measures. In view of
the difficulties of thorough advance grouting (due to the low permeability of
fine-grained soils and the constraints imposed by the tunnelling equipment on
the drillhole pattern), it is interesting to note that even a relatively small ground
improvement suffices to stabilize the face.
The TBM drive started March 2008 and finished without particular problems
in April 2009, about one month earlier than planned [18]. The evaluation
outlined above assisted in defining the criteria for decision-making during
construction with respect to advance probing, ground improvement and TBM
operational mode. The measures foreseen were: rotary percussive drilling for
routine advance probing and core drilling only for fault zones thicker than 2 m
consisting of ground of uncertain characteristics (fractured rock or soil that was
predominantly clayey or silty); stabilization grouting and/or closed mode TBM
operation in the case of fault zones thicker than 2 m consisting of low to medium
permeability soil (e.g. silty fault gouge, but not a practically impervious clay).

Figure 11 : Melen tunnel. (a) Problem layout; (b) Required support pressure s as a function of the
thickness d of the fault (undrained shear strength Su = 40 kPa); (c) Critical fault thickness dcr for an
unsupported tunnel face as a function of undrained shear strength Su (computation after [16])
Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 11

11

19/06/14 10:22

4 >> Lake Mead Intake No 3 Tunnel

4.1 PrOJECT
Lake Mead, behind the Hoover Dam, supplies about 90% of Las
Vegas valleys water. Over recent years, drought has caused the
lake level to drop by more than 30 meters. In order to maintain
water supplies, a third intake is under construction that is
deep enough to function at the lowest lake levels [7]. The main
structures of the project are a 170 m deep access shaft, a tunnel
approximately 4700 m long with a bored diameter of 7.22 m and
an intake structure in the middle of the lake.
The tunnel crosses metamorphic rocks and tertiary sedimentary
rocks (conglomerates, breccias, sandstones, siltstones and
mudstones of very variable quality), at a maximum depth of about
139 m beneath the current lake level. The rock cover decreases in
the last portion of the alignment, amounting locally to just 20 - 30 m
(Fig. 6d).
Due to the existence of several faults in the project area, the ground
at the elevation of the tunnel may be recharged directly from Lake
Mead, which implies the possibility of considerable water ingress
during construction. Given the high hydrostatic pressures and the
poor quality of the prevailing sedimentary rocks over long portions
of the alignment, attention was paid right from the start to the
potential hazards of a cave-in of the rock at the working face or
a flooding of the tunnel, and a decision was taken to construct
the tunnel using a convertible hybrid TBM [7, 19]. The TBM is
capable of boring in open- or closed-mode. In open mode, a screw
conveyor extracts the excavated rock from the working chamber. In
closed mode, the screw conveyor is retracted from the cutterhead
and the TBM operates as a closed shield by applying a pressurized
bentonite slurry which counterbalances the hydrostatic pressure
and stabilizes the tunnel face. The machine is designed to cope
with water pressures of more than 14 bar - the highest pressures
ever seen in closed shield tunnelling anywhere in the world (Fig. 12).
Although the TBM can bore in closed mode at this depth below
the water level, however, the high hydrostatic pressures make
inspection and maintenance in the working chamber extremely
demanding.
The inherent technological risk of such high-pressure closedmode TBM operation and the lack of experience with hyperbaric
interventions at 14 bar in tunnelling made it necessary to develop
fallback strategies involving open mode operation, potentially in
combination with advance grouting and/or drainage.
4.2 Assessment of ground behaviour
The main difficulty with assessing the behaviour of the prevailing
weak, water-bearing, low-permeability tertiary rocks is that their
response to tunnel excavation is time-dependent. This means
that the tunnel face might be initially stable but fail after a period
of time: the short-term behaviour of the ground (i.e. the behaviour

12

Figure 12 : Maximum slurry pressures in selected tunnel projects

with high hydrostatic pressure (after [24], revised)
under so-called undrained conditions) is characterized by a
constant water content and the development of negative excess
pore pressures which have a stabilizing effect. This effect is only
temporary, however, because the excess pore pressures dissipate
over the course of time. The long-term behaviour (so-called drained
conditions) is characterized by a fully developed seepage flow
towards the tunnel, which is unfavourable for face stability and may
necessitate stabilisation measures (depending on the shear strength
of the ground).
In the present case, it was expected that an unsupported face
would be stable under undrained conditions (i.e. in the short term),
but unstable under drained conditions (i.e. in the long term) for long
stretches through tertiary rocks [20]. The central questions were
thus: how long would the face remain stable? how certain was it
that favourable undrained conditions would prevail during ongoing
excavation or short standstills? The decisive parameter in this
respect was the permeability k of the ground, which governs how
rapidly the excess pore pressures dissipate and how rapidly the
transition from undrained to drained conditions occurs. In general,
an assumption of favourable short-term conditions will be more
reasonable the less permeable the ground, the more rapid the
excavation and the shorter the standstills.

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 12

19/06/14 10:22

4 >> Lake Mead Intake No 3 Tunnel

The stand-up time of the tunnel face can be estimated by coupled

hydraulic-mechanical numerical calculations that take account of
the time-dependent processes in the ground ahead of the tunnel
face [21, 22, 23]. Figure 13 shows the results of such an analysis.
For permeability values less than 10-8 m/s, the stand-up time
amounts to several days, which means that the conditions during
TBM advance (including short standstills of about 0.5 - 1 day) are
practically undrained. For higher permeabilities or longer standstills, however, unfavourable drained conditions must be taken into
account. For the expected range of permeability, the stand-up time
can be anything between a few hours and several days. A stand-up
time of several days would allow open mode TBM operation and
maintenance under atmospheric conditions. A stand-up time of a
few hours might allow TBM advance in open mode or at low slurry
pressure, but would probably necessitate hyperbaric interventions
for maintenance. The difference between a few hours and several
days is thus very significant from the construction point of view.
4.3 Control of the face
The computation of Figure 13 was carried out for a specific tunnel
section (in the so-called Red Sandstone formation), but the results
are also typical for the other tertiary rocks that prevail in the major
part of the Lake Mead tunnel. They indicate that long portions of
the alignment are in the geotechnically demanding transition zone
between drained and undrained conditions. In this zone it cannot
be said with certainty that favourable short-term conditions apply.
This introduces an element of uncertainty concerning tunnel face
stand-up time, with direct consequences for the operating mode.
Drained conditions in combination with low shear strength of the
ground would necessitate either closed mode operation or auxiliary
measures such as advance drainage or grouting of the ground
ahead of the tunnel face.

Figure 13 : Lake Mead Intake No 3. Stand-up time of the tunnel

face as a function of permeability (computation after [22])

Advance drainage decreases the pore pressures and thus also the
destabilizing seepage forces acting within the ground towards the
face. Drainage-induced pore pressure relief is significant even under
the practical limitations imposed by the construction equipment
with respect to the spacing and number of boreholes [25, 26] (Fig. 14).
As can be seen from Figure 15, pore pressure relief has direct
consequences in terms of the slurry or compressed air pressure
that is needed in order to ensure face stability. Advance drainage
increases the feasibility range of open mode operation (bottom of
Fig. 15).
In the case of particularly poor quality ground, advance drainage
either alone (Fig. 15, BC) or in combination with grouting (Fig. 15,
ABC) allows reductions to be made in operational pressure.

Figure 14 : Lake Mead Intake No 3. Contour lines of the hydraulic

head (dark blue: almost atmospheric pore pressure)

Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 13

13

19/06/14 10:22

4 >> Lake Mead Intake No 3 Tunnel

4.4 Experience in the field

The TBM started in December 2011, and after 18 months and a
950 m drive it reached the sedimentary rocks, in which it has so
far tunnelled 1800 m (in a period of 10 months). The sedimentary
rocks proved to be sufficiently stable at least in short term, making
it possible to operate the machine practically always in open mode,
always in combination with three 45 m long drillholes (overlapping
by 10 m) through the cutter-head, which were used for advance
probing and drainage. During longer standstills (more than two
days), 2 to 3 additional 10 m long drillholes were installed.

However, the biggest problem was the virtual impossibility of

accessing the excavation chamber for maintenance under
atmospheric pressure. In order to assess the feasibility of men
entering the working chamber, the quantity of water inflow was
estimated in advance by using the TBM as a large-scale constanthead permeameter, i.e. by reducing the slurry pressure stepwise
and monitoring the water inflow using a pre-defined procedure [27].
Attempts to lower the slurry pressure from the in situ hydrostatic
pressure (14 bar) to atmospheric pressure were often interrupted
because the water inflows reached hundreds of cubic metres per
hour even at relatively high slurry pressures (Fig. 16).

Considerable difficulties were encountered, however, in the first part

of the alignment through metamorphic rocks. These were caused
by the unfavourable combination of high water pressure, extremely
high rock permeability and the presence of an unexpected fault
zone. The fault, consisting of almost cohesionless material, was
oriented subparallel to the tunnel and therefore affected the works
for a considerable portion of the alignment. These conditions made
it necessary to operate the slurry shield in closed mode at 14 bar
for several hundred metres. This is a remarkable achievement; it
has never been done before anywhere in the world.

Conditions such as these leave two possible options for carrying

out maintenance work: hyperbaric intervention or ground
improvement by grouting. Although the necessary equipment and
logistics for saturation diving were available on site, the decision
was made to proceed with the second solution because of the
higher chances of success (hyperbaric intervention has never
been attempted in tunnelling at 14 bar). A series of pre-excavation
grouting campaigns succeeded in reducing water inflow to an
extent which allowed maintenance work to be carried-out at
least on the slurry lines, which was indispensable for continuing
excavation. Work could be performed at the cutterhead only later,
after reaching competent rock [27].

Figure 15 : Lake Mead Intake No 3. Slurry or compressed

air pressure s required for long-term stability as a function of
the effective cohesion c with and without advance drainage
(computation after [25])

Figure 16 : Lake Mead Intake No 3. Measured quantity of water

inflow as a function of the slurry pressure at different locations (solid
lines) and theoretical prediction for two values of the permeability
coefficient k (dashed lines)

14

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 14

19/06/14 10:22

5 >> Zurich Cross Rail

5.1 Project
The Zurich Cross Rail project includes a 4500 m bored doubletrack tunnel of 11.3 m diameter [5]. The last 280 m of the tunnel
crosses a glacial till overlain by lake deposits and fluvial gravels. An
86 m long part of the soft ground section runs at a minimum depth
of about 9 m below bed of the Limmat river and 13 m below river
level (Fig. 6b).
The tunnel was constructed using a shielded TBM, which was
operated in the soft ground section in closed mode as a slurry
shield. In order to reduce the impact of tunnelling on surface
structures, the last 140 m were excavated under the protection of a
large diameter pipe arch (Fig. 17a).
In the following we will focus on an interesting peculiarity of this
tunnel that arose from the rare combination of subaqueous
tunnelling and tunnelling underneath a building on the west bank of
the river (Fig. 17c).
5.2 Subaqueous tunnelling underneath a building
In the soft ground section, face stability was ensured through a
pressurized bentonite slurry that compensated the hydrostatic
pressure and exerted an excess pressure on the face (Fig. 17b).
Due to the small depth below the riverbed, the slurry pressure
acting upon the crown involved the risk of a blowout. During
TBM advance under the river, the slurry pressure required at the
crown was 20% lower than the critical blowout pressure and thus
not problematic. More critical in this respect were the hyperbaric
interventions needed for inspection and maintenance of the cutter
head, because of the inherently higher excess air pressure at the

crown. The risk of a blowout proved to be manageable, however,

by a partial lowering of the slurry level (up to the tunnel axis). The
conditions under the river were demanding but nevertheless rather
typical and well known for this type of work.
An interesting situation arose under the west bank of the river at
a specific phase in the construction where the cutterhead was
located underneath a commercial building, while a part of the shield
was still underneath the riverbed and with reduced cover due to existing bridge abutments. Here, the uplift resistance of the overlying
ground to the slurry pressure was slightly lower than before due to
the smaller depth of cover above the rear part of the shield.
At the same time, in order to limit the settlement of the building,
the TBM had to be operated at a higher slurry pressure than before
(Fig. 17c). In addition, there was a possibility that the TBM might
encounter obstacles in the form of old sheet pile walls.
In the worst case this would mean work at the cutterhead having to
be carried out under slurry support by divers due to the conflicting
criteria of settlement limitation and safety against blowout.
Tunnelling under such conditions resembles a tightrope walk.
The pressure required for settlement control is higher than that
required for stability alone, but, due to the presence of the pipe
arch, probably considerably lower than the theoretical maximum
pressure, i.e. the pressure corresponding to the condition at rest.
The experience from the TBM drive confirmed this expectation, thus
underlining the importance of the pipe arch. Non-stop operation,
precise slurry pressure control, continuous real-time settlement
monitoring and the excellent performance of the TBM during the
advance through the sheet piles were all essential to the successful
completion of the work.

Figure 17 : Zurich Cross Rail. Tunnel cross section (a), slurry pressure and critical blowout pressure underneath the river (b) and at the river
bank (c)

Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 15

15

19/06/14 10:22

6 >> Gibraltar strait tunnel

6.1 Project
The project under study for the Gibraltar Strait crossing foresees a bored
tunnel approximately 38 km long, of which about 28 km will be under
the sea [8, 9]. The largest part of the tunnel is located in Flysch. At the
central section of the alignment, where the sea is 300 m deep and the
overburden amounts to approximately 200 m, the tunnel will cross two socalled paleochannels filled with clayey breccias of extremely poor quality
(Figure 6e). The very low strength and high deformability of the breccias in
combination with the 50 bar pore pressures prevailing at tunnel elevation
will cause squeezing conditions, i.e. large deformations of the opening and
development of high loads upon the tunnel lining. This, in combination with
the probable length of the critical stretch, its great depth and the distance
from the European and the African coast (all of which limit the construction
options) makes crossing the breccias a key technical challenge of the
project [28, 29].

Figure 19 : Gibraltar tunnel.

Breccia specimen consisting of a clay
matrix with hard inclusions

6.2 The breccia section

Due to the great sea depth, the periodically strong sea currents and the
heavy ship traffic through the strait, even the geological exploration of the
breccias is difficult. Sea currents and ship traffic lead to frequent interruptions
in the probe drilling. Special devices were developed to allow re-entries of
equipment and restarts of probing (Fig. 18a).
In addition, the direction of the sea currents is different in the deeper areas of
the strait (Fig. 1), which, in combination with the great sea depth, may cause
twisting of the drill rods. In order to reduce hydrodynamic resistance and
avoid overstressing of the rods, aerofoil type fairings were developed (Fig.
18b).
The geotechnical testing of the breccias is also particularly demanding. Due
to the prominent role of pore water pressure, the mechanical behaviour of
the breccias is investigated by means of consolidated drained (CD) and
consolidated undrained triaxial (CU) tests [32]. The difficulties of testing are
associated with the structure and permeability of the material and the high
in situ pore pressures and effective stresses. As a consequence of the low
permeability of the clayey matrix of the breccias (measured k = 10-13 - 10-11
m/s), triaxial testing takes long time (4 - 8 weeks, sometimes more) even

Figure 20 : Back-pressure device

with step-motor for triaxial testing
of extremely low-permeability
materials

Figure 18 : Gibraltar tunnel. (a) Template for re-entry probe borehole; (b) hydrodynamically favourable fairing [30, 31]

16

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 16

19/06/14 10:22

6 >> Gibraltar strait tunnel

when using small specimens. The presence of hard, gravel-size inclusions

within a very soft clayey matrix (Fig. 19) in combination with low permeability
(which makes the testing of large samples impossible) also renders the
selection of representative specimens difficult.
Due to the hard inclusions, specimen preparation is very demanding. Satisfactory
results have been achieved by cutting the samples with air-flushing through the use
of a special electronically controlled diamond band saw.
Due to the high in situ pore pressures and effective stresses, testing equipment
with high nominal loads must be used. (It is almost a matter of soil-mechanics
testing under conditions which are more typically associated with rock
mechanics problems). On the other hand, in spite of the high nominal loads,
the strain rate has to be finely tuned within an extremely low range due to the
low permeability of the material. An electromechanical high precision stepmotor was developed specifically for controlling the pore pres-sure (Fig. 20).
According to the results of the CD and CU triaxial tests, the breccias exhibit
cohesion values of only a few 100s of kPa, friction angles in the order of
15 - 20 and Youngs moduli of a few 100s of MPa [32]. Figure 21 gives an
impression of the deformability of the material.

In the following, the practical significance of such a poor quality ground for
tunnelling will be illustrated by means of computational results concerning
the short-term behaviour of the ground, i.e. the behaviour under constant
water content. The short-term response of the ground to excavation, i.e. its
behaviour close to the tunnel face, is important from the constructional point
of view. Depending on the intensity of squeezing, it may be necessary to apply
large amounts of support close to the working face in order to control the
ground. This slows down tunnel advance considerably, as support installation
interferes with the excavation work. In mechanized tunnelling, which is an
important construction option in the present case due to the great tunnel
length, squeezing may result in the complete immobilization of the tunnel
boring machine [34, 35] (Fig. 22).
The black solid line in Figure 23 shows the relationship between the radial
displacement of the excavation boundary and the pressure supporting the
ground under the simplifying assumption of rotational symmetry (ground
response curve). For the given parameter set, which applies to the conditions
prevailing in the breccias in the deepest section of the alignment, the maximum
convergence of an unsupported opening is equal to 50% of the tunnel radius.
Even with a heavy support pressure of 1 - 2 MPa, the estimated convergence
reaches 20 - 30% of the tunnel radius. It should be noted that the ground
response curve was obtained by means of a large strain analysis; neglecting
the geometric non-linearity arising in the case of such large deformations
would lead to a convergence of more than 100% of the tunnel radius (Fig. 23,
dashed line), which is physically meaningless. Tunnelling experience shows
that drainage not only improves the stability of the tunnel heading (Section 3),
but is also favourable with respect to squeezing, because the consolidation
of the ground increases its resistance to shearing [37, 40], thus leading to
a dramatic reduction in the deformation of the opening (Fig. 23, blue line).
Advance drainage is in fact an important auxiliary measure for the breccia
section of the Gibraltar tunnel [41, 42]. In view of the very low permeability of
the breccias, complete pore pressure relief by advance drainage may take a
prohibitively long time. The effect of drainage is, nevertheless, remarkable even
with partial pore pressure relief (Fig. 23, red line).

Figure 21 : Gibraltar tunnel. Breccia specimen before and after

CD-testing [33]

Figure 22 : Shield jamming due to squeezing rock [36]

Figure 23 : Gibraltar tunnel. Short-term ground response curves

with and without advance-drainage (computation after [37, 38, 39])
Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 17

17

19/06/14 10:22

7 >> Closing Remarks / 8 >> References

Subaqueous tunnels are demanding projects and, as with every

demanding project, a motor for technological progress. They are
also a motor for advancing the science of tunnelling, which benefits
the entire industry. Thus, for example, the Storeblt project was
instrumental in improving our understanding of the mechanics of
face stabilization by EPB shields [14]. Or, to take another example,
the Seikan and Gibraltar tunnels triggered important research on
grouting and drainage [43, 44] with results that have been very
beneficial in the planning phase of the base tunnels through the
Alps.
Demanding projects also increase the attractiveness of the civil
engineering profession and provide a powerful motivation for our
students. Recent large-scale infrastructure projects that gained
public attention have been influential in increasing civil engineering
student numbers in Switzerland. Despite the wide range of
project conditions and construction methods in the case studies
discussed, there is a unity in the underlying physical principles.
The success of a project often depends on our respect for these
principles and our understanding of what lies behind phenomena
such as instability or large deformations. It is important to pass
this message on to students and to enable them to push back the
boundaries of knowledge even into areas where no experience
is yet available to draw on as they respond to the challenge of
increasingly difficult tasks and the rapid development of technology.
Acknowledgements : The author thanks Mr. Roberto Schuerch,
MSc ETH, for his suggestions and help to prepare this paper.

[1] Ostenfeld, K.H., Steenfelt, J.S., Pedersen, F.M., Srensen, C.S.

2003. Transcendent foundation solutions - the cross pollination
of scientific and industrial disciplines to overcome economic and
technical challenges in foundation of large bridge structures.
Jubilee volume in celebration of 75th anniv. of K. Terzaghis
Erdbaumechanik Vienna Technical University.
[2] elik, . 2009. Presentation on Marmaray project. Rep. of
Turkey, Min. of Transp., Gen. Directorate of Railways, Harbors and
Airport Construction (DLH).
[3] Anagnostou, G., Kovri, K. 2005. Tunnelling through geological
fault zones. Int. Symp. on Design, Construction and Operation of
Long Tunnels, Taipei, Taiwan.
[4] Biggart, A.P., Rivier, J.-P., Sternath, R. 1993. Storeblt Railway
Tunnel - Con-struction. Proc. Int. Symp. on Technology of Bored
Tunnels under Deep Water-ways, Copenhagen, Denmark, 63 - 93.
[5] Bosshard, M., Bolliger, J., Kobel, R. 2011. Zurich Cross Rail,
Weinberg Tunnel - the challenges of the loose ground section and
passing under the river. Geomechanics and Tunnelling 4, Volume 6,
651 - 663.
[6] Algan, M. 2007. Greater Istanbul Water Supply Project - Melen
System Stage I - Preliminary Engineering Geological Evaluation and
Geological Profile For Melen Bosphorus Crossing Tunnel Route.
ZEMAR Soil Site Investigation Ltd., Istanbul.
[7] Feroz, M., Jensen, M. & Lindell, J.E. 2007. The Lake Mead
intake 3 water tunnel and pumping station, Las Vegas, Nevada,
USA. RETC Proceedings, 647 - 662.
[8] Pliego, J.M. 2005. The Gibraltar Strait tunnel. An overview of the
study process. Tunnelling and Underground Space Technology, 20,
558 - 569.
[9] SNED & SECEG 2007. Project of the fixed link through the strait
of Gibraltar. Rabat: Socit Nationale dEtudes du Dtroit, Madrid:
Sociedad Espaola de Estudios para la Comunicacin Fija a travs
del Estrecho de Gibraltar.
[10] Darling, P. 1993. Coming clean over Storeblt. Tunnels &
Tunnelling, 10, 18-26.
[11] Sternath, R. 1994. Geotechnische Sonderverfahren fr
die Vortriebe des Eisenbahntunnels unter dem Grossen Belt.
Baugrundtagung, 175 - 195, Kln.
[12] Zell, S. 1995. Final Stage Of Storebaelt Railway Tunnel Summary And Appraisal. RETC Proceedings, Chapter 48.
[13] Krebs Ovesen, N. 1997. Invited Lecture Geotechnical
aspects of the Storeblt Project. Proc. 14th Int. Conf. Soil Mech.
Found. Engng., vol. 4, 2097 - 2104, Hamburg.
[14] Anagnostou, G., Kovri, K. 1996. Face stability conditions
with Earth Pressure Balanced shields. Tunnelling and Underground
Space Technology, 11 (2), 165 - 173.
[15] Dahl, T. S., Nilsen, B. 1994. Stability and rock cover of
Norwegian hard rock subsea tunnels. Tunneling and Underground
Space Technology, Vol. 9, No. 2, 151 - 158.

18

Some critical aspects of subaqueous tunnelling

20975-Muir Wood 2014-.indd 18

19/06/14 10:22

8 >> References

[16] Anagnostou, G., Kovri, K. 1994. The face stability of

slurry-shield driven tunnels. Tunnelling and Underground Space
Technology, 9 (2), 165 - 174.
[17] Nilsen, B. 2014. Characteristics of Water Ingress in Norwegian
Subsea Tunnels. Rock Mechanics and Rock Engineering, 47, 3,
933 - 945.
[18] Herr, R. 2009. Tunnel beneath the Bosphorus Strait to
safeguard Istanbuls Water Supply. Tunnel, 4, 69 - 74.
[19] McDonald, J. & Burger, W. 2009. Lake Mead Intake Tunnel No.
3. Tunnel, 4, 43 - 48.
[20] Anagnostou, G., Cantieni, L., Nicola, A., Ramoni, M. 2010.
Face Stability Assessment for the Lake Mead Intake No 3 Tunnel.
ITA-AITES World Tunnel Congress 2010 Tunnel Vision Towards
2020, Vancouver.
[21] Hfle, R., Fillibeck, J., Vogt, N. 2008. Time dependent
deformations during tunnelling and stability of tunnel faces in finegrained soils under groundwater. Acta Geotechnica, 3, 309 316.
[22] Schuerch, R., Anagnostou, G. 2012. Tunnel face stability under
transient conditions: stand-up time in low permeability ground.
22nd European Young Geotechnical Engineers Conference,
Chalmers.
[23] Schuerch, R., Anagnostou, G. 2013. Analysis of the standup time of the tunnel face. In: Anagnostou & Ehrbar (eds),
Underground the way to the future, World Tunnel Congress,
Geneva, 709 - 714.
[24] Holzhuser, J., Hunt, S. W., Mayer, C. 2006. Global Experience
with Soft Ground and Weak Rock Tunneling under Very High
Groundwater Heads. In: Ozdemir (ed.), North American Tunneling,
277 - 289, Taylor & Francis Group, London.

[32] Dong, W., Pimentel, E., Anagnostou, G. 2013. Experimental

investigations into the mechanical behaviour of the breccias around
the proposed Gibraltar Strait tunnel. Rock Mech Rock Eng, 46, 923
- 944.
[33] Anagnostou, G., Pimentel, E., Cantieni, L. 2008. Gibraltar
strait fixed link Tunnel project. Experimental investigations on the
strength and deformability of Breccia material. ETH Zurich, Report
Nr. 081301 to SECEG SA, Madrid, Spain, and SNED, Rabat,
Morocco.
[34] Ramoni, M., Anagnostou, G. 2010. Tunnel boring machines
under squeezing conditions. Tunnelling and Underground Space
Technology, 25, 139 - 157.
[35] Ramoni, M., Anagnostou, G. 2010. The effect of consolidation
on TBM shield loading in Water-Bearing squeezing ground. Rock
Mech Rock Eng., 44, 63 83.
[36] Courtesy Werner Burger, Herrenknecht AG.
[37] Anagnostou, G. 2009. The effect of advance-drainage on the
short-term behaviour of squeezing rocks in tunneling. Int. Symp. On
Computational Geomechanics, 668 - 679, Juan-les-Pins, Cote d
Azur, France.
[38] Vrakas, A., Anagnostou, G. 2013. Large deformation analysis
for a planned tunnel crossing heavily squeezing ground. In:
Anagnostou & Ehrbar (eds), Underground the way to the future,
World Tunnel Congress, Geneva, 746 - 753.
[39] Vrakas, A., Anagnostou, G. 2014. A finite strain closed-form
solution for the elastoplastic ground response curve in tunnelling. Int
J Num and An Meth in Geomech. DOI: 10.1002/nag.2250
[40] Kovri, K. 1998. Tunnelling in squeezing rock. Tunnel 5/98,
12 - 31.

[25] Zingg, S., Anagnostou, G. 2012. The effects of advance

drainage on face stability in homogeneous ground. ITA-AITES World
Tunnel Congress Tunneling and Underground Space for a Global
Society, Bangkok, Thailand.

[41] Floria, V., Fidelibus, C., Repetto, L., Russo, G. 2008. Drainage
and related in-crease of short-term strength of low permeability
rock mass. Building underground for the future; AFTES International
Congress Monaco, Monte Carlo, 281-284, AFTES Paris.

[26] Anagnostou, G., Zingg, S. 2013. On the stabilizing effect of

advance drainage in tunnelling. Geomechanics and Tunnelling, 6
(2013), No. 4, 338 - 354.

[42] Panciera, A. 2009. Gibraltar Tunnel Herausforderungen

bei der Planung eines Tunnels an der Grenze der Machbarkeit.
Colloquium Tunnelbau in druckhaftem Gebirge, ETH Zurich.

[27] Nickerson, J., Bono, R., Donadoni, N., Nicola, A., Anagnostou,
G., Schuerch, R., Zingg, S. 2014. Lake Mead Intake Tunnel No. 3
A step beyond the limits. Swiss Tunnel Congress, Lucerne.

[43] Soriano, A., Oteo, C.S., Serrano, J.M. 1989. Effects of

previous soil improve-ment with high pressure grouting in
subaquatic tunnels. Tunnels and Water, Ser-rano (ed.), 839-845,
Balkema, Rotterdam.

[28] Russo, G., Grasso, P., Bensaid, A. 2008. A framework for the
risk analysis of the Gibraltar strait Railway-link tunnel. World Tunnel
Congress 2008 - Underground Facilities for better Environment and
Safety - India, 1726 - 1735.
[29] Lombardi, G., Neuenschwander, M., Panciera, A. 2009.
Gibraltar Tunnel Project update the geomechanical challenges.
Geomechanics and Tunnelling, 2, No. 5, 578 - 590.

[44] Adachi, T., Tamura, T. 2005. Undersea Tunnel-Effect of

Drainage and Grouting. Ch. 12 in Ground Improvement Case
Histories (Eds: Buddhima Indraratna and Jian Chu), 327 - 355,
Elsevier, Amsterdam, Boston.

[30] Sandoval, N. 2010. Las investigaciones geolgicas dentro de

los estudios de factibilidad de un tnel excavado bajo el estrecho
de Gibraltar. SEGECSA, Presentation ETH Technical visit to Tarifa.
[31] Fugro Seacore. Strait of Gibraltar Site Investigation.
http://www.seacore.com/projects/strait-of-gibraltar

Muir Wood Lecture 2014

20975-Muir Wood 2014-.indd 19

19

19/06/14 10:22

longrine 04 90 14 48 48 - (20975 - 03/14)

ITA Secretariat c/o EPFL - Bt. GC - Station 18 - CH - 1015 Lausanne - Tel. : + 41 21 693 23 10 - Fax : + 41 21 693 41 53
Email : secretariat@ita-aites.org - Web : www.ita-aites.org

20975-Muir Wood 2014-.indd 20

19/06/14 10:22