Project Report 30 March 1996

Prediction and effects of ground

movements caused by tunnelling in
soft ground beneath urban areas
Prepared under contract to CIRIA by
Mott, Hay and Anderson, now Mott MacDonald

L M Lake MSc DIC PhD CEng MICE MIMM FGS

CIRIA
Project Report 30 March 1996

Prediction and effects of ground

movements caused by tunnelling in
soft ground beneath urban areas
Prepared under contract to CIRIA by
Mott, Hay and Anderson, now Mott MacDonald

L M Lake MSc DIC PhD CEng MICE MIMM FGS

W J Rankin MSc CEng MICE FGS
J Hawley MA CEng MICE

This report has been issued to Project fenders, CIRIA Core Programme
Sponsors, members of the Steering Group and those who made
technical contributions to the project.

CIRIA has no objection to any Funders Report being photocopied by

entitled recipients, provided that the contents are treated with the same
confidentiality as the original.
 Summary

The prediction of ground movements caused by tunnelling in soft ground is of renewed interest
within the United Kingdom as a result of an upsurge in tunnelling in urban areas, particularly
for transportation as well as water supply and waste water. Attempts at predicting displacements
associated with tunnel construction in soft ground using theoretical analytical techniques have
met with limited success to date. Traditionally, predictions have been based on empirically
derived relationships from case history information and this is likely to continue to be the case.

A wide ranging review of case histories was a fundamental component of the initial stages in
preparation of this report. The case histories are drawn from a diverse set of tunnelling projects
around the world, with many examples in urban areas especially under existing streets. The case
records cover a variety of ground conditions, although there is a preponderance of data from
tunnelling in stiff clays. The majority of cases relate to traditional shield tunnelling, describe
surface settlements only and the detail provided varies considerably. Often information relating
to a particular location is not collated and has therefore to be inferred, virtually done of the
records discuss or relate ground displacements to structure damage. Only a few examples of the
more recently introduced techniques such as earth pressure balance, slurry shields or
observational New Austrian Tunnelling Method (NATM) type techniques were available at the
time of study although more records of experience with these techniques are now becoming
available. The present evidence suggests that the same empirically based methods of prediction
of surface settlement are also applicable to these techniques, but this needs further
corroboration.

Tunnelling in soft ground is inevitably accompanied by sub-surface ground movements which

are manifest at the surface in a trough centred over the tunnel and extending in front of the
advancing face. Two distinct phases of ground movement are commonly identified. An initial
short term phase is associated with construction and occurs typically within a period of a few
weeks after the passage of the tunnel face. The subsequent phase of movement is time-
dependent and in clay soils is believed to be caused by consolidation of the clay stratum during
drainage of the surrounding soil to the tunnel. The data available confirm that the shape of the
initial surface settlement trough developed across and behind the tunnel face follows that of an
inverted normal distribution curve for a variety of ground conditions and tunnelling methods.
The lateral extent of the trough is primarily related to the depth of the tunnel, but there are
important differences between clay and granular soils.

The magnitude of the initial maximum settlement at the surface can be related to the volume
loss associated with ground movements into the tunnel face, around the shield (if used) and
behind the lining, as well as the depth and diameter of the tunnel. The shape of the surface
trough in many case histories approximates to a normal distribution curve which can be
conveniently expressed mathematically to provide a means of assessing settlement and slope at
any point across the trough. The prediction of the horizontal displacements that accompany
settlement is less satisfactory and requires an assumption of radial movement to the tunnel axis
to permit generalised formulae to be derived. However, this assumption deserves further study
since horizontal strain is of considerable significance in assessment of structural response. In
regard to sub-surface displacements the available case history records highlight significant
differences in the observed pattern depending on tunnelling method and performance even
though a normal distribution settlement trough might be observed at the surface.

Existing criteria for classification of damage to structures are based upon allowable settlement
beneath foundations and the extent and magnitude of cracking within plaster, brickwork or
masonry. These factors in turn can be related to structure deformation parameters of relative
rotation (8) and deflection ratio (A/L) of the structure or structural elements. Limiting values of
these deformation parameters are more severe for structures sited within zones where hogging
type deformation and tensile horizontal strain occurs, such as on the limbs of the settlement
trough. This report links these limiting deformation parameters to the more simply calculated
parameters of ground slope and settlement for an initial assessment of risk of damage. Predicted

2 FRCP/5
CIRIA Core Programme PR 0 1' ErT REPORT 30
Prediction and effects of ground movements caused by tunnelling in soft ground in urban
areas

CORRIGENDA

page 7 Delete from the list of figures

Figure 13 Initial versus one-year settlement from (deep) tunnels in London Clay 31
Figure 31 Relative rotation within a tunnel settlement trough 62

page 9 Delete from the list of tables

Table 24 Settlement data from a series of deep running tunnels in London Clay 118

page 29 Delete the whole of the paragraph beginning

`Details of settlements associated ....'.

page 30 Delete Figure 13

pages 60/61 Delete (last nine lines on page 60) from 'The appropriate values' to 'can be
identified' (first two lines on page 61) and replace with Tor further treatment of this
problem, reference should be made to Boscardin and Cording (1992) and New and
O'Reilly (1991)'.

page 62 Delete Figure 31

page 109 Delete paragraph (g) Data of Table 24

page 116 Delete Table 24

In relation to the corrections to Section 4, Risk assessment, i.e. on pages 60 to 62, the two
additional references are:

Boscardin, M.D. and Cording;E.J. (1992)

Building response to excavation-induced settlement .
J. Geotech Engng, ASCE, Vol. 118, No. 4, April, pp 636-637

New, B.M. and O'Reilly, M.P.

Tunnelling-induced ground movements; predicting their magnitude and effects
Invited review paper to 4th International Conference on Ground Movements and Structures. July
91. Cardiff University/ICE, London

FMJ/SLM
23.12.92
 ground movements can then serve as the basis for more detailed assessments of particular
structures, also taking into account ground/structure interaction where this is thought to be
significant, bearing in mind the other uncertainties involved.

The proposed risk classification puts forward various levels of risk associated with
corresponding limiting values of slopes or settlement that might occur. The classification is
intended to cover a broad spectrum of structures with shallow foundations, and also services.
The limiting values of slope or settlement adopted are based upon the assumption that the
structure does not significantly modify the predicted ground response, which is likely to be true
for many buildings commonly found in urban areas. The levels of risk of structural damage
range from Category 1 (negligible) where superficial damage is unlikely through to Category 4
(severe) where structural damage is expected. In the early stages of a project, during alignment
optimisation, the risk classification can thus be established and shown in tramline fashion either
side of on the plan alignment of the tunnels. The report emphasises the need for a staged
approach to risk assessment. Although slope and settlement criteria may provide a valid initial
assessment for alignment studies, it will be necessary in most cases, however, to assess the
likely response for particular structures to relative rotation and horizontal strain and to seek
expert assistance for particular structures requiring detailed assessment.

The report includes a comprehensive review of available precautionary and protective measures
to minimise or eliminate the effects of tunnelling upon structures and services, together with
comments on their range of applicability.

In addition to the extensive list of references cited in the text, the sources of case history data
are presented separately for various types of ground conditions, and a supplementary
bibliography of relevant references is provided.

Prediction and effects of ground movements caused by tunnelling in soft ground beneath
urban areas
Construction Industry Research and Information Association
Research Project 316, 1992

Keywords
Tunnelling, ground movements, deformation of structures, risk assessment

Reader Interest
Tunnelling and structural engineers, building owners, local authorities, public utility engineers.

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by
any means, including photocopying and recording, without the written permission of the copyright
holder, application for which should be addressed to the publisher. Such written permission must
also be obtained before any part of this publication is stored in a retrieval system of any nature.

CLASSIFICATION

AVAILABILITY Restricted

CONTENT Subject area review

CIRIA 1992 STATUS Committee guided

USER Tunnelling, structure and

geotechnical engineers

Published by CIRIA, 6 Storey's Gate, Westminster, London SW1P 3AU

FR/CP/5 3
 Foreword

This report presents the results of a research project carried out for CIRIA's then Advisory
Committee, AC6, Underground Construction. The project was undertaken by Mott, Hay and
Anderson, now Mott MacDonald, under contract to CIRIA. The authors of the report are
Dr L M Lake, Mr W J Rankin and Mr J Hawley of the Foundations and Geotechnics Division
of Mott MacDonald.

Following CIRIA's usual practice, the research was guided by a Steering Group which, during
the course of the project, comprised:

Mr D R Mead (Chairman) London Transport Executive

Professor P B Attewell University of Durham
Mr E R Bewick Vigers
Mr C L Hinsley British Gas
Dr M Howe British Gas
Mr P Hunter British Gas
Dr R J Mali Geotechnical Consulting Group
Dr B M New Transport and Road Research Laboratory
Mr G J Noblett Tarmac Construction
Dr A Owens Stress Engineering Services Ltd
Mr P B Rumsay Water Research Services Ltd
Professor C P Wroth University of Oxford

CIRIA's research managers for this project were Dr R W Poole and Mr F M Jardine.

The project was funded by CIRIA and the following organisations:

Department of the Environment, Construction Directorate,

Department of Transport, through Transport and Road Research Laboratory
Water Research Centre
British Telecom.

CIRIA and the authors are grateful for the help given to this project by the funders, by the
members of the Steering Group, and by the many individuals who were consulted. The authors
particularly acknowledge the advice and valuable comments received from Dr R J Mair of the
Geotechnical Consulting Group, Dr T O'Rourke of Cornell University and their colleague
Mr A Powderham.

4 FR/CP/5
 Contents

List of figures 7
List of tables 9
Notation 10
Glossary of terms 11

1 INTRODUCTION 13
1.1 Background 13
1.2 Terms of reference 13
1.3 Structure of the report 14

2 GROUND DISPLACEMENTS 16
2.1 Introduction 16
2.2 Causes of ground displacement 16
2.3 Extent of initial surface trough 16
23.1 General case 16
2.3.2 Granular soils 22
2.3.3 Catastrophic losses 23
2.4 Magnitude of initial displacement 23
2.5 Time-dependent displacements 27
2.5.1 General 27
2.5.2 Case history data 27
2.6 Initial surface settlement profile 29
2.6.1 Transverse profile 29
2.6.2 Transverse slope and curvature 30
2.6.3 Horizontal displacements 33
2.6.4 Longitudinal profile 34
2.7 Sub-surface displacements 35
2.8 Analytical techniques 39

3 INFLUENCE OF GROUND DISPLACEMENTS ON THE DEFORMATION OF

ADJACENT STRUCTURES 41
3.1 General concepts 41
32 Nature and implications of damage 42
3.2.1 Safety 42
32.2 Architectural or aesthetic damage 42
3.2.3 Functional damage 42
3.2.4 Structural damage 42
3.2.5 Prevention or repair 43
33 Damage criteria for structures 44
3.3.1 Crack width 44
3.3.2 Deformations 45
3.3.3 Horizontal deformation 48
3.4 Damage criteria for services 48

4 RISK ASSESSMENT 50
4.1 Historical 50
4.2 Preliminary appraisal 50
4.3 Sub-surface displacements 61

5 PRECAUTIONARY AND PROTECTIVE MEASURES 64

5.1 Site investigation 64
5.2 Condition surveys 64
53 Planning alternatives 65

FR/CP/5 5
 5.4 Protection of structures 66
5.4.1 General 66
5.4.2 Shielding 66
5.4.3 Compensation grouting 67
5.4.4 Underpinning 68
5.4.5 Jacking 68
5.4.6 Strengthening 69
5.5 Protection of services 69
5.6 Method of tunnel construction 69
5.7 Geotechnical processes 71
5.8 Observation/instrumentation 71
5.8.1 General 71
5.8.2 Measurement of displacement and deformation 72
5.8.3 Measurement of load and strain 74

6 DIRECTION AND NEED FOR FURTHER STUDIES 75

6.1 Case studies 75
6.2 Variations associated with tunnelling methods 75
6.3 Non-linear small strain soil behaviour 75
6.4 Long-term effects 75
6.5 Radial movement 76
6.6 Risk assessment 76
6.7 Horizontal strains 76
6.8 Behaviour of structures 76
6.9 Piles as ground reinforcement 76
6.10 Models 77

References 78

Appendix A Equations for magnitude of settlement on longitudinal profile 85

Appendix B Factors influencing allowable additional deformation 86
Appendix C Classification of visible damage 89
Appendix D Acceptable vertical deflection limits for structural elements 90
Appendix E National Coal Board (1975) classification of subsidence damage 91
Appendix F Damage criteria for services 92
Appendix G Derivation of points of limiting slope and settlement 95
Appendix H Recommended phased stages of site investigation 98
Appendix I Notes on geotechnical processes 99
Appendix J Rational approach to instrumentation after Dunnicliffe (1982 and 1988) 104
Appendix K Case history review, data analysis and summaries 105
Appendix L References for case history data 115
Appendix M Supplementary bibliography 123

6 FR/CP/5
 List of figures

Figure 1 Simplified procedure for risk assessment 14

Figure 2 Three-dimensional shape of surface trough and tunnel co-ordinate system
(after Yeates, 1985) 17
Figure 3 Transverse and longitudinal settlement profiles (after Mair, 1983) 18
Figure 4 Idealised transverse surface settlement profile with normal distribution form
(after O'Reilly and New, 1983) 18
Figure 5 Generation of transverse settlement profile (after Cording and Hansmire, 1975) 19
Figure 6 Settlement trough width characteristics — summary of case history data
(a) i and z,„ (b) i and half trough width 20
Figure 7 Idealised longitudinal surface settlement profile with cumulative normal
distribution form 21
Figure 8 Soil displacements around model tunnels in sand
(a) after Potts (1976), (b) after Cording et al. (1976) 22
Figure 9 Stability ratio versus volume loss for cohesive soils
(a) criteria and mode of behaviour
(b) various suggested relationships
(c) case history data: cohesive soils 25
Figure 10 Variation of load factor with volume loss (after Mair et al., 1982) 26
Figure 11 Comparison of change of distance to point of inflexion between initial and
longer term settlement (cohesive soils) 28
Figure 12 Comparison of change of maximum slope between initial and longer term
settlement (cohesive soils) 29
Figure 13 Initial versus one-year settlement for (deep) tunnels in London Clay 30
Figure 14 Normal distribution function for transverse settlement trough 31
Figure 15 Relationship between maximum settlement and depth for various values of
volume loss 32
Figure 16 Relationship between maximum slope and depth for various values of
volume loss 33
Figure 17 Idealisation of transverse surface displacement and strains shortly
after tunnelling 34
Figure 18 Comparison of empirical relationships, model tests and case history data
with aximum surface and crown settlement, tunnel depth and radius 36
Figure 19 Soil displacements around model tunnel in soft clay (after Kimura and
Mair, 1981) 38
Figure 20 Lateral displacement at axis level adjacent to tunnels in stiff clays 39
Figure 21 Two-dimensional components of deformation (after Geddes, 1984) 41
Figure 22 Proportion of differential settlement affecting cladding and finishes
(after Burland, Broms and de Mello, 1977) 43
Figure 23 Likely behaviour of different types of structures on the limbs of a surface
settlement trough experiencing hogging deformation
(a) framed buildings and reinforced load-bearing walls
(b) unreinforced load-bearing walls 47
Figure 24 Schematic response of various types of structure on the limb of a
settlement trough 51
Figure 25 Idealised deformation of various types of structure on limbs of surface
settlement trough 53
Figure 26 Examples of risk assessment for various tunnelling methods reflected in
different values of volume loss 55
Figure 27 Example of risk zoning at surface for a single tunnel 56
Figure 28 The inter-relationships between depth of tunnel, volume loss and
surface risk zones 57
Figure 29 Example of risk zoning at surface for complex of tunnels 59
Figure 30 Relationship of damage to relative rotation and horizontal ground strain
(after Boscardin and Cording, 1989) 62

FR/CP/5 7
 Figure 31 Relative rotation within a tunnel settlement trough (after Boscardin and
Cording, 1989) 62
Figure 32 Boundary modes of deformation of pipes (after O'Rourke and Trautman, 1982) 92
Figure 33 Case histories — transverse distance versus depth for different soil types
(a) cohesive soils, (b) granular soils, (c) residual soils/silts/fills,
and (d) mixed soils 106
Figure 34 Case histories — maximum settlement and maximum slope trends versus
depth for cohesive soils 107
Figure 35 Relationship between log settlement and transverse distance for
normal distribution 108

8 FR/CP/5
 List of tables

Table 1 Causes of ground displacement 17

Table 2 Summarised surface settlement trough data for a range of UK soils 24
Table 3 Comparison of slope curvatures, displacements and strains over transverse and
longitudinal profiles 37
Table 4 Comparative settlements at the crown and at the surface 44
Table 5 Classification of visible damage to walls 45
Table 6 Allowable total settlements of foundations 46
Table 7 Classification of deformation by different authors 48
Table 8 Relationships between damage due to tunnelling and maximum slope
near edge of settlement trough, for structures experiencing hogging deformation 56
Table 9 Typical values of maximum building slope or settlement for damage risk
assessment 56
Table 10 Action suggested for various risk categories 61
Table 11 Tentative limits of zone of undefined risk around tunnels in different soil types 65
Table 12 Planning Alternatives 68
Table 13 Possible adverse effects of geotechnical processes on ground displacements 73
Table 14 Categories of instruments for measuring displacements 75
Table 15 Typical pipe material properties for short-term static loading in direct tension 95
Table 16 Example 1: Calculation of critical trough width 98
Table 17 Example 2: Calculation of critical trough width 99
Table 18 Factors influencing selection of penetration grouts 101
Table 19 Summary of case history factor: cohesive soils 112
Table 20 Summary of case history factor: granular soils 114
Table 21 Summary of case history factor: residual soils, silts and fills 115
Table 22 Summary of case history factor: mixed faces 116
Table 23 Summary of case history factor: twin tunnels in cohesive soils 117
Table 24 Settlement data from a series of deep running tunnels in London Clay 118
Table 25 Observed data on sub-surface horizontal movements at tunnel axis level 119

FRCP/5 9
 Notation

Symbol Description

C cover to tunnel crown

c„ undrained shear strength
d distortion factor
D excavated tunnel diameter
D10 effective grain size
E Young's modulus
F factor of safety (inverse of load factor)
G Shear modulus
building height
trough width parameter (to inflection point)
coefficient of earth pressure at rest
K constant (as in i = Kz.)
k constant (as in i = kr (zi2r)°)
L overall building size
1 spacing for structure or foundation grid
N stability ratio
index constant (as in i = kr (z(, 2r)°)
P shield length
✓ tunnel excavated radius
s limit of integration of standardised normal distribution
T width of tail gap around shield
t variable in derivative of standardised normal distribution
u longitudinal forward displacement
✓ volume loss in tunnel or volume of trough
Vr percentage volume loss per metre
✓ lateral displacement
vmu maximum horizontal displacement
w vertical displacement
w, initial surface settlement
x distance ahead of tunnel face
xo point above tunnel face
y transverse distance from tunnel axis
z depth below ground surface
depth to tunnel axis

a angular strain
B relative rotation (angular distortion)
7 unit weight of soil
relative deflection
8„ differential settlement
e average strain
building slope
Og maximum ground slope in surface settlement trough
at tunnel support pressure
total overburden pressure
standardised normal distribution
effective angle of internal friction
X angle of draw
SZ overall tilt of building

10 F R/C P/5
 Glossary of terms

average strain (e) Ratio of increase in length to length of the structure (a negative
sign indicates shortening).

angular distortion (B) Rotation of a particular member relative to the structure as a

whole.

angular strain (a) The difference in rotation of two adjacent parts of a structure
(sagging positive, hogging negative).

axis Centre line of tunnel.

deflection ratio (e/L) Ratio of relative deflection to the size of the structure.

deformation Movements/distortions of or within a structure or service.

differential settlement (8„,) Difference in settlement (or heave) of two points.

displacement Movement of the ground in any direction, settlement being a

special case.

extrados External surface of tunnel lining.

intrados Internal surface of tunnel lining.

`locked in' stress Total stresses imparted to a pipe as a result of fabrication and
installation.

building slope (0,,) The maximum slope to which a structure may be subjected,
derived from the maximum differential settlement that occurs
anywhere within the plan area of the structure.

relative deflection (e) Deflection of part of a structure relative to the structure as a

whole.

rotation Amount a particular member rotates due to settlement.

settlement(w) Vertical displacement downwards (a negative sign indicates

heave).

soft ground Soft to hard clays, silts, sands, gravels, residual soils, fills and
very weak rocks.

structure All man-made structures including services and pipelines.

tilt (fl) Average or rigid body rotation of the structure.

F R/CP/5 11
 1 Introduction

1.1 BACKGROUND

Tunnel construction in urban areas is likely to increase in the long term as the need for more
direct and environmentally acceptable transportation corridors expands. Such underground
construction in soft ground which in this context covers the spectrum from very weak rocks to
soft clays, can cause initial modes and rates of deformation different to those normally
experienced by structures responding to ground displacements caused by their self weight and
imposed structural loading. These initial additional strains, which may occur relatively rapidly,
are superimposed upon the existing strains within the structure or service and could produce
effects out of proportion to their magnitude.

The spatial extent and magnitude of the initial surface settlement profile caused by tunnelling
can be predicted with a reasonable degree of confidence for 'green field' site conditions, which
in this context means undisturbed by urban fabric and where structures do not interact to
modify the profile. These relatively simple predictions are mainly based upon empirical
correlations derived from field observations. The presence of existing and past building
foundations and services will interact with the settlement trough and is likely to modify the
pattern and magnitude of the ground displacements. These effects are not well understood and
there are still very few well documented case records where both the ground and structure have
been appropriately monitored before, during and after tunnelling. Consequently, there are
limited data to support even empirical correlations in this area. To improve the present state of
knowledge, the existing database urgently needs to be expanded, and further research conducted.

Recent years have seen a growing awareness of the need to understand the interaction between
structures/services and the ground which supports or hosts them. The theoretical relationships
for materials with idealised properties are themselves complex, but add the many variables
introduced by inherent variations in ground properties, stress history, construction methods and
workmanship, then the complexity is increased many fold.

Attempts to analyse and model the displacements, associated with tunnelling using elastic
solutions and computer finite element methods have, to date, been relatively unsuccessful.
Commonly the models adopted are simplified to two dimensional plane strain and the results
obtained do not match experience from case records. It is considered that the three dimensional
effects around the advancing face dominate. The prediction of ground displacements is likely to
continue to rely predominantly upon case history information for various tunnelling techniques
in a variety of ground conditions.

1.2 TERMS OF REFERENCE

The Terms of Reference set by the Steering Group reflect long-term objectives and are as
follows:

(a) To provide guidance on estimating and monitoring the effects of underground construction
on nearby structures and services.

(b) To provide a framework for a general risk classification for the potential level of damage
and hence the appropriate level of investigation, analysis and protection measures
appropriate to the structures or services.

(c) To indicate that the effects on nearby structures and services need not be a barrier to
underground construction in urban areas, provided that appropriate precautions are taken.

In order to address these objectives this research project was divided into two phases. The first
phase was directed at reviewing the available literature and towards assembly of an adequate

FRCP/5 Previous page 13

is blank
 data base of records of ground movement associated with tunnelling in a variety of ground
conditions. This was achieved and the second phase was commissioned to develop a rational
basis for prediction of ground movements and to establish a framework for a general risk
classification which would be used in preliminary assessment of structures and services above
the proposed tunnels.

It is recognised by the Steering Grouop that the terms of reference are extensive and during the
production of this report it was agreed that the scope of the report could not realistically include
forms of underground construction other than soft ground tunnelling, nor could it adequately
cover detailed structural assessment. In addition it was recognised that this report may require
significant revision and updating as new and more detailed case history data become available.

1.3 STRUCTURE OF THE REPORT

The report has been structured so as to follow the sequence of actions undertaken when
assessing the effects of urban tunnelling in soft ground upon structures and services in the
vicinity. The logic diagram, Figure 1, is intended to provide the reader with guidance on the
manner and sequence in which this document can be used to assist engineers with the
assessment of the potential risk to structures and services.

Location in Text

Stages in assessment

Phased investigation/obtain sufficient information on ground conditions,

and position of the water table

1st part of Estimate likely zone of displacement in absence

Chapter 2 of urban fabric in relation to tunnel(s) depth,
diameter and initial plan and profile alignment

Is structure within or NO: END OF

linked to zone J a PROBLEM
1
YES

2nd part of Estimate magnitude of displacements

Chapter 2 and ground slope assuming no
modifying interactive effects

Compare predictions with case histories

and local precedent experience.
Modify predictions if appropriate

Chapter 4 Preliminary risk assessment based on ground

slope and settlement, zone structures
with near surface foundations into various
categories of risk

Optimise plan and profile alignment to reduce risk,

yet meet other operational or fixed criteria,
take into account any especially important or
sensitive structures

Not covered in Second stage evaluation of

detail in this structures at risk, review
report potential modes of structural
behaviour

Detailed evaluation of
Particular structures

I Condition Survey of structures at risk

Chapter 5 Protective or Precautionary Monitor and observe

Measures introduce contingency measures
if required

I Remedial Measures

Figure 1 Simplified procedure for risk assessment

14 FRCP/5
 Section 2: Describes how the extent of the surface settlement trough can be predicted and
then how the magnitudes of displacements, ground slopes and strains can be
estimated. Potential variations resulting from soil type, workmanship and
position of water table are discussed. The evidence from case histories is
summarised. The current limitations and potential for use of analytical
techniques are reviewed.

Section 3: Examines the susceptibility of structures and services to damage by

displacement and deformations, and discusses damage criteria and assessment.

Section 4: Assesses the degree of risk of damage associated with simply predicted
maximum ground slope and settlement criteria which rely only on knowledge
of the depth, shape and position of the surface settlement trough in relation to
the position of the structure or service. A tentative classification is provided
which can be used during preliminary design for planning and optimisation of
tunnel alignments. Supplementary classifications which address the important
influence of horizontal strain and angular distortion which can be used for
second stage evaluation are discussed.

Section 5: Describes the range of precautionary and protective measures available to

safeguard structures and services at risk. The options open at design and
construction stages to overcome potential or encountered difficulties are
outlined.

Section 6: Summarises the perceived needs for further studies to advance the state of
knowledge on tunnelling induced settlement and its effects.

Additional and supplementary material is provided in a series of appendices. An extensive list

of references including key sources is given together with a cross-referenced lists of case
histories used in the preparation of data plots for different soil types and an extended
bibliography.

The notation and a short glossary of terms used throughout this report, immediately follows the
list of contents. Attention is particularly drawn to the following terms, which are used
specifically in the senses defined, to avoid ambiguity and needless qualifying statements in the
text. They are:

`displacement' referring only to ground movements in any direction

`settlement' representing the vertical component of ground displacement

`deformation' referring to movement of or within structures or services,

`structure' subsequently covering all man-made structures, services and pipelines.

This report is not a substitute for sound engineering judgement nor can it replace experience
and understanding of geological structure and soil behaviour, the design and execution of
underground works and the response of structures to deformation and distortion. It should
enable competent engineers to ascertain the potential level of risk to apply to specific structures
and to recognise where it is appropriate to obtain expert assistance.

FR/CP/5 15
 2 Ground displacements

2.1 INTRODUCTION

The potential magnitude of ground displacements depends closely on the geological setting and
the nature of the ground surrounding the tunnel. In order to assess the potential for damage to
structures caused by tunnelling, it is necessary first to predict the zone of displacements,
,

secondly estimate the magnitude of soil displacements produced by the works in that zone and
finally indicate how these influence and may be modified by, the presence of structures.

The prediction of initial surface settlements caused by tunnelling in soft ground has been well
researched and reasonably accurate, simple, empirical formulae, developed from case history
data, are available. The prediction of long-term surface settlements and of sub-surface
displacements, is still uncertain and the few available observations highlight the variability in
actual experience, in particular with relation to soil type and tunnelling method.

The prediction methods discussed in this document are derived from specific countries or
localities, geology or ground conditions, employing specific control and supervisory practices.
Therefore, they may not be universally acceptable. Further, the data and hence the predictions
relate to older established tunnelling methods and machines and may not reflect the latest
techniques. Earth pressure balance machines and incremental support techniques such as the
New Austrian Tunnelling Method (NATM) are being more widely used and the case history
records associated with these different techniques are currently filtering into the technical press.
Indications are that the prediction methods described in this report continue to apply, but it may
be some years before the validity of these can be fully established for recently introduced soft
ground tunnelling techniques.

2.2 CAUSES OF GROUND DISPLACEMENT

Two phases of displacement arising from tunnelling in soils are recognised, (see Table 1). The
almost immediate phase which accompanies excavation and tunnel construction, which has
usually been considered dominant, is followed by a post construction phase of time-dependent
displacements. The latter phase is primarily of consolidation in clay soils following changes in
pore water pressure, and in other soils may result from recompaction and other factors. Whereas
the immediate movements can be predicted with reasonable confidence, this is not the case for
the subsequent time-dependent movements.

2.3 EXTENT OF INITIAL SURFACE TROUGH

2.3.1 General case

An initial surface trough develops to the front and sides of the advancing tunnel face but it is
not fully developed until some time after the passage of the working face. A diagrammatic
sketch of the three dimensional shape of this trough is given in Figure 2 with simplified
transverse and longitudinal settlement profiles in Figure 3. Both vertical and horizontal
displacements occur and the notation used to denote these displacements relative to the tunnel is
indicated on Figure 4.

16 FR/CP/5
 Table 1 Causes of ground displacement

Portion of total Nature of loss Cause of loss

displacement

Initial Lou of material into the face elastic and/or plastic deformations and tuns or flows
of soil

Lou between the face and leading Poling plates, over-cutten,

edge of lining (over the shield if or bead on shield
employed)

Over-excavation, ploughing, yawing or negotiating

curves

Pushing aside boulders

Build up of grout on shield tailskin

Lou during or after erection of lining When soil void not completely filled
(at tail of shield if employed)

Delays in erection of lining or in grouting

Time-dependent Loss with time as heading advances Void collapse

Lining deflection

Additional loss with time Compaction of soil

Consolidation of soil (due to long-term pore water

changes)

Y(tranverse
distance)
(depth to zo
tunnel axis

crown
shoulder
axis level
knee
invert
Tunnel

Figure 2 Three-dimensional shape of surface trough and tunnel co-ordinate system

(after Yeates, 1985)

FR/CP/5 17
 •

Settlement trough length

Surcharge pressure
Settlement profile
vavaveAv vAvAvAvAuvAvAv yam too Original
x= 0 4— ground level
Wrnart
Tunnel radius, r

r
2r = D Tunnel axis

a, = support pressure p Tunnel lining

(a) Longitudinal settlement profile

I. I Settlement trough width )

Original
Settlement profile ground level
C

(b) Transverse settlement profile

Figure 3 Transverse and longitudinal settlement profiles (after Mair, 1983)

half trough width

transverse distance from centreline

0 S
settlement volume (per unit advance) 02 if
maximum curvature 'hogginglir- I f"'"'.
-- VE V2nly wmax
d2w wmax 04
=0446 j yt
dy ` WSWmax eXa (-Y1/21y1) O
06
.11
*maximum horizontal strain (tensile)
0.8 0
dv , wmax
z) .0-446 zo
dy ty, wmax
average slope
point of inflexion(ysi,w.0-607wmax) wmax
dw wmax
maximum slope. —.0•607 3i
dy
*maximum horizontal displacement
Ft(hi) .0-6074-0 wmogym

maximum curvature sagging

d2w = wmax
Note,
horizontal displacements and de i y2
strains assume radial ground
maximum horizontal strain (compressive)
movements towards tunnel axis dv(y,z) = wmax(y,z)
dy zip
*not applicable to granular soils

tunnel axis level

Figure 4 Idealised transverse surface settlement profile with normal distribution form
(after O'Reilly and New, 1983)

18 F R/C P/5
 The general case considered in this section relates essentially to the development of the initial
trough when tunnelling in cohesive soils, any differences for granular soils being covered in
Section 2.3.2.

(a) Trough width/transverse profile

The cross-sectional shape of most surface settlement troughs correspond approximately to an

inverted normal (Gaussian) distribution form, as proposed by Peck (1969) (see Figure 4). Figure
4 also shows the key parameters, including horizontal displacements and strains assuming radial
ground movement towards the tunnel axis, as proposed by O'Reilly and New (1983). The
trough width can be represented by an empirical equation derived by Schmidt (1969).

21 z eqn 2.1
kko
D

where 1., = distance from tunnel centre-line to point of inflexion

D = diameter of excavation
zt, = depth from surface to tunnel axis

Peck (1969b) indicated that the value of n is normally around 0.8 and that k can vary from less
than one for hard clays and sands above the water table to more than one for sands below the
water table. A value of k =1 is often taken for soft to stiff clays.

An alternative estimate of trough width can be made using the approach proposed by Cording
and Hansmire (1975) in which the settlement trough is simplified to a triangular form as shown
in Figure 5, the width being related to the angle of draw (x), where x = 45°4/2. This approach
has been further developed by Attewell (1977).

2T= 2 (r sec ik+(C+r) tan IP

)1.

C
zo C+r ( l+sin ifr)ltan
r cos 17//
A
V
D A
r
V

Figure 5 Generation of transverse settlement profile (after Cording and Hansmire, 1975)

FR/CP/5 19
 The two methods, by Schmidt (eqn 2.1) and Cording and Hansmire, generate similar settlement
trough widths for a given depth to tunnel radius (z• r) ratio.

O'Reilly and New (1983) suggest that for most purposes:

iy • eqn 2.2

where K depends upon the soil type, varying from 0.4 for stiff clays to 0.7 for soft silty clays,
and that for practical purposes the total width of the trough can be taken as 6i, or approximately
equal to three times the depth to the tunnel axis.

The case history data which are presented in Appendix K and summarised in Figure 6 indicate
that the value of iy, for the surface trough for practical purposes appears to vary linearly with
tunnel axis depth. The assumption K = 03 appears reasonable for the clay soils, with values
K = 0.4 to 03 for the narrower trough widths associated with granular soils. At depths
exceeding 20 m the value of iy, is probably smaller than might be predicted using these
relationships. A best fit to all the results on Figure 6(a) appears to be K = 0.5. Field
measurements also show that the width of the trough is generally independent of the degree of
support within the tunnel and therefore independent of the tunnel construction technique.

transverse distance to point of inflexion, i Cm)

10

KEY
0 • Granular sods
x Cohesive soils
A Residual soils/silts and fills
o Mixed soils
KEY 0
• Granular soils x

x Cohesive soils
0
• Residual soils/silts and fills
a Mixed soils

x
x o
x y__cil • • half trough width
x/ Lr equal to Sty
• 06

•
general range of data
/1

•• r
•X x

10 20 30
discernible half trough width(m)

(b) (a)

Figure 6 Settlement trough width characteristics - summary of case history data

(a) i and zo, (b) i and half trough width

20 FR/CP/5
 For a complex of tunnels in stiff clays, a reasonable initial prediction of settlement profile can
be achieved by adding the individual settlement troughs. This method is not valid for tunnels in
soft to firm clays or granular soil, where the ground can be weakened significantly by the
passage of the first tunnel.

(b) Trough length/longitudinal profile

The idealised shape of the vertical displacement profile, parallel to the tunnel axis, is shown
theoretically in Figure 7. The first signs of displacement at the surface occur at a distance of
about one or two times the depth of the tunnel ahead of the face; a small heave has sometimes
been noted at this extremity. The initial displacement is normally 80% to 90% complete at a
similar distance behind the face.

Longitudinal distance Longitudinal distance

-x x
behind tunnel face ahead of tunnel face

face position
x=0 ix
;

x
0
(x =i w = 0.159 w max)
Maximum curvature (hogging)
d2 wmax
•
0.242 ,
d xw -

(x. - i w= 0.841 w max) *Maximum horizontal strain

Maximum curvature sagging (tensile)
d 2 w 0.242 wmax du wmax
. , = 0.242
dx 2 - l x- dx zo
*Maximum longitudinal strain Point of inflexion (x =0, w = 0 5 w max)
(compressive) dw wmax
Maximum slope — = 0.399
du = 0.242 w max d
dx zip *Maximum horizontal displacement

u =0.399 w max lx
zo

Direction of drive *Horizontal displacements and

strains assume radial
ground movements towards
Tunnel axis level 1 0
tunnel face

Figure 7 Idealised longitudinal surface settlement profile with cumulative normal distribution form

FR/CP/5 21
 The equations for the solution of the longitudinal profile, given in Appendix A, assume a
cumulative normal distribution function. This has been shown by Attewell and Woodman
(1982) to be consistent with the assumption of an inverted normal distribution for the settlement
trough cross-section.

Attewell and Hurrell (1985) have tested the assumption that the trough length parameter (i,) has
the same value as the trough width parameter (9, i = ix = iy, using a number of case histories
where access was available to the original detailed site measurements. Their comparison
indicates that in practice the profile ahead of the advancing tunnelling face is slightly flatter
than would be predicted, but the profile behind the face is substantially flatter with the point
x = 0 (i.e. where w = 0.5 w,„„,) representing a point between the face and the tail (for a shield
driven tunnel). The implication is that for most practical design cases the assumption of i„ = iy,
is reasonable, and results in predictions of slope or curvature parallel to the tunnel axis which
will tend to slightly over-estimate those that would occur in practice.

The concept of angle of draw has also been used by Cording and Hansmire (1975) for
prediction of development of subsidence ahead of the face. This relationship can be compared
with the approximate solution for the trough length being equal to 2.5i ahead of the tunnel and
for practical purposes the predicted longitudinal settlement trough lengths are similar.

Whereas in the transverse trough, settlement is less than 1% of the maximum at distances
greater than about 3i from the trough centre, the corresponding value for the longitudinal profile
occurs about 2.5i ahead of the tunnel face. This assumes that the trough length parameter (1„)
has the same value as the trough width parameter (ii).

2.3.2 Granular soils

The previous sections relate essentially to the behaviour of cohesive soils. To estimate the initial
settlement in granular soils, it is normally assumed that the shape of the surface settlement
trough also approximates to an inverted normal distribution curve. In fact Schmidt's original
stochastic analysis (1969) models a granular rather than a cohesive material. Frequently, a
surface settlement trough is observed with the inverted normal distribution form, but with a
central deepened zone of irregular form. This is attributed to loosening of the soil or ground
losses concentrated over the tunnel crown.

Other model studies and case records for granular soil indicate that particle displacements often
are not radial towards the tunnel axis, but frequently show a relatively narrow zone with
predominantly vertical displacements above the tunnel with inward (lateral) displacement largely
restricted to near the ground surface, see Figure 8

Figure 8 Soil displacements around model tunnels in sand

(a) after Potts (1976), (b) after Cording et al. (1976)

22 FR/CP/5
 The position of the tunnel with respect to the water table is of considerable significance within
granular soils as indicated by Peck (1969b) and will often determine the method of construction,
extent of pre-drainage and type of ground treatment. The magnitude of displacements in
granular soils is very dependent on the method of tunnelling and standard of workmanship.
Predictions can only be made if it is assumed that the ground is controlled at all times and that
no gross ingress of ground and water occurs.

O'Reilly and New (1983) have indicated that a linear relationship between i and zoalso exists in
granular soils. However their dnta are limited to tunnels above the water table and at depths of
less than 10 m for which they suggest that K varies between 0.2 and 0.3. For tunnels below the
water table, K is typically in a similar range to that for tunnels in clay.

2.3.3 Catastrophic losses

In underground construction there is always a possibility of catastrophic failure resulting from

large ground losses either at the face or through the ground sump ports; such an example is
given by Heuer (1976). These are characterised by singular, large, sudden and unrestrained
ground movements. Unforeseen geological conditions such as buried channels, fault zones,
former construction and water-bearing ground are common hazards. An important element in
design and construction is the consideration and provision of contingency measures, whereby
potential incidents can be averted or minimised, allowing the situation to be recovered with
maximum efficiency.

2.4 MAGNITUDE OF INITIAL DISPLACEMENT

In cohesive soils it is generally assumed that the volume of ground loss into the tunnel (i.e. in
excess of the notional excavated volume) equals the volume of the settlement trough per unit
length. This may not be valid for granular soils or stiff fissured clays subject to dilation
(bulking). The volume of the settlement trough per unit length is obtained by integration of the
area bounded by the transverse settlement trough, given approximately by
V = (27t)si .iy.w 2.5iyw eqn 2.3

and where Viis the percentage volume lost,

2 eqn 2.4
= 0.0125 r

Volume loss for even the most carefully controlled tunnelling operation in soft clays is rarely
less than about 2-3% of the notional tunnel volume, although with earth pressure balance (EPB)
shields this may reduce to less than 2%. In stiff London Clay with good workmanship at
cover/diameter CID ratios greater than 5, the initial volume loss varies little, falling in the range
1-2% with lower values achieved using special measures such as additional face support,
reduced beads, continuous working and grouting close to the shield. Further, the available
evidence produced by Craig (1975) suggests that in stiff clays different types of segmental
lining (bolted or expanded) have no significant influence on the amount of settlement.

In granular soils the volume loss determined from the settlement trough usually appears to be
less than that determined from the face yield. This is consistent with dilation or bulking of the
soil occurring above the tunnel.

There are indications that the percentage volume loss is higher for smaller diameter shield
driven tunnels, probably due to the sensibly constant clearance provided around the shields.
However, the total loss and the effects of displacement associated with a smaller tunnel would
usually still be less than those for a larger diameter.

FR/CP/5 23
 Use of the observational tunnelling techniques which apply incremental support known as the
New Austrian Tunnelling Method (NATM) has been extended to soft ground conditions in
Europe, South America and the Far East. The essence of the method is control over the
redistribution of stress in the ground as the tunnel advances by adjusting the construction
sequence, the stiffness of the primary support (shotatte) and the timing of its installation.
In weaker materials the method may assist through the early application of ground support over
the excavated section. Performance is monitored by routinely measuring convergence and
settlement within the tunnel, and settlement at the surface above the tunnel, and in some cases
by more extensive instrumentation including radial ground pressure, hoop stress in the lining,
piezometers, and multi-point extensometers. Experience in the Far East (Cater and
Shirlaw, 1985; Hulme et al., 1989) has shown significantly greater ground displacements
associated with these techniques than with similar shield-driven tunnels. However, in Europe
there is evidence (Muller-Salzburg et al. 1977; Schultz, 1975); Brem, 1981; Deix and
Gebeshuber, 1987) to suggest that in firm to stiff clays surface settlements are no greater than
might be predicted for more established soft-ground tunnelling methods. In granular materials
above the water table, very small ground movements occur in general as with most techniques,
but occasional catastrophic failures have been recorded. The observational approach and
flexibility of the technique allow optimisation to suit the prevailing conditions. The size of the
initial heading, control of the face and in particular the timing of invert closure have been
shown to be key parameters in minimising ground movements.

Table 2 summarises typical trough widths and volume losses for a range of soil types and
tunnelling methods, based on UK experience. However, much of this data depends upon surface
observations which are readily distorted or modified by the presence of road and pedestrian
pavements, shallow structures and utilities. Thus these values should be used with caution.

Table 2 Summarised surface settlement trough data for a range of UK soils (provided CID>1),
after O'Reilly and New (1982) and Yeates (1984)

Ground Tunnelling Trough width Volume loss Remarks

conditions method parameter Vi (%)
constant, K

Stiff fissured clay Shield or none 0.4-0.5 1/2-3 Considerable data

available, losses
normally 1-2%

Glacial deposits Shield in free air 0.5-0.6 2-21/2

Shield in 1-1V, 'Used to assist in

compressed air controlling ground
movements

Recent silty clay Shield in 0.6-0.7 2-10

deposits compressed air
(c,F10-40 kPa)

Granular material 0.2-0.3 1-5

above
the water table

Granular material Compressed 0.4-0.5 1-10

below the water air/slurry/EPB
table

With suitable engineering experience and using Table 2 as a guide, the likely volume loss
appropriate to the soil conditions, position of water table and tunnelling method can be
estimated. In addition, the effects of possible variations in K and VIcan be easily assessed on
the range of potential initial maximum surface settlements evaluated for conditions that ignore
the interaction and modifying effects of structures.

24 FR/CP/5
 For clay soils, knowledge of the stability ratio, N, is also helpful in assessing the face stability.
This has been defined by Broms and Bennermark (1967) for clays as:

N - (a7 at) eqn 2.5

c•

where cr„ = total overburden pressure at the tunnel axis (including any surcharge)
a, = tunnel support pressure (if any)
= undrained shear strength of the clay

In general, where N is greater than 6 there is liable to be general face instability, between 4 and
6 plastic yielding is likely, and localised limited plastic yielding between 2 and 4. For N less
than 2, the response is likely to be essentially elastic and the face stable. The mode of
behaviour related to stability number is shown in Figure 9(a).

a) Limiting criteria and mode of behaviour. b) Various suggested relationships.

7.0 L1.417.40, FAIL004

iluctize1-1.. (1985)

4
ikami 1.60-weJ60mAatc. (196.7)

MA12 (I 83
440

4.0
V6,1.3514-14
4Los3o.P(1974) 1.
4.4c er_..s 444 4, 4i
U.45'rA 41t
31
THE.082.710.41.12494,464 Qs
Umecoratcrco Loses
E3st..1TIALLy ELASTIC. LIA0I3 T Scamscer 0949)
v/.401, t eewoo-e. (1959 3
2 6. 8 to 12 14 14 if 20 0 2 4 O s to 12 '14 14 Is to
INITIAL VOLUME LOSS \PC. INITIAL VoLuA4E Loss VC (41/o)

c) Case history data cohesive soils.

7 14, • 14
40-
• 96
• le
• 13
■
•
•9
4•0- 414..
• 1r Pu4-15142.0 CAS.
41s-ratty 26.F.
e25
l. 54a T,46LE K .1.
*
2.0 •3
• IA 117
• C.

0 2 4 4 it to t1 1.4 K. If lo
iparnAL Vou).44. Loss Vt (0/..)

Figure 9 Stability ratio versus volume loss for cohesive soils

(a) criteria and mode of behaviour
(b) various suggested relationships
(c) case history data: cohesive soils

FR/CP/5 25
 Kimura and Main (1981) demonstrated with model tests in kaolin clay the dependence of
stability upon tunnel heading geometry and depth below the ground surface. They showed that
the critical values of stability ratio N could vary between 5 to 9 for many practical tunnelling
geometries. It should be noted that their investigations concentrated on relatively shallow
tunnels with CID ratios of less than 4. Ward and Pender (1981) claimed that for a deep tunnel,
stability ratios of between 1 and 2 are required to ensure that only small ground movements
occur and movements are strictly only elastic at stability ratios of less than one. Small
movements can be achieved in free air for stiff clays or with high support pressures in soft
clays. O'Reilly (1988) has shown that good agreement between predicted and measured values
of volume loss can be obtained in London Clay provided that the stability number is assessed
using the bulk shear strength of the clay. On this basis he indicates that small settlements can
be obtained provided that the stability number is below four in London Clay. These stability
criteria are strictly for undrained conditions and take no account of possible reduction in shear
strength of the exposed ground due to time-dependent softening or sensitivity. Additionally, the
relationships proposed by Schmidt (1969), Glossop (1978) and Hurrell (1985) are shown on
Figure 9(b) which. assumes:

(a) the tunnel is two dimensional (thus three-dimensional effect of tunnel heading is neglected);

(b) tunnel sufficiently deep that any variation of gravitational stress from crown to invert is
negligible;

(c) prior to tunnelling K. = 1;

(d) soil is elastic — perfectly plastic, i.e. no strain hardening or softening in the plastic zone.

Stability ratio versus volume loss is plotted in Figure 9(c) for published case history data in
cohesive soils, taken from Table 19 in Appendix K. Most results conform to the expected trend
except case history 17, where a high volume loss was experienced, which is attributed to poor
workmanship and unfavourable soil fabric. This example may indicate that published case
histories are selective, reporting being limited to the more favourable results. However, the plot
should provide a guide to the volume loss in cohesive soils that might be anticipated for a
known stability ratio and good workmanship.

Another useful concept generated by the model tests and finite element analyses carried out by
Kimura and Mair (1981) and Mair et al. (1982) is that of load factor, the reciprocal of factor of
safety defined as the ratio of stability number at working conditions to stability ratio for
collapse. Their results are plotted in Figure 10 and suggest that to achieve a volume loss not
exceeding 4%, the mean factor of safety against tunnel face collapse should be at least 1.5.
These results are broadly consistent with field observations but direct application of the results
should be undertaken with caution because ground loss is so dependent upon working methods
and workmanship, and the concept does not take account of the three-dimensional effect of the
tunnel heading.
I•O

../
.., ...., ----
, ..-- 1Z/104410s eeSuLTS ritom XPermi.
/ ../e
/ T1111'116 Oa Seri' WAes.14 Guy
,
../ MEA4 A.60 F.orrs tlILMOIT AaALysto
0
.
r
LL LOAD rAGT012 I • ■
3 (W•le14.)41 C•.forne.$)
0 )4 (AT Cou-Ae$4.)
1
0 04

F. 1.6 AT VOW...it 1.0.65 = 4%

0 2. 4 • le 12 14 Ir IS 2# as
VoLumg. L...ss (V.)

Figure 10 Variation of load factor with volume loss (after Mair of al., 1982)

26 FRCP/5
 2.5 TIME-DEPENDENT DISPLACEMENTS

2.5.1 General

It has often been assumed that all settlement effects are complete within a few weeks of the
tunnel face passing any reference point on the surface, such that the zone of immediate
influence has advanced beyond that point at which settlement is being considered. Such an
assumption is supported by the fact that complaints of settlement-caused damage to buildings
generally relate to this period. However, it is clear from field measurements that, in at least
some cases of tunnels in clay, settlement continues for long periods, perhaps years, after
completion. It is possible that this occurs in the majority of cases for tunnels in clay.

The construction of a tunnel in soils will induce changes in the porewater pressure distribution,
the effect of which is to change the effective soil stresses thereby causing time dependent
volume changes. In the permanent condition it is seldom practicable to make a tunnel watertight
and the tunnel may act as a deep drain. The resultant radial seepage flow will in turn increase
the effective stresses in the surrounding ground and in cohesive soils induce consolidation
settlement. To activate this process only drips or a slight seepage of water into a tunnel from a
low permeability clay would be required. Ward and Pender (1981) have verified that tunnels in
London Clay have behaved in this manner.

The evidence for tunnels in clay suggests that, during consolidation, the displacements are
predominantly vertical and distributed over a wider zone above the tunnel and hence further
significant lateral displacements and strains are unlikely and structures are more likely to exhibit
free body movement. The corresponding lack of complaints or records of damage in London in
the long term suggest that the effects of the long-term component of settlement are less
damaging than those associated with the initial phase of movement, although this may be
related to relatively deep tunnels. More records are required to examine differences between the
initial trough shape and that in the long term so that these additional time-dependent effects on
structures can be assessed.

The potential effects of long-term settlement associated with relatively shallow tunnels are
likely to be more damaging due to the rate of movement and more limited zone of movement
above the tunnel. The rate of consolidation of the clay is likely to be increased due to shorter
drainage paths and possibly by higher coefficients of consolidation as a result of fissuring near
the surface. The magnitude of consolidation settlement is unlikely to be significantly influenced
by the diameter of the tunnel, although for small tunnels the long-term consolidation settlement
may be of a similar magnitude to the initial settlement.

Methods of estimating the settlement caused by groundwater lowering and soil consolidation
using flow nets have been described by Glossop (1978), Howland (1980) and Fitzpatrick et al.
(1981). These methods can take account of the effects of compressed air working. During
compressed air working in water-bearing soils, the water pressure is commonly balanced
between knee and axis level, permitting some seepage in the invert. Case records have shown,
for example Glossop et al. (1979), that such seepage and consequential drawdown may have
initiated early consolidation settlement. It is apparent from a number of case histories of
tunnelling in Hong Kong, such as those described by Cater et al. (1984), that significant
displacements can be associated with dewatering, and this component needs to be considered
separately from those directly associated with the tunnelling.

It has been suggested that compaction or recompaction of granular soils can be.a major
component of surface settlement (Cording et al. 1976 Howland, 1981). Such compaction or
recompaction of granular soils may take place as a result of construction and/or traffic
vibrations as well as seepage or stress readjustments.

2.5.2 Case history data

Long-term, time-dependent settlements have been known to occur in association with tunnels in
London Clay. However, there are very few case history records to assist in an empirical

FR/CP/5 27
 assessment of the magnitude and effects of this phenomenon. The consolidation due to tunnels
acting as drains is influenced by the potential for recharge from the Thames Gravels above and
the degree of underdrainage as a result of previous groundwater abstraction in strata below the
base of the London Clay. The groundwater depression in Central London is continuing to lead
to consolidation settlement of the London Clay on a regional scale although groundwater levels
are starting to recover at depth. It is widely believed that for many of the earlier underground
railway tunnels in London, typically at depths of around 30 m below ground level, the
magnitude of long term settlements occurring more than a year after construction was relatively
small, diffused over a large area at the ground surface and for practical purposes complete. For
deep tunnels further long-term settlement may have been sufficiently widespread to allow free
body displacement of structures, resulting in comparable effects to those caused by other
seasonal or regional movements and/or occurring at rates that may be accommodated by the
existing structures.

The case history data for the initial and longer term changes in the transverse distance to the
point of inflexion (iy), and the change in the value of maximum ground slope (0s) for soft to
firm clays are compared in Figures 11 and 12 respectively. The data points are mostly within
the range 1.0-1.5 for the ratio of final to initial transverse trough width parameter (iy) and
between 1.0 and 2.0 for the ratio of final to initial transverse maximum slope (ed • As the time-
dependent settlement has usually been recorded within one year of tunnel construction the data
records are probably incomplete and therefore likely to represent a lower-bound value. Any
increase in maximum slope over the long term may be of significance to property damage. At
present there are limited data available on this effect.

., is- 1.4,n.AL
of
/
*- /
we
I / . 16 i Ch,gi5
. 1A;1
iarr
S 10- /
/
4 /•IA /..--
/
I
I- /
9 •Jo /.1'v
7
3 / v
/v
6- 5 "/
t 9c v"
`) /. ,tI
4
1--
.,-
/004A96
J /v4 b
4 7,
fr. 7
.2 0 ..**
15
F1.4,e1L. Tib20414 VhoT&I PARANCrtia

Figure 11 Comparison of change of distance to point of inflexion between initial and longer term
settlement (cohesive soils)

28 FRCP/5
 1•-
ri•JAL $51.4rTIJA. 9

1.7 FIAAL
/
c I-5 lyTIAL9
/
/ 98 9e. 7
• • ,...-
/ 7
/
/ •IC V ..-
/ 7
9
/ "" A 7- ---F"0"- 0
/ „- .7 11 ..-- *". .2.o LirrrAt.e

I
4 0.2 0.4 0.0 0-6 1.0 1.1 1.4 1.40
MAgIMuK SL4pm 9 (•)

Figure 12 Comparison of change of maximum slope between initial and longer term settlement
(cohesive soils)

O'Reilly et al. (1991) report a case history covering an 11 year period for a sewer tunnel at
Grimsby located in very soft alluvial clays which provides an insight on both the short and
long-term settlements. Over this period the half width of the settlement trough increased from
an initial value of 1.5 Z. to a final value in excess of 4 Z0. The maximum short-term
settlements above the tunnel varied, but the long-term settlements were typically of a similar
magnitude for this 3.0 m diameter tunnel, i.e. the total settlement after 11 years was around
twice the initial short-term construction settlement.

Details of settlements associated with some twenty deep railway tunnels in London Clay are
given in Appendix K (Table 24). These data have been evaluated separately to determine the
settlement trends for tunnels at 25 to 30 m depth within a relatively homogeneous medium,
where the number of variables is reduced. These data have not been duplicated in the cohesive
soil data Table 19 to avoid bias. A plot of the initial versus the one-year settlements is shown
on Figure 13 which suggests the relationship is non-linear. A closer inspection of the data
suggests that there is a change in the ratio of initial to one year settlement between the 2% to
3% volume loss contours. The case records indicate that where the volume loss is less than 2%,
the ratio of initial to one year settlement is between 1:1.5 and 1:2 and where the volume loss
exceeds 3% the ratio of initial to one year settlement increases to around 1:3. Some of the
results above 3% volume loss are noted as being accompanied by seepage which suggests that
larger initial settlement may have a significant effect on the soil fabric resulting in enhanced
drainage to the tunnels and thus correspondingly larger settlements after one year.
Unfortunately, there are no data beyond approximately one year and it is recognised that these
tunnels are generally concentrated in a zone around 30 m below ground level.

2.6 INITIAL SURFACE SETTLEMENT PROFILE

2.6.1 Transverse profile

The initial transverse settlement profile has been observed in many instances to take a form
similar to an inverted normal Gaussian distribution. Whilst there is some theoretical explanation
for this, given by Schmidt (1969), the adoption of the mathematical equation for this

FR/CP/5 29
 distribution is convenient and allows the shape of the profile to be easily generated with
knowledge of the trough width parameter, iy.

N. Apparent contours
of % volume loss
20 1 T'
„eab
.
e .(4)
\
i
Initial settlement, w„,, ,, (mm)

.14\I 4
\u- /....••••--:.i
+0 r

3
..,. r.d,
/....----I 07 (iii.
/.......---"°.
......---"". i m
I (15(; • .1 (6)..
..... . i
10 2 ..........4--'4
i 131
‘..„ - 1 .••• •"•• 1
I
f orc, .‘.... F I°E Key:
/0 •• 1 N (T I (3) 4`4.......... I
i i2) I ...--••••••-• ; Table 23 reference
1 ." 0 ; 1 • ........0I). H . • (% volume loss)
, 1 U'

zo
A. %,(1 .5)
....
5>or•G
"_, 1 (3) (5) • seepage
% ....t..--
o0).I.. I
12 51

11.51
14
(5)
.19135.r..°
1 1 1 I I 1 1 1 1 I 1 1 1 I

10 20 30 40
One-year settlement (mm)

Figure 13 Initial versus one-year settlement for (deep) tunnels in London Clay

The initial surface settlement at any point at distance, y, from the axis of the tunnel can be
calculated from the expression for the inverted normal distribution curve as follows:
( v2
W= wmaxexp --
-=
-. eqn 2.6
2i27 )

where w = settlement at point y from the axis of the tunnel

= maximum settlement above axis

iy = distance from tunnel centre line to point of inflexion

A dimensionless plot for the form of the inverted normal distribution (w/w,„„xversus y/iy) is
given in Figure 14, and given the settlement at intermediate points over the trough profile
can be quickly evaluated.

2.6.2 Transverse slope and curvature

To study surface behaviour the maximum slope within the initial surface trough is a potentially
useful indicator, as its position and value can be defined. At y = iythe maximum slope of the
settlement trough equals 0.607 winjiyand taking a typical value for many clay soils of
iy = 0 .
5 z 0,

30 FRCP/5
 yie y

o 1.0 2.0
o
02

04

08

1.0 shape of settlement trough

Figure 14 Normal distribution function for transverse settlement trough

maximum slope — 1.213w eqn 2.7

The ground loss V, approximates to 2.5.iy.w.

and where VIis the percentage volume lost

w 025•Vr (r2)
MX= 0• eqn 2.8

and maximum slope = 0.030.Vt. (212 eqn 2.9

z.

Using these equations relating tunnel depth (z0) radius (r), and typical trough widths for clay
soils, the maximum settlement and slope for a range of anticipated ground losses can be
estimated. Typical examples are given in Figures 15 and 16 where the maximum ground surface
settlement slope is denoted Ogit can be seen that

FRCP/5 31
 MAXIMUM IssITIAL SvrTLE.MLNT 44
/41Ax
(^e.• v^)

So 100 ISO

REDucE4 ComaciLli teeLy Ex e.e.ss,46 ger-rt-4Mea-'7.5

INI62% Ve•4%

LOSS

o.o2 s
MAx
(.zei-r1 )

Foot 1. , c O. S Z.

vn Irv% SO -w■
•,,
- 75 ws•ret

Figura 15 Relationship between maximum settlement and depth for various values of volume
loss

(a) for a given tunnel diameter and depth, the maximum settlement and slope both increase in
direct proportion to the increase in volume loss, but as iyis expressed as a function of zo
the trough width remains relatively unaffected. For a doubling of the volume loss, the
detectable trough width (based on settlement of 0.5% to 1% of the maximum) increases by
less than 10%;

(b) for a given tunnel diameter and volume loss, doubling the depth of the tunnel decreases the
maximum slope on the surface trough by factor of four and the maximum settlement by a
factor of two, but doubles the width of the zone affected.

(c) for a given volume loss, both maximum slope and maximum settlement increase rapidly for
tunnels at shallow depth as cover reduces.

Over the centre line the maximum sagging curvature is given by wi.../iy2 , with the more critical
maximum hogging curvature about half this value. Since iyis roughly proportional to z., the
radius of curvature is highly dependent on the depth of the tunnel.

The case history data indicated on the plot of maximum surface slope (0g) versus cover to
diameter ratio for cohesive soils shown in Figure 34(b), includes sufficient data points to enable
tentative lines of volume loss to be identified; these compare well with theoretically derived
lines given on Figure 16.

32 FRCP/5
 MAJLIMUM SLOPE. Oc lel CT 1 AL rj4.TTLEMEP4T Tf206,S
AT THE 6iro.p. ► ce. LJ Cos44.51vL SOILS, O (%)

0.5 1.0 1.5 9.0

kt.ouce,o Covc..r2./L i4 by ExCES51 44. 54.af E-6

I gee 1%
tett2%
.1
s/K4.%1

VoLume. Lo ss

O . I
Soo 2oo So

MAX SLOPE el 4) = 0.03054q%)(-r \

Zel 2
Foe 0•5Z0

Figure 16 Relationship between maximum slope and depth for various values of volume loss

There are virtually no data for maximum slope and maximum settlement at shallow depths
related to structural or pipeline damage; this is an unfortunate deficiency.

2.6.3 Horizontal displacements

During the initial phase of development of the surface profile with the passage of the face of
the tunnel, settlements are accompanied by horizontal displacements, which at the surface are
directed towards the centre of the trough. An idealised relationship between initial vertical and
horizontal displacements and horizontal strains for the transverse trough is shown on Figure 17.
This makes the assumption that ground movements are directed towards the tunnel axis, a
mathematically convenient assumption but one that represents a considerable simplification of
the varying behaviour in practice. There is evidence from finite difference modelling using a
non-linear variation of stiffness at small strains to suggest that this assumption may significantly
over-estimate the corresponding horizontal tensile strains.

FR/CP/5 33
 ▪

▪where vortical and horizontal

displacements are approximate
equal.
+1.0 g
E

0
3i
Y •
.02- Iy
,••••1(..--
EY
-1.0 •
•••• OA - • — horizontal strain
it max. tensile strain at iii•point
-- lateral movement
of maximum hogging' curvature ..."
0-6- , --- settlement
E of settlement curve. , 7
?.
A CI'
s typical horizontal strain profile
../
.4-.._ typical normal distribution form of
1.0 • tronsverse surface settlement profile

Figure 17 Idealisation of transverse surface displacement and strains shortly after tunnelling

The maximum horizontal displacements are developed near the point of inflexion (iv) on the
settlement profile and the ratio of maximum horizontal movement to maximum settlement
(v./mi.) is commonly between 0.25 and 0.40, see field studies described by Hansmire (1975)
and Attewell (1978).

The movements that are potentially the most damaging are the tensile strain and the hogging
curvature, both of which reach a maximum at a point about d3i from the centre of the trough
based on the radial movement assumption. The zone of hogging curvature and corresponding
tensile strain extends from the edge of the trough to the point of inflexion on the initial surface
settlement trough. The maximum compressive strain occurs over the tunnel centre line and is
two or three times the magnitude of the maximum tensile strain. Towards the edges of the
initial transverse surface trough the magnitude of lateral displacement may be similar to or even
greater than the settlement, and this factor is of significance in assessing risk to damage of
structures. Discernible lateral displacements have been recorded beyond the limit of discernible
settlement at the edges of the initial trough.

The measurement of horizontal displacements and corresponding strains is not simple and
although potentially key factors in assessment of damage to structures, these parameters are
often not measured and supporting case history data are sparse. Therefore the collection of a
body of data from reliable case records is of prime importance if a better understanding of the
relationship between both surface and sub-surface displacements and the corresponding strains
is to be obtained. In addition, the transfer of horizontal movements in the ground through either
layered strata or from a single soil type to the underside of a foundation is not well understood.
Having calculated horizontal strains in the ground based on a possibly conservative assumption
of radial movement, it may not then be appropriate to further assume that such ground strains
can be transferred in an unfactored manner to the underside of a foundation.

2.6.4 Longitudinal profile

Features of the longitudinal profile based on the assumption that this follows a cumulative
normal distribution are shown on Figure 7, where wm. is the maximum settlement in the fully

34 FRCP/5
 developed transverse settlement trough. In the derivation of the curve the face position, x = 0,
represents the single point at which ground is being lost into the tunnel. It could be expected
that in reality, where ground is being lost at a number of points at and behind the face, the
point of 50% settlement would occur somewhere behind the face and the longitudinal settlement
profile would be somewhat longer and flatter than calculated. This occurs in practice where in
firm to stiff clays the ratio w/w at the face position is commonly in the range 30-50%, as
shown by Attewell and Woodman (1982).

Using the equations and tabulated values of the cumulative normal distribution (see
Appendix A, page 85) it is possible to calculate the settlement at any point forward of the face
on the longitudinal profile. Settlement values on profiles parallel to the tunnel can be derived by
substitution of w by the value of settlement at that distance from the tunnel axis on the fully
developed transverse trough. Therefore, as the settlement at the face is less than half of the total
initial settlement, the average slope in the longitudinal direction is only about 60% of that in the
transverse direction. The maximum slope of the forward trough is two thirds of that in the
transverse direction.

The maximum slope, curvatures and horizontal strains for the longitudinal settlement profile are
thus significantly less than the corresponding values for the transverse settlement trough in the
same tunnelling situation as shown in Table 3.

Table 3 Comparison of slope curvatures, displacements and strains over transverse and
longitudinal profiles
Transverse (y) Longitudinal (x) Longitudinal/
transverse

Location value Location value Ratio %

Maximum slope iy 0.607.w„,./iy 0 0.300.w./i, 66

Maximum curvature 0 w„,j4 -1 0.24.w„,,,g, 24

(sagging)

*Maximum curvature 13.1, 0.45.w./4 +1, 0.42.w„,,„g, 54

(hogging)

Maximum horizontal iy 0.61.iyw„./z. 0 0.40.i..w./z. 66

strain displacement

'Maximum horizontal 13.iy 0.45.w, /z, 0.24.w./ze 54

strain (tensile)

Maximum horizontal 0.24.w„,„/4 24

strain (compressive)

Note: * Potentially most damaging movements

Consideration of the transverse trough will usually represent the highest risk of damage to a
structure. Longitudinal effects cannot be neglected though, as the form of the developing
settlement profile may be significant in some situations. In particular the development of the
forward trough may subject structures which are skew to the tunnel alignment to racking
movements. In addition structures above the centre line of the tunnel will experience maximum
differential settlements during the passage of the tunnel.

2.7 SUB SURFACE DISPLACEMENTS

-

Some observations suggest that the shape and magnitude of the sub-surface settlement profile
may be markedly different from that at the surface. Reliable observations of ground
displacements at depth, particularly close to the tunnel, are difficult to make and are sensitive to

FRCP/5 35
 tunnelling construction processes. Sensible caution must therefore be applied to the field data
used by various researchers to establish relationships for design.

Comparison of settlement at or near the tunnel crown with settlement at the surface carried out
by Cording et al. (1976) yielded an empirical expression for volume loss (V), i.e.

V = 2wc [r + (C z)] — eqn 2.10

where w, = settlement at point directly over crown
r = radius of tunnel
(C—z) = distance of crown to settlement point, 0<(C—z)<1/2r

The limited application, (C-z) between 0 and 0.5r should be noted. If iy= 0.5z., and
V = 2.5iy wmu, then where (C-z) = 0, = 1.6r/z..

Clough and Schmidt (1981) have suggested that there is an empirical relationship between
settlement at the surface and settlement at the crown as follows:
w _ (2r eqn 2.11
, z.

These two equations (2.10 and 2.11) are compared in Figure 18(a), again for convenience,
assuming i = 0.5z.; considerable discrepancy occurs near the crown.
4"'
.vtAVAJettevn4

0.2 0.4 •& o•

s.o /

/ /
CLooag • Sctimig• (19/10

eeeeeia et At (1 974)
C- Z0 • •/2
C - z, • 0

-.1i7L-
MAX C44P4114 L'APIegt.AL C2LLATto11•141P" ljaMAA ajelZ0 wN

0.2 0.4 0.6 e• O 0.2 0.4 0.1

yowl - 1- (0•11•2./r)
1 •3
(two+
13
a02 • O
14 23
o•
. •o 14•
24
6.0 w> se 20 0 22
21 ° 2 4
RAMC DECOP-MATfibi , +♦ . 00 •4 •
OF 5Au0 / 0 le 114 ze 17,5 II
r•,
F. ,..
lo. mA 1 al ED fib • •
x—— —OK
"4-.-______KAoLi:I
las •2 22

I TAW. 2.4 C,Ays

_,...21AL
4j, (0.0522'0)
J
c gm.* (1970
'4elova "7r-
1341

•CLAY
• SILT
WARD
r• O SAUD
a
r0 > 10 o 4AVII•
eevoea.
• (AV)
21 • MALY(

b. M0021 TEST RESULTS AFTER C. CASE STO‘t• DATA

A-maxISOW y, PeTTS (1911)

Figure 18 Comparison of empirical relationships, model tests and case history data with
maximum surface and crown settlement, tunnel depth and radius

36 FR/CP/5
 Shield tunnelling in free or compressed air is accompanied by settlement and lateral movement
towards the tunnel, whilst an earth pressure balance machine can initially cause heave and
outward lateral displacement. Since equations 2.10 and 2.11 were developed from conventional
shield tunnelling data, they are unlikely to apply in such cases.

Using model data from Atkinson and Potts (1977), Ward and Pender (1981) plotted field
measurement of surface/crown settlement, w,/we, against zir see Figure 18(b). The line wily, =
1 - (0.16z/r) is derived from model data and represents plane deformation of sand and it forms
a lower limit to the field data points. The upper line wfw, = 1 - (0.052zJr) relates to drained
overconsolidated kaolin used in model tests and from case records for overconsolidated London
Clay. Atkinson and Potts suggest that wfw, does not fall below 0.40 for values of zir greater
than 10. The case history data points are shown on Figure 18(c), but do not distinguish collapse
conditions or the use of bentonite shields. It is possible a more consistent picture which
distinguishes the behaviour of clays from sands and gravels could be obtained with better
screened data points.

Table 4 compares settlement at the crown with settlement at surface for open-face shield driven
tunnels in stiff clays and shows the former is 1.5 - 3 times the magnitude of the latter. These
data are included on Figure 18(b).

Table 4 Comparative settlements at the crown and at the surface,

(after Craig, 1975)
Depth to Settlement near crown
diameter compared with surface
ratio settlement
CID

Cast iron lining in 7 2.5 — 3 times

London Clay

Expanded concrete 5—8 2 — 2.5 times

lining in London Clay

Bolted concrete lining 3.6 — 2 times

in boulder clay

The assumption by O'Reilly and New (1983) of radial movement in clay soils, which may
overestimate the horizontal component of movement, implies that the width of the zone of
deformed ground decreases linearly with depth below the surface and a similar expression to
that used to determine trough width may be applied, i.e.

iZ = K(z„--z) eqn 2.12

where iz = trough width parameter at depth z
K = empirical constant

it then follows that,

v(y,z) .w(Yg) eqn 2.13

(zo-z)

where v(y,z) and w(y,z) are respectively the horizontal and vertical components of soil
displacement at a transverse distance y from the tunnel axis at depth z.

Formulae for generalised displacements have been developed by O'Reilly and New (1983) but
it is considered that the displacements calculated using these equations cannot be used with
confidence within approximately one diameter of the tunnel extrados; that is in the zone where
significant movements occur which are dominated by tunnel construction techniques and
practices.

FR/CP/5 37
 Model tests in clay by Mair (1979), see Figure 19, give some indication that the width of the
deformed zone tends to decrease with depth and that radial movement towards the tunnel would
be expected. These model tests do not take account of three-dimensional end effects. However,
results from instrumented sections around tunnels in clays in many of the case history referred
to highlight variations in ground movement actually experienced, which depart from the
idealised trend.

• • . k %%1 / I 0, • •
• • . . ‘ 4,
‘ \I 4i4i4iiii
i 4 4 4 4 il'
• f I • • •
. . .
\ .
**. 1
%, I I • •
% 1 ; LI 4 4 4iii /
• • •4 ‘ 1 % / • • I • •
• , aa‘
% ‘4 444 44i/ I f • • • •
•. a il, a 4 '4 444 4 44i , I • • • • •
. , .. a • %" 4444 44/ 1 I
• • • •
• , a, ' '14 4 4 4 4ii i I I
• • • •
a ' k 444 i 44i 1 I • • • • •
... a a a \a >4 44
'\4 H i
....a% % \>, pi'
,
P

• • •

Figure 19 Soil displacements around model tunnel in soft clay (after Kimura and Mair, 1981)

Although the well established correlation between surface settlement profiles and the inverted
normal distribution curve is apparently independent of the soil type and tunnelling method, this
characteristic surface shape does not reveal any of the considerable differences in the patterns of
sub-surface displacements that occur in reality. In practice the magnitude and direction of
subsurface movements appear to be closely related to the tunnelling method and soil type and
are likely to be linked to the stress history of the soil.

Few case histories are available where sub-surface displacements have been adequately
described and recorded; those identified are summarised in Table 25 listing separately data for
soft and stiff clay. Initial horizontal ground displacements at tunnel axis level in stiff clay are
plotted in Figure 20, and suggest that significant horizontal displacements are effectively
confined to a zone equivalent to one tunnel diameter either side of the tunnel axis.

The form of the sub-surface settlement profiles have been reported by Attewell and
Farmer (1974) for a single site in London Clay (C/D = 6.6). His data indicated that the profiles
were not of smooth inverted normal distribution form, but exhibited distinct peaks directly over
the crown and a much reduced trough width. Small lateral displacements towards the tunnel
were recorded, typically around 5 mm at axis level and within one half tunnel diameter.

Observations in soft clay show a wider zone of influence. At Grimsby, ground displacements
described by Glossop and O'Reilly (1982) associated with the construction of 3 m diameter
sewer tunnels (CID = 1.5) in soft silty clay (marine warp) driven with the use of compressed
air, indicated a near constant trough width with depth, but the shape of the profile below crown

38 FR/CP/5
 level was anomalous. Zones of increased displacement developed on either side of the tunnel at
axis level with measurable displacements extending up to 1.5 diameters. The volume loss in this
instance was 7%. In another case described by Glossop et al. (1979) the use of compressed air
to construct a 2.7m diameter tunnel (C/D = 1.2) through soft alluvial silt in Belfast resulted in a
2% volume loss. However, sub-surface outward displacements at axis level of 5 mm resulted,
the effect extending over one tunnel diameter. Glossop attributed this to grout build up on the
shield hood.

0
ZO.JE OA
Sita kitcle.itur
LATERAL.
bf SPLAt-i/AEOT
E ir:\
bg

0A D_

DATA TAKE-T-1 CROK.4 1741.6(.4 k. 7

E
0

ZONE OF 5t4i4IFICArrr
LA:ItersoL. D(Sel-AZEAAE.-LT

0
0 I-o 0 2.o D 3-o D
srA.-ice reasA E:x-ra A

Figure 20 Lateral displacement at axis level adjacent to tunnels in stiff clays

Clough et al. (1983) reported ground displacements associated with the use of earth pressure
balance (EPB) machines, for construction of a 3.7 m diameter tunnel in soft silty clay
San Francisco Bay Mud (C/D = 2). Sub-surface lateral ground displacements away from the
tunnel of up to 36 mm were recorded, extending over a distance of one half diameter from the
tunnel. As the machine advanced, the displacement direction reversed and up to 11 mm inwards
displacement was recorded. This was attributed to soil squeezing and filling the tail void. An
important variable in this case was the operating face pressure which affected directly the
ground displacements. Despite the reversals in the subsurface displacement arising from this
tunnelling method, the initial surface settlement profile followed the inverted normal
distribution.

2.8 ANALYTICAL TECHNIQUES

The previously described methods of prediction of the initial surface settlement profile are
essentially empirically based on observational data and generally simplified relationships.
Provided judgement is exercised and appropriate allowances are made for variations in
workmanship, construction method, soil type and other practical considerations, they are
reasonably reliable.

Finite and boundary element analyses offer the possibility of calculating ground deformation
and the interactive effects on structures and services. However, the application of these

FR/CP/5 39
 techniques to predictions of deformation has not yet reached the stage where they can be
applied generally.

Finite element analysis using an elastic soil model can be used in conditions where elastic
behaviour is approached, but the number of situations where this applies are likely to be few;
strictly only in stiff clays where N < 1. However, most soils exhibit distinctly non-linear
mechanical properties and are anisotropic and more complex. Predictions are very sensitive to
the assumed variation in the soil stiffness factor with depth and to the thickness of compressible
strata assumed to lie beneath the tunnel. Thus, contrary to experience, computations with
seemingly reasonable values of these parameters can indicate general heave resulting from the
stress relief caused by tunnel excavation. For clay soils, the trough shape predicted using an
elastic analysis is generally very much wider than that found in practice, with a tendency for the
edge of the trough to be limited by the model boundary.

Applied to cohesive soils, two-dimensional elasto-plastic analysis can give a more realistic
trough shape, though rather narrower than has been observed, as described by Rowe et al.
(1983). In his study, Rowe found that the magnitude of surface settlement was relatively
insensitive to the assumed ground properties, but was sensitive to the gap assumed between the
excavated tunnel and the ground support. The gap dimension depends upon shield design,
tunnelling method, over-excavation, ground type and characteristics and the method and timing
of primary lining-grouting; these factors are not readily quantified. More recently, Lee and
Rowe (1989) have studied the influence of anistropy and the importance of using appropriate
elastic soil parameters for shear and Young's modulus that have been obtained by testing high
quality, samples in compression or extension to match the appropriate stress paths. In granular
soils which dilate on excavation, additional assumptions for this and modified ground behaviour
near the tunnel are needed to obtain a realistic estimate of surface settlement.

Finite element analyses are often carried out assuming that the advancing tunnel can be
effectively modelled in two-dimensions with plane strain conditions. However, in the case of a
conventional shield driven tunnel there will be a 'bulb like' zone of displacement around the
advancing shield which may be of the order of one to two diameters about the face. In this zone
there will be movement into the face, over the shield and over ungrouted linings. It is suggested
that it may be more appropriate to model this situation and other tunnelling methods three-
dimensionally as a collapsing spherical cavity and research in this direction is warranted.

The use of an appropriate equation to model the markedly non-linear small strain stiffness
behaviour of soils shows some signs of improvement in two-dimensional modelling with more
realistic trough shape and extent. Further work is required in this area.

The more sophisticated methods of analysis can be useful in parametric studies such as those
initiated by Lee and Rowe (1989), where the sensitivity of ground movements to the likely
variations in the different variables can be assessed and compared with case records. However,
little progress can be made without both realistic and representative ground parameters. At the
present time, predictions made by analytical methods do not provide any greater accuracy or
dependability than the simpler empirical methods.

40 FR/CP/5
 3 Influence of ground displacements on the
deformation of adjacent structures

3.1 GENERAL CONCEPTS

Earlier studies of the effects of ground displacements on various types of structure have been
aimed primarily towards establishing criteria for self-weight induced building deformation and
settlement. However, underground construction can cause modes and rates of deformation
different from those normally experienced by structures responding to ground displacements
caused by their self weight and imposed structural loading. These new tunnelling-induced
strains are superimposed upon existing strains and it is possible that where the existing strains
in a structure are at a critical level, and this may not be apparent, the effect of additional strain
could be out of proportion to its magnitude. Factors which are likely to influence the allowable
additional deformation of existing structures include:

• type of deformation
• rate of deformation
• magnitude and distribution of deformation
• type, construction and condition of structure
• interactive soil/structure effects.

uniform + uniform tilt + uniform + extension + differential = total

settlement horizontal (or compression) settlement deformation
(or heave) deformation (differential (bending &
horizontal /or shear)
deformation) producing
differential tilt

rigid body deformation deformation

Figure 21 Two-dimensional components of deformation (after Geddes, 1984)

The components of deformation are shown diagrammatically in Figure 21 and further details of
the above factors are given in Appendix B but the key points are summarised here.

1. Differential horizontal deformation is often more damaging then differential vertical

deformation of an equal magnitude.

2. Displacements resulting in a hogging of a structure are potentially more critical than those
causing sagging.

3. The rate of deformation initially associated with tunnelling is rapid in comparison with self-
weight induced settlements.

4. Tunnelling-induced deformation is superimposed upon that already experienced by a

structure and usually will have not been foreseen in the original design.

5. Experience shows that structural alterations or repairs may introduce weaknesses or

structural discontinuities which are commonly found to be susceptible to ground movement
associated with tunnelling.

6. The presence of structures will generally modify the pattern of displacement.

FR/CP/5 41
 3.2 NATURE AND IMPLICATIONS OF DAMAGE

Recognising that ground displacements are an inevitable consequence of underground

construction, the deformations that can be allowed within the structure have to be assessed.
These can be considered under the following headings :

• safety
• architectural or aesthetic damage
• functional damage
• structural damage
• prevention or repair.

3.2.1 Safety

The safety of operatives, building occupants and the general public is of overriding importance
when assessing the effects of deformation on structures and services. The judgement of likely
effects and the protective measures to be adopted depend on the level of confidence in the
predictions and the consequences if wrong. There may be considerable doubt, not readily
resolved in advance, about the response of some buildings or utilities to ground displacement,
and in these circumstances, adoption of a conservative approach is essential. Visible measures
can be reassuring to users and the public, and measures such as monitoring, whether by
systematic visual assessment and/or instrumentation, provide important feedback on response,
providing progressive evaluation of response of the structure and a basis for rational decisions.

3.2.2 Architectural or aesthetic damage

Damage under this heading includes cracking or separation in infill panel walls, partitions,
floors and finishes and the general superficial deterioration to the exterior or interior of the
structure. In situations where such damage can be fairly readily repaired, such effects may be
tolerable to the user and function of the building. However acceptance of architectural damage
is subjective and may depend upon personal perception and the prestige, prominence or market
value of a structure. Limiting values of crack width in relation to ease of repair are standardised
but nevertheless acceptance of these criteria has to be established by designers, their clients and
the owners of a specific structure. Damage may be more acceptable to those accustomed to
such conditions, for example, in active mining areas.

The cladding, partitions and finishes are usually applied towards the end of construction, when
the structure has usually already undergone some settlement and partial bedding down, as
indicated in Figure 22. Consequently, limiting criteria for architectural damage derived from
case histories of buildings settling under their own weight or other time-dependent processes
may overestimate the tolerance of the architectural elements to deformation. However these
criteria need to be viewed cautiously before applying them to structures subject to additional
deformations which may occur rapidly. Dislodging of architectural finishes such as plaster
cornices may present some danger as a result of only small building deformations.

3.2.3 Functional damage

Functional damage impairs the use or serviceability of the structure, without affecting overall
integrity or safety. Certain plant or machinery (e.g. lathes, papermaking and printing plant, lifts,
automatic stacking systems, overhead cranes) may be sensitive to very small movements. The
consequences of interruption to the function or even a component of the function of a structure
may be commercially severe.

3.2.4 Structural damage

Structural damage relates to cracking or deformation in structural elements. It may lead to

failure of the elements, to the connections between them, or instability of the entire structure.
Evidence of such damage can be obscured by architectural finishes which may themselves
remain reasonably intact. However, plaster is usually a good indicator of crack propagation.

42 FR/CP/5
 ro
O

Time

• ... A (Cladding and finishes)

.....
v------A (Raft and lower levels
of structure)
Initial
settlement

E
(approximately)
T:. - N Relative deflection

°' Final settlement

z.
VI
A

Figure 22 Proportion of differential settlement affecting cladding and finishes

(after Burland, Broms and de Mello, 1977)

Applied to brick bearing wall structures, this definition requires modification because where
brick is the primary support, cracking or separation may not necessarily compromise overall
stability. In this case structural damage to brick bearing walls refers to deformation that can
lead to danger of instability of all or part of the structure.

Although cladding may be architectural, the connection of cladding panels to the building
should be considered as structural. These connections can be prone to poor design and may
have not been designed with tunnelling in mind, poor workmanship can also lead to dangerous
failures.

3.2.5 Prevention or repair

Small deformations that cause cracking with negligible risk of structural damage may generally
be easily repaired and at a much lower cost than the measures required to prevent them. Partial
underpinning is one example of a widely used preventative measure but one which may not
restrain more damaging lateral displacements. Deformations as a result of movement during

FR/CP/5 43
 installation or subsequent action may be as harmful as or more severe than those it is designed
to prevent. Therefore, there is a balance to be obtained between:

(a) the use of protective measures which are often costly, may result in considerable disruption
to the occupiers or function of the building and may cause some degree of cracking or
damage during their installation; and

(b) the adoption of an observational approach such as that described by Peck (1969a), whereby
the resulting behaviour of the ground and structure is monitored during progressive stages
of tunnelling. The observational method may allow the actual response of the structure to
be established during the first phase of sub-surface construction where the element of
associated risk is acceptably low. The remaining phases may then be reassessed on a more
informed basis. Contingency remedial strengthening or other protective measures are
introduced when and if required, with the final replastering and fmishing being =Tied out
after an appropriate interval following completion of the tunnelling.

3.3 DAMAGE CRITERIA FOR STRUCTURES

3.3.1 Crack width

The criteria for visible damage are usually related to crack width. A comprehensive
classification was developed by the Institution of Structural Engineers (1978) as an indicator for
damage to non-structural elements with respect to ease of repair. This is detailed in
Appendix C, with a summary provided in Table 5, for degrees of damage ranging from
negligible for hairline cracks less than 0.1 mm wide, to very severe cracks exceeding 25 mm.
This matter is also dealt with in relation to low rise buildings and ground floor slabs in BRE
Digest No.251 (BRE, 1981, revised 1990).

Table 5 Classification of visible damage to walls, after I.Struct.E (1978)

Category of Degree of damage Approx crack width

damage (mm)

Aesthetic 0. Negligible Not greater than 0.1

I. Very slight Not greater than 1.0
2. Slight Less than 5.0

Functional/ 3. Moderate 5-15, or a number of cracks >3 mm

serviceability 4. Severe 15-25, depends on number of cracks

Structural 5. Very severe Usually >25 nun but depends upon number of cracks

The degree of importance attached to cracks less than 5 mm is a subjective matter and there
may be situations where this is unacceptable to the user or owner or impairs the function. For
instance, where partial underpinning has been carried out to prevent deformation, the
development of cracking despite such treatment may be unacceptable.

The Institution of Structural Engineers (1989) in their publication on soil — structure interaction
have further qualified the degree of damage associated with crack widths and serviceability
limits for various types of structure. Although this approach is not adopted in this document for
reasons of simplicity, it does further emphasise the need to assess the degree of damage
associated with crack widths in relation to a range of factors such as national, historic and
economic importance, social and functional significance, and so on as well as the type of
construction.

44 FRCP/5
 3.3.2 Deformations

Settlement

A number of authors have suggested limitations on maximum allowable total vertical foundation
deformation resulting from settlement for a range of structures. These limits are summarised in
Table 6, but were intended to be indicative and to be used with caution. Project specific details
need to be taken into account, including the type of structure, or type or sensitivity of contained
machinery and the actual ground conditions.

Settlement resulting in uniform deformation and rigid body movement or tilt, does not
necessarily cause damage within the structure, although service connections, adjacent buildings
and paved areas could be damaged. In practice, however, deformation is rarely uniform except
for structures of unusual strength and rigidity. Differential settlement occurs due to variation in
ground properties, depth and magnitude of imposed loads over the area of the structure. In the
case of tunnelling, these variations are additional to the differential settlements caused by
tunnelling-induced ground movements.

Table 6 Allowable total settlements of foundations

Class of Type of building or structure Max. allowable total settlements, w(mm)

building and
structure lun and Polshln and Skempton and
Starzewski Tokar MacDonald
(1972) (1957) (1956)

1 Massive and robust 150-200 300 Clays 75-125

structures Sands 50-75

2 Statically determinate 100-500 100

structures

3 Statically indeterminate steel 80-100 150

structures, load-bearing
brickwork, and reinforced
frame structures, see note (e)

4 Stmctures of Class 3 but not 60-80 75-100 Clays 75

satisfying one of the stated Sands 50
conditions, see note (e) and
reinforced concrete structures
founded on isolated footings

5 Prefabricated structures 50-60

consisting of large slab or
block elements

Notes:

(a) The smaller values quoted relate to public buildings, dwellings, or buildings with structural members or finishes
particularly sensitive to differential settlement; larger values relate to taller buildings of considerable rigidity or to
structures which can accept such movements.
(b) In special cases (such as gantry beams, high-pressure boilers, special storage tanks, silos under non-uniform loading,
etc.) allowable maximum or differential settlement or both should be taken as specified by service or mechanical
engineers or by manufacturers.
(c) The maximum allowable settlement includes that which may occur during the construction period.
(d) The values given by Skempton and MacDonald which appear in the above table to be more conservative, were
criticised by Terzaghi (1956) as being too sweeping and that 'allowable' should be differentiated according to both
type of structure and individual regions where the geology is more or less uniform and not applied as a blanket
classification.
(e) Statically indeterminate load-bearing brickwork under Class 3 includes construction with reinforced concrete ring
beams at every floor level, with longitudinally reinforced concrete strip foundations and with cross walls of at least
250 mm thickness and spaced at not more than 6 m centres. Concrete frame structures with columns at less than
6 m centres and founded on strip or raft foundations.

FR/CP/5 45
 Distortion

A number of authors have used field observations to correlate structural distortion and tangible
damage and a selection is summarised in Table 7. Generally a clear distinction is made between
the behaviour of :

(a) framed buildings and reinforced load-bearing walls and

(b) unreinforced load-bearing walls.

Table 7 Classification of deformation by different authors, after I.Struct.E. (1978)

Buildings settling under their own weight Buildings subject to
rapid movement caused
by under-pinning

(a) Framed buildings and reinforced load-bearing walls

Limiting values of relative rotation (angular distortion, 13) given by:

Skempton Meyerhof Polshin & Bjerrum O'Rourke et at. (1976)

MacDonald(*) (1956) Taker (1957) (1963)
(1956)
Structural 1/150 1250 1/200 1/150
Damage

Cracking in 1/300 1/500 1/500 1/500 1/750

walls and (but 1/500
partitions recommended) (0.7/1000
to 1/1000
for end bays)

(b) Unreinforced load-bearing walls

Limiting values of deflection ratio (AIL) for the onset of visible cracking given by:
Meyerhof Polshin & Tokar (1957) Burland & Wroth O'Rourke et al. (1976)
(1956) (1975
Sagging 12500 1/3300-1/2500; <3 1/2503; Lill = 1 1/400004
1/2000-1/1430; 1,111 > 5 1/1250; L/H = 5 at LIff = 1

Hogging 1/5000; L/H = 1

112500; LIH = 5

Notes: (a) Studies of permissible deformation may not be applicable to tunnelling-induced movements
since the time scales are not comparable. In addition, deformations associated with
tunnelling may include a substantial component of horizontal strain.

(b) Given at (3 = 1/1000 and assuming a ratio of A/L to 11 of 1:4 consistent with circular
curvature on a limb of a trough, (O'Rourke et al., 1976)

Flexible framed buildings and reinforced load bearing walls tend to deform by shear as
rectangular panels distort to parallelograms or to bend at discrete positions where the structural
members have the tensile capacity to resist bending as shown in Figure 23(a). Relative rotation,
B, (angular distortion) measuring the rotation of a member relative to the building as a whole is
a useful deformation criterion. The magnitude of relative rotation depends upon the position of
the structure within the trough.

Tensile ground strains are generated along the projected tunnel axis ahead of the advancing
tunnel and also within the outer portion of the transverse settlement profile. Structures near the
edge of a trough may be subject to a damaging combination of horizontal tensile strain and
hogging deformations.

46 FRCP/5
 Unreinforced load bearing walls with little tensile strength may suffer bending deformation,
following more closely the ground displacement profile as shown in Figure 23(b). Deflection
ratio WL) has been used as a limiting criterion, and it can be shown that critical values for a
given ratio of building length to height relate to a limiting tensile strain as demonstrated by
Burland and Wroth (1975). The values quoted for 1,111 = 1 would relate to small buildings such
as houses, and correspond to higher tensile strains than the limiting values for longer buildings.
The value of deflection ratio is dependent upon the amount of rigid body movement or tilt.
Unreinforced walls with many openings, such as doors and corridors, become more flexible in
shear, therefore shear deformation and diagonal tension will control cracking.

(A)

shear distortion

(B) tensile cracking

Figure 23 Likely behaviour of different types of structures on the limbs of a surface settlement
trough experiencing hogging deformation
(a) framed buildings and reinforced load-bearing walls
(b) unreinforced load-bearing walls

Boscardin and Cording (1989) have reviewed the likely behaviour of masonry structures which
they indicate do not typically deflect solely in shear or bending but rather in a combined mode,
although dominated by shear. They also show that, both as a result of the likely ratio of
Young's modulus to Shear modulus (E/G) and due to the small initial ratio of deformed length
to height of structure (LIH) caused by the progressive tunnelling wave of settlement, shear
deformations are likely to responsible for most damage to masonry building walls. They
propose that relative rotation (B), a measure of shear strain, combined with horizontal strain are
appropriate parameters for assessing damage potential.

It may be concluded that for normal masonry structures on shallow spread foundations, shear
deformation is more common in response to tunnelling-induced ground movements than
bending, although the effects of bending if dominant can be more serious.

Engineers will be familiar with serviceability limit state deflection criteria for various structural
elements and a brief review of UK practice is useful for comparison. British Standard BS 8110:
Part 2 (1985) for the structural use of concrete requires that the final deflection (including the
effects of temperature, creep and shrinkage), measured below the 'as-case level of the supports
of floors, roofs and all other horizontal members in general, should not, exceed span/250. The
code further suggests that the part of the deflection occurring after construction of partitions or
application of finishes be limited to the lesser of span/350 or 20 mm, or to only span/500 where
partitions or finishes are of brittle materials.

FR/CP/5 47
 Acceptable deflection limits for various structural elements collated by Alexander and Lawson
(1981) are given in Appendix D. Assuming that the values given relate to total vertical
deflections subsequent to the erection of the particular element, then they must include
contributions from self-weight settlement of the element and settlement/deformations arising
during completion of the surrounding structure. The method of design for concrete elements
generally underestimates the true stiffness of the element, and usually the specified deflection
criteria under self weight and imposed loading are readily achieved.

3.3.3 Horizontal deformation

The National Coal Board (1975) is the main source of information on the damage associated
with horizontal displacements; (see Appendix E, p. 91).

Modern `Longwall' mining operates on a much larger scale than civil engineering tunnelling
works, the working coal face being measured in hundreds of metres and the resulting ground
subsidence amounting to 90% of the excavated thickness of coal. Consequently, the magnitude
of lateral displacements associated with mining, being much larger than those normally
associated with civil engineering tunnelling, range from 30 mm (slight) to 180 mm or more
(very severe). Some buildings in mining areas can remain functional and safe when subject to
large movements, and structures can be specifically designed to accommodate such movements.
Although the sub-surface excavation is the cause of ground displacement in both conventional
tunnelling and deep mining, the relative magnitudes are so different that the NCB classification
can have limited relevance in the context of this study. Further case studies of sensitivity of
various types of structure subject to differential horizontal movements are required to provide a
more meaningful framework for prediction. However, the classification of horizontal ground
strain used in connection with deep mines is used by Boscardin and Cording (1989) to provide
a basis for their relationship between horizontal strain and relative rotation (B). Their
relationship is likely to be conservative for classifying potential damage to buildings with a
length of less than 30 m parallel to the direction of straining, since they base the relationship on
buildings of 40 m length. In order to see the influences of this reference should be made to
their paper.

3.4 DAMAGE CRITERIA FOR SERVICES

In urban areas, tunnels are often sited below main streets, with tunnel(s) running parallel to the
main services but transverse to the side street connections. The situation may be much more
complex at street junctions, as shown by Kuesel (1972) on the San Francisco BART project,
where many mains and services interconnect, or where the tunnel has shallow cover and passes
under main services, not necessarily at right angles. Service connections to buildings are
particularly vulnerable to damage, due to differential settlement between the building and the
ground.

When the potential for damage arising from underground construction is assessed, this must
include both an assessment of the likelihood of damage and the nature of the consequences if
damage were to occur. The potential consequences of damage will relate to the function of the
pipeline/service, for example transmission of inflammable and/or noxious gases and high
pressure water or oil, mains drainage, and so on. It is common to take positive action to protect
gas mains, and in the UK it may be appropriate to relocate and/or renovate old sewers which
may now be susceptible to additional deformation. However, the type and spatial relationship of
adjacent structures and the type of ground may be important factors in assessment of
consequences. For example a burst water main in sandy soil may quickly erode the host ground
and undermine adjacent properties.

Prediction of the likely pipe response at the design stage abounds with uncertainties, particularly
in view of the wide range of possible sub-surface ground responses. Inevitably the assessment
process will involve considerable judgement and should include full participation of those
responsible for the buried services. In circumstances where the uncertainty cannot be resolved

48 FR/CP/5
 there may be little option but to adopt a cautious approach to safeguard the service(s) and all
other directly or indirectly interested parties.

Some details of the performance and limiting criteria for the assessment of the behaviour for
various pipe materials is given in Appendix F. For further details the reader is referred to Soil
movements induced by tunnelling and their effects on pipelines and structures by Attewell et al.
(1986) which provides a comprehensive and recent summary of knowledge and experience in
this area.

FRCP/5 49
 4 Risk assessment

4.1 HISTORICAL

It is of interest to look back briefly at the early days of underground construction in the UK to
gauge former attitudes to risk and damage. Such was the regard for the sanctity of property in
the Victorian period that until 1892 railway tunnels could not pass below a building without its
freehold being purchased. In 1892 a Joint Select Committee of the Houses of Lords and
Commons recommended that free wayleaves be granted beneath public highways, and
wayleaves (not freeholds) had to be purchased (subject to compensation for damage) when
passing under buildings. However the purchase of such property wayleaves was fraught with
problems due to the owner's concern about possible damage, and the majority of early
construction followed the existing street layout, often with one tunnel above the other and with
tight curves, as described by Jackson and Croome (1962).

Since 1910 it has been necessary to obtain Parliamentary authority to construct underground
railways and the limits of deviation that are permitted during construction are shown on plans
deposited with the Parliamentary Bill concerned. The extent of these limits are not necessarily
determined by consideration of the extent of likely displacements or associated damage, or
depth of the tunnel, but may take account of ancillary works associated with the main
construction. The vertical deviation is usually stated within the Act, the limit in 1913 was no
more than 5 feet upwards, and since 1955 it has been no more than 10 feet or 3 metres
upwards. It has been usual practice in London to produce a schedule of existing defects for all
structures falling within the zone of potential influence, although considerable judgement is
exercised in the selection of structures.

4.2 PRELIMINARY APPRAISAL

In the development and design of a project involving tunnelling it is necessary to assess the
likely damage that could be caused to existing, and sometimes planned, structures. It is
advantageous to make a preliminary assessment during the early design stages, when it is still
possible to rearrange the scheme to minimise the risk to structures; this is an iterative process.
For a large or complex scheme in an urban area with many structures potentially at risk, a
simple means of assessment is needed so that appropriate action can be applied to the individual
cases. Such a blanket classification has to be applied with discretion, and even at a preliminary
stage structures which are particularly sensitive, because perhaps of their importance, poor
condition, housing sensitive equipment or potentially high repair costs, should be identified and
considered separately.

For the preliminary risk assessment and classification, relatively simple criteria are desirable
with determinable parameters depending upon the position, depth and shape of the settlement
trough and on the position of the structures. Parameters such as relative rotation and deflection
ratio are not easily applied since, although they may offer a good guide to likely damage, they
rely on a detailed consideration and some knowledge of the individual structures, see Figure 24
Therefore a staged approach to ground movement prediction and likely building response can be
applied. A three step methodology has been progressively developed whereby a first-stage
assessment of structures at risk can be rapidly identified from simple predictions of `greenfield'
ground slope and settlement, based on knowledge of the ground conditions, tunnel geometry and
tunnelling methods, and the application of limiting criteria of both building slope and
settlement. A second stage assessment of particular structures identified in stage one including
any especially sensitive structures, not necessarily identified as at risk in stage one, can be made
based on knowledge of the main structural elements and predictions of both critical strain,
relative rotation (B) and mode of deformation of the structure. A third stage of more detailed
analysis which may include three dimensional effects and an expert appreciation of the
behaviour of the particular type of structure when subjected to settlement may be necessary, but
generally only in a limited number of cases for any particular tunnelling scheme.

50 FRCP/5
 GROUND SURFACE
BEFORE

I-
GROUND
w SURFACE AFFER
Ai %IMAM

a) Deformation of isolated pad/strip foundation

L = Length of structure
= Length of bay

GROUND
SURFACE BEFORE

GROUND
SURFACE AFTER

SLIP TYPE
DISPLACMENTS L A = o, A = o, p =o

b) Deformation of rigid raft

SLIP TYPE
DISPLACEMENTS

c) Deformation of basement/cellular foundation

Figure 24 Schematic response of various types of structure on the limb of a settlement trough

FR/CP/5 51
 In the first instance it is convenient to assume that the near-surface foundations of structures
follow the slope of the settlement trough, appreciating that this makes no allowance for restraint
created by the presence of structures or services. Boscardin and Cording (1989) consider that
most small to medium size masonry bearing-wall structures do little to modify ground
movements. This being so, the critical parameters of deformation depend on the slope of the
ground surface caused by settlement due to underground excavation. It is recognised that more
robust structures such as framed buildings will offer greater restraint and the application of
ground slope criteria may overestimate likely damage.

Whilst overall uniform settlement may not damage structures, actual settlements vary from the
predicted value due to such causes as inherent variations in ground properties and workmanship.
This irregular settlement also causes structural deformation, and may be assumed to be
proportional to the predicted magnitude of settlement associated with tunnelling at any given
point. In addition structures will experience differential settlements during the passage of the
tunnel, particularly those sited over the centreline of a tunnel. The development of the
longitudinal settlement profile causes the same magnitude of maximum differential settlement as
that in the transverse direction, although the maximum ground slope and maximum horizontal
strain are two thirds that experienced in the transverse trough.

Thus, the suggested criteria for classifying structures and services at risk during route alignment
studies are based on a combination of predicted critical values of initial settlement and ground
slope. The design values adopted may depend on ground conditions and on the type and
condition of the buildings in the area under consideration.

For tunnels in stiff clays it has been common practice to assess the potential risk of damage to
structures on the magnitude of the initial settlement and ground slope generated beneath the
building during and shortly after construction. Experience to date tends to confirm that for
deeper tunnels in stiff clays this has been acceptable for identifying buildings at risk and
generally long term settlements do not appear to have caused concern. This may also reflect the
timing of any repairs carried out, which in many cases may be a year or more after construction
at a time when the majority of potentially damaging movement has occurred.

For a preliminary assessment of the risk of damage due to tunnelling, it is convenient to

consider the maximum slope to which a structure will be subjected. This derives from the
maximum differential settlement that occurs anywhere within the plan area of the structure and
is not necessarily the same as the maximum slope on the settlement trough. Though maximum
ground slope is not a direct measure of building deformation, it enables the ground area above
the tunnel to be zoned into risk categories. Simple geometrical techniques can be applied to
derive approximate relationships between maximum slope and relative rotation (for framed
buildings and reinforced load-bearing walls) and deflection ratio (for unreinforced load-bearing
walls). The relationships vary depending upon the position of the building in relation to the
settlement trough. Considering buildings located in the area of the most potentially damaging
effects, near the edge of the settlement trough where there is a combination of tensile strain,
lateral ground movement, and hogging, the approximate relationships derived in Figure 25
typically apply. The maximum building slope (013) is commonly two to four times steeper than
the relative rotation (B) of the structure and six to eight times the deflection ratio (A/L). These
ratios have been used to derive the relationships between damage and maximum slope given in
Table 8.

52 F R/CP/5
 0 a
g *.z. , II
....
I
.....
-1...
3IF.,

11 g i.Z Z-73 II
1
(IIZ', gl-, D11
(P.
0
II
31-1 ci 6..—:
II 11
<I P o
1 11
1
4 1 3I — •—•
II
csi
II
G ..z, =. =`"

er,

O 31-3
II
Case (i) Gaussian curve, bending distortion, 2 bays over 2i

Figure 25 Idealised deformation of various types of structure on limbs of surface settlement

trough

FR/CP/5 53
 Table 8 Relationships between damage due to tunnelling and
maximum slope near edge of settlement trough,
for structures experiencing hogging deformation

Type of structure Range of values

(depends on LIH ratio) for maximum slope

a) Framed buildings and reinforced

load-bearing walls
Cracking 1/250 — 1/375
Structural damage 1175 — 1/125

b) Unreinforced loadbearing walls

Cracking 1/310 — 1/625

Typical values that have been used for planning and design purposes as well as for optimising
alignments are presented in Table 9. There are as yet insufficient case records to confirm their
general validity however and thus, for the present, they can only be considered as a guide to be
applied with experience and judgement.

Table 9 Typical values of maximum building slope or settlement for damage risk assessment

Risk Maximum slope Maximum Description of risk

category of building settlement of
building (mm)

1 < 1/500 < 10 Negligible: superficial damage unlikely.

2 1/500 — 1200 10 — 50 Slight: possible superficial damage which is unlikely to

have structural significance.

3 1200 — 1/50 50 — 75 Moderate: expected superficial damage and possible

structural damage to buildings, possible damage to
relatively rigid pipelines.

4 > 1/50 > 75 High: expected structural damage to buildings and

expected damage to rigid pipelines or possible damage
to other pipelines.

Note: The higher risk category from either slope or settlement consideration controls.

These criteria of slope or settlement or the derived zones of risk are readily marked up on a
plan of the tunnel route in tramline fashion. The method of doing so depends on the tunnel
layout. For a single tunnel, the predicted width of trough to the critical value of settlement or
slope can be calculated for various depths of tunnel, z,, and tabulated, an example is given in
Appendix G. The trough width to a given settlement, assuming a normal trough form applies,
has an algebraic solution, but the width to a value of slope has to be solved iteratively. Slopes
exceeding the critical value may occur in two bands either side of the trough centre line where
the slope reduces to zero, but it is suggested that the whole trough width to the extreme limits
of the critical slope be included. The tabulation can be produced relatively rapidly using a
programmable calculator, and can be repeated for different values of trough volume (i.e. volume
loss) in order to test the sensitivity of the zoning to the assumed value. As well as the
uncertainty in the likely volume of the settlement trough, different values of the volume loss
can reflect alternative tunnelling methods and may assist in selecting appropriate construction
methods.

Two examples of critical transverse distances for slope or settlement criteria are given in
Appendix G. The first example is calculated for a 5 m internal diameter sewer tunnel with a
3.05 m excavated radius located in silts and sands below the water table. The predicted ground
loss for a bentonite tunnelling machine was assumed to be 2%, but the table also gives values
for 4% volume loss, representing the use of a shield with compressed air. The results are shown
diagramatically on Figure 26 which, if drawn to an appropriate scale, could be used as a cursor

54 FR/CP/5
 to mark zones of a given risk category at the surface on plans of the works and a typical
example is given in Figure 27. The inter-relationships between depth of tunnel, volume loss and
surface risk categories for this example are shown diagrammatically in Figure 28 . It is worth
noting that whilst an increase in the volume loss gives a proportional increase in settlement or
slope at a given point, the increase in trough width to the point of limiting settlement or slope is
relatively small. A second example giving calculated values of slope and settlement for a tunnel
with an excavated radius of 2 m is also given in Appendix G.

240

15:0

10.0

a) volume loss = 2%
5 o 10.0 Q0.0
i5.o
TgAgsvgast. DisrAme4 ra
(L1TIGAL VALUE. OF SLOPE
OR SETTLEMENT (ws)

lo t'

kdy
Serat"Enir
Damets.SATe5

Q10•44 4ATE-Aogy

So
0 SLOPE>
cat 75,ww,
lo.o
0 ____ 51.0PL AJ
4-00
So
0— LOPE) too
)P to
• as
b) volume loss = 4%
5.0 10.0 15.0 20.0
TeAt.15W2SE DISTAMLE To
CRITICAL VALUE Or SLOPE.
OR 5c-rri..t.A4‘,.rr (-rn)

Figure 26 Examples of risk assessment for various tunnelling methods reflected in different
values of volume loss

FR/CP/5 55
 Figure 27 Example of risk zoning at surface for a single tunnel

56 FR/CP/5
 E
E
0

O
gr-

E
E O
E
0
0

E
E
0
".7
IF —

4sa
a

E
O
E
• 0
f`NI
E 0
O E
(NI 111 0

II

iv

Figure 28 The inter relationships between depth of tunnel, volume loss and surface risk zones
-

FR/CP/5
57
 Where the effect of a complex of tunnels is to be assessed, direct calculation of the extent of
the zones of various risk category zones requires the use of a computer using a specially
developed program. In practice, it is often easier to consider cross-sections at a number of
selected points, plot the settlements due to each individual tunnel and add them graphically to
plot the total settlement trough at each section. While performing this task, an experienced
engineer can make due allowance for areas where tunnelling is likely to be more than usually
difficult or prolonged and a greater proportional volume loss can be expected. Values of
predicted settlement can be plotted on a plan of the works, and settlement contours drawn
(see Figure 29). By this means, large numbers of buildings can then be classified quickly by
reading off the settlement and measuring the closest contour spacing to obtain the maximum
slope.

58 FR/CP/5
 De• mtton
f Se Illernenl

t ieny te
V
uts o
Be

a
-0-

8
z

.1i

nipinirtl opolinuritTiurrimlino
a
; S 2 g
 Having identified the extent of the zones of each risk category, it may be possible by adjusting
tunnel routes and depths to minimise the amount of damage, reduce the number of buildings
affected, or avoid sensitive structures. For example, a tunnel following the centre of a street
may put buildings on both sides at risk of superficial damage, but off-setting the tunnel might
remove the category of risk from one side of the street without increasing the risk level to the
buildings still affected. Some additional analysis would be needed to verify this, but simple risk
classification can demonstrate the possibilities. Any structures that straddle or have connections
to other structures within a more critical zone, should be included in the more critical
assessment zone.

Once buildings in an affected area have been classified, either on a plan or as a listing, and the
layout of the tunnel works finalised, a course of action is needed. Using the same risk
classification given in Table 9, action suggested could be as shown in Table 10. Particular
structures which have been identified as being in risk category 3 or 4 (and in some
circumstances even category 2) should be subject to a second-stage assessment.

Table 10 Action suggested for various risk categories

Risk Description of risk Description of action required

category

1 Negligible: superficial damage No action, except for any buildings identified as particularly
unlikely sensitive for which an individual assessment should be made.

Slight: possible superficial Crack survey and schedule of defects, so that any resulting
damage which is unlikely to damage can be fairly assessed and compensated.
have structural significance
Identify any buildings and pipelines that may be particularly
vulnerable to structural damage and assess separately.

3 Moderate: expected
superficial damage and possible Crack survey, a schedule of defects, and a structural assessment.
structural damage to buildings
and expected damage to rigid Predict extent of structural damage, assess safety risk, choose
pipelines whether to accept damage and repair, take precautions to control
damage or, in extreme cases, demolish.
4 (a) High: expected structural
damage to buildings Buried pipelines at risk — identify vulnerable services, and decide
whether to repair, replace with a type less likely to suffer
(b) Expected damage to rigid damage, or divert.
pipelines and possible damage
to other pipelines

Note: The above table relates to near-surface foundations or pipelines.

As part of a second stage evaluation, there is often a need to assess other aspects of certain
structures which may be more sensitive to particular modes of deformation. A method showing
promise for masonry and brickwork structures, which are sensitive to tensile strain from
horizontal ground movements or hogging deformations, has been put forward by Boscardin and
Cording (1989). A relationship of the likely level of damage to masonry structures is developed
between key parameters of horizontal strain in the building, which is assumed to be equal to
that developed in the ground, and maximum relative rotation (angular distortion, B) of the
building which is reproduced in Figure 30. The appropriate values of relative rotation (angular
distortion) depend upon the location and width of the structure and are shown in Figure 31 for
various combinations. The distortion factor (d) at the limits of the settlement trough varies
between 0.8 and 0.1. For buildings smaller than the half trough width, the magnitude of the
relative rotation (B) will be smaller than the average slope, and prediction of damage risk is
very similar to that obtained from. For buildings larger than the settlement half trough, the
relative rotation can be assumed to be equal to the average slope of the settlement half trough
and this is approximately half the value of the maximum ground slope at the point of inflexion
on the settlement trough, since max slope = 0.607 wraji, and average slope = 0.33 w.„Jiy.

60 FR/CP/5
 Using this assessment the dominant mode of deformation of the structure, that is bending or
shearing, can be identified.

In the detailed design stage of a project it may be necessary to carry out further more detailed
studies for those structures in the highest risk categories. This detailed assessment is likely to
include further evaluation of the following factors:

(a) structural continuity

(b) differential settlement
(c) racking
(d) soil/structure interaction.

As many aspects of the above are not amenable to precise calculations, the final assessment of
possible change will require engineering judgement and expert assistance should be sought.

4.3 SUB SURFACE DISPLACEMENTS

-

There is no established empirically based approach to estimating lateral displacements in ground

around tunnels during their construction. This is likely to remain the situation, particularly since
different tunnelling methods produce different patterns of sub-surface displacements.
Movements can be derived from finite element analysis, although the accuracy of such
predictions is in doubt due to the uncertainties associated with the selection of representative
ground properties, as well as the soil/structure interaction effects.

There are only a few case histories where such movements have been recorded; these results are
discussed in Section 2 and summarised in Appendix K, Table 25. The data for stiff clays have
also been plotted in Figure 20 and suggest rapidly decreasing initial lateral movements at axis
level with increasing distance away from the tunnel extrados and that only very small
displacements, less than 3 mm, are likely to occur at a distance in excess of one tunnel
diameter. However, all the examples plotted are from the London Clay and may only be valid
for either this material or a particular range or value of the ratio of horizontal to vertical stress
in the ground, 1C0. The following table, giving tentative limits for the zone of significant
displacements at axis level, is put forward as a possible framework from which others may test
its sensitivity and applicability.

Table 11 Tentative limits of zone of undefined risk around tunnels in different soil types

Ground conditions Zone of undefined risk at axis level

measured from tunnel extrados

Soft clays 2 diameters

Stiff clays 1 diameter
Granular soils 11/2 diameters*

Note: * (excluding mns or catastrophic losses)

For piled foundations close to the tunnel, consideration should be given to the effects on
bearing capacity of piles founded or gaining support by friction at levels above tunnel invert,
and to bending effects on the piles due to lateral ground movements. Even if the toes of end-
bearing piles are founded outside or below the zone of influence, both initial and long-term
displacements generated by the underground construction may contribute significant negative
skin friction upon shafts passing through this zone; either by increasing pile loads which could
result in additional settlement, or by putting pile shafts into tension. There may be situations for
developments constructed in advance of the tunnels where potential negative skin friction can be
eliminated by sleeving of piles in the zone likely to be disturbed. The provision of continuous
reinforcement throughout the length of the pile can be beneficial in minimising potential
tensional effects caused by differing settlement between top and toe of the pile shaft.

FRCP/5 61
 E1-AT I VE. Q OT ►T 1 40.4 C
Figure 30 Relationship of damage to relative rotation and horizontal ground strain
(after Boscardin and Cording, 1989)

Pow,' OP Wm-exit:y.4

I = 4%. )
14011‘.1-14
4 6A64146 MAX

War"' a. 5TaacTuag = Y1C

0.5 e
0•25 l
6.1 e
e. LitJ4T14
ST12.34r1J4ZE.

0 0•2 0.4 42•c. 0.S , 1.e

LocAT c>44 CaJTVE OF 5re

'MAX = -(2.906,4•1
MAK. 541.1%-6/-4 E-AT 04- 1
( A cm:Q. 60644.ep1,4 i coach.4 4, 19se,)

Figure 31 Relative rotation within a tunnel settlement trough (after Boscardin and Cording, 1989)

62 FRCP/5
 More often, assessments of the additional negative skin friction load and the pile settlement at a
safe working load have to be made. Where piles terminate at or close to tunnel level, the
bearing capacity of the piles may be reduced and/or additional load on the tunnel linings may
be caused.

Bending of a pile may, in the first instance, be assessed assuming that the pile follows the
predicted ground displacement, although this may be conservative. The vertical settlement at
depth can be estimated assuming a normal distribution trough shape, with its width
corresponding to the height above tunnel axis under consideration, and assuming that the ground
movements are directed towards a point at the crown of the tunnel. The latter assumption would
be consistent with the ground losses normally occurring over the shield. Thus, knowing the
magnitude of the resultant in the direction indicated, the magnitude of the horizontal component
can be estimated. Alternatively, generalised ground displacements at depth around a tunnel can
be estimated using the formulae developed by O'Reilly and New (1983), which assumes radial
movement towards the tunnel axis.

Morton and King (1979) report the results of model tests in which piles founded in granular
material are subjected to tunnelling settlements. The tests suggest that the pile behaviour in
granular soils may be influenced more by movement of the pile base than by the build-up of
negative skin friction, but the tests were carried out at low stress levels and it is not known in
what manner the results are affected by the model scale.

It is likely that the presence of piles within a settlement trough will significantly interact with
the ground and provide considerable stiffening. It may be possible to attempt to model piles as
a form of soil reinforcement and further study is required in this respect.

FRCP/5 63
 5 Precautionary and protective measures

5.1 SITE INVESTIGATION

In order to make sensible predictions of the ground displacements associated with underground
construction, it is essential to have a thorough understanding and site specific knowledge of the
grcund conditions and likely variations within the area of a particular tunnelling project. The
importance of site investigation in tunnelling has been well emphasised and the trend in recent
years of increased effort in phased site investigation works (see Appendix H) is reflected in
published case histories. The comparison of predicted and actual ground conditions carried out
by the Transport and Road Research Laboratory (IRRL) for several UK projects is valuable in
highlighting the differences and the need for further refinements (see the references of
Dumbleton and his co-authors, and those of West and his co-authors).

The essential process of pre-construction site investigation is likely to include the following:

• an appraisal of site history

• an understanding of regional and local geology, geomorphology, topography, and
hydrogeology
• knowledge of previous construction, including wells and previous investigatory works
• rigorous phased site investigation (possibly including large diameter boreholes, adits or shafts)
• location of artificial obstructions.

This process, if well conceived and effected, represents one of the most valuable precautionary
measures associated with tunnelling. Site investigation should not cease once construction
commences, but should be an integral part of it, particularly in the planning and implementation
of any of the construction expedients discussed below. Continuity of input by the same
engineers throughout the process of investigation and design is highly desirable.

Probing ahead of the tunnel face can provide advance warning of adverse conditions which
might cause additional settlement. At present, probing ahead is little used because the time
taken interrupts tunnel advance. Often a number of holes are needed to explore the tunnel cross
section adequately, and the limited distance that can be probed in an acceptable time gives only
days or even hours notice of an impending problem (BRETIRRL, 1975). In certain
circumstances probing will require a special valve/gland arrangement to prevent ingress of
groundwater and soil, both of which could result in additional settlement.

Open-hole techniques are often employed for probing ahead, since they are normally more rapid
than coring. However, these provide only a crude indication of ground conditions, and thus the
main application of this simple probing is to forewarn of potential inflows of water. In the
future it is possible that improved drilling methods, and instrumentation will allow more
information to be obtained with less disruption, and probing ahead may then be used more
widely. The current trend is, however, for tunnelling techniques to be developed to cope with
the full range of expected ground conditions, avoiding the need to probe ahead, except in
special circumstances.

5.2 CONDITION SURVEYS

To assess the effects of underground construction on nearby structures, it is essential that their
condition prior to construction is determined and a schedule of existing defects is recorded. This
may include detailed structural measurement, observation of cracks and verticality, history of
previous repair, location and type of strengthening measures, lists of defects and photographic
records. In some instances it will be desirable to inspect the nature and condition of
foundations. It is common practice in pre-tunnel surveys to broadly classify cracks as follows:

64 FR/C P/5
 • just visible to naked eye — hairline cracks
• up to 2 mm — minor cracks
• over 2 mm — major cracks.

In addition, local deviation of slope from the horizontal or vertical of more than 1/100 will
normally be clearly visible (Building Research Establishment, 1981).

In assessing any building, it is usual to include an inspection of the basement and the roof, and
for high risk structures to include a selection of intermediate floors — increased as necessary
depending on the defects found. To protect all the parties involved, it is good practice to record
the condition of all structures at risk of any level of damage, that is wholly or partially in Zones
of Risk Categories 2 to 4 of Table 9. There may be instances where particularly sensitive or
important structures which although apparently not 'at risk' by such a classification may lie
within the overall zone of influence of tunnelling and require condition surveys. Wherever
possible the condition survey should be directed at an assessment of the building response. This
may include taking account of the location, orientation, spacing and persistence of cracks as
well as identification of potential planes of weakness or existing strengthening measures that
may modify the structural behaviour.

During construction it is often necessary to record the interim condition of a property and
produce an updated schedule of defects, both internal and external. Photography is also often
used to supplement such an inspection. On occasion, some internal redecoration may be carried
out in order to maintain good relationships but also to assist in locating any defects that develop
subsequently. It is important that construction records are maintained and level surveys are
carried out regularly and systematically. Intermediate condition surveys during tunnel
construction at appropriate intervals may be required to ensure that building damage is correctly
attributed.

There are practical difficulties in assessing the pre-construction condition of services and it is
likely that assessment of age, type, the risk of possible damage and assessment of safety will
determine the need for diversion or isolation from the effects of movement, or acceptance of
such movement.

It has been suggested by Howe et al. (1981) that where calculations predict changes in pipe
strains in grey cast iron mains exceeding about 100 microstrain, the proposed scheme should be
appraised by the relevant statutory authority. Close links with all possible utility
owners/authorities are encouraged and these should be established in the planning stage to
enable systems potentially at risk to be identified and an agreed course of action to be
implemented concurrent with the scheme development.

A trial scheme of one-call 'Freefone' inquiries has been set up in Scotland since 1980 in an
attempt to reduce the risk of damage to utilities at the construction stage. This system which
deals with the location of buried pipelines has been backed up with computerised location
records and by patrolmen who visit sites to establish personal contact. This system may be
extended throughout the UK and be of general assistance during the construction stage.

5.3 PLANNING ALTERNATIVES

During the various stages through feasibility to final design, it will be necessary to assess the
effects of construction upon nearby structures and services, either initially in broad terms, or
finally in detail. In this design process there may be options open to the client/owner and the
designer to avoid potential damage, such as those suggested in Table 12.

The cost of direct protective measures has to be weighed against the cost and implications of
the more radical solutions which may include an observational approach and deal with damages
and claims as they arise. This has certainly been the approach adopted quite successfully in
some cases.

FR/CP/5 65
 Table 12 Planning Alternatives
Tunnel related Structure related Service related

re-alignment Purchase of property Relocation/diversion

(vertical or Demolition of property Replacement
horizontal) Relocation of property Strengthening
Strengthening Supporting
Supporting Isolating or shielding
Shielding

5.4 PROTECTION OF STRUCTURES

5.4.1 General

The main forms of protective measures currently available which directly modify the response
of the building to the effects of tunnelling fall into five broad groups as described below.

1. Measures that aim to shield the structure by the installation of a physical barrier between the
building foundation and the tunnel. This form of bather is referred to as a 'Curtain Walling
System'. The barrier is not structurally connected to the building foundations and so does
not provide direct load transfer. The intention is to modify the shape of the settlement trough
hence reducing the effects adjacent to the protected building.

2. Measures that compensate for settlements by injection of grout into the ground below the
building. The ground replacement techniques are variously known as fracture grouting,
squeeze grouting and compaction grouting.

3. Measures that attempt to negate the effect on the building of ground movements by the
introduction of an alternative foundation system that enables relative movement to be
eliminated. These techniques are generally referred to as underpinning.

4. Structural jacking to compensate for settlement.

5. Measures that strengthen the building in order that it may sustain the additional stresses
induced by ground movements.

The above forms of precautionary measures will only be applied after careful consideration of
the present building condition and the likely behaviour with and without the strengthening
works. Prediction of such behaviour is difficult. Where the structure is at low risk the most
practical and cost effective approach is generally to repair any damage caused, subsequent to
tunnelling once movements have substantially ceased.

5.4.2 Shielding

Structures close to tunnelling operations and underground construction may be shielded from
the effects of associated ground movements. Methods that can be considered include the
following:

• diaphragm walling
• soil/grout barriers
• contiguous bored pile walls
• sheet piling
• mini piles.

All of the above are relatively specialised fields, some of which cover a diverse range of
construction methods. The designer should be aware that most of the methods will not prevent
movement of the surrounding soil. The very processes of installation can induce movements
which may be as large as or of a more damaging nature than the movements they are designed
to prevent, a factor which must be carefully weighed against the expected benefits. Additionally,

66 FR/CP/5
 attempted shielding of structures against vertical displacement may not necessarily prevent
damage due to lateral displacements.

A mini pile curtain wall will minimise installation-induced settlements and can be undertaken
by compact rotary drilling units able to operate in confined spaces such as basements. With this
technique it may be possible to provide a perimeter curtain wall from inside a building thus
avoiding services and obstructions within the adjacent road and pavement area. The possible
limitations to this type of wall is the depth that piles can be installed (the practical limit for a
nominal 200 mm dilmeter pile is about 18 m) and the speed of installation because of the large
number of piles that are likely to be required.

A further shielding application of piles is their use as 'poling piles' to form a roof support
between the structure and tunnel. This technique can be beneficial for escalator shafts which
connect to the surface close to or within buildings.

5.4.3 Compensation grouting

Effective control of structural damage from tunnelling-induced settlements can be achieved by

`active' ground replacement techniques. These techniques compensate for ground loss by the
injection of grout into a zone of ground above the tunnel. As the settlement wave crosses the
building a compensating heave is induced which closely balances the settlements in both
location and magnitude. Successful application depends on accurate monitoring of ground and
building movement, precise timing and exact control of quantity and location of grout injection.
The two principal methods are fracture growing, which is generally used in cohesive soil, and
compaction grouting which is generally employed in granular materials.

Fracture grouting techniques, such as the `Soil-Frac' system, may be considered as a phased
operation which is summarised as follows:

(a) installation of monitoring system;

(b) installation of injection tubes, which may require the sinking of a shaft;

(c) initial injection of grout;

(d) interactive injection during tunnelling works to control settlements in accordance with
predetermined criteria and the monitoring system;

(e) injection of bentonite post-construction (optional);

(f) return to site for post-construction control of time-dependent settlement (optional).

Operations (a) and (b) must be undertaken well in advance of the tunnel construction to enable
correction of settlements that they themselves may generate and to establish the 'normal'
background movements that the protected structure experiences. Further details of this process
are given in Appendix I.

Compaction grouting involves the injection of stiff mortar, normally into granular material.
Once the material in the vicinity of the injection hole is compacted, subsequent injection results
in ground heave and hence jacking will occur. This technique can also be used to compensate
for volume loss during tunnelling. Where the tunnelling works themselves are within granular
soils, effective surface settlement control may be achieved by treating a narrow zone by
compaction grouting close to the crown of each tunnel. However, in the situation where
tunnelling takes place within clay and the affected buildings are founded on the overlying fills
and gravels (as is typical in London) compaction grouting is applied just below the foundations
with injection points at close centres (1-1.5 m). As with fracture grouting a comprehensive
monitoring system (as described above) is required in order to control grouting locations and
volumes.

FR/CP/5 67
 5.4.4 Underpinning

Underpinning as a precautionary measure requires the work to be undertaken prior to the

tunnelling works affecting the building and is only effective when taken below the zone of
influence or when jacking is used in conjunction with partial underpinning.

The main forms of underpinning are as follows:

1. Traditional mass concrete. This is installed by excavating below the foundation in short
sections, casting a new mass concrete footing. Depths are normally limited to 2-3 m and
further restricted by high water tables.

2. Underpinning by grouting. This includes compaction grouting and jet grouting as described
above.

3. Piling. This can be divided into large diameter and mini pile systems.

Large diameter piles are likely to be required in order to achieve a total underpinning where the
loads are taken down to below the tunnelling-affected zone for buildings close to the works.
However, they can only be installed externally with standard rigs and in order to connect the
piles to the existing structure would then require cantilevered pile caps to transfer the load.
Complete access around the perimeter of a building is likely to prove a severe limitation.

Mini piles can be installed using compact drilling units able to operate in confined basement
areas. Raking piles drilled through the existing foundations and thereby attached to it would
provide the load transfer mechanism. The length limitation referred to in Section 5.4.2 also
applies. Although most practical mini piling schemes may not be able to can-y the total building
load, the presence of close spaced mini piles within the soil mass will provide 'soil
reinforcement'. This is likely to result in the installation redistributing and to some degree
limiting the predicted settlements.

As part of any overall piling scheme other strengthening measures such as reinforced concrete
wall beams may be required in order to accommodate the new load paths created when
transferring the structural loads.

All forms of piling constructed within the zone affected by tunnelling-induced displacement will
be subjected to a significant negative skin friction causing down-drag, and, in some cases,
differential forces causing pile tension. Both of these effects may severely limit the load
carrying capacity of the piles.

5.4.5 Jacking

Jacking systems can take two basic forms:

(a) Used in conjunction with an alternative support system, usually piled underpinning, to
enable adjustment of structural level to be made in order to compensate for the behaviour
of the support system during and after the transfer of load. In this way adjustments are
possible if a partial underpinning system is subject to tunnel-induced ground movements.

(b) The existing foundations are utilised and adjustments to building level are achieved by the
partial or complete decoupling of the superstructure.

Jacking systems, as described above, comprising hydraulically operated jacks may be linked by
computer control in conjunction with electronic monitoring systems to form a complex 'state of
the art' method.

68 FR/CP/5
 5.4.6 Strengthening

Where a condition survey reveals potential structural weaknesses or discontinuities within a

building it may be appropriate, based on a risk assessment and cost benefit analysis, to
strengthen the building. The measures may include the following:

(a) Shoring and bracing of walls using timber or steel support frames to provide additional
stability and reduce induced stresses.

(b) Internal propping between walls and bracing of openings to limit pOtential distortion.

(c) Installation of tie bars to tie walls together using high tensile steel rod. This measure may
be applied in order to limit potential tensile cracking at the top of a structure through
bending mode distortion where the building lies on the outer hogging limb of the settlement
trough. Installation of the tie bars may involve significant disruption and damage within the
building as walls are drilled, sections of floors lifted and anchor beams fitted. Tie bar
schemes are normally unsuitable for increasing the 'shear' performance of buildings due to
the presence and requirements for continued usage of window and door openings.

5.5 PROTECTION OF SERVICES

Old mains can be replaced by new materials which are often more robust and flexible. Old cast
iron pipelines (particularly gas) can be renovated by 'trenchless' replacement with polyethylene
pipes of 75 mm and 150 mm diameter. The process involves a machine being drawn through
the iron main which expands and fragments the pipe as it progresses, simultaneously pulling
through a PVC sleeve. Subsequently a polyethylene main is placed within this protective sleeve.

Where horizontal movement is anticipated parallel to new pipes, a sleeve can be installed to
permit slip displacement between the pipe and the surrounding soil. In some situations the
pipeline can be exposed by removal of the surrounding soil and cradled across an excavation to
minimise the effects of ground movement. It is also possible to shield services using the
techniques described above for structures.

5.6 METHOD OF TUNNEL CONSTRUCTION

Various aspects of the method of tunnel construction will influence the associated ground
movements including the following:

• tunnelling method
• workmanship
• rate of advance/pauses
• tail void-grouting/sealing
• support system
• watertightness of lining.

The tunnelling method chosen must be consistent with general engineering considerations and
overall economics, but must always allow safe excavation of the tunnel and erection of an
appropriate tunnel lining. The choice is influenced by the tunnel size, depth and length, ground
conditions, programme and economics. One of the criteria for safe working is limiting damage
to structures and services. Although methods which allow safe and efficient working in the
tunnel normally provide adequate control of ground displacements, techniques can be adopted
with the object of reducing the magnitude of such displacements where this is necessary.

Tunnelling techniques are briefly discussed here in order of increasing displacement control.
However, no method is appropriate for all ground types, so control of displacement cannot
necessarily be the limiting consideration.

FR/CP/5 69
 (a) A heading, timbered as necessary, is feasible in cohesive soil with a low stability number
or in granular soil with some apparent cohesion above the water table. Generally headings
are limited to small short drives, typically not more than say 2 x 3 in cross-section. Ground
displacement is very sensitive to quality of workmanship and variation in ground
conditions.

(b) Heading excavation with early support by shotcrete and associated means, as in the NATM,
can give good control of settlement in some ground conditions. The excavation sequence,
primary support properties and timing of invert closure have to be carefully designed and
then adjusted as necessary on the basis of results of monitoring performance. Large cross-
sections and complex layouts have been constructed successfully using such methods in
stiff clays. In granular soils some cementing or apparent cohesion is necessary and water
inflows below the water table have to be controlled by ground treatment or groundwater
lowering.

(c) An open-faced shield or tunnelling machine provides control of radial ground movements
behind the face and is appropriate in conditions where the face is stable with minimal
support. Some face support can be provided by shelves or breasting boards and face rams
mounted in a shield, or by the cutting head of a machine.

(d) A closed-face or blind shield can be used in ground which is sufficiently soft and plastic to
allow the shield to be advanced, the soil being extruded through adjustable apertures in the
face. Displacement associated with this technique can be variable and unpredictable, and
better control can be obtained with closed-face tunnelling machines. Such machines are
available with variants which monitor the face support pressure and accept spoil through
variable width slots in the cutting head. They are used in soils where water inflow either is
not a problem due to low permeability or is controlled by other methods.

(e) Slurry machines provide ground support and control water inflow by using slurry under
pressure in the face. The slurry circulates to remove the excavated spoil and after passing
through a separation plant is recirculated. A thixotropic slurry such as bentonite is often
used for this purpose, which has the added advantage of being absorbed onto the excavated
surface to form a cake and assist in stabilising the face and hence reduce ground
displacement.

(f) The earth pressure balance machine, is a variant of the slurry machine, theoretically
providing a continuously varying pressure balance at the face. The rate at which spoil is
released from the chamber ahead of the near face bulkhead, is governed by a screw
conveyor. The consistency of the spoil may be controlled by the injection into the face
chamber of mud or slurry making agents. Many variants of this approach have been
developed in Japan, with a confusing variety of names.

There are insufficient case histories covering different tunnelling methods in comparable
conditions to quantify the influence on ground displacements of each method. Experience
indicates that the correct application of a method appropriate for the ground conditions and the
quality of workmanship are of paramount importance in attaining the minimum ground
displacement.

Ground displacement and hence settlement tends to be greater where the rate of advance is slow
or interrupted. Experience also shows that greater ground displacements tend to occur at the
start of a tunnelling contract, following a normal learning curve as the work force, machinery
and systems are toned up. Steering difficulties may also be associated with increased settlement
as well as the passage from a shaft or from a zone of treated to untreated ground. It should be
considered if drives can commence outside critical sections to allow for learning curve
development of the tunnelling operations, and whether continuous working (24—hour, 7-day
working) can be be adopted to minimise these effects.

To reduce the ground loss due to closure of the annular gap around the outside of the shield,
grouting should be effected as soon as possible after erecting the lining and advancing the

70 FRCP/5
 shield. For cases where the ground is liable to collapse onto the lining immediately, it is
possible to grout during advance of the shield, using a non-setting or slow-setting grout mix,
but a grout stop is needed in the tail of the shield. This technique has not been widely used and
problems of grout adhering to the tailskin and resulting in irregular or exacerbated movements
have been recorded by Glossop et al. (1979).

In certain circumstances, a pilot tunnel may be desirable, from which pre-drainage or pre-
treatment of the ground for the main tunnels drives can be completed. The direct knowledge of
the ground obtained during the driving of the pilot tunnel may suggest modifications to the
tunnelling machine and method or sequence of working. A pilot tunnel may also identify areas
of poor or difficult ground, thereby reducing unforeseen conditions and effect economies.

5.7 GEOTECHNICAL PROCESSES

The behaviour of various soil types, both above and below the water table, can be related to
stand-up time, the determination of which may indicate the need for some stabilising measures.
To control the ground and ground water and hence ground displacements during tunnelling,
there are several geotechnical processes available. The applicability of these processes in
various ground conditions is discussed more fully in Appendix I. However it should be
emphasised that these geotechnical processes can themselves induce ground displacements and
some of the adverse effects that have been noted are outlined in Table 13.

Table 13 Possible adverse effects of geotechnical processes on ground displacements

Process Adverse effects Reason/cause

Grouting Heave/lateral displacements tuns or High injection pressures, ungrouted zones

loss of soil/settlement

Dewatering Settlement, negative skin friction on Inadequate filter: possible loss of fines and/or
piles consolidation

Freezing Heave, runs or loss of soil Poor installation control, expansion of soil/water
ungrouted zones

Compressed air Settlement, heave, disruption at Water ingress/loss of fines/blow outs

surface

Earth pressure balance Radial displacements, reversals of Change in earth pressure on passage of shield
displacements

5.8 OBSERVATION/INSTRUMENTATION

5.8.1 General

The combination of observation and recording of construction procedures together with

measurement of displacements of the ground and the deformation of structures can in
appropriate circumstances provide a valuable means of assessing and improving performance.
High quality records may also provide evidence in the settlement of liabilities for damage and
repairs attributed to tunnelling.

The use of ground instruments to monitor tunnelling has increased over the past decade but few
well-documented accounts have been published. The amount of data generated by
comprehensive monitoring can become so voluminous as to be unmanageable unless the
programme has been carefully conceived and managed. To this end, the objectives must be
clearly identified. The data obtained must be processed quickly, ideally on site, and presented in
an intelligible form to be of immediate value to the contract. The use of instrumentation should
in no way diminish direct visual observations, as noted by Peck (1969a).

FR/CP/5 71
 Monitoring may be used to aid determination of in-situ properties for pre-contract design, to
verify design assumptions or to study and improve field performance, but in all instrumentation
programmes, the designer must be clear about the purpose of each instrument. These may
include:

• a check on the adequacy of design parameters

• control and optimisation of construction procedures or schedules
• potential for reduction of construction costs
• verification of long-term performance
• legal reasons
• verification of safety during and after construction or public reassurance
• advancing the state of the knowledge of ground response.

The need, location and selection of instrumentation is a matter requiring considerable

knowledge and expertise and a rational approach to such a programme is given for guidance in
Appendix J.

It is inevitable that instrumentation will be damaged during construction and allowance should
be made for this when designing the programme. A 40% instrument redundancy is often
adopted to allow for losses. Every effort should be made to protect the instruments and to
explain and involve the entire workforce in the programme; this can greatly improve survival of
the instrumentation and ancillary equipment. There is a growing body of opinion in the United
States that measurement of sub-surface displacements associated with tunnelling should be
specified in all cases for construction monitoring as stated by Hansmire and Cording (1985).

Installation of some instruments modifies the ground environment and both the instrument and
the host ground may require time to adjust and reach equilibrium. Consequently, installation and
critical monitoring may be required before the main construction commences. This enables such
factors as 'background' noise or scatter, operator variations, reading error, instrument error or
drift, seasonal variations, creep, temperature, etc., to be established and appropriate corrective
actions taken.

5.8.2 Measurement of displacement and deformation

Measurement of the ground displacements and the corresponding deformation of structures can
be determined by various means. The measurements may be made directly at the surface or
indirectly from sub-surface installations.

The range of instrumentation available and widely used in tunnelling projects is summarised in
Table 14. The advantages and limitations of these types of instruments have been detailed by
Dunnicliffe (1982).

In any system designed to monitor displacements, it is essential that the measurements are
related to a permanent, stable and accurate bench mark or reference point. For vertical
movement it may be necessary to establish a deep bench mark to depths below the seat of
vertical movements. However, even deep bench marks may be subject to significant movements
arising from regional changes, such as the lowering or recovery of ground water levels at depth.
Such changes are occurring for example in London, Bangkok and a number of Japanese cities.

Having estimated the range of ground losses and the associated pattern of likely displacements
for a particular tunnel, it is good practice to verify these at an early stage of a project. Ideally
an area representative of the ground conditions, but preferably a convenient 'open-space'
location should be selected.

72 F R/C P/5
 Table 14 Categories of instruments for measuring displacements, after Dunnicliffe (1982)
Category Type of measured displacement

4-► 1 r ❑ 0
Surveying methods • • •
Portable displacement gauges • •
Single point monuments • • •
Vertical pipe settlement gauges • •
Remote settlement gauges • •
Heave gauges • •
Inclinometers (electsolevels) • •
Borehole extensometers • •
Soil strain gauges • •

Notes: Horizontal displacement 4-►

Vertical displacement
Radial displacement
Surface displacement ❑

Sub-surface displacement Cl

The observations required may include the following:

1. Vertical displacements

(a) Sufficient cross-section surface settlement (or heave) points to define the shape and
extent of the transverse settlement trough.
(b) Sufficient surface settlement points along the centre line of the tunnel to define the
longitudinal surface settlement profile and to check the variation in settlement along the
tunnel drive.
(c) Sub-surface settlement measurements outside the zone directly affected by the method of
construction, both above and to the side of the tunnel using extensometers installed from
the ground surface in advance of construction.

2. Lateral displacements

(a) Surface movement points.

(b) Sub-surface horizontal movement ahead of or adjacent to the tunnel using inclinometers
or electrolevels depending upon the predicted magnitude of the movement.

It is critical that displacements recorded at significant stages as the tunnel passes the instrument
array are supplemented by comprehensive documentation of construction activities, such as
compressed air pressure, ground loss, stoppages, pressure grouting, progress of lining, etc.

The degree of further instrumentation may be dependent upon the size of the project as well as
the sensitivity of certain structures. Key locations should be selected for instrumentation and, in
addition to the above observations, measurement of both the structures and the intervening soil
could be required. It should be recognised that the instruments themselves can provide
preferential paths for loss of compressed air or ingress of water if located close to the tunnel, as
noted in Appendix I. The measurement of pore-water pressures around a tunnel- could be
required in some circumstances and is of considerable interest in understanding the
consolidation effects of tunnels in clay strata.

The number of different observations and the sophistication of the instrumentation within a
particular building will depend on proximity to the tunnel, the degree of risk and the sensitivity
of the structure to the predicted movements.

FRCP/5 73
 It is likely that a comprehensive range of monitoring vertical, lateral, surface and sub-surface
displacements will be justified in relatively few instances and these tend to fall into two main
categories, namely:

(a) protection of special buildings

(b) research studies.

The benefits of detailed monitoring may not be readily apparent or, in some instances, of
interest to the client. From the practical viewpoint there is considerable merit in increasing the
number of simple routine points along the centre line, and transversely at appropriate intervals,
that can be monitored by standard levelling surveys. The published results of these alone could
greatly improve knowledge of ground loss and settlement over a greater range of ground
conditions and depth of tunnelling.

5.8.3 Measurement of load and strain

Instruments for measuring load and strain fall into two groups, load cells and strain gauges. In
each case, the sensors measure small extensions or compressions. Strain gauges are attached
directly to the surface of or embedded within a structural member or pipeline. Load cells are
interposed in or between structural members in such a way that structural forces pass through
the cells.

The ability to instrument a building can be severely restricted either by difficult access or by
opposition of owners/occupiers. In the latter case this may be as a result of suspicion or a
feeling of insecurity, when to the lay observer the need to monitor may imply uncertainty.

It is unusual to require routine instrumented monitoring of pipelines. If there was any residual
doubt concerning safety and performance during construction this would normally have been
eliminated by positive measures and a conservative approach during the design process.

74 F Ft/C P/5
 6 Direction and need for further studies

6.1 CASE STUDIES

Whilst preparing this report it has become abundantly clear that there is a need for many more
case studies that provide comprehensive, well-documented, records of displacements, as well as
deformations and strains occurring in nearby structures, coupled with information on the levels
and consequences of distress or damage to those structures affected.

The application and reliability of current analytical methods is limited, due largely to the
paucity of field data against which to calibrate them. This situation will only be rectified by
high quality testing to define soil parameters adequately, and by well designed instrumentation
schemes with monitoring before, during and after construction coupled with accurate and
detailed records of tunnel construction and structural performance.

The more extreme cases of displacement and/or damage are of particular interest as they can
represent boundary or critical conditions, but unfortunately these are most likely to be
suppressed, often as a result of litigation. The reporting of well documented failures is of
considerable benefit to the profession, providing general understanding and help in avoidance of
repetition of similar events.

6.2 VARIATIONS ASSOCIATED WITH TUNNELLING METHODS

The measurement of ground displacements associated with earth pressure balance or slurry
machines has highlighted differences between the pattern of sub-surface displacement associated
with these machines and those associated with more traditional tunnelling methods, although the
shape of the surface settlement profile appears to be similar. It is likely that the use of these
machines will increase because of their versatility and suitability to a wider range of ground
conditions, including mixed faces, and their reduced dependence upon a specialised mining
work force. Therefore there is a need to confirm the surface expression of settlement and to
improve the knowledge of the relationship between face pressure and the pattern of sub-surface
displacements.

European experience of the observational tunnelling technique (NATM) in firm to stiff clays
suggests that surface settlements are no greater than those predicted for more traditional shield
tunnelling methods. However, the use of these techniques in such ground conditions is not well
established and further feedback from such construction in soft ground is required. Details of
the variations in the sequence of excavation and timing of invert closure and their influence on
accompanying settlements and subsurface displacements are particularly important.

6.3 NON LINEAR SMALL STRAIN SOIL BEHAVIOUR

-

The influence of the highly non-linear elastic behaviour of soils at small strains, which has been
recently well demonstrated by specialist triaxial testing, needs to be assessed. This behaviour
may assist in understanding the observed soil response around tunnels. In addition it may be
useful in obtaining better results from some finite element modelling.

6.4 LONG TERM EFFECTS

-

This report has highlighted the uncertainties associated with assessment of long-term effects
associated with tunnels in clay. There is a suggestion that the movements are more widespread
in the long term, such that rigid body displacement of structures may occur without significant
damage. However, the monitoring of ground movements and corresponding structural

FRCP/5 75
 performance in the long term above tunnels in clay is considered an important topic. There is
growing awareness of the potential for long-term movement and therefore a need for reliable
case history examples to demonstrate the magnitude and time relationship of such movements.

6.5 RADIAL MOVEMENT

The validity of the widely adopted assumption that volume loss, associated with initial
movements during construction, is generated by radial movements towards the centre of the
tunnel needs to be demonstrated. This may be achieved by observation of arrays of sub-surface
instruments that are able to distinguish accurately both horizontal and vertical components of
movement. The present assumption may result in unduly conservative predictions of horizontal
tensile strains especially in the hogging region of the settlement trough. This assumption also
influences the assessment of soil/structure interaction for piled foundations within the zone of
movement.

6.6 RISK ASSESSMENT

The risk assessment put forward in this document relates to two simply predicted ground
parameters, settlement and slope, which can then be used for a very generalised approach
during the initial alignment optimisation stage of a project. The Boscardin and Cording (1989)
approach identifies the importance of horizontal tensile strains and the nature of the structural
response — particularly in relation to the dominance of either shear or bending deformation.
There is a need to further develop the understanding of the behaviour of real structures to these
tunnelling related deformations, and to identify the prime factors which dominate their response
and control the resulting tilting, shearing or bending, and to robustly link these to limiting
criteria which can be reliably predicted.

6.7 HORIZONTAL STRAINS

The assumption that horizontal ground strains generated by tunnelling can be directly
transferred to structures may be unduly conservative. Knowledge of the development of strains
across the soil/structure interface or across boundaries between cohesive and granular soils in
multi-layered soils is a fundamental requirement to improving risk assessment of structures
based on this parameter together with the need to relate this to assumed levels for the neutral
axis in structures.

6.8 BEHAVIOUR OF STRUCTURES

There is a need to improve the understanding of the behaviour of different foundation types and
structural forms, including that of piled foundations. However, prestigious masonry structures of
historical significance may be particularly sensitive to ground movements and therefore of more
critical importance to the selection of alignment in cities.

6.9 PILES AS GROUND REINFORCEMENT

The presence of piles within the settlement trough are likely to provide an effect similar to soil
reinforcement and very significantly alter the ground response to tunnelling. It is likely that the
piles will smooth out the trough beneath a particular building, and case history data are required
to demonstrate this aspect of soil/structure interaction.

76 F R/C P/5
 6.10 MODELS

There is a need to develop models to demonstrate structural modes of response, particularly

with regard to varying E/G and L/H ratios, applied horizontal strain and the influence of
strengthening measures such as ties. In addition there is a need for models that can demonstrate
the influence of openings and existing planes of weakness within structures upon E/G ratios and
hence structural response.

FRCP/5 77
 References (4,indicates key reference)

ALEXANDER, S.J. and LAWSON, R.M. (1981) Design for movement in buildings. CIRIA
Technical Note 107.

ASH, J.L. and RUSSEL, B.E. ( 1974) Improved Subsurface Investigation for Highway Tunnel
Design and Construction. Federated Highway Administration, Office of Research HRS-11,
Washington, DC, USA.

ATKINSON, J.H. and POTTS, D.M. (1977) Subsidence above shallow tunnels in soft ground.
Proc. Am. Soc. Civ. Engrs. — J. Geotech. Engng. Div. , Vol. 103, No. GT4, pp. 307 — 25.

'ATTEWELL, P.B. (1978) Ground movements caused by tunnelling in soil. Proc. Conf. Large
Ground Movements and Structures, Cardiff, 1977. Published as Large Ground Movements and
Structures. Edited by J.D. Geddes. Pentech Press, Plymouth, pp. 812 — 948,

ATTEWELL, P.B. and FARMER, I.W. (1972) Determination and interpretation of surface and
sub-surface ground movements during construction of a tunnel in London Clay at Green Park,
London. Report to TRRL and Road Research Laboratory, Department of the Environment,
Contract No. ES/LN/842/101.

ATTEWELL, P.B. and HURRELL, M.R. (1985) Settlement development caused by tunnelling
in soil. Ground Engineering, Vol. 18, No. 8, pp. 17 — 20.

ATTEWELL, P.B. and WOODMAN, J.P. (1982) Predicting the dynamics of ground settlement
and its derivatives caused by tunnelling in soil. Ground Engineering, Vol. 15, No. 8,
pp. 13 — 36.

*ATTEWELL, P.B. and YEATES, J. (1984) Tunnelling in soil. In: Ground Movements and
their Effects on Structures. Edited by P.B. Attewell and R.K. Taylor. Surrey University Press,
pp. 132 — 212.

•ATTEWELL, P.B., YEATES, J. and SELBY, A.R. (1986) Soil movements, induced by
tunnelling and their effects on pipelines and structures. Blackie, Glasgow.

BOSCARDIN, M.D. and CORDING, E.J. (1989) Building responses to excavation-induced

settlement. J Geotech Engng, ASCE Vol. 115, No. 1, January, pp. 1 — 21.

BRITISH STANDARDS INSTITUTION (1982) Code of Practice for Safety in Tunnelling in

the Construction Industry. BS 6164.

BRITISH STANDARDS INSTITUTION (1985) Structural use of Concrete: Part 2, Code of

Practice for special circumstances. BS 8110.

BUILDING RESEARCH ESTABLISHMENT (1981, rev 1990) Assessment of damage in low-

rise buildings. BRE Digest No. 251, July, revised 1990.

BJERRUM, L. (1963) Evaluation of allowable settlements. Proc. 3rd Euro. Conf. Soil Mech.
and Found. Engg, Wiesbaden, Vol. 2, pp. 135-7.

BRANDT, C.T. (1970) A systems study of soft ground tunnelling. Report by Felix and Scisson
and A.D. Little U.S. DOT-FRA-OHSGT-231 Department of Transport, Washington, DC

BRE/TRRL WORKING PARTY (1975) Probing ahead for tunnels: a review of present
methods and recommendation for research. Transport and Road Research Laboratory,
Supplementary Report 171.

78 FR/CP/5
 BREM, G. (1981) U-Bahnhof Schweizer Platz in Frankfurt am Main. Berechnungs — and
Meperzebrisse Mr die dreischiffige unterirdisch aufgefalrene Bahnhofshalle. Forschung Praxis
27 STUVA

BROMS, B.B. and BENNERMARK, H. (1967) Stability of clay at vertical openings. Proc. Am.
Soc. Civ. Engrs — J. Soil Mech. and Found. Div., Vol. 93, No. SMI, pp. 71 — 94.

BRUCE, D.A. (1984) The drilling and treatment of overburden Geodrilling, August and
October, Paper originally presented at Drillex '84 Conference.

*BUBBERS, B.L. (1980) Safety aspects of tunnelling in an urban area. Proc. Int. Symp. The
Safety of Underground Works, Brussels, pp. 66 — 85.

*BURLAND, J.P. and WROTH, C.P. (1975) Settlement of buildings and associated damage.
Proc. Conf. Settlement of Structures, Cambridge, pp. 611 — 654 and discussion pp. 764 — 810.
(Reprinted as Building Research Establishment, Current Paper 33/75, 1975)

BURLAND, BROMS, B.B. and De MELLO, V.F.B. (1977) Behaviour of foundations and
structures. Proc. 9th Int. Conf. Soil Mech. and Found. Engg., Tokyo. Vol. II, pp. 495 — 546.

CATER, R.W. and SHIRLAW, J.N. (1985) Settlements due to tunnelling in Hong Kong.
Tunnels & Tunnelling, Vol. 17, No. 10 October, pp. 25 — 28

CATER, R.W. SHIRLAW, J.N., SULLIVAN, C.A. and CHAN, W.J. (1984) Tunnels
constructed for the Hong Kong Mass Transit Railway. Hong Kong Engineer, October,
pp. 37 — 49.

CLOUGH, G.W., SWEENEY, B.P. and FINNO, RJ. (1983) Measured soil response EPB shield
tunnelling. Journal of Geotech Engng. ASCE, Vol. 109, No. 2, February, pp. 131 — 149

*CLOUGH, G.W. and SCHMIDT, B. (1981) Design and Performance of Excavations and
Tunnels in Soft Clay. In: Soft Clay Engineering, edited by E.W. Brand and R.P. Brenner.
Elsevier Scientific, Amsterdam

CORDING, EJ. and HANSMIRE, W.H. (1975) Displacements around soft ground tunnels.
Proc. 5th Pan Am. Conf. Soil Mech. and Found. Engg., Buenos Aires, Vol. 4, pp. 571 — 633.

CORDING, EJ., O'ROURKE, T.D. and BOSCARDIN, M. (1978) Ground movements and
damage to structures. International Conference Evaluation and Prediction of Subsidence,
Pensacola Beach, Florida, pp 516-537.

CORDING, EJ., HANSMIRE, W.H., MacPHERSON, H.H., LENZINI, P.A. and

VONDEROHE, A.D. (1976) Displacement around tunnels in soil. Final report by University of
Illinois to Department of Transportation, Washington DC. (NTIS, PB, 267356).

CRAIG, R. N. (1975) Discussion. Tunnels and Tunnelling, Vol. 7, No. 6, pp. 61 — 65.

DEERE, D.U., PECK, R.B. MONSEES, J.E. and SCHMIDT, B. (1969) Design of Tunnel
Liners and Support Systems. Report for US Department of Transportation. OH SGT
Contract 3 0152, No. PB 183 799 NTIS, Springfield, Virginia, USA, p. 287.

DEIX, F. and GEBESHUBER, J. (1987) Erfahrungen mit det NOT beim U-Bahn-Ban in Wein.
Felsbau 5, No. 3.

DUMBLETON, M.J., COOPER J.N., FOWLER, P.P. and TOOMBS, A.F. (1978) Site
investigation aspects of the River Medway Cable Tunnels. Department of the Environment,
Department of Transport, TRRL Supplementary Report SR 451.

FRCP/5 79
 DUMBLETON, N.J. and TOOMBS, A.F. (1978) Site investigation aspects of the Empingham
Reservoir Tunnels. Department of the Environment, Department of Transport, TRRL Report
LR 845,

DUMBLETON, MJ. and PRIEST, S.D. (1978) Site investigation aspects of the River Tyne
Siphon Sewer Tunnel. Department of the Environment, Department of Transport, TRRL Report
LR 831.

DUMBLETON, MJ. KEDDIE, A.R. and PRIEST, S.D. (1978) Site investigation aspects of part
of the Tyneside South Bank Interceptor Sewer. Department of the Environment, Department of
Transport, TRRL Supplementary Report SR 454.

DUMBLETON, MJ. and WEST, G. (1976) A guide to site investigation procedures for tunnels.
Department of the Environment, TRRL Report LR 740.

DUMBLETON, MJ. and TOOMBS, A.F. (1976) Site investigation aspects of the Sydenham
Road Sewer Tunnel. Department of the Environment, TRRL Supplementary Report SR 235 UC,
TRRL.

DUMBLETON, MJ. and WEST, G. (1976) Preliminary sources of information for site
investigation in Britain. (Revised edn.) Department of the Environment, TRRL Report LR 403.

DUMBLETON, M.J. and WEST, G. (1974) Guidance on planning, directing and reporting site
investigations. Department of the Environment, TRRL Report LR 625.

*DUNNICLIFFE, J. (1982) Geotechnical Instrumentation for Monitoring Field Petformance.

National Co-operative Highway Research Program Synthesis of Highway Practice 89.
Transportation Research Board, National Research Council, Washington DC, April.

DUNNICLIFFE, J. (1988) Geotechnical Instrumentation for monitoring field performance

J. Wiley, Chichester.

FliA'ATRICK, L., KULHAWY, F.H. and O'ROURKE, T.D. (1981) Flow patterns around
tunnels and their use in evaluating problems. Proc. 6th Pan Am. Conf. Soil Mech. and Found.
Engng., Lima 1979, Speciality Session. Published in Soft-Ground Tunnelling. Edited by D.
Resendiz and M.P. Romo, A.A. Balkema, Rotterdam, pp. 95 — 103.

*GEDDES, J.D. (1984) Structural design and ground movements. In Ground Movements and
their Effects on Structures. Edited by Attewell, P.B. and Taylor, R.K. Surrey University Press,
London.

GLOSSOP, N.H. (1978) Soil deformation caused by soft ground tunnelling. PhD. Thesis.
University of Durham.

GLOSSOP, N.H., SAVILLE, D.R., MOORE, J.S., BENSON, A.P. and FARMER, I.W. (1979)
Geotechnical aspects of shallow tunnel construction in Belfast estuarine deposits.
Proc. Tunnelling '79, London, pp. 45 — 56.

GOLDER ASSOCIATES and JAMES F. MACLAREN LTD, (1976) Tunnelling Technology —

an appraisal of the state of the art for applications to transit systems. Ontario Ministry of
Transportation and Communications, May.

GRANT, R., CHRISTIAN, J.T. and VANMARCKE, E.H. (1974) Differential settlement of
buildings. Proc. Am. Soc. Civ. Engrs — J. Geotech Engng Div. Vol. 100, No. GT9,
pp. 973 — 991.

GUMBEL, J.E. (1983) Analysis and Design of Buried Flexible Pipes PhD. Thesis. University of
Surrey.

80 FR/CP/5
 HANSMIRE, W.H. (1975) Field measurements of ground displacements about a tunnel in soil.
PhD. Thesis, University of Illinois at Urbana.

HANSMIRE, W.H. and CORDING, E.J. (1972) Performance of a soft ground tunnel on the
Washington Metro. Proc. N. Am. Rapid Excavn. and Tunn. Conf. Chicago, Vol. 1, pp. 371 — 89.

HANSMIRE, W.H. and CORDING, E.J. (1985) Soil tunnel test section : Case History
Summary. Proc. Am. Soc. Civ. Engrs — J. Geotech. Engng. Div., Vol. 111, No. 11, pp. 1301 — 22

HARRIS, C.W. and O'ROURKE, T.D. (1983) Response of Jointed Cast Iron Pipelines to
Parallel Trench Construction. Geotechnical Engineering Report 83-5, March, Cornwell
Unversity, Ithaca, NY.

HEUER, R.E. (1976) Catastrophic ground loss in soft ground tunnels. Proc. Rap. Excay. Twin.
Conf. Las Vagas, pp. 278 — 95.

HOWE, M., HUNTER, P. and OWEN, R.C. (1981) Ground movements caused by deep
excavations and tunnels and their effect on adjacent mains. Proc. 2nd Int. Conf. Ground
Movements and Structures, Cardiff 1980. Published as Ground Movements and Structures.
Edited by J.D. Geddes. Pentech Press, Plymouth, pp. 812 — 840.

HOWLAND, A.F. (1981) The prediction of the settlement above soft ground tunnels by
considering the groundwater response with the aid of flow net construction. Proc. 2nd Int. Conf.
Ground Movements and Structures, Cardiff 1980. Published as Ground Movements and
Structures. Pentech Press, Plymouth, pp. 345 — 58.

HURRELL, M.R. (1985) The empirical prediction of long-term surface settlements above
shield-driven tunnels in soil. Proc. 3rd Int. Conf. Ground Movements and Structures, Cardiff
1984. Published as Ground Movements and Structures. Edited by J.D. Geddes. Pentech Press,
Plymouth, pp. 161 — 72.

HULME, T.W., POTTER, L.A.C. and SHIRLAW, J.N. (1989) Singapore Mass Rapid transit
system: construction. Proc. ICE. Part 1, August pp. 709 — 770 Vol. 86.

THE INSTITUTION OF STUCTURAL ENGINEERS (1978, 1989) State of the art report,
Structure — Soil Interaction, April 1978. (Revised and Extended in 1989) Inst. Strict. Engnrs.,
London.

JACKSON, A.A. and CROOME, D.F. (1962) Rails through the Clay. A History of London's
Tube Railways. George Allen and Unwin, London.

KERISEL, J. (1975) Old structures in relation to soil conditions. 15th Rankine Lecture.
Geotechnique, Vol. 25, No. 3, pp. 433 — 83.

KIMURA, T. and MAIR, R.J. (1981) Centrifugal testing of model tunnels in soft clay. Proc.
10th Int. Conf. Soil Mech. and Found. Engng. Stockholm, Vol. 1, pp. 319 — 22.

KUESEL, T.R. (1972) Soft ground tunnels for the BART project. Proc. N. Am. Rapid Excavn.
and Tunn. Conf., Chicago, Vol. 1, pp. 287 — 313.

LAKE, L.M. and NORIE, E.M. (1982) Application of horizontal ground freezing in tunnel
construction — two case records. Tunnelling '82, Instn. Mining and Metallurgy, London,
pp. 283 — 9.

LEE, K.M. and ROWE, R.K. (1989) Deformations caused by surface loading and tunnelling:
the role of elastic anistropy. Geotechnique Vol. 39, No. 1, pp. 125 — 40.

MAIR, R.J. (1979) Centrifugal modelling of tunnel construction in soft clay. PhD. Thesis,
Cambridge University.

FR/CP/5 81
 'MAIR, R.J., GUNN, M.J. and O'REILLY, M.P. (1982) Ground movement around shallow
tunnels in soft clay. Tunnels & Tunnelling, Vol. 14, No. 5, pp. 45 — 8.

°MEG AW, T.M. and BARTLETT, J.V. (1982) Tunnels — Planning, Design Construction.
Ellis Horwood, Chichester.

MEYERHOF, G.G. (1956) Discussion on paper by A.W. Skempton and D.H. MacDonald:
Allowable settlement of buildings. Proc. Instn. Civ. Engrs. Pt. II, S, PP. 774 — 5.

MORTON, J.D. and KING, K.H. (1979) Effects of tunnelling on the bearing capacity and
settlement of piled foundations. Tunnelling'79. Proc 2nd Int. Symp. Instn. of Min. & Metal.
BTS/IME/TRRL London March, IMM 1979

MOLLER-SALZBURG, L., SAUER, G., and CHAMBOSSE, G. (1977) Berechnungen,

Modellversuche and in-situ Messungen bei einen bergmannischen Vortrieb in tonigem
Untergrund. Bauingenieur, 52.

NATIONAL COAL BOARD (1975) Subsidence Engineers Handbook. NCB Mining

Department.

NATIONAL WATER COUNCIL (1977) Sewers and Water Mains — a national assessment.
Department of Environment Standing Technical Committee, Report No. 4. National Water
Council, June.

NEYER, J.C. (1984) Soft ground tunnel failures in Michigan. Proc. Int. Conf. Case Histories in
Geotech. Engng., Missouri, Vol. 3, pp. 1429 — 33.

NORGROVE, W.B., COOPER, I. and ATTEWELL, P.B. Site investigation procedures adopted
for the Northumbrian Water Authority's Tyneside Sewerage Scheme, with special reference to
settlement prediction when tunnelling through urban areas. Proc. Tunnelling '79, London,
pp. 79 — 104.

'O'REILLY, MP. and NEW, B.M. (1983) Settlements above tunnels in the United Kingdom,
their magnitude and prediction. Proc. Tunnelling '82, Brighton 1982, pp. 173 — 181. Report of
discussion. Trans. Inst. Mining Metallurgy, Vol. 92, Section A, pp. A35 — A48.

O'REILLY, M.P. (1988) Evaluating and predicting ground settlements caused by tunnelling in
London Clay. Proc. Tunnelling '88, London, The Institution of Mining and Metallurgy,
pp. 231 — 41.

O'REILLY, M.P., MAIR, R.J., ALDERMAN, G.H. (1991) Long-term settlements over tunnels:
an eleven-year study at Grimsby. Proc. Tunnelling '91, London. Institution of Mining and
Metallurgy/Elsevier Applied Science, London, pp. 55 — 64.

O'ROURKE, T.D. (1978) Tunnelling for Urban Transportation. A review of European

Construction Practice. Report No. UMTA-IL-06-0041-78-1, US Department of Transportation.
August.

'O'ROURKE, T.D., CORDING, E.J. and BOSCARDIN, M.(1978) Damage to brick-bearing

wall structures caused by adjacent braced cuts and tunnels. Proc. Conf. Large Ground
Movements and Structures, Cardiff 1977. Published as Large Ground Movements and
Structures. Edited by J.D. Geddes. Pentech Press, Plymouth, pp. 647 — 71.

O'ROURKE, T.D., CORDING, E.J. and BOSCARDIN, M. (1976) The Ground Movements
Related to Braced Excavation and their Influence on Adjacent Buildings. US Department of
Transportation, Washington, DC.

O'ROURKE, T.D. and TRAUTMANN, C.H. (1982) Buried pipeline response to tunnelling
ground movements. Proc. Europipe '82 Conference, Switzerland, Paper 1, pp. 9 — 15.

82 F R/C P/5
 OWEN, R.C. (1985) Vegetation and seasonal ground movement effects in buried mains.
Proc. 3rd Int. Conf. Ground Movements and Structures, Cardiff 1984. Published as Ground
Movements and Structures. Edited by J.D. Geddes. Pentech Press, Plymouth, pp. 145 — 60.

PECK, R.B. (1969a) Advantages and limitations of the observational method in applied soil
mechanics. Ninth Rankine Lecture, Geotechnique, Vol. 19, No. 2, pp. 171 — 187

*PECK, R.B. (1969b) Deep excavations and tunnelling in soft ground. Proc. 7th Int. Conf. Soil
Mech. and Found. Engng, Mexico City, Vol. 3, pp. 225 — 90.

PECK, R.B., HENDRON, AJ and MOHRAZ, B. (1972) State of the art of soft-ground
tunneling, Proc. lst North American Rapid Excavation and Tunnelling Conference, Chicago,
Vol. 1 pp. 259 — 86.

POLSHIN, D.E. and TOKAR, R.A. (1957) Maximum allowable non-uniform settlement of
structures. Proc. 4th Int. Conf Soil Mech. and Found. Engng, London, Vol. I, pp. 402 — 405.

POTTS, D.M. (1976) Behaviour of lined and unlined tunnels in sand. PhD Thesis, University of
Cambridge,

ROWE, R.K., LO, K.Y. and KACK, G.J. (1983) A method of estimating surface settlement
above tunnels constructed in soft ground. Canadian Geotech. J., Vol. 20, No. 1, pp. 11 — 22.

ROWE, R.K. and KACK, GJ. (1983) A theoretical examination of the settlements induced by
tunnelling: Four case histories. Canadian Geotech. J, Vol. 20, No. 2, pp. 299 — 314.

*SCHMIDT, B. (1969) Settlement and ground movements associated with tunnelling in soil.
PhD Thesis, University of Illinois, Urbana.

SCHULTZ, E.W. (1975) Methoden zur Verhinderung von schadlichen Setzungen

bei Unterfahrungen and verankerten Baugruben auhand von Beispielen aus dem U-und S-Bahn
Ban in Frankfurt/Main. Forschung ad Praxis, 19, STUVA 1975.

SKEMPTON, A.W. and MacDONALD, D.H. (1956) The allowable settlements of buildings.
Proc. Instn. Civ. Engrs, Vol. 5, No. 2, (Part III), pp. 727 — 68.

TERZAGHI, K. (1956) Discussion paper by Skempton and MacDonald on Allowable

settlements of buildings. Proc. Instn. Civ. Engrs, Vol. 5, No. 3, (Part III), pp. 775 — 77.

WARD, W.H. and PENDER, MJ. (1981) Tunnelling in soft ground — General Report. Proc.
10th Int. Conf. Soil Mech. and Found. Engng. Stockholm, Vol. 4, pp. 261 — 76.

WEST, G.W. (1980) Geophysical and Television Borehole Logging for Probing Ahead of
Tunnels. Department of the Environment, Department of Transport, TRRL Laboratory
Report 932.

*WEST, G.W. (1983) Comparisons between real and predicted geology in tunnels: examples
from recent cases. Q.J. Engng. Geol. Vol. 16, pp. 113 — 26.

WEST, G.W. and EWAN, VJ. (1981) Site investigation and construction of the Dinorwic
Diversion Tunnels. Department of the Environment/Department of Transport, TRL LR 984.

WEST, G.W. and McLAREN, D. (1981) Site investigation and construction of the Cardiff
Cable Tunnel. Department of the Environment, Department of Transport, TRRL LR 1012.

WEST, G.W. and TOOMBS, A.F. (1978) Site investigation and construction of the Liverpool
Loop and Link Tunnels. Department of the Environment, Department of Transport,
TRRL LR 868

F R/CP/5 83
 'WEST, G.W., CARTER, P.G., DUMBLETON, MJ. and LAKE, L.M. (1981) Site investigation
for tunnels. Int. J. Rock Mech. MM. Sci. & Geomech. Abrcts. Vol. 18(5), pp. 345 — 67.

WILUN, Z. and STARZEWSKI, K. (1972) Soil Mechanics in Foundation Engineering. Vol. 2

Theory and Practice. Intertext, London.

WOOD, J.G.M. (1985) Engineering assessment of structures with alkali-silica reaction.

Concrete Society Conference, London.

YEATES, J. (1985) The response of buried pipelines to ground movements caused by

tunnelling in soil. Proc. 3rd Int. Conf. Ground Movements and Structures. Cardiff 1984.
Published as Ground Movements and Structures. Edited by J.D. Geddes. Pentech Press,
Plymouth, 1985, pp. 145 — 60.

84 FRCP/5
 Appendix A Equations for magnitude of settlement
on longitudinal profile

Above the tunnel axis, the settlement at a point x, ahead of the tunnel can be derived by
considering the contribution of each point along the tunnel:

w f exp
4711, _
px-x0)21 .dx eqn (i)

where w is the maximum settlement in the fully developed tranverse settlement trough, as
before

The trough length parameter (4) has the same value as the trough width parameter. The integral
cannot be calculated by ordinary means but can be readily evaluated by numerical integration or
from statistical tables of the standardised normal distribution:

Cs) f exp H
-t2 .d, eqn (ii)

xo
w= * eqn
iF

FR/CP/5 85
 Appendix B Factors influencing allowable
additional deformation

(a) Type of deformation

Tunnelling and associated works commonly cause displacements with both vertical and
horizontal components acting in She transverse and longitudinal directions. The displacements
may occur in any direction relative to the position of the sub-surface excavation, though when
close to the tunnel drive may approximate to a radial pattern of displacement, either inwards or
outwards.

The response of structures to these displacements will vary depending substantially upon its
rigidity; this is illustrated schematically in Figure 24.

Differential horizontal deformation is often more damaging than differential vertical deformation
of an equal magnitude and even structures on shallow foundations, underpinned to protect
against vertical displacement, may be damaged by lateral displacement. Such movements can
occur during the passage of a shield or during underpinning works. Reversals of displacement,
are potentially more serious than displacements in only one direction. Displacements resulting in
hogging of a structure are potentially more critical than those causing sagging.

(b) Rate of deformation

The two phases of displacement recognised are the immediate phase which accompanies
excavation and tunnel construction and the post-construction phase. The rapid initial ground
displacement can be more harmful to structures than the longer term ground consolidation and
creep displacements or self-weight induced structure settlement. In the latter case, the rate of
displacement may be such that stress redistribution and creep in the structures may
accommodate deformation without necessarily exhibiting distress.

(c) Magnitude and distribution of deformation

The magnitude and distribution of deformation imposed upon a structure by underground works
is likely to be different from that generally experienced during and after the construction of that
structure. Tunnelling-induced deformations are superimposed upon those already experienced by
the structure; only in particular circumstances might a structure have been designed with this in
mind. Furthermore, newer structures may be less tolerant of deformation, and require more
careful assessment of their capacity to accommodate further deformations. The effect of
superimposition of deformation from tunnelling upon existing self-weight induced deformation
may appear disproportionately large where buildings are particularly sensitive either by virtue of
existing stress or strain levels, design or construction.

An assessment of the magnitude and distribution of displacement is greatly assisted by the use
of consistent and unambiguous descriptive terms. The definitions of ground and foundation
displacements proposed by Burland and Wroth (1974) are widely accepted, and are used in this
report. They are illustrated in Figure 24 for rafts, pad/strip, and cellular/basement foundations.
The generally accepted names, symbols and definitions of the displacements and deformations
are listed in the Glossary, but it should be noted that use of these terms in the published
literature is not always consistent. Further, whilst building distortion terms are readily defined
on paper, interpretation in practice may present many difficulties.

To quantify all the distortional parameters for a particular structure, a large number of
observation points are required together with detailed knowledge of the foundations and
superstructure. Such information is seldom available, particularly when the initial appraisal of
the effects of tunnelling adjacent to nearby structures is carried out. Consequently, to determine
the structures likely to be at risk, simple limiting criteria are required to describe the

86 FRCP/5
 characteristics of the potential settlement trough and the typical resultant distortions in relation
to general structure types. Those structures shown to be at risk can then be investigated
individually and a procedure is discussed in detail in Section 4.

(d) Type, construction and condition of structure

Factors having a bearing upon the deformation that would be acceptable in structures include :

• age, history and existing condition

• alterations and repairs
• the specific type and form of construction
• the type and quality of materials used in construction
• the quality of workmanship at the time of construction
• the quality or robustness of design and attention to detail.

Knowledge of the age, history and condition of the existing structure or facility assists in the
assessment of deformation already undergone and hence the capacity to accept additional
deformation. Geographical setting and age can be an indicator of the type of materials used in
construction. Time-dependent changes in the properties of construction materials may be
significant. Experience has shown that structural alterations or repairs introduce zones or planes
of weakness which are commonly found to be susceptible to ground displacements caused by
tunnelling. By way of illustration the following comments are a broad generalisation of
European and North American experience and practice:

1. Older structures (pre 1930) usually incorporate hardwood or better quality softwood, and
newer structures poorer quality softwood;
2. Pre 1900 brick structures commonly have a lime mortar rather than Portland cement
mortar, the latter having higher strength, stiffness and weathering resistance, but being less
tolerant of distortion;
3. Masonry and brick structures tend to become more brittle with age;
4. Older gas and water pipes are commonly grey cast iron which is a brittle material subject to
corrosion and not tolerant of tension or movement;
5. Cast iron and steel mains are subject to corrosion, concrete is attacked in some conditions
such as sulphates or alkali silica reaction and some plastics can be susceptible to
embrittlement.

(e) Interactive soil/structure effects

The presence of services and structures may significantly modify the soil behaviour and pattern
of displacement compared with that expected in a 'green field' site, either by reducing or
concentrating the displacements.

The extent to which a structure deforms and the mode of deformation exhibited, i.e. bending
and shearing in response to ground displacements resulting from underground construction, are
functions of its strength and structural stiffness. Depending upon proximity to underground
works and the magnitude and direction of ground displacement, structures may have sufficient
strength and stiffness to modify and smooth out the ground deformations and redistribute the
forces induced. In situations where the structure is on shallow foundations and the ground
displacements are small and essentially vertical, interaction can be assessed, although in
practice, this situation rarely occurs. Where the structure is founded at a deeper level and close
to the underground works, the pattern of ground deformation becomes more complex and lateral
displacements may become larger or more significant. Here, assessment of the interaction and
the resulting effects may be extremely difficult. In urban areas, piled foundations commonly
occur close to tunnel drives and the interactive effects are not readily resolved analytically.
However a practical approach to this situation is given in Section 4.

Within a particular structure there may be elements of differing stiffness or strength,

concentrating deformation into narrow zones within the structure or between structural elements.

FR/CP/5 87
 In some instances ground displacements may cause gaps to form between the structure and the
ground, or sliding may occur at an interface, thereby modifying the contact stresses.

(f) Orientation and location of structure

O'Rourke et al. (1977) highlights two prominent axes of structural deformation for brick
bearing walls:

(1) perpendicular to the bearing walls — which can result in loss of floor support

(2) parallel to the bearing walls — which can cause bending, shear and direct tensile
deformation.

In brick or masonry structures, simple end bearing is the most widely used type of floor or
beam support; the bearing length is commonly about 100 mm. For deformations perpendicular
to the bearing walls, O'Rourke et al. calculated that the maximum safe lateral displacement
between bearing walls would be of the order of 60% of this value.

88 FR/CP/5
 Appendix C Classification of visible damage

Category of damage Degree of damage Description of typical damage Approximate of Threshold

a)(ease of repair Is shown In crack width a) architectural
italics) (mm) damage - crack
width

Aesthetic Hairline cracks of less than >0.1

about 0.1 mm width are
classed as negligible

1. Very Slight Fine cracks which can easily be >1.0 In plastered walls
treated during normal 0.4 mm
decoration. Perhaps isolated
slight fracturing in building. In unplastered
Cracks in external brickwork block, brick or
visible on close inspection rough concrete
walls 0.8 m

2. Slight Cracks easily filled. <5.0 In joints of tiled

Redecoration probably floors 1.6 mm
required. Several slight
fractures showing inside of
building. Cracks are visible
externally and some re-pointing
may be required externally to
ensure weathertightness. Doors
and windows may stick slightly

3. Moderate The cracks require some 5 to 15 or a

opening up and can be patched number of cracks
by a mason. Recurrent cracks >3.0
can be masked by suitable
linings. Repointing of external
brickwork and possibly a small
amount of brickwork to be
replaced Doors and windows
sticking. Service pipes may
fracture. Weather-tightness
often impaired.

Functional service- 4. Severe Extensive repair work involving 15 to 25 but also

ability breaking out and replacing depends on number
sections of walls, especially of cracks
over doors and windows.
Windows and door frames
distorted, floors sloping
noticeably. Walls leaning or
bulging noticeably, some loss
of bearing in beams. Service
pipes disrupted

Structural 5. Very Severe This requires a major repair Usually >25w but
involving partial or complete depends on number
rebuilding. Beams lose bearing, of cracks
walls lean badly and require
shorting. Windows broken with
distortion. Danger of instability

Notes:

1. There is no simple relationship between serviceability and degree of visible damage.

2. It must be emphasised that in assessing the degree of damage, account must be taken of both the location and market value of the
building or structure.

3. Crack width is one factor in assessment and should not be used on its own as a direct measure of damage.

4. Classification of visible damage, based on I.Struct.E (1978) and O'Rourke et al (1976).

FRCP/5 89
 Appendix D Acceptable vertical deflection limits
for structural elements

Element Criterion Allowable deflection

Vertical Deflections
Beam Steel beam total deflection Span/200
Reinforced concrete beam Span/250 or 30 mm
Cracking of brick or blockwork partition* Span/500 or 15 mm
Cracking of lightweight partition* Span/350 to Span/360
or 20 mm
Live load visible deflection* Span/360
Upward deflection because of precamber Span/300
Floors or roofs Differential settlement Span/250 to Span/500
depending on cladding
Timber flooring Span/330
Paved or asphalt covering Span/250
Flexible short span roof sheeting Span/125
Movement of sensitive equipment (e.g. 1 in 750 slope
generator) (for example)
Cantilever Visible deflection* Span/180
Cracking on cladding Span/250 to Span 500,
(relative movement along edge) depending on cladding
Gantry girder Inefficient travel of overhead crane Span/700
Lateral deflections
Column Side-sway of multi-storey buildings* Height/1000
recommended
Failure of frame with diagonals 1 in 600
Racking or walls or infills of masonry Height/500
structure
Single storey or low-rise flexible frame ' Height/300
Visible deflection of canopy roof Height/250
Mullions Bending of support to glazing Span/175
Gantry girder Crane rail separation* Span/500

Source: After Alexander and Lawson, 1981.

Note: All deflections are serviceability limits under worst total loading except the following

* Installation after de-propping of floors.

+ Imposed short-term loading.

90 FFVCP/5
 Appendix E National Coal Board (1975)
classification of subsidence damage

Change of length Class of damage Description of typical damage

of structure (mm)

Up to 30 1. Very slight or Hair cracks in plaster. Perhaps isolated

Negligible slight fracture in the building, not visible
on outside.

30 — 60 2. Slight Several slight fractures showing inside the

building. Doors and windows may stick
slightly. Repairs to decoration probably
necessary.

60 — 120 3. Appreciable Slight fracture showing on outside of

building (or one main feature). Doors and
windows sticking; service pipes may
fracture.

120 — 180 4. Severe Service pipes disrupted. Open fractures

allowing weather into the structure and
requiring rebonding. Window and door
frames distorted; floors sloping noticably;
walls leaning or building noticeably. Some
loss of bearing in beams. If compressive
damage, overlapping of roof joints and
lifting of brickwork with open horizontal
fractures.

More than 180 m 5. Very severe As above, but worse, and requiring partial
or complete rebuilding. Roof and floor
beams lose bearing and need shoring up.
Windows broken with distortion. Severe
slopes on floor. If compressive damage,
severe buckling and bulging of the roof
and walls.

FR/CP/5 91
 Appendix F Damage criteria for services

(a) Performance of burled pipe

Research carried out by Gumbel (1983) on buried pipes has been related to pipe performance in
trenches and mainly considers the nature and compaction of the backfill and the ring stiffness of
the pipe. However, in relation to tunnelling, longitudinal deformation and bending of the pipes
and pipelines are of prime concern and these have been described by Harris and O'Rourke
(1983).

O'Rourke and Trautmann (1982) identified two boundary modes of deformation for pipes,
see Figure 32

(a) perfectly flexible: bending and flexural strain following the displacement which
may lead to rupture or intolerable deformation;

(b) perfectly rigid but individual rigid pipe lengths, with rotation and axial slip at
with flexible joints: joints — leading to leakage or disengagement.

+r I
N
1 11
Distributed settlement

a) Perfectly flexible pipeline

b) Perfectly rigid pipeline

Figure 32 Boundary modes of deformation of pipes

The performance of pipelines depend upon:

• history and existing condition

• relative stiffness of pipe and soil
• movement capacity of any joints
• location of any joints relative to shape of displacement profile
• resistance to shear between the soiVbackfill and the pipe.

A comparison by Attewell and Yeates (1984) of the properties of typical pipe material for
short- term loading in direct tension is given in Table 15. It shows that the elastic strain
equivalent to the design stress for cast iron is around 400 to 500 microstrain, but this increases
by a factor of around 20 for plastic pipes.

92 FR/CP/5
 Table 15 Typical pipe material properties for short-term static loading in direct tension

Material Ultimate tensile Typical max. design Elastic strain

stress (N/mm2) stress for working equivalent to design
loads (N/mm2) stress (microstrain)

Grey iron 110 — 215 27 — 54 400 — 500

spun (430 — 490)*
pit cast (370)

Ductile iron 420 (min) 155 940

(grade 420/12) (820)'

Mild steel 410 (min) 95 — 140 450 — 660

(grade 410)

Plastics
(a) UPVC 45 (min) 20 7000

(b) MDPE 30 at 50 nun/min 7 10000

(c) HDPE 32 at 125 mm/min 8 9000

(type 2)

Source: Attewell and Yeates 1984.

Notes:

1. Properties related to gradually applied, non-repeated loading without creep and at 20°C
2. Properties of grey iron depend on age of manufacture, lower values apply to material pit cast before year 1914
3. UPVC — unplasticized polyvinyl chloride
MDPE — medium density polyethylene
HDPE — high density polyethylene
4. Denotes values used by British Gas.

To assess the longitudinal bending of a pipeline and to estimate the corresponding extreme fibre
bending strain Attewell and Yeates (1984) presented a theoretical linear/elastic method of
analysis. It was claimed that the method overestimates pipe bending moments and that
prediction of pipe stress should be within 50%, provided the correct assumptions are made
about joint stiffness and position.

(b) Cast Iron pipes as an Indicator

The majority of older water mains and gas distribution pipes are of grey cast iron, and in urban
areas these may constitute up to 90% of the existing system. The National Water Council
(1977) estimated that there are some 450 000 kms of such pipe in the UK. Grey cast iron is a
brittle material with a failure strain significantly lower than that tolerated by the more modern
ductile iron, steel or plastic pipes. In addition, the joints of cast iron mains were commonly
caulked with lead or cement, and prior to caulking, the bell ends of the joints were often packed
with oakum. The wet oakum in water mains tends to expand when the joint deforms reducing
joints between older gas pipes tend to leak at less rotation and axial slip than the modern
equivalents, although shrinkage is now avoided in most areas by conditioning with
monoethylene glycol. Since cast iron mains are more susceptible to movement, and the limiting
criteria for failure are more stringent, they have been subject to considerable study in order to
provide design criteria for movement in relation to the effects of tunnelling and adjacent
excavations. Although most of the information available relates to cast iron, services constructed
of other materials may also be at risk and require individual assessment.

The following guidance on the behaviour of cast iron mains has been suggested by O'Rourke
and Trautmann (1982).

(a) allowable slip at joints = 25 mm

(b) allowable rotation = 0.5 — 1%

FR/CP/5 93
 (c) less than 200 mm diameter perform as relatively flexible
(d) more than 200 mm diameter perform as relatively rigid

The maximum design stress for grey iron at working loads is one quarter of the ultimate tensile
stress (U'S) as given by Attewell and Yeates (1984) and this level is set to limit non-elastic
strain, that is permanent set. In any assessment of damage potential, the objective should be to
determine if the pipe stress will exceed the maximum design stress, although for temporary
loading, Attewell and Yeates (1984) suggests that this could be increased to 30% UTS As yet,
there is no consensus of opinion. Limiting criteria for bending strain vary between countries and
statutory authorities, and it is possible that the limits suggested by Attewell could be raised for
some cast-iron mains.

Howe et al. (1981) have highlighted the importance of strain history of existing cast iron mains.
Most of the cast iron mains are old, installation was often poorly controlled, and the backfill
and bedding were poorly specified; it is likely that considerable post-installation deformation of
the mains has occurred. Owen (1985) has indicated that 'locked in' strains can also be
introduced during fabrication and result from seasonal temperature changes as well as ground
displacements caused by traffic and vegetation. From the figures in Table 15Table-F4, it may
be inferred that many older mains may already be strained to a critical level and incapable of
sustaining further tunnel-induced deformations. The susceptibility of metal pipes to damage is
also dependent on the existing level of corrosion.

O'Rourke and Trautmann (1982) approached this subject empirically and derived a tentative
relationship between cast iron pipe diameter and a measure of the slope of the settlement
trough. The limiting values above which damage may occur were identified as follows:

= 0.012 for relatively rigid pipes, diameter more than 200 mm (i.e maximum slope of
1/140)
= 0.02 to 0.04 for relatively flexible pipes, diameter less than 200 mm (i.e maximum
slope of 1/80 to 1/40)

However the data base for this relationship is extremely limited and additional information is
required before it can be applied with confidence.

94 FR/CP/5
 Appendix G Derivation of points of limiting slope
and settlement

The simplified expressions for slope and settlement in terms of volume loss, radius and trough
width parameter are given below. A programmable calculator can be used to solve the
expression for limiting values of slope and settlement for risk assessment and the results for two
examples are tabulated. These give calculated values of corresponding critical trough widths
from tunnel centre line for various tunnel depths, and for two different values of volume loss.
Figures 26 and 28 show the relationship between the zones of risk, depth of tunnel and volume
loss, and from these the dominance of either slope or settlement as the critical parameter can be
gauged.

Assuming the surface settlement trough corresponds to an inverted normal distribution curve:

settlement w -Y2
wmax exP

slope _ —YW
.2

maximum slope = 0.607 wmax

jy
It can be shown that:
r2
maK = 0.01251'
T

when VIis the volume of the settlement trough expressed as a percentage of the excavated
tunnel volume.

If it is assumed that i = 0.5zo, the following are obtained:

F2
maximum settlement, w = 0.025V, —
z„

maximum slope, 0.0301'i 1-2

Ze
-2y2 )
settlement = wpm= ( 2
zo
(Yvv.)exp( -2y2
slope
4z„

For a critical value of settlement less than the maximum, the corresponding values of y can be
obtained from equation Equation (iv) cannot be solved for y algebraically, but is readily
solved by iteration to give values of y for critical slopes.

F R/C P/5 95
 Table 16 Example 1: Calculation of critical trough width from tunnel centre line (y.t)
for limiting values of slope and settlement

Risk category
2 3 4
depth w..1 max w= slope = w = slope = w= slope =
c.(m) (mm) slope 10 mm 1/S00 SO mm 1/200 7S mm 1/50

For = 2%

4 117 1/28 4.4 6.0 2.6 5.2 1.9 3.7

5 94 1/44 5.3 7.0 2.6 6.0 1.7 3.4
6 78 1/63 6.1 8.0 2.8 6.6 0.8 -
8 58 1/113 7.5 9.6 2.2 7.3
12 39 11254 9.9 113 - -
IS 31 1/397 11.3 11.4 -
20 23 1/705 13.0 - -
25 19 1/1102 14.0 - -
30 16 1/1587 14.1 -

For III = 4%

4 234 1/14 5.0 6.5 33 5.8 3.0 4.6

5 187 1/22 6.1 7.7 4.1 6.8 3.4 5.0
6 156 1/32 7.0 8.8 4.5 7.7 3.6 5.2
8 117 1/56 8.9 10.9 5.2 9.1 3.8 -
10 94 1/88 10.6 12.6 5.6 10.0 3.3
12 78 1/127 12.2 14.0 5.7 10.4 1.7
15 62 1/198 14.4 15.6 5.0 8.2 -
20 47 1/353 17.6 16.4 - - -
25 37 1/551 20.3 - - - -
30 31 1/793 22.6 - - - -

Note: (a) Radius of tunnel excavation, r = 3.05 in

(b) Underlining represents dominance of w or slope
(c) Results plotted on Figure 26

96 FR/CP/5
 Table 17 Example 2: Calculation of critical trough width from tunnel centre line (yam;,)
for limiting values of slope and settlement

Risk category
2 3 4
depth w. MAX W = slope = w = slope = w= slope =
;(m) (mm) slope 10 mm 1/500 50 mm 1/200 75 mm 1/50

For VI = 2%

3 67 1/37 2.9 4.3 1.2 ' 3.7 - 2.4

4 50 1/66 3.6 5.3 0.2 4.4 - -
5 40 1/102 4.7 6.1 - 4.8 - -
6 34 1/48 4.7 6.8 - 4.8 - -
8 25 1/262 5.5 7.6 - - - -
10 20 1/140 5.9 7.4 - - - -
12 17 1/590 6.1 - - - - -
15 13 1/922 5.8 - - - - -
20 10 1/1639 - - - - - -
25 8 1/2561 - - - - - -

For VI = 4%

3 134 1/18 3.4 4.7 2.1 4.2 1.6 3.2

4 100 1/33 43 5.9 2.4 5.1 1.5 3.4
5 80 1/51 5.1 6.9 2.4 5.8 0.9 -
6 67 1/74 5.9 7.8 2.3 6.4 - -
8 50 1/131 7.2 9.3 0.4 6.8 - -
10 40 1/205 8.4 10.3 - - - -
12 34 1/295 9.4 10.8 - - - -
15 27 1/461 10.6 9.7 - - - -
20 20 1/819 11.9 - - - - -
25 16 1/1281 123 - - - - -

Note: (a) Radius of tunnel excavation, r = 2 m

(b) Underlining represents dominance of w or dope

FR/CP/5 97
 Appendix H Recommended phased stages of site
investigation

Stage I: Preliminary appreciation of site and ground conditions

(a) available information
(b) geological and engineering enquiries
(c) air photographs and surface reconnaissance
(d) interpretation and recommendations for next stage of investigation

Stage II: Ground investigation before construction

(a) preliminary ground investigation

(b) main ground investigation
(c) other investigations before construction
(d) interpretation and recommendations for ground investigation during construction

Stage III: Structure/service investigation before construction

(a) location and identification of structures/services

(b) general pre-condition survey/assessment within identified zone
(c) detailed assessment of particular structures or services potentially at risk
(d) identification of special precautions or preventative measures required

Stage IV: Ground investigation during construction

observation, and investigation where necessary, continued during the construction phase to
confirm and supplement the earlier investigations.

(a) observations on ground conditions during construction

(b) probing ahead in tunnels
(c) other investigations during construction
(d) review and amendments of plans and sections
(e) monitoring and feedback on performance during and after construction

Stage V: Structure/service investigation during and after construction

(a) interim assessment of condition of property where required

(b) monitoring and feedback on behaviour during and after construction.

98 FR/CP/5
 Appendix I Notes on geotechnical processes

(a) Compressed air

Compressed air has been used in tunnelling since 1879, but its use has been much reduced in
recent years due to the associated health risks and availability of alternative support systems
(e.g. slurry). The use of compressed air at pressures of 14 psi. X97 kPa) or above requires strict
and often statutory control over the duration of work and decompression.

Compressed air is generally effective in controlling water bearing sands and gravels provided
there is sufficient fine material to prevent large air losses. It may not be effective for controlling
isolated pockets of water bearing silt or sand contained within a relatively impermeable stratum
as the whole pocket becomes uniformly pressurised allowing water to flow. In addition it may
be used to provide additional face support in weak clays to assist in the reduction of settlement.

Air and groundwater pressure can only be balanced at one level of the face and a compromise
has to be obtained between water inflow below balance level and the possibility of air losses
which may lead to face instability above balance level. In addition the pressure within shallow
tunnels has to be limited to prevent heave or a blow in the ground cover to the tunnel. Similarly
care has to be taken to avoid loss of air along preferential paths due to the presence of services,
boreholes, instrumented holes, wells and piles. Therefore, in practice, it is necessary to provide
face support to maintain stability against some ingress of water or unexpected air losses.

(b) Grouting

There is a wide range of grouts and grouting methods available, the purpose of which may be
one or more of the following:

(i) to reduce the permeability of the soil

(ii) to increase the strength of the soil
(iii) to reduce the porosity of the soil (prior to compressed air work)
(iv) to densify loose soils
(v) to fill large voids
(vi) to provide support beneath adjacent buildings/services.

The objective of penetration grouting to fill inter-particle voids and the principal factors
governing the choice of 'penetration' grout mixtures and methods of application are indicated in
Table 18.

Table 18 Factors influencing selection of penetration grouts

Ground characteristics Grout properties

Permeability of the ground. Size of particles in suspension

Flow of ground water Viscosity
Groundwater pressure Setting time and rate of setting
Depth and accessibility of Grouting pressure and injection
zone to be treated rate
Chemical compatibility
Resistance to extrusion
Health and Safety aspects

Grout injection can be carried out from the surface, from a pilot tunnel or from within the main
tunnel. The latter method may halt the tunnel advance which can in some instances increase the
magnitude of ground movements. Grout injection pressures have to be carefully controlled to
prevent heave or excessive lateral movements adjacent to nearby structures or services. Kerisel
(1975) refers to examples during the construction of the Paris Metro where grout injection was

FRCP/5 99
 carefully controlled to produce a small heave prior to tunnelling which was nearly equal to the
settlement caused during the construction work.

Compaction grouting is not used to fill inter-particle voids but instead to densify fine grained
soils in situ. The grout used is a very stiff mortar-like grout (with a low slump) and the grout
generally remains as a homogeneous mass which grows in volume as the injection is continued,
displacing and compacting the surrounding soils. The method does produce considerable up-lift
forces, which in controlled circumstances have been used to raise settled structures. Low slump
(25-75 mm) cement/PFA/sand mortars are injected through 90 mm diameter tubes installed in
pre-drilled locations through the foundations. Grout spheres of up to 1200 litres can be created
in 20 minutes and pumping can be resumed in the same bulb within one hour. When pumping
at the same location takes place at a later time then a further sphere can be created above the
previous one. As grouting takes place close to building foundations a careful assessment of their
condition is necessary and strengthening works carried out if required in order that reaction
pressures from the grouting do not initiate local damage to the structure. Forces imposed on the
foundations from compaction grouting used to compensate actively for tunnelling settlement
should not be as great as in post-construction jacking to restore the building position. Access
around buildings and within basements in order to drill and place injection tubes beneath all
load-bearing walls is required with this technique which would necessitate the loss of access to
ground and basement levels.

One recent method of grouting that is becoming widely used in Japan and the Far East (now
available in the UK) for ground treatment prior to tunnelling works is a process called 'jet
grouting'. This method is usually carried out from the surface and uses high pressure air and
water jet to air lift soil from the base of a pre-bored hole. The combined jetting pressure may
approach a maximum 40 MPa, as reported by Bruce (1984). The void created is simultaneously
refilled with grout such that the water and some of the soil removed is replaced with injected
material to provide a soil/cement mixture that can be varied to suit requirements of
permeability, density or strength. This type of grouting can be used to provide panels using a
unidirectional jet and panels can be linked to provide continuous underground barriers. Columns
(up to 2 m diameter) can be formed by rotating the injection pipes and these can be contiguous
as walls or blocks. This process has been successfully used to control ground movements
associated with tunnelling through soft and loose alluvial soils to a maximum depth of about
45 m. However, it should be noted that the panels or columns cannot be regarded as free
standing or cantilever retaining walls as they are not reinforced and lack tensile strength and
bending resistance. In addition the process itself can cause considerable displacement in the
adjacent soil and surface heave particularly when treatment is attempted at depth within thick
deposits of soft clay. There have been recent uses of this process for treating ground ahead of a
tunnel face using boom mounted equipment.

limited area grouting (LAG) is a chemical grouting system developed by the Japanese, which
has been used for treating a wide range of ground conditions from clays to fine gravels from
the surface using rotary drilling equipment and is described by Bruce (1984). A flash setting
grout (5 seconds) is used and the two chemicals, silicate solution and re-agent, come into
contact and are ejected at the base of the hole from one point in the drill rod assembly. The
chemicals are intermixed with the soil during rotation and withdrawal of the rods. Grouted
ground strengths of 0.2 to 0.5 MPa are common and a treated diameter of 0.6 to 1.0 m per hole
is used for design purposes. This process is now widely used for tunnelling contracts in Asia
and in particular has been used on the MRT project in Singapore.

Double tube drilling and seepage (DDS) is another Japanese method of chemical grouting
which is similar in some ways to the LAG process also described by Bruce (1984). When the
drill rods have reached the required depth a plug within the rod system is activated to expose
six nozzles. Fast setting (10 — 30 seconds) grout (silicate plus re-agent) is ejected through these
nozzles with final mixing occuring at the nozzles and no rotation is required during rod
extraction. Grouting pressures of up to 1.5 MPa are used and the zone of influence may be up
to one metre.

100 FRCP/5
 Fracture grouting has been used in Europe to compensate for the volume loss associated with
tunnelling and comprises three important phases:

(i) Monitoring

A comprehensive monitoring system is an essential element of the scheme as it allows the

injection processes to be assessed and adjusted as the tunnelling work proceeds. A typical
monitoring system would be based on the overflow settlement cell principle. A series of cells
are connected to a master reserve which has a forced maintained content level. Floats within
these cells incorporate electric displacement measuring devices (LVDTs) which follow changes
in water level within the cell arising from the cell's movement relative to the fixed reference
water level.

The signals from the LVDTs are sent to a data logger where a complete set of readings may be
read every two minutes and then transmitted to a lap top computer for analysis.

Computer output in the form of three-dimensional graphs can be produced as well as a display
of up to four cells simultaneously on a real time basis, this being used to monitor heave during
injection.

The resolution of the LVDT is of the order of 0.01 mm, with the system as a whole having a
repeatability of 0.5 mm or better.

Other sophisticated computer-linked systems are also available such as electro-level

inclinometers.

Conventional levelling and surveying techniques together with regular inspection surveys are
also carried out in conjunction with the electronic monitoring system to check the overall
situation at specific times (e.g. once every month) before, during and after tunnel construction.

(ii) Installation of injection tubes

In order to control the settlement by injection it is necessary to install injection tubes to a

precise pattern. Grout injection must cover the whole area of the foundations and is generally
carried out from one or more shafts depending of the size of the building and the allowable
location and diameter of the shafts. Holes are generally horizontal and drilled in two levels to
allow treatment to a layer of thickness of up to 2 m, this also compensates for possible
deviation of holes during drilling. Holes are taken beyond the building to ensure the stressed
zone is totally treated.

It should be noted that the construction of the access shaft and tube installation will themselves
induce small settlements. The installation of the tubing may induce settlements in the order of
5 mm which can, if required, be corrected by initial injection immediately after installation of
each tube rather than at the end of the installation period.

Where possible shafts should be constructed away from buildings; however, this may not be
possible in a congested situation or where the maximum grout tube lengths of 50 — 60 m limit
application. Shafts can be constructed using either conventional segmental linings or, by
forming a ring of small diameter piles with a shotcrete skin. The shaft diameter is ideally
4 — 5 m in order to allow sufficient room for the drilling rig, grout pipe reels and an adequate
array of grout tubes.

(iii) Injection of grout

Injection of grout into the ground is carried out through pneumatic packers. In use the packer is
pushed into the sleeve to be injected and then compressed air is used to inflate the two end
seals. Grout is then injected at a set rate and pressure. Typically the rate of injection is limited
to 5 — 10 litres/minute at a pressures of up to 10 bars. The injection at each sleeve is monitored
and flow is stopped at preset volumes.

FRCP/5 101
 All aspects of the injection operation have to be rigorously controlled with pressure, rate of
injection and volume of injection being monitored and recorded for assessment in relation to the
building response.

Control accuracies of individual machine bases of better than 1 in 5000 for limiting tilt have
been achieved with this process enabling tunnelling projects to be undertaken beneath factories
in which high precision lathes continue to operate uninterrupted.

(c) Dewatering

Dewatering or lowering of the water table over large areas in fine grained soils will cause
consolidation and hence settlement of the ground. The associated surface effects on foundations
and services need to be carefully evaluated as well as the implications of settlement causing
down-drag (negative skin friction) on nearby piled foundations, or deterioration of timber piles
no longer submerged below the water table.

In soil conditions where major temporary modifications to the groundwater regime will not
cause surface settlement and thereby damage nearby buildings or services, dewatering may be
used to control the behaviour of soil at the tunnel face. The techniques include deep pumped
wells, well points, vacuum drainage and electro-osmosis. Where large volumes of soil are to be
dewatered the work is best done from the ground surface, if space is available, or from a pilot
tunnel. The use of deep pumped wells requires careful selection of suitable filters to prevent
loss of fines from the surrounding ground and hence settlement. Ordinary well points are
normally limited to a lowering to about 5 m below pump level and therefore are not commonly
used for normal tunnelling depths unless the jet eductor system is used.

In a situation when complete dewatering is either impracticable or uneconomic, partial lowering

can be used in conjunction with compressed air or slurry tunnelling machines.

In fine-grained soils the likelihood of damaging settlements caused by ground water lowering
must be evaluated at the design stage as the potential zones of influence may be very large.
Recharge wells to limit the zone of influence have been used successfully in special
circumstances.

(d) Freezing

Ground freezing is applicable to a wide range of soils and may offer a suitable method of
treatment in mixed ground conditions where grouting may be inappropriate, or instances where
compressed air would require high pressures. Rapid freezing usually using liquified nitrogen can
be suitable for emergency use in stabilising local conditions.

The presence of interstitial water within the soil is necessary for freezing to be effective. As the
ice is created the soil particles are bound together, an increase in the strength of the ground is
achieved, and an ice wall is formed in time.

Freezing can be achieved by :

(a) Indirect circulation — commonly brine as the secondary circulating coolant

(b) Direct circulation — commonly Freon
(c) Direct injection — nitrogen.

The effectiveness of freezing will be reduced by the presence of flowing ground water which
will supply additional heat, or by leaks when using brine which suppresses the freezing point of
the ground water.

Ground freezing may cause a layer of frost to form between the tunnel lining and the adjacent
soil. When thawing occurs, voids can be left in this zone which may permit lining deformation
or cause loss of ground. Carefully controlled multi-stage grouting during the thaw period has
been successfully used to remedy this problem, as described by Lake and Norie (1982).

102 FR/CP/5
 The process of freezing is associated with the expansion and heave of some types of soil, and
the subsequent thawing may produce proportionally greater settlements. Both these movements
are potentially damaging to nearby structures and services. Services can be isolated by
excavation of the surrounding soil in the freeze area. Lake and Norie (1982) indicate that
installation of freeze pipes can also be a source of ground settlement.

(e) Comparative costs

A review carried out by Brandt (1970) for tunnels up to 6 m diameter indicates that the
comparative costs of groundwater lowering and of using compressed air are approximately
similar and over an order of magnitude less than the cost of freezing for the same project. Work
on the London Underground described by O'Rourke (1978) has shown that when the
maintenance of freezing by brine is limited to five or six weeks, the costs of grouting and
freezing are similar for equivalent tunnelling conditions. However, costs of freezing for
tunnelling works are very sensitive to local practices and commercial conditions.

Bruce (1984) has indicated that the cost of jet grouting is approximately two to three times the
cost of conventional tubed manchette (sleeved tube) penetration grouting per cubic metre of
treated ground. He also indicates that the LAG process is about 20% cheaper and the DDS
system 30% cheaper than the tube (1 manchette process.

F R/CP/5 103
 Appendix J Rational approach to instrumentation
after Dunnicliffe (1982 and 1988)

Sequence of implementation Requirements

1. Defme project conditions geology/geomorphology/geotechnical properties;

groundwater conditions; status of nearby
structures/ services
2. Defme purpose of instrumentation
(see Section 5.8)
3. Select variables to be monitored Groundwater level/pore pressure/load;
vertical/horizontal deformation/strain; tilt
4. Predict likely behaviour Obtain range of likely response identify critical
values for construction control/safety
5. Assign tasks/responsibility Delineate personnel responsible
(owner/designer/specialist contractor) for all facets
of instruments — coordinated by one individual
6. Select instruments On the basis of:
—reliability
—maximum durability
—minimum sensitivity to climate
— proven performance
—self-verification
—cost and delivery

evaluate requirements for sensor, readout and

linkage separately
7. Identify additional observations Readings may be influenced by construction
required details, weather, installation method etc.
8. Select instrument locations Relate locations to predicted behaviour, obtain
data at earliest possible stage in construction (in
some cases prior to construction); maintain
flexibility; determine if fully instrumented
sections representative, check with simple, cheap
intermediate devices
16. List specific purpose of each Each instrument must have a viable purpose
instrument
17. Prepare instrument specifications There are a variety of methods and contractual
arrangements available, well described by
Dunnicliffe (1982)
9. Plan installation Carefully detailed procedures should be prepared
in relation to the specific site needs for
installation, together with comprehensive site
record sheets
10. Plan monitoring procedures Definite guidelines for frequency of monitoring,
calibration, checks on instrument correctness,
processing and interpretation should be prepared
so that the overall strategy is clear and that
contingency plans are available should the data
indicate unfavourable conditions

104 FR/CP/5
 Appendix K Case history review, data analysis and
summaries

(a) Introduction

A wide range of publications has been reviewed. The compilation of the data is given in the
following Tables 19 to 23, together with comments on the reliability, methods of interpretation,
shortfalls and omissions.

(b) Division of data

The case histories have been classified on the basis of material types as follows :

(i) cohesive soils (Table 19)

(ii) granular soils (Table 20)
(iii) residual soils/silts/fills (Table 21)
(iv) mixed soils (Table 22)

When reported, the groundwater level, the relative density or shear strength of the soil has been
given. The settlement trough characteristics for each soil type have been plotted with respect to
the following parameters:

(i) transverse distance ty versus depth to tunnel axis (zo)

(ii) transverse distance zyversus discernible half trough width
(iii) maximum slope of surface settlement trough (0g) versus cover/diameter ratio (C/D)
(iv) maximum settlement (wmax) versus depth parameter (zo/r2)

The results for various parameters are compared from data tables, 19 to 22, in Figures 33
and 34.

(c) Collection and sources of data

As a result of an earlier broad based request for unpublished data, a further case record has
been added together with some in-house data to a data base collected from a wide range of
publications. Details of this data base have been tabulated and are presented in Tables 19 to 23.
In addition some data from deep tunnels in London Clay have been collated separately in
Table 24, and some values of sub-surface movements listed in Table 25. Appendix L lists the
references for the case history data.

(d) Reliability

The process of this review has highlighted some of the difficulties experienced in assembling
such an apparently authoritative summary. It is considered important to draw attention to some
of the common shortfalls in published case histories and to some of the necessary
interpretations which had to be made.

It is stressed that the summaries of published data made in this report are of necessity
interpretations of data which has probably already been pre-selected. It was not feasible to
obtain the original raw data upon which these published records were based. -

FR/CP/5 105
 Transverse distance to point of inflexion, i(m) Transverse distance to point of irflexion, 1(m)
0 10 20 0 10 20
LI i i i 1 i I i I I I i i
1'
x• ow- .9A
-7Ace......4 17 Key: • Cu > 75 kPa Key: • below water table
4A %\ 3A
713 0----
- %\., 0 5 X Cu 35-75 kPa 2A
....gel- i3
. 0 C„ < 35 kPa 2C 7:::• 68
3A O. X0 . \ 3C _
-E. 2'''' ze \ 4 0r A
%\X
y° 1° ° >
10‘›‘ 10 d 13.7
12,1
_ 16082 \ °r1
R 1211 0‘ 0 0
‘:213
ea _ 2°1'1 )Z1,7%
7.) 9A X
C _ 25.AlA 1 X
C 3219 XXI.' ' \ ow \
O - \ v. 2,3t, X \
o 120(0\* \
13% %. \ \ \
n20 — 20
— X
\ X,, \
v ie\ 9° X 18 X
% • ‘
DE

8 _ \ . X \
\ \ 1 X
\ \ 1 \
, ■\\\ 14 0 X \
\
1.2 S. \ 1
30 — % ■x 30 % \
i = 0.4z0 0.5z0 0.6z,, i = 0.34 0.5z0

(a) Trough width parameter, i, and tunnel depth, (b) Trough width parameter, i, and tunnel depth,
zo, for cohesive soils (Table 19) zo, for granular soils (Table 20)

Transverse distance to point of inflexion, i(m) Transverse distance to point of inflexion, 1(m)
0 10 20 0 10 20
4, 1 1 1 1 1 1 1 1 1 II
-X
Key: R Residual soil 11B
— F Fill
— S Silt 11A. 106
_ ;R1 7C CX 78
7A
o(m)

Depth to tunnel axis, zo(m)

10 — \ 10 91
Depth to tunnel axis, z

g
A1
R5C. 'S V3C 9 \
— . Ft 9° X\aic __
R5D loA \ 0 \um
B5B • RSA 05
- SI Rsc A
8A \

20 — X \
20

• 2 .
S. \,0\ 4A
\ S. /tit 04E
\ X‘

30 30 \ \X \
i = zo 1= zo 0.5z0 0.6z0
(c) Trough width parameter, i, and tunnel depth, (d) Trough width parameter, i. and tunnel depth,
z0, for residual soils, fill, silt (Table 21) z0, for mixed soils (Table 22)

Figure 33 Case histories - transverse distance versus depth for different soil types (a) cohesive
soils, (b) granular soils, (c) residual soils/silts/fills, and (d) mixed soils

106 FRCP/5
 Maximum settlement, (mm) Maximum slope, 0 (%)
10 20 40 30 50 60 70 80 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
I I I I I I
a mi Data trend of • I Data trends of
03.2) l 1
19 • volume loss = 2% ___- volume
8.214
(o. IC loss = 1%
_....16... = 2%
1- MI) .•„,-- 4 (11
= 4%
21 • •20 __,,„„...,-,..--",„ ia 12)) (0. 4) (2 ,T 1201 ,....------
iv) _ (0.3, _ (2.4) 238. efir •. , • sa _...----
2 - 22 • . (2,) .....- 13 .5 (0.6)10.31- ,•„.•• •79...... .
( 4 ..32 •
.....•••- 98
(0.6) .31t.,/
(2....
M (0.7)1B )0203).... r.o)
4A /4(1%1 (4.31 •9A
•
.4 2.25 .4.1
• gc
Depth parameter, 41,2 (m-')

3 -6
12.11 • 7 (4.31
f (26)%,1S,
•
98
(24) 4),4
• •-.._,.. (11 .7) 14
17.01 12. (0.6) •• • 12 12.61 17 Ompetly demol.shed) (1.0)
.3)
10. / cn (15.9)
9.0 =
Ni3 24
• Y.2f..31
4- /25 (11.71 (0.3)7 •
(04)

5
.1
214 1 ••(1.9)
.c
O. 10
31
4

i 25 (2.4) 1
10
(5.0)

a) / osi
0
gill) I
6 -I. i II
10 11.61'
2 1

7 '-
(141•I 1;
1
(5.01
•

Key: •2
1 Key:
43 15 (1.4)
.....(1 16) 1:123..3C
40))
• Case history number
8 41 • Case history number
1
I
(volume loss, %,
calculated from data)
(volume loss, %)
9-
23A
1 • (3.91
10 - 1 20

(a) Maximum settlement, wr„.., and depth (b) Maximum slope of settlement trough, 0, and
parameter, z° / r 2 (Table 19) depth : radius ratio, zo /r (Table 19)

Figure 34 Case histories - maximum settlement and maximum slope trends versus depth for
cohesive soils

(e) Shortfalls in published data

1. Some records omit details of the more serious or alarming movements.

2. The information presented is seldom fully documented, thereby reducing its value.

3. Usually the information relating to one particular location is not collated and often has to be
inferred.

4. The majority of records give information only relating to surface settlements.

5. Virtually none of the case records discuss or relate ground displacements to building or
service damage.

6. It is noticeable in some cases that considerable effort was expended to obtain a small
number of supposedly very accurate measurements of very small movements, whereas a
greater number of routine readings over a wider range of conditions would probably have
been to be of greater benefit.

7. Detailed observations on construction operations or workmanship are usually omitted.

8. It is felt that there may be a tendency not to publish cases that apparently do not fit
established or pre-conceived trends.

9 Many cases of real interest (those representing extreme conditions) are probably withheld
because they are or have been subject to litigation.

FRCP/5 107
 (f) Interpretation of published data

(i) Wherever possible settlements associated with tunnelling have been assessed as short term
(initial) or long term (final) based upon the following criteria:

initial settlement — that which had occurred within the first three weeks after the passage of
the face past the measurement location.

final settlement — that additional settlement occurring which had occurred in a period of at
least one hundred days or more after the passage of the tunnel face past
the measurement location.

Where known the actual number of days has been noted.

(ii) Attempting to determine the trough width parameter (iy) by estimating it from the shape of
the settlement trough is very prone to error. Consequently the procedure adopted has been
as follows. The parameter, iy, has been estimated wherever possible by scaling off values
of settlement (w) for various distances from the tunnel centre line (y). Using this
information, values of (y2) and the natural logarithim of settlement (In w) have been
calculated and plotted as shown in Figure 35. If the points follow a normal distribution
curve, they would plot as a straight line where, for a value of settlement w = 0.607wmax,
the value of y2 = i4 . The best fit line is drawn through the actual data points, the intercept
on the log w axis read to give wmax•the value of y2scaled off and hence the value for iy
is determined. In order to reduce the errors associated with small settlements towards the
limit of the trough, values of settlement of less than 5 mm were generally ignored when
plotting in this manner.

(iii) Wherever possible an estimation of the trough-half width has been made directly from the
case history data, but this measurement is very prone to error because of the small
magnitude of displacement at the edges of the trough.

4-0

c
X 4. 1

= /0

I •

roo zoo 300

rp cued MCI f VeITG CII 04"1"4-Mc e i Si te4.

whim, Yit I fr/

(4.4,) (49 (4,9

/00 6 0 0 4-4
0.6 S '.5 412 436
11 r■ /0 leo 411
413 /.35 /3.5 /12.2 31)
20 if° If 0 324 .T. 0D
Figure 35 Relationship between log settlement and transverse distance for normal distribution

108 F R/C13/5
 (g) Data of Table 24

The data presented in Table 24 has been gathered from an earlier summary table held by Mott
MacDonald. It has not been possible to identify all the source data for this table and therefore
caution is required in the interpretation placed upon this data. In order not to bias published
data from tunnels in cohesive soils, the data from Table 24 has been plotted separately, such as
in Figure 13

(h) Errors and omissions

In view of the above comments, although every effort has been made to faithfully report data, it
is inevitable that some inaccuracies or misrepresentations have occurred, or that those made
during previous collections of case history data (by others) have been perpetuated.

FR/CP/5 109
 6 V Y

ga .
r ...1
2 i
gie
N . 01 i

4 t
N
4
04
INITIAL READINGSTAKENAl

. li

G ASYMETRICTROUGAIwirmt
saw:pc OE sovroi.or ramr.
If 0 'a ! 1111 07
11 i
'
V1
: + §
ii
0

a g !I 1 .
RE NARR S

t
t IA
7DAYS

.01 • ;
•
•
'Ei ii'
•• 1 •
1• 1• i• i
• °< th •
it
55 1i i XI ga 1 4
gf 4
4- §§ 21V1 52 5
42 1
I g I' ; T, N- : A i,
o 2 AAZ4 A
A R A
g; 0
n 1 :i,
.lo -
■ 4 .: g 0... « t, .1; .i $O m"7. .P
-6-kn ;
A Ina
A. A
F
*
.A^4A,4*A
sp 4 4 4°0N r
A
pa ga
A iN lo) A
.
O.,
12 1
E
^ t. !, *
A
• TAA,'
,0 9
qrn
, t0 *
4 4* r:
A ta
a gT S
• t:. -
lit

.11C A2E2 •7 :•: ; N '•

'4 Z : 2 :9 ! ler
6 Z ; Z A g : .1", : !9 ,
Cs t .. i., -a • • i% % 2 22:
11* 2 7 7 1,z. N« N
a T
vi T 71% g I' 0 : : .r
=. : : 1;
1r 5
5 i
A A ■ q
'
o
g 1 F. ! E I+ : ■ . E 0
•

-4 0
• .
tg ,0 .N 6we, ..0..
A .4 A .
rl
0 v
N
N
0 = 09
0 9I. . P'
• T 2
Uf bx
311! " ';' " «« :.' A g m A
h„ • t .4. •4- ; rr 4.a o.
9
: g :g
¢SP . '''
u- • I
o
A 4 2aM gZ -,' g 4 8 8 4, . a
t ■
A
X . a
ts1
111! ; ;A'A g
:t.iln ;5.1t g d , .: 0 0 2 2 ; t
. •
42:gi . 7. : : 2 : : 3 3 :% • 1 . ; ;
121:
flill 1, r.; z a 0 t A a n = 92 %x .pri x8 Ar.-AAPA ,% SI
f case history factor: cohesive soils

l
+: 2 : 1 :: .1
R ; ; 2 : :; : :aA go
PA
igP ; ; ; Z S
, g g 2 ; 2 g ; •
0
A• r°
'iiml
§04 : ! : :
--- A
'
,A
O : : : 2 :g :2 2 : 3 2. 2 : g ^ ; '„;
2 , .00 r; xPsg .91“ P ,,,, ;
—cy:
Ile .*■2 Q .; 2 z z : ri 2 0.40,4 6.41F ., , , -.0000 .;,
A

,4 i4
:2 ° '
Table 19 Summary o

.. I 1 , i A2gt 2g eti
4 ng g a,
0 : t life IR 1 1 11 i I"
g ig .. e
2 1 4
a
xw X 11 hi 1 R 11 / 1 i i ' I; ii$ it
1 Pg L0 e I,g ih 21 tg
i
.•

g nit ie5
STIFFEONESNEGRO0ND

04RCONSCUDATED•

I
14 1! III IT
GROUNDCONDI TIONS

SIVF. FissuRED. RENALY

STIFFCONESIvE SOIL
WY CORFU*SOIL

34
CLAYVv 170Whe

84 111, 51ii DI
5] Oi il Si li lj •
601.11/71•

Ylg 1 180 joi , i!

is
X2 ij ail lug 17 vi Ilix .g 5W PP 117
51; Oi il bi 44
W L; 11
it k,2 C? i J 1/ kid i § lifd 1.A5
. wo , d n g A .0%1 0 46- 2 t:n 0
84 s2 '4 '1 1 " ' '

Q d A e,.
2,4 $
iI
g lin °11
il' 't11.14wV 11 2
1 41 1A1111 11
.11 11 b 1 1111 411
la 1 U ii 1II il /1 i ig 1g g
 f c
Table 19 Summary o ase history factor: cohesive soils
TOOLE 2. OF2.

GS

6f
ME N430OF

A
"i6
h

a
'2

8.
GROUND CONDITIONS

to

1"X

§5
..7;

.•

'6
'g
3
2
EXCAVATION

1.9 '

.L..
265

8C/
g1
I- N
...

.
40

,,_
...Z
,....

...0a;
3Zi
lr
1*

i "
igix

$N1
4";

bx_
2,.

gH:
$1 !
obl..
i$I

lq.
.7
26.1e
.. .!

t; .
i
igir

ii
i

R ;-
=1
I

.:.
4

a
4411•
n4

a
t
4
2
2
FIRM COHESIVESOIL NANOMINED

•
n
0
2
. aDAY SE TTL EMENT •ETER

4,
.,
.1- .?
I

i
1

p
•A Anima SHIELD 1ST SHIELD PASSE D
.71

1
1

2'
4
Q
ao

el

a
n
•

..
MEDIUM CLAY a

7::
PECV.

c,
4,

T • 2
s2
CHIC

,
. .

2
CHICAGO L GOAAGO CLAY .
C.. • 10-12 Anb.•

1
,..

2
....
V
4

•
:.1.

n
-
a

1.

e
2
T

1e
ii M
z
1

i
.
.
S
9
2z4.12i Igo..,14."-...i
..
.-.
ill° !

.
s

aA
GLACIAL

4
.

n
.
:-:,
,
%
s.4

1
FIu ISTIETTILLMTVCLAY PARTIAL. MACHINE"FACE

'.
C RMZOO. ANA.

y
STIFF FISSURED

a
I

4
99

N
I

aT

:.

\I
•,.

§z ;"
LONDON

■
1
g
a

1 .
:
Cu • ISO CLAY
AN/...11

.
4:- -

0
FIRM 771 STI F F

1 2
4

t
4o

z.
19G
T

1
WEATHERED LONDON

pi4,

=tia

,.. 1
2

0 1

. ;
.,

,
CLAY tii
.

; !
Cu 90 AN/A,I.

(..
FIRM TO STIFF

..464.
4
*
t...
c
LONDON

rT
Z

I
WEATHERED

.1',
l'•

wt-
I,

,...
.1
q
0

,
CERT
Cu 90 ANAn'

2
g
'r—Z
STI FISSURED CLAY

N
N
N
94

r
0t

,E1
Z

a
48 rl

..1
CuF- F200- 400AN/..

.
,

A
..,a

4
I

7.
40

t,
,.
01
.L.o
. 00
ao 2V

.
oi
2
GROUTIDEVERY3 RINGS

0d
A...
1

n A
1 2

"
.32 04%.Vs 0t
1

A
.

I
w
g
..
a
4
A

a.■
 11.131.7111CIENtDETAILS
OFGROUNDIVITER
..,
A
ON WI

LYMR
n%d .

IN CAS ERE LOAD

i g
F

,
TED
W L A T CEN TRE OF TU NNEL

*, i

GROUNDWATER
„.1

GROUND WATER1.2m B GL

cfEFNT
A i

!FIRST OF TWOTUNNELS
11'

GROUND WATERGL
1 1601
A
•

R* AFTERABOUF12 DAYS

, r, rt
Pi

IN C&
yva K nvicivCROWN

2,
nt/ 9 1
ir F

2 Flits li -
:..,
.2
REMA RK S

2 Ett w x.,
Eltio;•Irtn

IN4" 1-----
;

MAIDSTONE
..1.."' A ci c EDI
K-1.70, ..S 2.-i5 i
/11.1 IIERe Rata A 9.4 Itc 0
N
it

.c
_s ee 4 ,.se..,97,
e _s_s ---, %
a.
ili IgIA
,---.
.---,
i- 0 L
t• au
it\ 1 2 kRZ 'Z A rt 1 ...
E 31 ii33 i 1 4 1 E
.t ..„
. ... .,c,‘ .,
. V
1.-z7:, iza tz2 ,,, -- : P-2 2: .,,
..qo °1 L. 0 2 . .r . . . 4 . w. 4 . ...., z ....2 ; 17 ;2
,..
._ i
:!F ; : „ %.1: '2 '2 74! ',,' g 2 c?, t; :Z : : ; 1c.), 2 1
'- 4 '4 ^ 12 ::..
=
c. c4 .4,9..4
410 =z; v: 1' ..P. 7 :::,, 9 :,?, 4, 3 :4' .'.: 7, ,-,-, N 1.2 z z T.4 y. 0 4 4. O. 0 0 0
ta

F+ a
a g 0
1 0 ?
.
dii 2 / eh o■
°
O9
44ii.
;
Mr! ;
fr
- i r.g..'t, •
I-1 .1 Z; ;:o r ti

I "!: '4"; C; X VO' :2Z e, Z a zz. .t, !:','_‘7,«Z.T.; `.; 2 '2 7.

9
6r _ ‘.• v. t.. 0 -
9 4 A 4 I co 4
i 1i =
, a IS f cfc 9 . .:*
VI 'I. 0 co
ea N
• ,3N i g 0 ct
. 0% - T""?'r
. *yr* *
N
r • • t 2
* LL
..i56
10:
c-0
..: M g A :7,.' g A ,e.
t of 0 o 0 o 9 -- `41
t.F.1 •:, ;',"; :'i.k.4' •
N
- 0
tA 4 t., V., ti . , t A et I ' 1 il a .iltr .
... th
11._ TI wl

e ta N 6? N9
iTai ,?, :', 4 1' ::, 1. 2. : 73 ; : :',4 2 2 ..' 2 Z : : .
ige5--
; t, • w .
Igi; t r, ,!...twa . iso, u, . t 1.. = :., To !V 44 2 rrearl?..2 t 2 4 9
2 ;1 1 •
1
hp — -2 . .
Li!
toe..
O T77o,ctoT 0 I,
2===.4 4 4.
1.
c'•
's. n.
-
9 "
co
0
1 is
Z
t.,
cli
1:1
0
co
,
to 44, 991 '3.1 9
.4. 4 0 0 '
0 2... -OS 9 9 ,
o ,' Vi: re.3 th".;
' .,'
2! - z"
=2,
01.4 .;•11,
__: !4!..
4_tr i.., ..
. .1,. N- :,.?,
cN N ,- rt.VI 9
KI
;i;N 99 M1
1 4
ti
9
u
9 N q. 99
utolocato :...-
O' i 9
0 9
- 2 0 r, : a " ' '
000 P-CV1
B,,[fc _
ilia - ,,, ,',' ••mmot;
V ,..; <4
Ti ; Zt4, 2cO
i4 el en 0 Co
. 110
co
4
9 41 4 0
vl ,..cl 29
u .I
.;, 0
4
9
ta
90g
el . vi . . e.
0 9 9 °
P.
.210
(7211

4 4
TUNN(1)1

9
IrirreaMs'ffit
1MOLE1

.."2 3
N ATICSL7
:4t2=1

LLYSHIELDED

9
lz
a.

VERYDINSE.SIAPITLY wECTUSAEs

4'—.v.§ > ot0 9 Jr 0

., I§ 0 '1
MEIt+30 OF

, 41
t s
EXCAVATION

*A i 1 1 4 i Iii 21 8E Pi
'' 2 2 d 1 Sc 41e
„
A PM g a ii it k. FA $.i
LOOSE SILTY SAVD (243)I
S P/ ('N'J

[
ti,
13040)

STIOSTOMMIS 9STJ

SAND 0 US)
COURSE SANu

rt-...4
. iti e. i7. ... k
?NKr° Wff7ER
re

lz4

il,,4
GROJNOCONDITIONS

GR AV ELSILTY SAND

DENSESANORGRAVEL

1 ze4 „
iE
• SONOS.CRAv
E GRAVEL

- :, A :
1511Als1216AD

IA1 I =k- -i
CEMENTEDSAND

s.9.6-03 i 2 1 el:4,
S
f

e5 *' ., `214
SILT YCLAYS
Jo,

Flauf

..*z,).=
SAN DS SH E

(.7
1t" 4 .
c
SOME

ce""Mg, ea :411
van( MSOM

iI51 A EI
CROWN
RARIR

.. i2 a kt 14...5 ,i,PIA
4
MEDIU

j `i .5.z..rf, ;
.

4 vi
4i.? Z or. nsa ..4 wa99 a.,,,, C a?
-1 Z el 0.%&?. re!

''''' . " a l Ax N ; :u 2 'a r, 1 z F, :,3 5.1 r_ ; l'j , ,1 :,',. ,%1 r9, r, r' z.
- N '"11
bl
7ff13 7/41191 ON
1

111.7 7006I001
e

WWI ST1SW1117

1 t 0 ,.
rrewnpreeit

.e llIrtdieNtraN

v._4 A3. .,
N019/07511WA
47.13 47113710
,gnomon=

2 it:
; 11 4 gt st i 11 , :2_
XR-1 •v'''
2171415

3 Pil gil ;3 2 ze
ig
ill
 PI
eR ,,
2
1
11

h2t f V w
4
$: • i
4 ■ 4
0
5 W, 55 V ed2 11 ,4 :
ltW 1%4 )Q 4
- ol 4 '•
., 1:1 4
cr,i 11;
h 11 '111
1 bgg
1 1 .1. ,:, 1b4
" r if .1 Qt.' 4
1 )10: W is,,1 li
5 %.
1t4
ihlin ill .. I*a l,
'.I. N41-
:Et1 "gil 41 i .%kl ,
L:: 1,4 Zt u .1. •4
,tz
:1,),
9 „ 41 , . 4 R
1 -:4" ••\'' ; !
7
■
Z *AZ * Al -kk ag +

-.14‘ ih4
J

4 4 ; 4• i ! 1; 4
V P
.40 4,o 2* t ! : IS : i
° '! gg (14 Z Z Z '' Z l
'
0. 10 !I * , i 11
I5 ./ '41 4, el. t ; M .40
2 1
't g
4. 7 fl T .01) „ . t 2
2 ' '
3 "(401 N 4 .4 4 .. .4 z 4 z I
S5

rrYt 4° l
0- :9i o 4.,I ; .44444 .4
c gte !0
I
-..t_
Rin
E. /
4 z
,
: ,- g oN,„ .,
.11!! m,
, ..
.,4 sa .4.♦ 6^ :
- .
11 : • •
CI 6 t ..
a ll
bi= z
,
.
..$qg
. i
bap T •• .,, • •
14 T 4: 4 4 t t ; a 4 '
1 o ., a t, 4
4 4 4 4 0 0 1. 1
§:St 1.-I -
,
i *
42
11
:', a4 x eAlo tttZtA A 4 4 •
v 12
O ”.:;
..'k.,
4 7 X Om ;
..,=
.
z. .., •
ih2 t t P q i e: 19t4tt8 41 ; .
,9 ::
eo ^ o
tV41S i " ." c.4 IN ? 'Y VI W 4 1 ' 4 0
l
N Lit
!IV i
l 1 ! si : : : :'
'•' •' ; • ""
.
q / 0 ' '1 4
EE 42
00 i tl i o 2 A 17 / ;
j ‘a X S $ 4 % V‘ 4 a
cri
Z '!4 ' 0!7. v Q1 i : 4 . 1, _
k '4'u 10i d i i a .1 1, ;:t 1 .4zi le
11 ;
2 A 4 4S .1: , a
„i
t 1 1?
v 1 4 I q
`i
g 0 42 PI
04 4 4 y v l 4 . ;
1,4
! 2 t 1! q0 11w a 't 4 n qi ill :A. 34
a 4 ci;,-..$1q
ne0 lo - Y
t,4, ;l
!
zt4414; li. •
.4 .
+
4A ),
:4
k3.4.V
3 ;i W. 3 id s.
4J
An
%
4 .1 3 i 4 4i 3Z4
3 i 4!
4
Ili
1 i 3 1 ta4 ILI d1 5a,, 710
..ii W :7' ': Z :1111111 1 N
4 "-ta
,

,.. ."4- :”
is —
ZFI L 1/0 4A t
'7 !
,!..
q
b J6 6 111
-all" A i'''Ll ::
., It. 47%" /11 1)1
)
I:
I gy

77,4
 DAMAGE1011111/C11111ES

.:. ifq .
0v '-
12,41.60AS

MI 4i I Oge if I !• I
sutfAra ,U111/0144, 64,14n,

ISA.,iblykKfa,

111111 iiti il i" tit li ! I,

1.

I, .131
It'll iill;
4 ail ellill 1 i ' • 'n
iza 11 ,14
1.1r10••••44. •

14
. -1 ' lt t •
O 1, 4

lii _ii 1111 •

U., MM.*

, 4 11

fiiii iii
iii

mcrur M Ac3>ogNI-D
1
7..
lae! 11111
foq
5,. . 7 f.-• ..-%:- . 1 • -'' ,:2i P.'
‹ a °
a 1 t_A g a.
-_,a .A_ ..-:- •
il .1, N $M,
Cf• N -:. 1"" 9.. 1""
. ef N r - -A.4
-1. i ,.. 4 I 7, 't;, N .4 N- ev cm r4 m ii A t -7, *f ,`;,*
4,, .,.,. „, ,..
.1., ,,,, ,. so ,„, ,,, , er
4.• *# ',' T .4 t
N N N . ,i, .. ,t, *7 , I.. 11. l
« Z I; 1 t ,t
G ,,
M• - ..„ ..., N 0 N " 1 .4. si ..
! ;".. 4 Z si r4. s.4 ," t.,.', ,.:, 4 ••
, .4
u, ; ; S i.%
N
1 '' --
.- r 0• 4:-
4110 N ::. a ..? .4. m -, • •4
It, e, .:4 - 1 m$
..i. ry NN N el .:4 r. i „. CN Vi.
Cg 4 N
to •0
1*
CR
.4 4 .6 24

in ° :
.
'
o o a 4•
2
iir
I !
$

g i
O i
-- N .. C " ‘4, 4. g., r.. to r,v. SI V
'es ,
il '' .. N

.gels - ,v,; ;!'..

A ...-O LL
dig
1 999 9
Az 0
?,_ .- t
9 0
N
9
9 .
41
:-.. r. e.- s..... i .•*9 v.- N

ig i
2 •
■

1 1I
Ilt
en 0
±:
.i.
OS ••• .4
.:,
■ 4 N .1
M. "" 'r
r• a % fp
Lr 4. M
C:4 cr.
S■
1 li.
US is
AA
4 .4 n
/gill
INI r-• "' fi 17.1 Li 2 ,4 " I.N r■t 2 N MS a *
g .1 d
8
gb11.
161! ‘r ; 2 vi « : 1'/41 `,1; C, '2 Zi 1 : 1 1, E 9 9 n 1.1
2 4 ‘,A kit
V8 9
N !' 2 o 9,
.- 1 .41 9 .f. ■ 0.,.14 v. r
0
0
ifqi z 4 7. 1 Z1 : 4 4 4 t' 0 4- 4- 1-•
2v
311 ■-• 4 :,- ;.t. 4 ,, i;N .:1t 4.4 sr.,u, so 4.." M tF, r, 41 4
J.. ,..,
4 44 .4.
i aa r'. 4 4 . 4 a 4 s. .4 .4 N N N .4 .44

A•

EC, 2 I21 1 1
1 I 11 i'l
?71 9 ; ; 1 ;
a 11 it li. Wig! I 111
U)
N il X i i g I
2
.0 • , 4 1 ats Qt. q .1 1 a 31 t t
CUM, 5..11/SP1. 1-15

-•
ti. 35 gla p4 l'iTo. ..fl i7 i
cut

■
§
.,

AIII
' slow

11
§; I. rIip 4. 11 !
...
$-
11111:11. Of maul n. .

si
MAW tkm mcpun to

t-
if
410411WA$101

$
.- 1001J1W

5;! 11 likit-il! 1
:1

;I J tii 0.3 :01 7,1 1'3 P.,

i=., i1 ir
1111.1: 10%50

1 i Avl * 111 11 !I ItNi 11 : 4

11

li Willa III i;
tAl Vili Ill '011 tfitil:t Mt 11
C,
Cl

.(1 ...) a 'M 4' '4 .../

il.. - In * ..., .,
..,-) .< ... ... '2. •-•
_ - N N 1 N ri... , o- r r
',--.
ku -- .
0 i '''' 3,, e,1 g. $ ._
A 5;= lel ig ill 11 ill
i 11 I 1
 i • •
g
!
1
ill. II igst I
g 21 11 gip
1 L;
ow As n 1 E A,- aA •
I 2 1 k it °I
Ta ble 23 Summary of case history factor : twin tunnels incohesive soils

80
P g a
VI p E I: illib-
411
1§111
PP
g /g
• 211
1.7.
;HE i l ok higi
=ALLi-' 2i _.; 2
„ _
N. r ... ; ..
use ,.... ....- ....- .4- 4.ml 94 40 ".4....• " 0., t;
w r -.
■
1 r.■ / : x 2 74. e ''' V r P... 1 IV .S.

0- i 5 t
'II t t t"
•
, 10 .7, 7,-, 7:,7. n 7 7 : 1 : 7, 2 2:::;:«Zii;:«7, 1 :42•Z2 ;
ge
II=
a.,;8?BWE2
wil ““01 " 2 2 2 2 .4 4: :2 :2 : ::61 1 2 A 2
!nil
o 1
SUMMARYOF CASE HISTORY DATA

II
s=_
do-
ssii t• a
. 2
an. i I N

O
t/
.' 2 a Li Li = 2 '"" .= it E I a : m R$ rz * ,. - P. r.
. ..
. 4
ill! 5' 1 -
oba ,
IS
IgliP 5:113 A 2
1WaiII 9 9 0 9 9 0 9 0 9 9 9 9 9 9 9 9 .4 2r,.4 4 9 9 v 4
2 249 2 2 2 229 2 2 4mC.20halaioft22enag
9
N aa 9 000
4
! :1
0-
0 4 4 4 9 9 9 V 9 4 9 43
143334ZCZCZ2± 9 9.
t o. - 4 A m -..=
- ioallo P. 41 4 = r4
4, N f4o
.
"
i.7 r; ..0... r..' '.4: '
i'c
- «"« m«T;
m
.77.44. . NTs• 4. . . . —r2 4: 1 t
... ..Anc,,t52 .. 22m2451314 iv' 0
igli " " il "
2t 8KNNNK
8 8 448K 0p. ZeSsi.%°;22&10&&
ltii 8 4 88S 24 7.: .... . g2 ,.. ?4
pp. 4 4

5
1 1 i 1 1 i i llf i 110.22
I
1
1
' ME UCCI OF
EXCAVATION

11 1 1 1111 11 1111 1 2 2 2.4

iI 1 25
2 t v 11 illiiiiiii I a isi
i4:,22 I4 3 'a 2 a at a 2 a r.. 82z 22822 -ZWgiz.ligV w 4gb
88 'II st : SV 04 1 11)0 11 10413
I gi gi 1$1i EY .; aigigig:1011t 11! ly 101 4141
u gR

1 tiki i
' 11111
1 3 1 a 1i R atilp s15/1
a k -- I i ij 5 lilaill I Wigigi Illth
h/
_ N rn .4. 41 . r- . • el 2. 2. 14 S • r. 2 t
. . Q
F. .ii .,.
• iiWA
ll
 %n in un %n

1.5
un
.2 E 0 d en en c3 to In 10 %n r- *0 N N •- ■ •-• en -er In
a .

N N N N ch N O. O. N N cz, 41 0 0 0 0 en ■
11 %n 00 0
CI
6. .c
■ ..t3 r.: s r: ei rz N r- t-: t4 vi N N r.: vi vi v-t v3 vi v3

22.3
In n
7.0

1.5

24.0
15.5

12.3

15.7

34.0

28.2
r": r": o:; CT 0. 00 00 .-I 4n
settlement

00 o en o O.
One year

— vl In
N cn -
C
— N 50
A
(mm)
final
Settlement

vl
3.8

14.0
3.5

00
0.8

7.5

../4 en ..0 un 00
6.8

.--. N .--.
14.7
9.3

o
O In 0
- 13.3 ,-.
-. v:i N vf, 4 0 ...
.... 0:
V..
vet
Initial
max
(mm)

o
7.4

7.4

'0 o en
8. 1

7.2

7.2
Z.L

5.3

5.8
0 0% N
"s' '-' -- 't N
r.: t-: ad oi cc; N N r- v3 vi vi sd sd
et.
- 4'

t
i ....... a: a: as a: O. a: 0: 0: 0: 0' a: a: O. a: a: a: a: O. a: ci: a:
N N oN N en N CV IV .... CT CO ,,C4 00 00 00 0 •-• N 4, n -4 •
C.) C ) E N N N enen enen en 01NNNNNNNNNNNN

ctS
C3 3
0
C
0 I.. . . Fri n
a1i g g A ..ge
L 35

1.1 ,
%.-4 le..1
31

'4' C.1 •-, •-• en

ON
26
1£

C,1 Crl C,1 N M N "

0
.J
0
trength

C
Shear s

C
Id's)
cu (

390

185
200

320
240

270

240
081

170

8 h 8
C V
C
3
a_
Fissured, sand partings, seepage

C
C u
Silty, seepage above tunnel

q
0
conditions at axis level

Organic traces, seepage

round

Sand partings, fissured

Sand partings, fissured

Sand partings, seepage

C
Silty, somefissures

Silty, sand partings

ng

Fissured, seepage

to
Comments o

co
Silty fissured
Silty fissured

E
Fissured

2
Sandy

em
.5

E
C
I

C .. -.
2.8

1.8

,C) t•-• .er ce, o .er vl oe

! .-; ...; -; ..e. .4 ei e..4 e.-4 .4 ri ei
Ti•
csi
C
i
U ..t co 0 A iii it. C) W .... .., SG .4 Z 0 a.. 0' c4 w I-

116 FR/CP/5
 add

's 3 Ts' N N •d• vt 00

d 0 ci 0 d d ci ci
CO VI 00 0

E
el VI
o 0 0 w.■ O
I 0 O.
tV Cl V";
WI 0 VI
C ei er; ni c:i
0

3 t4)
O
E
a 4)
4 Ls_ VI •-• N N

e
CA ta
g
§ 0
•

as >.
es =
U U 17) 7.1
a.
Observeddata on sub-surface horizontal movement at tunnel axis level

v-) %.1
oo
et:

Is en
Eisenstein et al. (1981) Canada

Glossop et al. (1979) Belfast

0,

g s
.9ws
1 o3
;
2
"1*
I
a, U A ul 1
4
;
as
Table 25

. •

CO
* s

FRCP/5 117
 Appendix L References for case history data

Tunnels in cohesive soils

Case Reference
history
No.

1. HANYA, T. (1977) Ground movements due to construction of shield-driven

tunnel. Proc. 9th Int. Conf. Soil Mech. and Found. Engng. Tokyo, Case History
Vol., pp. 759 — 90.

2. BARRATT, D.A. and TYLER R.G. (1976) Measurements of ground movement

and lining behaviour on the London Underground at Regents Park. Transport
and Road Research Laboratory. Laboratory Report 684.

3. ATTEWELL, P.B. and FARMER, I.W. (1974) Ground deformations resulting

from shield tunnelling in London Clay. Canadian Geotech. J., Vol. 11, No. 3,
pp. 380 — 95.

4. GLOSSOP, N.H., SAVILLE, D.R., MOORE, J.S., BENSON, A.P. and

FARMER, I.W. (1979) Geotechnical aspects of shallow tunnel construction in
Belfast estuarine deposits. Proc. Tunnelling 79, London, pp. 45 — 56.

5. TOOMBS, A.F. (1980) Settlement caused by tunnelling beneath a motorway

embankment. Transport and Road Research Laboratory. Supplementary
Report, 547.

6. WEST, G., HEATH, W. and McCAUL, C. (1981) Measurements of the effects

of tunnelling at York Way, London. Ground Engineering, Vol. 14, No. 5,
pp. 45 — 53.

7. ATTEWELL, P B , GLOSSOP, N.H. and FARMER, I.W. (1978) Ground

deformations caused by tunnelling in a silty alluvial clay. Ground Engineering,
Vol. 11, No. 8, pp. 32 — 41.

8. MUIR WOOD, A.M. and GIBBS, F.R. (1971) Design and construction of the
cargo tunnel at Heathrow Airport, London. Proc. Instn. Civ. Engrs, January, Vol.
48, Paper 7357, pp. 11 — 34.

9. GLOSSOP, N.H. and O'REILLY, M.P. (1982) Settlement caused by tunnelling

through soft marine silty clay. Tunnels and Tunnelling, Vol. 14, No. 9,
pp. 13 — 16.

10. PALMER, J.H.L. and BELSHAW, DJ. (1980) Deformations and pore pressures
in the vicinity of precast, segmented, concrete-lined tunnels in clay. Canadian
Geotech. J., Vol. 17, No. 2, pp. 174 — 84.

11. EDEN, WJ. and BOZOZUK, M. (1968) Earth pressures on Ottawa outfall sewer
tunnel. Canadian Geotech. J., Vol. 6, No. 1, pp. 17 — 32.

12. HENRY, K. (1974) Grangemouth tunnel sewer. Tunnels and Tunnelling, Vol. 6,
No. I, pp. 25 — 29.

13. KUESEL, T.R. (1972) Soft ground tunnels for the BART project. Proc. N. Am.
Rapid. Excavn. and Tunn. Conf. Chicago, Vol. I, pp. 287 — 313.

118 FR/CP/5
 14. MORETTI:), 0. (1969) Deep excavations and tunnelling in soft ground. Proc.
7th Int. Conf. Soil Mech. and Found. Engng. Mexico City, Vol. 3 discussion
pp. 311 — 75.

15. RODRIGUEZ, L.B., and RUELAS, S.A. (1981) Ground settlement from the
excavation of tunnels in soft clay. Proc. 6th Pan Am. Conf. Soil Mech. and
Found. Engng. Lima, Speciality Session. Published in Soft-Ground Tunnelling.
Edited by D. Resendiz and M.P. Romo. A.A. Balkema, Rotterdam, pp. 75 — 8.

16. CLOUGH, G.W., SWEENEY, B.P. and FINNO, RJ. (1983) Measured soil
response to EPB shield tunnelling. Proc. Am. Soc. Civ. Engrs J. Geotech. Engng
Div., Vol. 109, No. 2, pp. 131 — 49.

17. ANON, Gateshead. Private communication.

18. SAENZ, J.T. and VIETTEZ, L. (1971) Settlement around shield driven tunnels.
Proc. 4th Pan Am. Conf Soil Mech. and Found. Engng. San Juan, Vol. 2,
pp. 225 — 41.

19. HANYA, T. (1977) Ground movements due to construction of shield-driven

tunnel. Proc. 9th Int. Conf. Soil Mech. and Found. Engng. Tokyo, Case History
Vol. pp. 759 — 90.

20,21. PECK, R.B. (1969) Deep excavations and tunnelling in soft ground. Proc. 7th
Int. Conf. Soil Mech. and Found. Engng, Mexico City, Vol. 3, pp. 225 — 90.

22-24. O'REILLY, M.P. and NEW, B.M. (1983) Settlements above tunnels in the
United Kingdom, their magnitude and prediction. Proc. Tunnelling '82, Brighton,
pp. 173 — 81. Report of discussion, Trans. Inst. Mining & Metallurgy, Vol. 92,
Section A, pp. A35 — A48.

25. ATTEWELL, P.B. (1978) Ground movements caused by tunnelling in soil. Proc.
Conf. Large Ground Movements and Structure, Cardiff 1977. Published as
Large Ground Movements and Structures. Edited by J.D. Geddes. Pentech Press,
Plymouth pp. 812 — 948,

Tunnels with mixed face conditions

1. GLOSSOP, N.H. and O'REILLY, M.P. (1982) Settlement caused by tunnelling

through soft marine silty clay. Tunnels and Tunnelling, Vol. 14, No. 9,
pp. 13 — 16.

2. SOZIO, L.E. (1978) Settlements in a Sao Paulo shield tunnel. Tunnels and
Tunnelling, Vol. 10, No. 7, pp. 53 — 5.

3. ROWE, R.K., LO, K.Y. and KACK, G.J. (1983) A method of estimating surface
settlement above tunnels constructed in soft ground. Canadian Geotech. J.,
Vol. 20, No. 1, pp. 11 — 22.

4. SCHM1TTER, JJ. and RENDON, R. (1981) Tunnelling under compressed air in

Mexico City. Proc. 6th PanAm. Conf. Soil Mech. and Found. Engng. Lima,
Published in Soft-Ground Tunnelling. Edited by D. Resendiz and M.P.Romo.
A.A. Balkema, Rotterdam, pp. 45 — 55.

5. SAUER, G. and LAMA, R.D. (1973) An application of New Austrian

Tunnelling Methods in difficult builtover areas in Frankfurt,Main Metro. Symp.
Rock Mech. and Tunnelling Problems. Kurukshetra, India, pp. 72 — 92.

FRCP/5 119
 6. YOSHIKOSHI, W., WATANABE, 0. and TAKAGI, N. (1978) Prediction of
ground settlements associated with shield tunnelling. J. of Japanese Soc. Soil
Mech. and Found. Engng. Vol. 18, No. 4, pp. 47 — 59.

7. O'REILLY, M.P., RYLEY, M.D., BARRATT, D.A. and JOHNSON, P.E. (1981)
Comparison of settlements resulting from three methods of tunnelling in loose
cohesionless soil. Proc. 2nd Int. Conf. Ground Movements and Structures,
Cardiff. Published as Ground Movements and Structures. Edited by J.D. Geddes.
Pentech Press, Plymouth, pp. 359 — 76.

8. NORGROVE, W.B., COOPER, I. and ATTEWELL, P B (1979) Site

investigation procedures adopted for the Northumbrian Water Authority's
Tyneside Sewerage Scheme, with special reference to settlement prediction when
tunnelling through urban areas. Proc. Tunnelling '79, London, pp. 79 — 104.

9. MORTON, J.D. and DODDS, R.B. (1979) Ground subsidence associated with
machine tunnelling in fluviodeltic sediments. Tunnels and Tunnelling, (Part 1)
Vol. 11, No. , pp. 13 — 17 (Part 2) Vol. 11, No. 9, pp. 23 — 8.

10. CLOUGH, G.W., SWEENEY, B.P. and FINNO, RI (1983) Measured soil
response to E.P.B. shield tunnelling. Proc. Am. Soc. Civ. Engrs J. Geotech.
Engng, Div., Vol. 109, No. 2, pp. 131 — 49.

11. NORGROVE, W.B., COOPER, I. and ATTEWELL, P.B. (1979) Site

investigation procedures adopted for the Northumbrian Water Authority's
Tyneside Sewerage Scheme, with special reference to settlement prediction when
tunnelling through urban areas. Proc. Tunnelling '79, London, pp. 79 — 104.

Twin tunnels

1 to 16. HANYA, T. (1977) Ground movements due to construction of shield-driven

tunnels. Proc. 9th Int. Conf. Soil Mech. and Found. Engng. Tokyo. Cage History
Vol. pp. 759 — 90.

17. MacPHERSON, H.H. (1978) Settlements around tunnels in soil:three case

histories. Final report by University of Illinois to Department of Transportation,
Washington, DC Report No. UMTA-IL-06-0043-78-1.

Tunnels in granular soils

1. BODEN, J.B. and McCAUL, C. (1974) Measurement of ground movements

during a bentonite tunnelling experiment. Transport and Road Research
Laboratory, Laboratory Report 653.

2. BUTLER, R.A. and HAMPTON, D. (1975) Subsidence over soft ground tunnel.
Proc. Am. Soc. Civ. Engrs. — J. Geotech. Engng. Div. Vol. 100, No. Gil,
pp. 35 — 49.

3,4. O'REILLY, M.P., RYLEY, M.D., BARRATT, D.A. and JOHNSON, P.E. (1981)
Comparison of settlements resulting from three methods of tunnelling in loose
cohesionless soil. Proc. 2nd Int. Conf. Ground Movements and Structures,
Cardiff. Published as Ground Movements and Structures, Edited by J.D. Geddes.
Pentech Press, Plymouth, pp. 359 — 76.

5. CORDING, EL and HANSMIRE, W.H. (1975) Displacement around soft

ground tunnels. Proc. 5th Pan Am. Conf. Soil Mech. and Found. Engng. Buenos
Aires, Vol. 4, pp. 571 — 633.

120 FRCP/5
 6. EADIE, H.S. (1977) Settlements observed above a tunnel in sand. Tunnels and
Tunnelling, Vol. 9, No. 5, pp. 93 — 4.

7. YOSHIKOSHI, W., WATANABE, 0. and TAKAGI, N. (1978) Prediction of

ground settlements associated with shield tunnelling. J. of Japanese Soc. Soil
Mech. and Found. Engng, Vol. 18, No. 4, pp. 47 — 59.

8. O'REILLY, M.P., RYLEY, M.D., BARRATT, D.A. and JOHNSON, P.E. (1981)
Comparison of settlements resulting from three methods of tunnelling in loose
cohesionless soil. Proc. 2nd Int. Conf. Ground Movements and Structures,
Cardiff. Published as Ground Movements and Structures, Edited by J.D. Geddes.
Pentech Press, Plymouth, pp. 359 — 76.

9,10. NORGROVE, W.B., COOPER, I. and ATTEWELL, P.B. (1979) Site

Investigation procedures adopted for the Northumbrian Water Authority's
Tyneside Sewerage Scheme, with special reference to settlement prediction when
tunnelling through urban areas. Proc. Tunnelling '79, London, pp. 79 — 104.

11. PECK, R.B. (1969) Deep excavations and tunnelling in soft ground. Proc. 7th
Int. Conf. Soil Mech. and Found. Engng. Mexico City, Vol. 3, pp. 225 — 90.

12,13. MacPHERSON, H.H. (1978) Settlements around tunnels in soil : three case
histories. Final report by University of Illinois to Department. of Transportation,
Washington DC, Report No. UMTA-1L-06-0043-78-1.

14. HANYA, T. (1977) Ground movements due to construction of shield-driven

tunnel. Proc. 9th Int. Conf. Soil Mech. and Found. Engng. Tokyo. Case History
Vol. pp. 759 — 90.

Tunnels In residual soils/silts/fills

R.1 SOWERS, G.F. (1981) Lost-ground subsidences in two shallow tunnels. Proc.
6th Pan Am. Conf. Soil Mech. and Found. Engng, Lima 1979. Speciality
Session. Published in Soft-Ground Tunnelling, Edited by D. Resendiz and M.R.
Romo. A.A. Balkema, Rotterdam, pp. 75 — 8.

R.2A THOMSON, S. and EL-NAHHAS, F. (1980) Field measurements in two tunnels

in Edmonton, Alberta. Canadian Geotech. J., Vol. 17, No. 1, pp. 20 — 33.

R.2B EISENSTEIN, Z. and THOMSON, S. (1978) Geotechnical performance of a

tunnel in till. Canadian Geotech. J., Vol. 15, No. 2, pp. 332 — 45.

R.3 ATKINS, K.P. and ABRAMSON, L.W. (1983) Tunnelling in residual soil and
rock. Proc. Rapid. Excavn. and Tunn. Conf. Chicago, Vol. 1, pp. 3 — 24.

R.4 BRAHMA, C.S. and KU, C.C. (1982) Ground response to tunnelling in residual
soil. Proc. Speciality Conf. Engng and Constrt. in Tropical and Residual Soils.
Am. Soc. Civ. Engrs. Geotech. Engng. Div. Honolulu, pp. 578 — 87.

S.5 ENDO, K. and MIYOSHI, M. (1978) Closed-type shield tunnelling through soft
silt layer and consequent ground behaviour. Proc. Int. Tunnel Symposium '78,
Tokyo, B-3-12-1 to B-3-12-6.

S.6 HARRIS, G.M. (1974) Ground settlement above a tunnel in silt — a case record.
Tunnels and Tunnelling, Vol. 6, No. 4, pp. 50 — 3.

FR/CP/5 121
 F.7 McCAUL, C., DOBSON, C., COOPER, I. and SPENCER, I.M. (1983)
Ground movements caused by tunnelling in loose fill. Transport and Road
Research Laboratory. Supplementary Report, 781.

Note:

R = Residual soil
S Silt
F Fill

122 FR/CP/5
 Appendix M Supplementary bibliography

ATKINSON, J.H., ORR, T.L.L. and POTTS, D.M. (1975) Research studies into the behaviour
of tunnels and tunnel linings in soft ground. Transport and Road Research Laboratory,
Laboratory Report 176 UC.

ATKINSON, J.H. and MAIR, R.J. (1981) Soil mechanics aspects of soft ground tunnelling.
Ground Engineering. Vol. 14. No. 5. pp. 20 — 26 and 38.

ATTEWELL, P.B. and BODEN, J.B. (1971) Development of stability ratios for tunnels driven
in clay. Tunnels and Tunnelling, Vol. 3, No. 3, pp. 195 — 8.

ATTEWELL, P.B. and FARMER, I.W. (1974) Ground deformation resulting from shield
tunnelling in London Clay. Canadian Geotech, J, Vol.11, No. 3, pp. 380 — 95.

ATTEWELL, P.B. and FARMER, I.W. (1975) Ground deformation caused by shield tunnelling
in silty alluvium at Willington Quay, North East England. Report to Transport and Road
Research Laboratory, Department of Environment.

ATTEWELL, P.B. (1978) Large ground movements and structural damage caused by tunnelling
below the water table in a silty alluvial clay. Proc. Conf. Large Ground Movements and
Structures Cardiff, Edited by J.D. Geddes. Published as Large Ground Movements and
Structures. Pentech Press, Plymouth 1978, pp. 307 — 56.

ATTEWELL, P.B., GLOSSOP, N.P. and FARMER, I.W. (1978) Ground deformation caused by
tunnelling in a silty alluvial clay. Ground Engineering, Vol.11, No. 8, pp. 32 — 41.

ATTEWELL, P.B. and FARMER, I.W. (1974) Ground disturbance caused by shield tunnelling
in stiff overconsolidated clay. Engineering Geology, Vol. 8, No. 4, pp. 361 — 81.

ATTEWELL, P.B. and FARMER, I.W. (1975) Ground settlement above shield driven tunnels in
clay. Tunnels and Tunnelling, Vol. 7, No. I, pp. 58 — 62.

ATTEWELL, P.B., FARMER, I.W., GLOSSOP, N.H. and KUSZNIR, N.J. (1975) A case
history of ground deformation caused by tunnelling in laminated clay. Proc. Conf. Subway
Construction, Budapest, pp. 165 — 78.

BARR. M.V. (1977) Downhole instrumentation — a review for tunnelling ground investigation.
CIRIA Technical Note No. 90.

BARRATT, D.A. (1978) Rapid data reduction of ground movement measurements. Tunnels and
Tunnelling, Vol.10, No. 3, pp. 46 — 47.

BARTLETT, J.V. and BUBBERS, B.L. (1970) Surface movements caused by bored tunnelling.
Proc. Conf. Subway Construction, Budapest, pp. 513 — 89.

BARTLETT, J.V. NOSKIEWICZ, T.M. and RAMSAY, J.A. (1965) Soft ground tunnelling for
the Toronto Subway. Proc. Instn. Civ. Engrs. September, Vol. 32, pp. 53 — 75.

BRAHMA, C.S. (1977) Geotechnical aspects of tunnelling in soft clay. Proc. Int. Symp. Soft
Clay, Bangkok, pp. 571 — 86.

BRAHMA, C.S. and KU, C.C. (1982) Ground response to tunnelling in residual soil. Proc.
Speciality Conf. Engineering and Construction in Tropical and Residual Soils. Am. Soc. Civ.
Engrs. Geotcch. Engng Div., Honolulu, pp. 578 — 87.

FRCP/5 123
 BRETH, H. and CHAMBOSSE, G. (1974) Settlement behaviour of buildings above subway
tunnels in Frankfurt clay. Proc. Conf. Settlement of Structures, Cambridge, pp. 329 — 36.

BURKE, H.H. (1957) Garrison Dam — tunnel test section investigation. Proc. Am. Soc. Civ.
Engrs - J. Soil Mech. and Found. Div., Vol. 83, No. SM4 Paper 1438.

BURLAND, J.B. and MOORE, J.F.A. (1973) The measurement of ground displacement around
deep excavations. Proc. Symp. Field Instrumentation in Geotech Engng. Brit. Geotech. Soc.,
London, pp. 70 — 84.

CARDER, D.R., RYLEY, M.D. and SYMONS, I.F. (1985) Ground movements caused by deep
trench construction in boulder clay and their effect on a adjacent service pipeline. Proc. 3rd Int.
Conf. Ground Movement and Structures, Cardiff. Published as Ground Movements and
Structures, Edited by J.D. Geddes. Pentech Press, Plymouth5, pp. 3 — 23.

CHAMBOSSE, G. (1972) The deformation behaviour of Frankfurt Clay during tunnel driving.
Mitt. Vers. — Anst. Bodenmach, u. Grundg. No. 10 TH Darmstadt.

CORDING, EJ., HENDRON, AJ., HANSMIRE, W.H, MAHAR, J.W., MacPHERSON, H.H.,
JONES, R.A. and O'ROURKE, T.D. (1975) Methods of geotechnical observations and
instrumentation in tunnelling. Department of Civil Engineering. University of Illinois. (NTIS,
PB 252585).

CROFTS, J.E., MENZIES, B.K. and TARZI, A.I. (1977) Lateral displacement of shallow
buried pipelines due to adjacent deep trench excavations. Geotechnique, Vol. 27, No. 2,
pp. 161 — 79.

D'APPOLONIA, DJ. (1971) Effects of foundation construction on nearby structures. Proc. 4th
Pan Am Conf. Soil Mech. and Found. Engng., San Juan, Vol.!, pp. 189 — 236.

DAVIES, R. and HENKEL, DJ. (1980) Geotechnical problems associated with construction at
Chater Station, Paper J3. Mass Transportation in Asia Conference, Hong Kong.

DAVIES, E.H., GUNN, MJ., MAIR, R.J. and SENEVIRATNE, H.N. (1980) The stability of
shallow tunnels and underground openings in cohesive material. Geotechnique, Vol. 30, No. 4,
pp. 397 — 416.

DAVIES, T.P. CARTER, P.G., MILLS, D.A.C. and WEST, G. (1981) Kielder aqueduct tunnels
— predicted and actual geology. Department of Environment/Department of Transport.
TRRL SR 676.

DUNNICL1FFE, J. and DEERE, D.U. (1984) Judgement in geotechnical engineering: the

professional legacy of Ralph B. Peck. J. Wiley, New York.

MINTON, C.E., KELL, J. and MORGAN, H.D. (1965) Victoria Line : Experimentation,
design, programming and early progress. Proc. Instn. Civ. Engrs., (May), Vol. 31, Paper
No. 6845.

EISENSTEIN, Z. and THOMSON, S. (1978) Geotechnical performance of a tunnel in till.

Canadian Geotech. J, Vol. 15, No. 2, pp. 332 — 45.

EATHERLEY, MJ. (1977) The design and construction of Bush Lane House. Structural
Engineer, Vol. 55, No. 2, pp. 75 — 86.

ESCARIO, V., GARCIA GONZALEZ, J.M., MOYA, J.F., OTEO, C.S. and SAGASETA, C.
(1977) Geotechnical problems during the construction of Madrid Subway Extension. Proc. 9th
Int. Conf. Soil Mech. and Found. Engng., Tokyo, Case History Vol., pp. 727 — 56.

124 FR/CP/5
 FARMER, I.W. (1978) Case histories of settlement above tunnels in clay. Proc. Conf. Large
Ground Movements and Structures, Cardiff. Published as Large Ground Movements and
Structures. Edited by J.D. Geddes., Pentech Press, Plymouth, 1978. pp. 357 — 71.

FOLLENFANT, H.G., CUTHBERT, E.W. (1969) The Victoria Line. Proc. Instn. Civ. Engrs.
Supplementary Vol., Paper 7270S, pp. 337 — 475.

GEDDES, J.D. and KENNEDY, D. (1985) Structural implications of horizontal ground strains.
Proc. 3rd Int. Conf. Ground Movements and Structures, Cardiff. Published as Ground
Movements and Structures. Edited by J.D. Geddes. Pentech Press, Plymouth 1985,
pp. 610 — 29.

GEDDES, J.D. (1981) Ground movements and structures. Proc. 2nd Int. Conf. Ground
Movements and Structures, Cardiff 1980. Published as Ground Movements and Structures
Edited by J.D. Geddes. Pentech Press, London.

GHABOUSSI, J. and GIODA, G. (1977) On the time-dependent effects in advancing tunnels.

Int. J. for Numerical and Analytical Methods in Geomechanics, Vol. 1, pp. 249
59. —

GLOSSOP, N.H. and O'REILLY, M.P. (1982) Settlement caused by tunnelling through soft
marine silty clay. Tunnels and Tunnelling, Vol. 14, No. 9, pp. 13 — 16

GOLDER, H.Q. (1971) The allowable settlement of structures. Proc. 4th PanAm Conf. Soil
Mech. and Found. Engng., San Juan, Vol. 1, pp. 171 — 87.

GARCIA GONZALEZ, J.M. (1972) Spain : Madrid metro, current works. Tunnels and
Tunnelling, Vol. 4, No. 2, pp. 147 — 53.

GOLSER, J., HACKL, E. and JOSTL, J. (1977) Tunnelling in soft ground with the new
Austrian Tunnelling Method (NATM). Ground Engineering Review. Highlights of the IXth Int.
Conf. on Soil Mechanics and Foundation Engineering. Tunnelling in soft ground. Information
from the Speciality Session No. 1 in Tokyo, pp. 16 — 20.

GRAF, E.D. (1969) Compaction grouting technique and observations. Proc. Am. Soc. Civ.
Engrs J. Soil Mech. and Found. Div., Vol. 95, No. SMS, pp. 1151 58.
— —

GRESCHIK, G., PALOSSY, L. and SCHARLE, P. (1982) Investigation of the active tunnel
supporting system's effect on the surface settlements. Rock mechanics: caverns and pressure
shafts. ISRM Symposium, Vol. 2 (AACHEN), pp. 599 — 606.

GUMBEL, J.E., O'REILLY, M.P., LAKE, L.M. and CARDER, D.R. (1982) The development
of a new design method for buried flexible pipes. Europipe '82 Conference, Basel, Switzerland,
Paper 8, pp. 87 — 99.

HARRIS, J.S. and NORIE, E.H. (1982) Construction of two short tunnels using artificial ground
freezing. Proc. 3rd Int. Symp. Ground Freezing, Hanover, New Hampshire, pp. 383 — 8.

HARRISON, T.A. (1981) Early age Thermal Crack Control in Concrete. CIRIA Report 91.
-

HOLMSEN, G. (1953) Regional settlements caused by a subway tunnel in Oslo. Proc. 3rd Int.
Conf. Soil Mech. and Found. Engng, Zurich, Val, pp. 381 3. —

HOOPER, J.A. (1984) Raft analysis and design — some practical examples. Structural Engineer,
Vol. 62A, No. 8, pp. 233 — 44.

HOUSEL, W.S. (1943) Earth pressure on tunnels. Trans. Am. Soc. Civ. Engrs, No. 108,
pp. 1037 — 58. Discussion pp. 1059 — 109.

FR/CP/5 125
 HOWAT, M.D. and CATER, R.W. (1983) Ground settlements due to tunnelling in weathered
granite. Proc. Int. Symp. on Engng. Geol. and Underground Construction. Int. Assoc. of Engng.
Geol. Lisboa, Portugal, Vol.1,

HOWE, M. and HUNTER, P. (1983) Tunnelling '82. Discussion Session 4 — Geotechnics

(settlement). Discussion of O'Reilly and New (1983) by Howe, M. Hunter, P. and Leach, G.
Extracts from Transaction/Section A, Instn. Mining of Metallurgy, Vol. 92, pp. A28 — A61.

HOWE, M. (1982) Damage Control for Distribution Systems. First steps in risk assessment.
Gas Engineering and Management, October. pp. 377 — 401

HUDSON, J.A. and RYLEY, M.D. (1978) Measuring horizontal movement of the ground.
Tunnels and Tunnelling, Vol.10, No. 2, pp. 55 — 8.

HUDSON, J.A. and PRIEST, S.D. (1975) Instrumentation and monitoring. Tunnels and
Tunnelling, Vol. 7, No. 5, pp. 64 — 70. Discussion, Vol. 7, No. 6, pp. 59 — 65.

THE INSTITUTION OF CIVIL ENGINEERS (1977) Ground Subsidence. Thomas Telford,

London.

THE INSTITUTION OF STRUCTURAL ENGINEERS (1975) The Design and Construction of

Deep Basements. Instn. Struct. Engrs, London

THE INSTITUTION OF STRUCTURAL ENGINEERS (in association with ICE.) (1984) Soil
Structure Interaction, November, Instn. Struct Engrs, London.

THE INSTITUTION OF STRUCTURAL ENGINEERS (1981) Supplementary Report on the

Design and Construction of Deep Basements, Instn. Struct Engrs, London.

THE INSTITUTION OF STRUCTURAL ENGINEERS (1971) Report on Stability of Modern

Buildings. September, Instn. Struct Engrs, London.

THE INSTITUTION OF STRUCTURAL ENGINEERS (1980) Appraisal of Existing Structures

July, Instn. Struct Engr, London.

JANIN, J.J. and LE SCIELLOUR, G.F., (1970) Chemical grouting for Paris Rapid Transit
Tunnels. J. Constn Div. Proc. ASCE, June. Vol. 96, No. COI, pp. 61 — 74.

KEOGH, G.S. and SCOTT, C.R. (1978) Ground movements associated with the failure of a
tunnel lining in the London Clay. Proc. Conf. Large Ground Movements and Structures, Cardiff
1977. Published as Large Ground Movements and Structures. Edited by J.D. Geddes. Pentech
Press, Plymouth, pp. 411 — 26.

KITAMURA, M., SUMIKICHI, I. and FUGIWARA, T. (1981) Shield tunnelling performance

ad behaviour of soft ground. Proc. Rap. Excay. Tuna. Conf., San Francisco, Vol.!,
pp. 201 — 20.

LACROIX, Y. (1966) Observational approach and instrumentation for construction on

compressible soils, Utilization of Sites for Soft Foundations, Highway Research Record No. 133,
Washington DC.

LITTLEJOHN, G.S. (1974) Observation of brick walls subjected to mining subsidence. Proc.
Conf. Settlement of Structures. Cambridge, pp. 384 — 93.

LO, K.Y., NG, M.C. and ROWE, R.L. (1984) Predicting settlement due to tunnelling in clays.
Proc. Geotech '84 — Tunnelling in Soil and Rock., Atlanta, Georgia, pp. 47 — 76.

126 FR/CP/5
 McCANN, D., BARTON, K.J. AND HEARN, K. (1981) Geophysical Borehole Logging, with
Special Reference to Altnabreac Caithness. Radioactive Waste Disposal Research Series:
Altnabreac, August. Environmental Protection Unit ENPU 81-11. IGS-NERC.

McCAUL, C., MORGAN, J.M. and BODEN, J.B. (1976) Measurement of Ground Movement
Due to Excavation of a Shallow Tunnel in Lower Chalk. Transport and Road Research
Laboratory. Supplementary Report 199, UC.

McCAUL, C. (1978) Settlements Caused by Tunnelling in Weak Ground at Stockton on Tees.- -

Transport and Road Research Laboratory. Supplementary Report 383.

MacLEOD, I.A. and ABU-EL-MAGD, S. (1980) The behaviour of brick-walls under conditions
of settlement. Struct. Engr, Vol. 58A, No. 9, pp. 279 — 86.

MacPHERSON, H.H. (1978) Settlement Around Tunnels in Soil: The Case Histories. Final
report by University of Illinois to Department of Transportation, Washington DC Report No.
UMTA-1L-06-0043-78- 1.

MAGATA, H., NAKAUCHI, S. and SOGABE, H. (1982) Subsidence measurements associated

with shield tunnelling in soft ground. Proc. Tunnelling '82, Brighton, pp. 231 — 40.

MAIR, R.J. (1983) Geotechnical Aspects of Soft Ground Tunnelling. International Seminar on
Construction Problems in Soft Soils. December, Singapore.

MATHESON, D.S. (1970) A tunnel roof failure in till. Canadian Geotech. J. Vol. 7, No. 3,
pp. 313 — 17.

MAYNARD, T.R. and O'ROURKE, T.D. (1977) Soil movement effect on adjacent public
facilities. Proc. Am. Soc. Civ. Engrs. Fall Convention Exhibit, San Francisco. Reprint 3111,
pp. 1 — 30.

MIKI, G., SAITO, T. and YAMAZAKI, H. (1977) Theory and praxis of soft ground tunnelling
by the slurry mole method. Ground Engineering Review. Highlights of the IXth Int. Conf. on
Soil Mechanics and Foundation Engineering. Tunnelling in soft ground. Information from the
Speciality Session No. 1. pp. 6 — 11

MILLIGAN, V. (1977) The Uncertain Equation between Design and Construction in Soil
Engineering. Int. Soc. SMFE 9th Int. Conf., Tokyo. Speciality Session No. 3 (Relationship
between design and construction in soil engineering).

MITCHELL, R.J. (1983) Earth Structure Engineer. Allen and Unwin, London.

MITCHELL, J.M. and TREHARNE, G. (1976) Pile foundations designed for nearby tunnelling.
Proc. 6th Euro. Conf. Soil Mech. and Found. Engng., Vienna, pp. 523 28.—

MUIR WOOD, A.M. (1969) Deep excavation and tunnelling in soft ground (panel discussion).
Proc. 7th Int. Conf. Soil Mech. and Found. Engng., Mexico City, Vol. 3, pp. 363 5.—

MUIR WOOD, A.M. and GIBBS, F.R. (1971) Design and construction of the cargo tunnel at
Heathrow Airport, London. Proc. Instn. Civ. Engrs. January, Vol. 48, Paper 7357, pp. 11 — 34.

MULLER-SALZBURG, L. and SPAUN, G. (1977) Soft ground tunnelling under buildings in

Germany. Proc. 9th Int. Conf. Soil Mech. and Found. Engng, Tokyo, Voll, pp. 663 8. —

MUNCHENER RUCKVERSICHERUNGS GESELLSCHAFT (1980) Underground Railways

-

Construction and Insurance. Caarl Gerber, Munchen.

NATIONAL WATER COUNCIL (1983) Model Consultative Procedure for Pipeline.

Construction involving Deep Excavations. January, NWC, London.

FR/CP/5 127
 NEEDHAM, D. and HOWE, M. (1979) Why Pipes Fail. Communication 1103, The Institution
of Gas Engineers, November, London.

NEW, B.M. (1978) The effects of ground vibration during bentonite shield tunnelling at
Warrington. Transport and Road Research Laboratory. Laboratory Report 860 Crowthorne.

NEW, B.M., WILD, P.T. and BISHOP, C.J. (1980) Bentonite tunnelling beneath major services
in loose ground. Tunnels and Tunnelling, Vol. 12, No. 5, pp. 14 — 16.

O'ROURKE, T.D. and KUMBHOJKAR, A.S. (1984) Field Testing of Cast Iron Pipeline.
Response to Shallow Trench Construction. Geotechnical Engineering Report 84-3, June. Cornell
University, Ithaca, New York.

O'ROURKE, T.D. (Ed). (1984) Guidelines for Tunnel Lining Design. Geotechnical Engineering
Division. UTRc, ASCE, New York.

ORCHARD, R.J. (1982) Vitrified Clay Pipes in Areas of Mining Subsidence. Clay Pipe
Development Association Ltd, London.

OTEO, C.S. and MOYA, J.F. (1979) Estimation of the soil parameters of Madrid in relation to
tunnel construction. Proc. 7th Euro. Conf. Soil Mech. and Found. Engng, Brighton, Vol. 3,
pp. 239 — 47.

OTEO, C.S. and MOYA, J.F. (1979) Settlements induced by a tunnel in Miocenic soft rocks of
Madrid. Proc. 4th Congress Int. Soc. Rock Mech. Montreux, Vol.1, pp. 715 — 22.

OTTAVIANI, M. and PELLI, F. (1983) Influence of depth and of distance between the axes on
surface displacements due to the excavation of twin shallow tunnels. Proc. Int. Symp. on Engng.
Geol. and Underground Construction, Lisboa-Portugal, Vol.l.

PADFIELD, C.J. and SHARROCK, M.J. (1983) Settlement of Structures on Clay Soils.
PSA/CIRIA Publication. PSA Civil Engng Tech. Guide No. 38. CIRIA Special Publication 27.

PARISH, W.C.P., BAKER, W.I.T. and PURBRIGHT, R.M. (1983) Underpinning with
Chemical Grout. Civil Engng/ASCE, August.

PECK, R.B. (1984) State of the Art : Soft Ground Tunnelling, 'Tunnelling in Soil and Rock'.
Proc. Geotech. '84. Atlanta, Georgia, May. (Comments on sink holes interaction of structures.)

PECK, R.B., DEERE, D.U. and CAPACETE, J.L. (1956) Discussion of building damage. Proc.
lnstn. Civ. Engrs, Vol. 5, Part III, pp. 778 — 81.

PECK, R.B., DEERE, D.U., MONSEES, PSTKRT, H.W. and SCHMIDT, B. (1969) Some
design considerations in the selection of underjground support systems. Department of Civil
Engineering University of Illinois, Urbana. Contract No. 3-0152 for Office of High Speed
Ground Transportation and Urban Mass Transportation Administration. US Department of
Transport Washington DC.

PHAROAH, D.F.H. (1976) Small bore tunnelling for a sewer in difficult ground conditions.
Proc. Conf. Underground Engng — The Next Decade, London.

PRIEST A.V. and ORCHARD R.J. (1957 — 1958) Recent Subsidence Research in Notts and
derby coalfields. Transactions, Inst. of Mining Engrs. Vol. 117, pp. 503-513.

RUMSEY, P.B. and DORLING, C. (1985) The prediction of ground movement and induced
pipe strain caused by trench excavation. Proc. 3rd Int. Conf. Ground Movements and Structures,
Cardiff 1984. Published as Ground Movements and Structures. Edited by J.D. Geddes. Pentech
Press, Plymouth, pp. 24 — 49.

128 FRCP/5
 RYLEY, M.D., JOHNSON, P.E., O'REILLY, M.P. and BARRATT, D.A. (1998) Measurements
of ground movements around three tunnels in loose cohesionless soil. Transport and Road
Research Laboratory. Laboratory Report 938.

SELBY, A.R. (1985) Tolerance of highway bridges to ground movements induced by tunnelling
in soil. Proc. 3rd Int. Conf. Ground Movements and Structures, Cardiff 1984. Published as
Ground Movements and Structures. Edited by J.D. Geddes. Pentech Press, Plymouth 1985,
pp. 630 — 44.

SYMONS, I.F., CHARD, B. and CARDER, D.R. (1982) Ground Movements Caused by Deep
Tunnel Construction. Restoration of Sewerage Systems. Thomas Telford, London, pp. 87 104.—

SZECHY, K. (1970) Surface settlement due to the tunnelling method in cohesionless soils.
Proc. Conf. on Subway Construction. Budapest, pp. 615 25.—

SZECHY, K. (1973) Surface settlements due to tunnelling activities. In The Art of Tunnelling,
(2nd English Edition). Akademiaikiado, Budapest, pp. 1044 — 78.

TAN, D.Y. and CLOUGH, G.W. (1980) Ground control for shallow tunnels by soil grouting.
Proc. Am. Soc. Civ. Engrs J. Geotech. Engng Div, Vol. 106, No. GT9, pp. 1037 57.
— —

TAKI, H. and O'ROURKE, T.D. (1984) Factors affecting the performance of cast iron pipe.
Geotechnical Engineering Report 84 1, March. Cornell University, Ithaca, New York.
-

TATTERSALL, F., WAKELING, T.R.B. and WARD, W.H. (1955) Investigations into the
design of pressure tunnels in London Clay. Proc. Instn. Civ. Engrs, (Part 1), Vol. 4, No. 4,
pp. 400 — 71.

TERZAGHI, K. (1950) Geologic aspects of soft ground tunnelling. In Applied sedimentation.

Edited by P.D. Track, John Wiley, New York, pp. 193 — 209.

THOMAS, H.S.H. (1983) Observations on the behaviour of a pilot tunnel in London Clay as
the main tunnel was driven. Proc. lnstn. Civ. Engrs, (Part 1), Vol. 74, pp. 15 — 24.

VINNEL, C. and HERMAN, A. Shield tunnelling in Brussels sand. Proc. 7th Int. Conf. Soil
Mech. and Found. Engng, Mexico City, Vol. 2, pp. 487 — 94

WARD, W.H. (1969) Discussion of Peck R.B., Deep excavations and tunnelling in soft ground.
Proc 7th Int. Conf. Soil Mech. Foundn. Engng, Mexico City, Vol. 3, pp. 320 25.—

WARD, W.H. and THOMAS, H.S.H. (1965) The development of earth loadings and
deformations in tunnel linings in London Clay. Proc. 6th Int. Conf. Soil Mech. and Found.
Engng, Montreal, Vol. 2, pp. 432 — 36.

WARNER, J. (1982) Compaction Grouting — The first thirty years. Grouting in Geotechnical
Engineering. Proc. Conf. on Geotechnical Engineering, New Orleans, February, Pub. ASCE.

WATANABE, T. (1979) Subway tunnelling in soft ground. Proc. 6th Asian Regional Conf. Soil
Mech. and Found. Engng, Singapore, Vol. 2, pp. 77 — 121.

WEST, G., HEATH„ W.G. and McCAUL, C. (1981) Measurements of the effects of tunnelling
at York Way, London. Ground Engineering, Vol. 14, No. 5, pp. 45 — 53.

FR/CP/5 129
 CIRIA
sharing knowledge
■
building best practice

6 Storey's Gate, Westminster, London SW1 P 3AU

Tel: (+44) (0) 171 222 8891 Fax: (+44) (0) 171 222 1708
is h.• Jo a ors.
TROMNIVATAWN01111310101111111