Geotechnical Aspects of Underground

Construction in Soft Ground:

Proceedings of the Tenth International
Symposium on Geotechnical Aspects
of ... Cambridge, United Kingdom, 27-29
June 2022 1st Edition Mohammed
Elshafie (Editor)
Visit to download the full and correct content document:
https://ebookmeta.com/product/geotechnical-aspects-of-underground-construction-in-
soft-ground-proceedings-of-the-tenth-international-symposium-on-geotechnical-aspec
ts-of-cambridge-united-kingdom-27-29-june-2022-1st-edition/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Geotechnical Aspects of Underground Construction in

Soft Ground: Proceedings of the Tenth International
Symposium on Geotechnical Aspects of ... Cambridge,
United Kingdom, 27-29 June 2022 1st Edition Mohammed
Elshafie (Editor)
https://ebookmeta.com/product/geotechnical-aspects-of-
underground-construction-in-soft-ground-proceedings-of-the-tenth-
international-symposium-on-geotechnical-aspects-of-cambridge-
united-kingdom-27-29-june-2022-1st-edition/

Structural Information and Communication Complexity

29th International Colloquium SIROCCO 2022 Paderborn
Germany June 27 29 2022 Proceedings Merav Parter

https://ebookmeta.com/product/structural-information-and-
communication-complexity-29th-international-colloquium-
sirocco-2022-paderborn-germany-june-27-29-2022-proceedings-merav-
parter/

Integer Programming and Combinatorial Optimization 23rd

International Conference IPCO 2022 Eindhoven The
Netherlands June 27 29 2022 Proceedings Lecture Notes
in Computer Science 13265 Karen Aardal (Editor)
https://ebookmeta.com/product/integer-programming-and-
combinatorial-optimization-23rd-international-conference-
ipco-2022-eindhoven-the-netherlands-june-27-29-2022-proceedings-
lecture-notes-in-computer-science-13265-karen-aardal/

Intelligent Tutoring Systems 18th International

Conference ITS 2022 Bucharest Romania June 29 July 1
2022 Proceedings Scott Crossley

https://ebookmeta.com/product/intelligent-tutoring-systems-18th-
international-conference-its-2022-bucharest-romania-
june-29-july-1-2022-proceedings-scott-crossley/
Practical Aspects of Declarative Languages 24th
International Symposium PADL 2022 Philadelphia PA USA
January 17 18 2022 Proceedings 1st Edition James Cheney

https://ebookmeta.com/product/practical-aspects-of-declarative-
languages-24th-international-symposium-padl-2022-philadelphia-pa-
usa-january-17-18-2022-proceedings-1st-edition-james-cheney/

Practical Aspects of Declarative Languages 24th

International Symposium PADL 2022 Philadelphia PA USA
January 17 18 2022 Proceedings 1st Edition James Cheney

https://ebookmeta.com/product/practical-aspects-of-declarative-
languages-24th-international-symposium-padl-2022-philadelphia-pa-
usa-january-17-18-2022-proceedings-1st-edition-james-cheney-2/

Cambridge IGCSE and O Level History Workbook 2C - Depth

Study: the United States, 1919-41 2nd Edition Benjamin
Harrison

https://ebookmeta.com/product/cambridge-igcse-and-o-level-
history-workbook-2c-depth-study-the-united-states-1919-41-2nd-
edition-benjamin-harrison/

Formal Aspects of Component Software 17th International

Conference FACS 2021 Virtual Event October 28 29 2021
Proceedings Lecture Notes in Computer Science Gwen
Salaün Editor Anton Wijs Editor
https://ebookmeta.com/product/formal-aspects-of-component-
software-17th-international-conference-facs-2021-virtual-event-
october-28-29-2021-proceedings-lecture-notes-in-computer-science-
gwen-salaun-editor-anton-wijs-editor/

Proceedings of the 23rd International Symposium on

Advancement of Construction Management and Real Estate
Fenjie Long (Editor)

https://ebookmeta.com/product/proceedings-of-the-23rd-
international-symposium-on-advancement-of-construction-
management-and-real-estate-fenjie-long-editor/
M.Z.E.B. Elshafie, G.M.B. Viggiani & R.J. Mair, editors
GEOTECHNICAL ASPECTS OF UNDERGROUND CONSTRUCTION IN
SOFT GROUND
PROCEEDINGS OF THE TENTH INTERNATIONAL SYMPOSIUM ON GEOTECHNICAL
ASPECTS OF UNDERGROUND CONSTRUCTION IN SOFT GROUND, IS-CAMBRIDGE
2022, CAMBRIDGE, UNITED KINGDOM, 27-29 JUNE 2022

Geotechnical Aspects of
Underground Construction in
Soft Ground

Editors
Mohammed Z.E.B. Elshafie
Qatar University, Doha, Qatar

Giulia M.B. Viggiani

University of Cambridge, UK

Robert J. Mair
University of Cambridge, UK
Cover top photo credit: Tideway
Bottom cover photo: © Copyright Crossrail

CRC Press/Balkema is an imprint of the Taylor & Francis Group, an informa business

© 2021 ISSMGE, London, UK

Typeset by Integra Software Services Pvt. Ltd., Pondicherry, India

All rights reserved. No part of this publication or the information contained herein may be
reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, by photocopying, recording or otherwise, without written prior
permission from the publisher.

Although all care is taken to ensure integrity and the quality of this publication and the
information herein, no responsibility is assumed by the publishers nor the author for any
damage to the property or persons as a result of operation or use of this publication and/or
the information contained herein.

Library of Congress Cataloging-in-Publication Data

Applied for
Published by: CRC Press/Balkema
Schipholweg 107C, 2316XC Leiden, The Netherlands
e-mail: Pub.NL@taylorandfrancis.com
www.routledge.com – www.taylorandfrancis.com

ISBN: 978-0-367-33733-9 (Hbk)

ISBN: 978-1-032-02769-2 (Pbk)
ISBN: 978-0-429-32155-9 (eBook)
DOI: 10.1201/9780429321559
https://doi.org/10.1201/9780429321559
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

Table of contents

Preface to the current edition xiii

Organisation xv

Sponsors xvii

Field case studies and sensing technologies

Ovalisation of cast-iron tunnels in response to nearby tunneling work 3
M. Alhaddad & K. Soga
Lubrication characteristics of pipejacking in soft alluvial deposits 12
W.-C. Cheng, G. Li & D.E.L. Ong
A recent subway construction incident in soft alluvial deposits of Taiwan 19
W.-C. Cheng, G. Li & M.M. Rahman
Settlements due to tunneling in the City of São Paulo 26
A. Lopes dos Santos, W. Bilfinger & H.C. Rocha
Evaluation of geological/geotechnical geostatistical models for tunneling applications 35
R. Gangrade, W. Trainor-Guitton, M. Mooney & J. Grasmick
Pillar uplift observed in braced excavation of a subway station 43
Y.B. Gao, Z.Y. Yang, T.J. Yao & X.M. Tang
Solution implementation for a tunnel collapse in a weak embankment soil: The case of the
Ali Boumendjel Metro station, Algiers, Algeria 51
R. Hebib, Z. Derriche, B. Alloul & D. Belhai
Design of large twin-wall cofferdams for ship impact 60
B.D. Jones, E. Murphy & P.J. Astle
Design of large temporary works ship impact protection structures 68
B.D. Jones, E. Murphy & P.J. Astle
Large-scale improvement work of subway stations in soft ground 74
S. Konishi, T. Moriya, R. Fukuda, T. Murakami & K. Okanoya
Comparison between predicted and measured performance of a deep excavation in soft clay
in Gothenburg, Sweden 83
J. Langford, K. Karlsrud, E. Bengtsson, C. Hof & R. Oscarsson
Grouting for the rescue of stuck TBM in conglomerate boulder layer 91
H.-J. Liao, S.-J. Weng, C.C. Ho & R. K.N. Wong

v
Tunnelling in Urban Areas and Pile Interception Challenges – A Case Study: Bank Station
Upgrade Project (BSCU) 98
A. Nasekhian, C. Anthony, B. Haig, M. Dewhirst, J. Ares & C. Barker
Impacts of new development on existing underground assets using greenfield model 108
K. Neaupane & Y. He
Cementitious systems with carbon nanomaterials for underground infrastructure 115
I. Papanikolaou, A. Al-Tabbaa, M. Goisis & T. Embley
Pipe jacking tunnel construction crossing the Bang Pakong River 123
K. Piriyakul, S. Pochalard & E. Rungrueng
Nine Elms Station substructure – implementation of the Observational Method by
progressive modification 129
Nigel Pye, C. Lile, Anthony O’Brien & C. Penh
Evaluation of InSAR data for measuring the surface settlement during shield tunnel
construction of the North-South Line in Amsterdam 137
K.J. Reinders, F.J. van Leijen, R.F. Hanssen & M. Korff
Deep excavations for buildings in the Sabana Formation Bogota 153
J.A. Rodriguez
EPB tunneling and shaft construction in soft deltaic deposits for a railway link to Barcelona
airport 145
S. Sánchez, E. Silva, J. Izquierdo, A. Gens & E. Alonso
Use of jet grouted columns for deepening of basements and underpinning existing structures
at Valkyrien in Oslo 161
T. Sandene, K. Karlsrud & A. Worren
Settlements induced by EPB TBMs tunneling, a case study of theoretical and monitoring
values 169
A. Siemińska-Lewandowska & R. Kuszyk
Monitoring of an existing concrete-lined tunnel at CERN excavated in the molasse rock 177
K. Soga, V. Di Murro, L. Pelecanos, C. Kechavarzi, L. Scibile, J.A. Osborne & R.F. Morton
Achieving sustainability in tunnelling through innovation 184
A.H. Thomas
Measured post-construction ground response to EPBM tunnelling in London Clay 191
M.S.P. Wan, J.R. Standing, D.M. Potts & J.B. Burland
Gilgel Gibe II Hydropower Project in Ethiopia; TBM Tunnelling, when the rock turns into
mud: Analysis of a major collapse, its causes and solutions 199
E. Zoppis & A.M. Baldi

Modelling and testing of tunnels and deep excavations in soft ground

Torque estimation for tunnel boring machine design in soft ground 209
M. Ambrosi, K.B. Glab & W. Broere
Variability of soil stress-strain non-linearity for use in MSD analyses evaluated using
databases of triaxial tests on fine-grained soils 217
M.E.W. Beesley & P.J. Vardanega

vi
Asymmetric pressure distribution in EPB shields: Evaluation of measurements and numerical
simulations 226
A. Bezuijen, T.S. Dang & G. Meschke
Pore pressures in front of a slurry TBM, the influence of plastering mechanisms 234
A. Bezuijen
Face pressure and volume loss relationships for pressurized tunneling in Granular soils 242
S.J. Boone & J.N. Shirlaw
Albert Embankment: Design of deep excavations in the River Thames foreshore 250
O. Brown, P. Stewart, S. Thomson, B. Patel, A.M. Waller, S. Sismondi & F. Quesada
Centrifuge and numerical modelling of the influence of structural stiffness on basement
heave in over-consolidated clay 259
D.Y.K. Chan & S.P.G. Madabhushi
Influence of TBM geometry on lining loads of deep tunnels 266
V. De Gori, A. de Lillis & S. Miliziano
Development of a new excavation technique for centrifuge testing in sand 275
N.E. Faustin, R.J. Mair & M.Z.E.B. Elshafie
Interpretation of soil parameters used for numerical analysis with small strain model for
deep excavation in loose to medium dense sand 283
H.B. Bin-Chen, P. Khac Hai & C. Hung
Prediction of damage intensity of reinforced concrete tunnels and soil against blast loading 291
K. Senthil, S. Rupali & L. Pelecanos
EPB-TBM tunnel under internal pressure: Assessment of serviceability 300
N.A. Labanda, A.O. Sfriso, D. Tsingas, R. Aradas & M. Martini
A risk assessment of downdrag induced by reconsolidation of clays after upwards pipe
jacking 309
N.A. Labanda, A.O. Sfriso, D. Tsingas, R. Aradas & M. Martini
Small-scale modelling of pile drilling in sand – investigation of the influence on surrounding
ground 317
E.J. Lande, S. Ritter, E.J. Lande, H. Tyvold & S. Nordal
Excavation of an artificial tunnel using compressed air 325
N. Losacco, M. Cafaro & R. Marazzita
The use of adaptive smoothed finite-element limit analysis to seismic stability of tunnels 330
H.C. Nguyen
Upper bound analysis of seismic stability of tunnels using cell-based smoothed finite element 337
H.C. Nguyen
Investigation of the seismic performance of the complicated tunnel sections with
non-uniform heights 343
H. Nitta, S. Ito, T. Otsuka, S. Konishi, K. Tsuno, S. Tsuchiya & K. Maekawa
Influence of the annulus grout on the soil-lining interaction for EBP tunneling 350
M. Ochmański, G. Modoni & G. Spagnoli

vii
Long term additional load on a shield tunnel in soft clay due to clay consolidation with
water leakage 357
S. Oka, J. Saito, Y. Ito, W. Li, S. Kaneko, A. Afshani & H. Akagi
A study of a strut-free excavation system in deep excavations 365
C.Y. Ou, A. Lim, P.G. Hsieh & S.C. Chien
Increasing the passive resistance of deep excavations in very soft soils to mitigate ground
movements through centrifuge modelling 371
J.P. Panchal & A.M. McNamara
Machine learning algorithms applied to the blowout susceptibility estimation around
pressurized cavities in drained soil 379
F. P.-Ramirez & C. Arson
Specifying and Testing fibre Reinforced sprayed concrete: Advantages and Challenges of
some testing methods 385
Benoit De Rivaz
Undrained seismic response of tunnels 395
E.A. Sandoval & A. Bobet
Preliminary evidences on the influence of grains micro-structural features on the TBM tools
wear 401
D. Sebastiani, S. Miliziano, G. Guida & F. Casini
Front-face pressure drop during the standstill phase for EPB mechanized tunnelling in
coarse-grained soils 408
D. Sebastiani, S. Miliziano & A. Bezuijen
A miniature EPB TBM for use in a geotechnical centrifuge 415
C.J. Shepheard, A.S.N. Alagha, G.M.B. Viggiani & S.K. Haigh
Urban tunnelling in glacial soils: Tunnel de Champel, Geneva 421
W. Steiner, T. Witschi & A. Ferrari
Simplified stress-strain models applied to data from triaxial and pressuremeter tests on
London Clay 430
P.J. Vardanega, M.D. Bolton, S.K. Haigh, R.W. Whittle, A. Klar & M.G. Williamson
Propped cantilever wall stability design with the ‘What You Design Is What You Get’
method – background and development 438
C.K.S. Yuen
Multi-objective optimisation design for composite tunnel linings using non-dominated
sorting genetic algorithm 444
W. Zhai, D. Chapman, A. Faramarzi, H. Huang & D. Zhang
Support pressure transfer at a slurry supported tunnel face due to time dependent decrease of
soil permeability 451
C. Zhao, Z. Zizka, B. Schoesser, M. Thewes & A.A. Lavasan
SEM deformation prediction and observation by 3D numerical analysis 459
H. Zheng, M. Mooney, M. Gutierrez & C. Bragard
Face stability of slurry shield-driven tunnel in an aquifer 467
T. Xu, W.H. Zhou & A. Bezuijen

viii
Numerical analysis of Double-O-Tube shield tunneling in Shanghai 475
D. Zhou & L. Zdravković
Physical modelling of transient processes at the slurry supported tunnel face during shield
excavation 482
Z. Zizka, B. Schoesser & M. Thewes

Ground movements, interaction with existing structures and mitigation measures

Interaction between a newly excavated underground ramp and deep existing tunnels 491
A. Afshani, G. Hassan, H. Akagi & K. Endou
Pile driving interaction with existing tunnel 501
K.J. Bakker, R. Spruit & F.C.M. van Overstraten Kruijsse
Numerical modelling of framed structures with masonry infills affected by tunnelling­
induced deformation and damage 510
D. Boldini, N. Losacco, A. Franza & S.M. Miraei
Analytical investigation for the circumferential behavior of the shield-driven tunnel adjacent
to a braced excavation 517
H.Z. Cheng, R.P. Chen, H.N. Wu & F.Y. Meng
Winkler model for axial deformation of compressible piles due to tunnelling 523
J.J. Crispin
Bayesian inference for deep excavations 529
W.J. de Wolf, W.J. de Wolf, M. Korff, A. van Seters, & J.H. van Dalen
An underground excavation in Barcelona and its interaction with existing structures 536
A. Di Mariano, A. Varga & A. Gens
Equivalent frame model for the assessment of tunnel-induced damage to masonry buildings 545
D.B. Gulen, S. Acikgoz & H.J. Burd
Use of underpinning, horizontal jet grouting and ground freezing for ground stabilization to
control settlement of existing MRT tunnels during construction of a link-way and railway
tunnels for a new mrt line in Singapore 554
Kaoru Hashida, Tadashi Hashimoto, Yong Kwet Yew, Ramesh Nair & John Busbridge
Tunnel lining behaviors of close proximity large diameter parallel shield tunnels 563
T. Hashimoto, J. Nagaya, M. Isa & K. Fujiwara
Bolu NATM Tunnel, changes to the support system due to soft ground conditions and
deformation 569
S. Işık & Z. Buket
Use of spile umbrellas to reduce deformation 576
A. Klinger & J. Fillibeck
Modified gap method for prediction of TBM tunnelling-induced soil settlement in sand - a
case study 584
B.T. Le, N.T. Nguyen, S. Divall, R.J. Goodey & R.N. Taylor

ix
Long-term behaviour of ground around tunnel due to groundwater level fluctuations 590
Wei Li, Shigeaki Oka, Alireza Afshani & Hirokazu Akagi
Numerical investigation into time-dependent effects on short-term tunnelling-induced
ground response in London Clay 597
A.R. López, A. Tsiampousi, D.M.G. Taborda, J.R. Standing & D.M. Potts
Semi-coupled modelling of soil-structure interaction during tunnel construction: Two case
studies from Bank Station Capacity Upgrade 605
A. Luciano, M.N. Pascariello, E. Bilotta, S. Acikgoz & R. Mair
Centrifuge tests on tunnel-building interaction in liquefiable soil 613
G. Miranda, V. Nappa, E. Bilotta, S.K. Haigh & S.P.G. Madabhushi
The effect of deep excavation on existing railway tunnel 620
M. Mitew-Czajewska
Simplified modelling of the transient response of underground structures due to dynamic
loads generated from underground tunnels 627
L. Pelecanos, K. Senthil & S. Rupali
Mapping the risk of building damage due to excavation-induced displacements 632
S. Ritter, L. Piciullo, A.O. Kydland Lysdahl, M. Kahlström, J. Langford & F. Nadim
A new approach for compensation grouting in highly permeable gravel 640
M. Sailer, J. Fillibeck & S. Geuder
A case study on the effects of anchor drilling in soft, low sensitive clay and sandy, silty soils 647
T. Sandene, S. Ritter & E.J. Lande
Prediction of long-term settlement in shield tunnel using GA-BP neural network 656
Yi-Ming Shen, Dong-Mei Zhang, Jie Zhang, Dong-Mei Zhang & Jie Zhang
Tunnelling through a piled foundation: Interaction effects 664
Davor Simic & Belén Martínez-Bacas
The use of protective structures to reduce tunnelling induced damage to buildings 673
G. Song, A.M. Marshall & C.M. Heron
Evaluation method on ground movement using continuum ground model 681
M. Sugimoto, J. Chen, P.T. Anh, K. Manabe, L.G. Lam & S. Chaiyaput
The response of the Europier Terminal Building to the excavation of the T2B Basement at
Heathrow Airport 689
G.R. Taylor
Longitudinal structural deformation of shield tunnels induced by overlying excavation 696
H.N. Wu, S. Chen, R.P. Chen, Y. Liu, H.Z. Chen, F.Y. Meng & S.L. Shen
Simulating the water-soil leakage induced deformation around the shield tunnel with
material point method 702
X.C. Xie, D.M. Zhang, M.L. Zhou, S.J. Feng, D.M. Zhang & S.J. Feng
Influencing factors and protection technologies of underground diaphragm wall and deep
foundation pit construction on metro station 710
H.F. Xing, L.L. Liu & H. Zhang

x
A centrifuge modelling study on the effect of foundation configuration on tunnel-frame
interaction 719
J. Xu, A.M. Marshall & A. Franza
Modelling ground response to TBM tunnelling with active face support 727
T. Xu, W.H. Zhou & A. Bezuijen
Tunnel face stability considering drainage and surface settlements 732
C. Yi, S. Senent & R. Jimenez
Impact of subsoil spatial variability on deformations of immersed tunnel 738
X. Zhang, X. Wu & W. Broere
Coupled elastoplastic analysis of the soil-pile foundation interaction induced by deep
excavations 746
C. Zheng, A. Franza & R. Jimenez

Design and application of ground improvement for underground construction

Microcapsule-based self-healing cement stabilised clay 753
B. Cao, C. Litina, L. Souza & A. Al-Tabbaa
Application of biopolymer hydrogel for ground hydraulic conductivity control under
pressurized conditions 761
I. Chang, G-C. Cho & A.T.P. Tran
Soil enhancement via microbially induced calcite precipitation 765
C. Konstantinou & G. Biscontin
Base grouting against uplifting water for a deep excavation in Taipei Basin 773
H.J. Liao, S.J. Weng, S.H. Cheng & R.K.N. Wong
The largest tunnels in freshly consolidated soft soil: Tuen Mun Chek Lap Kok Link Subsea
tunnels, Hong-Kong 780
T. Lockhart
Soil conditioning for EPB tunnelling in coarse grained soils based on laboratory model tests 788
A.S. Merritt, S.A. Jefferis & R.B. Storry
Influence of mineral or polymeric modification on bentonite-based tunnel face support 796
P. Mianji, W. Baille, T. Wichtmann, R. Verst & M. Pulsfort
Compensation grouting for conventional tunnelling with low overburden at the Oberau
Bypass Tunnel 804
E. Neun & S. Sabew
Parameter selection for ground improvement works: Lessons from an instrumented site and
numerical analysis 812
G.A. Pittaro, S.H. Goh, C.F. Leung & N. Mace
Jet grouting for tunnelling at London Victoria Station 821
P. Rutty, C. Prangley, I. Heath & F. Mimnagh
Unconfined compressive strength of sand-fines mixtures treated with chemical grouts 829
G. Spagnoli, W. Seidl, E. Romero, M. Arroyo, R. Gómez & J. López

xi
Permeability characteristics of coarse-grained soil conditioned with foam for EPB shield
tunnelling 836
S. Wang, S. Huang, Q. Hu & Z. Liu
Stabilisation of Singapore soft marine clay using a novel sustainable binder for underground
construction 836
H. Yu, Y. Yi, R. Liu & N. Jiang
Experimental study of pore water pressure development in soil when foam infiltrates into
saturated sand 848
D. Zheng, A. Bezuijen & M. Thewes

Author index 855

xii
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

Preface to the current edition

Technical Committee 204: “Underground Construction in Soft Ground” of the International

Society for Soil Mechanics and Geotechnical Engineering (ISSMGE) was first established as
TC28 in 1989 to provide a forum for interchange of ideas and discussion amongst representatives
from different countries with an active interest in tunnelling and deep excavations in the urban
environment. In 1994, under the Chairmanship of Prof. Keiichi Fujita, TC28 organised its first
symposium as a satellite event to the ISSMGE International Conference in New Delhi. Since
then, the host society of TC28 was handed over from the Japanese Geotechnical Society to the
British Geotechnical Society (under the Chairmanship of Prof. Robert Mair), then to the French
Geotechnical Society (under the Chairmanship of Prof. Richard Kastner) and then to the Nether­
lands Geotechnical Society (Prof. Adam Bezuijen is the current Chairman of TC204). Over the
years TC28 (and TC204) has always kept its commitment towards publishing cutting edge
research and practice concerning geotechnical aspects of the design, construction and analysis of
deep excavations, tunnels and large underground structures in the urban environment. Particular
emphasis has traditionally been placed on the development, effects and control of ground move­
ments, their interaction with existing structures, mitigation measures and risk management. The
success of the New Delhi symposium in 1994 led to the organisation of eight more International
Symposia on “Geotechnical Aspects of Underground Construction in Soft Ground” since then.
These were held every three years in London (1996), Tokyo (1999), Toulouse (2002), Amsterdam
(2005), Shanghai (2008), Rome (2011), Seoul (2014) and Sao Paulo (2017).

The 10th symposium (IS-Cambridge 2020) was planned to be held in June 2020 in Cambridge,
United Kingdom. The call for papers drew an overwhelming response and 278 abstracts were
received, resulting in 112 technical papers accepted for publication in the proceedings from
over 25 countries. However, the COVID-19 outbreak in January 2020 and the very rapid
spread of the virus to all corners of the globe, with all the associated health risks, eventually
prompted the symposium organisers to postpone the conference to June 2022. The organisers
also decided to publish two versions of the proceedings. A first version (Version 1), this
volume, which contains only the accepted technical papers (112 in total), is published almost
1.5 years prior to the actual symposium to ensure that the hard and valuable work of the
authors is duly recognised. In 2022, just before the actual symposium, a second version of the
proceedings (Version 2) will be published. This new version will contain all the components
traditionally found in TC204 symposium proceedings including: all papers in Version 1 plus
the written versions of (i) the Fujita Lecture, (ii) the Special Lectures, (iii) the Session Reports,
and (iv) the Bright Spark Lecture. Version 1 will go out of print as soon as the complete Ver­
sion 2 is published.

The themes for the two versions of the proceedings are in line with the terms of reference of
Technical Committee TC204 and include:

• Field case studies

• Sensing technologies and monitoring for underground construction in soft ground
• Physical and numerical modelling of tunnels and deep excavations in soft ground
• Seismic response of underground infrastructure in soft ground
• Design and application of ground improvement for underground construction
• Ground movements, interaction with existing structures and mitigation measures

xiii
The success of publishing this version of the proceedings must be attributed to the authors of
the papers who shared their valuable work, the reviewers who diligently worked through the
papers providing useful technical advice and to the teams at Cambridge, which conducted the
editorial review of the papers and helped manage the complex logistics through very uncertain
times. The work presented here would not have been possible without the efficiency and tre­
mendous support of Mrs. Tian Wu, Mrs. Anama Lowday and Dr Njemile Faustin who sup­
ported the conference organisers every step of the way. The strong support from the staff at
Taylor and Francis, and from the Chairman and Secretary of TC204 (Prof. Adam Bezuijen
and Prof. David Chapman) is also acknowledged.

Mohammed Elshafie, Giulia Viggiani & Robert Mair (Editors)

xiv
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

Organisation

The Symposium of Technical Committee 204 on ‘Geotechnical Aspects of Underground Con­

struction in Soft Ground’ was organised by the Geotechnical and Environmental Research
Group at University of Cambridge under the Auspices of the International Society for Soil
Mechanics and Geotechnical Engineering.

LOCAL ORGANISING COMMITTEE

Professor Lord Robert Mair (Chair)
Professor Giulia Viggiani
Professor Gopal Madabhushi
Professor Abir Al-Tabbaa
Dr Stuart Haigh
Dr Mohammed Elshafie (Secretary)
Dr Giovanna Biscontin

TECHNICAL COMMITTEE 204

Prof. Adam Bezuijen, Chair
Prof. Chungsik Yoo, Vice Chair
Prof. David Chapman, Secretary

Weidong Wang Jochen Fillibeck Nagyoung Kim

Hirokazu Akagi Lou Areias Altaf Usmani
Mitsutaka Sugimoto Fabrice Emeriault Jamie Standing
Kiwamu Tsuno Jacques Robert Mandy Korff
Ilkka Vähäaho Irawan Firmansyah Wout Broere
Chang-Yu Ou Mete Erdemgil Alessandra Sciotti
Yung-Show Fang Mustafa Nalcakan Marco Barla
Michael Mooney Mustafa Koc Daniela Boldini
Richard Pang Gang Zheng Ilias Michalis
John Endicott David Masin Yuepeng Dong
Hervé LeBissonnais Petros Fortsakis Giovanni Spagnoli
Rui Carrilho Gomes Stavroula Schina Dongmei Zhang
Markus Thewes Atul Nanda Sergio Sánchez Rodríguez
Achim Hettler Ashish Juneja Rafael Jiménez
César Sagaseta Alejo Oscar Sfriso Jay Lee
Salvador Senent Hugo Acosta-Martinez Stephan Jefferis
Anna Siemińska- Marlísio Oliveira Jr Long Phung
Lewandowska Jorge Gabriel Laiun Huy Hung Tran
Monika Mitew-Czajewska Christopher Menkiti Tuan Nghia Do
D.E. Razvodovsky Graham Taylor Shuying Wang
Tony Ho Carlos Lam

xv
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

Sponsors

Conference Sponsor

Session Sponsor

xvii
Field case studies and sensing technologies
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

Ovalisation of cast-iron tunnels in response to nearby tunneling work

M. Alhaddad
Transport for London, London, UK

K. Soga
Department of Civil Engineering, University of California, Berkeley, USA

ABSTRACT: The behaviour of existing cast-iron (CI) tunnels subjected to ground movements induced by
new tunnelling works is a much discussed topic. In many cases new tunnels and shafts, particularly in highly
populated urban areas, will need to be constructed adjacent to existing tunnels. This inevitably results in
ground deformations that are transmitted to these structures. This paper investigates the response and the
radial tolerance of CI lining to these imposed movements. This paper highlights the common ‘damage assess­
ment’ procedures in practice and emphasizes the need of paying equal attention to historical case studies
when carrying out often complex analyses. A recent case study from the construction of Crossrail Line
(Queen Elizabeth Line) in London has been presented to demonstrate this. In this example the CI lining is
subjected to significant deformations caused by nearby tunnelling and excavation works.

1 INSTRUCTION calculate the magnitude of these imposed deformations

so they can assess whether the lining can accommodate
1.1 Cast-iron (CI) tunnels subjected to them ‘safely’. Safely here means a risk based approach
deformations to assess the likelihood of damage and the conse­
quences of deformations, or failures at extreme cases.
For near two centuries CI tunnels have been built for
a variety of reasons including for pedestrian usage,
transports, mail delivery and to carry utilities. Most 1.2 Damage assessment
of today’s London Underground tunnels are made of
The response of existing CI tunnels to tunnelling­
cast-iron.
induced deformations is a complex issue and prac­
A typical CI transport tunnel consists of a track­
ticing engineers are faced with significant uncertainties
bed that lies on the tunnel invert supporting the
about assessing their tolerances. The assessment is
tracks. The lining is made of several segments, typic­
often numerically complex and therefore empirical
ally of the same size, in addition to a segment that is
methods are alternatively implemented to (i) predict
of a smaller size called key-segment. Each segment
these ground deformations and also to (ii) examine the
is made of a panel and four flanges, which are bolted
tolerance of the affected existing tunnels. Depending
to the adjacent segments, forming the rings trans­
on the outcomes, further complex assessments may
versely and the body of the tunnel longitudinally. To
deem to be necessary.
note is that CI tunnels are not always uniform
(i) The ground movement caused by the tunnel-
throughout their body and the diameter of the tunnel
ling work is a fairly understood topic. It has been
could change, when approaching the platforms for
shown in numerous case studies as well as experi­
example. The area of separation is formed by
mental works (often using centrifuge tests) that soil
a concrete block called the head-wall. There are also
behaves in a Gaussian looking shape transversely to
other features that disturb the uniformity, such as
the tunnelling direction, and in a bow-wave looking
ventilation shafts and cross passages.
shape in the longitudinal direction. The shape and
When external forces are imposed on these tunnels,
the extent of the movements are pre-determined
they get accommodated partly by the segments moving
based on given parameters that have been arrived at
flexibly against each-other and partly by the deform­
empirically over the years. These predictions are par­
ation of their structural elements: namely, the segments
ticularly more accurate for settlement measurements,
and the bolts. This paper focuses on the radial deform­
especially for cities such as London, thanks to the
ation of the body of the tunnel. Designers carry out
abundance of case studies over the decades. How­
‘damage assessments’ to predict the nature and to
ever, there are more uncertainties with the prediction

DOI: 10.1201/9780429321559-1

3
of horizontal movements and even more so at the
longitudinal direction ahead of the face of the tunnel-
ling. Alhaddad et al 2017 and 2020 describe these in
more details. The focus of this paper is not the accur­
acy of predicting these movements but rather the
response of CI tunnels to accommodate them at
radial directions.
(ii) It is often conservatively assumed that CI tun­
nels deform freely to follow these calculated ground
movements from stage (i). In many cases, this has
been shown to be a reasonable assumption along the
longitudinal axis, where the cast iron rings are more
flexible to slide and bend against each other to accom­
modate those movements (Alhaddad 2016). However,
transversely, this is subject to debate for mainly two
reasons: 1- the calculated horizontal movements
derived from the empirical methods are not as reliable
as the settlement predictions and 2- the CI tunnel rings
are less flexible to accommodate movements radially
than they are longitudinally. Hence, the importance of
case studies to provide evidence based ‘assessments’.
Underground standards (London Underground 2007
and 2009). Li et al., (2015a) shows an example of this
envelop (Figure 1) for a typical geometry of LU North­
ern Line CI tunnels (Diameter = 3.8 m and Depth = 19
mbgl at Euston Station area). When the lining is
defected (visually inspected), reduction factors are
applied.
2.2 Determine existing hoop forces in the lining
Hoop forces are determined using the ‘Elastic Con­
tinuum’ method after Duddeck & Erdmann (1985)
where the maximum and minimum hoop thrusts are
calculated using Equation 1 and Equation 2, respect­
ively. Existing hoop forces are calculated based on
given input parameters. The imposed hoop forces
are assumed to be zero unless it is thought that sig­
nificant loading changes occur.

1.3 Case studies

There are a limited number of cases in which CI tun­
nels have been influenced by large enough deform­
ations to examine the extent of the longitudinal and
radial deformations imposed on them. Crossrail (2007)
lists the case studies recorded in the literature and
Alhaddad (2016) presents several new case studies
from recent years. Rarely damages are reported, and Where σv and σh = vertical and horizontal stres­
when they are, it is not clear if they have been caused ses of the ground; R = Radius of the tunnel; ν = Pois­
by the longitudinal deformations or the ovalisation of son ratio of the tunnel; E = Young’s modulus of the
the linings. This paper selects one of the new case tunnel lining; Ec = Stiffness of the ground; A = Sec­
studies where the authors have detailed information to tion area; J = Second moment of area; K = Earth
derive reasonable conclusions. The case study is Pressure Coefficient; and Ir = Reduced second
a disused Royal Mail Tunnel (RMT) that is built from moment of area of the tunnel lining in accordance
cast iron lining. The tunnel was influenced by nearby with LU standards and is equal to J(4/n)2 where n is
Sprayed Concrete Lining (SCL) excavations (tunnel­ the number of segments.
ling and shafts) at close proximities. It is the radial dis­
tortion of the CI lining that is the topic of this paper.

2 DAMAGE ASSESSMENT OF RADIAL

DISTORTIONS

Radial movements are usually assessed in the form

of imposed ovalisation on the lining. The ‘damage
assessment’ process is described briefly in the fol­
lowing five steps. These steps have been adopted for
the assessment of CI tunnels influenced by the con­
struction of the new Crossrail Line, also known as
Elizabeth Line (Crossrail 2010 and 2011) and are
generally used in similar recent assessments.

2.1 Determine permissible tunnel lining capacity

envelope
This is the moment-thrust permissible envelop of
tunnel lining and is measured by the given parameters Figure 1. Tunnel lining permissible capacity envelop (Li et
of CI lining, such as those provided by London el., 2015a).

4
2.3 Determine existing bending moment in the
lining
The bending moment within the lining is calculated
using Equation 3 after Morgan (1961). The existing
bending moment is measured by assuming an exist­
ing ovalisation. For LU tunnels this is specified to
be 0.6% of the tunnel diameter (the applicability of
this assumption is debated in this paper). For
example, a 3.56 m diameter tunnel will have
a maximum radial deflection of 10.69 mm. This
deflection is assumed to be either constant (Figure 2
right) or variable (Figure 2 left). For the variable
case, the assessment considers both scenarios:
squatting (horizontal egg shape) and elongation
(vertical egg shape).
Where δ = Deflection (actual radius - nominal
radius)
2.4 Determine increase in bending moment from
lining distortion due to the new construction
The total bending moment due to the construction
work is measured by adding the existing deflection
value to the maximum imposed value. In tunnelling
related cases this value is calculated from assuming
that the lining follows the imposed ground move­
ments freely and equates to the differential soil
movements radially along the extrados of the
lining.
2.5 Check the axial force and bending moment
remain within the permissible capacity
envelope
The envelope is determined at stage 1 and if the
imposed stresses calculated at following stages are
outside the envelope further analyses (e.g. FE mod­
elling) and/or assessments will be needed. Mitigation
measures are also recommended accordingly.
An important input into specifying the tolerable
ovalisation on the lining is the existing circularity of
the lining and the assumptions that are made about its
nature.
Figure 2. Deflection Shape of the Lining; squat (left) and
uniform (right).
5
2.6 Circularity of the lining
Wright (2009) states that typically CI tunnels have
a squatting of 0.5% to 1%. When the existing circular­
ity of the tunnel is unknown, the LU standard requires
the use of a value of 0.6%. If the assessment for 0.6%
ovalisation fails, it is recommended to repeat the pro­
cedure and to adopt a more realistic approach taking
into account that some of the existing ovalisation has
been caused by the as-built imperfection of the lining
during its initial construction. Assuming a correct dir­
ection of this imperfection (squat or elongation) is
fundamental.
The TLL Tube Line Report (2005) has carried out
a comprehensive and an extensive laser scanning and
FE modelling analysis of 33% of running tunnels. 93%
of these experience around 1% squatting, which is
believed to be caused by the imperfection during
installation.
When looking at a number of CI tunnels that are
built adjacent and one after the other, the data show
that one of these tunnels shows more squatting than
the other (Wright, 2009; 2010). This suggests that at
least this differential squat between the two tunnels
(assuming the one that was squatted more was installed
first) might have been caused by the changes in the
stresses within the lining rather than by the as-built
imperfections. This magnitude should not be
disregarded.
Almost all of the linings scanned by TLL had
experienced squat (Wright, 2010) while FE model-
ling and soil parameters at typical LU tunnel depths
should result in elongation of the tunnels (earth pres­
sure coefficient ‘K’ values are thought to be 1-1.25).
This indicates that changes in horizontal stresses
during the excavation and/or the drainage of soil in
the long-term caused by the higher permeability of
the tunnel has lowered the surrounding pore water
pressure and resulted in the coefficient earth pressure
‘K’ values to be lower than 1. If we accept this
hypothesis, then it is more likely that the recorded
non-circularities are largely due to the external stres­
ses and less likely to be caused by the as-built imper­
fection. It will be shown in the presented case study
that tunnel linings are more susceptible to imposed
squatting than they are to elongation (cracks hap­
pened where rings experienced squatting).
This indicates that current damage assessment
procedures might need revising, especially where the
CI lining is subjected to additional squatting. Aas­
suming that the non-circularity is a result of as-built
imperfection is not completely justified. On the other
hand, assuming that the ring is a rigid ring (con­
tinuum theorem) is not perfectly representative too.
Li (2014) carried out a 3D FE modelling paramet­
ric study and suggests that bending moment is pri­
marily sustained by the contact of the circumferential
flanges with little influence from the joints while
hoop thrust is mainly transmitted through the joints.
This should be taken into account and a ‘continuum
theorem’ that assumes a rigid ring is therefore a less
realistic approach. When modelling a flexible ring,
Li (2014) observes that the ovalisation is first caused
by the pressure of the ring’s self-weight and con­
tinues to change by soil loading to a value as high as
1% in the absence of imperfections. A value which is
close to what has been witnessed in the majority of
circularity surveys of tunnels. At complex scenarios
such as cross-passage openings, this modelling
becomes especially less relevant (Li et al., 2015b).
It is noteworthy to mention that the scanned tun­
nels have not had concerning signs of damages and
therefore have not necessarily exceeded their stress
state envelopes. In other words, assuming an elastic
continuous ring is not a realistic approach, although
it is surely a conservative one.
However, this becomes less conservative when
the existing ovalisation is treated to be an as-built
imperfection. More importantly, the direction of ova­
lisation (squat or elongation) becomes more relevant
to making a more evidence based decision, as will
be demonstrated in the case study below. Deploying
mitigation measures, such as monitoring, become
a pragmatic tool to manage the uncertainties and the
risks associated with such assessments.
2.7 Monitoring ovalisation
Monitoring helps with managing the risks during the
construction work but also provides an insight that
can be used for future assessments. Conventionally
systems such as ‘Bassett Convergence System’,
prisms read by automatic total stations and chain of
tilt meter sensors have been used to monitor the ova­
lisation in real-time and more recently new technolo­
gies such as shape arrays and digital image
correlation (DIC) techniques have been deployed
successfully to measure these deformations. Alhad­
dad et al. (2020) describes deployment of a high pre­
cision DIC system within tunnel environments.
Figure 3. RMT’s Rings Condition and Geometry.
of on-going water ingress could be seen. This was
confirmed by the many visits of the author to the
tunnel. A typical schematic of the missing bolts and
tunnel cross-section is shown in Figure 3. Note that
the location of the key-segment varies and some­
times staggers between the adjacent rings.
The invert is made of an in-situ cast concrete
block. The quality and consistency of the concrete is
not known and historic drawings and records show
that there is an underlying layer of gravelly material
beneath the concrete surface.
3.2 Description of Crossrail work
The Crossrail tunnels were constructed using SCL
techniques and excavated in two stages: a pilot
tunnel of approximately 6 m diameter followed by an
enlargement of the diameter to approximately 11 m.
The invert of the RMT is at an elevation of
between +92 to +86 m ATD, falling from west to
east over the extent of the interface with the Cross-
rail works. The RMT is located within London Clay,
while the Crossrail works extend into the top of the
Lambeth Group. The interface between the London
Clay and Lambeth Group is at around +76 m ATD.
The new tunnelling work crossed RMT at four
locations (Figure 4). They are described below
(based on a chainage reference at RMT):

3 RMT AT LIVERPOOL STREET STATION

(RMT) – CROSSRAIL WORK

3.1 Description of the tunnel

The Royal Mail Tunnel (RMT) extends beneath cen­
tral London. It is formed from bolted CI segments
and was constructed between 1914 and 1917. Since
2004, the RMT has not been in use unless on occa­
sions for the purpose of running battery-operated
locomotives to support inspection or maintenance
visits and more recently as a tourist attraction. It has
an internal diameter of 2.74 m (Figure 3). The lining
is built of 0.5 m wide rings bolted to each other at
their flanges.
Before the Crossrail work started, an inspection
carried out by the contractor confirms that the tunnel
was in a good condition and no obvious cracks or
deformations were reported, although many bolts Figure 4. RMT Layout and its Proximity to Crossrail
were recorded missing and in many locations signs Work.

6
1) Chainage 223 to 243: Construction of the SCL
westbound platform tunnel (PTW) with 6m
clearance.
2) Chainage 263 to 288: Construction of the SCL
central concourse passageway with 5m
clearance.
3) Chainage 298 to 313: Construction of SCL
Finsbury Circus access shaft excavated around
2 m to the south of the RMT. The shaft was
approximately 40 m deep; varying between
12.5 m and 15.7 m in diameter.
4) Chainage 343 to 485: Construction of the SCL
eastbound platform tunnel. To the east of chai­
nage 343 the clearance between RMT and the
enlarged Crossrail platform tunnel is approxi­
mately 2 m decreasing to less than 0.5 m Figure 5. Propping Installed over 15 m of RMT.
towards the end. From around chainage 460 to
480, the RMT gradually enlarges towards a step After construction had finished, an inspection by
plate junction. The new tunnel was constructed the author and Crossrail’s Asset Protection team as
parallel and directly beneath these chainages for well as the limited monitoring data in place con­
more than 100 m. The photograph in Figure 4 firmed that the props did not get engaged (they did
shows the enlargement directly beneath the not take any load meaning elongation of all of the
RMT at these chainages. rings were less than 10 mm or 0.4% ovalisation).
Other relevant sections to note are as follows:
3.4 Monitoring layout
1) RMT Enlargement Chamber: This chamber,
bordered by chainages 243 and 263, consists of Figure 6 shows the conventional monitoring that was
13 rings of 3.7 m diameter with an opening at carried out inside RMT. This included the use of auto­
the northern side to an unused shaft. The con­ mated total stations (ATS) monitoring 3- and 5-point
nection between the larger rings and the smal­ radial prism arrays at generally 20 m centres. In add­
ler diameter rings is a concrete head wall ition, levelling points (LP) were installed in the invert of
which is typical for such transitions. the tunnel (track-bed) and monitored manually on
2) Blomfield Box: From chainage 440 to 475, a weekly basis. Looking west the arrangement of the
a 36m deep box-retained cut excavation was prisms were; P1 at Crown, P2 and P4 at right-axis and
formed around 5m to the south of the RMT. right-invert and P3 and P5 at left-axis and left-invert,
3) Propping: In accordance with the Damage respectively. LP at invert level was monitored manually.
Assessment for the POT, propping was Visual inspection of the RMT was also carried out
installed to mitigate radial deformation; along generally on a weekly basis to check for any signs of
a 15 m section between chainage 375 to 390. distress or change in tunnel condition.
Over a short section of the RMT above the con­
struction of eastbound platform tunnel, further moni­
3.3 Damage assessment and mitigation measures toring was installed by Cambridge Centre for Smart

The damage assessment of radial and longitudinal

movements was carried out as explained in Section
2, which concluded that some cracks might occur at
the track-bed (invert) of the tunnel due to the exces­
sive imposed longitudinal curvature (not radial).
Devriendt & Alhaddad (2015) explain the damage
assessment and general observed movements.
Given the level of uncertainty relating to the ground
movement and structural impact assessment calcula­
tions, to mitigate radial impacts, a short 15 m section
of the tunnel was propped at axis level (Figure 5). This
took place where the tunnel was expected to first
experience elongation and the propping was installed
such that the loads would only be taken up if radial
convergence at axis level exceeded 10 mm. Additional
props were kept on standby in case they needed to be
Figure 6. Examples of Prism (P), Manual Monitoring (LP)
installed elsewhere (subjected to future observations
and Automatic Total Station (ATS) Arrangement inside
and assessments). RMT.

7
Infrastructure and Construction (CSIC). This work is
described in Alhaddad et al. (2014). The results of this
monitoring are not relevant to this study except that
no damage or cracks were seen in this section of the
tunnel. The lining there was mainly subjected to
elongation (0.4%) and longitudinal curvatures.
3.5 Construction progress and monitoring results
Figure 7 shows sketches of the construction progress
at each of these stages. The significant movements to
the RMT occurred at the following times:
1) Stage 1: Excavation of the Finsbury Circus
Shaft (FCS) – Completed before June 2012.
2) Stage 2: Excavation of the Central concourse
pilot tunnel and enlargement – Completed by
early April 2013.
3) Stage 3: Excavation of the Eastbound and
Westbound pilot tunnels – Completed in
October 2013.
4) Stage 4: Excavation of the Eastbound and
Westbound enlargement and influencing Cross-
passages pilot tunnels – Completed by mid-
March 2014.
Figure 7. Main Construction Stages Influencing the RMT.
8
5) Stage 5: Enlargement of the influencing cross-
passages and the start of Blomfield Box exca­
vation – May 2014.
6) Stage 6: End of the majority of the construction
including Blomfield Box and the arrival of the
Eastbound TBM – February 2015.
The settlements of the crown level and invert level,
the horizontal convergence of the RMT measured from
points 2 and 3 and the vertical convergence measured
from points 1 and LP for each of these phases are
shown in Figure 8. Vertical convergence is measured
from differential values collected from two different sys­
tems (ATS and manual monitoring) with an estimated
measurement error of up to 2 mm. Also, LPs only meas­
ure the settlement and are installed on the track bed and
are assumed to follow the lining at the invert level.
A maximum settlement of approximately 100 mm
was measured for the construction of the westbound
platform tunnel and approximately 90 mm for the
construction of the eastbound platform tunnel. This
Figure 8. Monitoring of Settlement and Convergence for
RMT; (a) Settlement at Invert Level (Manual Monitoring),
(b) Settlement at Crown Level (ATS Monitoring), (c) Con­
vergence of the Tunnel Horizontally Measured Between P2
and P3, (d) Convergence of the Tunnel Vertically Between
P1 and LP.
exceeded the preconstruction assessment calculating
a maximum of 70mm settlement and further assess­
ments had to be carried out during the works. This
included carrying out a circularity survey to measure
the tunnel’s existing circularity (Crossrail 2013).
The horizontal and vertical convergences give an
indication of the magnitude of the ovalisation
imposed on the lining. Concluding from the direction
of the movements, RMT experienced up to 10 mm
of elongation (with no signs of damage) and up to
10mm of squat (where cracks were reported). These
were above the 7 mm value calculated during the ini­
tial damage assessment procedure.

3.6 Summary of cracks

First reported cracks were between chainages 303
and 323 (Figure 9) and were most probably caused
by the excavation of FCS and/or the excavation of
the SCL cross-passage directly beneath the RMT.
The 6 cracks were all on the circumferential flanges
at bolt locations and were seen around January 2013
(eight months after the completion of FCS and days
after the breakage of cross-passage enlargement).
The cracks were all approximately on the same
longitudinal line near the crown and emanate from
the bolt position. The cracked flanges were all on the
weaker segment (every other neighbouring segment
on the next ring that is not cracked is a key segment).
Figure 8 (d) shows that around 10 mm squat (0.4%
ovalisation) was imposed on the tunnel by the exca­
vation of FCS. The deformed shape appears to be an
egg-shape pointing towards the bottom of FCS and
the approaching excavation of the cross-passage.
The second series of the cracks were observed
after stage 4 of the construction and were located to
the west of the RMT enlargement chamber at around
CH250. They can be seen on the CI lining and also
on the invert and the connecting head-wall between
the RMT enlargement chamber and the running

Figure 10. Cracks above PTW Excavation at CH 250.

tunnels (Figure 10). It is unclear when exactly they

Figure 9. Cracks Adjacent to FCS at CH 303.
occurred, but the majority of the imposed movements
were caused by the enlargement of PTW westwards
from the end of January to early February 2014.
The mechanism of deformation and the arrange­
ment of the cracks are very similar to those seen
around FCS. The deformed shape is likewise an egg-
shape pointing towards the approaching excavation
of PTW from the east. Also, here cracks occurred on
the same position adjacent to the crown and on the
weaker segments at approximately 0.5% ovalisation.

9
 When looking at the convergence history of the
closest monitored ring at CH243 (7 rings away from
the nearest crack – where the furthest crack is one
ring away from the headwall and is roughly at CH
250), it may be seen that there is a rapidly increasing
trend of movements when PTW enlargement excava­
tion approaches the tunnel (Figure 11). The values
are maximum on 28th of January 2014 when the tun­
nelling face was recorded to be directly beneath the
RMT. The trends start to decrease as the tunnelling
face crosses RMT and proceeds to the other side.
Data are baselined on 15th of July 2013, a few
days before the chainage was influenced by the PTW
pilot excavation (the construction of the pilot tunnel
then was paused until 1st of October 2013).
It should also be noted that the divergence of axes
1-5 and the convergence of axes 3-4 start to return to
their original shape after a peak corresponding to the
PTW crossing underneath. The trend of returning to
the original shape appears to be taking on a plastic
Figure 11. History of Monitoring of CH 243: Settlement (a),
Horizontal Convergence (b) and Vertical Convergence (c).
10
deformation shape. This trend of movements implies
that the cracks or some of them may have happened
around the peak (the invert cracks had been spotted
on a visit on 7th of February 2014, hence the cracks
must have happened before this date).
4 CONCLUSION
CI tunnels deform longitudinally and/or transversely
to accommodate external movements (such as those
caused by nearby construction activities). The focus
of this paper is the response of CI tunnels to those
movements transversely and along their radial rings.
This paper demonstrates the ‘damage assessment’
stages that are often followed to assess whether CI
tunnels can accommodate those movements radially
and briefly describes the mitigation measures where
the deformation are assessed to be excessive; namely
the use of props inside the tunnel, when possible.
Mitigation measures are not often practical (for
examples props are not possible within running tun­
nels) and it is important to understand the likely
response of the tunnels to be able to evaluate the
risks in advance of the start of any given construc­
tion work. Numerical assessments are complex, and
results of such assessments are not guaranteed to be
correct. Historical case-studies provide an important
insight to enable evidence-based decision makings.
This paper presents a case study where the CI
tunnel was subjected to significant radial movements.
It is argued that where constructed within London
Clay, typical CI linings (uniform tunnels not influenced
by other structures such as existing cross passages) are
likely to be more susceptible to squatting than they are
to elongating deformations. This is because the evi­
dence suggests that CI linings have experienced squat­
ting over their lifetime and as a result they would have
less bending tolerance to accommodate squatting than
elongation.
The cracks in this case study have evidently been
the result of radial deformation when the rings were
squatting with about 0.4-0.5% ovalisation. The rings
had elongated with similar ovalisation values with­
out any sign of damage. This reinforces the argu­
ment that CI rings experience squat over their
lifetime. In this case the alignment of the segments
and the staggering of the keys appear to also contrib­
ute to the radial tolerance of CI rings (by perhaps
concentrating the loads over the weak axes).
Designers must be more cautious and should not
downgrade the existing circularities as ‘as built
imperfections’ especially when the new constructions
are likely to impose more squatting on the tunnel.
This is, for example, when excavating shafts nearby
or when carrying out tunnel excavations with
a parallel and an offset axis from the existing linings.
On the other hand, accepting that there is the risk
of cracks happening on the radial flanges will open
the conversation between the stakeholders early on
and will provide a more pragmatic approach to
mitigating the risks. The cracks on their own are not
likely to impose a critical risk to the operation of the
rail-tunnels and at extreme scenarios they can be
fixed/replaced after the construction work is com­
plete. Hence providing more extensive and better tar­
geted (locally targeted) monitoring solutions for
such cases could become a more favourable
approach.
ACKNOWLEDGEMENT
This work was carried out in CSIC and would not
have been possible without the support of Arup,
CH2M Hill (now Jacobs), London Underground and
Royal Mail Group and thanks to Engineering and
Physical Sciences Research Council (EPSRC) for
their financial support. The authors would like to
namely acknowledge Robert Mair, Mohammd Elsha­
fie, Michael Devriendt, Frances McDonnell, Mathew
Wilcock and Chang Ye Gue for their contribution
throughout this work.
REFERENCES
Alhaddad, M. 2016. Photogrammetric monitoring of cast-
iron tunnels and applicability of empirical methods for
damage assessment. PhD Thesis, Department of Engin­
eering, University of Cambridge, Cambridge, UK.
Alhaddad, M. et al., 2014. Multi-Suite Monitoring of an
Existing Cast Iron Tunnel Subjected to Tunnelling­
induced Ground Movements. Tunneling and Under­
ground Construction - Geo-Shanghai. American Society
of Civil Engineers: 293–307.
Alhaddad, M. et al., 2017. Imposed Longitudinal Settle­
ment on a Cast-iron Tunnel from the Excavation of
a New Tunnel Beneath. 9th International Symposium on
Geotechnical Aspects of Underground Construction in
Soft Ground, IS – Sao Paulo: 343–353.
11
Alhaddad, M. et al., 2020. Cast-Iron Tunnels’ Tolerance to
Imposed Longitudinal Settlement Curvature Geotechni­
que Symposium in Print: Linear infrastructure.
Crossrail, 2007. London Underground Interface, Case His­
tories, Prepared by Geotechnical Consulting Group for
Crossrail Project, 1D0101-G0G00-01022A.
Crossrail, 2010. Civil Engineering Design Standard – Part
7, Ground Movement Prediction, Version 6.0, CR-STD
-303-7 (CEDS 7).
Crossrail, 2011. Assessment of ground movement effects
on the Post Office tunnel at Liverpool Street Station
(PO/05), Rev 3.0, C122-OVE-C2-RAN-CR101-00016.
Crossrail, 2013. RESULTS FROM LASER SCANNING
THE POST OFFICE TUNNEL AT LIVERPOOL
STREET STATION (PO/05), Rev 2.0, C122-OVE-C2-RG
-50031.
Devriendt, M. & Alhaddad, M., 2015. Construction impacts
of Crossrail Liverpool Street Station on the Royal Mail
Tunnel. XVI ECSMGE Geotechnical engineering for
infrastructure and development 1: 831–836.
Duddeck, H. & Erdmann, J., 1985. Structural design models
for tunnels. International Journal of Rock Mechanics and
Mining Sciences & Geomechanics 20(1): 83–91.
Li, Z., 2014. Long-term Behaviour of Cast-iron Tunnel
Cross Passage in London Clay. PhD thesis, University
of Cambridge, Department of Engineering.
Li, Z., Soga, K. & Wright, P., 2015a. Behaviour of cast-iron
bolted tunnels and their modelling. TUNNELLING AND
UNDERGROUND SPACE TECHNOLOGY 50: 250–269.
Li, Z., Soga, K. & Wright, P., 2015b. Long-term performance
of cast-iron tunnel cross passage in London clay. Tunnel-
ling and Underground Space Technology 50: 152–170.
London Underground, 2007. Manual of Good Practice,
Civil Engineering – Deep Tube Tunnels and Shafts,
G-055, Version A1, (LU-G055).
London Underground, 2009. Standard 1-50 Civil Engineer­
ing Common Requirements, Version A1, (LU-1-50)
Morgan, H.D., 1961. A contribution to the analyses of
stress in a Circular Tunnel. Geotechnique 11(1): 37–46.
Wright, P., 2009. Assessment of London Underground tube
tunnels investigations, monitoring and analysis. Smart
Structures and Systems 6(3): 239–262.
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

Lubrication characteristics of pipejacking in soft alluvial deposits

Wen-Chieh Cheng & Ge Li
Xi’an University of Architecture and Technology, Xi’an, China

Dominic E.L. Ong

Griffith University, Queensland, Australia

ABSTRACT: The impacts of the lubrication upon the pipejacking works are significant and necessary
to be considered in tunnelling design. This study described a method that can be used to evaluate the
lubrication performance by the reduction in the frictional coefficient μ as a function of injection type,
soil and lubrication natures, and misalignment. The results of an application of the proposed method to
a pipejacking project in soft alluvial deposits were presented. The effect of misalignment increased the
frictional stress τld to 12.5 kPa, reducing the reduction in the μ value to 71%. The combined effects of
misalignment and varying face resistance contributed to the τld value of 4.0 kPa and were deemed as
the main cause leading the low reduction of 84%. The misleading reduction of 60% was attributed to
the inability of the occasional gravel to develop lower face resistance. The effects had greater influence
on the lubrication performance than the injection type.

1 INTRODUCTION frictional stress was reduced from 45 to 90%. How­

The pipejacking technologies are specially favour­
able for urban pipeline systems construction because
of their low cost and short construction period.
Inappropriate thrust force and/or inadequate jacking
capacity is often seen while pipejacking, causing
damages to jacked pipe string and adjacent proper­
ties. This would hold true especially for long-
distance pipejacking in coarse soils. Effective lubri­
cation enables long-distance pipejacking to be con­
ducted. The friction resistance usually constitutes the
major component of jacking loads, which can be
approximated from the minimum bound to the total
jacking load and exhibits a cumulative nature while
pipejacking in coarse soils. There have been many
factors affecting the friction resistance and amongst
the factors, lubricant plays a vital role in achieving
effective lubrication. Generally, lubricant should be
maintained within overcut annulus, and the percent­
age reduction of frictional stress appears to be
closely linked to the volume of injected lubricant.
Cui et al. (2015) reported that due to loss of lubrica­
tion fluid into surrounding fissures, the effectiveness
of slippery film was decreased, leading that the
thrust force of Line 2 of a pipe ramming project con­
veying water from southern to northern Jiangsu was
much higher than the other three lines (Lines 1 and
3-4). Pellet-Beaucour & Kastner (2002) indicated
that for volumes varying from 25 to 170 litre/m,
ever, the distribution characteristics of injected lubri­
cant and their effects on the reduction of the
frictional resistance has rarely studied by literature.
The objectives of this study are (1) to present the
distribution characteristics of injected lubricant with
reference to the presented case history of pipejack­
ing, (2) to evaluate the lubrication performance of
pipejacking in soft alluvial deposits using the base­
line technique and (3) to investigate the impacts of
the distribution characteristics of injected lubricant,
soil and lubricant natures and pipe deviation on the
lubrication performance.
2 METHODOLOGY
2.1 Lubricant distribution characteristics
The injection of lubricant can be categorised into three
types which are single-point injection, two-point injec­
tion and multi-point injection. The injection into the
overcut annulus through a single injector is termed
single-point injection. The injection into the overcut
annulus by two injectors at two different distances
from the face is termed two-point injection, as depicted
in Figure 1. Injector 2 at a distance more close to the
face would work with Injector 1 as the friction during
pipe ramming gets higher than a single injector can
manage. While jacking RCP 14, the overcut annulus at

DOI: 10.1201/9780429321559-2

12

Figure 1. Distribution characteristics of injected lubricant
for two-point injection.
4 m and 11 m distances from the face is saturated
through Injectors 2 and 1 by injecting 0.13 m3
(20.9 %) and 0.50 m3 (79.1 %) lubricant, respectively.
Figure 1 also provides the details about the ramming of
RCPs 15 and 16. Since effective lubrication cannot be
easily sustained for long-distance pipe ramming, multi-
point injection may be used to prevent inadequate
lubrication from occurring.
2.2 Lubrication performance
Lubrication can largely affect pipejacking works,
which can be assessed through the reduction of the
frictional coefficient μ. Pellet-Beaucour & Kastner
(2002) indicated that local variations of total jacking
load are generally linked to the varying face resist­
ance. A line constituted by the minima of total jacking
load is thus referred to as baseline of jacking loads.
The jacking forces from both ends of a baseline sec­
tion divided by the section length leads to the average
jacking force favg. The average jacking force favg div­
ided by the circumference of pipe yields the frictional
stress τld. There are many national standards available
for calculating the pressure acting upon the pipe’s
crown (termed normal contact pressure σz hereafter)
but based upon different safety concepts.
Dividing the τld value by the calculated σz value
gives the frictional coefficient μld for lubricated
drives. The percentage reduction can be calculated
by (μnd-μld)/μnd ×100 where the value of μnd is
13
recommended by Stein et al. (1989) for soil-pipe
(unlubricated) interface. The described method that
can be used to evaluate the lubrication performance
can be briefed in short as follows: (1) preparing the
baseline of jacking loads, (2) calculating the fric­
tional stress τld for lubricated drives, (3) calculating
the normal contact pressure σz, (4) evaluating the
frictional coefficient μld for lubricated drives, and (5)
assessing the percentage reduction in the μ value.
3 PROJECT BACKGROUND
3.1 Background
The slurry shield was adopted to perform excavation
of the two pipejacking drives in soft alluvial deposits.
The alignment for all the drives is straight. During
pipe ramming, the bentonite slurry with unit weight
of 10.6 kN/m3 was used to stabilise the excavation
face and transport tunneling soil spoils to decantation
chambers. Since the tunnels were excavated via the
cutter wheel of 1500-mm in diameter, an overcut
annulus of 30 mm was formed by using the smaller
concrete pipe of 1440 mm in diameter. The two
drives characteristics are detailed in Table 1.
3.2 Engineering geology
The stratigraphic profile is established with reference
to five 15-m deep geological boreholes penetrating
through the 6-m thick silty sand layer into the
7.5 m thick poorly-graded to well graded sand and
gravel layer. Based upon the standard penetration
test and triaxial test results, the soil physical and
mechanical properties are listed in Table 2.
4 ANALYSIS AND DISCUSSIONS
4.1 Development of jacking loads baseline
At Drive C, the shield spanned at a depth of
10.8 m and the associated activities are presented in
Figure 2. The baseline of jacking forces consisted of
the first 2-8 m section (gravel), 8-21 m (gravel) and
21-40 m (gravel) sections, and final 40-75 m (clayey
gravel) section was established through the minima of
Table 1. Characteristics of two pipejacking drives.
Cutter Soil Face V.
Length Depth dia. cover resist. inj
Parameters m m m m kPa l/m
Drive C 75 10.8 1.5 10.1 555 552
Drive D 102 10.8 1.5 10.1 555 534
Note: Cutter dia = cutter wheel diameter, Face resist. = face
resistance, V.inj = volume of injected lubricant.
Table 2. Summary of soil physical and mechanical F 1962 (ASTM 2011), and GB 50332 (MOC 2002) are
properties. deemed to be the well-developed models that can be
used to calculate the normal contact pressure σz. The
Thickness SPT-N γ c’ ϕ’ qu models were modified with reference to the Terzaghi
arching model founded upon active trap-door experi­
Layer m kN/m3 kPa ° kPa ment (Terzaghi 1936) where the shear bands arise from
the outside of tunnel cross sections along oblique lines,
Backfill 2.0 17.6 with a horizon included angle equal to 45°+ϕ/2, and
Silty sand 2.8 2-9 18.1 0 28 then they turn to vertical lines after passing the tunnel
Gravel/Sand >4.5 >100 20.1 0 35 6.7e3 crown’s level and finally arrive at the ground surface,
Silty sand >3.0 9-18 19.1 0 30 as illustrated in Figure 4.
Equation 4 in the modified models was derived
Note: γ = unit weight, c’ = cohesion, ϕ’ = friction angle, from the limit equilibrium of a horizontal slide (Ter­
qu = uniaxial compression strength.
zaghi 1943):

Figure 2. Pipejacking activities at Drive C.

where B1=silo width, γ=unit weight of soil, K=soil
pressure ratio, c=soil cohesion, τf=shear strength of
jacking forces, with the average jacking forces favg at soil, and σz=normal contact pressure. Equation 4 is
1.6, 56.5, 0.5, and 12.3 kN/m, respectively. While their a single order ordinary differential equation. Inte­
frictional stresses τld were calculated as being 0.4, grating Equation 4 and considering boundary condi­
12.5, 0.1, and 2.7 kPa, respectively. The pipe ramming tion σz=0=q at the surface deduces the σz value at any
of Drive D traversed at 10.8 m depth and its activities level:
are shown in Figure 3. The pipejacking results deter­
mined the baseline of jacking forces corresponding to
the first 5 m (gravel) section (from 6 to 11 m), the sub­
sequent 11-24 m (clayey gravel) and 24-31 m (clayey
gravel) sections, and the final 31-102 m (clayey
gravel) section, with the average jacking forces favg at In fact, the σz value should be calculated one stra­
2.0, 18.1, 49.0, and 3.5 kN/m, respectively. The favg tum by one stratum from the surface to the bottom.
values divided by the circumference of pipe yielded Equation 5 can thus be rewritten as Equation 6 for
the frictional stresses τld equal to 0.4, 4.0, 10.8, and 0.8
kPa, respectively.

4.2 Evaluation of lubrication performance

JMTA (JMTA 2013), ATV A 161 (German ATV Rules
and Standards 1990), BS EN 1594 (BS 2009), ASTM

Figure 3. Pipejacking activities at Drive D. Figure 4. Terzaghi arching model (after Terzaghi 1943).

14
Table 3. Parameters used in the modified models.

Silo width δ K c’

Parameters m ° kPa

Terzaghi (1943) De×[1+2tanα] ϕ 1 C

JMTA (2013) (De+0.08)×(secα+tanα) ϕ 1 C
ATV A 161 (1990) 1.732De (ϕ=30°) ϕ/2 K0=0.5 None
BS EN 1594 (2009) De×(1+2tanα) ϕ K0 c (with verification)
ASTM F 1962 (2011) 1.5De ϕ/2 tan2(45°-ϕ/2) None
GB 50332 (2002) De×(1+tanα) Katanϕ=0.19 None

Note: δ = friction angle in shear plane, α = 45°-ϕ/2, De = outer pipe diameter, Db = tunnel bore diameter, ϕ = soil friction
angle, c = soil cohesion, K0 = soil pressure ratio at rest.

calculation of the σz value in multistratum

environment.

where i=stratum numbering from 1, 2,.to n,

hi=thickness of ith stratum. Parameters used in the
modified models are summarised in Table 3. Equa­
tions to calculate the normal contact pressure acting
upon pipeline should refer to the national standards.
The modified models assume the fully developed
shearing bands. Zhang et al. (2016) measured the
normal contact pressure acting upon the 17th, 24th
and 26th pipe respectively during the construction of
Gongbei Tunnel to verify their proposed new calcula­
tion model. The stratum properties utilised in the cal­ Figure 5. Comparison of normal contact pressure between
culation of the σz value against the national standards Terzaghi, JMTA, ATV A 161, BS EN 1594, ASTM F 1962,
(Terzaghi, JMTA, ATV A 161, BS EN 1594, ASTM GB 50332 and measured result.
F 1962, GB 50332) are shown in Table 4. Compari­
sons of the normal contact pressure σz on the 17th
pipe between the calculated normal contact pressures
and the measured ones were conducted, as shown in measurements for jacked pipes. Thus, ATV A 161
Figure 5. The soil prism weight would give most safe, and ASTM F 1962 were selected as the preferred
robust but uneconomical pipeline designs, as indi­ models for calculating the normal contact pressure
cated in Figure 5. σz. Dividing the frictional stress τld by the calculated
It was evident that ATV A 161 and ASTM F 1962 normal contact pressure σz led to the frictional coef­
give estimations in good agreement with field ficient μld. The percentage reduction was obtained
using , as listed in Table 5.

Table 4. Soil properties for 17th pipe (after Zhang et al. 4.3 Effect of lubricant injection type
2016). Two drives (Drives C & D) were analysed because
of their good data completeness. The cumulative vol­
Thickness γ c’ ϕ’
umes of injected lubricant for Drives C & D were
retrieved from Figures 2 and 3 and estimated relying
Parameters m kN/m3 kPa ° only upon lubricant injections into overcut annulus
of the analysed baseline section (Tables 6-7). For
Medium to coarse sand
with clay
9.3 17.6 0 30.2 instance, the analysed baseline section for Drive
Silty clay 17.6 18.3 10.6 6.9
C was 2-8 m (Section 1), 8-21 m (Section 2), 21­
40 m (Section 3) and 40-75 m (Section 4). While the

15
Table 5. Summary of reduction in μ value against each Table 7. Summary of cumulative volume at Drive D.
baseline section.
RCP RCP RCP RCP RCP RCP RCP
favg τld σz μld Red. μ 02 09 19 29 38 48 58

kN/m kPa kPa % 4.05

(100)
Drive C 1.6 0.4 111.1 0.004 98 2.90 1.45 0.25
56.5 12.5 111.1 0.1 71 (63) (31.5) (5.5)
0.5 0.1 111.1 0.0009 99 3.34 1.67 0.29
12.3 2.7 102.0 0.03 88 (63) (31.5) (5.5)
14.67 8.79 3.48 3.48 3.48 1.66 1.66
Drive D 2.0 0.4 111.1 0.004 98
(36.2) (21.7) (8.6) (8.6) (8.6) (4.1) (4.1)
18.1 4.0 102.0 0.04 84
49.0 10.8 102.0 0.1 60
3.5 0.8 102.0 0.008 97

Note: Each drive has four baseline sections and values Table 7. (cont’d). Summary of cumulative volume at
shown here are associated to the sections. favg = average Drive D.
jacking force, τld = frictional stress, σz = calculated normal
contact pressure, μld = backanalysed frictional coefficient, RCP RCP RCP
Red. μ = percentage reduction in μ value. 68 78 88

1.66 0.81 0.81

(4.1) (2) (2)
analysed baseline section for Drive D was 6-­
11 m (Section 1), 11-24 m (Section 2), 24­ Note: Number in bracket indicates the percentage volume
of injected lubricant. RCP = reinforced concrete pipe.
31 m (Section 3) and 31-102 m (Section 4). RCP02 is farther from the face than RCP88 and thus has
A 30 mm overexcavation outside the 1.44-m larger percentage volume of injected lubricant.
diameter pipe justified the overcut ratio of 0.02 cor­
responding to the theoretical overcut annulus of
0.138 m3/m. The lubricant injection type for the four saturate the overcut annulus to sustain effective
sections of gravel (first three sections at Drive C and lubrication conditions, but also showed a significant
first section of at Drive D) refers to Tables 6-7. The loss of lubricant while pipe ramming. Their percent­
averaged injection volume can be calculated as age reductions in the μ value from ATV A 161 were
being the cumulative injection volume divided by calculated as being 98% (0.004 vs. 0.35), 71% (0.1
the length of analysed baseline section. Their injec­ vs. 0.35), 99% (0.0009 vs. 0.35), and 98% (0.004 vs.
tion volumes averaged 0.139 m3/m (0.84 m3/6 m), 0.35), respectively. The buoyancy of 17.3 kN greater
0.466 m3/m (6.06 (=4.8+1.26) m3/13 m), 0.619 m3/ than the pipe self-weight of 12.6 kN and the satur­
m (11.77 (=7.06+2.35+1.18+1.18) m3/19 m), and ated overcut made a 71% reduction (the second
0.811 m3/m (4.05 m3/5 m), respectively, and were lowest in this study) from 0.35 (the average μ value
greater than 0.138 m3/m. The excessive injection recommended by Stein et al. (1989) for gravel-pipe
volumes not only indicated a strong intention to interface) to 0.1. The main cause to lead to
the second lowest reduction was attributed to the
excessive pipe deviation (Figure 2). The μld value of
0.1 matched the lower limit recommended by Stein
Table 6. Summary of cumulative volume at Drive C.
et al. (1989) for lubricated drives, suggesting that the
RCP RCP RCP RCP RCP RCP RCP 8-21 m section of gravel was well-lubricated. Also,
02 09 19 29 38 48 58 the overcut full of lubricant and the enough buoy­
ancy made the reductions high enough for the other
0.84 three sections, with the μld values smaller than 0.1,
(100) indicating that lubrication during pipe ramming of
4.8 1.26 the other three sections was very effective.
(79.2) (20.8) Tables 6-7 details the lubricant injection type for
7.06 2.35 1.18 1.18 the other four sections of clayey gravel (final section
(60) (20) (10) (10) at Drive C and final three sections at Drive D). Their
9.08 4.54 2.73 2.73 1.82 0.91 0.91 injection volumes averaging 0.649 m3/m (22.72
(40) (20) (12) (12) (8) (4) (4) (=9.08+4.54+2.73+2.73+1.82+0.91+0.91) m3/35 m),
Note: Number in bracket indicates the percentage volume 0.354 m3/m (4.60 (=2.90+1.45+0.25) m3/13 m),
of injected lubricant. RCP = reinforced concrete pipe. 0.757 m3/m (5.30 (=3.34+1.67+0.29) m3/7 m), and
RCP02 is farther from the face than RCP58 and thus has 0.570 m3/m (40.5 (=14.67+8.79+3.48×3+1.66×3
larger percentage volume of injected lubricant. +0.81×2) m3/71 m), respectively, were also in excess

16
of the theoretical overcut annulus of 0.138 m3/m.
The excessive injection volumes made the percent­
age reductions to reach to 88% (0.03 vs. 0.25), 84%
(0.04 vs. 0.25), 60% (0.1 vs. 0.25), and 97% (0.008
vs. 0.25), respectively. The enough buoyancy and
the lubricant-saturated overcut caused an 84% reduc­
tion from 0.25 suggested by Stein et al. (1989) for
clay-pipe interface to 0.04. The varying face resist­
ance induced by jacking into the clayey gravel and
the excessive pipe deviation were deemed to be the
main cause to lead to the reduction of 84%. The μld
value equal to 0.04 was far less than the lower limit,
which also indicated adequate lubrication. It is worth
to note that jacking into gravel at 31 m distance
caused the reduction of 60% (the lowest in this
study). This phenomenon was most likely because of
the gravel not being long enough to develop lower
face resistance. The use of the excessive injection
volumes accompanied with the justified buoyancy
made the percentage reduction far less than 0.1 for
the other two sections, which also indicated adequate
lubrication. To short, the effect of misalignment and
the varying face resistance contributed to the low
percentage reductions despite the excessive injection
volumes. The occasional gravel led to the misleading
reduction. The effects on the lubrication perform­
ance outweighed the effect of injection type.
4.4 Effect of soil and lubricant natures
The permeation of lubricant into the surrounding
geology would mitigate the effort of establishing
a lubricating layer at soil-pipe interface. This is most
likely because of the inability of the lubricant to
develop a filter cake of low permeability. Such
a permeable overcut could also result in injection
volume in excess of the theoretical overcut. The phe­
nomena discovered in this study resulted in the rela­
tively large injection volume of 0.552 m3/m at Drive
C (including mostly the gravel) than 0.534 m3/m at
Drive D (including mostly the clayey gravel). On the
other hand, the continuous injection of the lubricant
with Marsh cone viscosity of 38 mins into the over-
cut annulus effectively reduced the friction resist­
ance to viscous resistance. This led the reductions in
excess of 88%, which is consistent with Staheli et al.
(2006).
4.5 Effect of misalignment
The effect of misalignment significantly increased
the friction resistance and had implications on the
lubrication performance. There was an increase of
the jacking force of 771 kN while spanning between
8 and 21 m distance at Drive C (Figure 2). This
increase in the jacking force caused the percentage
reduction in the μ value to reduce by 27% from
98%. The main cause was not because of the overcut
not full of lubricant, but because of the exaggerated
frictional stress τld of 12.5 kPa induced by the pipe
deviation in excess of 60 mm, which is almost 30
17
times larger than 0.4 kPa of the 2-8 m section in the
same gravel. While traversing through the 11­
24 m section of gravel at Drive D, the jacking force
increased by 637 kN to 1617 kN (Figure 3). The
increase of 637 kN reduced the percentage reduction
to 84% from 98%, most likely because of the exag­
gerated frictional stress τld of 4.0 kPa, which is 5
times larger than 0.8 kPa of the 31-102 m section in
the same clayey gravel. The main cause to lead to
the reduction of 84% was attributed to the combined
effects of misalignment and varying face resistance,
resulting from tunnelling into the clayey gravel.
5 CONCLUSIONS
The method that can be utilised for assessing the lubri­
cation performance using the percentage reduction in
the frictional coefficient was described. The effect of
lubricant injection mode, soil and lubricant natures and
misalignment on the lubrication performance was
investigated. Some main conclusions can be drawn as
follows:
(1) The excessive volumes of injected lubricant
made the overcut full of lubricant and led to
enough buoyancy. Despite the excessive vol­
umes, the significant pipe deviation and/or the
varying face resistance had implications on the
percentage reduction in the frictional coefficient
μ. The occasional gravel could lead to the mis­
leading reduction. The said effects on the lubri­
cation performance outweighed the effect of
injection type.
(2) The excessive volumes of injected lubricant were
either attributed to permeable ground or to the
inability of the injected lubricant to develop
a filter cake of low permeability. The highly vis­
cous lubrication with Marsh cone viscosity of 38
mins reduced the friction resistance to viscous
resistance, leading to the reductions greater
than 88%.
(3) The types of injection and the distribution charac­
teristics of lubricant may vary between pipejack­
ing projects. Notwithstanding that, this study
provides an access of evaluating the lubrication
performance for pipejacking works and the pre­
sented results would be useful in managing the
lubrication performance for upcoming pipejack­
ing project.
REFERENCES
ASTM. 2011. F1962-11 Standard Guide for Use of Maxi-
Horizontal Directional Drilling for Placement of Poly­
ethylene Pipe or Conduit under Obstacles Including
River Crossings. West Conshohocken: PA.
British Standards. 2009. BS EN 1594-09 Gas Supply
System-Pipelines for Maximum Operating Pressure
over 16 Bar-Functional Requirements. Brussels: UK.
Cui, Q.L., Xu, Y.S., Shen, S.L., Yin, Z.Y. &
Horpibulsuk, S. 2015. Field performance of concrete
pipes during jacking in cemented sandy silt. Tunneling
and Underground Space Technology 49: 336–344.
German ATV Rules and Standards. 1990. ATV-A 161
E-90. Structural Calculation of Driven Pipes. Hennef:
Germany.
Japan Microtunnelling Association (JMTA). 2013. Micro­
tunnelling Methods Serious II, Design, Construction
Management and Rudiments. Tokyo: Japan.
Pellet-Beacour, A.L. & Kastner, R. 2002. Experimental and
analytical study of friction forces during microtunneling
operations. Tunneling and Underground Space Technol­
ogy 17(1): 83–97.
Stein, D., Möllers, K. & Bielecki, R. 1989. Microtunneling:
Installation and Renewal of Nonman-Size Supply and
Sewage Lines by the Trenchless Construction Method.
Berlin: Germany.
18
Staheli, K. 2006. Jacking Force Prediction: An Interface
Friction Approach Based on Pipe Surface Roughness.
Ph.D. Thesis. Georgia Institute of Technology.
Terzaghi, K. 1936. The shearing resistance of saturated
soils and the angle between the planes of shear. In Pro­
ceeding of the 1st International Conference on Soil
Mechanics and Foundation Engineering, June 1936.
Cambridge: MA.
Terzaghi, K. 1943. Theoretical Soil Mechanics. New York:
USA.
The Ministry of Construction of the People’s Republic of
China (MOC). 2002. GB 50332-02 Structural Design
Code for Pipeline of Water Supply and Waste Water
Engineering. Beijing: China.
Zhang, H., Zhang, P., Zhou, W., Dong, S. & Ma, B. 2016.
A new model to predict soil pressure acting on deep
buried jacked pipes. Tunneling and Underground Space
Technology 60: 183–196.
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

A recent subway construction incident in soft alluvial deposits of Taiwan

Wen-Chieh Cheng & Ge Li
Xi’an University of Architecture and Technology, Xi’an, China

Md Mizanur Rahman
University of South Australia, Mawson Lakes, Australia

ABSTRACT: A recent water leak incident taken place throughout the parallel tunnels LUO09 construction in
the soil alluvial deposits in Kaohsiung, Taiwan was analysed and discussed in this study. Dumping quick-set
cement was intended to ease the water leak incident but in vain. The water leak was initiated by the piping and
the associated ground loss caused two large surface cave-ins. The existing underpass caused the jet-grout col­
umns installed not exactly in plumb. Their overlapping was estimated to be less than the design value of 60 cm,
developing seepage-prone weak zones. The hydraulic gradient being equal to 12.1 and existence of seepage-
prone weak zones were deemed as the main cause initiating the water leak incident. The pinhole test results
highlighted not only the nonplastic nature of the Kaohsiung silt but also its vulnerability to piping under large
hydraulic gradients. Some bullet points that indicate engineers should do or avoid were summarised.

1 INTRODUCTION incident often occurs, accompanied with significant

Braced excavation and/or shield tunnelling in soft
ground with high piezometric levels leads to a high
potential to trigger water leak incident. Such water leak
into excavation pit or sewage pipeline can result in
ground collapse and damages to adjoining facilities.
Prevention of such leaking incident has thus been
deemed to be the key in developing sustainable infra­
structure in urban areas. Jo et al. (2016) indicated that
the migration of the fine particles mainly contributed to
the ground sinking incident occurred in Seoul, Korea.
Hou et al. (2015) declared that the combined effects of
the sewage pipeline corrosion, soil strength deterior­
ation, and construction disturbance triggered a 20-m
long, 14-m wide and 12-m deep cave-in throughout
soft soil tunnelling of Beijing metro Line 10. Chen
et al. (2015) indicated that the ground collapse occurred
in a 15.7-m deep excavation in very sensitive clay in
Hangzhou, China caused damages to the retaining
structure and the water main and led to ground sinking
of a road next to the excavation. Feng and Lu (2016)
reported that the bi-slurry grouting failed to stop the
seepage causing a failure of retaining structure installed
in very thick sand layers with a high phreatic surface
during a metro station excavation in Nanchang, China.
Tan & Lu (2017) indicated that a flawed slab connector
embedded in the silt and sand layers has been deemed
as being the main cause to lead to a sudden outburst of
groundwater while deepening a subway excavation in
Shanghai, China. The above studies indicate that
casualties and economic losses, while developing infra­
structure in urban areas and relates mostly the phenom­
enon of water leak despite several protective measures
available. In spite that serious attention has been
drawn, consensus on prevention and mitigation meas­
ures has not reached yet.
The objectives of this study are: (1) to analyse
and discuss the water leak incident occurred while
deepening the middle sump pit of the parallel tunnels
LUO09, (2) to reveal the triggering mechanism, with
reference to the testimony from the men at work and
the field measurements, and (3) to suggest preventive
measures against similar leaking incident.
2 BACKGROUND
Two 837-m long, 6.1-m wide parallel tunnels LUO09
were excavated using the shield tunnelling method.
A 6.24-m diameter earth pressure balance (EPB) shield
was responsible for tunnel boring operations. Their
ends are linked to the subway stations O7 and O8 of
Kaohsiung metro system, and a cross passage with
a middle sump pit at 32.6 m depth was built for safety
purpose. Several boreholes that penetrate through a 42­
m thick alternating layer of soft clay and silty sand and
into a more than 30-m thick silty gravel were installed
providing a detailed description of the field geological
conditions. Figure 1 shows the soil properties profile.
The groundwater level was at 5-6 m depths below the

DOI: 10.1201/9780429321559-3

19
 Figure 2. Scenario for progressing the water leak incident.

Figure 1. Soil properties profile.

ground surface. A series of 3.2-3.5 m diameter jet-

grout columns at the depth of 35 m, with a mutual
overlapping of 60 cm, were installed in the vicinity of
the cross passage using the super jet-midi (SJM)
method because of the surrounding silty sand of high
permeability. The grouting tubes, while boring oper­
ations, were installed with horizontal included angles
because of the existing overlying vehicle underpass.
Upon completion of the cross passage, the
3.3-m diameter sump pit was excavated.

3 INCIDENT, DAMAGE AND REMEDIAL

MEASURE Figure 3. Two surface cave-ins.
It was noticed that an outburst of mud water presented
while excavating to near the bottom of the sump pit.
Authorities urgently responded to the water ingress by Subsequently a 2-year rehabilitation depicted in
dumping sand bags and quick-set cement but in vain. Figure 4 was performed to resume the traffic and its
Men at work soon heard sounds of breaks at liner activities are detailed as follows:
joints of the tunnel on the south side, with water (1) Two watertight plugs in the tunnels at some dis­
leaked into the tunnel from the ripped liner joints. The tance from the point of water ingress were to
water also carried away fine particles and the tunnel secure the watertight effectiveness with add­
due to this reason lost contact with the ground leading itional jet-grout blocks;
to stepwise surface settlement in the longitudinal dir­ (2) Porewater within the surrounding soil was
ection. More soil, while the settled cone expanded frozen using the ground freezing method;
upwards, fell by slipping along the vertical wall of the
vehicle underpass and passing through the ripped liner
joints into the tunnel. Not only serious ground loss,
but also two undermined water mains transferred the
soil into more flowable debris, causing a large surface
cave-in on the south side. The ground loss also
caused the tunnel on the north side to sink, accompan­
ied by the offsets of the liners. The soil fell into the
tunnel mainly from the liner offset at the junction
between the tunnel and the cross passage, leading to
another surface cave-in. Figures 2-3 show the scenario
for progressing the water leak incident and the two
surface cave-ins respectively. The amount of earth and
quick-set cement dumped to fill the surface cave-ins
was estimated to be as much as 12,000 m3, whereas
the water inflow, resulting from the two undermined Figure 4. Schematic illustration for progressing the
water mains, was approximately 2,000 m3. rehabilitation.

20
(3) Diaphragm walls of 1.5 m in thickness to retain
the excavation-induced lateral loads were con­
structed at the depth of 60 m;
(4) Dewatering wells screened at the bottom of the
tunnels were constructed to secure the excavation
face dry;
(5) The removal of the undermined vehicle under­
pass was conducted upon completion of Region
“II” excavation;
(6) New tunnel liners were erected following com­
pletion of Region “I” excavation;
(7) Controlled low strength materials were back­
filled to the bottom of new vehicle underpass,
allowing for new vehicle underpass installation; Figure 5. Layout of monitoring instruments and locations
of surface cave-ins.
(8) Controlled low strength materials were back­
filled to surface to resume traffic.

4 FIELD INSTRUMENTATION DATA

Such significant ground loss caused two surface cave-

ins to be developed on the south and north sides,
respectively. The tunnel on the south side and the
vehicle underpass due to loss of the contact with sur­
rounding ground significantly settled 2.7 m and 1 m,
respectively. The tunnel on the north side showed
a relatively small settlement of 0.16 m. Additionally,
the adjacent railway also measured some settlements
and track relative displacements. As discussed, the
diaphragm walls with the frozen ground and soilcrete
blocks aimed to secure the watertight effectiveness
while progressing the rehabilitation. The rehabilitation
including the removal of undermined vehicle under­
pass and the erection of new tunnel liners as well as Figure 6. Layout of monitoring instruments and locations
the installation of new vehicle underpass was per­ of surface cave-ins.
formed when the water inside the walls was drained
using the pumping wells and when the earth inside
was removed. Thus, the proposed layout of monitor­
ing instruments depicted in Figure 5 intended not only Table 1. Details for the monitoring instruments used in
to observe the groundwater level and surface settle­ the group-well pumping test.
ment variations while progressing the rehabilitation,
but to assess the associated environmental impacts. Water well Instrument Depth
Location type number m
To verify the watertight effectiveness of the dia­
phragm walls, frozen ground and soilcrete blocks, Inside d-wall OW OW-S1 40
a group-well pumping test was performed using eight Inside d-wall OW OW-S2 45
pumping wells installed inside the diaphragm walls Inside d-wall OW OW-02 45
prior to progressing the rehabilitation. Figure 6 pre­ Inside d-wall OW OW-04 45
sents the variations of the groundwater levels and the Inside d-wall OW OW-08 45
number of pumping well opened during the group- Inside d-wall OW OW-10 45
well pumping test. The group-well pumping test con­ Outside
sisted of the pumping phase and recovery phase and OW OW-14 40
d-wall
each phase continued for 7 consecutive days. Table 1
lists the monitoring instruments used in the test. It Note: d-wall = diaphragm wall, OW = observation well.
can be seen from Figure 6 that the groundwater levels Depth indicates where monitoring instrument is installed.
in the observation wells OW-S1, OW-S2, OW-04,
and OW-10 declined sharply to the 33-35 m depths in
the very beginning of the pumping phase and main­ OW-08. There was a sudden increase in the ground­
tained almost constant until the end of the pumping water levels right before the completion of the pump­
phase. A similar tendency, but with smaller variation, ing phase, most likely because of the recharge effect
was observed from the observation wells OW-02 and resulting from a rainfall. The recovery rates of the

21
observation wells OW-S1, OW-S2, OW-04, and OW­
10 gradually declined from 1.2 m/day in the begin­
ning of the recovery phase to 0.1-0.2 m/day in the
end of the recovery phase. Except the duration of
rainfall, the groundwater level in the observation well
OW-14 remained almost constant, indicating that the
group-well pumping appeared to have negligible
influence on the groundwater level outside the walls,
which also indicated good watertight effectiveness.
Figure 7 shows the variations of the ground sur­
face settlements while progressing the rehabilitation.
The surface settlement point SM-T1 measured the
smallest settlement, compared with the surface
settlement points SM-T2, SM-T3, SM-T4, SM-A,
SM-B, and SM-C. Similarly, the surface settlement
points SM-01, SM-02, SM-04, SM-06, and SM-07
measured the relatively small settlements, compared
with the surface settlement points SM-03 and SM­
05. This phenomenon is due to the fact that the indi­
cated surface settlement points were all close to the
short side of excavation pit that possesses larger
bending stiffness than the long side, exhibiting smal­
ler lateral deflection and surface settlements. Despite
the nonnegligible contribution to the ground surface
settlements from the effects of the diaphragm wall
construction and the strut removal, the associated
increases were typically 4.8-9.4 mm and
5.5-7.5 mm, respectively, for SM-A to C and SM-T1
to T4 and 1-13 mm and 8.2-13.1 mm, respectively,
Figure 7. Variations of ground surface displacements for
progressing the rehabilitation: (a) SM-A to C and SM-T1
to T4 and (b) SM-01 to 07.
22
for SM-01 to 07. While the pit excavation appeared
to have minimal influence on the ground surface
settlements. It is worth to mention that when the
ground was unfrozen prior to the installation of new
vehicle underpass, the ground surface settlements
increased very quickly, especially for SM-01 to SM­
07. Despite the increases in the surface settlement,
the surface settlements then soon reached a steady
state condition.
5 ANALYSIS AND DISCUSSIONS
5.1 Triggering mechanism
The piping phenomenon was responsible for initiating
the water leak incident, with reference to the testi­
mony of workmen. Notwithstanding that, the initiation
of piping was conceived variously. The existing
vehicle underpass, in fact, led to some difficulties in
installing the jet-grout columns exactly in plumb.
Thus, the overlapping of jet-grout columns was esti­
mated to be less than the design value of 60 cm and
seepage-prone weak zones were thus developed. On
the other hand, as the groundwater level was at 5-­
6 m depths below the ground surface, the head differ­
ence between the groundwater level and the
34 m depth where the piping initiated was 29 m. The
length of shortest path for groundwater seepage was
from the bottom of the soilcrete body to the point of
water ingress and measured at about 2.4 m. This
came out with the hydraulic gradient i being equal to
about 12.1 (i=29 m/2.4 m=12.1). The critical
hydraulic gradient icr for triggering piping within gap-
graded sand-gravel mixture could be as low as 0.2-0.3
as compared to icr=0.9-1.0 for clean sands (Skempton
& Brogan 1994). In spite that this might not be true
for the Kaohsiung soil with the finer particle size
range, the soil particles could easily be detached or
washed away as subjected to such high hydraulic gra­
dient i of 12.1. It is evident that the seepage-prone
weak zones allowed the water to flow through the jet-
grout columns and into the sump pit and that this high
hydraulic gradient made the water inflow even greater.
The combined effects of high hydraulic gradient and
existence of seepage-prone weak zones were deemed
as the main cause to initiate the water leak incident.
5.2 Soil erodibility
Sherard et al. (1976) described the pinhole test where
water flows through a 1-mm diameter hole punched
in a specimen of compacted clay and the water emer­
ging from dispersive clay carries a suspension of col­
loidal particles (Bell & Walker 2000). Sherard et al.
(1977) then utilised the pinhole test to investigate ero­
sion in clay-silt mixtures in response to the problem
emerging from occasional failure of low-height earth
dams which occurred upon water filling in Australia
and U.S. The pinhole test provides a direct qualitative
measurement of the dispersibility or deflocculation
and consequent erodibility of clay soils as described that there is no necessity to perform the test through
in ASTM standard D4647-93 (ASTM 2006). There to 1020 mm of head. The purpose to further increase
were seven 33-mm diameter, 25-mm height speci­ the head is to classify dispersive soils, which is not
mens tested in the pinhole test for which their proper­ within the scope of this study. The effluent turbidity
ties are summarised in Table 2. The average grain and size of pinhole at the end of each test were
size D50 for the specimens was about 0.075 mm. recorded. If the effluent remains clear and the pinhole
A steel nipple was pushed into the specimen and the size unenlarged, then the soil is non-dispersive. Con­
hole was punched through the nipple as a guide hole trarily, the effluent is turbid and the pinhole size is
using a 1-mm diameter steel needle. Wire screen and enlarged for dispersive soil.
pea gravel were placed in the pinhole cell on either Table 3 summarises the results of the pinhole test.
side of the specimen, as shown in Figure 8. As the For the specimens which were categorised as D, they
pinhole cell assembled, distilled water was allowed to were dispersive soils. There was also a common char­
percolate through the specimen under constant heads acteristic. Most of the specimens primarily included
of 50, 180, 380, and 1020 mm, in accordance with silt with a fines content varying from 52 to 96% (Lu
ASTM standard D4647-06. The principal differenti­ et al. 2019) and their Unified Soil Classification
ation between dispersive and non-dispersive soils, System (USCS) symbols were ML and SM-ML. The
however, is given by the test results under 50 mm of pinhole size enlarged in a range of 2-4 mm after the
head, as suggested by Sherard et al. (1976), indicating test. The specimen S-19 categorised as ND was non-
dispersive soil with a fines content of 94% and
mainly consisted of clay. The USCS classified this
specimen as CL, which was an indication of low plas­
Table 2. Summary of soil properties in the pinhole test. ticity clay. The pinhole size for the specimen S-19
remained unenlarged after the test. The test results
Sand Silt Clay Water content LL PI
revealed a fact indicative of the tendency of the silt
from Kaohsiung, Taiwan to possess high dispersion,
Specimen % % % % % or very low strength, or both. The particle-size distri­
bution analysis indicated that the Kaohsiung soil
S-14 19 76 5 22.4 27 4.1
belonged neither to the gap-graded gravelly sands sus­
S-15 48 52 0 25.6 21.4 0.8
ceptible for segregation piping nor to the dispersive
S-16 15 82 3 24.4 27.5 1.2
clay prone to internal erosion. In fact, the particle size
S-17 4 86 10 19.5 35.2 6.4
for the Kaohsiung soil lay midway in the range of silt
S-18 20 74 6 21.7 26.5 3.3
and sand. The pinhole tests were regarded as an
S-19 6 56 38 42.9 43.9 20
effective means to identify not only the nonplastic
S-20 15 84 1 25.1 26.1 1.2
nature of the Kaohsiung silt contained in the sand but
Note: LL = liquid limit, PI = plasticity index. also its high vulnerability to piping or internal ero­
sion. It may be conclusively mentioned that the seep-
age-prone weak zones and high hydraulic gradient
initiated the water leak incident and that water from
the two undermined water mains transferred the Kao­
hsiung silt into more flowable debris and its nonplas­
tic nature aggravated the collapse even further.
Despite an apparent inadequacy in dealing with simi­
lar incident in other sites, the local sandy silt was
found to be prone to the piping at deep depths under
large hydraulic gradients and the two undermined
water mains enhanced its flowable nature, thereby
enlarging the scale of the water leak incident.

5.3 Lessons learned

Quite often subway construction failures are not
explained nor published, which makes the same mis­
takes to be made another time with great casualties
and economic losses. Some bullet points that indi­
cate what engineers should do or avoid are learned
and summed up as follows:
(1) Quality control for grouting: Since in-ground
obstructions such as wastewater and electricity
mains can result in some difficulty in installing
Figure 8. Layout of monitoring instruments and locations
of surface cave-ins. jet-grout columns exactly in plumb, a trial

23
Table 3. Summary of the pinhole test results.
Final head Colour cloudiness
Specimen mm
S-14 50 Moderately dark
S-15 50 Barely visible
S-16 50 Moderately dark
S-17 50 Dark
S-18 50 Dark
S-19 180 Clear
S-20 50 Moderately dark
Note: Pinhole size = Pinhole size measured after the test, D = dispersive soil, ND = non-dispersive soil.
grouting should be performed prior to the
formal one to verify the design grouting
parameters.
(2) Seepage-prone weak zone: Except for trial
grouting, additional soilcrete columns, result­
ing from chemical grouting, should be con­
sidered during design phase and constructed
next to jet-grout columns, which not only
increases the length of path for the ground­
water seepage, but also prevents formation of
the seepage-prone weak zone.
(3) Local soil nature: The Kaohsiung silt contained
in the sand has proved to possess high disper­
sion, or low strength, or both. There is
a necessity to investigate the nature of local
soils while progressing upcoming subway con­
struction project if they are found to exhibit
peculiar behaviour differentiated from common
sense in soil mechanics.
6 CONCLUSIONS
The main cause to lead to the water leak incident
was investigated. The pinhole test results highlighted
not only the nonplastic nature of the Kaohsiung silt
but also its vulnerability to piping under large
hydraulic gradients. Some main conclusions can be
drawn as follows:
(1) The piping was responsible for initiating the
water leak incident. Dumping sand bags and
quick-set cement failed to ease the water leak.
Mud water carried away fine particles in the
ground and together flowed into the tunnels
through the ripped liner joints leading to serious
ground loss. Such ground loss caused two sur­
face cave-ins. The surface cave-ins not only
impeded traffic, but also caused damages to
adjacent properties.
(2) The hydraulic pressure at the point of water
ingress was estimated to be about 300 kPa.
The length of shortest path for groundwater
Pinhole size Specific surface area Dispersion classification
mm m2/g
3 3.20 D
2 3.27 D
3 3.11 D
4 2.95 D
4 2.84 D
1 6.18 ND
3 3.07 D
seepage was from the bottom of the soilcrete
body to the point of water ingress; that is,
2.4 m, corresponding to the hydraulic gradi­
ent i of 12.1.
(3) The existing vehicle underpass led to some dif­
ficulties in installing the jet-grout columns
exactly in plumb and some seepage-prone weak
zones were thus developed. The seepage-prone
weak zones allowed the water to flow through
the jet-grout columns into the sump pit and
such high hydraulic gradient made the water
inflow even greater. The combined effects were
deemed as the main cause initiating the water
leak incident.
(4) The results of the pinhole test identified not
only the nonplastic nature of the Kaohsiung silt
but also its vulnerability to piping or internal
erosion under large hydraulic gradients. Water
from the two undermined water mains and the
nonplastic nature of the Kaohsiung silt aggra­
vated the collapse even further. More detailed
investigation is considered to be necessary to
address the explored issue with different aspects
of view.
REFERENCES
ASTM. 2006. D4647-93 Standard Test Method for Identifi­
cation and Classification of Dispersive Clay Soils by the
Pinhole Test. West Conshohocken: PA.
Bell, F.G. & Walker, D.J.H. 2000. A further examination of
the nature of dispersive soils in Natal, South Africa.
Quarterly Journal of Engineering Geology and Hydro-
geology 33: 187–199.
Chen, R.P., Li, Z.C., Chen, Y.M., Ou, C.Y., Hu, Q. &
Rao, M. 2015. Failure investigation at a collapsed deep
excavation in very sensitive organic soft clay. Journal of
Performance of Constructed Facilities 29(3): 04014078.
Hou, Y.J., Fang, Q., Zhang, D.L. & Wong, L. 2015. Exca­
vation failure due to pipeline damage during shallow
tunnelling in soft ground. Tunnelling and Underground
Space Technology 46: 76–84.
Jo, Y.S., Cho, S.H. & Jang, Y.S. 2016. Field investigation
and analysis of ground sinking development in
24
 a metropolitan city, Seoul, Korea. Environmental Earth
Sciences 75: 1353.
Lu, J., Wang, T.H., Cheng, W.C., Yang, T. & Luo, Y. 2019.
Permeability anisotropy of loess under influence of dry
density and freeze-thaw cycles. International Journal of
Geomechanics 19(9): 04019103.
Sherard, J.L., Steele, E.F., Decker, R.S. & Dunnigan, L.P.
1976. Pinhole test for identifying dispersive soils. ASCE
Journal of Geotechnical Engineering Division 102: 69–85.
25
Sherard, J.L., Dunnigan, L.P. & Decker, R.S. 1977. Some
engineering problems with dispersive clays. In:
Sherad, J.L. & Decker, R.S. (eds.), ASTM STP 623; Pro­
ceedings of Symposium on Dispersive Clays, Related
Piping, and Erosion in Geotechnical Projects, Chicago,
27 June–2 July 1976.
Tan, Y. & Lu, Y. 2017. Forensic diagnosis of a leaking acci­
dent during excavation. Journal of Performance of Con­
structed Facilities 31(5): 04017061.
 Geotechnical Aspects of Underground Construction in Soft Ground – Elshafie, Viggiani & Mair (eds)
© 2021 ISSMGE, London, UK, ISBN 978-0-367-33733-9

Settlements due to tunneling in the City of São Paulo

A. Lopes dos Santos
Navier Géotechnique, École des Ponts ParisTech, Champs-sur-Marne, France

W. Bilfinger
Vecttor Projetos Ltda., São Paulo, Brazil

H.C. Rocha
Companhia do Metropolitano de São Paulo - Metrô-SP, São Paulo, Brazil

ABSTRACT: Major tunneling works in the city of São Paulo have started in the 1970´s, with the construc­
tion of the 1st Metro Line. Since then, tunnels with different cross sections have been constructed regularly,
using either mechanized or conventional tunneling methods. From a geological point of view, the city is
mainly located in a sedimentary basin, with its subsoil partly composed by residual soils overlaying bedrock.
Experience has shown that the induced settlements in the residual soils are higher and more variable. Historic
registers of tunneling induced settlements measured for different constructive techniques and soil types are
presented on this paper. Focus is given to the measurements during the construction of Metro Line 4, where
approximately 5 km of line tunnels were excavated using NATM in residual soils, saprolite and rocks, and
around 8 km were excavated in tertiary sediments, using an EBM TBM. Typical values for volume losses on
this context are proposed.

1 INTRODUCTION were excavated using NATM in residual soils, sapro­

São Paulo is one of the so-called Megacities of the
world and, for this reason, the use of tunnel is the
only lasting solution to solve several problems,
including transportation and utilities. Historical
registers show that in the early 20th century some
tunnels were built. But major tunneling started in the
1970´s, with the construction of the 1st Metro Line.
Since then, tunnels with different cross sections have
been built regularly, using either mechanized or con­
ventional tunneling methods.
São Paulo is located, from a geological point of
view, mainly in a sedimentary basin, with tertiary
clays and sands, locally overlain by quaternary soils
and manmade ground. Partially, the subsoil consists
also of residual soils overlaying bedrock. Tunnel
construction in these completely different environ­
ments generates the need for different constructive
techniques and experience has shown that the
induced settlements in the residual soils are propor­
tionally higher and more variable.
The paper will present historic registers of tunnel­
ing induced settlements and analyses of volume
losses measured for different constructive tech­
niques. Focus will be the settlements measured
during the construction of Line 4 of Metro São
Paulo, where approximately 5 km of line tunnels
lite and rocks, and around 8 km were excavated in
tertiary sediments, using an EBM TBM.
Finally, typical values for volume loss evaluated
in different geological environments and different
constructive methods will be proposed.
2 SETTLEMENTS DUE TO TUNNELLING
2.1 Fundamental aspects
The first fundamental aspect about settlements
related to tunneling is that they will certainly occur.
Any relevant ground excavation will result in ground
movements, of which the causes, magnitude and
consequences depend on many factors, constituting
a very complex phenomenon briefly discussed in the
following paragraphs.
The main causes of settlements associated to tun­
neling are related to soil stress state disturbance due
to the excavation process. Every part of this process
may have an influence on the induced ground move­
ments, from groundwater lowering, to the cross-
section opening itself. Ground pre-conditioning,
lining stiffness and its installation distance from the
face also have an influence.
In conventional tunneling, the settlements induced
ahead of the tunnel face are directly related to the

DOI: 10.1201/9780429321559-4

26
Another random document with
no related content on Scribd:
freqüent pero enérgica y brusca. En los segons, la acció del Estat pert
en intensitat lo que guanya en extensió. Es que la lluyta s’ha
generalisat entre tots los membres á la mida de la generalisació y
aument de las necessitats, perdent al meteix temps lo carácter de
guerra oberta pera pendre lo de la competencia. La idea abstracta de
justicia, traduhida en dret positiu al punt que la defineix una
agregació social, está subjecta á la lley de evolució que presideix al
desenrotllo aixís de la vida moral com de la vida material. Per aixó,
en un Estat avansat en cultura, la lley ha de ser complicada, fundada
en sos preceptes, y suau en la sanció.

L’Estat cumpleix sa missió de definir y aplicar lo dret ab tanta mes

ventatja pera sos elements, en quant més respecta la llibertat y
fomenta la iniciativa de tots aquestos. En una agrupació política ben
organisada, cada individuo, al despendres d’una part de la seva
independencia en pró del conjunt, ha de rebre en cambi molt més de
lo que dona. La seguretat de son dret y la garantía de las llibertats
que se reserva, han de deixarlo de fet molt més lliure que no fora
sens haverse imposat aquellas limitacions. Los debers socials que’s
contreuhen, en una paraula, han d’estar exuberantment compensats
ab la seguretat y facilitat que’s trobin en l’exercici dels drets y en la
expansió de totas las iniciativas. La lley més perfecta es la que’s basa
en lo fecundo principi de llibertat; la que no posa al dret de cada hu
més límits que’l deber correlatiu de respectar lo dret dels altres.

La branca del dret més difícil de fixar y definir es la que’s refereix á la

organisació de la familia y regulació de la propietat, que son las bases
de la llegislació civil. Contats, contadíssims, si es que algún n’ha
existit, son los pobles en que s’hagi ofegat la societat familiar y
prescindint de la propietat individual. Baix los régimens més
tiránichs, la llar doméstica ha merescut cert respecte, y fins quan la
col·lectivitat, representada per un ó per molts, s’ha cregut árbitra de
disposar á son capritxo dels bens dels subdits ó ciutadans, s’ha deixat
á aquestos la potestat de fer entre ells transaccions y contractes. Pot
haverse establert lo sistema de castas; pot haverse fundat la
organisació del Estat en la esclavitut del major número: la classe
priviligiada, la que ha representat l’element actiu, ha gosat sempre
d’un grau de llibertat que li ha permes establir la competencia entre
sos membres. Moltas vegadas, las institucions políticas han estat en
contradicció ab las de dret civil, establint las unas lo despotisme y
reconeixent las altras la llibertat, ó vice versa, pero fins ara no hi ha
hagut en cap Estat un poder prou potent pera ofegar totas las
iniciativas. Sempre s’ha hagut de deixar oberta una válvula per la
qual pogués escaparse lo vapor de la llibertat, producte fatal y
necessari de la naturalesa del home. L’Estat pot contrariar la lley
natural, pero no destruhirla.

Si la missió propia del Estat en la definició y aplicació del dret, ó

sigui en lo cumpliment de son fi jurídich, no es altra que presidir y
regular la lluyta entre’ls interessos distints y fins oposats dels
elements associats, un cop lo meteix Estat entra á fomentar la
civilisació y la cultura general, no deu reduhirse á presidir y regular
la lluyta, sinó que ha de pendre en ella part activa. Al establir serveys
públichs, com per exemple los de correus, telégrafos, etc., etc., tota la
forsa social se utilisa com element actiu de competencia. Si crea
universitats y escolas, y organisa museus, y subvenciona institucions
científicas, artísticas, de beneficencia ó de qualsevol altra classe, no
porta altre objecte que aumentar las aptituts de sos elements pera
que puguin obtenir victoria ó defensarse al menys en la gran lluyta de
la vida. Al desempenyar l’Estat las funcions que tenen per objecte
fomentar la civilisació y la cultura, reparis com moltas vegadas no’s
reduheix á pendre part en la competencia entaulada, sinó que la
provoca. Pera provehir los empleos, convoca á oposició, es á dir, á
lluyta; pera adjudicar travalls públichs, posa en competencia als que
poden executarlos, y n’encarrega als que surten victoriosos. Si vol
protegir las arts ó las ciencias, crida á certamen. No té altre camí que
la lluyta. Tot lo que sens ella naix, surt anémich y difícilment arriva á
poder viure. Lo que no es fill de lluyta ho es de la imposició, y la
imposició es lo més contrari al progrés y á la millora.

Lo fins aquí indicat basta pera demostrar que la missió altíssima del
Estat no es pas suprimir la lluyta, sinó regularla, al objecte de que sos
resultats siguin civilisadors y progressius. Aquestos están en relació
directa de la activitat y energía de la lluyta, per qual motiu l’Estat,
lluny de restringirla ó contrariarla, ha de procurar per tots sos medis
fomentarla y exténdrela. Per medi de la direcció de las relacions
exteriors, ha de darli camp franch; per medi de la definició, fixació y
aplicació del dret, ha d’evitar que degeneri en batalla, mantenintla
dins dels límits de la competencia; per medi de las atribucions que se
li concedeixen pera la producció y aument de la cultura, ha de pendre
en ella part activa, provocantla, avivantla y extenentla á tots los
terrenos civilisadors.

Cumplirá l’Estat sa missió tant més perfectament, quanta més

importancia dongui al element de la varietat, ó, en altres termens,
quant més s’acosti al extrem de la llibertat. Mes, puig volem estar
sempre dins del terreno positiu, reconeixem de bon grat que
l’element contrari, ó sigui’l de la igualtat, ha de ser també tingut en
compte, y se li ha de respectar lo lloch que li pertoca. La armonisació
dels dos elements en tal forma que l’un no contrarihi al altre, sinó
que de la combinació dels dos resulti l’avens social, es precisament
l’objecte del particularisme.

En lo present capítol no hem fet més que presentar deslligadas,

generalitats de las que pensem fer aplicació en lo curs del nostre
travall. Al arribar á precisar las bases en las quals ha de apoyarse la
organisació particularista, aquellas generalitats anirán lligantse y
combinantse de manera que se presentin formant un sistema
complet.
Capitol II.
 Tendencia científica actual.
Procediment científich.– Lo especialisme.– Método de observació directa.–
Decahiment dels arguments de autoritat.– Positivisme en las brancas físicas
de la ciencia.– L’exemple es imitat per las metafísicas.– Aliansa entre unas
y altras.– Exemples trets de varias d’ellas.– Unió entre totas pel camí de la
varietat.– Dos grupos de sistemas filosófichs.– Conformitat del nostre
particularisme ab la tendencia científica actual.– Dificultats que ha de trovar
en son camí.

Proposantnos, com nos proposem, demostrar que’ls nostres punts de

mira son perfectament científichs, hem de donar una ullada á las
manifestacions de la ciencia en lo punt en que avuy se trova.

La primera observació que salta á la vista al dirigirla al camp

científich, es que sos procediments actuals son eminentment
particularistas. Vivim en plena época dels especialistas. La ciencia se
divideix y subdivideix: lo qui la cultiva concentra tota la activitat no
ja en una de sas cent brancas, sinó en un punt concret d’un dels mil
brots de que aquestas están formadas. A pesar de la facilitat ab que
cada qual disposa de tots los elements d’estudi desitjables, los
homens enciclopédichs, que van caracterisar lo período del
renaixement filosófich, van sent ja rara avís en los nostres temps.
Agafis una branca qualsevol de la ciencia física ó metafísica, y si’s
volen recordar los noms tan sóls de las especialitats á que donan
naixensa, se necessita una memoria prodigiosa. L’especialisme
invadeix totas las profesions científicas. Lo metje se reduheix á
estudiar una sola classe d’enfermetats; l’advocat se dedica á una sola
especie de qüestions; lo filosoph se fixa sóls en un ordre reduhit
d’especulacions. Dels que estudian lo llenguatje, per exemple, los uns
son filólechs, los altres lingüistas, los de més enllá políglotas, etc.,
etc.. Tan particularistas son avuy los procediments científichs, que
l’especialisme arriva potser á constituhir lo defecte de la época.

Los procediments son sempre fills del método, de manera que pot
ben assegurarse que sent aquells particularistas, particularista ha de
ser tambe aquest. Y en efecte, lo método científich tendeix avuy
directament cap al particularisme. La observació directa ha vingut á
ser la base de las ciencias, aixis de las que buscan las lleys dels sers
materials, com de las que estudian lo món metafísich. Las hipótesis
sens fonament real en fets demostrats no captivan avuy las
intel·ligencias. Los llibres de filosofía en que l’autor no hi va vuidar
més que’ls productes d’una admirable gimnástica de cervell, han
perdut quasi tota la autoritat de que han gosat en altras épocas. Tals
esforsos d’intel·ligencia admiran pero no convensen.

Lo procediment especialista y’l método d’observació directa portan

directament cap al positivisme científich. La varietat es la regla y lo
medi: lo fi, l’enllás entre las varietats. La observació directa,
practicada per especialistas, fa que’ls judicis hagin de ser fills de la
realitat, y que’s tendeixi á presentar tot lo que es objecte d’estudi tal
com es y no tal com podria imaginarse. Alguns exemples nos
donarán la demostració completa de la transformació que ha portat
al camp científich lo procediment actualment empleat, en
contraposició al que ha estat en predicament en altras épocas.
Fixemnos en las brancas de la ciencia que tenen per objecte lo
coneixement del home aixis en sa part física com en sas parts moral é
intel·lectual. En los temps en que predominava lo procediment de las
abstraccions, y en que totas las qüestions estavan implicitament
resoltas en quatre principis brillants, que eran considerats altras
tantas veritats absolutas é indiscutibles, lo coneixement exacte y
precis del organisme físich del home no tenia cap interés pera la
ciencia, que temerosa y encongida no s’hauria atrevit á obrir un
cadáver per temor á la profanació. Lo principi uniformador del
autoritarisme era la base de tot procediment científich, y dant
torment als textos dels autors predilectes, girantlos y capgirantlos, se
lograva ab penas y ab dolors aplicarlos á la qüestió que s’havia de
resoldre de conformitat ab los prejudicis admesos. Lo fet real y
observat no feya cap pes, y la observació directa y repetida devia
callar y prosternarse devant de una cita en llatí de Sant Tomás ó de
Aristóteles. La ciencia llavoras tendia á la concentració y era oposada
á tot particularisme.

Si de tal manera se resolvian las qüestions referents al home físich,

consideris com debian tractarse las del ordre moral é intel·lectual.
Los que las cultivavan, aceptavan los principis que se’ls davan com
inmutables, y llurs afanys se reduhian á acomodarhi las
conseqüencias. Tot particularisme era un absurdo científich, y per
mica que’s acostés al terreno teológich era calificat d’heretgia, en
qual cas son autor havia de temer fins per sa propia persona.

Las brancas físicas de la ciencia foren las primeras que feren la

evolució cap al positivisme, tardant molt temps á seguir llurs passos
las metafísicas. Havian passat molts anys desde que Leonard de
Vinci, Copérnich, Galiléo, Newton y cent altres havian ja restaurat la
observació directa com á base del estudi del món material, y’l moral
seguia encara regulat cientificament per prejudicis y abstraccions
sens base comprobada. En va fou que la teología més ortodoxa
hagués d’acabar per reconeixe que la Biblia, si es d’autoritat
indiscutible en lo terreno relligiós, no prejutjá cap qüestió en lo
científich, entregat á la investigació del home; puig á pesar de la
protesta de molts, las classes predominants van seguir aprofitantse
de las abstraccions y prejudicis, reduhintse á cambiarlos en lo precís
pera acomodarlos á llur interés. Lo moviment filosófich que va
precedir á la revolució francesa va declararlos guerra á mort; pero
tan prompte com los hagué destruhit, no’ls va suplir per los principis
d’un ordre oposat d’ideas, sinó per los prejudicis contraris, dins del
meteix ordre. Precindint de tota observació directa, va acceptar la
uniformitat com á base de son sistema, y cap á la uniformitat va
encaminar sas novas teorías. Al voler la revolució, per exemple,
aplicar sas Constitucions radicalment igualatarias lo meteix als
habitants de Paris ó de Lyó que als negres de las Guayanas, de cop y
volta transformats d’esclaus en ciutadans d’una democracia, obehia
sens repararho al meteix principi que havia impulsat sigles avans als
castellans á condemnar com heretje al Inca peruá per practicar la
relligió propia de la seva terra. En lo terreno polítich social los dos
extrems no podian deixar de trovarse en un resultat semblant. Un y
altre eran fills de la imposició autoritaria. Los pobles no havian fet
mes que cambiar d’amo, passant del absolutisme d’un rey al de una
massa; de la intolerancia teológica á la intolerancia filosófica.

Avuy, si las brancas de la ciencia que estudian la vida material,

persistint en lo camí que venen seguint fa ja sigles, van perfeccionant
y ampliant cada dia lo sistema de la observació directa, las que
s’ocupan de la vida moral han adoptat per fí aquesta via, y establint
unas y altras estreta aliansa, han obtingut ja grans ventatjas y
prometen avansar molt més encara. L’escarpell y la balansa son los
instruments capitals del observador, que pera multiplicar la potencia
y finura dels seus sentits disposa d’aparells de tota clase. Ab tan
poderosos medis s’ha proseguit l’estudi del home, y á cada pas que
s’avansa en son coneixement se van modificant las ideas que
predominavan, pronunciantse més y més en pro del particularisme.

En efecte: com mes se divideixen y subdivideixen las brancas de la

ciencia; com més se multiplican los camps d’observació, més se va
posant de manifest, que la varietat es regla de la naturalesa y
condició necessaria de la vida. Fixantnos sols en los estudis que més
directament nos interessan, ó siguin los que s’ocupan del home, aixis
considerat aisladament, com en relació ab los demés sers, veurem
que á cada pás que s’avansa se van descubrint diferencias fins en los
punts en que més s’havia afirmat la uniformitat. La antropología, per
exemple, amidant las cavitats del crani y pesant y examinant la
massa del cervell, nos ensenya á caracterisar y á classificar las
varietats y rassas, assignant á cada una d’aquestas lloch distint en la
escala de la perfecció, per medi de la observació individual
comparada. La etnografía y la etnología, brots de la branca de la
ciencia que estudia al home en general, lo consideran formant
agrupacions y pobles, y tot llur afany es cercar las diferencias que
caracterisan á cada grupo ó col·lectivitat al objecte de poder
classificarlos y distingirlos. La lingüística ve en auxili d’aquestos
estudis, y fixantse en las transformacions que ha anat experimentant
la expressió de las ideas, acaba de posar de relleu las diferencias
establertas per las brancas similars de la ciencia. Gracias á aqueix
conjunt d’observacions, sabem avuy que cada agrupació d’homens té
sa personalitat propia, que per herencia se transmet de generació en
generació, y que las circunstancias de lloch y temps que las envoltan
tenen en ellas influhencia incontrastable. Los avensos fets fins ara,
ab tot y que no s’ha arrivat de molt encara al punt á que lo sistema de
observació directa dels fets pot portar á totas las brancas de la
historia natural, nos demostran ja que es absurdo tot sistema que
pretengui subjectar á reglas idénticas á sers y col·lectivitats que
viuhen en tan distintas condicions, y’s trovan en tan diferents estats
de desenrotllo y de cultura. Avuy, lo sentimentalisme extraviat que
vol igualar al salvatje del cor del Africa ó de las islas occeánicas al
home blanch refinat de las grans poblacions europeas ó americanas,
se presenta ja com un romanticisme esbravat y passat de moda, y lo
qui pretengués portar á las societats imperfectas de la Terra de Foch
ó de la Micronessia las llibertats y drets de que disfruta un nort
americá ó senzillament lo més incult proletari del Estat més atrassat
d’Europa, no lograria altra cosa que fer esclafir en una riallada
general als que han aprés no més que’ls rudiments de la ciencia
sociológica. Essent absurdo lo més, ha de serho tambe lo menys dins
del meteix ordre d’ideas. De igual manera, donchs, que los grupos
d’homens que’s troban molt separats en la escala del desenrotllo han
de regirse per reglas distintas, distintas han de ser també, encara que
no en grau tan marcat, las que regeixin als que’s troban mes acostats,
puig tota diferencia per petita que sigui, es atendible, y necessitats
variadas han de ser satisfetas per medis variats.

Aqueix moviment cap á la observació directa no es exclussiu de las

brancas de la ciencia que estudian la vida del home y dels pobles,
sinó que influheix fins en las que s’ocupan de las materias mes
abstractas. La estética, la psicología, la moral, la meteixa teología
s’han fet també observadoras y experimentals. La bellesa no está ja
subjecta á un patró fixo, sinó que es admesa sigui la que sigui la
vestidura ab que se adorni. La obra artística no ha de enmotllarse á
reglas arbitrarias preconcebudas, y’l que trova una nova manera de
fer sentir impressions, mereix general aplauso. La originalitat es una
de las qualitats que més s’estiman, y la originalitat no es res més que
la consagració de la varietat. La psicología dona cada dia major
importancia á las sensacions, ó sigui á la comunicació de l’ánima ab
lo món exterior, y si la moral ha de reconeixer que está tambe
subjecta á la lley de la evolució, essent més ó menys perfecta segons
lo grau de desenrotllo y de cultura dels sers capassos de practicarla,
la teología mes ortodoxa, influhida per la corrent general, no pot
evitar lo entrar en estudis comparatius, reconeixent per aquest sol fet
los drets de la varietat com element de progrés, y de millora.

La estreta aliansa establerta entre las brancas materials de la ciencia

y las abstractas, es la que ha produhit que unas y altras prenguessin
lo carácter que estem analisant, treyentne abduas gran benefici. Ella
es la que ha donat á totas las manifestacions científicas l’aspecte
positivista que avuy las distingeix, y las fa progressar depressa.
L’element de la varietat del que la ciencia s’havia privat en altras
épocas, la vivifica y vigorisa. La llibertat científica es la mes hermosa
y tracendental conquista dels temps moderns.

Y notis be que la varietat, filla de la llibertat, no pren en lo camp de la

ciencia cap aspecte de mesquinesa ni conduheix al exclussivisme,
sinó al contrari. La aliansa estreta establerta entre las brancas del
saber, lluny de rompres á mida que las va fent á totas més
particularistas, se va refermant y estrenyent, fins al punt de que
avuy, mes que aliansa, es ja germanor perfecta. Lo carácter general
científich es precisament la unitat de propósits entre’ls elements que
en lo moviment prenen part. Cada especialista travalla afanyós en lo
terreno que cultiva, pero jamay abandona lo punt de mira de
contribuhir á resultats de tracendencia general. Es que lo
particularisme es l’únich camí que porta cap á la unió verdadera. Pel
camí de la experimentació, la ciencia ha arribat ja á entreveure la
identitat de la materia per entremij de las inumerables varietats de
forma ab que’s presenta, y no está lluny de poder fixar, al menys en
sos carácters generals, la filosofía de la historia, ó sigui la lley
constant que presideix á la marxa de la humanitat. Tals resultats y
esperansas son fills llegítims y naturals del método científich que’s
basa en lo particularisme.

Tots los sistemas filosófichs que fins ara ha concebut l’home y tots
los que concebirá mentres sols disposi de las facultats que
actualment lo posan en relació ab los móns exteriors, poden ben bé
calificarse en dos grans grupos: idealisme y positivisme. Quan se
precindeix de la realitat, y las qüestions se resolen en virtut de
teorías preconcebudas y fillas de generalisacions autoritarias, los
resultats han de ser purament convencionals, y s’imposan més á la
imaginació que á la rahó. Quan s’atén á la realitat y no se avansa un
pás que no pugui fundarse en la observació directa y comprobada, las
conseqüencias, menys brillants que sólidas, s’imposan á la rahó més
que á la imaginació. Los sistemas idealistas produheixen entusiasme,
pero llurs efectes per lo meteix que son molt intensos, no tardan á
decaure. Los sistemas positivistas no arrivan quasi jamay á
entusiasmar, pero en cambi son d’efectes mes durables, puig que
satisfán á la rahó y donan convenciment. Uns y altres sistemas tenen
un gran perill en l’exclusivisme. L’idealisme arriva moltas vegadas á
prescindir completament dels fets, aixis com lo positivisme
precindeix moltas altras de las abstraccions. Reduhintnos als
sistemas basats en la observació, afegirem que sols es llegítim lo
positivisme, quan pera sas resolucions té en compte totas las
impressions de que son capassas las variadas facultats del home,
aixis las reflexivas com las imaginativas.

Lo positivisme més científich y filosófich, donchs, es aquell, que

partint de la base de la observació directa y comprobada, té per
objectiu arrivar á la generalisació. Per fortuna aquest caracter, si be
que no prou marcat tal vegada encara, es lo distintiu de la ciencia en
la nostra época.

Baix aquest punt de vista, lo particularisme que defensem es

perfectament científich. Eminentment positivista, allí ahont trova
una varietat, no sols la respecta, sino que la fomenta pera que arrivi á
ser un element de progrés. Estudia y observa cuidadosament lo fet
particular, pero ab l’objectiu de agrupar tot lo estudiat y observat en
un sistema general. Arrenca de la llibertat, y camina cap á la unió.
Posa en moviment tots los interessos, pero al meteix temps los
dirigeix cap á la armonía, imitant aixis á la naturalesa, que donant á
la materia innumerables formas, y dotant als sers de facultats y
aptituts indefinidament variadas y variables, assenta en aquesta
varietat las lleys generals de la vida. La naturalesa, si se’ns permet
copiar una frase antigua de la física, te horror á la uniformitat. Dels
milions y milions d’homens que poblan y han poblat la terra, no n’hi
ha dos que siguin idéntichs; tant, que en aquesta falta d’identitat
s’apoya la personalitat de cada individuo, que fora impossible de
reconeixer sens la may desmentida varietat, que es la major
maravella de la naturalesa. Tots los homens, á pesar de llurs
diferencias individuals, forman la humanitat, ab virtuts y vicis
generals que no’s reuneixen en cap de sos membres, com en cap
d’ells no’s condensa lo prototipo general humá, y tots plegats estan
subjectes á lleys fatals, que son com lo llas d’unió de tota la especie.
Lo positivisme té també horror á la uniformitat, essent aixis fidel
imitador de la naturalesa. Lo nostre sistema, donchs, no sols
s’acomoda al positivisme, fill de la observació comprobada, que es lo
distintiu de la ciencia en sa evolució actual, sino que l’esperit
cientifich que l’anima se posará més y més de relleu á mida que la
ciencia anirá avansant camí en lo descubriment de las lleys naturals y
purificant lo seu positivisme.

Per desgracia lo positivisme particularista ha de trobar grans

dificultats pera poder avansar camí en lo camp polítich social. Los
que’l cultivan están impulsats per l’interés propi, y al ocuparse de la
cosa pública los va molt be lo sistema de las generalisacions y dels
idealismes. Los polítichs d’ofici no tenen més objectiu que la
possessió del poder, y quantas més facultats en aquest se concentrin,
més utilitats treurán d’ell al exercirlo.

Difícil, donchs, ha de ser substituhir lo empirísme que avuy ocupa’l

lloch dels verdaders principis científichs. No hi ha pitjor sort que’l
que no hi vol sentir, y la gent política es sorda per conveniencia.
No es estrany, donchs, que’l particularisme, ab tot y basarse en los
principis més rigurosament científichs, no hagi lograt per ara influir
directament en la marxa de la major part de las nacions.

Mes no per aixó hem de desanimarnos. Comensém per demostrar

que tenim rahó: un dia ó altre lograrem que se nos reconegui, á pesar
dels interessos illegítims que han de sortirne perjudicats.
Capitol III.
 La llibertat
Dificultat de definir la llibertat, y facilitat de explicar sos efectes.– Distints
conceptes de la meteixa segons los graus de civilisació y cultura.–
Concepte en los pobles clássichs.– Concepte en la revolució francesa.–
Definicions oficials dadas per aquesta.– Concepte modern de la llibertat.–
Escolas francesa y anglo-saxona.– Llibertat personal, civil y política.–
Aquesta última es no sóls un fí sinó també un medi.– Lo self government.–
Lo self-government institucional.– Caracterisació de la llibertat moderna.–
Lluyta entre’l poder y las iniciativas particulars.– Armonisació dels
interessos contradictoris.– Garantías sólidas d’equilibri.

Comensem ja á baixar de las alturas á que hem degut enfilarnos, y

acostemnos més á ran de terra, preguntantnos ¿que es la llibertat?

Lo més difícil de definir es quasi sempre lo que tots sentim. En

aquest cas deu trovarse la idea de llibertat, quan un dels mes
il·lustres representants de la rassa que en los nostres temps millor la
comprén en la práctica, comensa un capítol destinat á explicarla ab
las següents paraulas: “Moltas definicions s’han donat de la llibertat.
La major part ni mencionarse mereixen”.

Mes, si es difícil definir la llibertat, no ho es de molt tant explicar sos

efectes. Aquestos són los que més interessan al nostre punt de vista
positivista.

Pera basar lo sistema polítich social del particularisme en la llibertat,

no tenim cap necessitat de desentranyar lo concepte metafísich
d’aqueixa paraula. Deixem de bon grat que’ls teólechs moralistas
fassin esforsos de penetració intelectual pera agermanar la
responsabilitat humana, sanció de tot sistema religiós, ab la
omnipotencia y preciencia de la divinitat, sens las quals cap teología
tindria base. Deixem enhorabona que’ls filosops de la escola idealista
imaginin com únich subjecte digne de la llibertat, una abstracció que
jamay ha existit en lo món real, y considerin que sóls es
verdaderament lliure l’home que no está subjecte á cap de las
pasions inherents á sa naturalesa. Pera’l nostre objecte hem de
contar ab l’home tal com es, ab tots sos vicis y virtuts, ab sas pasions
bonas y dolentas, y dadas aquestas condicions, ferlo tan amo de si
meteix com sigui possible.

La llibertat es un desitj, un sentiment imposat al home per sa propia

naturalesa. Desde’l moment que’l home pot fer us de sas facultats
intel·lectuals, té idea de sa propia personalitat, y vol disposar d’ella.
La llibertat es una aspiració tan humana com la justicia, com la
bellesa, com cent altras del meteix ordre, y sa noció, com la de totas
aquestas aspiracions, va aclarintse, perfeccionantse y complicantse á
mida que las facultats intel·lectuals que’n donan esment al home van
cultivantse y progressant. L’objecte de qualsevol aspiració en l’ordre
moral ó intel·lectual es una de las forsas conservadoras de la
naturalesa, que com las que regulan la vida física, estan latents y no’s
mostran á la mirada del home, fins que aquesta te prou forsa pera
descubrirlas y aprofitarlas. En la naturalesa, per exemple, sempre ha
existit la forsa del vapor y la aptitut llumínica de la electricitat; pero
durant molts sigles han estat ocultas als ulls del home, que ni tenia
idea de que fossin aprofitables. De la meteixa manera han existit
sempre las forsas morals de que’ns ocupem, encara que no hagi
arrivat l’home á aprofitarlas fins que’l desenrotllo de sas facultats li
ha permés descubrir llur utilitat y ventatjas. Las manifestacions de
las forsas naturals son indifinidament variadas, y per moltas que
arrivi l’home á descubrirne, sempre’n quedarán de latents y ocultas á
sa mirada: d’aquí que la evolució de las ideas hagui de ser constant é
indefinida, aixis en l’ordre físich com en l’intel·lectual y moral.

Bastan aquestas indicacions pera explicar los distints conceptes que

s’han dat á la idea de llibertat en los diferents estats de civilisació y
de cultura que han atravessat los pobles. Pera l’assiátich, subjecte al
régimen de castas, la llibertat no vol dir mes que pertanyer á una de
las superiors, ab tot y que aquestas meteixas visquin baix lo jou del
despotisme. Pel negre de Cuba, la llibertat es sóls qüestió de color,
puig que pera ell, lo blanch no esclau es lliure. Los ilotas de la
república de Esparta consideravan lliure al poble de Lacedemonia, á
pesar de que estigués subjecte á la voluntat dels ciutadans de la
capital, puig que en ell hi veyan un grau menys de subjecció. A la
antigua Roma, durant molt temps, llibertat no volgué dir mes que
gobern republicá y abolició de la monarquía; sentit que per
reminicencia donan avuy per avuy encara á aquella paraula molts
sectaris, que ab tal de no viure baix l’autoritat d’un rey, no tenen
reparo en reconeixer com amos absoluts á un dictador ó á una
Cámara. La revolució francesa va confondre la llibertat ab la igualtat,
fins al extrem de que en un diccionari polítich se defineixi aquella
paraula dihent: “llibertat es igualtat; igualtat es llibertat.” Portada
per aquest afany, no es d’extranyar que hagui destruhit tots los
elements de varietat á que ha pogut arrivar en sa furia destructora,
establint damunt de las ruinas l’autoritarisme del Estat omnipotent.
Avuy encara tot Europa se ressent d’aquest fals concepte de la