J ' Shuu..

IR Bu<l>nJ�"'· '< Yi 15)

Consolidation Settlements over Tunnels: A Review

J.N. Shirlaw, J.R. Busbridgo and X. Y1

GolderAssoctates
Mtsslssauga, Onrano, Canadn

t roduction term development of settlement over tunnels

than there is for \hon term �ttlemem.
The prediction of the masnitudt" and ionn of
jt>ttlemcnt\ resulting from wnncllin� i� a major The development. )hapc, magnit u de and, to
OIS?CCt of the dcsiS•' of runnel� 111 an urhan <Ome extent, the cau�� of the immediate
environment. Prcdictivns arc u� to dctcm1inc: M:Jtlemcnt� are generally reasonably well
·tqUJrcmcnts fur the unck·rpinnin�. or other understood. The settlements develop in a
iOrms of protection. of adjac ent buildings. reverse S shaped curve: as t11e tunnel passes
Protection work� can fonn a \ignificant under the monitonng point. (Lo ct al., 1987).
,roportion of the co5t of major urban tunnelling Measurements made orthogonally to the line of
�o;hemcs, <o the accurncy of the l)rediction\ c:m the tunnel follow an 'error function' or Gaussian
'lave a <li;llificant effe('t on proJcx't cost curve. w11h the width of the curve directly
rclatccl to the d.:pth to the tunnel springlin e
PredictiOn!> for sculcut�:nt gcncmll) have �)me (l·igure 1).
empirical b:hh. using d.1t:1 obtained I rom earlier
unncls drt\tn under sunilar conduion\. The In general tenns it is known that long term
cmp•rk.U data can either be used directly a\ a sculem en ts can be significant, particularly when
t>asis for the prC<Itction. or as a \\ay oi tunnelling in or under compres.�ible soil�.
confirming the assumption< and �•mphfication� Shirlaw ( 1993) records examples of long term
mhcrcnt in numerical modelling. seltlcmcntc. where the long t= component had
the et feet of increasing the short term
�luch of the empirical clata h)r �ulcmcnt� over settlements by up tO a factor of ten, although
r only the p.!riod where th.:rc
tunnds cove<. "1lS this was unu�ual and mt�rc: typically the incre.,M:
active tunnelling in progrc<s. ·1hts data on the due to long term settlements has been in the
short term or 'unmcdiate' \ettlcmcnt forms the: order of 30% to 100%. Generally it has been
basis of lll0$1 prediction). It is l.nu\\n that th.:rc: ro::cordad that the effect of long term settlements
can be a long term comlxm ent or �ulcml!nt. hut has been to wtclcn the �ulcmcnt trough. It has
there is rdativcly much lc�s cldta on the long also been noted that the settlement. tf plotted on
a lo!!nrilhmic scale:. typically follows a straight
 1 BI.E I

E.�Qmpk." or lrnmt-di<ltr nnd Cun,olid.uion �1tlton'"'llts Rl'COrdcd On..- Tunnels

c..� 1\umcllinc Soli T>tw s. s :-

...
t. lrth
o d hn•n l nt�.&.
, urcd 11t n(un )

lS

f"'·
'..

"""Y 25 II)

•
<'3

53 �

(.0

l<irollt-h.Y. Arra� 9

.. 11<-.-c.
IL.t��>o�<.,. It.:Ki
' Tuoucf•l

•
Tt.an�b &crt�pot-.nl)' s.t,..,l)\·41 "'tb '"""' hl\olnJa:. "•h.�r en��.·n•• Nnn-.-1
- Nn•r �<"rno.-.11y C�t1.,.,hd..,<'-'·
- la$Qtcod..,.lt �-tdC'fD<DI
- Co. �.. �ttku;...,, ·
 �
... _.,, ..
eS.=J

•
...
•

.L

�-
figure I hupe nnd dt • t'lnpm('nt ur inunedintc nnd cortc;olidntlon �lllrmcnt
• (fumiS of
Consolirlatinn $('lllcuwn t nfler· nuh nc el al. 19911)

hnc. a good indication that th•< ..cuk-m�nt•� due value) hnmcdiately folio"ing tunneUrng to the
to conMllidato
i n. long term condition . II i� therefore imponam to
un dc:r)tand both the �:I teet of tunnelhn g induced
Ho\\t:\Cr, compared with thl! immcdrat�' :.c:ttle­ )tram� on pore: p re��u� and the �t.:ady �tate
rncnt tht're ·� a rclauvel)' �mall dat:r ha<e of condiuon after completion oi th�: t unnell ing .
field mcasur�mcnts ofC00'41 h d:uion :.c:ttlenlCrlts. The re�ults of �ome po� pressure mea�urcmcnts
or dc:tailed undcr�tanding of their C:IU..C\. fh� d unng t unnelling will be revJewcd, and th!
purpo)G of this paper i� to comptlt! <nmc ot th.: generaI pattern comp3rcd "� th the surf.tce
e'•idcncc on the nat ure :ltld dc,·cto�m�t?nt of ..cttlc:ment patterns. Theoretical work on the
consolidauon settlements. The maJorit)- of the pore pressure changes associated "llh tu nnelhog
evidence for conrolid:ni�111 around t unnel� will be brielly covered for co mparison with the
derives from long term me.,<urcmcnt� of surface ticld mc:a�u rc mc:m�.
sculcmcnh. 'l11e n ature: .md developme nt of
these sculemc:nts will theret(\ro! he r.:\lt\\ed \ln)!,nilude of Cnrl�lidat ion 5<-lllemetltS
first. am! gcncral pauern� c�tablbhed.
t..;xamplcs of mc:a�ured consolidation scHiements
Consolidation �ttlcmcnt� re�ult from incrl"'<rSed lire p rovided in Tahlc I. The mnj ority of thesc
effective �tr���- in turn due to reduction� in exnmplcs relate, not surprisingly. to tunnels
pore prc�urcs. Pore prc��ure m.:a�uremem� con�tructcd in near normally con)Qiidntcd clays.
arc therefore the b�t me.'n� of �tud� in g the GcncmUy the monitoring continued for up to a
cau:.c:� of con ohd:rtion around lllnncb.
� year alter completion of the tunnels. but thh
Strain� �)OCiatcd with tun ncll rng dir�.-�tl) atteet period wa� prohahly i n�uffic r c nt to measure the
pore prcs�ur� arourld the tumh::l: fvtlcl\\lllg full rnngnilllde nf the consoli d;Hion sculement.
con�t ruetion oi t he: tunnel the groundwater In many c-ases 'ICttlcment was still contmumg at
reg ime then changes to a lont-: tcrrn (St.:<td)' <.tate the time v. hen monitoring cca\Cd. An
seepage) cond itil'" · C'on�lid:uron �rtlc:mem.\ c.�cepllonal case i� that of the Grimsby tunnel.
result from the change: 10 pore pr�surc: lrnm the As rccorlled by O" Rc illy et al
{1991) monitoring
256 ConU)Iilhuon Sctllemcntt over Tunnd� A Itt\ 1ev.

continued for up to II years. and the r�:ad ing� con5idcrcd as a �parate trough superimposed on
stabilised after 7 to 10 years. the immediate �ettlement trough. Hulme et al
(1990J sugg�l>l that there arc at least three forms
There is some inconsistency in the defin11ion of that the cons<•liclation trough can lllke:
immediate and consolidation sc:ulement� a�
presented in Table 1. In some �� the Form I • due to seepage into the tunnel
consolidation settlement is ll1kcn a.\ all cau�ing a general reduction in
settlement measured from seven day� after groundwater levels out�id� the
tunnelling, in others only the �ttlcmcnt alter tunnel. The Form I scnlcrnent
tunnel decompression is included. The trough i� typically extremely
difference in the data is largdy due to the wide, and genernlly extends
nature of th.: published data. and probably tends beyond the area being monitored
to err on the side of not [ncluding all of the for scnlemcnt
consolidation settletnc:nt.
Form ':! - due to tunnelling at a pressure
Although the values for eon�olidallon \Cttlemcnt le�s than the tott�l overburden
given in Table I probably underMate the full pressure. The Form 2 settlement
value of the seulemcnt. lor the reasons given trough is typically significantly
ahove, these examples are <·a:.e� where wider than the immt'.diate
consolidation settlement was a concern. and sctllcmcnt trough
where significant movement� did develop.
These cases should not he considered as typical Form 3 - due to tunnelling at a pressure
of the behaviour over all tunnels. a� the procc:�s over the total overburden
by which they were selected iavourcd the pres5ure, and with a shape and
inclusion of ca�e� where co1t�olidation width �irniiM to the immediate
scrt1crncnts were rela ti\'dy high r.tther than scnlcmcnt troug h.
those cases where it w;b rclntivdy low. Dcsp11e
this caveat, it b clear from Table I tln\t These three iorms of the conwlidation
consolidation sculemcnts arc potentially a settlement tro ugh can he identified in published
significant part of the total scllkment due to easc.s. The full width of the Form l trough h
tunnelling. Undemanding the cau:...:' and rarely measured. simply because of its width.
effects of consolidation sclllcmcllls is therefore Within the rclauvcly limited selllcmcrt
unportant to be ahlc to minimise those monitoring normally undertaken. Form I is seen
movements and improve design methods. as virtually tlat. Shirlaw and Doran (198i·
record a case oi a tempornry stopp.1ge in a
Shape of the l..;ttt't'a1 s<·llll'nu·ut t rough du(' In tunnel due to excessive tlow of water at the
Consolidation face. During the temporary stoppage
munitoring continued on poinLs installed ahead
It has genernlly been rcpMted that the ct'fcct of of the face. and senlemenb were recorded up to
consolidation selllement i� tv d�eJ><:Il �nd 180 m (30 tunnel diameters) ahead of the tunnc'
broaden the initial scll1cmcnt trough (Hurrell (Figure 2).
1985. Attcwcll et al. 1986). However. it h a�
been pointtd out by Shirlaw ( 19941) and Fujita
(1994), the consolidation sclllcmcm can he
 2.57

tunnc:ls varied �ignificantly. from almost

• • •
isotropic to highly ani�otropic. If the
con\Oiidation settlement was due to seepage into
the tunnel dcp rc��ing the general water table.
: l l• the w1dth of the lateral con\Oiidation tr oug hs

L.:.:1..�i
should al\0 vary significantly. In isotropic
cond•uon� the trough should be much narrower
than in anhot ropic conditions (Fitzpat ri ck ct aJ.
1981). The con$iSttnt pauern of consolidation
tigure 2 Settlement� ;llu·:td or tuuncl trough illu�tratcd in Figure 4 would sugge�>t that
race, Sin�:apor� l\IRT Con tnct these settlement s arc due to the dis_�tpauon of
J 05 (adapted fruu1 Shirln" nnd \lratn induced pmitivc excess pore pre))UI'el>,
Duran 1988) rather than seepage.

Figure 3, from Shirlaw.I99J. \u rnmanscs a

number of
consolidation settlement trough� L>e•·elop111cnt of Consoliclnl ion Sett lements
11 it h Time
' nor mali�cd and
measured over lllnn e ls in c ia',
compared with a typical immediate �ttlement
trough. The muncd•ate settlement trough shown It ha� been reported by a number of a uth on Lhat
h an error fu nction curve with i =0.5 depth to consolidation �uh:ments follow a straight line
tunnel spri ngline, a� rerom mendt<J for clay$ by when ploucd on a logari thmic time scale. 'fhh
Mair et al { 1993). It can be seen that i n the two has been reported for tunnc:ls con�tructed in soli
cases of tunnel� con\tr uc tcd hy Earth Pressure clay mtng co mp rc�\Cd air (Giossop 1980), for
Balance Shield (!:PH) with ��� execs� face a lllnnel 1n �ill (Morton and Dodds, 1979) and
prc.m�rc (Singapore MRT and l'umngJian�). the for runnel� in hard over con solid ated clay
consol idauon settlement trough hM a �umlar {Forbes et al.. 1994). A slightly di ff erent
form to the typicaltmmt>diate $<!tllemcnt trough. pauem was reported for both comprc.�scd air
This confirms the Form :l consolidation and EPB tunnelling i n m arin e clays 111
�ttlement trough. Singapore. where: Shirla" (19941) and Hulme ct
aJ (1991) report S<lttlement/log time cun·es
The lateral consolidation trough fur tht! which have the chamcteriMic reverse ·s· shape
conventiunally dnvcn tunnels, shown in rigure of a 'C.:3 consolid ation curve. as dtfined by
•

3 increa�) m proporuon to Lhc tunnel diameter, TcrL.aghi and Peck (1967). The differenC'!
rather than the depth of the tlllllh!l. In Ftgurc J bet\\<!eO Log/linear and 'C.:3' is not signi ficant
the con�olidation \cttlcm c nts for the cJ.Cept in the early and la1t stages of Lhe
conventionally dri,·cn t unnels has been plotted consolidation settlement .
with the X axis normah� agamst tunnel
dtamctcr, rather than depth. It can be )Cell that
the curves approxunatcly follow an error ron! Jlrrs�ure Chn n�:cs ;\!;SOciat ed wit h

function curve wi th i. the distancc to the poi nt 'J'u n nell in �

of inflection. equal to 2.5 tunnel diameters.
There has been a tendency w a..cnhc thi� type There:: ha) been r e latively hule puhh�hed
of settlement to the cffccl.li oi .sctpage i nto the:: evidence on the pore pressure changes that arc

tunnel, or to the de\\-dtt:ring eff«h of the direct cause: of the consolidation settlements
compre-;.<:ed air. The ratio of horz i ontal to discus� aho,·c. To truly identify both Lhc way
vertical per meabili t y of the w1b fur the)�: the different tunnelling activitie� affect pore
 Cvmuhduuon Seule�nl\ o\·cr lUnncb• \ Wt\11:'\\

015T�CE FROII CEJITREU�E

112 DEPTH TO TIJN� EL SI'RI�GUkE

0 2 • s ' 7
� 0 ' '

.... §
a ...

02

�� 0. 4
� ....
.... ...
..,�

.. � G.'

��
5!�
o.a
()�-;.�s.n�L.M l!.:ltt.'.J)"'D)
g� lO � �'� :.ki:J .,'-: ·J.l'!·l1l. £N, u•2.1..
':: hU.7' '-'l JJtW .. t.r..)J._
� ..

..l'lts,&0.U:UT• I)"L\,..�,)CO.�
� l2
....
14' "111-H'" .u;ou,r6f, 0'0SittCl' • tcWit[S.:.CD A& ,1.00. �-
• <;�;tot"t �
'!
..' 11 *:U CC otl"\�IU'
(:.nf.��t' I <I )I)It �f.,-.\ Tl.l\.«ll
A :JIIn.\C•'"lJ�VtC3:,llW6(.�JS;t.).._�
• c,._ :u.. •• \ T\J\_1(\.,

.........(,lJUUCJ'.'{ • TH 1•�1 • tEJ-·il�SP� ..a.•

Figure 3 Con.solidnlion St'lllcml'nls. with cli!>tance norruali:r.cd by tunul'l dCI>lh (afH•r Shirl:w
1994)

DISTAHC£ fROII CENTRfUNE

SHIELD DW!ETER
0 z ' 7 •

0
*
O.t 0 •

0 ,4

0 ( ·':o:l' •ilae I • 2JD

0.
O __ TliUH!l£R 1!.4Y ARRAY I.
0, 1

«---'R! II.IBY lARA'/ A.

A � SU!JWA'I UJ« TU4·100.

t::._ /o\PA!I IDIUI JIWI'Cll 82·100.

Figure 4 Consolidation selllemrnts, "ilh distance uonnali'lccl b�· tunnel diameter.

 J N l>IIJri•"· J R Ru•Ntd�c. X Yl

prcssurc:s and the lateral extent of the pore condition, wa.s the rc:sult of a complex sequence
pressure chang�s. it i � nccc��ry to install a of chanscs res ultin g irom the various stage�
number of pictomcter� capahlc of mpid nwolvcd in shield tunnel li ng.
response. and to moni tor th�m rrcqucn tly a� the
tunnel progresses. Tiu� was done d u rin� the The l'uron&,jiang tunnel as one of the e>.amplcs
monitoring of the Furongp ang tunnel 111 C:hi na u(Cd to demon�trate in Figure 3 that the lateral
(\'1 et al., 1993). and the result� provtde a good con�olidation �ellle m e n t trough for
illu.mation of the elfccts of a complex tunnelling ovcrpress u ritcd EPU tunnel lint has a completely
proce-�\ {in thb cnsc EPU tunnelltng) on pore d1ffercnt fo rm to that over conventionally driven
pressure change�. tunne ls . Because of the difference in th e shape
of the \urfacc settlement p rofile it is likely that
Figure 5 show� the r�pon\e of Pic1ometcr No. the di�tribut10n of pore prt:S\ur c chang� around
60. placed at Spring li Qe levo:l 0.5m from the •conventionally" driven tunnel� has been
extrados of the: tunnel. to tunnelling. It Cl\11 be different from the trtangular pattern of positive
�'Cfl that the piezometer rcactc:d to the excess pore prc-;.\urc me�ured at Furongjiang.
tunnelling i n a complex JMtlc!rn that l"dll be
�i mpli fi ed as: lllcasuremcnt� of pore pres.(ure changes due to
t unnelling as extensive as those of Furo ngjiang
• Ini tial ly, a negative cxet·s� f)I\�(\Urc in arc rare in the publisheld hte rarure. However.
response to the itpproaching. m<�dune mca.surcmenb oblllincd from a ln i e of
piezometers placed vrthogonally [O the tunnel
• The-n a po�itive prc�ure as the
Cl(CC:.\\ drh e have been recorded for a number of
face approachc!tl and passed. in rL�ponsc tunnels. including:
to face pressure at above total
overburden pressure. The Thu nder B�y �wcr tu nnc:l 1n Canada
(}>-.timer and Bclsha", 191\0, Ng et al. 1986)
• f-ollowed by " reduction to �))()ut LCro I he East Oelfa�t !.ewer tunnel in N. Ireland
cxCQS prc�\ure. itS the tall of the (Glo\<;O() et at., 1979)
maclunc p3.\'>t.'d the pic1omctcr .lltd the A tunnel in Doom clt�y (lllatr and Taylo r, 1992)
tail void closed.
The re�ults of the piezometer measurement.\ at
• Thc:n au inc rease: w po\ttive excess the� three t unnel s as �ummnriscd in Fi gure 7.
pressure. prohahly dut" tn �routing In each ca� the origiml.lly published data ha)
around the tunnel linmg. I>L-en ad apted. w1th the distance from the tunnd
no rmalised by the tunnel diameter. All of the
• Finally. di ((ipauon of the po�itt\ c e.>.cess measurcmcn� at th� three sites. and at the
pore pressure rc�ulting tn a final Furonruiang sne recorded in F1gure 6, w�::re
mc:a�urcd pore prc:s�urc clo!oe to the obHtined fmm pneumatic or vibrating wire
101tial value. pletometer� . except the three outer reading s at
the Thu nde r Ray sewer tunnel. where standpipe
The: distnbuti o n of c-Xl't:S) Pl're prc��urC.\ with pic7o mcters were installed. The u � of three
di�tancc from the tunnd, for each of these fi,e slower r�ponding standpipe pic:1.orne�ers may
�tagcs. is )ho"n tn Fig ure 6 It can be x-cn e.>.pl;un the apparently dda)c:d respon\C to
that the pore pr�wre at fourth �t:1gc. when t he tunnelling compan:d wnh the thrc:e inner
dtrect effect' of the tunnelling ''as co mplete but pictometcrs.
pnor to final c.lhSIJl:lhon to the stc�dy 'tate
2�)

'0

No.60
•

20

0
AU,U$l
------C1(.j'' ·I·
SEPIEMB(R PO IIT I O. or "(lO�
tiiRJ
__ _,
tJ z • l6 u lO Jl 2 • ' e 10 12
-IU ·I�� I.I .u ·II ·U 14 ·l.l ·U 0 U ll tJ IU �· ""l.'CtflOIIPlllOOillLliO
lllllOfact I,.)

A C I< II I i I'I EZOME !( It

At i'!IO

IISlltUIIEli!(O SECTION

l•lgurr 5 Pore pre!.-�11 re re.�pou�r 1 o lunnclliug, Piezomeler 110, Ftwougj ian�,; (nfler Yi cl n I
1993)

.
!
,
] (!)

CD'-"'' •._
1'\IWL :;:ttl!l.,;.JrOI
; •
� C c.wttP
!' .. • u ..

'
!
....

d
l
,. '0 lh 41 lh .,, lh il)

Fi gul"\! 6 Chnngc iu por·l' J>ressu� dur 10 lunnellinj!. Furon�jinng (ad:.pled for Vi cl a11993}
 J.� Shirlow. J I< Bu.<l>rid�<· X Yi 261

(t)
DISTANCE FROII. C EH TRELINE
S H IELD DIA Y ETER

2 l • s

"]
,_ '

�
�

!i ·l�
0
�
:J .U St41(LDPWto
�
�
"' /\ >l><ts MT(R S>!JQ.D 'I.SSU)
.. .
�
0
� -
� 0 II D.OYHHEII ...
UD,A>S£0
�
�
<>
><
�

i�

tl

DISU.'iCE FR® CE_Jm£UJ£ DIS1UCE FROII CEKIAEUM£

---sii ELO 01Al!£TUI EXC.WATEO OWCEII
•

"]
si
- ..
•
t: . .
•
'
0
. .. 1 '. 1
.
E
� "

• ••
::'
.. .
I!
�
,.� ..
• t.a•C .S:ILUD

00 Of t-CUfiiXTK.•
"
•
•
·-
;;: u.
,
..
•
� cJ

Figure 7 Change in pm·e prrssur� measured at a) Thunder Bay Sewer Tunnel (adapted from
data in N1: ct al. 1986) b) Dtsl Belfast Sewer (adapted from Glossot> ct nl, 1979),
c) Boom <.: lay tunnrl (nclaptcd f1·om �Jail' nncl Taylor, 1992).

Of these measurements. the ones for the East The pit:70meter measurements for Array 2 in
Bel fast sewt:r can be auributcd to dnnnagc 111tn Thunder Bay arc for an area where relatively
the tunnel. The piezometer mc�uremcnts given lillie consolidation seul ement occurred. There
1n Figure 7b relate to a period when the were significant consolidation sclllcments at
tunnelling was temporarily stopped, du.: to a array I over this tunnel (see Table 1), but much
labour di�'j}ute. Dunng this penO<I the rae.: wns lc� pic1.0metcr data.However, the pattern or
simply boarded, and the comprt::Sl."Cd air turned the pore pre�ure changes recorded at Array 2
off. h panicuhtrly inter�ting. It can bt: l>t!t:n that
262

the paltem of excess pore prc$.�urcs after driving �ft clay has been studied by. among others,
the tunnel wa.� very different to that n:coru�t at Schmidt (19&9), Mair and Taylor (1992) and
Fur ongjian g. '11tc c�tcnt of the positive excess Samarasckara and �iscn�tein (1992). The three
pore pressures, to over six tunnd uiameter� Mudics \how quite different patterns of pore
from the tunnel centre-line is reasonably pr�ure response. Schmidt shows a pattern of
consh1ent with the wiuth of the Form 2 nega tive exce� pore pre��ures clo�c: to the
settlement protile as shown in r-igure 4. Lon g tunnel. wuh positive excess pore pressures
term measurements of 'steady state' piezometric further away, similar to the pattern mc.asurcd at
levels have been repo rted for the Grimsby the Thunder ll<ly wnnel. Mair and Taylor show
tunnel (II years) and Thunder Ilay tu nnel (-l a pattern of nega tive excess pore pre�ures,
years), and showed pore pre��ures similar to similar to those measured at the in !3oo m clay
those prior to tunnellin g. indicating little or no .
tunnel. Samar1sekara and Eisenstein show a
effect on pore pressure� due to �page into the paucrn oi excess positive pore pressure�. similar
•

tunnels. to tho!JC measured at the Furongjian g tunnel.

While all three solutions show one of the
The licld measurements of excess pore pressures pattern� iden11fied in the licld measurements, for
around the soft clay tunnels �ummarised above the same basic assumptions the three solutions
support the conclu �ion . hascd on sur face arc mlllually contradictory.
scttlemcnL�. that there arc at least three different
patterns of pore pressures that can rc.�ult from ror the �tud y of the effects of seepage into the
tunnellin g, and thn:e difft:n:nt forms or tunnel. Fit 7.pa trick ct al. ( 1981) have puhlishcd
consolidation trough related to those pore a simple pa rame t ric � tud y. This is very
pres�ure
. changes. informative. particularly on the effects of high
permeability path� on !low patterns around
tunnd�. They show that the presence of a
Tl•eo•·ctical Work on Pn•·c J•··��urc Chnn,::cs single:: lens of material one hundred rimes �

lnduced by Tunnelling pcrmcahle as the main soil ca n res ul t in pore

pr�sure changes as far as 20 diameters from
from the empirical evidence presented above. the tunnel. This corresponds well with the
both the pore pressure response to tunnelling observed sc::ttlements in anistropic condition�
induced strains and the linal groundwater shown in Figure 2.
regime around the tunnel affect the m�gnitude
of the consolidation settlement due to tunnelling.
Tlt corc tical
work on both of these a�()L"'Ct�
should therefore help in understanding the
causes of consolidation sc:ttlemt:nt�. The evidenc-e presented above mongly sugges:�
that there arc at least the
r e ha.\ic patterns o·
for most tunnds the dominant eifc:ct of the pore pn!$'ure ch�ngc.\. and associate.;
tunnelling on the surrounding soil is a pattern of consolidation settlements, due to nmnellin�
inward yield ing , (Lee et al 1 992) . either due to White the theoretical ha.�is for thc.�c pattern� s
movement at the face (typically for NAT.\I and not fully devdojJ<!d. som e genc:n1l conclu sions
hand excavated tunnel�) or at the ta1l�J.in or the can he made on the �oil s in which �uch chango
tunnelling machine (TBM and EPB driven occur, and on the factors controlling L�
tunnels), or a c.:ombinatl\111 of the...: t\\V (l.lJJCn mag nitu de.� of both the pore preSl;ure chan�o
face shield mnncls). The effect of a simple , and the settlement�. Generally, the ma gnituue
single Mage of inward movement for tutmd� in of t·onsolidation sc:ttlemt:nts wilt be contro'at
 J ' �hul�w. J.R. Rt.,hfld�. X Yl

both by the magnitude: M the chao:;t in pore hydrofracture the >Oil, ""hich is typacally
pre.��ure and tht comprc�thtllly oi the 'mi. rdathcly low in nonnally con�hdatcd clay.
The penncabiliry of the <;Oil contruh the rate of
change of the pore prt'>.�urc�. Thl) ttnpMcs a The available evidence suggests that the
limit on th e type of :.oil ''here oon:.ohdnuon m:�gnuude of Form 3 ;enlement •s proportional
:.eulemcnts can practically be :.epnrntcd from the to the immcdi<ue �ttlemcnt sa'ed by over
immediate seulement!>. For reawn> of pre��urising the face or grout. In other words.
compr��ahility and permeability. �·gnificant pushmg harder may reduce rmmcdiatc
conSC�Iid auon scnlcment> arc mo>t h�el� to �ttlcments at the expense of incrca.�ing the kmg
occur where tunndhng occur> 111 or below term �nlcmcnt>. The precise relauonvup will
normally consolidated clay>, "hatcvcr lorm the depend on thc: magnitude both of the pore
conwlidation senlemcnts take. Some of the pre�sure parameter and the compn:�sihility of
panicular factors aff�ctwl� the inti" adunl form; the S<•il.
arc:

J·orm I in' olve� dramage rnto the tunnel and Couclu�ions

therefore isa functt<'ln of the relative
permeability or the lining e�nd the �oil. Con�lidation settlements are potentially a
significant source of the total settlement over
Form 2 invoh es the gcnt:nstinn oi exec:�> pore tunnels drhen through or under compressible
pr�urc� at some dtstancc from the tunnel. For soih. The changes an pore pressures associated
this to ue<:ur. the re;pon....: of the soiI II• ch31tgcs "tth such �ulcmcnts can he complex, and are
in shear stress muq b.: to gcncrnre JXI>IIJ\'C not fully under>tood. From the available data
excess pore pressure>. The relatiun�hips on both :.eulemcnrs and pore prc.ssure changes
developed by Schmidt sugge.�t thilt the it is s uggested that there are at least three
magnitude oi the po�1th e excess pore prc\�uncs. different patterns of both SCHlcment :md pore
and therefore the \CIIIcmc:nt. i� dircctl) pre»llrc change associated with tunnelling.
controlle<l by the magnitude llf utward Bccau>t the three dafferent patterns have
movement dunng tunnelling. Thas iunhcr chff.-:rent causes, the appropriate method to
suggem that the magnatudc oi the Form 2 minim1>e the con\Ohdation sctdcment� is
con:.ohdntion scttlcmc:nt� i� dirc.:tly prnllMiional cliffcrcnt for each of the three forms.
to the magmtudc of the ttnmcdiatc C ) llh:ments.
although also dependant on the m�gnitude both
of the pore pressure p.1rnmctcr and the
compressibtliry of the !>Oil.
All\\t:ll. P.R., Ycal��. J. and Selby. A.R.,
rorm 3 invol ve� tile genemlton of pos111ve 1986. Soil Movement� Induced by Tunnelling
excess pore prc:��urcs around the tu nnel due to and th.:ir Effccb on l'ipclincs and Structures.
over pressu ring the: face and/or the grout nt the Blackie. Gla�gow.
rail void. For this to occur. the n:�pon:.e of the
soil to undntinc:d loading mu>t be to �enemte Oc:hha". D.J. and Palmer, J.H.L. 1978.
poslll\e excess pore prc.��urcs. I Ius type of Rc�1tlt' of a program of anstrumentation
respon>e is typically a'>.<;Oeiatcd with ncar anvolving a precast scgmemed cMcretc-lined
normally consolidated clay>. For shallow tunnd in clay. Canadian Gcotcch. J, 15. pp.
tunnels the amount ol J){lsith·c excc:s� J)('lrc 573-583.
pressure b limited by the pre't\un: rcq111red to
21>4

Cater, R.W Shirlaw. J.N., Sullivan. C.A &

.• llan>"· T. 1977. Ground movemcnb due to
Chan, w:r 1984. Tunnels con�tructtd for the constru..:uon of h
) 1cld dmen runnels. Proc. 9th
Hong Kong Mass ·rransn l{.ulway. Hong J.:onr lnt Conf. So1l �tech. and Found. t:ngmg.
Engineer, October, 37-49. TOJ..}O pp. 759 10 790.

Ftt.tpatricl... 1.. . Kulha....y. F H.. and O'Rourl..e, Hulme, I.W., Chapman. T'.H J.. Pol.. Shcuog
T.D.. 1981. Flow pauems around tunnch and Foo, Cop�y,J.P.. Kr.tlt. 8., Snpathy, P .•

their usc tn evaluating oon�ruction problems. rouer, L.A C.. and Shirla..... J.N. 1991.
Soft-ground tunnelling. failure) and �•ngapore Mass Rapid Tran)it Sy)tem •

displacements, i:d. D. Rc�ntht and M.l'. Homo l>1SCU�)IQn. Procccdmg�. lnstuuuun of Ctvil
Publ. A.A. llalkcma, Ronerdam. p p. 95-103. F.ngin.:ers. Part I. 90. April, pp 46:\ 502. •

Forbes. 1. . uasscn. R.H .. and I atham. 1\I.S. l!ulme, T.W .. Slurlaw, J.N. and Hwang. R.N.
Monitoring and lntc:rpretauon oi Movement of 1990. Sculc:mcnt) during the underground
the Man�•on due 10 tunncllin1'.
House conwucuon of the StnJ:aporc MRT
Procccxhn&s of ln ,tll ution of C1v11
the: P�'dmg� of the tenth Southeast ASlan
F.ngmeer�. GMuxhn1cal Fnginccnng. Apnl Gootc:chntcal Conference, fa1pe1. Republic of
1994, pp. 89·98. C'hma. pp. 5�1-5�6.

Fujita. 19<J4. Soft ground tunnd hng and Hurr�ll. M.R. 1985. The empirical prechcuon
buried structures. Procccthn�s XIII ICSMH.- of long term �urt'a� scttlc:mcnts above �•eld­
1994, New L>elht, India, puhl. 0\tord IBH dnvcn tunnels m soil. Procc:edmJ:) of the 3rd.
Pub!. Co. Pvt. Ltd .. Vol. V. pp 89 • 101!. International Confc:rence on Ground Movements
and Strucrur�. UWIST. puhl. Pentech Pre�\.
Glossop, N.ll., Saville, l>.R .. Moore.J.S. . Vol 3. pp 161 172.
•

Denson, A.P. and farmer. I.W, 1979.

Geotechnical a�pcct� of shallow tunnel Lee. K.M., Rowe. R K. and Lo , K. Y
conMrucuon 1n Uelf;m F.�tuarinc !Xpo�ll\. 1992 Subl.tdcnce O"mg to tunnelli ng. I.
Proceedings of 'Tunnci!J nj; 79'. London. U.K fuumating the gap p.trnmeter Canadian
pubI. lcutitution of Mming dlld Mc:tallurg�. pp G(Otcchmcal Journal, 29. 9�9-9J0.
45 • 56.
Lo. 1.: \\ .,14"C. S.T.., Mal.. 1no. H .. Ch.lng.
Glossop N.H. 19SO. Ground dctonnat1on l.K.. Leung. C. F and Mihara. T. 198:
caused by Tunnelling m o;oft ground at Grim�hy. Tunnch 1n clo:.e proxinuty. Prtx·ccxhn g� of the
lntemal Rc:port. Univer)lly of Newcal>tle upon Siny.tpore Ma\5 Rapid Transit Conference.
Tync, Department of l\lmmg Fnginc:cring . April 1987, puhl. lnsutution of Eflgincer),
Singapore, pp 275 · 282.
Glossop N.H.
and O'Heilly M.P. 191!2.
Seulcmcnt caustd by t unnelling through soft l\lair K.J. and Taylor, R.N. 1992. l'rcdJCUOII
marme clay, Tunnds and Tunnelling, Octohcr or clay behaviour around tunnc:l� U\1ng pla.�ticit}
1982, pp 13 • 16 �IUII{liH. Proc Wroth l\lc:morial Symposium.
.•

O�ford. U K l'llOm3) Telford.

Glossop, N.H. and 1-armer. I.\\'. 1979,
Seulement 3SMIC13lcd with the rc:mO\aJ ot Ma1r, R.J.. Taylor. R N. and Draccgirdlc:. A.
compi"CSM:d air durtng tunnclhn� tn ••lluvli!l 1993. Subsurf;.cc settlement protilc� aho\e
clay. Gcotechnique 29, No. I, pp 67 7'1.

---························
 J N. �hlrl>��>', J K llu<hrod8< X Yl

tunnels in cia)•�. Cicmcchnique J.3. No. :!. Jl5 • Shirlaw. J.N. 199-11• Ob!�erved and calculated
320. pore pressures and deformations induced by an
earth prt:-)wre balance �hidd: Discussion.
Monon, J.D.• and Dodd�. R.R. 1979. Ground Canathan Gcotcch. J. (in pre�)
)
�ubsidcnce a�iated with machine tunndling in
I'Juvio-dcltaic scchments - Part 2. Tunnels and Shirlaw, J.N. and Doran, S. 1988. Ground
funnclling, November, 1979. movcmcnLS and settlements caused by tunnelling
for the: Singapore Mnss 1'ransil System.
Ng. R.M.C., l..o, K.Y.. and Rowe. R.K. 1986. Proceedings. International Symposium,
Analysis of F1cld Performance The Thunder • Tunnelling '!!!!, London. UK, publ. rnstitution
Bay Tunnel Canadian Gcotechnic.t l Journal Vol. of M1nmg and Metallurgy. pp 295 - 314.
:?3, PI> 30 • 50.
. Tcrzaghi, K. and Peck. R.l'. 1967. Soil
O'Reilly, 1\f.f>., Mair, R.J. and Aldcrmann, mechanics in engineering practise. 2nd. ed.
G.H. 1991. Long term l>Cttlcmcnts over John Wiley and Sons, lnc., New York.
tunnel�; an eleven ycar stud) at Grimshy.
Proceedings, Tunnelling '91 Sympo�tUm, X. Yi. Rowe. KK. and Lee, K.M. 1993.
Institution of Mining and ;\tetallurgy. London. Ob)ei'V�;d and calculated pore prc.��ures and
lJ.K. deformt1tion� induced by an earth prc'\Sure
b:llance �hicld. Canadian Geotechnical Journal.
Palmer. J.H.L. and Bc:bhaw. D.J. I'JSO. 30. pp -176-490.
Dcfonnauons and pore pres.,urc\ m thc 'icinity
of a precast. concrete-lined tunnel in clay.
Canadian Gcot.:chnical Journal, 17, pp 174 ·

184.

Samara.�lrcra, 1.. and Eisenstein Z. 1992. Pore

pressures around tunnd� 111 clay. Canadian
Geotechnkal Journal. 29 pp. 819-831.

Schmidt. B.. 1989. Consolidauoo Settlement

due to �ft ground nmnclling. l'roc. 12th Int.
Conf. on Soil Mcch. nncl found. t:ngrg. . R1o de
Janerio. Pub!. Sal l-ema. Volume 2. pp 797-800.

Shirlaw. J. N. I 993. l'ore prc�'ure� around

tunnels 111 clay: Di�us.�on. Can. Cicntcch. J.
30. pp l(}.l4 1046. •

Shirlaw, J.N. 1994'. Subsiderlce �1\\ing to

tunnelling II, evaluation of a predictiOn
tcchnique:Discus)ion. Canadian Geotcch. J. (in
press)

View publication stats