water
Article
Failure Analysis and Treatments of Tunnel Entrance Collapse
Due to Sustained Rainfall: A Case Study
Long-Long Chen 1 , Zhi-Feng Wang 1,2, * and Ya-Qiong Wang 1,2
Citation: Chen, L.-L.; Wang, Z.-F.;
Wang, Y.-Q. Failure Analysis and
Treatments of Tunnel Entrance
Collapse Due to Sustained Rainfall: A
Case Study. Water 2022, 14, 2486.
https://doi.org/10.3390/w14162486
Academic Editor: M. Levent Kavvas
Received: 29 June 2022
Accepted: 9 August 2022
Published: 12 August 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Water 2022, 14, 2486. https://doi.org/10.3390/w14162486
1 School of Highway, Chang’an University, Xi’an 710064, China
2 Key Laboratory for Highway Bridge and Tunnel of Shaanxi Province, Xi’an 710064, China
* Correspondence: zhifeng.wang@chd.edu.cn
Abstract: Rainfall is a crucial issue affecting the entrance slope stability of mountain tunnels, as
it decreases the shearing strength of soil and reduces the stability of tunnel entrance. This paper
presents a case history of the collapse failure of a tunnel entrance in Yunnan Province under rainfall
conditions, in which the failure mechanism and potential factors and treatment measures were
discussed by field investigation, theoretical analysis, and in-situ monitoring. The analysis results
show that the decrease of soil shear strength was mainly attributed to the decline of matric suction
value of soil caused by the increase of soil water content. The decrease of the soil shear strength
reduced the sliding resistance of the entrance slope and then triggered the collapse. Based on the
results, three treatment measures to prevent a secondary tunnel entrance collapse due to rainfall
are adopted, including anti-slide pile, grouting, and slope reinforcement. Combined with the field
monitoring data, the effects of treatment measures were investigated. Lessons in this case study
facilitate prevention and treatment of tunnel entrance constructions under rainfall conditions.
Keywords: tunnel entrance; rainfall; collapse failure; shearing strength; treatment
1. Introduction
Railway tunnels are playing an increasing role in mountainous roads with complex
terrain [1,2]. However, the entrance section of highway tunnels is shallowly buried and the
surrounding rock is broken. In some mountainous areas with complex geological conditions
and poor hydrological conditions, collapse disasters at the cavern entrance occur frequently,
causing a large number of casualties and economic losses to tunnel construction [3–6].
There are many factors that cause construction collapse at the tunnel entrance section,
such as poor stratum stability, poor construction quality, and continuous rainfall [7–9].
Among them, rainfall is an important factor causing tunnel collapse, and most of the tunnel
entrance section collapse cases are caused by rainfall [10,11]. Therefore, it is important
to study the abnormal characteristics before the occurrence of rainfall−induced collapse
of tunnel entrance under complex geological conditions, and to analyze the causes and
mechanisms of collapse for the prevention and treatment of tunnel entrance construction
collapse under complex construction conditions [12–15].
There have been many existing methods for the prediction of tunnel construction
collapse under adverse geological conditions, all with their own focus and advantages.
However, because the causes of tunnel collapse are not only numerous but also complex, it
is difficult for all these methods to accurately and timely predict the tunnel construction
collapse during the construction phase. Monitoring and measurement, as a common
method to determine the stability of the surrounding rock during tunnel construction, has
been adopted by a large number of constructors. However, since this method can only
determine the amount and rate of deformation of the surrounding rock, the damage to
the lining structure and the surrounding rock cannot be obtained. The main causal factors
of tunnel collapse occurrence are selected as evaluation indexes, and the risk assessment
https://www.mdpi.com/journal/water
Water 2022, 14, 2486 2 of 17

method can be applied to obtain the risk level of tunnel collapse occurrence [16,17]. Based
on the functional mutation theory and related damage criterion, the tunnel damage criterion
and collapse prediction model can be derived [18–20]. In addition, collapse mechanism
studies based on indoor model tests and numerical simulations can study and understand
the tunnel collapse mechanism and evolutionary process [21–23], and then make predictions
for collapse. However, the anomalous features of the surrounding rock or tunnel structure
that appear before the collapse of the tunnel entrance, such as longitudinal cracks in the
initial support, localized collapse of the excavation face, and slope spalling, are aspects that
are easily ignored in collapse prediction studies [24]. Moreover, these anomalous features
from the construction site can directly present the stability state of the tunnel and the
surrounding rock. They are of high value for judging the stability of tunnels and predicting
tunnel collapse [25,26]. So far, there are almost no studies and summaries on the anomalous
characteristics before the occurrence of construction collapse in the tunnel entrance section
under the rainfall action [27].
The mechanism of rainfall−induced tunnel collapse is also very complex. However, a
large number of studies on the analysis of tunnel collapse mechanisms have shown that
low strength of surrounding rock, the softening effect of water, and the synergistic effect
between geological formations are the main causes of tunnel collapse [28–30]. The collapse
of the shallowly buried section of the tunnel entrance is often sudden destabilization
damage caused by accumulated deformation of surrounding rock [31–33]. Therefore,
timely follow−up of the primary support and improvement of the bearing capacity of the
support are very important to circumvent collapse [34,35]. Research shows that the collapse
damage of the tunnel opening section is caused by the combination of the destabilization
of the palm face and the destabilization of the loose rock body in the upper part of the
tunnel [36,37]. The erosion and softening of the surrounding rock by the seepage action of
surface water causes a decrease in the strength and an increase in the self−weight of the
rock, which is the main cause of the instability of the surrounding rock [38,39].
After analyzing the specific causes of tunnel collapse, it is necessary to select a rea-
sonable and effective collapse treatment method to prevent the occurrence of secondary
collapse in a targeted manner [40]. The existing tunnel collapse treatment methods mainly
include: grouting and anchoring method, large pipe shed grouting and advanced support,
and collapse pit grouting method [41–43]. The main mechanism of these methods is to
improve the integrity and strength of loose or weak surrounding rocks and reduce the
deformation of surrounding rocks to prevent collapse, and the treatment effect has been
verified in a large number of engineering applications [44]. In addition, methods such
as changing the excavation method and improving the strength of the initial support are
often used in the treatment of collapse [45,46]. However, the location of the tunnel en-
trance is special, and the treatment of preventing collapse should not only improve the
strength and stability of the slope, but also prevent the seepage erosion of the slope by
continuous precipitation.
In this paper, a tunnel entrance collapse accident was analyzed based on collapse
observation, monitoring, and theoretical analysis, which aims at exploring the mechanism
of collapse under the rainfall condition. Besides, the damage characteristics before large-
scale collapse and treatment measures were summarized and analyzed. This research can
provide a reliable reference for the prevention and control of construction collapse at tunnel
entrance section during continuous rainfall.

2. Project Background
2.1. Tunnel Location
The Fada tunnel in this case study is located in the northeast of Yunnan Province in
China, as shown in Figure 1. The tunnel is a part of the Yibin−Zhaotong highway project
with a design speed of 80 km/h. The construction of this project began on 30 October 2017.
The left bound tunnel is from stake K161 + 739 to K160 + 983, with a total length of 756 m.
It connects the Heping Bridge and the Zhafang Tunnel, and the Haizi River is located
Water 2022, 14, x FOR PEER REVIEW 3 of 18

Water 2022, 14, x FOR PEER REVIEW 3 of 18

Water 2022, 14, 2486 3 of 17
The left bound tunnel is from stake K161 + 739 to K160 + 983, with a total length of 756 m.
It connects the Heping Bridge and the Zhafang Tunnel, and the Haizi River is located near
thisThe left bound
tunnel. tunnel is from
The elevation of thestake
tunnelK161 + 739
area to K160
ranges from+ 983,
1160with
to a1720
totalm,
length
withofan756 m.
elevation
near
It this tunnel.
connects the The elevation
Heping Bridge and ofthe
theZhafang
tunnel area ranges
Tunnel, and from
the 1160River
Haizi to 1720
is m, withnear
located an
difference of 560 m. It is a tectonically denuded mountainous terrain area with large topo-
elevation
this tunnel.difference of 560ofm. It tunnel
is a tectonically denuded mountainous terrain area with
graphic
large
relief The
andelevation
topographic
a relief
naturaland
the
slope
a natural
area ranges
of 20–60°.
slope of
The
20–60
from 1160
tunnel to 1720 m, with
◦ . Theistunnel
excavated
is
usinganthe
excavated
elevation
upper
using the
and
difference
lower bench of 560 m. It ismethod,
excavation a tectonically denuded
supported by mountainous
anchor−plate terrain area with
retaining, with large
a topo-
longitudinal
upper
graphic and lower
relief andbench excavation
a natural slope ofmethod,
20–60°. supported
The tunnelby is anchor
excavated−plate retaining,
using the upperwith a
and
slope of 2.5%. the
longitudinal slopeheight
of and
2.5%. the the width
height and ofthe
the excavated
width of the section are
excavated 9.59 are
section m and
9.59 m13.7 m
lower bench excavation method, supported by anchor−plate retaining, with a longitudinal
respectively.
and
slope13.7
of m respectively.
2.5%. the height and the width of the excavated section are 9.59 m and 13.7 m
respectively.

Figure 1. Location of the tunnel.

Figure 1. Location
Figure ofofthe
1. Location thetunnel.
tunnel.

2.2. Geological Characteristics

2.2. Characteristics
2.2. Geological Characteristics
When the
When the tunnel
tunnel was
was excavated
excavated to to 10
10 m
mfrom
fromthethetunnel
tunnelexit,
exit,aalarge
large−scale tunnel
−scale tunnel
When
collapse the tunnel
occurred.
collapse occurred. Thewas excavated
geological to 10
conditions m from
where the
the tunnel
collapse exit,
occurred a large−scale
are shown
shown in tunnel
in
collapse occurred.
Figure 2.
Figure The geological conditions where
2. It can be found that the slope of the tunnel entrance the collapse
entrance slopeslope is occurred
is close
close to 40°.. Besides, in
to 40 ◦ are shown
Figure
the 2. It can slope
the entrance be found thatto
happens
happens the
to slope
have
have of the
aalarge
large tunnel
amount
amount ofofentrance
silty
siltyclay
clayslope
with
witha is close toof40°.
thickness
a thickness 3 m.
of 3Besides,
In
m.
theIn
entrance
summer,
summer, slope
duedue happens
toto to have
thewashing
the washing aerosion
large of
anderosion
and amount ofdebris
ofrainfall,
rainfall, silty clay
debris flows
flowswith
andaother
thickness of 3 m. In
geological
geological
disasters
disasters
summer, are
areto
due prone
the to
prone to occur
occur in
washing this
this area.
inand area.
erosion of rainfall, debris flows and other geological
disasters are prone to occur in this area.

Figure 2.
Figure 2. Geological conditions.
conditions. (a) Strata at the collapsed site; (b) measured soil parameter profiles.
profiles.

According
According to to the
the geological
geological survey,
survey, thethe silty
silty clay
clay (S1),
(S1), soil
soil aggregate
aggregate (S2),
(S2), and
and mass
mass
rock
Figure and
rock2.and soil (M)
Geological all have
soil (M)conditions. a high rate
(a) Strata
all have a high of water
rate ofatwater content,
the collapsed
content, site;up to
up to(b) 40%. The
measured
40%. porosity of each
soil parameter
The porosity of each profiles.
soil
soil
layerlayer is also
is also high.
high. AsAs thethe sandstone
sandstone atatthis
thislocation
locationisisheavily
heavily weathered,
weathered, the
the uniaxial
uniaxial
compressive
Accordingstrength is less thansurvey,
to the geological 3 MPa. the The silty
mineral
claycomposition of the silty(S2),
(S1), soil aggregate clayand
was mass
rock and soil (M) all have a high rate of water content, up to 40%. The porosity of each soil
layer is also high. As the sandstone at this location is heavily weathered, the uniaxial
Water 2022, 14, x FOR PEER REVIEW 4 of 18

Water 2022, 14, 2486 4 of 17

compressive strength is less than 3 MPa. The mineral composition of the silty clay was
obtained by X−ray diffraction test, as shown in Table 1. The main mineral components of
red clay at the tunnel site are quartz and kaolinite, accounting for 67%. The mass fraction
obtained by X-ray diffraction
of montmorillonite test, as shown
is 14%. Moreover, in Table 1. has
montmorillonite Theamain mineral
greater components
influence of
on the wa-
red clay at the tunnel site are quartz and kaolinite, accounting for 67%. The mass
ter content of soil than other minerals. Therefore, the construction of a tunnel entrance fraction
of montmorillonite
section is 14%.
in rainy season Moreover,
is prone montmorillonite
to geological disasters.has a greater influence on the water
content of soil than other minerals. Therefore, the construction of a tunnel entrance section
in rainy
Table season
1. The is prone
mineral to geological
composition disasters.
of the silty clay.

Table 1. The mineral composition of the silty clay. Montmo-

Mineral Name Quartz Gibbsite Kaolinite Illite Goethite Anatase
rillonite
Mineral Name Quartz Gibbsite
Mass fraction Kaolinite Illite Montmorillonite Goethite Anatase
27 9 6 40 14 3 1
Mass fraction (%) 27 (%)
9 6 40 14 3 1

2.3. Precipitation
2.3. Precipitation
The climate
The climate of
of Zhaotong
Zhaotong City, where the
City, where the tunnel
tunnel isis located,
located, isis subtropical
subtropical monsoonal,
monsoonal,
with the southeast monsoon from the Pacific Ocean and the
with the southeast monsoon from the Pacific Ocean and the southwest monsoon southwest monsoon fromfrom the
the
Indian Ocean bringing large amounts of rainfall in summer, with an average
Indian Ocean bringing large amounts of rainfall in summer, with an average annual rainfall annual rain-
fall1110
of of 1110
mm. mm. Rainfall
Rainfall throughout
throughout the yeartheisyear
mainlyis mainly concentrated
concentrated betweenbetween
May andMay and
October,
October, accounting
accounting for 65.6–91.5%
for 65.6–91.5% of annual
of the total the totalrainfall.
annual Therainfall. The seasonal
uneven uneven seasonal dis-
distribution
tribution
of rainfallofresults
rainfall
inresults in frequent
frequent geological geological
hazards such hazards such as landslides
as landslides and debrisandflow
debris
in
flowmountainous
the in the mountainous
areas ofareas of the region
the region in rainy in season.
rainy season.
Figure Figure 3 shows
3 shows the character-
the characteristics
istics
of theofdistribution
the distribution of rainfall
of rainfall in 2018.
in 2018. It can
It can bebe seen
seen fromFigure
from Figure3a 3athat
thatthe
the maximum
maximum
monthly rainfall was 242.5 mm in August, and the rainfall was mainly concentrated from
June to September.
September. The distribution of monthly rainfall from June to September in Yunnan Yunnan
Province where the tunnel is located is shown in Figure 3b. It can be seen from the figure
Yunnan Province
that most areas of Yunnan Province have
have abundant
abundant rainfall
rainfall from
from June
June toto September.
September. In
addition, due to the topographic relief In this region, it is not conducive to the excavation
of the mountain tunnel entrance section section in in summer.
summer.

Figure 3. 3.
Figure Rainfall
Rainfall description
description in 2018: (a) in 2018:rainfall
monthly (a) of monthly rainfall
Zhaotong City of Zhaotong City
(https://www.qweather.com/
(https://www.qweather.com/historical/zhaotong-101291001.html, accessed on
historical/zhaotong-101291001.html, accessed on 5 June 2022); (b) rainfall distribution 5 June 2022); Province
in Yunnan (b) rain-
fall distribution in Yunnan Province (http://www.geodata.cn,
(http://www.geodata.cn, accessed on 5 June 2022). accessed on 5 June 2022).

3. Collapse Overview and Failure Mechanisms

3.1. Collapse Overview
3.1. Collapse Overview
The
Thetunnel
tunnelsite experienced
site experiencedcontinuous rainfallrainfall
continuous from 18from
to 21 August 2018.
18 to 21 On 212018.
August August
On 2018,
21
when
Augustthe2018,
tunnel
whenface the
of the left bound
tunnel face oftunnel was
the left nearlytunnel
bound 10 m away from the
was nearly 10tunnel
m away exit, the
from
entrance
the tunnelslope
exit,was
thescoured
entrance byslope
continuous heavy rainfall,
was scoured and a large
by continuous amount
heavy of surface
rainfall, and a water
large
seeped into the slope. Figure 4a shows a photograph of the tunnel before the
amount of surface water seeped into the slope. Figure 4a shows a photograph of the tunnel collapse. On
22 August, a large−scale tunnel collapse occurred, and the tunnel structure was completely
damaged. The tunnel was buried by the collapsing soil above, as shown in Figure 4b.
According to the investigation on the collapse site, the length of the collapse landslide
 as shown in Figure 4b. According to the investigation on the collapse site, the length
the collapse landslide along the longitudinal direction of the tunnel was about 27 m, a
Water 2022, 14, x FOR PEER REVIEW 5 of 18
the collapsing soil was mainly silty clay.

Water 2022, 14, 2486 5 of 17

before the collapse. On 22 August, a large−scale tunnel collapse occurred, and the tunnel
structure was completely damaged. The tunnel was buried by the collapsing soil above,
as shown in Figure 4b. According to the investigation on the collapse site, the length of
along the longitudinal
the collapse landslide direction
along theof the tunnel was
longitudinal aboutof27the
direction m, tunnel
and thewas
collapsing
about 27soil
m,was
and
mainly silty clay.
the collapsing soil was mainly silty clay.

Figure 4. Photographs of the tunnel: (a) before collapse; (b) after collapse.

Before the large−scale collapse, many obvious abnormal phenomena appeared in t

tunnel site, such as primary support cracking, local collapse of tunnel face, and cracki
Figure4.4.Photographs
Figure
and spalling Photographs ofthe
of
of slope protection. thetunnel:
tunnel:However,
(a) before
(a) beforecollapse;
collapse;
these(b)(b) aftercollapse.
after
abnormal collapse.phenomena did not attract t
attention Before
of construction
Before the
thelarge
large−scale personnel.
−scale collapse, many obviousobvious abnormal
abnormal phenomena
phenomena appeared appearedin inthe
the
As thesite,
tunnel
tunnel surface
site, such
such as water
asprimary continues
primary supportto
support flow into
cracking,
cracking, localthe
local slope,
collapse
collapse of the
oftunnel
tunnelsoil water
face,
face, and content
cracking rose ra
andcracking
and
andspalling
idly. The shear strength
spalling of
of slope
slope protection.
of the soilHowever,
protection. was also
However, these abnormal
weakened
these abnormaldue phenomena did
didnot
to the increase
phenomena notattract the
of water
attract the conte
attention
attention
resulting of construction
of construction
in local personnel.
collapse personnel.
of the tunnel face. Figure 5a shows the local collapse on t
As
As the surface water
the surface watercontinues
continuestotoflow flowinto
intothethe slope,
slope, thethesoilsoil water
water content
content roserose
rap-
upperrapidly.
bench of the tunnel face at K161 + 739 before the large−scale collapse. The cracki
idly. TheThe shear shear strength
strength of the ofsoil
thewas
soilalso
wasweakened
also weakened due to due to the increase
the increase of waterofcontent,
water
of initial support
content,
resultingresulting
concrete
in local in local also
collapse collapse reduced the Figure
of the face.
of the tunnel tunnel stability
face. 5a of the
Figure
shows 5a showstunnel
the local
structure
thecollapse
local collapse during co
on the
struction.
on theThe
upper upper
bench development
bench of longitudinal
of the tunnel
of the tunnel face at face
K161at+ K161 +cracks
739 before739 the indicated
before that
−scale
the largecollapse.
large−scale thecollapse.
distribution
The The
cracking of su
cracking
rounding of initial
rocksupport
of initial pressure support concrete
had also
concrete changed, also reduced
reducedresulting the
the stability stability
in of of
insufficient the tunnel structure
bearing during
the tunnel structure during
capacitycon-of the p
construction.
struction. The The development
development of of longitudinal
longitudinal cracks
cracks indicated
indicated that that
the the distribution
distribution of of
sur-
mary support. Before the large−scale collapse of the left bound tunnel, many cracks a
surrounding rock pressure had changed, resulting in insufficient
rounding rock pressure had changed, resulting in insufficient bearing capacity of the pri- bearing capacity of the
peared on the
primary primary
support. support.
Before The maximum
−scale length ofbound
longitudinal crack reached 3
mary support. Before thethe large
large−scale collapse
collapse of
of the the
left left
bound tunnel,
tunnel, manymany cracks
cracks ap-
and the maximum
appeared width of crack reached 3 mm. In
peared on the primary support. The maximum length of longitudinal crack reached 3 m,
on the primary support. The maximum length addition,
of longitudinal the concrete
crack reachedof 3them, pipe ro
at theand the
themaximum
tunnel
and entrancewidth
maximum also of
width ofcrack
had reached
reached33mm.
longitudinal
crack mm. In
In addition,
cracks, with athe
addition, the concrete
concreteof
maximum the
thepipe
ofcrack roof
width
pipe roof of 5 m
at
as shown the
at theintunnel
Figure
tunnel entrance also
5b. Before
entrance had
also had longitudinal
the cracks,
collapse,cracks,
longitudinal with
the shotcrete a maximum
with a maximum crack
on the entrance width
crack width of
slope5 mm,
also crack
of 5 mm,
as shown in Figure 5b. Before the collapse, the shotcrete on the entrance slope also cracked
as shown
and spalled in Figure
in ainlarge 5b. Before
range, the collapse, the shotcrete on the entrance slope also cracked
and spalled a large range, almost losing
almost losing itsits
rolerole in preventing
in preventing the infiltration
the infiltration of surfaceof surfa
and spalled in a large range, almost losing its role in preventing the infiltration of surface
water, as shown
water, as shown ininFigure
Figure 5c.
5c.
water, as shown in Figure 5c.

Figure 5. Phenomena before collapse: (a) local collapse of tunnel face; (b) lining cracking; (c) shot-
crete spalling.
FigureFigure
5. Phenomena
5. Phenomenabefore collapse:
before collapse: (a)collapse
(a) local local collapse of (b)
of tunnel face; tunnel face; (b)
lining cracking; (c)lining
shotcretecracking;
spalling. (c) sh
crete spalling.
3.2. Collapse Failure Mechanism Analysis
3.2. Collapse Failure Mechanism Analysis
The soil
The soil of
of the
the tunnel
tunnel front
front slope
slope isis mainly
mainly silty
silty clay.
clay. There
There are
are many
many factors
factors that
that
3.2. Collapse
affect Failure
the shear Mechanism
strength of Analysis
silty clay, such as the mineral composition, water content, and
affect the shear strength of silty clay, such as the mineral composition, water content, and
density,
The among
soilamong
density, of the which
tunnel
which the water content
frontcontent
the water slopehas has the most
is mainly obvious
silty clay.
the most obvious effect on
effectTherethe shear
on the are strength
shearmany
th factors
strength
parameters of
parameters of soil.
soil. Moreover,
Moreover, the the shear
shear strength
strength determines
determines thethe stability
stability of
of the
the soil
soil mass.
mass.
affect the shear strength of silty clay, such as the mineral composition, water content, a
Therefore, rainfall is the main cause of the collapse. Through the analysis of the abnormal
density, among which the water content has the most obvious effect on the shear streng
characteristics of the tunnel structure observed on site before the collapse, the main failure
parameters of soil.
mechanism of theMoreover, the shear
accident induced strength
by rainfall can determines the stability
be obtained. Firstly, of the soil ma
the excavation
Water 2022, 14, 2486 6 of 17

of the tunnel caused unloading of the entrance slope, changing the original equilibrium
conditions of the slope. Further, unloading caused slight deformation of the slope, resulting
in cracks in the shotcrete of the tunnel slope protection. During the continuous rainfall, the
surface water continuously scoured the entrance slope, causing the shotcrete cracks of slope
protection to expand. As a result, the water content of slope soil was increasing. As the
water content of soil increased, the cohesion of soil decreased [47]. Thus, the shear strength
of the soil decreased, and the sliding resistance of the slope decreased. At the same time,
due to the infiltration of surface water into the slope, the weight of the slope soil increased
and the sliding force of the slope increased. During the continuous infiltration of surface
water, the shear strength of the clay on the excavation face of the tunnel decreased due
to the continuous increased of water content, and local collapse occurred. The abnormal
phenomena such as longitudinal cracks of lining observed before the large-scale collapse
intuitively show that the stress state of the primary support was changing. Finally, when
the strength of the primary support structure of the tunnel was insufficient, excessive
deformation occurred under the gravity of the upper soil mass, the stress balance was
broken, and the entrance slope of the tunnel also collapse.
Relevant studies show that the mechanism of the decrease of soil shear strength caused
by the increase of water content can be explained by the change of matrix suction between
soil particles. The water content directly affects the matrix suction between soil particles.
The smaller the water content of clay, the greater the matrix suction and cohesion between
soil particles. On the contrary, the greater the water content of clay, the smaller the cohesion.
Therefore, the matrix suction decreases with the increase of soil water content, which is an
important factor leading to the decline of soil shear strength [48,49]. When the soil changes
from unsaturated state to saturated state, the matrix suction between soil particles will
gradually decrease to 0. A large number of landslide accidents were induced by the change
of soil moisture content, so the matrix suction is closely related to the stability of soil [50].
As shown in Figure 6a, before the rainfall, the soil on the entrance slope of the tunnel
was in an unsaturated state with low water content, and all meniscus between soil particles
were independent of each other. At this time, the matrix suction between soil particles
caused the stress Pi (i = 1, 2, 3, . . . , 6) that cannot be transferred between soil particles,
which inhibited the sliding of soil particles. Because the plastic deformation of soil is
interpreted as the sliding between soil particles, the matrix suction resisted the plastic
deformation of soil when the soil water content was low. At this water content, the entrance
slope was in a good state of stability. When the rainfall started, the water content of the
soil mass of the entrance slope gradually increased from the tunnel to the surface [51].
The matrix suction between soil particles also decreased gradually from the tunnel to the
surface. Therefore, the shear strength of soil presented the change law of decreasing from
the tunnel to the surface. At this time, the soil can be roughly divided into saturated soil
and unsaturated soil along the vertical direction. Due to the increasing water content of
the unsaturated soil, the unsaturated state between soil particles changes, as shown in
Figure 6b. The meniscus in the soil pores was gradually connected. The pressure caused by
matrix suction can be transmitted between soil particles, which can cause the compression
of soil and reduce the stability of entrance. When the soil mass in a certain depth below the
surface of the tunnel entrance slope reached the saturated state, the matrix suction of the
soil mass dropped to 0, and this part of the soil mass was prone to sliding.
With the continuous rainfall, the surface water continuously seeped into the entrance
slope, resulting in the continuous increase of the water content of the whole tunnel entrance
slope, including the soil around the tunnel structure. Therefore, the water content of clay
on the tunnel face was also rising, resulting in the decrease of soil shear strength. This
then, resulted in local small−scale collapse on the tunnel face, as shown in Figure 5a. The
stress state of tunnel support structure was also changed due to rainfall and excavation,
which caused abnormal deformation of tunnel primary support, longitudinal cracks. When
the surrounding rock pressure exceeded the bearing capacity of the primary support, the
 Water 2022, 14, 2486 7 of 17

Water 2022, 14, x FOR PEER REVIEW

tunnel collapsed instantaneously. At the same time, the entrance slope also lost its balance
and slid.

Figure The
Figure6. 6. relationships
The between
relationships rate of water
between ratecontent andcontent
of water matric suction of silty clay
and matric [50]. (a)
suction of low
silty clay [
water content; (b) high water content.
low water content; (b) high water content.
3.3. Factors Analysis
Withwere
There the many
continuous rainfall,
factors that led tothe
the surface water continuously
tunnel collapse, seeped
which can be divided into the e
into
intrinsic and extrinsic
slope, resulting in factors. The intrinsic
the continuous factors were
increase poor
of the geological
water conditions
content of theand whole tu
structural defects of the tunnel, while the extrinsic factors were rainfall and
trance slope, including the soil around the tunnel structure. Therefore, the water construction
disturbances. A large number of tunnel construction collapse accidents occur during the
of clay
rainy on the
season, andtunnel facetime,
at the same wastunnels
also rising, resulting
with collapse in theduring
accidents decrease of often
rainfall soil shear s
Thispoor
have then, resulted
geological in local such
conditions, small−scale
as severe collapse
weatheringon of the tunnel face,rock,
the surrounding as shown
loose in Fi
The stress
internal stateand
structure, of tunnel supportofstructure
easy destruction the internalwas also when
structure changed duetotowater.
exposed rainfall and
tion, which caused abnormal deformation of tunnel primary support, longitudina
3.3.1. Rainfall
When the surrounding rock pressure exceeded the bearing capacity of the prima
The collapse area belongs to the subtropical monsoon climate, with heavy and concen-
port, rainfall
trated the tunnel collapsed
in summer. instantaneously.
Continuous rainfall caused At the same
the entrance time,
slope to bethe entrance slope
continuously
its balance
washed and slid.
by surface water, and the water content of the soil rose, leading to hydrogenic
softening, and the shear strength of the soil decreased. At the same time, the weight of the
soil
3.3.increased
Factors due to the increase of water content, which led to the continuous adjustment
Analysis
of the forces acting on the tunnel lining structure, and the stability of the soil decreased
greatly,There
whichwere
was themany
main factors that
reason for the led to collapse.
tunnel the tunnel collapse,
In the whichsection
shallow buried can be divid
intrinsic
of andthe
the tunnel, extrinsic factors.
infiltration Thewater
of surface intrinsic factors
into the werethe
soil made poorwatergeological
content of conditi
the soil near the tunnel lining increase quickly, and the soil reached
structural defects of the tunnel, while the extrinsic factors were rainfall saturation when theand cons
precipitation was large. At this time, the pressure acting on the primary support of the
disturbances. A large number of tunnel construction collapse accidents occur du
tunnel was not only the gravity of the soil, but also the hydrostatic pressure and seepage
rainy resulting
forces, season, inand at the same
a significant time,
increase tunnels
in the load onwith collapse
the primary accidents
support during
compared to rainf
have
that poor
before thegeological conditions,
rainfall. In the such strength
primary support as severe weathering
weak position, theofprimary
the surrounding
support roc
cracking occured, especially the longitudinal cracks were the most harmful
internal structure, and easy destruction of the internal structure when exposed to [52]. As the soil
outside the liner approached saturation, water in the soil leaked into the tunnel along the
liner fissures (Figure 5b) and caused erosion of the shotcrete for the primary support. At
3.3.1. Rainfall
the location of the tunnel face, if the strength of the shotcrete enclosed on the soil surface of
Theface
the tunnel collapse area belongs
was insufficient, to strength
the shear the subtropical monsoonwith
of the soil decreased climate, withofheavy a
the increase
water
centrated rainfall in summer. Continuous rainfall caused the entrancewater
content, which led to a local collapse of the tunnel face (Figure 5a). As surface slope to be
uously washed by surface water, and the water content of the soil rose, leading to
genic softening, and the shear strength of the soil decreased. At the same time, the
of the soil increased due to the increase of water content, which led to the con
adjustment of the forces acting on the tunnel lining structure, and the stability of
Water 2022, 14, 2486 8 of 17

continued to seep underground, the pressure on the lining structures increased, and when
its bearing capacity was exceeded, the tunnel collapsed instantaneously.

3.3.2. Topography and Geology

The topographical conditions of the tunnel area lead to the decline of the stability of
the entrance slope, which has the basic conditions for the collapse disaster.
The location of the collapse in the left bound tunnel was covered with 1–4 m thick
powder clay below the surface, which was conducive to the downward seepage of surface
water. The clay content of powdered clay was higher compared to the general unsaturated
soil, and the clay content was closely related to the size of matrix suction and water storage
capacity of the soil. Extensive research found that the higher the clay content, the stronger
the water storage capacity, and the water content of silty clay was affected by the increase
of clay content, the degree of which first increased and then decreased [53,54]. Therefore,
silty clay with high clay content have a higher saturated water content compared to soils
with low clay content. Moreover, the physical state and mechanical properties of silty clay
are obviously affected by its water content, softening significantly in water, and mechanical
parameters such as cohesion and elastic modulus drop significantly, so its water content
directly affects the stability and safety of the tunnel [55,56]. When continuous rainfall
occurred, the shear strength of the silty clay soil decreased continuously as the water
content of the soil rose, and when the water content of the soil was close to saturation, the
shear strength of the soil was the lowest and the stability of the entrance slope was the
worst. At the same time, the silty clay layer tended to slide along the top surface of the
mass rock and stone. The tunnel face of the left bound tunnel passed through the interface
between the silty clay layer and the mass rock and soil as well as the sandstone with severe
weathering. In the process of continuous downward seepage of surface water, the stability
gradually decreased, causing the soil layer to slide along the soil interface.

3.3.3. Unacceptable Construction

The construction of the tunnel entrance section does not meet the requirements of
the design, which can lead to a decrease in the stability of the tunnel structure and induce
collapse damage to the tunnel. Insufficient primary support strength, untimely support
follow−up, and insufficient slope strength are all construction factors that affect the stability
of the tunnel [57,58]. The reasons for the insufficient strength of the primary support struc-
ture include insufficient thickness and strength of the shotcrete, too large spacing between
steel arches, etc. Because the strength of the primary support was insufficient, the shotcrete
produced cracks under the pressure in the surrounding rock. Under rainfall conditions, it
was also accompanied by water leakage from the lining and erosion of the primary support
structure. After tunnel excavation, the surrounding rock stress redistribution occurred. If
the support does not follow up in time, it will lead to further expansion of the plastic zone
and cause collapse [59]. Besides, the shotcrete of the slope protection was not thick enough
and cracks existed, therefore, the surface water continuously seeped down into the slope
body, causing the water content of the slope body to rise continuously and the weight of
the soil body to increase. A landslide occurred when the sliding force of the slope exceeded
the sliding resistance of the slope. Therefore, in areas with high precipitation, improving
the quality of slope protection in the tunnel entrance section is one of the effective ways to
ensure the safety of the tunnel structure.

4. Countermeasures after Collapse

After the collapse occurred, the tunnel was completely buried by the upper sliding
soil. Before excavation and removal of the collapsed body, the slope near the collapse that
had not experienced a collapse landslide was reinforced. Figure 7 shows the layout of
reinforcement measures. The reinforcement measures used for the secondary excavation of
the tunnel entrance consisted of three main types: grouting, anti−slide pile, and shotcrete
slope protection. Grouting reinforcement mainly includes tunnel surface grouting and
 4. Countermeasures after Collapse
After the collapse occurred, the tunnel was completely buried by the upper sliding
soil. Before excavation and removal of the collapsed body, the slope near the collapse that
had not experienced a collapse landslide was reinforced. Figure 7 shows the layout of
Water 2022, 14, 2486 reinforcement measures. The reinforcement measures used for the secondary excavation 9 of 17
of the tunnel entrance consisted of three main types: grouting, anti−slide pile, and shot-
crete slope protection. Grouting reinforcement mainly includes tunnel surface grouting
and grouting
grouting in tunnel.
in tunnel. Anti− Anti−slide
slide pilespiles areat
are set setthe
at tunnel
the tunnel entrance
entrance location
location to reinforce
to reinforce the
the foundation of the tunnel. At the same time, in order to stop the tunnel
foundation of the tunnel. At the same time, in order to stop the tunnel from deforming from deformingtoo
too after
fast fast after excavation,
excavation, leading
leading to cracking
to cracking of theoflining
the lining structure,
structure, temporary
temporary steel steel sup-
supports
portsinstalled
were were installed
near the near the excavation
tunnel tunnel excavation
surface.surface. Four inclinometer
Four inclinometer pipe were pipe were
placed
placed
on on the surface
the surface to obtainto obtain the deformation
the deformation data ontodatathe
onto the upper
upper slope slope
of theoftunnel,
the tunnel,
and
and together
together withdisplacement
with the the displacement monitoring
monitoring data ontodatathe
onto the cavity,
tunnel tunnel they
cavity,
were they
usedwere
as
used as the basis of tunnel stability judgment. The direction of the displacement
the basis of tunnel stability judgment. The direction of the displacement monitoring was monitor-
ing Xwas
the the X direction
direction pointed at pointed
by the at byarrow
red the red
in arrow
Figurein 7b.Figure 7b.

Figure 7.
Figure 7. Layout
Layout of
of treatment
treatment measures.
measures. (a)
(a) Vertical
Verticalsection;
section;(b)
(b)vertical
verticalview.
view.

4.1.
4.1. Anti −Slide Pile
Anti−Slide Pile
The
The foundation
foundationunder underthethetunnel
tunnelopening
opening section mainly
section includes
mainly includespelitic siltstone
pelitic and
siltstone
silty mudstone. They were severely weathered, with a large number
and silty mudstone. They were severely weathered, with a large number of fissures, bed- of fissures, bedded
crushing, and low
ded crushing, androck
low strength. In order
rock strength. to prevent
In order the foundation
to prevent from the
the foundation fromtunnel from
the tunnel
sliding underunder
from sliding the rainfall and to
the rainfall andensure the stability
to ensure of theoftunnel
the stability entrance
the tunnel section,
entrance the
section,
foundation of the tunnel was reinforced by anti − slide piles. The anti − slide
the foundation of the tunnel was reinforced by anti−slide piles. The anti−slide piles were piles were set at
the arch foot on both sides of the tunnel entrance, with a pile length of 8 m
set at the arch foot on both sides of the tunnel entrance, with a pile length of 8 m and a and a rectangular
cross −section
rectangular of 2 × 2 m. Pre
cross−section of 2−×excavated pile holes were
2 m. Pre−excavated pile excavated
holes wereon both sides
excavated onatboth
the
arch footing position and the reinforcement cages were placed, then the
sides at the arch footing position and the reinforcement cages were placed, then the piles piles were poured
using C20 concrete,
were poured as shown
using C20 in Figure
concrete, as shown8. Finally, connecting
in Figure 8. Finally,bars were tiedbars
connecting between
were the
tied
Water 2022, 14, x FOR PEER REVIEWtwo anti−slide piles and concrete was poured to form a concrete beam to connect the 10 two
of 18
between the two anti−slide piles and concrete was poured to form a concrete beam to
anti −slide
connect thepiles
twoas a whole and
anti−slide pilesfurther increase
as a whole andthe stability
further of thethe
increase foundation.
stability of the foun-
dation.

Figure 8.
Figure 8. Anti
Anti−slide
−slide pile.
pile.

4.2. Grouting
Grouting reinforcement is divided into two parts: surface grouting and grouting in
tunnel [60]. Surface grouting reinforcement can not only enhance the integrity of the up-
per strata of the tunnel, but also form a curtain for cutting off water to stop the downward
seepage of surface water and fundamentally cure the problem of slope erosion by surface
 Figure 8. Anti−slide pile.
Water 2022, 14, 2486 10 of 17
4.2. Grouting
Grouting reinforcement is divided into two parts: surface grouting and grouting in
4.2. Grouting
tunnel [60]. Surface grouting reinforcement can not only enhance the integrity of the up-
Grouting
per strata of the reinforcement
tunnel, but also is divided into twofor
form a curtain parts: surface
cutting off grouting
water to and
stopgrouting in
the downward
tunnel [60]. Surface grouting reinforcement can not only enhance the integrity
seepage of surface water and fundamentally cure the problem of slope erosion by surface of the upper
strata of the tunnel, but also form a curtain for cutting off water to stop the downward
water. The surface reinforcement grouting holes are shown in Figure 9a. The grouting
seepage of surface water and fundamentally cure the problem of slope erosion by surface
pipe is made
water. of steel
The surface pipe with agrouting
reinforcement diameter of are
holes Φ50shown
mm in and pipe9a.
Figure thickness of 5 pipe
The grouting mm. The
front part of
is made of the
steelgrouting
pipe withpipe was drilled
a diameter of Φ50with
mm8andmmpipediameter slurry
thickness of 5 outlets
mm. The with
fronta spac-
ingpart
of 100 mm.
of the Therepipe
grouting waswasno slurry
drilled outlet
with 8 at
mm 500 mm atslurry
diameter the end of the
outlets grouting
with a spacingpipe.
of The
depth of the
100 mm. surface
There was nogrouting hole at
slurry outlet was5005mmm. at
The
theplane
end oflayout of thepipe.
the grouting grouting hole was
The depth
of the surface
arranged accordinggrouting hole
to 1.5 wasm.
× 1.5 5 m.
TheThe planefor
slurry layout of the
surface groutingused
grouting hole ordinary
was arrangedPortland
according to 1.5 × 1.5 m. The slurry for surface grouting used ordinary
cement slurry with a water: cement ratio of 1:1, and the initial grouting pressure ranged Portland cement
slurry
from 0.5 with
to 1.0a MPa
water:andcement ratio of 1:1, and
the termination the initial
pressure grouting
ranged frompressure
1.5 to 2.0ranged
MPa.from
Advance
0.5 to 1.0 MPa and the termination pressure ranged from 1.5 to 2.0 MPa. Advance small
small conduits were used for grouting in the tunnel, with a diameter of 30 mm and a
conduits were used for grouting in the tunnel, with a diameter of 30 mm and a length of
length of 5 m9b).
5 m (Figure (Figure 9b). Thepipes
The grouting grouting pipes were
were spaced 500 mm spaced
along500 mm along
the tunnel the tunnel cir-
circumferential
cumferential
direction withdirection withinsertion
an external an external
angleinsertion angleand
of 15 degrees, of 15
thedegrees, andlap
longitudinal thelength
longitudinal
of
laptwo
length of two adjacent rows of grouting pipes was 1 m. The grouting
adjacent rows of grouting pipes was 1 m. The grouting slurry was ordinary cement slurry was ordi-
nary cement
slurry, and theslurry,
cementandwasthe cement
PO42.5 was cement.
Portland PO42.5WhenPortland cement.
the hole When the
was grouted, theinitial
hole was
pressure
grouted, was
the 0.2 MPa
initial and the
pressure termination
was 0.2 MPa pressure
and the was 0.6 MPa. pressure was 0.6 MPa.
termination

Figure 9. Photographs
Figure 9. Photographsof
ofgrouting. (a) Surface
grouting. (a) Surfacegrouting;
grouting;(b)(b) grouting
grouting in tunnel.
in tunnel.

4.3. Slope Reinforcement

In the entrance section of the tunnel, reducing the seepage of surface water into the
tunnel entrance slope is the key to ensuring the stability of the slope. The entrance slope of
the tunnel was protected by shotcrete and grouting bolts. Specifically, the grouting bolts
were 22 mm in diameter and 4 m in length, arranged in quincunx shape with a spacing of
0.8 m. After the grouting was completed, the reinforcing mesh was laid on the slope surface
of the tunnel entrance slope, which was made of a steel bar with a spacing of 0.2 m × 0.2 m.
The diameter of the steel bar is 8 mm. After the grouting and reinforcing mesh layout were
completed, the tunnel entrance slope was covered with C20 concrete with a thickness of
0.15 m. In order to further reduce the infiltration of surface water into the slope during
rainfall, an interceptor ditch was excavated along the outer edge of the slope protection,
and surface water flowing to the interceptor ditches was drawn into the interceptor ditch
and drained away.

4.4. Analysis on Effects of the Countermeasures

Figure 10 shows the variation curves of vault settlement with the time of two moni-
toring sections before the tunnel collapse. From the vault settlement monitoring curves
of the two monitoring sections in Figure 10, it can be seen that the vault settlement at
 protection, and surface water flowing to the interceptor ditches was drawn into the inter-
ceptor ditch and drained away.

4.4. Analysis on Effects of the Countermeasures

Water 2022, 14, 2486 Figure 10 shows the variation curves of vault settlement with the time of two moni- 11 of 17
toring sections before the tunnel collapse. From the vault settlement monitoring curves of
the two monitoring sections in Figure 10, it can be seen that the vault settlement at K161
+K161
746 started to grow
+ 746 started toat a significantly
grow slower slower
at a significantly rate onrate
the 5on
th day
the after excavation,
5th day indicat-
after excavation,
ing that the
indicating tunnel
that surrounding
the tunnel the rock
surrounding the had
rock stabilized at this
had stabilized point.
at this However,
point. However, from 18
from
August,
18 August,thethe
vault settlement
vault increased
settlement increasedsharply, andand
sharply, after 4 days
after before
4 days thethe
before collapse on on
collapse 22
22 August,
August, thethe average
average amount
amount ofof vaultsettlement
vault settlementwas
wasabout
about48 48mm.
mm.TheThevault
vault settlement
settlement
at K161 ++ 743 also showed a sudden increase on 18 August. It indicates indicates that the tunnel
structure−bearingcapacity
structure−bearing capacitydecreased
decreasedfrom
fromthisthisdate
datebecause
because of ofthe
the rainfall.
rainfall.

Figure
Figure 10.
10. Vault settlement before
Vault settlement before collapse:
collapse: (a)
(a) K161
K161 +
+ 746
746 section;
section; (b)
(b) K161
K161 ++ 743
743 section.
section.

Although the the tunnel

tunnelwaswasreinforced
reinforced byby thethe above
above measures,
measures, it was
it was in rainy
in rainy seasonseason
with
frequent
with rainfall,
frequent and there
rainfall, and wastherestill
wasthe possibility
still of collapse
the possibility again. The
of collapse possible
again. causes
The possible
of collapse
causes were aswere
of collapse follows: (1) The(1)
as follows: soil
Theatsoil
thisatlocation still had
this location stillahad
higha water content
high water after
content
the erosive
after action
the erosive of theofdisturbance
action the disturbancethat caused this collapse.
that caused (2) After
this collapse. the collapse,
(2) After the
the collapse,
soil in the tunnel entrance section became looser than before. (3) The
the soil in the tunnel entrance section became looser than before. (3) The above reinforce-above reinforcement
measures
ment were were
measures not necessarily the most
not necessarily suitable
the most treatment
suitable measures
treatment measuresfor this
for collapse and
this collapse
werewere
and subject to verification.
subject In order
to verification. to grasp
In order the displacement
to grasp changes
the displacement of surrounding
changes rock
of surround-
during
ing rockthe secondary
during excavationexcavation
the secondary and the deformation characteristics
and the deformation of the entrance
characteristics slope,
of the en-
displacement
trance test elementstest
slope, displacement were placed were
elements insideplaced
the tunnel
insideand
the inclinometer tubes were
tunnel and inclinometer
installed
tubes were inside the slope.
installed insideInclinometer tubes were tubes
the slope. Inclinometer laid atwere
the outer
laid atedge
the of the edge
outer interceptor
of the
ditch and inserted vertically downward into the soil layer.
interceptor ditch and inserted vertically downward into the soil layer. The tunnel The tunnel displacement
monitoring data of the monitoring section ZK161 + 729 was selected for reinforcement
effect analysis; the test elements and monitoring lines of this section are shown in Figure 11.
A total of seven pressure boxes were deployed to monitor the surrounding rock pressure,
numbered Pi, i = 1, 2, . . . , 7. A total of seven stress gauges were laid at the outer edge of
the steel arch to measure the size and distribution of the internal forces in the steel arch,
numbered Si, i = 1, 2, . . . , 7. There were three displacement monitoring lines, among which
one horizontal convergence monitoring line was laid at the upper and lower step positions.
The surrounding rock deformation monitoring curve of ZK161 + 729 is shown in
Figure 12. The vault settlement and convergence deformation grew faster in 10 days after
the excavation was completed, and gradually slowed down as time increased, and basically
stabilized on the 25th day. The stable value of vault settlement was about 24 mm, while
stable values of convergence deformation were less than 25 mm. It shows that the tunnel
treatment measures played a good role in improving the stability of the entrance slope
and tunnel structure during the secondary excavation. Besides, rainfall occurred on the
13th–15th day after the excavation of this section, but the monitoring data found no
significant impact on the deformation trend of the tunnel structure, so the surface grouting
and slope protection can effectively prevent the infiltration of surface water into the slope.

Water 2022, 14, 2486
shown in Figure 11. A total of seven pressure boxes were deployed to monitor
rounding rock pressure, numbered Pi, i = 1, 2, ..., 7. A total of seven stress gaug
laid at the outer edge of the steel arch to measure the size and distribution of the
forces in the steel arch, numbered Si, i = 1, 2, ..., 7. There were three displacemen
toring lines, among which one horizontal convergence monitoring 12line of 17was lai

upper and lower step positions.

Water 2022, 14, x FOR PEER REVIEW 13 of 18

Figure
Figure 11.11. Layout
Layout and and photographs
photographs of test elements
of test elements at ZK161 + at ZK161
729. + 729.

The surrounding rock deformation monitoring curve of ZK161 + 729 is show

ure 12. The vault settlement and convergence deformation grew faster in 10 days
excavation was completed, and gradually slowed down as time increased, and b
stabilized on the 25th day. The stable value of vault settlement was about 24 mm
stable values of convergence deformation were less than 25 mm. It shows that th
treatment measures played a good role in improving the stability of the entran
and tunnel structure during the secondary excavation. Besides, rainfall occurred
13th–15th day after the excavation of this section, but the monitoring data found
nificant impact on the deformation trend of the tunnel structure, so the surface g
and slope protection can effectively prevent the infiltration of surface water into th

(a) (b)
Figure12.
Figure 12.Displacement
Displacementmonitoring
monitoringcurve
curve
ofof primary
primary (a)(a)
lining.
lining. variation
variation curve
curve of vault
of vault settle-
settlement;
ment;
(b) (b) variation
variation curve of convergence
curve of convergence deformation.deformation.

The
Thedisplacement
displacementofofsoil
soilmass
massinside
insidethe
theentrance
entranceslope
slopecancanbebeobtained
obtainedbybyusing usingthe
the
inclinometer,
inclinometer,andandthen
thenthe
thepossible
possiblesliding
slidingsurface
surfaceand
anddisplacement
displacementofofthe theslope
slopecancanbebe
judged.
judged.Figure
Figure1313shows
showsthethedistribution
distributionofofdisplacement
displacementsize sizeofofthe
thesoil
soillayer
layeratatthe
thefour
four
inclinometer tubes locations when the monitoring data were stable. At this
inclinometer tubes locations when the monitoring data were stable. At this time, the tun-time, the tunnel
was
nelexcavated to ZK161
was excavated + 712 +position.
to ZK161 The displacement
712 position. valuesvalues
The displacement of the of
soil
thelayer
soilat the at
layer four
the
monitoring locations were all within 4.46 mm, and the soil layer of the entrance
four monitoring locations were all within 4.46 mm, and the soil layer of the entrance slope slope did
not
didproduce too much
not produce displacement.
too much displacement.The monitoring resultsresults
The monitoring show show
that the
thatreinforcement
the reinforce-
measures are effective
ment measures for the tunnel
are effective for the entrance collapse.
tunnel entrance collapse.
Figure 14 shows the distribution of steel arch stress and surrounding rock pressure
at the basic stability of the location of the monitoring section ZK161 + 729. Because the
thickness of the upper soil layer of the tunnel at this location was only about 15 m, the steel
arch stress and the surrounding rock pressure were not very large. The maximum value
 (a) (b)
Figure 12. Displacement monitoring curve of primary lining. (a) variation curve of vault
Water 2022, 14, 2486 ment; (b) variation curve of convergence deformation. 13 of 17

The displacement of soil mass inside the entrance slope can be obtained by us
of the surrounding rock pressure was 22.3 kPa at the vault, and it shows the distribution
inclinometer, and then the possible sliding surface and displacement of the slope
feature of “large at the top and small at the bottom”. In general, the pressure on the right
judged. Figure
side of the tunnel13wasshows
slightlythe distribution
higher ofthe
than that on displacement sizearch
left side. The steel of the
wassoil
underlayer at t
inclinometer tubes
pressure at the five locations
monitoring when with
positions, the monitoring
a maximum stressdataofwere
5.954 stable.
kN, andAtthethis
size time, t
distribution characteristics were similar to those of the surrounding rock
nel was excavated to ZK161 + 712 position. The displacement values of the soil layepressure. The
steel arch stress and the surrounding rock pressure were within reasonable limits and no
four monitoring locations were all within 4.46 mm, and the soil layer of the entranc
abnormalities occurred, which reflected that the tunnel was in a stable state at this location,
did notdemonstrating
further produce toothat much displacement.were
the countermeasures Theeffective
monitoring results
in improving theshow that
stability of the rei
ment measures are effective for the tunnel entrance collapse.
the tunnel structure and entrance slope of the secondary excavation.

Water 2022, 14, x FOR PEER REVIEW 14 of 18

and the size distribution characteristics were similar to those of the surrounding rock pres-
sure. The steel arch stress and the surrounding rock pressure were within reasonable lim-
its and no abnormalities occurred, which reflected that the tunnel was in a stable state at
this location, further demonstrating that the countermeasures were effective in improving
13.Variation
Figure 13.
Figure
the stabilityVariationcurve of horizontal
curve
of the tunnel
displacement
of horizontal
structure
at various at
anddisplacement
entrance slope
depths.
various
of depths. excavation.
the secondary

Figure 14 shows the distribution of steel arch stress and surrounding rock p
at the basic stability of the location of the monitoring section ZK161 + 729. Beca
thickness of the upper soil layer of the tunnel at this location was only about 15
steel arch stress and the surrounding rock pressure were not very large. The ma
value of the surrounding rock pressure was 22.3 kPa at the vault, and it shows the
bution feature of “large at the top and small at the bottom”. In general, the press
the right side of the tunnel was slightly higher than that on the left side. The ste
was under pressure at the five monitoring positions, with a maximum stress of 5.9

Figure 14. Stress

Figure 14. Stressmonitoring
monitoring results:
results: (a) (a) surrounding
surrounding rock pressure;
rock pressure; (b)arch
(b) steel steel arch stress.
stress.

There are many other tunnel construction sites in southwest China with geological and
There are many other tunnel construction sites in southwest China with geological
climatic conditions similar to those of the tunnel site in this study. These projects have many
and climatic conditions similar to those of the tunnel site in this study. These projects have
commonalities, such as geological conditions, precipitation, and construction methods.
many commonalities,
Therefore, such as geological
it is of good application conditions,the
value to summarize precipitation,
experience ofand construction
tunnel entrance meth-
ods. Therefore,
section collapse init is of area
this good application
and to achievevalue to prevention
collapse summarizeand theeffective
experience of tunnel en-
treatment
trance
for similar projects by engineering analogy method. After the study of this accident, an treat-
section collapse in this area and to achieve collapse prevention and effective
ment for similar
excavation collapseprojects by engineering
prevention analogy
and treatment processmethod. After
applicable the study
to areas of this
with high accident,
and
concentrated
an excavation seasonal rainfall
collapse and complex
prevention andgeological
treatmentconditions
process in the tunnel to
applicable inlet section
areas with high
is summarized,
and concentrated as shown in Figure
seasonal 15. and complex geological conditions in the tunnel inlet
rainfall
section is summarized, as shown in Figure 15.
 many commonalities, such as geological conditions, precipitation, and construction meth-
ods. Therefore, it is of good application value to summarize the experience of tunnel en-
trance section collapse in this area and to achieve collapse prevention and effective treat-
ment for similar projects by engineering analogy method. After the study of this accident,
an excavation collapse prevention and treatment process applicable to areas with high
Water 2022, 14, 2486 14 of 17
and concentrated seasonal rainfall and complex geological conditions in the tunnel inlet
section is summarized, as shown in Figure 15.

Figure15.
Figure 15. Prevention
Prevention and
and treatment
treatmentprocess
processof
oftunnel
tunnelentrance
entrancecollapse
collapseunder
underrainfall
rainfallcondition.
condition.

5.
5. Conclusions
Conclusions
This
This paper
paper analyzed
analyzed the
the failure
failure mechanisms
mechanisms and
and treatment
treatment measures
measures ofof entrance
entrance col-
col-
lapse due to sustained rainfall in a highway tunnel as a case study. Also, the characteristics
lapse due to sustained rainfall in a highway tunnel as a case study. Also, the characteristics
of tunnel before collapse was summed up by field investigation, and the main factors of the
collapse were analyzed. Based on the analysis results, treatment measures for tunnel en-
trance collapse were adopted. Further, the effectiveness of the collapse treatment measures
was verified by field monitoring. The main conclusions can be summarized as follows:
(1) In the process of construction at tunnel entrance, rainfall should be emphasized. It
reduces the shearing strength of the soil and damages the structure of tunnel, which
often causes collapse failures. Other factors including unacceptable construction,
topography, and geology also contribute to tunnel entrance collapse. Particularly,
improving the quality of slope protection is highly recommended.
(2) The basic causes of the collapse were the infiltration of surface water and the increase
of water content of silty clay. The matrix suction decreased with the increase of soil
water content, which led to the decline of soil shear strength.
(3) Field investigation showed that abnormal phenomena appeared before large−scale
tunnel collapse, such as local collapse of tunnel face, primary support cracking, and
shotcrete spalling of slope protection. Timely discovery of abnormal phenomena is of
great significance for the prevention of large−scale tunnel collapse accidents. When
these phenomena occur on the construction site, it is strongly recommended that the
construction personnel evacuate the tunnel immediately.
(4) To prevent a secondary collapse in construction, targeted treatment measures were
implemented at this tunnel, such as anti−slide pipe, grouting, and slope reinforce-
ment. Grouting can prevent the seepage of water in soil and improve the stability of
surrounding rock. After the collapse treatment was completed, the vault settlement of
the monitoring section was only about 24 mm. The countermeasures are effective for
collapse at tunnel entrance under rainfall conditions based on the monitoring results
and can be applied to similar tunnel entrance slope. Lessons learned from this case
provide a reference for the reinforcement of tunnel entrance slope.
Water 2022, 14, 2486 15 of 17

Author Contributions: Writing—manuscript, L.-L.C. and Z.-F.W.; methodology, L.-L.C. and Y.-Q.W.,
data collection, L.-L.C., Y.-Q.W., and Z.-F.W.; case analysis, L.-L.C.; project management, Y.-Q.W. and
Z.-F.W.; funding, Z.-F.W. and Y.-Q.W. All authors have read and agreed to the published version of
the manuscript.
Funding: The authors gratefully acknowledge the financial support by the National Key R&D
Program of China (No. 2021YFB2600404, 2021YFF0501101, 2021YFA0716901 & 2021YFB2601402) and
the National Natural Science Foundation of China (No. 51978059 & 52178310).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data used to support the findings of this study are available from
the corresponding author upon request.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Lai, J.; Wang, X.; Qiu, J.; Chen, J.; Hu, Z.; Wang, H. Extreme deformation characteristics and countermeasures for a tunnel in
difficult grounds in southern Shaanxi. Environ. Earth Sci. 2018, 77, 706. [CrossRef]
2. Aberle, J.; Henry, P.Y.; Kleischmann, F.; Navaratnam, C.U.; Vold, M.; Eikenberg, R.; Olsen, N.R.B. Experimental and Numerical
Determination of the Head Loss of a Pressure Driven Flow through an Unlined Rock-Blasted Tunnel. Water 2021, 12, 3492.
[CrossRef]
3. Lai, J.; He, S.; Qiu, J.; Chen, J.; Shi, S.; Wang, L.; Wang, K.; Wang, J. Cause analysis of soft and hard rock tunnel collapse and
information management. Environ. Earth Sci. 2018, 76, 94. [CrossRef]
4. Soga, K.; Laver, R.G.; Li, Z.L. Long-term tunnel behaviour and ground movements after tunnelling in clayey soils. Undergr. Space
2018, 2, 149–167. [CrossRef]
5. Zhang, N.; Zheng, Q.; Elbaz, K.; Xu, Y. Water Inrush Hazards in the Chaoyang Tunnel, Guizhou, China: A Preliminary
Investigation. Water 2020, 12, 1083. [CrossRef]
6. Kim, J.; Kim, C.; Kim, G.; Kim, I.; Abbas, Q.; Lee, J. Probabilistic tunnel collapse risk evaluation model using analytical hierarchy
process (AHP) and Delphi survey technique. Tunn. Undergr. Space Technol. 2022, 120, 104262. [CrossRef]
7. Zheng, X.; Zhang, L.; Wu, S.; Song, K. Study on the Shape of the Aerator of High-Head Discharge Tunnel with Mild Bottom Slope.
Water 2021, 13, 2128. [CrossRef]
8. Farghaly, A.A.; Kontoni, D.P.N. Train induced dynamic response of a pedestrian tunnel under a four-track surface railway for
different soil water contents. Geomech. Eng. 2019, 16, 341–353. [CrossRef]
9. Gattinoni, P.; Consonni, M.; Francani, V.; Leonelli, G.; Lorenzo, C. Tunnelling in landslide areas connected to deep seated
gravitational deformations: An example in Central Alps (northern Italy). Tunn. Undergr. Space Technol. 2019, 93, 103100.
[CrossRef]
10. Xue, Y.; Zhang, S.; Zhou, M.; Zhu, H. Novel SfM-DLT method for metro tunnel 3D reconstruction and visualization. Undergr.
Space 2021, 6, 134–141. [CrossRef]
11. Guo, C.; Guo, Q.; Zhang, T.; Li, W.; Zhu, H.; Yan, Z. Study on real-time heat release rate inversion for dynamic reconstruction and
visualization of tunnel fire scenarios. Tunn. Undergr. Space Technol. 2022, 122, 104333. [CrossRef]
12. Fraldi, M.; Guarracino, F. Analytical solutions for collapse mechanisms in tunnels with arbitrary cross sections. Int. J. Solids Struct.
2010, 47, 216–223. [CrossRef]
13. Sun, W.; Liang, Q.; Qin, S.; Yuan, Y.; Zhang, T. Evaluation of groundwater effects on tunnel engineering in loess. Bull. Eng. Geol.
Environ. 2021, 80, 1947–1962. [CrossRef]
14. Wang, J.; Lin, G.; Xu, G.; Wei, Y.; Li, S.; Tang, X.; He, C. Face stability of EPB shield tunnels in multilayered ground with soft
sand lying on hard rock considering dynamic excavation process: A DEM study. Tunn. Undergr. Space Technol. 2022, 120, 104268.
[CrossRef]
15. Prendes-Gero, M.B.; Lopez-Gayarre, F.; Menendez-Fernandez, C.; Llardent, M.R.A. Forensic analysis of the failure of the
foundations of a tunnel built to channel the course of a river. Eng. Fail. Anal. 2013, 32, 152–166. [CrossRef]
16. Xu, Z.; Cai, N.; Li, X.; Xian, M.; Dong, T. Risk assessment of loess tunnel collapse during construction based on an attribute
recognition model. Bull. Eng. Geol. Environ. 2021, 80, 6205–6220. [CrossRef]
17. Wu, B.; Qiu, W.; Huang, W.; Meng, G.; Huang, J.; Xu, S. Dynamic risk evaluation method for collapse disasters of drill-and-blast
tunnels: A case study. Math. Biosci. Eng. 2021, 19, 309–330. [CrossRef]
18. Han, K.; Ju, J.W.W.; Kong, H.; Wang, M. Functional catastrophe analysis of progressive failures for deep tunnel roof considering
variable dilatancy angle and detaching velocity. Rock Mech. Rock Eng. 2019, 52, 3987–3997. [CrossRef]
19. Zhang, C.; Han, K. Collapsed shape of shallow unlined tunnels based on functional catastrophe theory. Math. Probl. Eng. 2015,
2015, 681257. [CrossRef]
Water 2022, 14, 2486 16 of 17

20. Ma, E.; Lai, J.; Xu, S.; Shi, S.; Zhang, J.; Zhong, Y. Failure analysis and treatments of a loess tunnel being constructed in ground
fissure area. Eng. Fail. Anal. 2022, 134, 106034. [CrossRef]
21. Xu, S.; Lei, H.; Li, C.; Liu, H.; Lai, J.; Liu, T. Model test on mechanical characteristics of shallow tunnel excavation failure in gully
topography. Eng. Fail. Anal. 2020, 119, 104978. [CrossRef]
22. Yu, L.; Zhang, D.; Fang, Q.; Cao, L.; Zhang, Y.; Xu, T. Face stability of shallow tunnelling in sandy soil considering unsupported
length. Tunn. Undergr. Space Technol. 2020, 102, 103445. [CrossRef]
23. Hu, X.; Fu, W.; Wu, S.; Fang, Y.; Wang, J.; He, C. Numerical study on the tunnel stability in granular soil using DEM virtual air
bag model. Acta Geotech. 2021, 16, 3285–3300. [CrossRef]
24. Wu, Y.; Wang, K.; Zhang, L.; Peng, S. Sand-layer collapse treatment: An engineering example from Qingdao Metro subway tunnel.
J. Clean Prod. 2018, 197, 19–24. [CrossRef]
25. Xu, H.; Shao, Z.; Chen, C.; Wang, Z.; Cai, L.; Li, Z.; Huang, C. Characteristics and evolution of tunnel collapse in fully-weathered
coastal red sandstone strata. J. Coast. Res. 2022, 38, 289–301. [CrossRef]
26. Xu, H.; Wang, Z.; Chen, C.; Cai, L.; Li, Z.; Deng, Y.; Xia, Y. Model tests on characteristics and evolution of tunnel collapse in
soil-sand interbedded strata. Chin. J. Geotech. Eng. 2021, 43, 1050–1058. (In Chinese)
27. Wang, Z.; Xie, Y.; Lai, J.; Xie, Y.; Su, X.; Shi, Y.; Guo, C. Designing an innovative support system in loess tunnel. Geomech. Eng.
2021, 24, 253–266. [CrossRef]
28. Senent, S.; Jimenez, R. A tunnel face failure mechanism for layered ground, considering the possibility of partial collapse. Tunn.
Undergr. Space Technol. 2015, 47, 182–192. [CrossRef]
29. Wang, B.; Li, T.; HE, C.; Zhou, Y. Analysis of failure properties and formatting mechanism of soft rock tunnel in meizoseismal
areas. Chin. J. Rock Mech. Eng. 2012, 31, 928–936. (In Chinese)
30. Toulkeridis, T.; Rodriguez, F.; Jimenez, N.A.; Baile, D.S.; Martinez, R.S.; Addison, A.; Freyre, D.C.; Mato, F.; Perez, C.D. Causes
and consequences of the sinkhole at El Trebol of Quito, Ecuador-implications for economic damage and risk assessment. Nat.
Hazards Earth Syst. Sci. 2016, 16, 2031–2041. [CrossRef]
31. Liu, X.; Liu, F.; Song, K. Mechanism analysis of tunnel collapse in a soft-hard interbedded surrounding rock mass: A case study of
the Yangshan Tunnel in China. Eng. Fail. Anal. 2022, 138, 106304. [CrossRef]
32. Wang, Y. Collapse mechanism and preventive measures of mountain tunnel. Chin. J. Rock Mech. Eng. 2012, 30, 2376. (In Chinese)
33. Wang, Z.; Shen, S.; Modoni, G.; Zhou, A. Excess pore water pressure caused by the installation of jet grouting columns in clay.
Comput. Geotech. 2020, 125, 103667. [CrossRef]
34. Jiang, Z.; Duan, S.; Zhao, X. Manipulation of tunnelling in a quantum dot array. J. Phys.-Condes. Matter 2005, 17, 4207–4222.
[CrossRef]
35. Niu, J.; Sun, Y.; Wang, B.; Zhang, K.; Huang, Y.; Huang, S.; Yu, J.; Qiu, L. Grouting Treatment of Water and Mud Inrush in Fully
Weathered Granite Tunnel: A Case Study. Disaster Adv. 2020, 2020, 8838769. [CrossRef]
36. Zhang, T.; Taylor, R.N.; Zheng, G.; Sun, J.; Fan, Q.; Diao, Y.; Zhou, H. Modelling ground movements near a pressurised tunnel
heading in drained granular soil. Comput. Geotech. 2018, 104, 152–166. [CrossRef]
37. Wu, D.; Deng, T.; Zhao, R.; Wang, Y. THM modeling of ground subsidence induced by excavation of subway tunnel. Comput.
Geotech. 2018, 94, 1–11. [CrossRef]
38. Wang, Z.; Cheng, W. Predicting jet-grout column diameter to mitigate the environmental impact using an artificial intelligence
algorithm. Undergr. Space. 2021, 6, 267–280. [CrossRef]
39. Yang, Z.; Wu, S.; Gao, Y.; Jin, A.; Cong, Z. Time and technique of rehabilitation for large deformation of tunnels in jointed rock
masses based on FDM and DEM numerical modeling. Tunn. Undergr. Space Technol. 2018, 81, 669–681. [CrossRef]
40. Yuan, J.; Chen, W.; Tan, X.; Yang, D.; Wang, S. Countermeasures of water and mud inrush disaster in completely weathered
granite tunnels: A case study. Environ. Earth Sci. 2019, 78, 576. [CrossRef]
41. Zhang, N.; Shen, J.; Lin, C.; Arulrajah, A.; Chai, J. Investigation of a large ground collapse and countermeasures during mountain
tunnelling in Hangzhou: A case study. Bull. Eng. Geol. Environ. 2019, 78, 991–1003. [CrossRef]
42. Zheng, G.; Cui, T.; Cheng, X.; Diao, Y.; Zhang, T.; Sun, J.; Ge, L. Study of the collapse mechanism of shield tunnels due to the
failure of segments in sandy ground. Eng. Fail. Anal. 2017, 79, 464–490. [CrossRef]
43. Zheng, H.; Li, P.; Ma, G.; Zhang, Q. Experimental investigation of mechanical characteristics for linings of twins tunnels with
asymmetric cross-section. Tunn. Undergr. Space Technol. 2021, 119, 104209. [CrossRef]
44. Qiu, J.; Wang, X.; Lai, J.; Zhang, Q.; Wang, J. Response characteristics and preventions for seismic subsidence of loess in Northwest
China. Nat. Hazards 2018, 98, 1909–1935. [CrossRef]
45. Wang, Z.; Shen, S.; Modoni, G. Enhancing discharge of spoil to mitigate disturbance induced by horizontal jet grouting in clayey
soil: Theoretical model and application. Comput. Geotech. 2019, 111, 222–228. [CrossRef]
46. Koyama, T.; Nishiyama, S.; Yang, M.; Ohnishi, Y. Modeling the interaction between fluid flow and particle movement with
discontinuous deformation analysis (DDA) method. Int. J. Numer. Anal. Methods Geomech. 2011, 35, 1–20. [CrossRef]
47. Yang, Q.; He, J.; Luan, M. Comparative study on shear strength of unsaturated red clay and expansive soils. Rock Soil Mech. 2003,
24, 13–16. (In Chinese)
48. Shiau, J.; Al-Asadi, F. Revisiting circular tunnel stability using Broms and Bennermarks’ original stability number. Int. J. Geomech.
2021, 21, 06021009. [CrossRef]
Water 2022, 14, 2486 17 of 17

49. Cho, S.E. Stability analysis of unsaturated soil slopes considering water-air flow caused by rainfall infiltration. Eng. Geol. 2016,
211, 184–197. [CrossRef]
50. Lin, H.; Li, G.; Yu, Y.; Lu, H. Influence of matric suction on shear strength behavior of unsaturated soils. Rock Soil Mech. 2007, 28,
1931–1936. (In Chinese)
51. Mahmood, K.; Kim, J.M.; Khan, H.; Safdar, M.; Khan, U. The probabilistic stability analysis of saturated-unsaturated MH and CL
soil slope with rainfall infiltration. KSCE J. Civ. Eng. 2018, 22, 1742–1749. [CrossRef]
52. Lan, W.; Deng, X.; Sutton, M.A. Investigation of crack tunneling in ductile materials. Eng. Fract. Mech. 2010, 77, 2800–2812.
[CrossRef]
53. Stark, T.D.; Jafari, N.H.; Zhindon, J.S.L.; Baghdady, A. Unsaturated and Transient Seepage Analysis of San Luis Dam. J. Geotech.
Geoenviron. Eng. 2017, 14, 04016093. [CrossRef]
54. Zhang, Y.; Pu, S.; Li, R.T.Y.M.; Zhang, J. Microscopic and mechanical properties of undistributed and remoulded red clay from
Guiyang, China. Sci. Rep. 2021, 10, 18003. [CrossRef] [PubMed]
55. Lu, C.; Qin, W.; Zhao, G.; Zhang, Y.; Wang, W. Better-Fitted Probability of Hydraulic Conductivity for a Silty Clay Site and Its
Effects on Solute Transport. Water 2017, 9, 466. [CrossRef]
56. John, D.; Jaye, C.; Robert, L.; Kemp, G.P. Sediment Deposition at the Caernarvon Crevasse during the Great Mississippi Flood of
1927: Implications for Coastal Restoration. Water 2016, 8, 38. [CrossRef]
57. Zhang, W.; Cai, F.; Zhou, J.; Hua, Y. Experimental Investigation on Air-Water Interaction in a Hydropower Station Combining a
Diversion Tunnel with a Tailrace Tunnel. Water. Stud. 2019, 9, 274. [CrossRef]
58. Xue, Y.; Li, X.; Li, G.; Qiu, D.; Gong, H.; Kong, F. An analytical model for assessing soft rock tunnel collapse risk and its
engineering application. Geomech. Eng. 2020, 23, 441–454. [CrossRef]
59. Huang, F.; Zhao, L.; Ling, T.; Yang, X. Rock mass collapse mechanism of concealed karst cave beneath deep tunnel. Int. J. Rock
Mech. Min. Sci. 2017, 91, 133–138. [CrossRef]
60. Niu, F.; Cai, Y.; Liao, H.; Li, J.; Tang, K.; Wang, Q.; Wang, Z.; Liu, D.; Liu, T.; Liu, C. Unfavorable Geology and Mitigation Measures
for Water Inrush Hazard during Subsea Tunnel Construction: A Global Review. Water 2022, 14, 1592. [CrossRef]