Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

A Case Study of Embankment Retaining Wall in Ontario: October 2016

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/309014331

A case study of embankment retaining wall in Ontario

Conference Paper · October 2016

CITATIONS READS

0 803

3 authors:

Laifa Cao S.M. Peaker


Ryerson University WSP Canada Inc.
33 PUBLICATIONS   281 CITATIONS    15 PUBLICATIONS   10 CITATIONS   

SEE PROFILE SEE PROFILE

Shaheen Ahmad
Abdul Wali Khan University Mardan
47 PUBLICATIONS   852 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

undrained shear strength of soil View project

Using in-situ tests in glacial till View project

All content following this page was uploaded by Laifa Cao on 12 October 2016.

The user has requested enhancement of the downloaded file.


A case study of embankment retaining wall in Ontario
Laifa Cao, Scott Peaker & Shaheen Ahmad
WSP Canada Inc., Toronto, Ontario, Canada

ABSTRACT
This paper presents a case study of an embankment retaining wall. The retaining wall was designed to be up to 6m in
height comprising six layers of rock-filled gabion baskets with base width of up to 4m founded on a very stiff to hard silty
clay to clayey silt till. Three years after wall construction, vertical settlement of up to 138mm and lateral movement of up
to 413mm were recorded at the face of the retaining wall. The significant lateral movement induced cracking in the road
pavement structure. The wall stability and bearing capacity were reviewed. It was found that the wall was underdesigned
for overturning, sliding and global stabilities; and soil bearing capacity was overestimated. The settlement was found mainly
due to the consolidation and overstressing of the founding soils, whereas wall rotation and sliding as well as compaction
and expansion of the stone baskets contributed to the significant lateral movement. Improvements for such retaining wall
design are recommended.

RÉSUMÉ
Cet article présente une étude de cas d’un mur de soutènement de talus. Le mur de soutènement a été conçu pour
atteindre 6 m de hauteur formée de six couches de gabions remplis de roches avec largeur de base jusqu'à 4 m, fondée
sur un très raide d’argile limoneuse dur à limon argileux jusqu'à. Trois ans après la construction du mur, règlement vertical
jusqu'à 138 mm et un mouvement latéral de 413mm ont été enregistrés à la face du mur de soutènement. Le mouvement
latéral significatif induite par fissuration dans la structure de chaussée routière. La stabilité du mur et portance ont été
examinées. Il a été constaté que le mur a été underdesigned pour des stabilités renversement, coulissantes et mondiales;
et sol capacité portante a été surestimé. La colonie est trouvé principalement en raison de la consolidation et sursollicitation
des sols fondateurs, tandis que le mur rotation et glissement ainsi que compactage et l’expansion des ensembles de Pierre
a contribué à l’important mouvement latéral. Améliorations pour une telle structure de mur de soutènement sont
recommandées.

1 INTRODUCTION contributed to the significant lateral movement. Possible


remedial options for the retaining wall are discussed.
Roadway embankments approach a bridge are often Improvements for such retaining wall design are
constrained by retaining walls due to limited space in the recommended.
right of way. Flexible retaining walls such as reinforced
earth wall and gabion wall are generally used as they are
considered to be tolerant of the ground settlement under 2 SITE CONDITIONS AND RETAINING WALL
the weight of embankment fill. However, if the ground CONSTRUCTION
conditions are not properly assessed, ground settlement
and lateral movement of such retaining walls could The site of the approach embankments of a road bridge
significantly affect the serviceability of the roadway. (road over rail) is located in Ontario, Canada. Due to the
This paper presents a case study of an embankment road bridge replacement and road widening, the approach
retaining wall. The retaining wall was designed to be up to embankments needed to be reconstructed. Since the east
6m in height comprising six layers of rock-filled gabion side of the embankment was constrained by private
baskets with base width of up to 4m founded on a very stiff property, a retaining wall was required to be constructed to
to hard silty clay to clayey silt till over firm to very stiff silty retain the embankment fill. A geotechnical investigation (by
clay till. Three years after wall construction, vertical others) including borehole drilling, standard penetration
settlement of up to 138mm and lateral movement of up to tests (SPT) and laboratory testing of soil index parameters
413mm were recorded at the face of the retaining wall. The was carried out to support the design of the bridge and
significant lateral movement induced cracking in the road retaining wall.
pavement structure. The wall stabilities, soil bearing In general, the subsurface conditions encountered at
capacity, ground settlement under the embankment fill, and the site consist of 3.0 to 5.6m thick, loose to compact sandy
wall lateral movement were reviewed. It was found that the silt fill extending to El. 96.0 to 94.4m, underlain by about
wall was underdesigned for overturning, sliding and global 5.5 to 7.6m thick, very stiff to hard silty clay to clayey silt till
stabilities; and soil bearing capacity was overestimated. extending to El. 90.6 to 86.8m, which in turn is underlain by
The settlement was attributed mainly to the consolidation firm to very stiff silty clay till to El. 79.7 to 79.2m. The firm
and overstressing of the founding soil under the weight of to very stiff silty clay is underlain by hard clayey silt till or
embankment fill, whereas the wall rotation and sliding as very dense silt till to sandy silt till. The water contents were
well as compaction and expansion of the stone baskets 9% to 17% for the sandy silt fill, 12% to 16% for the very
founded on the very stiff to hard silty clay to clayey silt till at
El. 95.5 to 94.1m. Soil parameters including friction angle,
unit weight, and coefficient of lateral earth pressure were
also recommended in the geotechnical investigation report.
Based on the soil bearing capacity and parameters
recommended in the geotechnical investigation report, a
gabion retaining wall was designed to be up to 6m in height
comprising six layers of gabion baskets with base width of
up to 4m founded at El. 94.3m on the very stiff to hard silty
clay to clayey silt till. A typical cross section of the gabion
wall is shown in Figure 2. The wall face and base had an
inclination angle of 6o (i.e. 1H:9.5V) in order to increase
the base resistance and resisting moment. The wall was
designed with 1m height of fill in front of the wall toe. A
drainage pipe was also proposed behind the wall. It is
noted that the wall was designed to retain a 1V:2H slope
backfill above the wall. The new fill height at the road centre
was about 2.8m (above the existing level prior to bridge
replacement) and the maximum new fill height at the edge
of the road was about 7.8m.
The gabion retaining wall was constructed in the
summer of 2007.
Figure 1. Soil profile and SPT N-values at the site

stiff to hard silty clay to clayey silt till, and 17% to 18% for 3 FIELD OBSERVATIONS
the firm to very stiff silty clay till. It is noted that the pocket
penetrometer tests on the firm to very stiff silty clay till only Shortly after the gabion wall was constructed and the road
showed the undrained shear strengths of about 25 kPa, was opened, longitudinal pavement cracks were observed.
indicating a relatively weak condition of the firm to very stiff In response, a movement monitoring program of the gabion
silty clay till. Field vane shear strength testing was not retaining wall was initiated. The monitoring points were
undertaken by the geotechnical consultant. The steel bars driven into rock-filled gabion baskets along the
groundwater level was at El. 87.8 to 86.9m during the top of the retaining wall. The measurements started in
investigation. Figure 1 shows soil profiles and SPT N- November 2007, a few months after the completion of the
values. retaining wall construction. The development of vertical
The geotechnical report recommended an allowable settlement of up to 138 mm and horizontal deflection of up
bearing capacity of 200 kPa and factored ULS (ultimate to 413mm had been recorded by July 2010.
limit states) bearing capacity of 300 kPa for a retaining wall

Figure 2. A typical section of designed gabion wall


Photo 3. Tensile cracks above wall and tilted guide rail
Photo 1. Wall tilt near abutment (the top of fence post was
pushed laterally) weights was applied. The unit weight of the foundation soil
and backfill was taken as 20 or 22 kN/m3. The friction angle
A field visit of the site was carried out in June of 2010 was taken as 30o or 32o for the foundation soil, 40o for the
to examine the condition of retaining wall. There were friction between gabions, 15o or 16o for the friction between
visual signs that both vertical and lateral movements were the wall and the backfill, and 22o or 29o for the friction
being experienced by the retaining wall and the approach between the wall and foundation soil. The lower bound
embankment. The maximum height of the wall was values were generally similar to those proposed in the
measured as 4.8 m instead of 5.0 m above the natural geotechnical investigation report for this project.
ground level. The tilt of wall was observed as shown in Analyses of the following four cases were carried out:
attached Photo 1. Uneven settlement was observed in the (1) 6m high gabion wall as designed with lower bound
CSP culvert underlying the north approach fill as shown in of soil parameters;
attached Photo 2. Longitudinal cracks had developed in (2) 6m high gabion wall as designed with higher
the pavement and granular fill along the side of the gabion bound of soil parameters;
wall as shown in attached Photo 3. (3) 5.8m high gabion wall measured on July 23, 2010
with lower bound of soil parameters;
(4) 5.8m high gabion wall measured on July 23, 2010
4 REVIEW OF RETAINING WALL DESIGN with upper bound of soil parameters.
For each case, two groundwater level (GWL) scenarios
The stability of the east retaining wall including bearing, were modelled, one with GWL assumed at El. 87.8m
sliding, overturning and global stabilities, vertical (lower GWL, the groundwater level measured in the
settlement and lateral movement were reviewed using the monitoring well during the geotechnical investigation) and
available data. A parametric approach to the analysis, another with GWL at the base of wall (higher GWL).
using a plausible range of soil shear strengths and unit
4.1 Bearing Capacity

It is well known that the soil bearing capacity of an earth


retaining structure is not only controlled by the founding soil
properties but is also affected by the lateral load acting on
a retaining structure. The soil bearing capacity for a
foundation with a lateral load will be significantly smaller
than that without a lateral load. However, this is often poorly
communicated in geotechnical investigation reports.
The soil ultimate bearing capacity was estimated based
on the formula proposed by Vesic (1973, 1975), in which
the influence of lateral load on the bearing capacity is
considered. Table 1 summarizes the soil ultimate bearing
for the four cases under higher GWL and lower GWL
conditions. Taking the vertical bearing resistance factor of
0.5 as recommend in Canadian Foundation Engineering
Manual (2006), the factored ULS bearing capacity was
calculated and is shown in Table 1. It was found that the
Photo 2. Separated and sagged culvert below approach fill factored ULS bearing capacity was lower than 300 kPa
Table 1. Soil bearing capacity

Case Ultimate Factored ULS Allowable


Lower Higher Lower Higher Lower Higher
GWL GWL GWL GWL GWL GWL
1 351 256 176 128 117 85
2 534 395 267 198 178 132
3 338 248 169 124 175 83
4 512 381 256 191 173 127
Unit: kPa.

as recommended in the geotechnical investigation report


for all cases. At the higher GWL, the estimated ULS
bearing capacity with the consideration of the lateral earth
pressure was only 43% to 66% of that recommended in the
geotechnical investigation report, in which the influence of
lateral load to the bearing capacity was probably not
considered. At the serviceability limit states (SLS), the
bearing capacity is typically defined by settlement criteria
under the load and resistance factor design (LRFD).
However in common practice, the bearing capacity at the Figure 3. A typical result of overall stability analysis
SLS is often taken as the same as the allowable bearing
capacity in the traditional working stress design (WSD). Table 2. Factor of safety against stability failures
Taking a global factor of safety of 3, the allowable bearing
capacity was calculated and is shown in Table 1. Again the Case Min. FOS Min. FOS Min. FOS against
estimated allowable bearing capacity with the against against overall
consideration of the lateral earth pressure was only 43% to overturning sliding
66% of the recommended allowable bearing capacity in the Lower Higher
geotechnical investigation report. GWL GWL
It is noted the vertical stress at the retaining wall base 1 1.81 1.36 1.39 1.35
level ranged from 176 to 194 kPa at the location of the road 2 1.85 1.96 1.48 1.44
centre and ranged from 116 to 132 kPa at the location of 3 1.69 1.15 1.38 1.34
the retaining wall. If the influence of the lateral earth 4 1.72 1.51 1.47 1.38
pressure was not considered, the allowable bearing
capacity of 200 kPa recommended in the geotechnical designed wall as the FOSs for the wall in 2010 were less
report met the vertical load requirement. However, under than the designed wall under the same condition. This is
the action of the lateral earth pressure, the average working primarily due to the loss of batter, i.e. the wall became
load on the effective base of the retaining wall ranged from vertical in 2010.
206 to 212 kPa, which was greater than the allowable The factor of safety against basal sliding for the wall in
bearing capacity recommended in the geotechnical report. 2010 was in the critical condition with a FOS of 1.15 when
It seems that the working load was underestimated and the groundwater level was modelled to lie at the base of the
soil bearing capacity was overestimated without the gabion wall if the soil parameters proposed in the
consideration of the influence of lateral earth pressure on geotechnical report were used.
the work load and the soil bearing capacity during the It is noted that the outlet of the drainage pipes for the
retaining wall design for this project. wall in 2010 was located close to the existing ground
surface. When the groundwater level at the wall toe
4.2 Stabilities reaches the ground surface, local sliding failure would
occur. In the case of the groundwater table rising up at the
A retaining wall should be designed for overturning, sliding back of wall due to clogging of filter fabric at the back of
and overall (global) stabilities. Although the LRFD has been wall, local sliding failure would also occur. This could be
widely used for structure design, the WSD is still used for one reason that a lateral movement of 413mm was
retaining wall design in the checking of the overturning, recorded. Bulging of gabion baskets may also be a
sliding and overall (global) stabilities. Table 2 summarizes contributing factor.
the factor of safety (FOS) against these stability failures
under the working stress condition for the four cases. 4.3 Settlement and Lateral Movement
Figure 3 shows a typical result of overall stability analysis
using commercial limit-equilibrium software. The monitoring results for vertical settlement and lateral
The analyses indicate that the overturning and overall movement for monitoring points 37 to 45 are shown in
stabilities of the gabion wall are marginally acceptable, but Figures 4 and 5, respectively. Monitoring points 37 to 42
less than the generally accepted requirements of a FOS of were installed at the wall top with the wall height of less
2 for overturning and FOS of 1.5 for overall stability. The than 5m. Monitoring Point 43 was installed at the wall top
stability of the gabion wall in 2010 was worse than the with the wall height of 5.0m, Point 44 was positioned at the
Figure 4. Vertical movement measured at wall top Figure 5. Horizontal movement measured at wall top

wall top with the wall height of 5.8m, and Point 45 at the
corner of the wall near the bridge abutment wall. The
measurements began in November 2007, a few months
after retaining wall construction. The development of
vertical settlement of up to 138mm and horizontal
deflection of up to 413mm was recorded in July 2010,
relative to the November 2007 baseline. The monitoring
results show that the settlement of the wall due to soil
consolidation could be considered as being essentially
complete by August 2008, one year after wall construction
(See Figure 4). The maximum consolidation settlement
was 96mm, which was close to the settlement of the north
approach embankment of about 100 mm. From August
2008 to July 2010, the increment of vertical movement at
points 37 through 42 (wall height of less than 4.5m) was
less than 10mm and remained comparatively stable;
whereas the increment of vertical settlements at points 43 Figure 6. Comparison between predicted and measured
through 45 (wall height of 5.0 to 5.8m) ranged from 22 to settlements
50mm and did not slow down. The significant vertical
settlement of the highest wall sections may be partially due height. Many baskets appeared to be deformed under the
to overstressing of the founding soils which was induced by self-weight load, suggesting that they were not tightly
the loss of wall batter. packed with stones.
From November 2007 to July 2010, a maximum lateral Since no consolidation testing was carried out for this
movement of 413 mm was recorded at monitoring point 44 project, an accurate estimation of the consolidation
installed in the highest retaining wall section (see Figure 5). settlement is impossible. An approximate estimation of the
Similar to the vertical movements, the lateral movements vertical settlement can be conducted using empirical data
at points 37 to 42 have slowed down since January 2009, for the glacial tills in Ontario as recommended by Cao, et
whereas the lateral movements at points 43 and 44 in the al. (2015). Assuming that the compression index (Cc) is
highest wall section did not decline. A relatively small 0.037 for the very stiff to hard silty clay till and 0.1 for the
lateral movement of 50mm at point 45 installed at the firm to very stiff silty clay till, the recompression index (Cr)
corner of the wall is reasonable as the gabion wall was is 0.081 for the very stiff to hard silty clay till and 0.016 for
butted against 3m high RSS (retained soil systems) wall at the firm to very stiff silty clay till, the initial void ratio (eo) is
the corner. The significant lateral movement could not be 0.36 for the very stiff to hard silty clay till and 0.46 for the
purely due to the vertical settlement and the wall rotation. firm to very stiff silty clay till, and the coefficient of
Sliding of the gabions may contribute part of the lateral consolidation (cv) is 20 m2/year, the total ground settlement
movement. Some component of the apparent lateral was 220mm including 90mm in the first three months after
movement may also be a result of expansion (bulging) of the construction of 2.8 to 7.8m height of the embankment.
the gabion baskets. Workmanship in construction of the Even through the selected soil parameters were based on
gabion baskets might not have been of the highest quality. the empirical data, the predicted settlements are in good
The measurements in 2010 showed that individual basket agreement with the measured settlements as shown in
heights varied from 800mm to the design filled height of Figure 6. The back-analysis indicates that 95% of the
1000mm, but most baskets were 850 mm to 900 mm in consolidation settlement would be reached by July 2010.
5 DISCUSSIONS ON REMEDIAL WORKS The last option is to do nothing as the majority of the
consolidation settlement of the underlain silty clay till has
The monitoring data show that the vertical settlement and been reached, the soil strength has been improved under
lateral movement of the highest wall sections have not the embankment fill, and the retaining wall is marginally
attenuated. Review of the wall stability indicates that the stable. Continuous monitoring of the wall movement is
local bearing shear failure and sliding failure may occur required if this option is chosen. More frequent
when the groundwater level reaches the ground surface at maintenance of the road pavement will be required as the
the toe of the wall. These suggest that remedial work of tensile cracks in pavement structures could continuously
the highest wall section may be required. develop due to the lateral movement of the retaining wall.
Overstressing of the foundation soils beneath the
highest sections of the gabion wall appears to be the main
cause of settlement and tilting. The monitoring did not show 6 CONCLUSIONS AND RECOMMENDATIONS
an abating trend in settlement or deflection in the highest
sections of the wall. From the review of an embankment retaining wall, the
If the horizontal component of the load on the retaining following conclusions can be provided:
wall (i.e. the lateral earth pressure) can be significantly • The working pressure on a retaining wall base
reduced, then the inclination of the resultant force acting on was higher than the vertical load, due to the
the base of the wall can be rendered essentially vertical, contribution of the lateral earth pressure. For a
without significant eccentricity. preliminary analysis of a gravity retaining wall, the
Three methods of picking up or reducing the lateral working pressure can be assumed as 150% to
earth pressure include: 200% of the vertical load.
(i) Soil nailing; or • The bearing capacity for the retaining wall design
(ii) Installation of closely spaced layers of horizontal was significantly underestimated, probably
geogrid; or without the consideration of the lateral earth
(iii) Replacement of the backfill in the active wedge pressure. For a preliminary analysis of a gravity
with light weight fill (expanded slag, expanded retaining wall, the bearing capacity can be taken
polystyrene or equivalent). as 50% of that without consideration of a lateral
The gabion wire baskets and gabion stones within the load.
baskets present some practical difficulties as far as the • The significant vertical settlement is attributed to
advancement of horizontal soil nails through the wall is not only the soil consolidation, but also the
concerned. Although it might be technically feasible to overstressing of the founding soils which was
vibro-drive steel bar through the gabions, without the induced by the loss of wall batter.
benefit of field-trials to assess the feasibility of advancing • The significant lateral movement is contributed to
soil nails through the cages and tension tests to assess not only the wall rotation and sliding, but also the
their capacity, this method is not considered contractually compaction and expansion of the stone baskets
viable. For this reason, the second alternative – installation under the self-weight load.
of geogrid in the backfill soils behind the gabion wall in the For the preparation of a geotechnical investigation
north approach fill is more reliable. report for a retaining wall, a recommendation for a global
Option (ii) will require reconstruction of most, if not all, stability analysis is required, and a settlement analysis
of the north approach fill and roadway infrastructure. must be carried out if the founding soils are underlain by
The benefit of the geogrid fill reinforcement option is relatively weak soils. Since details regarding retaining wall
that the existing gabion wall is essentially converted from a design are often not know at the time of preparation of a
gravity retaining structure to a facing wall only. A secondary geotechnical investigation report, it is critical that the
benefit is that the upper sloped portion of the approach fill geotechnical engineer be engaged at the design stage to
(which lies above the top of the gabions and below the reassess the recommended soil parameters and to assess
asphalt road surface), can be reinforced. This area the wall stability.
exhibited creep and surface erosion, resulting in tension
cracking of the pavement and tilting of the guide rail.
Reinforcement of the north approach fill would require REFERENCES
road (bridge) closure during the full course of the work. The
existing approach fill material can be salvaged and reused Canadian Geotechnical Society, 2006. Canadian
to create the reinforced fill. The face of the fill abutting the Foundation Engineering Manual, 4th ed., BiTech
existing gabions wall will require a filter fabric lining, Publisher Ltd, Richmond, BC, Canada.
followed by a wrap of the geogrid. In this way, the Cao, L.F., Peaker, S. and Ahmad, S., 2015. Engineering
reinforced fill will remain independent of the gabion wall. characteristic of glacial tills in GTA, 68th Annual
The potential conflict with existing reinforcing strips to Canadian Geotechnical Conference, Quebec, Canada.
the bridge abutment might also apply to Option (iii). This Vesic, A.S. 1973. Analysis of ultimate loads of shallow
option involves the sub-excavation of existing fill behind the foundations, JSMFD, ASCE, 99: 45-73.
retaining wall within the active wedge zone and replacing Vesic, A.S. 1975. Foundation Engineering Handbook, 1st
this fill with compacted light weight fill, such as expanded ed., Winterkorn and Fang, Van Nostrand Reinhold.
slag. Alternatively, the use of ultra-light weight fill
(expanded polystyrene) could also be considered.

View publication stats

You might also like