Road submergence during flooding and its effect on subgrade strength

Abstract

Deterioration of road structural integrity because of flooding may cause huge expenditure for rehabilitation and maintenance of roadway. In principle, the design of pavement structure is based on the strength of compacted soil known as the subgrade or road foundation. Therefore, subgrade is a significant part of the road structural system. When roads are inundated for a long time or repeatedly, the materials in each layer of road structure become saturated, and the original condition of subgrade soils will be compromised. This study investigated the effect on sub-grade strength and properties due to road submergence period and repeated submergence of the road structural systems. Two types of soil that are normally used as the embankment material in road construction, which can be categorized as cohesive and cohesionless materials, were used in this study. California Bearing Ratio (CBR) test and consolidation settlement test were carried out on various categories of inundation and loading conditions including repeated inundation. The findings indicated that the strength of subgrade soil further decrease when they are inundated for a longer period. Similarly, consolidation test also shows that a quick and higher settlement could occur when the soil is inundated for a longer period. These findings are useful for road design and maintenance strategies of flood affected road links.

Full text

1848
International Journal of GEOMATE, May, 2016, Vol. 10, Issue 21, pp. 1848-1853
Geotec., Const. Mat. & Env., ISSN: 2186-2982(Print), 2186-2990(Online), Japan
ROAD SUBMERGENCE DURING FLOODING AND ITS EFFECT ON
SUBGRADE STRENGTH
Abdul Naser Abdul Ghani1, Nur Izyan Roslan2and Ahmad Hilmy Abdul Hamid3
1,2,3 School of Housing Building and Planning, Universiti Sains Malaysia, Malaysia
ABSTRACT: Deterioration of road structural integrity because of flooding may cause huge expenditure for
rehabilitation and maintenance of roadway. In principle, the design of pavement structure is based on the
strength of compacted soil known as the subgrade or road foundation. Therefore, subgrade is a significant part of
the road structural system. When roads are inundated for a long time or repeatedly, the materials in each layer of
road structure become saturated, and the original condition of subgrade soils will be compromised. This study
investigated the effect on sub-grade strength and properties due to road submergence period and repeated
submergence of the road structural systems. Two types of soil that are normally used as the embankment
material in road construction, which can be categorized as cohesive and cohesionless materials, were used in this
study. California Bearing Ratio (CBR) test and consolidation settlement test were carried out on various
categories of inundation and loading conditions including repeated inundation. The findings indicated that the
strength of subgrade soil further decrease when they are inundated for a longer period. Similarly, consolidation
test also shows that a quick and higher settlement could occur when the soil is inundated for a longer period.
These findings are useful for road design and maintenance strategies of flood affected road links.
Keywords: Flooding, Road Sub-grade, Pavement
1. INTRODUCTION
Floods have resulted in many undesirable
outcomes on human and the environment. This is
mainly due to their adverse impacts on humans,
properties, environmental surrounding, road
structures and many forth. Additionally, human
activities can also contribute to the flooding event
which include: (i) farming and deforestation that
exposes the soil to erosion and increases surface
runoff, (ii) urbanization by unplanned building
construction in vulnerable areas without following
the regulations of town planning, poor watershed
management and failure to control the flooding
promptly, and (iii) obstruction of natural flow of
water through drainage modification [1]. The impact
of flood disaster can be significant after the event
since it may impact the whole infrastructure
involved as well as can have long term effect in
terms of the maintenance work. In the long run,
flooding can bring the deterioration to road
pavement foundation when the phenomenon keeps
repeating. Continuous flood submersion of roads
could bring damage on large part of the road
infrastructure, thus affecting the stability of asphaltic
concrete pavement layer [2].
Damages of roads structure due to flood event
are commonly causing a huge expenditure for the
rehabilitation and maintenance works of roads.
Recently, Malaysian federal government has
allocated RM42 million to repair the embankment
collapsed in federal roads damaged by floods in the
state of Kelantan. Previously, Bernama [3] reported
that the state of Terengganu spent more than RM 74
million for flood damaged roads of only one flood
event.
Similarly in town of Sibu in the state of Sarawak
it was reported that the city requires RM 500 million
to repair road related infrastructure damaged by
flooding [4]. Malaysian federal government has also
allocated RM 106 million [5] to repair federal roads
damaged by flash floods in between October 2012
and January 2013 (monsoon season). Figure 1 is a
typical road inundated during flash flood.
Monsoon season will always come every year
and it is only logical to expect the same or more
amount of money will be required to reinstate the
damaged roads annually. On the other hand, when
the roads are inundated for a long time, the materials
in each layer for road pavement become saturated,
Fig.1: Typical road inundation during flash flood

International Journal of GEOMATE, May, 2016, Vol. 10, Issue 21, pp. 1848-1853
1849
and then as floodwaters drained, the sub-grade soils
began to shrink and subside. The excessive water
can drain into the foundation reducing its load
bearing efficiency. Thus, this situation can cause the
strength of road pavement systems to be
compromised.
1.1 The Sub-grade
The sub-grade is a foundation for the pavement
structure to support the load from upper layer to the
beneath soil. Basically, sub-grade must be stable in
performance to carried load in any weather
conditions. Generally in road engineering, CBR test
is performed to determine the strength of sub-grade
soil and these CBR values will be used to design the
thickness of flexible pavement [6]. As soil is a
highly variable engineering material due to its
composition and the dependence of its properties on
environmental conditions, it is logical to evaluate the
effect of the variability associated with sub-grade
strength on pavement design and performance [7],
[8]. Generally, no systematic study has been
performed to evaluate the effects of weak, variable
sub-grade conditions on pavement design,
construction, and performance prediction [9], [10].
Change of the properties of sub-grade soil is
definitely affecting the performance of pavement
structure.
Fairweather and Yeaman [11] recently studied
the influence of flooding on road pavement
deterioration and recommended further research to
better predict pavement failure.
2. EXPERIMENTAL SETUP
The samples of sub-grade soil were taken from
two different quarries that supply soil which is
usually used as the embankment soil in road works.
The soil samples were categorized as cohesive
material and cohesionless material according to
Standard Specification for Road Works by Public
Work Department (JKR), Malaysia. The properties
of soil are shown in Table 1.
Table 1 Properties of Soil
2.1 Sample Preparation
Initially, the experiments such as grain size
distribution and index properties were conducted to
find out the different properties of soil. After that,
the CBR test was performed on unsoaked and
soaked specimens with different days of
submergence. Additional set of specimens were kept
submerged repeatedly under water for certain period
to simulate repeated flooding. Finally, the Odometer
consolidation apparatus were used to determine the
progressive settlement of soil specimens.
2.2 CBR Test
CBR test was carried out according to the
BS1377. 4300g of soil was compressed in the mould
and assign to unsoaked, soaked and repeated
submerged condition. The unsoaked soil sample is
tested immediately after the soil being compressed
into mould, while soaked soil sample is tested for its
strength after being soaked for 1, 3 and 7 days.
Furthermore, to simulate the effect of repeated
inundation the samples were kept in water for 1 hour
on Day 1, Day 3 and Day 7. The penetration was
measured using a dial gage which has accuracy
0.01mm.
2.3 Oedometer Consolidation Test
The Odoemeter test was carried out according to
the ASTM D2435. In this research, the Oedometer
was modified by using data logger for data recording.
The standard Oedometer test is carried out on a
cylindrical specimen of saturated soil with the
dimension of 75 mm diameter and 20 mm thick. The
soil sample is enclosed in a metal ring and is placed
on a porous stone. The soil samples were prepared
for 1-day submerged and 3-days submerged before
tested. The test involves applying increments of 1kg,
2kg, 4kg, 8kg, 16kg and 32kg of vertical static load
to the sample and recording the corresponding
settlement. The time intervals were 6s, 15s, 30s, 60s,
120s, 240s, 480s, 900s, 1800s, 3600s, 7200s, 14400,
28800s and 86400s.
3. RESULTS AND DISCUSSION
The mechanical and physical properties for the
two types of the tested soils were determined and the
particle size distribution for the tested soils,
according to American Association of State
Highway and Transportation Officials (AASHTO),
is shown in Fig. 2 indicating that the percentage
passing no. 200 sieve for Soil 1 and Soil 2 are 0.6%
Soil Properties Values
Soil 1
(Well-graded sand
with clay and gravel)
Liquid limit
Plastic limit
Plasticity
index
78.5
34.2
44.3
Soil 2
(Silty clay and
gravel)
Liquid limit
Plastic limit
Plasticity
index
88.8
38
50.8

International Journal of GEOMATE, May, 2016, Vol. 10, Issue 21, pp. 1848-1853
1850
and 1% respectively. Based on the AASHTO soil
classification system, both soils belongs to soil
group which usual types of significant constituent
materials are silty or clayey gravel and sand.
Moreover, according to the Unified Soil
Classification System (ASTM D2487), the tested
soils were found and can be classified in group name
of “well-graded sand with clay and gravel” for the
Soil 1, whereas for the Soil 2 is “poorly graded silty
clay and gravel”.
Fig. 2 Particle Size Distribution Curve
3.1 CBR Performance
Soil 1
The result of CBR strength presented in Fig. 3
illustrated the comparison of soil strength for
unsoaked and soaked condition. The soil samples
were inundated for 1, 3 and 7 days for soaked
condition. From the bar chart, it shows the CBR
value for unsoaked condition relatively higher than
CBR value for soaked condition due to the saturated
period for soaked soil samples. It shows that the
CBR value for unsoaked condition was 35.7% and
on submerging the soil samples for 1, 3 and 7 days,
the CBR values were 15%, 12.2% and 8.6%
respectively. Generally, the soil strength has been
reduced by 76% from the condition of unsoaked
sample to the 7-day soaking sample. Obviously, the
presence of water when the soil had been soaked for
1, 3 and 7 days contributes to the decreasing of soil
strength. Soil had been loss strength starting on 1-
day soaking when it compared to the unsoaked
condition. Soil sample in unsoaked condition show
its capability to sustain the higher load since it is
evident that there have no subsequent loss of
strength. The unsoaked sample basically showed
better performance on their strength and the CBR
strength probably can be increased with well
compacted on soil tested.
However, the CBR value for soaked condition
decreased with the strength accordingly due to the
number of inundation days for each soaked soil
samples. It was found that further increase in the
number of days of soaking decreases the CBR value
gradually and it is also observed that the loss of
CBR value between conditions of 1 day until 7 days
soaking. Significant loss of strength was observed
caused by inundation and subgrade soil becomes
saturated within the soaking period. From the results,
it is concluded that the value of CBR for the given
soil sample decreases rapidly from unsoaked
condition to 1 day of soaking. Additionally, it is also
observed that the variation between 1-day and 7-day
soaking values are quite different. Soil had been loss
more strength on 7-day soaking compared to 3-day
soaking since the percentage of CBR value decrease
from 12.2% to 18.6% respectively. The volume of
soil has been changed effect from the soaking
condition, thus the strength of soil become less due
to number of inundation days.
Fig. 3 Comparison of CBR values of inundated
samples (Soil 1)
Meanwhile, the bar chart in the Fig. 4, for the
repeated submerged condition has shown different
result of unsoaked and soaked condition. The soil
samples were submerged for 1 hour only on day-1,
day-3 and day-7. It shows that the CBR value for
unsoaked condition was 35.7% and on repeated
submerging for 1 hour on day-1, day-3 and day-7,
the CBR values were 25%, 15.9% and 18.5%
respectively. Basically, the result shows that
unsoaked condition still have the higher CBR
strength value when it compared to the repeated
submerged condition of soil samples. In the repeated
submerged case, the CBR strength was reduced on
day-1 after submerged for 1 hour and subsequently
the CBR value also reduced on day-3 compared to
the unsoaked sample. However, on the day-7, the
soil sample was gaining its strength again when
inundated for 1 hour after on the day 3. The CBR
strength was increased by 16% after submerged for
1 hour on day-7.
Soil sample in unsoaked condition show its
capability to sustain the higher load since it is
evident that there were no subsequent loss of
strength. Moreover, the CBR values for repeated
submerged condition is higher than soaked condition
since the soil was only inundated for short period
when it compared to the soaked condition which the
soil has been inundated for a longer period. CBR

International Journal of GEOMATE, May, 2016, Vol. 10, Issue 21, pp. 1848-1853
1851
values were strongly affected by the long-term
inundation compared to the case of repeated
submerged condition. Since the soil 1 was
categorized as well-graded sand with clay and gravel,
the possibility of soil to gain the strength on day-7
after submerged on 1 hour is easier because of the
soil particle and lower pore water pressure itself.
Fig. 4: Comparison of CBR of repeated
submergence samples (Soil 1)
Soil 2
Figure 5 presented the result of CBR value for
the second soil samples which comparing between
unsoaked and soaked condition of soil samples. The
test and soil conditions were conducted similar with
the first soil samples. From the bar chart, it shows
the CBR value for unsoaked condition relatively
higher than CBR value for soaked condition due to
the saturated period for soaked soil samples. It
shows that the CBR value for unsoaked condition
was 22.9% and on submerging the soil samples for 1,
3 and 7 days, the CBR values were 10.7%, 6.84%
and 3.42% respectively. Generally, the CBR value
for both conditions on second soil quite different
from soil 1 since the soil 2 was categorized as silty
clay and gravel. CBR strength has been reduced by
85% from the condition of unsoaked sample to the
7-day soaking sample. Basically, the second soil
sample in unsoaked condition shows its capability to
sustain the higher load similar with the first soil
samples since it is evident that there has no
subsequent loss of strength. However, the CBR
strength of soil samples for soaked condition was
decreased due to submerging time.
The unsoaked sample basically showed better
performance on their strength and the CBR strength
probably can be increased with better compaction
before the soil will be tested. Meanwhile, soil that
soaked for 7 days show the deterioration of its
strength performance compared to the 1 and 3 days
of soaking condition. Significant loss of strength
was observed caused by inundation and subgrade
soil becomes saturated within the soaking period.
From the results, it is concluded that the value of
CBR for the given soil sample decreases rapidly
from unsoaked condition to 1 day of soaking. Soil
had been loss more strength on 7-day soaking
compared to 3-day soaking.
Fig. 5: Comparison of CBR values of inundated
samples (soil 2)
On the other hand, the bar chart in the Fig. 6
shows the repeated submerged condition seen
differently compared to the unsoaked and soaked
condition. The soil samples were submerged similar
with soil 1 condition which is the soil samples were
submerged for 1 hour only for 1, 3 and 7 days. It
shows that the CBR value for unsoaked condition
was 22.9% and on repeated submerging for 1 hour
on day-1, day-3 and day-7, the CBR values were
13.5%, 8.05% and 5.25% respectively. Basically, the
result shows that unsoaked condition still have the
higher CBR strength value when it compared to the
repeated submerged condition of soil samples. In the
repeated submerged case, the CBR strength was
reduced on day-1 after submerged for 1 hour and
subsequently the CBR value also reduced on day-3
and day-7 compared to the unsoaked sample. The
reduction of strength was 61.1% from day-1
submerged for 1 hour to day-7 submerged for 1 hour.
This condition occurs probably due to clay condition
which its properties consist of small particle size
which tends to be very dense. The density of clay
that thicker and heavier than other soil types will
takes longer time to clay particles absorb this water,
and further slowing the flow of water through the
soil.
Fig. 6 Comparison of CBR values of repeated
submergence samples (Soil 2)
3.2 Consolidation Performance

International Journal of GEOMATE, May, 2016, Vol. 10, Issue 21, pp. 1848-1853
1852
Consolidation is a process by which soils
decrease in volume. In this study, the focus is to
observe the settlement effect from various loading
capacity. Figure 7 and Figure 8 show the graph of
settlement against time for soil condition of 1-day
(24 hours) submerged and 3-days (72 hours)
submerged respectively. For 1 day submerged
condition, as shown in Fig. 7, the settlement
increases due to the increment of loading. Generally,
the load of 1kg, 2kg and 4 kg obtain quick initial
settlement at 60s before the soil reached at the
optimum settlement. In addition, the load for 8kg,
16kg and 32kg take longer time to reach the constant
settlement. The average time to soil reached the
constant settlement was about 2 hours. The higher
settlement takes places when 32kg load was applied
and reach the constant settlement at 1.52mm.
Fig. 7 Graph of settlement against time for 1-day (24
hours) submerged sample
Fig. 8 Graph of settlement against time for 3-days
(72 hours) submerged sample
Figure 8 shows the settlement of soil against time
for the 3-days submerged condition. The pattern line
of graph can be seen quite different when compared
to soil samples of 1-day submerged condition since
the initial settlement occurs for each load is higher
than 1-day submerged condition. This is because the
longer period of inundation cause the quick and
higher initial settlement. The higher settlement takes
places when 32kg load was applied and reach the
constant settlement at 1.70mm which higher than the
1-day submerged condition but the consolidation
occur in a relatively short time to reach the constant
settlement.
The test provides a reasonable estimate of the
amount of settlement on soil samples However, the
rate of settlement is often underestimated, that is, the
total settlement is reached in a shorter time than that
predicted from the test data. This is largely due to
the size of sample which does not represent soil
fabric and its profound effect on exact conditions.
Besides the natural condition of the sample,
sampling disturbance will have a more pronounced
effect on the results of the test done on small
samples. Furthermore, the boundary effect from the
ring enhances the friction of the sample. The friction
reduces the stress acted on the soil during loading
and reduces swelling during unloading.
4. CONCLUSION
Experimental study has been carried out to
determine the strength of soil samples when tested in
different inundation conditions. The CBR strength
for both soils samples indicated the decreases of its
strength due to higher increment number of
inundation days. It can be concluded that the
strength of soil further decrease when they are
inundated for a longer period. Similarly,
consolidation test also shows that a quick and higher
settlement could occur when the soil is inundated for
a longer period. A more extensive testing will
provide the basis for the inclusion of inundation
effect in the road design procedures.
5. ACKNOWLEDGEMENTS
The authors gratefully acknowledge financial
support from FRGS research grant no.
USM/203/PPBGN/6711257 and NAPREC grant no.
USM/401/PPBGN/510059/I-124.
6. REFERENCES
[1] Adedeji, A. A. and Salami A. W., (2008):
Environmental hazard: Flooding and Its Effects
on Residential Buildings in Ilorin, Nigeria.
http://www.scribd.com/doc/9268022/ . Date
Accessed Jan 29 2015.
[2] Bowoputro, H. et al. (2009): Temperature
Influence and Lapindo Mud Submersion
towards the Stability Value of Asphaltic
Concrete Mix, Rekayasa Sipil Journal, Volume
3, No.3 1978 – 5658.
[3] Bernama (2009): RM 21 Juta Untuk Baiki Jalan
Rosak Akibat Banjir. Bernama.Com. December
22, 2009.
[4] mySarawak (2009): RM 500 juta bina
infrastruktur rosak akibat banjir di Sibu.
mySarawak, February 8 , 2009
[5] The Star (2013): RM 106 Juta Diperuntuk
Untuk Baik Pulih Jalan Persekutuan Rosak
Akibat Banjir. The Star, Jauary 11, 2013.
[6] Naagesh, S., Sathyamurthy, R., Sudhanva, S.
(2013): Laboratory Studies On Strength and

International Journal of GEOMATE, May, 2016, Vol. 10, Issue 21, pp. 1848-1853
1853
Bearing Capacity Of GSB-Soil Sub-Grade
Composites.
[7] Deshpande, V. P., Damnjanovic, I., and Gardoni,
P. (2008): Modeling The Effects of
Rehabilitation Actions on The Reliability of
Flexible Pavements, Proc., Transportation
Research Record 87th Annual Meeting,
Transportation Research Board, Washington,
D.C.
[8] Tarefder, R. A., Saha, N., Hall, J. W., and Ng, P.
T. T. (2008): Evaluating weak subgrade for
pavement design and performance prediction: A
case study of US 550, Journal of
GeoEngineering, 3(1), 13–24.
[9] Khogeli, W. E. I. and Mohamed, E. H. H.
(2004): Novel Approach for Characterization
Of Unbound Materials. Journal of the
Transportation Research Board, No. 1874,
Washington, D.C., 38-46.
[10] Theyse, H. L., Hoover, T. P., Harvey, J. T.,
Monismith, C. L. and Coetzee, N. F. (2006): A
Mechanistic-Empirical Subgrade Design Model
Based On Heavy Vehicle Simulator Test Results,
Geotechnical Special Publication No. 154,
Pavement Mechanics and Performance, B.,
Huang, R., Meier, J., Prozzi, E., Tutumluer, Geo
Institute, ASCE, 195-202
[11] Fairweather, H. and Yeaman, J. (2014). “A
study of the parameters affecting the
performance of roads under extreme rainfall
event”. Int. J. of GEOMATE, Sept., 2014, Vol.
7, No. 1 (Sl. No. 13), pp.955-960
International Journal of GEOMATE, May, 2016,
Vol. 10, Issue 21, pp. 1848-1853.
MS No. 5151 received on Ju , 2015 and
reviewed under GEOMATE publication policies.
Copyright © 2015, Int. J. of GEOMATE. All rights
reserved, including the making of copies unless
permission is obtained from the copyright
proprietors. Pertinent discussion including authors’
closure, if any, will be published in Jan 2017 if the
discussion is received by July 2016.
Corresponding Author: Abdul Naser Abdul
Ghani