Proceedings of the 20th International Conference on Soil Mechanics and Geotechnical Engineering, Sydney 2021
Geotechnical and seismic design considerations for dikes and coastal protection for
reclaimed lands in Manila Bay
Considérations de conception géotechnique et sismique pour les digues et la protection côtière
des terres récupérées dans la baie de Manille
Gian Paulo D. Reyes, Roy Anthony C. Luna, & John Michael I. Tanap
AMH Philippines, Inc., Philippines, gian.reyes@amhphil,com
Mark Albert H. Zarco & Antonio J. Reyno
Geotechnical Engineering Group, Institute of Civil Engineering, College of Engineering, University of the Philippines
Diliman, Philippines
ABSTRACT: The paper presents the various geotechnical considerations in the design of dikes and coastal protection for reclaimed
areas in Manila Bay. The embankment stability analysis model should best represent the subsurface condition underlying the site,
properties of the fill, and applied loads. Slope stability analysis using computer programs based on limit equilibrium approach, as
well finite element methods are utilized. Seismic analysis to evaluate long-term stability of slopes and embankments is carried out
through earthquake-resistant design through the selection of appropriate ground motion parameters.
A case study presenting various options for ground improvement, aimed at ensuring the long-term stability of the dikes and coastal
protection is discussed.
RÉSUMÉ : Le document présente les diverses considérations géotechniques dans la conception des digues et la protection côtière des
terres récupérées dans la baie de Manille. Le modèle d'analyse de la stabilité du remblai doit représenter au mieux l'état du sous-sol du
site, les propriétés du remblai et les charges appliquées. L'analyse de la stabilité des pentes à l'aide de programmes informatiques basés
sur une approche d'équilibre limite, ainsi que des méthodes par éléments finis sont utilisées. L'analyse sismique pour évaluer la stabilité
à long terme des pentes et des remblais est réalisée grâce à une conception antisismique grâce à la sélection de paramètres de mouvement
du sol appropriés.
Une étude de cas présentant diverses options d'amélioration des sols, visant à assurer la stabilité à long terme des digues et la protection
côtière, a également été discutée.
KEYWORDS: Reclamation, coastal protection, soft clay, seismic hazard analysis, non-linear time history analysis
1
INTRODUCTION.
In recent years, the demand for more property development on
reclaimed areas in the Philippines has escalated as real estate and
infrastructure projects within the nation’s capital, Manila,
become more and more essential in boosting the country’s
economy and its drive to become a ‘developed’ nation. To date,
there are more than twenty (20) proposed and on-going
reclamation projects along Manila Bay comprising of various
types of developments and infrastructures. One that will be
discussed in this paper is a proposed 2,000+ hectare major
infrastructure development located northwest of Manila Bay.
As with all land reclamation projects, one key component is
the design and engineering of the land platform, dikes, and
coastal protection structures that will ensure that the
superstructures can be built safely on top of the reclaimed area.
Additional design considerations are made to account for the
engineering consequences of the country’s geographical
location. Being situated along the Pacific Ring of Fire, the
Philippines accounts for 3.2% of the world’s seismicity, posing
threats of ground shaking, landslides, and liquefaction all over
the country, and has more than twenty (20) typhoons passing
over it every year. This paper focuses on the geotechnical
engineering and seismic hazard assessment of such reclamation
and coastal structures. A design approach is discussed which is
specifically tailor-made to design against the unique hazards
present in the country due to its geographical location using state-
of-the-art analyses and technologies in geotechnical and
earthquake engineering such as seismic velocity logging (SVL)
geophysical test, probabilistic seismic hazard analysis (PSHA),
and nonlinear dynamic time-history analysis (NLTHA) using
finite element numerical method (FEM). Details of which are
discussed using the case study project.
2
PROJECT BACKGROUND.
The proposed 2,000+ hectare land reclamation project in Manila
Bay is one of the latest flagship infrastructure projects of the
country that aims to spearhead the economic growth of a
particular group of provinces adjacent to Metro Manila. Project
information shall not be disclosed at this point in time. Phase 1
of the project is the reclamation of the land where the
infrastructures will be built on; this will be the main focus of this
paper. The structures requiring engineering design and solutions
are the land platform and the coastal protection comprising of
armor rock slopes that will confine the entire area. Given the very
soft subsurface conditions at the project area, a ground
improvement scheme of preloading with prefabricated vertical
drains or PVDs was deemed to be the most economical solution.
The general construction methodology is to backfill the entire
area until the pre-determined PVD installation levels, install
PVDs, backfill until required surcharge level including
settlement compensation heights, allow consolidation to occur,
then finally construct the armor rock protection bund. About 8m
Proceedings of the 20th International Conference on Soil Mechanics and Geotechnical Engineering, Sydney 2021
to 10m high fill embankments are going to be placed to reclaim
the overall land footprint and attain a specified target final
ground level. After consideration of ground settlements, the final
height of the armor rock slope protection ranges from 4m to 6m.
3
DESIGN APPROACH.
In consideration of the project requirements and the various
factors affecting the engineering design of the reclamation
structures, the following design approach for geotechnical
engineering and seismic hazard assessment was developed:
3.1
Identify all potential geohazards
The first step is to identify the potential geohazards that may
occur at the project area in order to plan and develop the
appropriate design schemes and types of analyses that will
mitigate and address such engineering disasters. The geohazards
identified for the case study project are:
1) settlement of ground due to presence of very thick soft soils;
2) settlement or movement of ground due to seismic events;
3) slope instabilities of staged embankments and armor rock
slope protections;
4) liquefaction of both in-situ and fill soil materials; and
5) flooding due to extreme rainfall, wave action, and seismic
events.
This paper covers discussions on items 1 through 3. The
proposed ground improvement scheme was designed to address
liquefaction and is not of interest in this paper. The same is the
case for flood and coastal studies.
3.2
Obtain and develop engineering parameters
From the geotechnical and geophysical investigation programs
and related literatures, obtain and derive the necessary material
parameters to be used for the analyses. This includes both
geotechnical and seismic engineering parameters.
3.3
Perform seismic hazard analysis (SHA) and spectral
matching
Being situated in a seismically active zone, conducting advanced
analysis to quantify the seismic hazard of the project area is
almost always necessary. A probabilistic seismic hazard
assessment (PSHA) is performed for the project to provide
structural and geotechnical designers with the response spectra
at different levels of ground motions where seismic design
parameters and inputs can be extracted from. This provides sitespecific parameters that are not provided by local and/or
international guidelines and manuals on earthquake engineering.
Spectral matching and ground motion selection is then carried
out for a pre-selected ground motion suite to come up with input
ground motions for the nonlinear time-history analysis
(NLTHA).
3.4
Perform geotechnical analyses
With the identified geohazards and derived engineering
parameters, geotechnical analyses shall be performed to come up
with a design that ensures the safety and stability of the
reclamation structures. The following analyses are performed for
the project:
1) settlement analysis of land filling;
2) slope stability analysis by limit equilibrium method (LEM)
of staged embankments and armor rock slopes; and
3) nonlinear time-history analysis (NLTHA) by finite element
method (FEM) of armor rock slopes to check deformations
during seismic loadings.
4
GEOTECHNICAL CONDITIONS.
In order to characterize the geotechnical subsurface conditions at
the project area, various geotechnical investigations were carried
out comprising of more than 160 boreholes with standard
penetration tests (SPT), triaxial strength tests (UU, CU, CD),
field vane shear tests, one-dimensional consolidation tests, and
routine geotechnical laboratory tests; more than 60 cone
penetration tests with pore pressure (CPTu); and a few number
of seismic velocity logging (SVL) tests. One particular test that
this paper would want to highlight is the SVL. Seismic velocity
logging is an intrusive non-destructive method used to determine
the physical properties of the underlying soil or rock surrounding
a borehole and the speed with which seismic waves propagate
through the strata. The test is conducted by lowering a PS
suspension logger probe that has an acoustic wave source and
geophone receivers into the borehole. The source, which is
located bottom of probe, will apply waves that will then travel
through the soil/rock material and the geophones, located top of
probe, will receive and record the signal. The resulting outputs
are the mechanical and dynamic properties of the soil such as
shear wave velocity, Vs, modulus of elasticity, E, shear modulus
of elasticity, G0, and Poisson’s ratio, , which are important
parameters for the subsequent seismic hazard analysis (SHA) and
nonlinear time-history analysis (NLTHA).
From the results of the field and laboratory tests, it was
observed that the existing subsurface conditions at the project
area consist of thick layers of very soft clays, extending from 5m
to 30m below the existing ground level. These layers are
underlain by stiff to very stiff clays and pockets of very dense
sands. Subsurface idealization was performed and parameters
were subsequently derived for the geotechnical analyses
conducted. All parameters (listed below) are derived from field
tests results and laboratory tests conducted at soil samples
collected from the >160 boreholes and are checked and validated
with widely known references and correlations (i.e., Bowles, Das,
Lambe & Whitman). Additionally, dynamic properties such as
E50ref are derived from references provided by PLAXIS and are
also checked with established literatures (i.e., Bowles) and
previous project experiences. Tables 1 to 3 present the general
summary of soil parameters.
Table 1. Summary of settlement properties of soil
Cc
Cα
Cv
[-]
[-]
[m2/y]
Very soft clay
~0.3 – 0.8
~0.020 – 0.045
~2.0
Stiff clay
~0.4 – 0.6
~0.015 – 0.020
~2.0
Soil layer
Table 2. Summary of strength properties of soil
γ
Su
[kN/m3]
[kPa]
Very soft clay
~15
~10 – 30
Stiff clay
~16
~30 – 70
Very stiff clay
~19
~80 – 120
Hard clay
~20
~150 – 250
Soil layer
Table 3. Summary of dynamic properties of soil
Soil layer
Very soft clay
Stiff clay
Vs
G0
E50ref
[m/s]
[MPa]
[MPa]
~90 – 140
~20 – 30
~5 – 10
~140 – 200
~30 – 60
~10 – 20
Proceedings of the 20th International Conference on Soil Mechanics and Geotechnical Engineering, Sydney 2021
Very stiff clay
~230 – 280
~60 – 90
~30 – 40
Hard clay
~300 – 430
~140 – 370
> ~40
These are discussed in the following sections.
6
5
SEISMIC HAZARD ANALYSIS.
Seismic hazard analysis (SHA) is the process of quantifying the
overall seismic hazard of an area in terms of acceleration. The
probabilistic approach (PSHA) in performing SHA quantifies
seismic hazard at different levels of risk depending on the
recurrence interval or return period of the design ground motion.
PSHA also considers multiple seismic sources simultaneously
and accounts for uncertainties related to distance, time,
recurrence, and size (magnitude).
In performing SHA, empirically-formulated attenuation
models are utilized to determine the expected surface
acceleration by estimating how seismic waves propagate and
travel from source to site. Attenuation models are commonly
referred to as Ground Motion Prediction Equations (GMPE), and
these equations were formulated using globally-acquired
earthquake information (e.g. epicenter location, depth, and
magnitude). The New Generation Attenuation West2 (NGAWest2) GMPE’s developed by the Pacific Earthquake
Engineering Research (PEER) Center were used for fault
systems, and the BC Hydro GMPE (Abrahamson et al., 2016,
2018) was used for subduction zone sources.
For the case study project, the required earthquake return
periods are 475-year or 10% probability of exceedance in 50
years, also known as Design Basis Earthquake (DBE), and 2,475year or 2% probability of exceedance in 50 years, also known as
Maximum Credible Earthquake (MCE). The output target
response spectra at rock outcrop for each return period at 5%damping are presented in Figures 1 and 2.
SPECTRAL MATCHING AND GROUND MOTION
SELECTION
To carry out a nonlinear time-history analysis (NLTHA) for the
armor rock slopes, a ground motion suite of records of at least
three (3) ground motion records should be developed. The
ground motion records selected must have similar characteristics
to the seismic source with the most contribution to the overall
seismic hazard at the project area and shall be spectrally matched
with the target response spectra discussed in the previous section.
Spectral matching is the process of modifying the amplitude
and/or frequency content of a certain ground motion record such
that the original record’s response spectrum matches the
response spectrum obtained from the SHA. By doing so, the
modified record can be surmised to be an event that may happen
on-site given the nature of the potential earthquake generators,
their respective recurrence parameters, and their rupture/focal
mechanisms.
Spectral matching was done using SeismoMatch 2018 by
Seismosoft. SeismoMatch performs spectral matching using the
wavelet algorithm of Al Atik and Abrahamson (2010), which is
an update of the original algorithm proposed by Abrahamson
(1992). This wavelet algorithm utilizes an improved tapered
cosine adjustment function that prevents drift in the modified
velocity and displacement time series without baseline correction.
The quality of the matching is assessed by looking at how wellmatched ground motions compare with the original records’
time-histories (velocity and displacement). The primary criterion
would be to check if the non-stationary parameters of the original
ground motion is well-maintained even after matching. The
matched response spectra (dashed plots) of the 3 ground motions
are shown in Figure 3.
Figure 1. 5%-damped DBE response spectra for rock outcrop.
Figure 3. Response spectra of matched ground motions.
7
Figure 2. 5%-damped MCE response spectra for rock outcrop.
These rock site response spectra are then used to derive input
ground motions for the nonlinear time history analysis (NLTHA)
and design of armor rock slopes considering seismic conditions.
Furthermore, the response spectra at the surface of the assumed
consolidated or post-ground improvement soil conditions are
also derived to provide seismic parameters for the limit
equilibrium slope stability analysis for a simplified approach.
SETTLEMENT ANALYSIS.
Given that the existing subsurface conditions consist of thick
layers of very soft clays, it is prudent to consider the settlement
of this layer in the design of the land reclamation. The ground
will potentially subside in large magnitudes due to the sheer
volume of backfill materials that will be placed on top of it. With
these types of highly plastic clayey soil materials, long-term
primary consolidation settlement is expected after the application
of 8m to 10m high fill embankments due to the very low
permeability of cohesive soils. Also, it is anticipated that the finegrained soil materials will undergo plastic deformations over
constant effective stress and cause further settlement in the soil
mass, known as secondary consolidation or creep settlement.
In order to improve the strength characteristics of the
underlying soft layer and address both settlement and
Proceedings of the 20th International Conference on Soil Mechanics and Geotechnical Engineering, Sydney 2021
liquefaction hazards, ground improvement by means of
preloading with prefabricated vertical drains (PVD) were
considered, which will induce settlement and densification of the
soil and at the same time accelerate the time of consolidation by
a significant amount of time. With PVDs, consolidation time was
reduced from hundreds of months (> 150 months) to only 8 to 12
months.
Staged construction and phasing were employed in the
analysis to cater the target completion dates of each specified
project area. This meant overlapping and sequencing of filling
and consolidation periods across the entire land area. Note that
these are only analysis considerations. Land filling was staged by
placing a series of 3.0m high embankments which were
progressively placed until the required fill thickness or final
surcharge level was achieved. PVDs were installed during the
first stage of filling. Additional surcharge heights of about 2.0m
to 3.0m were included to account for the effects of future loads
as well as mitigate secondary settlement. This excess layer was
later removed or unloaded after the consolidation period in order
to overconsolidate the soil and induce the ‘rebounding effect’ in
the soil layers, reducing the calculated values of creep
settlements by a significant amount (Terzaghi, Peck, and Mesri
1996). Sufficient resting periods after each filling stages were
included such that the underlying soft soil layers can attain
adequate increase in shear strength prior to the installation of the
next stage, which were checked in the stability analysis. Table 4
presents the general staged construction sequence considered.
Table 4. General staged construction sequence
Fill
height (m)
Cumulative.
fill (m)
Activity description
1
3
3
1st filling; Installation of
PVDs
RP1
-
-
Resting period for stage 1
2
3
6
2nd filling; starts after
resting period of stage 1
RP2
-
-
Resting period for stage 2
~2 – 3
~8 – 9
3rd filling; starts after
resting period of stage 2
-
-
Resting period for stage 3
~1 – 2
~9 – 10
Final filling; includes 23m excess surcharge; starts
after resting period of stage 3
~9 – 10
Consolidation period;
allow the soil to achieve
desired degree of
consolidation
Stage
3
RP3
4
5
6
-
[-]2
~7 – 8
Removal of 2-3m excess
surcharge and compaction
The time rate of settlement was estimated using Terzaghi’s
one-dimensional theory. Since the underlying soil layers mostly
consists of clay, resulting settlements were found to be long-term.
The calculated settlements range from ~2.0m to 3.0m, with total
fill heights ranging from ~8m to 10m. Estimated consolidation
period to achieve a specified degree of consolidation range from
8 to 12 months. Post-construction settlements, which are mainly
attributed to the remaining primary settlement and the entire
secondary settlement, were also calculated considering the
design life of the project and are found to range from ~10mm to
30mm. Table 5 shows the summary of the results of the
settlement analysis.
Table 5. Summary of settlement analysis results
~8 – 10 m
Total fill height
~2.0 – 3.0 m
Total primary settlement
~8 – 12 months
Consolidation period
~10 – 30 mm
Post-construction settlement
The resulting post-construction settlements were below the
threshold settlement limits for the project. Nevertheless,
importance of settlement monitoring and test embankments
before and/or during the construction were emphasized in order
to verify and refine the calculations and construction
methodology as the land filling works progress.
8
SLOPE STABILITY ANALYSIS BY LIMIT
EQUILIBRIUM METHOD.
Slope stability analysis by limit equilibrium method (LEM) is
performed for the various filling stages of the land platform
perimeter slopes. The definition of LEM was intentionally left
out of this paper as it is already a well-known and established
approach in geotechnical engineering. The analysis includes both
the temporary filling slopes and the final armor rock protection
slopes described in the previous sections. From the construction
methodology, the most critical slope configurations throughout
the entire construction period are determined and modeled in the
limit equilibrium analysis. From the settlement analysis, the
development and increase in shear strength of the underlying soft
soil layers throughout the consolidation process have been
estimated and incorporated in the slope models. For the
temporary slopes, only static loading conditions are considered.
For the permanent armor rock protection slopes, both static and
seismic loading conditions are applied. The seismic coefficients
used are obtained based on the peak ground acceleration (PGA)
value of the SHA response spectra as described in the previous
section. A post-consolidation soil condition was estimated in the
modelling of the armor rock slope sections. Bishop Simplified,
Janbu Simplified, and Morgenstern-Price methods for both
circular and non-circular failure planes are adopted. A general
summary of the results is shown in the Table 6.
Table 6. Summary of slope stability analysis results
Reqd.
factor of
safety
Calc.
factor of
safety
3m high temporary embankment
1.2
~1.5
6m high temporary embankment
1.2
~1.4
9m high temporary embankment
1.2
~1.3
10m high temporary embankment
1.2
~1.3
4m high armor rock slope – static
1.5
~2.0
4m high armor rock slope – seismic
1.1
< 1.0
Slope section
Figure 4. A typical armor rock section. Slope gradients and thicknesses
vary.
Results would show that adequate factors of safety (FoS)
have been attained for all static loading cases. In all cases, non-
Proceedings of the 20th International Conference on Soil Mechanics and Geotechnical Engineering, Sydney 2021
circular failure planes provided the lowest FoS. On the other
hand, all seismic loading cases did not meet the required factor
of safety. This means that the shearing resistance of the soil
layers have been overcame by the driving seismic forces and
have caused the slope to displace at certain magnitudes. This,
however, does not conclude that the slope is unsafe and unstable.
In recent years, the design of earth structures using numerical
deformation analysis have gained popularity amongst
geotechnical engineers wherein displacements are being allowed
to occur until certain limiting values. This is guided by the fact
the earth structures are flexible in nature and can maintain
stability even after experiencing displacements or deformations.
This analysis, also known as nonlinear time-history analysis
(NLTHA), is discussed in the following section.
2007) was used to model the soil materials which involves
dynamic parameters such as the shear wave velocity, Vs, shear
modulus of elasticity, G0, and modulus of elasticity, E50, Eoed, and
Eur. The HSS model has the ability to capture hysteretic behavior
of soils at large and small strain levels. A compliant base
boundary condition was set to account for the outcrop ground
motions and a factor of 0.5 was applied to the time-histories to
consider only the upward travelling seismic waves. The finite
element model was extended such that there will be no ‘confining
effects’ that will impact the results.
A general summary of the results of the NLTHA is presented
in Table 7. The magnitudes of each time-history record are also
indicated. Figures 7 and 8 shows a typical deformed mesh and
deformation contour outputs of the NLTHA in PLAXIS 2D.
Extent of Armor Rock Section
Figure 5. (Left) Snapshot of typical LEM results under static conditions
showing adequate FoS (=1.535) and potential non-circular failure plane
passing through the soil just below the armor rock. (Right) Snapshot of
typical LEM results under pseudo-static conditions showing inadequate
FoS (=0.420) and non-circular failure planes that are deep-seated. Figure
is much more zoomed out compared to figure on the left.
Figure 6. Spectrally-matched acceleration time-history of one of the
selected ground motions (TH-1).
Table 7. Summary of NLTHA results. ux and uy are horizontal and
vertical deformations, respectively.
Maximum calculated deformation
Ground motion
9
NONLINEAR TIME-HISTORY ANALYSIS.
Nonlinear time-history analysis (NLTHA) measures the response
of a soil-mass over the period of during and after the application
of a full ground motion time-history record. With NLTHA, the
stress state of materials is allowed to exceed the linear-elastic
region and behave with nonlinear material properties (plastic or
elastoplastic behavior). As such, moving past the limitation of
being within the linear-elastic region may approximate or realize
a more realistic behavior of the soil-mass (e.g. plastic hinge
manifestation, stiffness degradation, hysteretic damping).
NLTHA utilizes earthquake records that must capture the
expected hazard on-site as described in the previous section. In
geotechnical analyses, NLTHA is performed via numerical
modeling, where the process is often referred to as “Site
Response Analysis” (SRA). SRA is a process that measures
accelerations and deformations by means of deconvolutionconvolution cycles. Deconvolution and Convolution refer to the
attenuation and amplification of ground accelerations,
respectively, as seismic waves propagate across the subsurface.
The deconvolution-convolution cycle is generally performed in
a multi-step manner. SRA was carried out using a finite element
method (FEM) numerical analysis tool, PLAXIS 2D. FEM is a
numerical technique used for finding approximate solutions by
continuum-based methods wherein the governing equations
describing the state of stress of the soil/rock mass are derived on
the principles of conservation of mass, momentum, and energy.
These equations, together with the prescribed boundary
conditions, result in a nonlinear boundary value problem.
NLTHA was performed for the armor rock protection slopes
using the 3 spectrally matched ground motions discussed in the
previous section. See Figure 6 for the acceleration time-history
record of one of the selected ground motions. In PLAXIS 2D, the
Hardening Soil Small-strain (HSS) constitutive model (Benz
Crest
Toe
ux
uy
ux
uy
TH-1 [Mw=7.37]
~40
~220
~120
~8
TH-2 [Mw=7.13]
~30
~215
~100
~1
TH-3 [Mw=6.19]
~40
~210
~100
~1
The resulting deformations are found to be within the
allowable limits set for the project. It is noted that the horizontal
deformations, ux, were assessed in terms of relative
displacements. Time-history record no. 1 (TH-1) incurred the
most deformation at the crest and toe of the armor rock slope
having the highest magnitude among the three ground records.
However, further assessment of the deformations at points well
inside the land platform, i.e., 5 to 100m from the armor rock slope,
would show that the other ground motion records (TH-2 and TH3) induced higher deformations in those areas. While these are
not critical areas and the magnitudes are well within the tolerable
limits, it just shows that certain time-history records will have a
unique effect to the soil-mass.
Figure 7. A snapshot of the deformed mesh of armor rock slope. Figure
is scaled up 20.0 times; full extent of model not shown. Lighter shaded
soil materials represent the unimproved conditions while the darker
shaded soil materials represent the improved conditions.
Proceedings of the 20th International Conference on Soil Mechanics and Geotechnical Engineering, Sydney 2021
Figure 8. A snapshot of the deformation contours of the armor rock slope.
Full extent of model not shown. Colors represent the deformation
contours with red (warmer colors) as the highest in magnitude and blue
(colder colors) as the lowest. Magenta line at the bottom represents the
ground motion application. Blue vertical arrows represent the loads
applied on top of the land platform.
In terms of the surface accelerations that resulted from the
site response analysis, there has been an observed reduction in
the peak ground acceleration (PGA), i.e., acceleration at period
equals zero, at the land platform surface when comparing it to
the PGA value at rock site. However, acceleration amplifications
were also observed in the mid-to-long period range. See Figure
9. This is expected as the seismic waves travel from the rock up
to the soft-to-stiff soil layers. These are important values in the
design of the superstructures which is part of the next phase of
the development.
history analysis (NLTHA), the armor rock slopes have been
designed in such a way that it will be allowed to displace at
certain limiting values. This provided an alternate engineering
solution that is much more cost-effective whereas if the
traditional limit equilibrium method (LEM) was to be employed,
there is a potential for overdesigning the armor rocks.
Furthermore, with the site-specific response spectra calculated
from the NLTHA, as well as the large set of outputs that it
provides including accelerations, stresses, and deformations at
any point in the finite element model, optimizations, realistic
simulations, and performance-based designs are even more
attainable.
The design of dikes and coastal protection structures for
reclamation projects in the Philippines has been undergoing
continuous improvement and development throughout the years
as more and more reclamations projects are being envisioned and
realized in the country. The constant rise of state-of-the-art
geophysical and geotechnical engineering tests and analyses has
led to the advancement in data gathering and design analysis of
such structures. A combination of classical soil mechanics
theories and advanced nonlinear numerical analyses has helped
in providing engineering solutions that are more accurate,
elaborate, realistic, and at the same time cost-effective. However,
it is extremely important to remember that all of these analyses
only provide the best estimates and simulations of the behavior
of the structures and its surrounding. It is still best practice to
conduct validation or proof tests, monitoring measurements, and
recalibration analysis to verify all theoretical results with the
actual observations.
11
Figure 9. Comparison of calculated response spectra at platform level vs
rock site level for 3 ground motions. Response spectra at platform level
is derived from SRA while response spectra at rock site is derived from
SHA.
When comparing the results of NLTHA/FEM with LEM, it
can be observed from Figure 5 (Right) that the failure planes
corresponding to the minimum FoS under pseudo-static
conditions pass through 5m to 10m below the armor rock and
extends horizontally for almost the entire layer/s, which is quite
indicative of the FEM findings. These failure planes represent the
area in the soil mass where the largest strains develop as the
overturning forces become significantly larger than the resisting
forces, thus, yielding large displacements. It is noted that LEM
can provide an indication of the qualitative behavior of slope
failure under earthquake conditions. However, it does not
represent the actual deformation pattern nor can provide
displacement magnitudes due to limitations in its methods.
Actual deformation behaviors and magnitudes should still be
based on the more advanced numerical analysis, especially when
results of the LEM are below satisfactory.
10
CONCLUSIONS
This paper presented the design approach and considerations for
the engineering of a proposed reclamation land and its coastal
protection structure and were specifically developed to address
the geohazards present at the project area. Excessive settlements
and slope instabilities were handled through classical soil
mechanics theories and were still found to provide reliable
results. On the other hand, seismic analysis was carried out using
advanced numerical methods. By performing nonlinear time-
ACKNOWLEDGEMENTS
The authors would like express their utmost gratitude to the
different agencies/institutions for the tremendous effort they
have put into research which provided essential information to
this study (e.g. MGB, PHIVOLCS, and USGS). For without any
of which, this study may not have been completed; to GEM and
PEER, for their extensive research in seismology and for
providing the necessary tools (e.g. OpenQuake, GMPE’s) in
performing SHA.
The authors would also like to thank AMH Philippines, Inc.
and the organizers of the 20th ICSMGE for providing the
opportunity to present this study.
12
REFERENCES
Abrahamson N., Silva W., and Kamai R. 2014. Summary of the ASK14
ground motion relation for active crustal regions. Earthquake Spectra,
30(3), 1025-1055.
Baker J.W. 2015. Introduction to Probabilistic Seismic Hazard Analysis.
Bowles J. E. 1996. Foundation Analysis and Design 5th Edition.
McGraw-Hill, Pennsylvania.
Das B. M. 2010. Principles of Geotechnical Engineering 7 th Edition.
Cengage Learning, Stamford.
Duncan J. M and Wright S. G. 2005. Soil Strength and Slope Stability.
John Wiley & Sons, Inc., New Jersey.
GEM Foundation. 2020. The OpenQuake-engine User Manual.
Lambe W.T. and Whitman R. V. Soil Mechanics. 1969. Massachusetts
Institute of Technology. John Wiley & Sons, New York.
R. Mote, T. Manlapig, R., and Zamora C. 2009. Probabilistic Seismic
Hazard Assessment for Central Manila in Philippines. AEES.
Philippine Institute of Volcanology and Seismology. (2019). Distribution
of Active Faults and Trenches in the Philippines.
Thenhaus P.C., Hanson S.L., Algermissen S.T., Bautista B.C., Bautista
L.P., Punongbayan B.J., Rasdas A.R., Nillos J.T.E., and Punongbayan
R.S. 1994. Estimates of the Regional Ground Motion Hazard in the
Philippines. Proceedings of the Conference on Natural Disaster
Mitigation in the Philippines.
United States Geological Survey. ANSS Comprehensive Earthquake
Catalog Database.