PROJECT REPORT

A STUDY ON WAVE-CURRENT INTERACTIONS DURING
EXTREME WEATHER EVENTS, USING COUPLED
ADCIRC+SWAN MODEL
A THESIS SUBMITTED TO THE GRADUATE FACULTY
in partial fulfilment of the requirements for the degree of MASTER OF SCIENCE
By:
George Victor Emmanuel
OST-2013-21-01
M.Sc. Physical Oceanography and Ocean Modeling
Department of Physical Oceanography
School of Ocean Studies and Technology
Kerala University of Fisheries and Ocean Studies
Under the guidance of
Dr. P. Vethamony
CSIR - National Institute of Oceanography,
Dona Paula, Goa – 403 004

P a g e | 1
DECLARATION BY STUDENT
I, George Victor Emmanuel (OST-2013-21-01), a student of M.Sc. Physical
Oceanography and Ocean Modelling, Department of Physical Oceanography,
KUFOS hereby declare that the work entitled “A study on wave-current
interactions, using coupled ADCIRC+SWAN model” is my original work. I
have not copied from any other students’ work or from any other sources except
where due reference or acknowledgement is made explicitly in the text, nor has any
part been written for me by another person.
Name: George Victor Emmanuel Date submitted: 01.09.2015
(Name of the student)
Student’s signature _______________

P a g e | 2
DECLARATION BY SUPERVISOR
Myself, P. Vethamony, Chief Scientist, CSIR-National Institute of
Oceanography, Goa hereby certifies that the work entitled “A study on wave-
current interactions, using coupled ADCIRC+SWAN model” was prepared by
the above named student, and was submitted to the “FACULTY” as a partial
fulfilment for the conferment of M.Sc. Physical Oceanography and Ocean Modeling,
and the aforementioned work, to the best of my knowledge, is the said student’s
work.
Name: P. Vethamony Date: 30.08.2015
(Name of the Supervisor)
Supervisor’s signature: _______________

P a g e | 3
ACKNOWLEDGEMENT
I would like to thank all those who contributed in one way or other to the work described
in this thesis. First and foremost, I thank my thesis advisor, Dr. P. Vethamony for accepting me
into his group. He was a great support to me by giving intellectual freedom, engaging me in new
ideas and demanding a high quality of work. I was fortunate to have the chance to work under his
guidance, and I am sincerely grateful to him.
Samiksha S.Volvaiker and I worked together on several different phases of the modeling
process, and without her help and efforts my job would have undoubtedly been more difficult. I
am greatly benefited from her keen scientific insight, her knack for solving seemingly intractable
practical difficulties in modelling, and her ability to put complex ideas into simple terms. I am
fortunate to have met Suneel.V, Veerasingam.S and Soumya here, and I thank them for their
friendship, love, and unyielding support. I’ve my heartfelt thanks to Sherin.V.Raju,
A.V.S.Chaithanya and Ravish Naik, who made my time here at NIO, Goa a lot more fun. I would
like to acknowledge CSIR-NIO for providing the facility of High Performance Computing (HPC)
system Pravah.
I owe a debt of gratitude to Dr. C.V.K.Prasada Rao (Academic Consultant, KUFOS) and
Mary Jisha Francis (Assistant Professor, KUFOS) as my post-graduate experience benefited
greatly from the courses I took and the opportunities I had under them.
Finally, I would like to acknowledge my friends and family who supported me during my
time here. I’m so grateful to my Mom, Dad and brother for their constant love and support.

P a g e | 4
S.No Figures and Graphs Page No.
Figure 1 IMD provided cyclone track for THANE cyclone. 25
Figure 2 IMD provided cyclone track for PHAILIN cyclone. 26
Figure 3 Bathymetry plot of the study domain, for both
THANE and PHAILIN cyclones.
29
Figure 4 Unstructured mesh generated for the study
domain, using SMS.
29
Figure 5 Schematic diagram on ADCIRC+SWAN coupling
mechanism
33
Figure 6 Surge residual at Puducherry coast (79.855E,
11.933N).
39
Figure 7 Comparison of significant wave Height at
Puducherry, b/w model & observed.
40
Figure 8
Time and spatial variation of atmospheric pressure
simulated along the track of Thane Cyclone, using
coupled ADCIRC+SWAN model from 26th
December, 2011-12:00pm at 12hour interval, until
30th
December, 2011-12:00pm.
40-42
Figure 9
Spatial-Time plot of wind stress simulated along
the track of Thane Cyclone, using coupled
ADCIRC+SWAN model from 26th
December,
2011-12:00pm at 12hour interval until 30th
42-43

P a g e | 5
Figure 10 Maximum elevation model output (maxele.63) for
THANE cyclone from ADCIRC+SWAN model run.
44
Figure 11
Time and spatial variation of significant wave
height (Hs) simulated along the track of Thane
Cyclone, using coupled ADCIRC+SWAN model
from 26th December, 2011-12:00pm at 12hour
interval, until 30th December, 2011-12:00pm.
44-46
Figure 12
Comparing SWAN generated and ADCIRC+SWAN
generated significant wave height for THANE
cyclone at landfall time.
46
Figure 13
The significant wave heights time series at Ganjam
coast (85.07139ºE, 19.35299ºN) generated by
ADCIRC and ADCIRC+SWAN model runs.
47
Figure 14
Time series of significant wave height (Hs)
simulated at Gopalpur coast (84.969ºE, 19.280ºN),
using coupled ADCIRC+SWAN model.
47
Figure 15
Time and spatial variation of atmospheric pressure
October, 2013-12:00pm at 12hour interval until
13th October, 2013 -12:00pm.
48-49
Figure 16
Time and spatial variation of wind stress
October, 2013-12:00pm at 12hour interval until
13th October, 2013 -12:00pm.
49-50
Figure 17 Maximum elevation model output (maxele.63) for
PHAILIN cyclone from ADCIRC+SWAN model
51

P a g e | 6
run.
Figure 18
Time and spatial variation of significant wave
height (Hs) simulated along the track of Thane
Cyclone, using coupled ADCIRC+SWAN model
from 8th
October, 2013-12:00pm at 12hour
interval until 13th
October, 2013-12:00pm
51-52
Figure 19
Comparing SWAN generated and ADCIRC+SWAN
generated significant wave height for PHAILIN
cyclone at landfall time.
52
Figure 20
Water surface elevation time series from different
model runs, with inputs being specified, for
THANE Cyclone.
55
Figure 21
Time series of velocity magnitudes of currents
under ADCIRC and ADCIRC+SWAN model runs,
for THANE Cyclone.
56
Figure 22
Significant wave height comparison between two
model run outputs. The parameters included in
each run are specified.
58

P a g e | 7
CONTENTS
Unit Topic Page No.
1
2
ABSTRACT
INTRODUCTION
1.1 WAVE CURRENRT INTERACTIONS
1.2 NUMERICAL MODELING
1.2.1 ADCIC MODEL
1.2.2 SWAN MODEL
1.2.3 ADCIRC+SWAN COUPLED MODEL
1.3 STUDY AREA
LITERATURE REVIEW
2.1 UNDERSTANDING WAVE-CURRENT
INTERACTIONS
2.1.1 EFFECT OF CURRENTS ON WAVE
2.1.2 EFFECT OF WAVES ON CURRENTS
2.2 EXTREME EVENTS
2.3.1 MESHES
2.3.2 ADCIRC
2.3.3 SWAN
2.3.4 ADCIRC+SWAN
2.4 CASE STUDIES
2.4.1 CYCLONE THANE
2.4.2 CYCLONE PHAILIN
10
12
13
13
14
15
15
16
18
19
19
21
22
22
23
23
24
24
25
25
26

P a g e | 8
3
4
5
METHODOLOGY
3.1 DOMAIN AND MESHES
3.2.1 ADCIRC
3.2.2 SWAN
3.2.3 ADCIRC + SWAN
3.2.4 SURFACE WATER MONITORING
SYSTEM (SMS)
3.3 SIMULATION
3.3.1 PARAMETERS
3.3.2 MODEL RUNS
RESULTS
4.1 THANE
4.1.1 VALIDATION
4.1.2 ATMOSPHERIC PRESSURE
4.1.3 WIND STRESS/ VELOCITY
4.1.4 MAXIMUM ELEVATION
4.1.2 SIGNIFICANT HEIGHT
4.2 PHAILIN
4.2.1 VALIDATION
4.2.3 WIND STRESS/VELOCITY
DISCUSSIONS
5.1 THANE CYCLONE
5.1.1 SURFACE ELEVATION
5.1.2 CURRENST
5.1.3 WAVES
28
29
30
31
32
33
34
34
34
35
38
39
39
40
42
44
44
46
46
48
49
51
51
53
54
54
56
57

P a g e | 9
6
7
5.2 PHAILIN CYCLONE
5.2.2 CURRENTS AND WAVES
SUMMARY
REFERENCES
57
57
58
60
63

P a g e | 10
ABSTRACT
Funakoshi et al. (2008), Dietrich et al. (2010), Xie et al. (2001, 2003) and many others
have shown that the sea surface elevations and currents affect the propagation of waves and the
location of wave-breaking zones. It is also been observed that water levels can be altered due to
wave-current interaction. This work aims at applying the wind-wave-current models to study the
wave-current interaction, and thereby estimating the change in wave parameters, primarily the
significant wave heights during the two extreme weather events, namely, cyclones Thane (2011)
and Phailin (2013). Both the cyclones affected the east coast of India, specifically Puducherry,
Tamil Nadu and Odisha coasts.
The unstructured-mesh SWAN spectral wave model and the ADCIRC shallow-water
circulation model have been integrated into a tightly-coupled ADCIRC+SWAN model. The
model components were applied to an identical, unstructured mesh and run sequentially in time.
Wind speeds, water levels, currents and radiation stress gradients are vertex-based, and therefore
can be passed through memory or cache to each model component. The coupled
SWAN+ADCIRC system is highly scalable and allows for localized increases in resolution
without complexity, cost of nested meshes or global interpolation between heterogeneous
meshes. This model setup is used for the study.
Different model runs under different oceanic and meteorological forcings such as
meteorological forcing alone, tidal forcing alone and tidal and meteorological forcings combined
together were carried out. The outputs generated, namely, water surface elevation, currents and
wave heights were analysed, and compared with model outputs generated from SWAN (wave
model) and ADCIRC+SWAN coupled model (circulation -wave model). The modelled
significant wave heights (Hs) were compared with the available buoy data off Puducherry. The
ADCIRC+SWAN coupled model produced a significant wave height of approx. 3m near
Puducherry coast during the cyclone Thane before the landfall which matches well with the
observed values. From the model run, it was observed that the significant wave height increased
rapidly during landfall, which is due to cyclone effect. However, the model over-predicts wave
heights during landfall. Former studies suggest that this over-prediction may be due to missing
physics or poor numerics. It was also noted from the ADCIRC+SWAN coupled model outputs
that the significant wave heights were higher on the right side of the cyclone track in the
direction of propagation, compared to those on the left side.

P a g e | 11
Whatever work presented in this dissertation is to be considered as preliminary results, as
two models have been independently as well as coupled way used in a short period of time. For
obtaining accurate results of sea surface elevation, currents and waves, and studying wave-
current interaction, and predicting oceanic parameters during extreme weather events, these
models need to be tuned further.

P a g e | 12
UNIT–1
INTRODUCTION

P a g e | 13
1.1 WAVE-CURRENT INTERACTION
In the energy spectrum, wind waves and circulations are well separated, however they
interact with each other. Wind-driven waves affect the vertical momentum mixing and bottom
friction, which in turn affect the circulation (Dietrich et al., 2010). Mastenbrok et al. (1993) also
presented the same idea that the wind waves can indirectly affect the coastal ocean circulation by
enhancing the wind stress.
Water levels and currents affect the propagation of waves and the location of wave-
breaking zones. Wave transformation generates radiation stress gradients that drive currents. It is
observed that water levels can be increased by 5–20% in regions across a broad continental shelf,
and as much as 35%in regions of steep slope (Funakoshi et al., 2008; Dietrich et al., 2010) due to
wave-current interaction. Xie et al. [2001, 2003] studied the interactions between waves and
currents and found that the overall circulation can play a significant role on the wind waves, in
coastal regions and vice-versa also. In combined wave and current environments, nonlinear
interactions can play a significant role and we cannot simply superpose the two components.
Nonlinear interactions have non-negligible impacts on the hydrodynamics of a wave–current
system, especially in the turbulent boundary layer with high roughness (Davies et al.,
1988 and Grant and Madsen, 1979). Thus in most coastal applications, a coupled approach on
waves and circulation should be preferred.
Even rogue waves, which are surface waves with approximately twice the size of
surrounding waves, being very unpredictable, are known to be formed by the focusing of wave
energy. This can cause the waves to dynamically join together, forming very big 'rogue' waves.
In the numerical regime, models for wave–current interactions have been developed for
deep or finite-depth waves (Nwogu, 2009, Swan and James, 2001 and Swan et al., 2001) and
long waves (Benjamin, 1962, Freeman and Johnson, 1970 and Shen, 2001).Spectral, spatial and
temporal resolutions are the limiting factors of any wave-circulation model. Structured mesh
nesting, which was developed as a solution to this has now been replaced with the unstructured
mesh. Resolution varies over a range of scales within the same mesh from deep water to the

P a g e | 14
continental shelf to the channels, marshes and ﬂood plains near shore (Westerink et al., 2008), in
an unstructured mesh model. Unstructured meshes allow for localized resolution, where solution
gradients are large and correspondingly coarser resolution where solution gradients are small,
thus minimizing the computational cost relative to structured meshes with similar minimum
mesh spacings (Dietrich et al., 2010).
In this work, “SWAN” wave model and the “ADCIRC” circulation model are integrated
and coupled tightly, and are run on the same global unstructured mesh. Thus the physics of
wave-circulation interactions are resolved accurately in both models and that too efficiently, by
eliminating the need for costly interpolation and extensive global communication as information
is passed between the models. The SWAN +ADCIRC model is suited ideally to simulate waves
and circulation and their propagation from deep water to complicated nearshore systems. The
coupled model is highly scalable and can be integrated.
1.2.1 ADCIRC MODEL
“ADCIRC model or ADvanced CIRCulation model” is a system of computer programs
for solving time dependent, free surface circulation and transport problems in two and three
dimensions (http://adcirc.org/).
CHL (Computational Hydraulics Laboratory) along with the “University of Carolina” and
the “University of Texas” has developed the ADCIRC finite element based coastal ocean
circulation code. The initial developers of the code were Rick Luettich (University of North
Carolina at Chapel Hill) and Joannes Westerink (University of Notre Dame). This ocean model
was developed as a part of the DRP (Dredging Research Programme), for the sake of generating
harmonic constituents database for tidal elevations and currents at discrete locations along the
east, west and the Gulf of Mexico coast of US and to utilize tropical and extra-tropical global
boundary conditions to compute frequency indexed from storm surge hydrographs along the US
coast.
The model finds its applications in predicting storm surges and flooding, modeling tides
and wind driven circulations, near shore marine operations, dredging feasibility and material
disposal studies and many more.

P a g e | 15
1.2.2 SWAN MODEL
A number of studies were carried out to understand wave generation and wave growth
during hurricane (Young, 2006; Prasad Kumar and Stone, 2007; Xu et al.,2007; Chu and Cheng,
2008; Soomere et al., 2008; Fan et al.,2009; Babanin et al., 2011). “SWAN model (Simulated
WAves Nearshore model)” is a third generation wave model, developed at the Delft University
of Technology, The Netherlands that computes random, short crested wind generated waves in
coastal regions and inland waters (http://www.swan.tudelft.nl/).
It models the energy contained in waves or the wave radiation stress as they travel over
the ocean surface towards the shore. Several issues associated with the wave-circulation model
coupling in the past have been resolved by the use of the unstructured mesh version of SWAN.
Earlier, heterogeneous meshes were employed, in which a separate sub-mesh and solution
information is interpolated and exchanged between the models via external files or a generic
framework for solving each model. The wave model is a fully implicit finite difference method
recently extended to unstructured grids that employs a sweeping Gauss–Seidel technique to
compute the numerical solution (Zijlema, 2010).
SWAN model is very popular and widely accepted as it makes us possible to model
waves over a large area, for any boundary input, in a cost-effective method, that too in a
relatively short period of time. Rather than working as an individual model, it is widely used in
coupling with other models.
1.2.3 ADCIRC + SWAN COUPLED MODEL
Coupling wave and circulation models is vital in order to define shelf, nearshore and
inland hydrodynamics during a hurricane (Dietrich et al., 2011). The tightly coupled
SWAN + ADCIRC paradigm allows both wave and circulation interactions to be solved on the
same unstructured mesh resulting in a more accurate and efficient solution technique. It has been
widely recognized as a successful strategy for modeling storm surge applications (Dietrich et al.,
2011a and Dietrich et al., 2012).

P a g e | 16
We employed the tightly coupled Simulating Waves Nearshore (SWAN) model and
ADvanced CIRCulation (ADCIRC) model to simulate the evolution of waves and storm surge
from deep water to the coastal region. ADCIRC and SWAN wave models are run in series on the
same local mesh and core. The two models “leap frog” through time, each being forced with
information from the other model. The procedure is stable at any time step and allows for local
mesh refinement in areas of interest. Since the scale of individual wave phenomena is too small
to be resolved on large domains, SWAN computes the wave action density in geographic,
spectral and temporal spaces. SWAN is driven by wind speeds, water levels and currents
computed at the vertices by ADCIRC. While ADCIRC is driven partly by momentum flux
gradients, which are computed using SWAN outputs.
On each coupling interval, ADCIRC is run first, because we assume that in the nearshore
and coastal flood plain, wave properties are more dependent on circulation (Dietrich et al.,
2010).The SWAN model provides the link between offshore wind and wave conditions and the
nearshore conditions that drive the longshore transport (Margaret L. Palmsten, 2001). The
ADCIRC+SWAN coupling eliminates many efforts as compared to individual or separate
working. This coupling shares the work among the model components in a way that can speed up
the combined run time. SWAN time step is more than that of ADCIRC, so coupling interval is
taken to be same as the SWAN time step.
1.3 STUDY AREA
There are numerous reports on the casualties and damage to life and property due to
coastal flooding reported from various global ocean basins (Nicholls et al., 1999).With an
increasing surge in human populace staying along coastal zones, the risk levels have multiplied
as of late. In the Indian setting, there is an expanding pattern of population density in the
maritime states, and the death toll and property also in the same proportion because of flooding
during an extreme weather conditions. But, however, these have been reduced in the recent years
because of improvement in forecasting the events, and subsequent precautionary measures
adopted by the Government.
The north Indian Ocean is one among those tropical cyclone vulnerable oceans in the
world, where the cyclones can be of serious threat to many countries like India, Myanmar,
Bangladesh, and Gulf countries bordering the Arabian Sea. The flooding caused by coastal

P a g e | 17
storms is an occasional threat to people living in coastal regions (Huiqing Liu et al.,
2006).Coastal floods associated with storm surges are a matter of concern for the east coast of
India (P K Bhaskaran et al, 2014). Even the very fast moving winds cause severe damage.
Providing accurate ocean state forecasts during an extreme event is challenging. However, it is
also important to fore-warn the users such as fishermen community, oil and shipping industry,
ports and harbours, navy, coast guard and coastal communities on the possible hazardous
conditions during such extreme weather events (T. M. Balakrishnan Nair). Thus, it turns out to be
a necessity to have proper and enough hindcast studies, as well as forecasting of waves, storm
surges, ocean circulation and their interactions, for efficient preparedness and planning of any
extreme coastal event.
In the sections following, the component ADCIRC and SWAN models are described, and
their tight coupling is introduced. The coupled model is then used to study the interaction of
currents on waves during two extreme events, taking cyclones THANE (India, 2011) and
PHAILIN (India, 2013) as case studies.

P a g e | 18
UNIT-2
LITERATURE REVIEW

P a g e | 19
2.1 UNDERSTANDING WAVE-CURRENT INTERACTION
In coastal waters and nearshore regions, wave–current interactions have always been a
topic of interest. The dynamic interaction between waves, currents and tides has been a subject of
studies since many years. Wave–current interactions can occur over a wide range of both wave
and current conditions. Former studies like those done by J Wolf and D Prandle., (1999) have
shown that the wave-current interactions have significant influence on the wave and current
behaviours independently.
2.1.1 EFFECT OF CURRENTS ON WAVES
There are broadly two ways that currents affect waves: first, the effective wind of
generating waves is modified by the ocean current, and the effective fetch also changes in the
presence of current (for example as in Kara et al., 2007); the other, the current modifies the wave
action balance equation through changes in propagation velocity of wave energy (see for
example, Fan et al., 2009 and Benetazzo et al., 2013).
Refraction, bottom drag modification and blocking due to the presence of currents can
affect the waves. Past studies showed that the effects of depth refraction can be seen easily,
turning the mean wave direction towards shore-normal. Current refraction has a more subtle
effect, depending on the spatial variation of currents, whether decreasing or increasing towards
the coast. Generally the tidal amplitude will increase with shoaling depths, towards the coast
until friction reverses this trend. The waves will tend to turn towards the direction of the current
axis.
The steady currents have a crucial effect on the relative wave frequency. Longer will be
the intrinsic period for waves of the same apparent (absolute) period, in a following current and a
shorter intrinsic period in an opposing current. Waves steepen on an opposing current, related to
this Doppler shift, due to shorter wavelength and increased wave height from wave action
conservation.
Former studies, like those by Mellor., (2008) and RL Soulsby., (1990) on various
empirical theories on wave–current interaction in the bottom boundary layer has put forward that
the friction coefficient experienced by waves in a current regime will be larger, compared to that
in no current. This also applies to the effective current friction factor in the presence of waves.

P a g e | 20
The wave intrinsic (angular) frequency, σ is related to the wave number, k by the
dispersion relation:
𝝈 = √𝒈𝒌. 𝒕𝒂𝒏𝒉. 𝒌𝒉,
in water depth h, whereas the observed or apparent (absolute) frequency, ω is Doppler-shifted:
𝝎 = 𝝈 + 𝒌. 𝑼
where k is the wave-number vector and U the current vector, e.g., Phillips (1977).
The time variation of absolute frequency is given by:
𝒅𝝎
𝒅𝒕
=
𝝈𝒌
𝒔𝒊𝒏𝒉𝟐𝒌𝒉
𝝏𝒉
𝝏𝒕
+ 𝒌.
𝝏𝑼
𝝏𝒕
= 𝒌 {𝒄
𝝏𝒉
𝝏𝒕
+
𝝏𝑼 𝒄
𝝏𝒕
}
Where,
𝑪 =
√𝒈𝒌.𝒕𝒂𝒏𝒉.𝒌𝒉
𝐚𝐧𝐝 𝑼 𝒄 = 𝑼𝒄𝒐𝒔(𝜹 − 𝜶) (Jonsson, 1990)
Here, time is t, current direction, δ, wave direction α. This shows that the variation with time is
related to the time variation of depth and relative current. In deep water (h>L/2, wave length,
L=2π/k; L is the wave length) the depth-related component disappears. The time derivative of
the intrinsic frequency is
𝒅𝝈
𝒅𝒕
= −
𝝈𝒌𝒉
𝛁. 𝑼 − 𝒄 𝒈 𝒌.
𝝏𝑼
𝝏𝒔
Where, cg is the wave group speed and ∂/∂s the space derivative in the direction of wave
propagation.
In order to look for the classical steepening effect of waves on an opposing current we
need to use the intrinsic or relative frequency (which is directly related to wave length). The
significant steepness parameter,
𝑺 𝑺 =
𝟐𝝅𝑯 𝑺
𝒈𝑻 𝟐
𝒁
gives a good approximation to steepness HS/LM (LM is the mean wave length which is equal to
2π/∫kE(f)df ) if the relative (intrinsic) TZ is used. Hs is the significant wave height Tz is the wave
period.

P a g e | 21
Current refraction is governed by Snell's law: k sin β=constant, where β is the angle between
wave direction and the normal to current direction, i.e.
𝒔𝒊𝒏 𝜷 𝟐
𝒔𝒊𝒏 𝜷 𝟏
=
𝑳 𝟐
𝑳 𝟏
(Jonsson, 1990). In this case, the values β1 and L1 represent the incident angle and wavelength in
one current regime (U1), β2 and L2 represent the angle with the normal and wavelength in a
second current regime (U2).
2.1.2 EFFECT OF WAVES ON CURRENTS
The wave transformation generates radiation stress gradients that drive set-up and
currents. The wind-driven waves affect the vertical momentum mixing and bottom friction,
which in turn affect the circulation. Wolf et al., (1988) and Janssen (1989) explains that the
effective surface drag coefficient for wind-driven surge currents may change with wave age.
However, in practice it is still difficult to demonstrate the importance of this effect. When waves
are present, the bottom friction coefficient of currents will be modified.
Works of Kemp and Simons (1982, 1983) and Kolpman (1994, 1997) showed that near
surface velocity of current is increased if waves are opposed and vice-versa. Fine sediments on
the bottom of shallow lake or sea can be re-suspended by waves and transported by currents.
Bakker and van Doorn (1978) and Mathisen and Madsen (1996a, 1996b) found that a
current followed by waves experience a reduction of speed near the bed, hence an increase of
apparent roughness. Zengrui Rong (2014) recognized that the wave enhanced bottom stress can
dominate over the shelf, and can trap more freshwater to the nearshore region. In the cross-shore
direction, the vertical imbalance between the depth-uniform pressure gradient due to wave setup
and the depth-varying momentum flux generates a near-bed seaward under-current (Svendsen,
1984). While in the long-shore direction, the spatially non-uniform wave momentum flux
provides a new forcing of a wave-driven long-shore current (Longuet-Higgins, 1970).

P a g e | 22
2.2 EXTREME WEATHER EVENTS
It has been noticed that the wave-current interactions discussed above are common to all
kind of weather systems, normal or extreme. However, they get intensified in high magnitudes,
only during an extreme event. An event is called extreme in this sense if it is from the tails of the
climatological distribution, occurring, for example, only 5% or less of the time. The exact choice
of the cut-off climatological probability value used in the definition is somewhat arbitrary.
Extreme events occur naturally in physical systems.
The same physical processes that are responsible for generating non-extreme events are
contributing to the occurrence of extremes also. It is the unusual and/or unexpected nature of
extreme weather that makes these events significant to society. The effect of rare events like
floods in forming the natural environment can also be catastrophic and much larger than the
cumulative effect of near normal weather between the rare events. Thus it turns out to be a
necessity to have a better and deeper understanding on waves and their interactions with currents
and that too especially during extreme events.
A tropical cyclone is such an extreme event, which is rotating, organized system of
clouds and thunder storms that starts over tropical or subtropical waters and has a closed low-
level circulation. Tropical typhoons pivot counter-clockwise in the Northern Hemisphere.
Tropical cyclones forming between 5º and 30º North latitudes generally move towards the west.
However the winds in the centre and upper levels of the atmosphere may change sometimes and
which may steer the cyclone in the north and northwest directions. As the tropical cyclones reach
30º north, they tend to move northeast.
Numerical modeling of the ocean is concerned, there are always trade-offs to be made in
terms of computational efficiency, spatial and temporal resolutions, and sophistication of the
physics and numerical schemes used. Any ocean modeling is completely based on making
appropriate choices and assumptions to tackle any specific issue. With a specific end goal to
appropriately and efficiently use the model outputs, it is necessary for us to have an

P a g e | 23
understanding of the choices that have been made in developing and implementing that specific
model.
2.3.1 MESHES
Until recently, circulation and wave models, which have been limited by their spectral,
spatial and temporal resolutions were widely applied for the waves, currents and their interaction
studies. Nesting structured meshes to enhance resolution in specific regions by employing
meshes with progressively finer scales helped in overcoming their limitations. Relatively fine
nearshore wave models, such as STWAVE and SWAN were nested inside relatively coarse deep-
water wave models, such as WAM and WaveWatch III (WAMDI Group, 1988; Komen et al.,
1994; Booij et al., 1999; Smith et al., 2001; Thompson et al., 2004; Gunther, 2005; Tolman,
2009). However, these nearshore wave models were not efficient enough when applied to large
domains, while the deep-water wave models may not contain the necessary physics or resolution
for nearshore wave simulation.
Unstructured circulation models were developed recently to provide localized resolution
of gradients in geometry, flow processes and bathymetry or topography. Large velocity space
requires efficient filtering of small scales. Varied resolution within the same mesh over a range
of scales from deep water to the continental shelf to the channels, marshes and flood plains
nearshore is its highlight.
2.3.2 ADCIRC MODEL
ADCIRC is a system of computer programs, for solving time dependent, free surface
circulation and transport problems in two (depth integrated) and three dimensions, based on
hydrodynamic primitive equations.
These equations have been formulated using the traditional hydrostatic pressure and
Boussinesq approximations and have been discretized in space using the finite element (FE)
method and in time using the finite difference (FD) method. These programs utilize the finite
element method in space allowing the use of highly flexible, unstructured grids.

P a g e | 24
2.3.3 SWAN MODEL
SWAN model or Simulated WAves Nearshore model” was developed at Delft
University of Technology.It computes random, short crested wind generated waves in coastal
regions and inland waters and models the energy contained in waves or the wave radiation stress
as they travel over the ocean surface towards the shore.
SWAN model is a very popular and widely accepted as it makes possible to model waves
over a large area, for any boundary input, in a cost-effective method.
2.3.4 ADCIRC + SWAN MODEL
The modeling system used for the present study is a coupled wave, current and
astronomical tide model (ADCIRC + SWAN) that uses the same computational grid for all the
physical processes. The tightly coupled SWAN + ADCIRC paradigm allows both wave and
circulation interactions to be solved on the same unstructured mesh resulting in a more accurate
and efficient solution technique. It has been widely recognized as a successful strategy for
modeling storm surge applications (Dietrich et al., 2011a and Dietrich et al., 2012).
The ADvanced CIRCulation (ADCIRC) model that solves for water levels and currents at
a range of scales is coupled with SWAN model, which models the energy contained in waves as
they travel over the ocean surface. These models together compute the storm surge, depth-
averaged currents and net still water level elevation. A parallel mode of running is opted over the
series mode, since it works on multiple processors and hence is time saving.

P a g e | 25
2.4 CASE STUDIES
2.4.1 CYCLONE THANE
Figure 1: IMD provided cyclone track for THANE cyclone.
Thane cyclone occurred in the Bay of Bengal during December, 2011, and created havoc
along the Tamil Nadu coast (Bhaskaran et al, 2014). It is regarded as one among the most
devastating severe cyclones in the North Indian Ocean basin. In the early hours of 30th
December, 2011 cyclone Thane had its landfall between coastal Cuddalore and Pondicherry.
Fig.1 shows the track of cyclone THANE obtained from IMD (India Meteorological
Department). The life span of Thane lasted from 25 to 31 December, 2011.
It originated as a depression in the Bay of Bengal, on 25th
evening of December, 2011.
Later, IMD (India Meteorological Department) on 28th
December, 2011 rated Thane as a very
severe cyclonic storm having a 1-minute sustained wind speed of 120kmph. In the early hours of

P a g e | 26
29th
December, 2011 Thane intensified with sustained wind speed reaching 150 kmph.
Thereafter, on 30th
December, 2011 the wind speeds varied between 120 and 140 kmph, during
the landfall time. The system weakened later, and stayed as a well-marked low pressure area over
north Kerala and neighbourhood in early morning of 31st December, 2011. The timely and
accurate warning helped the disaster managers to initiate appropriate action.
2.4.2 CYCLONE PHAILIN
Figure 2: IMD provided cyclone track for PHAILIN cyclone.
Cyclone ‘Phailin’ (in Thai“ไพลิน” meaning “sapphire”) is one among the strongest
tropical cyclones ever recorded over the northern Indian Ocean basin (Bernhard et al., 2013).
Phailin made its landfall on the Odisha coast.

P a g e | 27
On 4th
October, 2013 cyclone Phailin originated as a tropical depression in the Gulf of
Thailand. Later on 6th
October, 2013, Phailin crossed over Malay Peninsula and entered into
Andaman Sea (eastern Indian Ocean) and moved west–northwest direction. ‘Phailin’ intensified
rapidly and became a cyclone equivalent to a category 1 hurricane on October 10, 2013. Phailin
developed into a super cyclone on the next day, moving over warm tropical Indian Ocean. Within
a span of 24 hours, wind speed increased from 83 kmph to 213 kmph. . About 215 kmph was the
maximum sustained wind speed during landfall at Gopalpur on 12th
October, 2013, with a central
pressure of 940mb.
A wide swath of damaged infrastructure, flooding of agricultural farmlands, wide spread
death of livestock and a few instances of human loss was left along coastal Odisha and parts of
northeastern Andhra Pradesh, by cyclone Phailin. Timely warnings, alertness and massive
evacuation efforts were very effective in minimizing human loss, compared with the death toll
during 1999 Odisha Super Cyclone. Fig.2 shows the track of cyclone PHAILIN obtained from
IMD (India Meteorological Department).
Table 1: Cyclones and their specifications
S.No
Cyclone
Name
Date of
Formation
Max. Wind
Speed
(kmph)
Lowest
Pressure
(mb)
Impact/
Landfall
Region
Date of
Dissipation
IMD Cyclone
Category
1 THANE
25 December
2011
140 969 Tamil Nadu
30 December
2011
Very Severe
Cyclonic
Storm
2 PHAILIN
9 October
2013
215 940 Odisha
12 October
2013
Super
Cyclonic
Storm

P a g e | 28
UNIT-3
METHODOLOGY

P a g e | 29
Over the past few decades, there have been a number of studies focusing on the wave -
current interaction. Studies conducted by researchers like Mastenbrok (1993), Signell (1990), LJ
Cote (1960) and many others have already proved evidently that the wind waves play a
significant role in the overall circulation in the coastal regions.
The key question to be addressed here is how these processes affect each other and how
they can be properly included in numerical models. Broadly, these processes influence one
another in several ways: (i) wind stress which is changed by incorporating the wave effect
(Donelan et al., 1993); (ii) radiation stress can be incorporated into wave models by invoking
wave action conservation (Komen et al., 1994, Lin and Huang, 1996); (iii) bottom stress, which
is a function of wave-current interaction in the near bottom layer when the water depth is
sufficiently shallow for wave effects to penetrate to the bottom (Signell et al., 1990); (iv) the
depth variation and current conditions which are given as input into wave models (Tolman,
1991).
3.1 DOMAIN AND MESHES
Generating a flexible finite element mesh for the study region is the first step. The semi-
circular shape of study domain is preferred to the rectangular one for this specific case, since the
Figure 4: Unstructured mesh generated for the study domain,
using SMS.
Figure 3: Bathymetry plot of the study domain, for
both THANE and PHAILIN cyclones.
Figure 3
Figure 4

P a g e | 30
rectangular domain is computationally more intensive (due to larger number of nodes). However,
an issue of computational instability at the corner points (commonly seen with semi-circular
domain) that connects the mainland with offshore boundaries do exist.
The model domain covers the region in the Bay of Bengal and the coast, between 80-90ºE
and 10-20ºN (Fig.3). The bathymetry has been generated from the modified Etopo2 datasets by
Sindhu (Sindhu et al., 2007). Improved shelf bathymetry derived by digitizing the depth contours
and sounding depths less than 200m from the hydrographic charts, published by the Naval
Hydrographic Office, Dehra Dun for the Indian Ocean region is included in it. The existing
ETOPO2 datasets for depths less than 200m is modified by the digitized data that are gridded.
The computational grid was generated using Mesh Generation Package SMS (Surface Water
Modeling System, http://www.aquaveo.com/products).
The physical phenomenon of tides and storm surge waves can be resolved using a coarse
grid in deep waters, whereas the resolution is critical and needs to be higher in coastal and
nearshore waters for better estimates (Blain et al., 1994; Luettich and Westerink, 1995). For the
mesh generated, the grid resolution is refined in shallow waters and relaxed in deeper waters. The
element size can be larger in deep waters due to large bathymetric depth allowing a simple
geometry (Bhaskaran et al, 2014).Thus the unstructured mesh used have the capability to resolve
sharp gradients in bathymetry, especially in the near-shore areas and thereby providing a better
resolution of wave transformation(Dietrich et al., 2011).
Variable grid resolution features the optimization of computational time. Under this
consideration, the minimum and maximum grid spacing were specified as 1km along the shallow
coast and 5km along the off-shore boundary. Rao et al. (2009) showed that a grid resolution of 1
km nearshore is sufficient and good enough for precise computation of surge heights along east
coast of India. The generated mesh covers the east coast of India, from Kodikkarai (Tamil Nadu)
to Dosinga (Odisha) and comprise of 44,575 nodes and 87364 triangular elements. This same
finite element mesh was used for both the case studies, THANE (2011) and PHAILIN (2013).

P a g e | 31
3.2.1 ADCIRC MODEL
ADCIRC is a continuous–Galerkin, finite–element, shallow-water model that solves for
water levels and currents at a range of scales (Westerink et al., 2008; Luettich and Westerink,
2004; Atkinson et al., 2004; Dawson et al., 2006).
ADCIRC unstructured coastal ocean model is applied to compute water levels by solving the
Generalized Wave Continuity Equation (GWCE). Reformulating the primitive equations into a GWCE
form gives highly accurate, noise-free, FE-based solutions to the shallow-water equations (Lynch and
Gray 1979; Kinnmark 1984). ADCIRC can be forced with elevation boundary conditions, normal flow
boundary conditions, surface stress boundary conditions and tidal potential earth load/self-attraction tide
𝝏 𝟐
𝜻
𝝏𝒕 𝟐
+ 𝝉 𝟎
𝝏𝜻
𝝏𝒕
+
𝝏𝑱̃ 𝒙
𝝏𝒙
+
𝝏𝑱̃ 𝒚
𝝏𝒚
− 𝑼𝑯
𝝏𝝉 𝟎
𝝏𝒙
− 𝑽𝑯
𝝏𝝉 𝟎
𝝏𝒚
= 𝟎
Where,
𝑱̃ 𝒙 = −𝑸 𝒙
𝝏𝑼
𝝏𝒙
− 𝑸 𝒚
𝝏𝑼
𝝏𝒚
+ 𝒇𝑸 𝒚 −
𝒈
𝟐
𝝏𝜻 𝟐
𝝏𝒙
− 𝒈𝑯
𝝏
𝝏𝒙
[
𝑷 𝒔
𝒈𝝆 𝟎
− 𝜶𝜼] +
𝝉 𝒔𝒙,𝒘𝒊𝒏𝒅𝒔 + 𝝉 𝒔𝒙,𝒘𝒂𝒗𝒆𝒔 − 𝝉 𝒃𝒙
𝝆 𝟎
+ (𝑴 𝒙 − 𝑫 𝒙) + 𝑼
𝝏𝜻
𝝏𝒕
+ 𝝉 𝟎 𝑸 𝒙 − 𝒈𝑯
𝝏𝜻
𝝏𝒙
𝑱̃ 𝒚 = −𝑸 𝒙
𝝏𝑽
𝝏𝒙
− 𝑸 𝒚
𝝏𝑽
𝝏𝒚
− 𝒇𝑸 𝒙 −
𝒈
𝟐
𝝏𝜻 𝟐
𝝏𝒚
− 𝒈𝑯
𝝏
𝝏𝒚
[
𝑷 𝒔
𝒈𝝆 𝟎
− 𝜶𝜼] +
𝝉 𝒔𝒚,𝒘𝒊𝒏𝒅𝒔 + 𝝉 𝒔𝒚,𝒘𝒂𝒗𝒆𝒔 − 𝝉 𝒃𝒚
𝝆 𝟎
+ (𝑴 𝒚 − 𝑫 𝒚) + 𝑽
𝝏𝜻
𝝏𝒕
+ 𝝉 𝟎 𝑸 𝒚 − 𝒈𝑯
𝝏𝜻
𝝏𝒚
and currents from the vertically integrated momentum equations:
𝝏𝑼
𝝏𝒕
+ 𝑼
𝝏𝑼
𝝏𝒙
+ 𝑽
𝝏𝑼
𝝏𝒚
− 𝒇𝑽 = −𝒈
𝝏
𝝏𝒙
[𝜻 +
𝑷 𝒔
𝒈𝝆 𝟎
− 𝜶𝜼] +
𝝉 𝒔𝒙,𝒘𝒊𝒏𝒅𝒔 + 𝝉 𝒔𝒙,𝒘𝒂𝒗𝒆𝒔 + 𝝉 𝒃𝒙
𝝆 𝟎 𝑯
+
𝑴 𝒙 − 𝑫 𝒙
𝑯
&
𝝏𝑽
𝝏𝒕
+ 𝑼
𝝏𝑽
𝝏𝒙
+ 𝑽
𝝏𝑽
𝝏𝒚
− 𝒇𝑼 = −𝒈
𝝏
𝝏𝒚
[𝜻 +
𝑷 𝒔
𝒈𝝆 𝟎
− 𝜶𝜼] +
𝝉 𝒔𝒚,𝒘𝒊𝒏𝒅𝒔 + 𝝉 𝒔𝒚,𝒘𝒂𝒗𝒆𝒔 + 𝝉 𝒃𝒚
𝝆 𝟎 𝑯
+
𝑴 𝒚 − 𝑫 𝒚
𝑯

P a g e | 32
The unstructured mesh used, provide high localized grid resolution where solution
gradients are large, and low grid resolution where solution gradients are small. Thus, minimizing
both local and global error norms for a given computational cost.
ADCIRC can be run either with a single processor as a series mode, which consumes a
great deal of time or with multiple processors in a parallel mode, which is time saving.
3.2.2 SWAN MODEL
The SWAN (Simulating WAves Nearshore) model is developed at the Delft University of
Technology, The Netherlands. SWAN is the third generation shallow water spectral wave model
that includes wave propagation, refraction due to currents and depth, generation by wind,
dissipation (white-capping, bottom friction, depth induced breaking), and nonlinear wave–wave
interactions (Booij et al., 1999). SWAN models the energy contained in waves as they travel over
the ocean surface. The model incorporates the height, shape and direction changes associated
with the waves, due to winds, white capping, wave breaking, energy transfer between waves and
variations in the ocean floor and currents.
The cost of collecting in situ wave measurements by installing multiple wave buoys, may
be cost prohibitive. SWAN makes it possible to model waves over a large area, for any boundary
input, in a cost-effective method. The model reproduces the field conditions well when the model
is properly initialized.
SWAN predicts the evolution in geographical space 𝑥⃗ and time t of the wave action
density spectrum N(𝑥⃗,t,σ,θ), with σ the relative frequency and θ the wave direction, as governed
by the action balance equation (Booij et al., 1999).
𝝏𝑵
𝝏𝒕
+ 𝛁 𝒙⃑⃑ ∙ [(𝑪 𝒈
⃑⃑⃑⃑ + 𝑼⃑⃑ )𝑵] +
𝝏𝒄 𝜽 𝑵
𝝏𝜽
+
𝝏𝒄 𝝈 𝑵
𝝏𝝈
=
𝑺 𝒕𝒐𝒕
𝝈
The unstructured mesh version of SWAN implements an analogue to the four-direction Gauss
Seidel iteration technique employed in the structured version and it maintains SWAN’s
unconditional stability (Zijlema, 2010). At the vertices of an unstructured triangular mesh, the
wave action density spectrum, N(𝑥⃗,t,σ,θ) is computed by SWAN. It orders and sweep through the

P a g e | 33
mesh vertices and update the action density information from neighbouring vertices. It then
reverses the direction until sufficient wave energy has propagated through geographical space in
all directions.
3.2.3 ADCIRC+SWAN COUPLED MODEL
In the coupling of SWAN and ADCIRC, the unstructured-mesh version of SWAN is
applied, so that both models run on the same mesh, thus eliminating the need for interpolation
between models (Zijlema 2010; Dietrich et al. 2011). The two models “leap frog” through time,
each being forced with information from the other model. Because of the sweeping method used
by SWAN to update the wave information at the computational vertices, it can take much larger
time steps than ADCIRC (Dietrich et al., 2010). Hence, the SWAN time step and the coupling
intervals is same.
The coupling of SWAN+ADCIRC allows both models to utilize the same global and
local meshes so that information is passed between models at the mesh vertices, without the need
for interpolation between heterogeneous meshes (Dietrichetal.,2011a) Water levels and currents
are computed by ADCIRC and passed at each SWAN time step. SWAN solves the action
balance equation for the wave action (Booij et al., 1999; Ris et al., 1999) and passes the data
back to ADCIRC, which derive wave radiation stress from this for its next step. In this way, the
radiation stress gradients used by ADCIRC are always extrapolated forward in time, while the
wind speeds, water levels and currents used by SWAN are averaged over each of its time steps
(Dietrich et al., 2010).
The solution technique employs boundary conditions, input parameterizations, wetting
and drying of elements, unstructured mesh refinement, and efficient parallel communication
(Antonia Sebastian., 2004).
ADCIRC
WATER LEVELS AND AMBIENTS
CURRENTS
SWAN
WAVE RADIATION STRESS
GRADIENTS
Figure 5: Schematic diagram on ADCIRC+SWAN coupling mechanism.

P a g e | 34
3.2.4 SURFACE WATER MONITORING SYSTEM (SMS)
SMS or Surface-water Monitoring System is a complete program for building and
simulating surface water models. The graphical user interface and analysis tool provided by SMS
allows researchers to visualize, manipulate, analyse and understand numerical data and their
associated measurements. It supports various models like ADCIRC, BOUSS-2D, CMS-Flow,
PTM, STWAVE, TABS and TUFLOW.
SMS was initially developed by the Engineering Computer Graphics Laboratory (later
renamed in September, 1998 to Environmental Modeling Research Laboratory or EMRL) at
Brigham Young University, in the late 1980’s.
SMS gather background data from a variety of sources from GIS to CAD and access
online data from numerous databases of maps, images, and elevation data. SMS allows to interact
with models in true 3D taking advantage of optimized OpenGL graphics and to create photo-
realistic renderings and animations for PowerPoint, print, and web presentations. It features 1D
and 2D modelling and a unique conceptual approach. The interface to SMS has been developed
in such a way that separate modules are used for each data type. As the user switches between
two modules, the available menus and tools change. Within a module, the user associates a
numeric model with a mesh or grid and when the grid is active, the associated tools and menus
are also enabled.
3.3 SIMULATIONS
3.3.1 PARAMETERS
The appropriate mesh that was generated for the study domain is semi-circular, and that
was used throughout the study.
A constant 2DDI (Two dimensional depth integrated) quadratic friction coefficient of
0.0025 is used in the coupled model, over a bathymetry with minimum water depth taken as
0.05m. A constant Coriolis parameter of value 0.0001 is applied to it. Ramp function value,
which is the duration that the model takes to move from no circulation to 4 tidal amplitudes, was
set as 2 days.The wetting-drying algorithms were set off and the value for spatially constant

P a g e | 35
horizontal eddy viscosity coefficient for momentum equations was set 2. Time-step was set as
10s with data saved in every 1h interval. These parameters are kept uniform for all the model
runs.
3.3.2 MODEL RUNS
The model was then run (cold start) for dates from 25th
December, 2011 (1200h) to 30st
December, 2011 (1200h) for Thane cyclone and from 7th
October, 2013 (1200h) to 13th
October,
2013 (1200h) for cyclone Phailin. The model was run with 5 conditions for both the cyclones.
The modelling specifications of Thane and Phailin cyclones are provided in Table 2.
In the first model run, only ADCIRC was run. The effect of tidal currents alone on the sea
surface state was intended to be studied. Along the model boundary domain, time varying water
level data were obtained from ‘Le Provost’ tidal database, which is represented by 13 tidal
harmonic constituents (K1, M1, N2, O1, P1, S2, K2, I2, N2, MU2, NU2, Q1 and T2) based on
the finite element solution version 95.2 (Le Provost et al., 1998). Out of the 18 input files
suggested for any ADCIRC model run, only the 2 necessary files were provided for the run. They
are “fort.14” which is the grid and boundary information file and “fort.15” which is the model
parameter and periodic boundary condition file. Water surface elevation time series file (fort.63)
and depth averaged velocity time series file (fort.64) are the extracted outputs. This scenario
represents surface elevation and currents due to tides.
The second model run, which was intended for the study of sea surface elevation due to
both wind and tidal forcing, was run with ADCIRC alone. Besides the two necessary input files
(fort.14 and fort.15), the meteorological input file was also provided. The wind/meteorological
input file (fort.22) was applied in the dynamic Holland wind field model, which utilizes the best
track information from the JTWC (Joint Typhoon Warning Centre). The Holland model
calculates the wind field, and provides information on sea-level pressure distribution and gradient
wind within the tropical cyclone. The wind speed in terms of surface stress was then specified to
ADCIRC model based on the relation proposed by Garrett (1977). Besides the meteorological
input, the 13 tidal constituents were also provided from Le Provost. The atmospheric pressure
time series (fort.73) and wind stress/velocity time series (fort.74) files were also extracted as the

P a g e | 36
outputs, along with water surface elevation and water surface velocity time series. This scenario
represents surface elevation and currents due to winds provided along the cyclone track and tides.
In the next stage, the ADCIRC model was run alone for sea surface elevation with
meteorological/wind forcing obtained from the dynamic Holland wind field model as explained
above. Water surface elevation time series (fort.63), water surface velocity time series (fort.64),
atmospheric pressure time series (fort.73) and wind stress/velocity time series (fort.74) files were
generated as outputs. This scenario represents sea surface elevation and currents due to winds
provided along the cyclone track.
In the fourth conditional model run, wave forcing was incorporated by additionally
providing the SWAN file (fort.26) as an input, besides the boundary information, boundary
condition and meteorological file inputs. This is a run with ADCIRC and SWAN. Sea surface
elevation resulting from tidal, meteorological and waves were generated. This conditional system
is the most realistic one. Time series files of water surface elevation, velocity, wind stress and
atmospheric pressure can be derived from the run. This scenario represents sea surface elevation
and currents due to winds provided along the cyclone track, currents and waves.
In the fifth and the last run, only SWAN was run only with cyclone winds (no currents
and tides). The outputs generated are time series files of sea surface elevation and currents due to
only winds and waves (no tidal or currents effects).

P a g e | 37
Table 2: Modelling specifications associated with the THANE and PHAILIN cyclones
THANE AND PHAILIN CYCLONES
Input files
Input Parameters (in
respective order as input
files)
Output files
Output Parameters (in
respective order as input
files)
Model
1st
model
run
fort.14,
fort.15
The grid and boundary
information file, the
model parameter and
periodic boundary
condition file
fort.63,
fort.64
Water surface elevation
time series, water
surface velocity time
series
ADCIRC
2nd
model
run
fort.14,
fort.15,
fort.22
model parameter and
periodic boundary
condition file,
meteorological input file
fort.63,
fort.64,
fort.73,
fort.74
time series, water
series, atmospheric
pressure time series and
wind stress/velocity time
series
ADCIRC
3rd
model
run fort.14,
fort.15,
fort.22
model parameter and
periodic boundary
condition file,
meteorological input file
fort.63,
fort.64,
fort.73,
fort.74
time series, water
series, atmospheric
series
ADCIRC
4th
model
run
fort.14,
fort.15,
fort.22,
fort.26
model parameter and
periodic boundary
condition file,
meteorological input file,
SWAN file
fort.63,
fort.64,
fort.73,
fort.74,
SwanHS.63,
maxele.63
time series, water
series, atmospheric
series, Significant height
time series, Maximum
surface elevation
ADCIRC+
SWAN
5th
model
run
fort.14,
fort.15,
fort.22,
fort.26
model parameter and
periodic boundary
condition file,
meteorological input file,
SWAN file
fort.63,
fort.64,
fort.73,
fort.74,
SwanHS.63,
maxele.63
time series, water
series, atmospheric
series, Significant height
time series, Maximum
surface elevation
SWAN

P a g e | 39
The ADCIRC, SWAN and ADCIRC + SWAN model simulations were carried out under
different conditions for the cyclones THANE and PHAILIN, and the results obtained for sea
surface elevation, currents and effect of currents on waves are presented in this chapter.
4.1 THANE CYCLONE
4.1.1 VALIDATION
The predicted storm surges for the Thane cyclone by the India Meteorological
Department (IMD., 2011) varied between 1 and 1.5 m height above the astronomical tide for
Puducherry, Tiruvallur, Villupuram, Chennai and Kancheepuram districts of north Tamil Nadu.
However, post-cyclone survey conducted by IMD, marked a storm surge of maximum1 m height,
inundating the low lying coastal areas of Cuddalore, Puducherry and Villuparam districts, at the
time of landfall of the cyclone, THANE. IMD also reported a gale wind speed reaching 120-140
kmph that prevailed along and off north Tamil Nadu and Puducherry coast. Puducherry reported
a maximum wind speed of 68 knots (125 kmph) while Cuddalore reported maximum wind speed
of 76 knots (140 kmph) at the time of landfall.
Figure 6: Surge residual at Puducherry coast (79.855E, 11.933N)
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
29-12-201100:00
29-12-201102:00
29-12-201104:00
29-12-201106:00
29-12-201108:00
29-12-201110:00
29-12-201112:00
29-12-201114:00
29-12-201116:00
29-12-201118:00
29-12-201120:00
29-12-201122:00
30-12-201100:00
30-12-201102:00
30-12-201104:00
30-12-201106:00
30-12-201108:00
30-12-201110:00
30-12-201112:00
Surgeresidual(m)
currents residual currents+waves residual

P a g e | 40
The simulated model outputs shows that the entire coastline of Tamil Nadu was affected
by storm surges, with varying magnitudes. The model computed water level elevation above
astronomical tide (Fig.5) for Puducherry coast is about 0.6m, which matches fairly with the IMD
observations (IMD).
The buoy data available in the Puducherry coast was also compared with the model
outputs and was found to be in good match, close to the landfall period.
Figure 7: Comparison of significant wave height at Puducherry, between model & observed.
0
0.5
1
1.5
2
2.5
3
3.5
4
29-12-201100:00
29-12-201101:00
29-12-201102:00
29-12-201103:00
29-12-201104:00
29-12-201105:00
29-12-201106:00
29-12-201107:00
29-12-201108:00
29-12-201109:00
29-12-201110:00
29-12-201111:00
29-12-201112:00
29-12-201113:00
29-12-201114:00
29-12-201115:00
Significantwaveheight(m)
observed buoy data ADCSWAN

P a g e | 42
Figure 8(above): Time and spatial variation of atmospheric pressure simulated along the track of Thane
Cyclone, using Holland model from 26th
December, 2011-12:00pm at 12hour interval, until 30th
December,
2011-12:00pm.
4.1.3 WIND STRESS/ VELOCITY

P a g e | 43
Figure 9: Spatial-Time plot of wind stress simulated along the track of Thane Cyclone, using Holland model
from 26th
December, 2011-12:00pm at 12hour interval until 30th

P a g e | 44
`
Figure 10: Maximum elevation model output (maxele.63) for THANE
cyclone from ADCIRC+SWAN model run.

P a g e | 46
Figure 11(above): Time and spatial variation of significant wave height (Hs) simulated along the track of
Thane Cyclone, using coupled ADCIRC+SWAN model from 26th December, 2011-12:00pm at 12hour
interval, until 30th December, 2011-12:00pm.
Figure 12: Comparing SWAN generated and coupled ADCIRC+SWAN generated significant wave height for
THANE cyclone at landfall time.
4.2PHAILIN
4.2.1 VALIDATION
As per the post cyclone survey conducted by IMD, a maximum storm surge of 2-2.5 m
height has been estimated along the low lying areas of Ganjam district of Odisha. As per the
IMD reports, at the time of landfall on 12th
October, 2013 maximum sustained surface wind

P a g e | 47
speed during cyclone PHAILIN was about 115 knots (215 kmph) and estimated central pressure
was 940hPa with pressure drop of 66hPa at the centre compared to surroundings.
Figure 13: The significant wave height at Ganjam coast (85.07139ºE, 19.35299ºN) generated by ADCIRC and
coupled ADCIRC+SWAN model runs.
The significant wave heights time series at Ganjam coast (85.07ºE, 19.35ºN) generated by
ADCIRC and ADCIRC+SWAN model runs is compared. The model generated surge value of
about 1.2m which fairly satisfied with the IMD reported values, considering the demerits and
flaws associated with the modelling tools like Holland wind model, which do not incorporate the
normal prevailing winds, but only along the cyclone track and it is also to be noted that 2-2.5m
surge, mentioned in the IMD report 2013, is specifically only for the low lying areas of Ganjam
district.
Figure 14: Time series of significant wave height (Hs) simulated at Gopalpur coast (84.969ºE, 19.280ºN), using
coupled ADCIRC+SWAN model.
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
12/12/201300:00
12/12/201301:00
12/12/201302:00
12/12/201303:00
12/12/201304:00
12/12/201305:00
12/12/201306:00
12/12/201307:00
12/12/201308:00
12/12/201309:00
12/12/201310:00
12/12/201311:00
12/12/201312:00
12/12/201313:00
12/12/201314:00
12/12/201315:00
12/12/201316:00
12/12/201317:00
12/12/201318:00
12/12/201319:00
12/12/201320:00
12/12/201321:00
12/12/201322:00
12/12/201323:00
SurgeResidual(m)
adcirc residual adcswan residual
0
1
2
3
4
5
6
7
11/10/201306:00
11/10/201309:00
11/10/201312:00
11/10/201315:00
11/10/201318:00
11/10/201321:00
12/10/201300:00
12/10/201303:00
12/10/201306:00
12/10/201309:00
12/10/201312:00
12/10/201315:00
12/10/201318:00
12/10/201321:00
13/10/201300:00
13/10/201303:00
13/10/201306:00
13/10/201309:00
13/10/201312:00
Significantheight(m)

P a g e | 48

P a g e | 49
Figure 15: Time and spatial variation of atmospheric pressure simulated along the track of Thane Cyclone,
using coupled ADCIRC+SWAN model from 8th October, 2013-12:00pm at 12hour interval until 13th
October, 2013 -12:00pm.
4.2.3 WIND STRESS/VELOCITY

P a g e | 50
Figure 16: Time and spatial variation of wind stress simulated along the track of Thane Cyclone, using
coupled ADCIRC+SWAN model from 8th October, 2013-12:00pm at 12h interval until 13th October, 2013-
12:00pm.

P a g e | 51
Figure 17: Maximum elevation model output (maxele.63) for PHAILIN cyclone from ADCIRC+SWAN model
run.
4.2.5 SIGNIFICANT WAVE HEIGHT

P a g e | 52
Figure 18: Time and spatial variation of significant wave height (Hs) simulated along the track of Thane Cyclone, using
October, 2013-12:00pm at 12hour interval until 13th
October, 2013-12:00pm.
Figure 19: Comparing SWAN generated and ADCIRC+SWAN generated significant wave height for
PHAILIN cyclone at landfall time.

P a g e | 53
UNIT-5
DISCUSSIONS

P a g e | 54
The model generated outputs were validated with available data and a comparative study
of various model generated outputs was done. To maximize the model execution speed and
minimize the required memory, ADCIRC 2DDI includes the non-linear terms explicitly.
Therefore, numerical instabilities are expected to be amplified where a time step is used, that
gives a Courant number based on wave celerity, Cr, [Cr = ∆t√𝑔ℎ/Δx] of order unity or larger
(Luettich et al., 1991).
The tight coupling of SWAN+ADCIRC enables waves, water levels and currents to
interact in complex problems and in a way that is accurate and efficient. The model generates
waves in deep water, and wave energy will be dissipated due to changes in wave-wave
interaction, bathymetry and bottom friction; the radiation stress gradients create set-up and wave-
driven currents in the circulation model, and then return those water levels and currents to the
wave model (Dietrich et al., 2011).
Atmospheric pressure gradient is a major component in deep water; the wind stress
becomes important on the shelf and continues to be a major driving force on the shore. Bottom
stress is a major dissipative force in shallow water; the wave radiation stress is a contributing
force limited to the shallow water breaking zone; it decreases its relative importance further
inland and offshore.
5.1 THANE CYCLONE
The outputs from 3 model runs as detailed in Table 2 were compared in order to assess
the impact of waves, currents and tides on the total surface elevation during the cyclone Thane.
The second model run was set up with ADCIRC model, with tides and cyclonic winds and
atmospheric pressure as inputs which has given us the maximum storm surge generated during
Thane. The third model run is also an ADCIRC model run during the cyclone time with only
cyclonic winds and atmospheric pressure forcing generated by the Holland Asymmetrical model
and no tidal effects. While the fourth experiment includes two-way coupling of ADCIRC and
SWAN i.e. the model run combined with winds, tides and waves.

P a g e | 55
Figure 20: Water surface elevation time series from different model runs, with inputs being specified, for
THANE Cyclone; for details please refer to Table 2.
Referring to Fig 19, we find that the two-way coupling shows a change in the total water
surface elevation during the Thane cyclone which implies that the inclusion of waves, tides and
currents (indirectly winds) can give good approximation in generating realistic sea surface
elevation using the model. Simultaneously, it has been noted that the wave-current interaction
need not necessarily give rise or fall of sea surface elevation, instead it seems that the kind of
interaction between tides, winds, waves and currents is what that determines the resultant.
It was also observed from the Fig 19, that tidal effects are crucial during any cyclone for
the sea surface elevation. If the storm makes the landfall during high tide, the water level is
observed to be higher than when it is during a low tide. It is to be noted that that storm tide is the
combination of the storm surge with the astronomical tide.
Dietrich et al. (2011) explained that as the waves move on to the continental shelf and
further nearshore, they break depending on the conditions, and exert a stress on the water column
which changes the water levels and/or drives the water currents. Change in the mean water level
is a result of the transfer of wave momentum to the water column. As wave momentum increases
in the presence of non-breaking waves, the mean water level lowers. As breaking commences,
the wave energy and momentum decrease, resulting in a reduction of the radiation stress carried
28-12-201100:00
28-12-201103:00
28-12-201106:00
28-12-201109:00
28-12-201112:00
28-12-201115:00
28-12-201118:00
28-12-201121:00
29-12-201100:00
29-12-201103:00
29-12-201106:00
29-12-201109:00
29-12-201112:00
29-12-201115:00
29-12-201118:00
29-12-201121:00
30-12-201100:00
30-12-201103:00
30-12-201106:00
30-12-201109:00
Watersurfaceelevation(m)
tides tides+winds winds tides+winds+waves

P a g e | 56
by the waves. These stresses are imparted into the water column. The rapid reduction of wave
radiation stress near the coast forces a rise in mean sea level.
The discharged momentum from the waves pushes against the water column, and
produces an opposing hydrostatic pressure gradient. During storm events, the resulting rise in
water level can play a major role in storm surge. According to linear theory, the effective change
in water level from a steady train of linear waves approaching normal to the shore on a gently
sloping bottom is about 19% of the breaking wave height (Dean and Dalrymple, 1991). This may
increase or decrease as we take into account nonlinear effects, dissipative forces, and wave
obliquity. The amount of wave set-up is also affected by the bottom contour of the nearshore and
beach face (Rao, 2009, Natural hazards).
5.1.2 CURRENTS
Figure 21: Time series of velocity magnitudes of currents under ADCIRC and ADCIRC+SWAN model runs,
for THANE Cyclone. For details please refer to Table 2.
The directions of alongshore currents during the flood and the ebb phases of tides
were observed to remain unchanged both in magnitude and direction at all stations from
Nagapattinam to Paradip along the ECI. (Misra et al., 2013). Studies by Mishra (2010), Mishra
(2011) and Panigrahi et al. (2010) explain that the maximum current speed along the east coast of
India varies from 0.2 to 0.5 m/s. Present study (Fig 20) records model simulated current speeds
upto 0.3 m/s which very well match with the former studies, in the absence of any extreme
0
0.2
0.4
0.6
0.8
1
1.2
1.4
25-12-201113:00
25-12-201118:00
25-12-201123:00
26-12-201104:00
26-12-201109:00
26-12-201114:00
26-12-201119:00
26-12-201100:00
27-12-201105:00
27-12-201110:00
27-12-201115:00
27-12-201120:00
28-12-201101:00
28-12-201106:00
28-12-201111:00
28-12-201116:00
28-12-201121:00
29-12-201102:00
29-12-201107:00
29-12-201112:00
29-12-201117:00
29-12-201122:00
30-12-201103:00
30-12-201108:00
Velocity(m/s)
tides+winds tides+winds+waves

P a g e | 57
weather event. However, when the Thane cyclone made landfall near Puducherry, the current
speed increased to 1.3 m/s from 0.3 m/s, which was followed by a sudden drop in current speed
to 0.1 m/s, after few hours from the landfall. The current speed predicted by the coupled
ADCIRC+SWAN model is higher than the ADCIRC prediction by ~0.2m/s.
5.1.3 WAVES
The modelled significant wave heights (Hs) were compared with the available buoy data
off Puducherry. ADCIRC+SWAN coupled model generated significant wave height was approx.
3m near Puducherry coast (Fig 7), before the landfall which matched well with the observed
values. From the model run, it was observed that the significant wave height increased rapidly
during landfall, which is due to cyclone effect. However, the model over-predicted wave heights
during landfall. This over-prediction may be due to missing physics (such as the warm-core
eddy) or poor numerics (such as the coarseness of the mesh). It was also noted from the
ADCIRC+SWAN coupled model outputs that the significant wave heights were higher on the
right side of the cyclone track in the direction of propagation (while on the left side of it,
comparatively lower values).
Comparison of the two model results indicates that currents have an effect on wave
characteristics. Wave period also becomes longer (shorter) when propagating following
(opposing) the direction of the current, but this could not be covered in this present study.
Changes in water depth also lead to changes in wave propagation, producing shoaling, refraction
or wave breaking, which in turn, changes the wave characteristics.
5.2 PHAILIN CYCLONE
The outputs from three model runs were compared in order to assess the impact of waves,
currents and tides on the total surface elevation during the cyclone Phailin. The first was set up
with ADCIRC model, with tides and cyclonic winds and atmospheric pressure as input which
gave us the maximum storm surge generated during Phailin cyclone. The second is also an
ADCIRC model run during cyclone time with only cyclonic winds and atmospheric pressure

P a g e | 58
forcing generated by the Holland Asymmetrical model, and no tidal effects. While the fourth
experiment included two-way coupling of ADCIRC and SWAN, i.e. the model run combining
wind and pressure fields with tidal and wave forcing.
The model generated outputs for Phailin cyclone was very much similar to Thane
cyclone, the two-way coupling showed a change in total water surface elevation during the
cyclone which implies that the only the inclusion of wave forcing along with the currents can
give good approximations in generating realistic model results. Simultaneously, it has been noted
that the wave-current interaction is not necessarily to give rise or fall of water surface elevation,
instead it seems that the kind of interaction between waves and currents is what that determines
the resultant.
Cyclone Phailin case also stressed on the idea that tidal effects are crucial during any
cyclone for the surge heights. If the storm makes landfall during high tide, the water level is
observed to be higher than when it is during a low tide. It is to be noted that that storm tide is the
combination of the storm surge with the astronomical tide.
5.2.2 CURRENTS AND WAVES
Figure 22: Significant wave height comparison between 2 model run outputs. The parameters included in each run are
specified.
0
1
2
3
4
5
6
7
11/10/201300:00
11/10/201303:00
11/10/201306:00
11/10/201309:00
11/10/201312:00
11/10/201315:00
11/10/201318:00
11/10/201321:00
12/10/201300:00
12/10/201303:00
12/10/201306:00
12/10/201309:00
12/10/201312:00
12/10/201315:00
12/10/201318:00
12/10/201321:00
13/10/201300:00
13/10/201303:00
13/10/201306:00
13/10/201309:00
13/10/201312:00
Significantwaveheight(m)
Chart Title
Tide+wind+wave Waves

P a g e | 59
Similar to Thane cyclone, it was observed that the current speed increased from the
normal levels when the Phailin cyclone made landfall near Ganjam, Odisha which was followed
by a sudden drop in current speed, after a few hours from the landfall.
Two wave modelling runs were successfully completed using SWAN and
ADCIRC+SWAN coupled model. The inputs like meteorological, bathymetry and parameters
were given identical to previous runs.
The modelled significant wave heights (Hs) were compared with different model runs
(Fig 21). From the ADCIRC+SWAN coupled model run it was observed that the significant
wave height increased rapidly during landfall time which is due to cyclone effect (Fig 21). The
model generated over-prediction for significant heights during landfall is expected to be because
of missing physics (such as the warm-core eddy) or poor numerics (such as the coarseness of the
mesh) as stated earlier.
It was also noted from the ADCIRC+SWAN coupled model outputs that the significant
wave heights were higher on the right side of the cyclone track in the direction of propagation,
while on the left side of it, they are comparatively lower values.

P a g e | 61
The importance of wave-current interaction in coastal regions has been asserted by many
former studies. Coastal region being a domain with high population density, developmental
activity is also picking-up momentum. Studies show that wave-current interaction during
extreme weather event can alter the coastal environment to a larger extent, and hence wave-
current interaction turns out to be very crucial factor to any coastal activity.
Even though many studies have been completed, the depth of understanding of wave-
current interaction is still in an infant stage, especially, during extreme weather events. But, the
lack of sufficient data at the time of extreme events and the difficulties in obtaining data during
such rare events, encourages us to use numerical ocean models for studying the above
phenomenon.
In this present work, ADCIRC circulation model, SWAN wave model and coupled
ADCIRC + SWAN models after tight coupling were used to compare and study the interactions
between waves and currents. The models were run on the same domain with same global
unstructured mesh. The east coast of India and the coastal region of Bay of Bengal were choosen
as the study area.
Different model runs under different oceanic and meteorological forcings such as
meteorological forcing alone, tidal forcing alone and tidal and meteorological forcing’s
combined together were carried out. The outputs generated, namely, water surface elevation,
currents and wave heights were analysed, and compared with model outputs generated from
SWAN (wave model) and ADCIRC+SWAN coupled model (circulation -wave model).
The model run was completed using a parallel mode of running. The model tuning factors
like the ramp function, bottom friction coefficient, constant Coriolis force were set to be the same
for all the model runs. The model runs were performed for the two cyclones, namely, Thane
cyclone of 2011 and Phailin cyclone of 2013 that hit the east coast of India (Puducherry and
Ganjam, respectively).
. The modelled significant wave heights (Hs) were compared with the available buoy data
off Puducherry. The ADCIRC+SWAN coupled model produced a significant wave height of
approx. 3m near Puducherry coast during the cyclone Thane before the landfall which matches
well with the observed values. From the model run, it was observed that the significant wave
height increased rapidly during landfall, which is due to cyclone effect. However, the model
over-predicts wave heights during landfall. Former studies suggest that this over-prediction may

P a g e | 62
be due to missing physics or poor numerics. It was also noted from the ADCIRC+SWAN
coupled model outputs that the significant wave heights were higher on the right side of the
cyclone track in the direction of propagation, compared to those on the left side of it.
Whatever work presented in this dissertation is to be considered as preliminary results, as
two models have been independently as well as coupled way used in a short period of time. For
obtaining accurate results of sea surface elevation, currents and waves, and to study wave-current
interaction, and use it for predicting oceanic parameters during extreme weather events, these
models need to be tuned further.

P a g e | 63
UNIT-7
REFERENCES

P a g e | 64
1. J.C. Dietrich , M. Zijlema, J.J. Westerink, L.H. Holthuijsen, C. Dawson, R.A. Luettich Jr.,
R.E. Jensen J.M. Smith, G.S. Stelling, G.W. Stone., 2011. Modeling hurricane waves and
storm surge using integrally-coupled, scalable computations. Coastal Engineering 58 (2011)
45–65.
2. Prasad K. Bhaskaran, R. Gayathri, P.L.N.Murty, SubbaReddy Bonthu, Debabrata Sen., 2014.
A numerical study of coastal inundation and its validation for Thane cyclone in the Bay of
Bengal. Coastal Engineering 83 (2014) 108–118.
3. P.L.N. Murty, K.G. Sandhya, Prasad K. Bhaskaran, Felix Jose, R. Gayathri, T.M.
Balakrishnan Nair, T. Srinivasa Kumar, S.S.C. Shenoi., 2014. A coupled hydrodynamic
modeling system for PHAILIN cyclone in the Bay of Bengal. Coastal Engineering 93 (2014)
71–81.
4. IMD, 2013. Very severe cyclonic storm. Phailin over the Bay of Bengal (8–14 October
2013): a report India Meteorological Department, New Delhi, p. 43, (2013).
5. J Wolf, D Prandle., 1999. Some observations of wave–current interaction. Coastal
Engineering Volume 37, Issues 3–4, August 1999, Pages 471–485.
6. Dietrich, J.C., 2005. Implementation and Assessment of ADCIRC's Wetting and Drying
Algorithm. (M.S. Thesis) University of Oklahoma 223.
7. Dietrich, J.C., 2010. Development and Application of Coupled Hurricane Wave and Surge
Models for Southern Louisiana. (Ph.D Thesis) University of Notre Dame 337.
8. Dietrich, J.C., Kolar, R.L., Luettich, R.A., 2004. Assessment of ADCIRC's wetting and
drying algorithm. Dev. Water Sci. 55 (2), 1767–1778.
9. Bhaskaran, P.K., Nayak, S., Bonthu, S.R., Murty, P.L.N., Sen, D., 2013. Performance and
validation of a coupled parallel ADCIRC–SWAN model for THANE cyclone in the Bay of
Bengal. Environ. Fluid Mech. 13 (6), 601–623.

P a g e | 65
10. Rao A D, Indu Jain, M V Ramana Murthy, T S Murty, and S K Dube, 2009 “Impact of
cyclonic wind field on interaction of surge-wave computations using finite-element and
finite-difference models”, Natural Hazards, 49, 2, 225-239, 2009.
11. Amrutha, MM and Kumar, VS and Anoop, TR and Balakrishnan Nair, TM and Nherakkol, A
and Jeyakumar, C (2014) Waves off Gopalpur, northern Bay of Bengal during Cyclone Phailin.
Annales Geophysicae, 32 (9). pp. 1073-1083.
12. J.C. Dietrich, S. Tanaka, J.J. Westerink, C.N. Dawson, R.A. Luettich Jr., M. Zijlema, L.H.
Holthuijsen, J.M. Smith, L.G. Westerink, H.J. Westerink., 2012. Performance of the
Unstructured-Mesh, SWAN+ ADCIRC Model in Computing Hurricane Waves and Surge. J
Sci Comput (2012) 52:468–497
13. Dietrich, J.C., Bunya, S.,Westerink, J.J., Ebersole, B.A.,Smith, J.M.,Atkinson, J.H., et al.,
2010.A high resolution coupled riverine flow, tide, wind, wind wave and storm surge model
for southern Louisiana and Mississippi: Part II — synoptic description and analyses of
Hurricanes Katrina and Rita. Monthly Weather Review 138, 378–404.
14. Hong Zhang.
, Ole S. Madsen, S.A. Sannasiraj, Eng Soon Chan., 2004. Hydrodynamic model
with wave–current interaction in coastal regions.Estuarine, Coastal and Shelf Science,
Volume 61, Issue 2, October 2004, Pages 317–324
15. Dietrich, J.C., Tanaka, S., Westerink, J.J., Dawson, C.N., Luettich Jr., R.A., Zijlema, M.,
Holthuijsen, L.H., Smith, J.M., Westerink, L.G., Westerink, H.J., 2012. Performance of the
unstructured mesh SWAN + ADCIRC model in computing hurricane waves andsurge. J. Sci.
Comput. 52 (2), 468–497.
16. Liliana Rusu, C. Guedes Soares., 2011.Modelling the wave–current interactions in an
offshore basin using the SWAN model. Ocean Engineering, Volume 38, Issue 1, January
2011, Pages 63–76.
17. Nair, Balakrishnan T. M. ; Remya, P. G. ; Harikumar, R. ; Sandhya, K. G. ; Sirisha, P. ;
Srinivas, K. ; Nagaraju, C. ; Nherakkol, Arun ; Prasad, Krishna B. ; Jeyakumar, C. ;

P a g e | 66
Kaviyazhahu, K. ; Hithin, N. K. ; Kumari, Rakhi ; Kumar, Sanil V. ; Kumar, Ramesh M. ;
Shenoi, S. S. C. ; Nayak, Shailesh (2014) Wave forecasting and monitoring during very
severe cyclone Phailin in the Bay of Bengal Current Science, 106 (8). pp. 1121-1125. ISSN
0011-3891
18. L. Rusu, M. Bernardino, C. Guedes Soares., 2011. Modelling the influence of
currentsonwave propagation at the entrance of the Tagus estuary. Ocean Engineering,
Volume 38, Issue 10, July 2011, Pages 1174-1183.
19. Luettich, R.A., Westerink, J.J., 2004. Formulation and numerical implementation of the
2D/3D ADCIRC Finite Element Model Version 44.XX. http://adcirc.org/adcirc_theory_
2004_12_08.pdf2004.
20. Westerink, J.J., Luettich, R.A., Feyen, J.C., Atkinson, J.H., Dawson, C., Roberts, H.J., et al.,
2008. A basin to channel scale unstructured grid hurricane storm surge model applied to
southern Louisiana. Monthly Weather Review 136 (3), 833–864.
21. Bhaskaran, P.K., Nayak, S., Bonthu, S.R., Murty, P.L.N., Sen, D., 2013. Performance and
validation of a coupled parallel ADCIRC–SWAN model for THANE cyclone in the Bay of
Bengal. Environ. Fluid Mech. http://dx.doi.org/10.1007/s10652-013-9284-5.
22. Beven, J. L., II, and Kimberlain, T., 2009. “Tropical cyclone report, Hurricane Gustav
(AL072008), 25 August–4 September, 2008.” National Hurricane Center, National
Oceanographic and Atmospheric Administration, Washington, DC.J. Geophys. Res.,
104(C4).
23. Zijlema, M. (2010). “Computation of wind-wave spectra in coastal waters with SWAN on
unstructured grids.” Coastal Eng., 57(3), 267–277.
24. Bunya, S., Dietrich, J.C., Westerink, J.J., Ebersole, B.A., Smith, J.M., Atkinson, J.H., et al.,
2010. A high resolution coupled riverine flow, tide, wind, wind wave and storm surge model

P a g e | 67
for southern Louisiana and Mississippi: Part I — model development and validation. Monthly
Weather Review 138, 345–377.
25. Liliana Rusu and C.Guedes Soares, (2011). Modelling the wave–current interactions in an
offshore basin using the SWAN model. Ocean Engineering 38 (2011) 63–76.

PROJECT REPORT

More Related Content

What's hot (20)

Similar to PROJECT REPORT (20)

PROJECT REPORT