Journal of Coastal Research 26 3 523–534 West Palm Beach, Florida May 2010

Coastal Vulnerability Assessment for Orissa State,

East Coast of India
T. Srinivasa Kumar{, R.S. Mahendra{, Shailesh Nayak{, K. Radhakrishnan{, and K.C. Sahu1
{ { 1
Indian National Centre for Ocean Vikram Sarabhai Space Centre (VSSC) Department of Marine Sciences
Information Services (INCOIS) Thiruvananthapuram—695022 India Berhampur University
Hyderabad—500 055 India Berhampur—760 007 India
srinivas@incois.gov.in

ABSTRACT
KUMAR, T.S.; MAHENDRA, R.S.; NAYAK, S.; RADHAKRISHNAN, K., and SAHU, K.C., 2010. Coastal vulnerability
assessment for Orissa State, east coast of India. Journal of Coastal Research, 26(3), 523–534. West Palm Beach (Florida),
ISSN 0749-0208.

Coastal areas of Orissa State in the northeastern part of the Indian peninsula are potentially vulnerable to accelerated
erosion hazard. Along the 480-km coastline, most of the coastal areas, including tourist resorts, hotels, fishing villages,
and towns, are already threatened by recurring storm flood events and severe coastal erosion. The coastal habitats,
namely the largest rookeries in the world for olive Ridley sea turtles (the extensive sandy beaches of Gahirmatha and
Rushikulya), Asia’s largest brackish water lagoon (the ‘‘Chilika’’), extensive mangrove cover of Bhitarkanika (the wildlife
sanctuary), the estuarine systems, and deltaic plains are no exception. .The present study therefore is an attempt to
develop a coastal vulnerability index (CVI) for the maritime state of Orissa using eight relative risk variables. Most of
these parameters are dynamic in nature and require a large amount of data from different sources. In some cases, the
base data is from remote sensing satellites; for others it is either from long-term in situ measurements or from numerical
models. Zones of vulnerability to coastal natural hazards of different magnitude (high, medium, and low) are identified
and shown on a map. In earlier studies, tidal range was assumed to include both permanent and episodic inundation
hazards. However, the mean of the long-term tidal records tends to dampen the effect of episodic inundation hazards
such as tsunamis. For this reason, in the present study, tsunami run-up has been considered as an additional physical
process parameter to calculate the CVI. Coastal regional elevation has also been considered as an additional important
variable. This is the first such study that has been undertaken for a part of the Indian coastline. The map prepared for
the Orissa coast under this study can be used by the state and district administration involved in the disaster mitigation
and management plan.

ADDITIONAL INDEX WORDS: Coastal vulnerability index, Tunami N2 Model, MIKE-21, GLOSS/SRTM data,
geographic information systems, Orissa, erosion hazard area.

INTRODUCTION resources as well as increasing people’s exposure to coastal

hazards. At least 200 million people were estimated to live in
Tremendous population and developmental pressures have the coastal floodplain in 1990 (in the area inundated by a 1 in
been building in the coastal areas for the last four decades.
1000 year flood), and it is likely that their number will increase
According to the estimates of the United Nations in 1992, more
to 600 million by the year 2100 (Mimura and Nicholls, 1998).
than half of the world’s population lives within 60 km of a
Furthermore, global climate change and the threat of acceler-
shoreline. Also, urbanization and the rapid growth of coastal
ated sea-level rise exacerbate the already existing high risks of
cities have been dominant population trends over the last few
storm surges, severe waves, and tsunamis. Over the last 100
decades, leading to the development of numerous mega cities in
years, global sea level rose by 1.0–2.5 mm/y. Present estimates
all coastal regions around the world. There were only 2 mega
of future sea-level rise induced by climate change range from 20
cities in 1950 (New York and London), whereas there were 20
to 86 cm for the year 2100, with a best estimate of 49 cm. It has
mega cities in 1990. It has been projected that there will be 30
been estimated that a 1-m rise in sea-level could displace nearly
mega cities by 2010, having a population of 320 million people
7 million people from their homes in India (Ipcc, 2001).
(Nicholls, 1995). The ratio of people living in coastal zones
Scientific study of the natural hazards and coastal processes
compared with available coastal lands further indicates that
there is a greater tendency for people to live in coastal areas of the Indian coast has assumed greater significance after the
than inland. According to the United Nations Environment December 2004 tsunami because the country learned lessons
Programme (UNEP) report, the average population density in on the impact of natural hazards in terms of high damage
the coastal zone was 77 people/km2 in 1990 and 87 people/km2 potential for life, property, and the environment. The nation’s
in 2000, and a projected 99 people/km2 in 2010 (Unep, 2007). rapidly growing population of coastal residents and their
Collectively, this is placing both growing demands on coastal demand for reliable information regarding the vulnerability
of coastal regions have created a need for classifying coastal
DOI: 10.2112/09-1186.1 received 14 January 2010; accepted in lands and evaluating the hazard vulnerability. Government
revision 10 February 2010. officials and resource managers responsible for dealing with
524 Kumar et al.

natural hazards also need accurate assessments of coastal vulnerability to sea-level rise in tide-dominated, sedimentary
hazards to make informed decisions before, during, and after coastal regions.
such hazard events. Thieler and Hammer-Klose (1999) used coastal slope,
Disciplines such as geography, physical, urban, or territorial geomorphology, relative sea-level rise rate, shoreline change
planning, economics, and environmental management helped rate, mean tidal range, and mean wave height for assessment
to strengthen what can be called an applied science approach to of coastal vulnerability of the U.S. Atlantic coast. The result
disasters. Maps became more and more common because of the showed that 28% of the U.S. Atlantic coast is of low
ever greater participation of geologists, geotechnical engineers, vulnerability, 24% of the coast is of moderate vulnerability,
hydrologists, and other experts. They were able to provide 22% is of high vulnerability, and 26% is of very high
required data for the adequate identification of the danger or vulnerability.
hazard zones, according to the area of influence of the natural Pendleton, Thieler, and Jeffress (2005) assessed the coastal
phenomena. Computer science tools such as geographic vulnerability of Golden Gate National Recreation area to sea-
information systems (GIS) have facilitated this type of level rise by calculating a coastal vulnerability index (CVI)
identification and analysis. This type of study or analysis of using both geologic (shoreline-change rate, coastal geomor-
risk has increasingly been presented with the intention of phology, coastal slope) and physical process variables (sea-level
contributing data on threats or risk to physical and territorial change rate, mean significant wave height, mean tidal range).
planning specialists as an ingredient of the decision making The CVI allows the six variables to be related in a quantifiable
process (Bankoff, Frerks, and Hilhorst, 2003). manner that expresses the relative vulnerability of the coast to
physical changes due to future sea-level rise.
PREVIOUS WORK
STUDY AREA
Hegde and Reju (2007) developed a coastal vulnerability
index for the Mangalore coast using geomorphology, regional Orissa, located in the northeastern coast of India, is a
coastal slope, shoreline change rates, and population. However, maritime state with immense potential in natural resources. It
they opined that additional physical parameters like wave is located between 17u499 N and 22u349 N latitudes and 81u279 E
height, tidal range, probability of storm, etc., can enhance the and 87u299 E longitudes. Orissa State covers an area of
quality of the CVI. 156,000 km2 and has a total population of 36.7 million (2001
Gornitz (1990) assessed the vulnerability of the east coast of census). The state has a population density of 236 persons/km2
the United States with emphasis on future sea level rise. (2001 census) covering 30 districts including six coastal
Dominey-Howes and Papathoma (2003) applied a new districts, viz., Balasore, Bhadrak, Kendrapada, Jagatsinghpur,
tsunami vulnerability assessment method to classify building Puri, and Ganjam, spanning a coastline of 480 km (Figure 1).
vulnerability (BV) by taking the worse case of the tsunami The total population of these six coastal districts is 8,975,581
scenario of 7 February 1963, for two coastal villages in the Gulf and is distributed in an area of 21,887 km2 with a population
of Corinth, Greece. The result showed 46.5% of all buildings are density 410 persons/km2 (2001 census). The study area enjoys
classified as highly vulnerable (BV) and 85% of all businesses international importance and is one of the sites of world
are located within buildings with a high BV classification and heritage attracting tourists and pilgrims. It is gifted with Asia’s
13.7% of the population is located within buildings with a high largest brackish water lagoon, the Chilika; a 672 km2 extensive
BV class. mangrove forest and wetland, the Bhitarkanika wildlife
Rajawat et al. (2006) delineated the hazard line along the sanctuary; and the world’s largest known nesting beaches of
Indian coast using data on coastline displacement, tide, waves, olive Ridley sea turtles, the Gahirmatha and the Rushikulya.
and elevation. It is pitiable that Orissa is also vulnerable to multiple
Pradeep Kumar and Thakur (2007) assessed the role of disasters such as tropical cyclones, storm surges, and tsuna-
bathymetry in modifying the propagation of the tsunami wave mis. The threat of the coastal vulnerability to such hazards has
of 26 December 2004 and concluded that undersea configura- increased manyfold with the growing population. The economy
tion has an important role in enhancing the wave height of the state has received tremendous setbacks because several
because some coastal stations on the eastern margin of India natural hazards occurred in succession. The coastal districts of
have suffered maximum damage wherein bathymetry has Orissa have experienced major surges in the past. Severe
shown an anomalous pattern. flooding caused by storm surges during the 1999 super cyclone
Dinesh Kumar (2006) used sea-level rise scenario for caused massive destruction to life and property. Extreme sea
calculating the potential vulnerability for coastal zones of levels are major causes of concern for coastal flooding in this
Cochin, southwest coast of India, and concluded that climate- region. The loss of land to the sea has now become a more
induced sea-level rise will bring profound effects on coastal recurrent phenomenon. Identification of vulnerable areas and
zones. effective risk mapping and assessment is the need of the hour.
Belperio et al. (2001) considered elevation, exposure, aspect, Damage can certainly be minimized if extreme sea levels are
and slope as the physical parameters for assessing the coastal forecast well in advance. The government of Orissa, after
vulnerability to sea-level rise and concluded that coastal witnessing the alarming situation in the state, has decided to
vulnerability is strongly correlated with elevation and expo- start with an ‘‘Integrated Coastal Zone Management Plan
sure, and that regional scale distributed coastal process Development Project’’ financed by the World Bank to the tune
modeling may be suitable as a ‘‘first cut’’ in assessing coastal of INR 100 crores. The present study is an attempt to develop

Journal of Coastal Research, Vol. 26, No. 3, 2010

 Coastal Vulnerability Assessment 525

Table 1. Data used for the study on coastal hazard.

Parameter Data Used Resolution Period

Shore-line change rate LANDSAT MSS 57 m 1972

LANDSAT TM 30 m 1991, 2000
Sea-level change rate GLOSS data — 1900–2000
Coastal slope GEBCO data 1 min —
Significant wave height Numerical model 0.25u 2005
Tidal range Prediction tool –— 2006
Coastal regional SRTM data 90 m —
elevation
Coastal geomorphology IRS P6 LISS IV 5.8 m 2005
Tsunami run-up Numerical model 1 min —

— 5 not applicable.

ogy, rate of sea-level change, past shoreline evolution, and

other factors by following the method of estimating the CVI.
This approach combines the coastal system’s susceptibility to
change with its natural ability to adapt to changing environ-
mental conditions and yields a relative measure of the system’s
natural vulnerability to the effects of sea-level rise (Klein and
Nicholls, 1999). All these parameters have been related in a
quantifiable manner that expresses the relative vulnerability
of the coast. This method uses a rating system that classifies
the coastal area based on degree of vulnerability as low,
medium, and high according to the CVI value of that area.
The method of computing the CVI in the present study is
similar to that used in Pendleton, Thieler, and Jeffress (2005);
Thieler (2000); and Thieler and Hammar-Klose (1999). In
Figure 1. Study area. addition to the six parameters used by earlier researchers, the
present study uses an additional geologic process variable, i.e.,
coastal regional elevation, and an additional physical process
coastal vulnerability indices for Orissa, which can facilitate the variable, i.e., tsunami run-up. The eight relative risk variables
state and district administration involved in disaster mitiga- used are shoreline change rate, sea-level change rate, coastal
tion and management. slope, mean significant wave height, mean tidal range, coastal
regional elevation, coastal geomorphology, and tsunami run-
METHODOLOGY up. This is the first such study that has ever been undertaken
for a part of the Indian coastline.
Vulnerability may be defined as an internal risk factor of the Most of these parameters are dynamic in nature and require
subject or system that is exposed to a hazard and corresponds to a large amount of data from different sources to be acquired,
its intrinsic predisposition to be affected, or to be susceptible to analyzed, and processed. They are derived from remote
damage. In general, the concept of ‘‘hazard’’ is now used to refer sensing, GIS, and numerical model data. Data sets used for
to a latent danger or an external risk factor of a system or the present study in deriving each of these parameters is
exposed subject. This can be in mathematical form as the presented in Table 1.
probability of occurrence of an event of certain intensity in a The importance of each of the considered parameters and the
specific site and during a determined period of exposure. On the procedure to generate the same for use in assessment of CVI
other hand, vulnerability may be understood, in general terms, are given in the following section.
as an internal risk factor that is mathematically expressed as
the feasibility that the exposed subject or system may be
Shoreline Change Rate
affected by the phenomenon that characterizes the hazard.
Thus, risk is the potential loss to the exposed subject or system Coastal shorelines are always subjected to changes due to
resulting from the convolution of hazard and vulnerability. In coastal processes, which are controlled by wave characteristics
this sense, risk may be expressed in a mathematical form as the and the resultant near-shore circulation, sediment character-
probability of surpassing a determined level of economic, social, istics, beach form, etc. From the coastal vulnerability point of
or environmental consequence at a certain site and during a view, coasts subjected to accretion will be considered as less
certain period. vulnerable areas as they move toward the ocean and result in
Although a viable, quantitative predictive approach is not the addition of land areas, whereas areas of coastal erosion will
available, the relative vulnerability of different coastal envi- be considered as more vulnerable because of the resultant loss
ronments to sea-level rise may be quantified at a regional to of private and public property and important natural habitats
national scale using basic information on coastal geomorphol- such as beaches, dunes, and marshes. It also reduces the

Journal of Coastal Research, Vol. 26, No. 3, 2010

526 Kumar et al.

distance between coastal population and ocean, thereby of tsunami hazard for the coast (Dotsenko, 2005). Coastal slope
increasing the risk of exposure of population to coastal hazards. characteristic is an important parameter in deciding the degree
Ortho-rectified Landsat MSS and TM images covering the to which coastal land is at risk of flooding from storm surges
Orissa coastline for the years 1970, 1980, and 2000 were and during a tsunami (Klein, Reese, and Sterr, 2000). Coastal
downloaded from Michigan State University (2008). The data locations having gentle land slope values have great penetra-
have been projected to the Universal Transverse Mercator tion of seawater compared with locations with fewer slopes, and
(UTM) projection system with WGS-84 datum. The shoreline resulting land loss from inundation is simply a function of
along the Orissa coastline was digitized using ArcMap 9.2 and slope: the lower the slope, the greater the land loss ((Klein,
ERDAS Imagine software using the on-screen point mode Reese, and Sterr, 2000). Thus coastal areas having gentle slope
digitization technique. The near infrared band that is most were considered as highly vulnerable areas and areas of steep
suitable for the demarcation of the land–water boundary has slope as areas of low vulnerablity.
been used to extract the shoreline. The digitized shoreline for General Bathymetric Chart of the Oceans (GEBCO) data of
the years 1970, 1980, and 2000 in the vector format were used one-minute grid resolution coastal topography and bathymetry
as the input to the Digital Shoreline Analysis System (DSAS) to have been used to get the regional slope of the coastal area. It
calculate the rate of shoreline change. The inputs required for also incorporates land elevations derived from the Global Land
this tool are shoreline in the vector format, date of each vector One-kilometer Base Elevation project data set. GEBCO data
layer, and transect distance. The rate of shoreline change is are useful in deriving the coastal slope values on both land and
calculated for the entire study area, and risk ratings are in the ocean. The slope values in degrees are calculated using
assigned. the Environmental Information System software package. The
slope is calculated for the entire study area, and risk ratings are
Sea-Level Change Rate assigned.

Sea-level rise is an important consequence of climate change,

both for societies and for the environment. Mean sea level at Significant Wave Height
the coast is defined as the height of the sea with respect to a
Heights of the waves depend on characteristics of the wind
local land benchmark, averaged over a period, such as a month
responsible for generating them (Ashok Kumar, Raju, and
or a year—long enough that fluctuations caused by waves and
Sanil Kumar, 2005). Significant wave height is the average
tides are largely removed. Changes in mean sea level as
height (trough to crest) of the one-third highest waves valid for
measured by coastal tide gauges are called relative sea-level
the indicated 12-hour period. Mean significant wave height is
changes (Church and Gregory, 2001). Sea-level rise can be a
used here as a proxy for wave energy, which drives coastal
product of global warming through two main processes:
sediment transport (Usgs, 2005). In general, wave heights are
thermal expansion of seawater and widespread melting of land
considered to demarcate the vulnerability line all along the
ice. Global warming is predicted to cause significant rises in sea
coast. The vulnerability study based on wave height is an
level over the course of the twenty-first century. Thus it
important step in setting up an all-hazards warning and
becomes necessary to study the effect of sea-level rise on the
management system (Usgs, 2005). Wave energy increases as
coastal areas. From the coastal vulnerability point of view,
coast subjected to a high rate of sea-level rise is considered as a the square of the wave height; thus the ability to mobilize and
high vulnerable area and vice versa. transport beach/coastal materials is a function of wave height
(Usgs, 2001). The wave energy increases with increase in the
The tide gauge data set of the Global Sea-level Observing
System (GLOSS) during the past century is used as the wave height, which results in loss of land area due to increased
primary source of information for sea-level trend in the study erosion and inundation along shore, so those coastal areas of
area. Tide gauge data recorded in 15 stations around the Indian high wave height are considered as more vulnerable coasts and
Ocean, including Paradip, Sagar, Visakhapatnam, Chennai, areas of low wave height as less vulnerable coasts.
etc., during the period from 1900 to 2005 is used to form the sea- In the present study, MIKE 21 SW software, a new third-
level–rise rate contour. The rate of sea-level change is generation spectral wind–wave model was used to estimate the
calculated for the entire study area, and risk ratings are significant wave height in the study area. MIKE 21 SW, which
assigned. simulates the growth, decay, and transformation of wind-
generated waves and swells in offshore and coastal areas,
solves the spectral wave action balance equation formulated in
Coastal Slope
either Cartesian or spherical coordinates. The model includes
Slope is used to describe the measurement of the steepness, the following physical phenomena: wave growth by action of
incline, gradient, or grade of a straight line. A higher slope wind, nonlinear wave–wave interaction, dissipation by white
value indicates a steeper slope and vice versa. The coastal slope capping, dissipation by wave breaking, dissipation due to
is defined as the ratio of the altitude change to the horizontal bottom friction, refraction due to depth variations, and wave–
distance between any two points on the coast. Coastal slope current interaction. Daily significant wave height data were
(steepness or flatness of the coastal region) is linked to the generated using this software forced with wind data from
susceptibility of a coast to inundation by flooding (Thieler, European Center for Medium-Range Weather Forecast
2000). The run-up of waves on a coast is the most important (ECMWF) for the year 2005, and the mean values were
stage of a tsunami from the viewpoint of evaluation of the level calculated and risk ratings were assigned.

Journal of Coastal Research, Vol. 26, No. 3, 2010

 Coastal Vulnerability Assessment 527

Tidal Range coastal areas (Ipcc, 2001). Rising sea level will bring about the
redistribution of coastal landforms comprising subtidal bed-
Forced by the gravitational attraction of the moon and the forms, intertidal flats, salt marshes, shingle banks, sand dunes,
sun, tides are periodic and highly predictable. Tidal range is the cliffs, and coastal lowlands (Pethick and Crooks, 2000). This
vertical difference between the highest high tide and the lowest evolution in geomorphology will determine not only the quality
low tide. Tidal range is linked to both permanent and episodic and quantity of associated habitats and the nature of their
inundation hazards. From the vulnerability point of view, it is ecosystem linkages but also the level of vulnerability of wildlife,
an obvious tendency to designate coastal areas of high tidal people, and infrastructure in coastal areas.
range as highly vulnerable. This decision was based on the Coastal geomorphology is a result of prevailing geomorphic
concept that large tidal range is associated with strong tidal processes that were forced to attain the present morphology.
currents that influence coastal behavior. For the current study, Hence the geomorphic units are the indicators of the coastal
coastal areas with high tidal range are considered as high processes that act on it. The term ‘‘coastal vulnerability’’ as
vulnerable and low tidal range as low vulnerable. used in this study refers to the (geomorphic) vulnerability of
In the current study, predicted tide data from WXTide coastal landforms to hazards such as wave erosion, tsunami,
software for the year 2006 is taken as the base data, and the and storm surge flooding, etc. The study on coastal vulnerabil-
maximum amplitudes of the tide in a year for the Indian coastal ity assessment described here identifies coastal areas that are
locations are calculated, and risk rates are assigned. in many cases already vulnerable to coastal hazards under
present-day conditions but that are likely to become increas-
Coastal Regional Elevation ingly vulnerable in future as a result of climate change and sea-
level rise.
Regional elevation is referred to as the average elevation of a Indian Remote Sensing Satellite (IRS) P6 Linear Imaging
particular area above mean sea level. It is important to study Self-scanning Sensor-IV (LISS-IV) data and the Digital Terrain
the coastal regional elevation detail for the study area to Model (DTM) have been used to extract the coastal geomor-
identify and estimate the extent of land area threatened by phology. LISS-IV satellite data were imported into the ERDAS
future sea-level rise. These coastal elevation data are also used Imagine 9.1 image processing software package. The satellite
to estimate the land potentially available for wetland migration data were geo-corrected using the reference image and
in response to sea-level rise and the sea-level rise impacts to the projected to the UTM projection system. Then coastal geomor-
human built environment (Anderson et al., 2005). From the phic classes were extracted based on the visual interpretation
coastal vulnerability point of view, coastal regions having high keys using the on-screen digitization technique. The coastline
elevation will be considered as less vulnerable areas because geomorphology has been classified based on the dominant
they provide more resistance for inundation against the rising geomorphic class representing the section of coastal zone
sea level, tsunami run-up, and storm surge. Those coastal (500 m). Coastline representing the geomorphology has been
regions having low elevation are considered as highly vulner- overlaid on the DTM using ESRI 3D Analyst. Using the
able areas. topographic information from the DTM, cliff areas were
In the present study, Shuttle Radar Topography Mission identified and classified. The classes recorded in the study
(SRTM) data are used to derive the coastal regional elevation. area include sandy beach, delta, mangrove, cliff, estuary, mud
The 90-m resolution SRTM raster data are resampled to 1 km flat, spits, aquaculture and salt pans, and inundated coasts.
and risk rates are assigned to the entire coastline based on the Further, these geomorphic classes were assigned the risk
elevation values. rating as high vulnerable (sandy beaches, deltas, mangroves,
spits), medium vulnerable (estuaries), and low vulnerable
Coastal Geomorphology (cliffs, aquaculture and salt pans, and inundated coasts).

Geomorphology is defined as the study of landforms and

Tsunami Arrival Height
landscapes, including the description, classification, origin,
development, and history of planetary surfaces. Geomorphol- Tsunamis result in generation of waves of different periods
ogy seeks to identify the regularities among landforms and and height. These wave parameters depend on earthquake
what processes lead to patterns. Geomorphology includes source parameters, bathymetry, beach profile, coastal land
endogenic processes—volcanism, tectonics, flooding, cyclones, topography, and presence of coastal structures. These surges
tsunami, faulting and wrapping—and exogenic processes— cause flooding of seawater into the land as much as 1 km or
weathering, mass wasting, erosion, transportation, and depo- even more, resulting in loss of human life and damage to
sition. The processes responsible for this are alluvial and property.
fluvial, glacial, aeolian, and coastal. The Indo–Burma–Sumatra subduction zone is known to
Coastal geomorphology provides a basic understanding of trigger large undersea earthquakes that are capable of
the coastal environment. With the predicted rise in sea level as generating tsunamis in the Indian Ocean. Indicators suggest
a result of global warming, there has been increasing a high potential for giant earthquakes along the coast of
speculation and concern as to the impact on coastal geomor- Myanmar (Cummins, 2007) that could be especially dangerous
phology. Sea-level rise and changes to wave conditions will for the east coast of India. In the current study, a magnitude
likely bring changes in the dimension and function of the 9.5-Moment Magnitude (Mw) earthquake with epicenter in the
coastal habitats as well as increased risk to those living in the Andaman subduction zone has been considered to cause the

Journal of Coastal Research, Vol. 26, No. 3, 2010

528 Kumar et al.

Table 2. Risk rating assigned for different parameters.

Risk Rating

Variable Low (1) Medium (2) High (3)

Shoreline change rate (m/y) .0 (accretion) $210 and ,0 (erosion) ,210 (severe erosion)
Sea-level change rate (mm/y) ,50 .0 and #1.0 .1.0 and #2.0
Coastal slope (degrees) .1.0 .0.2 and #1.0 $0 and #0.2
Significant wave height (m) — 1.25–1.40 —
Tidal range (m) #2.5 .2.5 and #3.5 .3.5
Regional elevation (m) .6.0 .3.0 and #6.0 $0 and #3.0
Geomorphology Inundated coasts, cliffs Estuaries, vegetated coasts Sandy beaches, deltas, spits,
(other than mangroves) mangroves, mud flat
Tsunami arrival height (m) $0 and #1.0 .1.0 and #2.0 .2.0

worst-case tsunami scenario for Orissa state. This translates to g 5 risk rating assigned to coastal geomorphology
a source segment of 1200 km in length extending from h 5 risk rating assigned to tsunami run-up
Myanmar in the north to Car Nicobar in the south, 300 km in
The CVI is calculated based on the risk values assigned to
width, 15 m slip and strike angle parallel to the plate boundary
input parameters using the simple vector algebraic technique
of the subduction zone.
using ESRI ArcMap software. The CVI values thus generated
The TUNAMI N2 model has been used, which basically takes
for different segments of the coastline are categorized into
the seismic deformation and bathymetry as input to predict the
three CVI classes, viz., low, medium and high vulnerable
run-up heights and travel times of a tsunami wave for different
corresponding to ,25th percentile, 25th–50th percentile, and
parts of the coastline for any given earthquake. GEBCO
.50th percentile, respectively.
bathymetric data have been used as input in the model. The
seismic deformation for an earthquake has been computed
using the earthquake parameters like location, focal depth,
strike, dip and rake angles, length, width, and slip of the fault
plane (Mansinha and Smylie, 1971). Based on the run-up
estimated along the entire study area, the risk ratings are
assigned.

Calculation of CVI
The CVI is determined by combining the relative risk
variables to create a single indicator. For the purpose of the
current study, the entire coastline is divided into grids of 1 km
3 1 km. Each of the eight input relative risk variables are then
assigned appropriate risk classes 1, 2, and 3 based on its ability
to cause low, medium and high damage, respectively, for a
particular area of the coastline. After this process, each coastal
grid will have risk ratings for all eight variables under
consideration. The risk rating assigned for each variable is
given in Table 2.
Once each section of coastline is assigned a risk value for
each variable, the CVI is calculated as the square root of the
product of the ranked variables divided by the total number of
variables (Pendleton, Thieler, and Jeffress, 2005). The CVI is
represented by the Equation (1).
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
CVI ~ ða  b  c  d  e  f  g  hÞ=8, ð1Þ

where

a 5 risk rating assigned to shoreline-change rate

b 5 risk rating assigned to sea-level change rate
c 5 risk rating assigned to coastal slope
d 5 risk rating assigned to significant wave height
e 5 risk rating assigned to tidal range Figure 2. Risk classes for shoreline change rate.
f 5 risk rating assigned to coastal regional elevation

Journal of Coastal Research, Vol. 26, No. 3, 2010

 Coastal Vulnerability Assessment 529

Figure 3. Risk classes for sea-level change rate. Figure 4. Risk classes for coastal slope.

The CVI is an indication of the relative vulnerability of the of more than 1.0 mm/y along the coastal stretches of Ganjam,
various segments of the Orissa coast to coastal inundation Chilika, Puri, Jagatsinghpur, and southern Balasore districts.
hazards. The map prepared for the Orissa coast under this About 166 km of coastline has a medium risk rating with sea-
study can be used by state and district administrations level change rates between 0.1 and 1.0 mm/y along the coastal
involved in the disaster mitigation and management to take stretches of Kendrapara and Bhadrak districts. About 23 km of
advance action to mitigate the effects of impending disasters coastline that has not recorded change in sea level along the
and to prioritize areas for evacuation. northern Balasore district has a low risk rating (Figure 3).

RESULTS
Coastal Slope
Shoreline Change Rate
The present study revealed that most of the study area
The present study revealed that about 55 km of coastline has (429 km of coastline) has a high risk rating, recording coastal
a high risk rating, recording erosion rates of more than 10 m/y slopes of less than 0.2u. Only 41 km of coastal stretch covering
along the coastal stretches north of Puri, of central Kendra- parts of Ganjam and Puri districts are in the medium risk
para, and south of Bhadrak. About 194 km of coastline has a category, with a coastal slope between 0.2u and 1.0u (Figure 4).
medium risk rating with erosion rates between 0 and 10.0 m/y
along the coastal stretches near Chilika Lake, north of Significant Wave Height
Kendrapara, and north of Bhadrak. About 231 km of coastline
that recorded accretion along the coastal stretches of Ganjam, The present study revealed that the mean significant wave
Jagatsinghpur, Bhadrak, Balasore, south of Puri and south of height ranges between 1.25 and 1.40 m. The entire coastline is
Kendrapara has a low risk rating (Figure 2). in the medium vulnerability class (Figure 5).

Sea-Level Change Rate Tidal Range

The present study revealed that about 292 km of coastline fell The present study revealed that about 37 km of coastline
has a high risk rating, recording historical sea-level change rates that has recorded tidal ranges more than 3.5 m along the

Journal of Coastal Research, Vol. 26, No. 3, 2010

530 Kumar et al.

Figure 5. Risk classes for significant wave height. Figure 6. Risk classes for tidal range.

coastal stretches of northern Balasore district has a high risk there are a number of intermittent extensions of sand spits
rating. About 302 km of coastline has a medium risk rating northward and repeated destruction of the same. The majority
with tidal range between 2.5 and 3.5 m along the coastal of the geomorphic classes along the Orissa coastline (367 km)
stretches of northern Puri, Jagatsingpur, Kendraparha, Bha- comprised sandy beaches, deltas, spits, mangroves, and
drak, and southern Balasore. About 141 km of coastline has a mudflats that have a high risk rating. About 74 km length of
low risk rating, recording tidal ranges of less than 2.5 m in coastline comprising estuaries and nonmangrove vegetated
Ganjam, Chilika Lake, and southern Puri (Figure 6). coasts have a medium risk rating. About 39 km of coastline
comprising inundated coasts and cliffs along the Chilika region
Coastal Regional Elevation has a low risk rating (Figure 8).

The present study revealed that about 207 km of coastline Tsunami Run-up
has a high risk rating, recording coastal regional elevation
between 0 and 3 m along the coastal stretches of Jagatsingh- The present study revealed that about 121 km of coastline
pur, Bhadrak, and northern Balasore. About 182 km of has a high risk rating, recording tsunami run-up of more than
coastline has a medium risk rating with coastal regional 2.0 m along the coastal stretches of Ganjam, Chilika and
elevation between 3.0 and 6.0 m along the coastal stretches of southern Puri. About 327 km of coastline has a medium risk
Kendraparha and southern Balasore. About 91 km of coastline rating with tsunami run-up between 1.0 to 2.0 m along most of
that has recorded coastal regional elevation of more than 6.0 m the coastal stretches of Jagatsinghpur, Kendraparha, Bha-
along Ganjam, Chilika Lake, Puri, and mid-Balasore has a low drak, and Balasore districts. About 31 km of coastline that has
risk rating (Figure 7). recorded tsunami run-up between 0 and 1.0 m was accorded a
low risk rating along the mid-Balasore coast (Figure 9).
Coastal Geomorphology
Coastal Vulnerability Index (CVI)
The Orissa coast forms a very wide arc with an overall
concavity toward the sea, maintaining a general trend of The coastal stretches of Orissa are classified as low, medium,
southwest to northeast. It is observed that at the river mouths, and high risk based on their vulnerability to the eight relative

Journal of Coastal Research, Vol. 26, No. 3, 2010

 Coastal Vulnerability Assessment 531

Figure 7. Risk classes for regional elevation. Figure 8. Risk classes for coastal geomorphology.

risk variables under study (Table 3). The resultant CVI is effective in that it highlights coastal areas where the various
calculated and the vulnerability zones along the coastal effects of sea-level rise may be the greatest. In addition to the
shoreline are delineated on the map (Figure 10). six variables used by earlier researchers, the present study
The CVI value along the study area of Orissa coastline varied uses two additional variables to represent vulnerability more
from 2.1 to 19. The 25th and 50th percentiles of CVI value are precisely; an additional geologic process variable, i.e., coastal
4.75 and 9.5, respectively. Those parts of the coastline having regional elevation and an additional physical process variable,
CVI values ranging from 2.1 to 4.75 are considered to be low i.e., tsunami run-up. The imperative for using these additional
vulnerable, those ranging from 4.75 to 9.5 are considered to be variables is discussed.
medium vulnerable, and the remaining parts having CVI In the earlier studies, tidal range was assumed to include
values of more than 9.5 are high vulnerable. Accordingly, about both permanent and episodic inundation hazards. However,
76 km of the coastal stretch of Orissa state, covering parts of the mean of the long-term tidal records tends to dampen the
Ganjam, Chilka, southern Puri, and Kendraparha, is low effect of episodic inundation hazards such as tsunamis. For this
vulnerable. About 297 km of the coastal stretch of Orissa state, reason, in the present study, tsunami run-up has been
covering northern Ganjam, Chilika, central Puri, Jagatsingh- considered as an additional physical process parameter to
pur, Kendraparha, southern Bhadrak, and northern Balasore, calculate the CVI. Similarly earlier studies used coastal slope
is medium vulnerable. About 107 km of the coastal stretch of as one of the parameters to calculate CVI with low coastal slope
Orissa state, covering northern Puri, parts of Jagatsinghpur, representing high risk and vice versa. Such an assumption does
Kendraparha, northern and southern Bhadrak and southern not always hold. For instance, areas with low coastal slope
Balasore, is high vulnerable (Figure 11). falling in areas of high coastal regional elevation are not as
vulnerable as similar areas falling in low coastal regional
DISCUSSION elevation. Such inconsistencies could be effectively addressed if
coastal regional elevation is also considered as an additional
The CVI presented in this study is similar to that used in parameter that represents the vertical level of the terrain. The
Pendleton, Thieler, and Jeffress (2005); Thieler (2000); and integration of these parameters makes the present study much
Thieler and Hammar-Klose (1999). This method is very more comprehensive.

Journal of Coastal Research, Vol. 26, No. 3, 2010

532 Kumar et al.

Figure 9. Risk classes for tsunami run-up. Figure 10. CVI classes along Orissa coast.

This study revealed that 22% of the Orissa coastline is in CONCLUSIONS

the high vulnerable category, 62% in the medium vulner-
able category, and 16% in the low vulnerable category. The present study conclusively proves the usefulness of
Results showed that coastal areas in the districts of Puri remote sensing data, in situ observations, numerical modeling,
and Jagatsinghpur are in the high vulnerable class. These and GIS analysis tools for coastal vulnerability studies. The
areas are known to be historically vulnerable to coastal coastal vulnerability maps produced using this technique serve
flooding especially from storm surges, thus validating the as a broad indicator of threats to people living in coastal zones.
credibility of this study. Many coastal areas in Ganjam, This is a objective methodology to characterize the risk
Chilka, and southern Puri fall in the low vulnerable associated with coastal hazards and can be effectively used by
category. coastal managers and administrators for better planning to
The vulnerability maps derived from this study depict mitigate the losses due to hazards as well as for prioritization of
vulnerable areas as per the eight parameters considered. areas for evacuation during disasters.
These maps are therefore not maps of total vulnerability but of
essential aspects constituting overall vulnerability. They Table 3. Risk classes for different parameters and resultant CVI.
depict the problematic regions, and therefore further attention
Length (km)
should be directed to these regions to analyze their vulnera-
bility in the context of nested scales and on higher resolution S. No Parameter Low Medium High
than the 1000 m 3 1000 m grid. Evolving technologies in 1 Shoreline change rate 231 194 55
remote sensing, GIS, and numerical modeling are making 2 Significant wave height 0 480 0
accurate data available at better spatial and temporal scales for 3 Sea-level change rate 23 166 292
4 Tidal range 141 302 37
all the considered variables. Use of such data sets might throw 5 Coastal regional elevation 91 182 207
better light on coastal vulnerability aspects at a much more 6 Coastal slope 10 41 429
local level. Use of additional parameters such as cyclone, storm 7 Tsunami run-up 31 327 121
surge, and coastal flooding will add an additional dimension to 8 Coastal geomorphology 39 74 367
9 CVI 76 297 107
the current study.

Journal of Coastal Research, Vol. 26, No. 3, 2010

 Coastal Vulnerability Assessment 533

Figure 11. Enlarged portions of the CVI classes in the area 1–5 shown in the Figure 10.

ACKNOWLEDGMENTS Ecosystem Services (GOES), Michigan State University,

for the Landsat data, Global Sea Level Observing System
The authors thank Dr. P.S. Goel, former secretary, (GLOSS) for the sea level data, and USGS for the making
Ministry of Earth Sciences for his encouragement. The available the Digital Shoreline Analysis Software (DSAS) on
authors would like to thank Global Observatory for their website.

Journal of Coastal Research, Vol. 26, No. 3, 2010

534 Kumar et al.

LITERATURE CITED Climate Change in the Mediterranean: Socio-economic Perspective

of Impacts, Vulnerability and Adaptation. Camberley, UK: Edward
Anderson, K.E.; Cahoon, D.R.; Guitierrez, B., and Thieler, E.R., 2005. Elgar Publishing, pp. 245–278.
The Physical Environment. Public Review Draft. Washington, DC: Mansinha, L. and Smylie, D.E., 1971. The displacement fields of
US Climate Change Science Program, Environmental Protection inclined faults. Bulletin of the Seismological Society of America, 61,
Agency, pp. 55–60. 1433–1440.
Ashok Kumar, K.; Raju, N.S.N., and Sanil Kumar, V., 2005. Wave Mimura, N. and Nicholls, R.J., 1998. Regional issues raised by sea-
Characteristics off Visakhapatnam Coast during a Cyclone. Annual level rise and their policy implications. Journal of Climate
Report. Goa, India: National Institute of Oceanography, Ocean Research, 11, 5–18.
Engineering Division. Michigan State University. 2008. Global Observatory for Ecosystem
Bankoff, G.; Frerks, G., and Hilhorst, D., 2003. Mapping Vulnerabil- Services. www.landsat.org (accessed July 2, 2008).
ity: Disasters, Development and People, chapter 3. London: Earth- Nicholls, R.J., 1995. Coastal megacities and climate change. Geojour-
scan Publishers. nal, 37(3), 369–379.
Belperio, T.; Bourman, B.; Bryan, B., and Harvey, N., 2001. Pendleton, E.A.; Thieler, E.R., and Jeffress, S.W., 2005. Coastal
Distributed process modeling for regional assessment of coastal Vulnerability Assessment of Golden Gate National Recreation Area
vulnerability to sea-level rise. Environmental Modeling and to Sea-Level Rise. USGS Open-File Report 2005-1058.
Assessment, 6(1), 57–65.
Pethick, J.S. and Crooks, S., 2000. Development of a coastal
Church, J.A. and Gregory, J.M., 2001. Climate Change 2001: Working
vulnerability index: a geomorphological perspective. Environmen-
Group I: The Scientific Basis, chapter 11. Location: International
tal Conservation, 27, 359–367.
Panel on Climate Change.
Pradeep Kumar, A. and Thakur, N.K., 2007. Role of bathymetry in
Cummins, P.R., 2007. The potential for giant tsunamigenic earth-
tsunami devastation along the east coast of India. Current Science,
quakes in the northern Bay of Bengal. Nature, 449, 75–78.
92(4), 432–434.
Dinesh Kumar, P.K., 2006. Potential vulnerability implications of sea
Rajawat, A.S.; Bhattacharya, S.; Jain, S.; Gupta, M.; Jayaprasad, P.;
level rise for the coastal zones of Cochin, southwest coast of India.
Tamilarasan, V.; Ajai, and Nayak, S., 2006. Coastal Vulnerability
Environmental Monitoring and Assessment, 123, 333–344.
Dominey-Howes, D. and Papathoma, M., 2003. Tsunami vulnerability Mapping for the Indian Coast. Second International Symposium on
assessment and its implication for coastal hazard analysis and ‘‘Geoinformation for Disaster Management’’ (Dona Paula, Goa,
disaster management planning, Gulf of Corinth, Greece. Natural India, International Society for Photogrammetry and Remote
Hazards and Earth System Sciences, 3, 733–747. Sensing).
Dotsenko, S.F., 2005. Run-up of a solitary tsunami wave on a sloping Thieler, E.R., 2000. National Assessment of Coastal Vulnerability to
coast. Journal of Physical Oceanography, 15(4), 211–219. Future Sea-Level Rise. USGS Fact Sheet, fs-076-100.
Gornitz, V., 1990. Vulnerability of the east coast, U.S.A. to future sea Thieler, E.R., and Hammar-Klose, E.S., 1999. National Assessment of
level rise. Journal of Coastal Research, 9, 201–237. Coastal Vulnerability to Sea-Level Rise, U.S. Atlantic Coast: U.S.
Hegde, A.V. and Reju, V.R., 2007. Development of coastal vulnera- Geological Survey Open-File Report 99-593, 1 sheet.
bility index for Mangalore coast, India. Journal of Coastal UNEP (United Nations Environment Programme), 2007. Physical
Research, 23, 1106–1111. Alteration and Destruction of Habitats. www.unep.org (accessed
IPCC (Intergovernmental Panel on Climate Change). 2001. IPCC February 27, 2008).
Report, Working Group-I, Climate Change–2001: The Scientific USGS (U.S. Geological Survey), 2001. National Assessment of Coastal
Basis. Cambridge, UK: Cambridge University Press. http://www. Vulnerability to Sea-Level Rise: Preliminary Results for the U.S.
ipcc.ch/ipccreports/tar/wg1/408.htm (accessed April 10, 2007). Gulf of Mexico Coast. http://pubs.usgs.gov/of/2000/of00-179/pages/
Klein, R. and Nicholls R., 1999. Assessment of coastal vulnerability to risk.html (accessed August 20, 2008).
climate change. Ambio, 28(2), 182–187. USGS, 2005. The Digital Shoreline Analysis System (DSAS) version
Klein, R.J.T.; Reese, S., and Sterr, H., 2000. Climate change and 3.0, an ArcGIS Extension for Calculating Histrionic Shoreline
coastal zones: an overview of the state-of-the-art on regional and Change, Open-File Report 2005-1304. http://woodshole.er.usgs.gov/
local vulnerability. In: Giupponi, C. and Shechther, M. (eds.), project-pages/DSAS/version3/ (accessed February 27, 2008).

Journal of Coastal Research, Vol. 26, No. 3, 2010

Copyright of Journal of Coastal Research is the property of Allen Press Publishing Services Inc. and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.