Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 116 (2015) 381 389

8th International Conference on Asian and Pacific Coasts (APAC 2015)

Department of Ocean Engineering, IIT Madras, India.

A study on the shoreline changes and Land-use/ land-cover along

the South Gujarat coastline
Misra, Aa and Balaji, Ra*
a

Department of Civil Engineering, IIT Bombay, Powai,Mumbai 400076 India

Abstract
Shoreline change dynamics along the coastal districts of South Gujarat is studied by the application of multi-temporal remote
sensing datasets. The coastal zone along the three districts of Surat, Navsari and Valsad in South Gujarat, India are reported to
be facing serious environmental challenges, especially due to anthropogenic impacts. This study evaluates the decadal changes
in historical shoreline changes, using satellite images of Landsat TM, ETM and OLI. The analysis of shoreline changes in these
districts revealed significant variations in the form of shoreline erosion. Further, the Land-use/ landcover(LULC) maps for
each district for the year 2014 is prepared using band ratios as a pre-classification step, followed by implementation of a
combination of supervised and unsupervised classification called hybrid classification. An accuracy assessment is carried out
for each of the datasets and the overall accuracy ranges between 90 - 95%. The LULC study is carried out in order to identify
the various classes in this region which are vulnerable to the environmental degradation. The maps prepared in this research
will contribute to both the development of sustainable land use planning as well as shoreline protection measures.
2015
2014The
The
Authors.
Published
by Elsevier
B.V.

Authors.
Published
by Elsevier
Ltd. This
is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of organizing committee of APAC 2015, Department of Ocean Engineering, IIT Madras.
Peer- Review under responsibility of organizing committee , IIT Madras , and International Steering Committee of APAC 2015

Keywords: Coastal Erosion; Shoreline Change; South Gujarat; Band Ratios; LULC

* Corresponding author. Tel: (+91)22 2576 7321 (off), Fax.: (+91)22 2576 7302.
E-mail address: rbalaji@iitb.ac.in

1877-7058 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer- Review under responsibility of organizing committee , IIT Madras , and International Steering Committee of APAC 2015

doi:10.1016/j.proeng.2015.08.311

382

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

1. Introduction
The shoreline or coastline is the interface between the land and water which gets altered due to the various
coastal processes that govern it such as, wave characteristics, near-shore circulation, sediment characteristics and
beach forms etc. Shoreline change is a result of a process called littoral transport, which is responsible for moving
eroded material in the coasts by means of waves and currents in the near shore zone. Apart from the natural
processes, human induced factors such as dams or reservoirs, dredging, mining, and water extraction etc (Coastal
Engineering Manual, 2002) can also be responsible for erosion or accretion. The shoreline has a tendency to
changes it configuration in an attempt to reach a state of equilibrium (Pandian et al., 2004) in terms of its sediment
budget which can be observed over a period of time. In the context of coastal vulnerability, accreting coastlines are
considered less vulnerable, as they result in the addition of land areas by moving towards the ocean. On the other
hand, eroding coastlines are considered highly vulnerable because of the resultant loss of natural as well as manmade resources associated with it. Thus, healthy beaches and shorelines are considered essential to the quality of
life along the coast, and also provide buffers for storms and critical habitats for many species of plants and animals.
Further, monitoring changes in the shoreline enables to identify the nature and processes that cause these changes
in a particular area, assess the anthropogenic impacts and accordingly suggest management measures.
The delineation of shoreline is a pertinent exercise from the point of view of coastal zone management,
watershed definition, and flood prediction. Traditional ground surveying techniques although are, relatively more
accurate, often prove to be time consuming and almost impossible for a large coastal belt (Cracknell, 1999).
Remotely sensed data on the other hand, can provide valuable information by virtue of their spatial and temporal
scales with reasonable accuracy. This can be analyzed in a geographic information system (GIS) environment by
measuring differences in past and present shoreline locations.
Several studies have been carried out to study the shoreline change and related subjects using geo-informatics as
a tool. Muttitanon and Tripathi (2005) used Landsat 5 TM data in order to find out land use/land cover changes in
coasts of Ban Don Bay, Thailand. Siddiqui and Maajid (2004) evaluated a multi-temporal principal component
analysis (PCA) on Landsat Multispectral Scanner (MSS) and TM data to evaluate coastal changes between 1973
and 1998 in Pakistan. Ghanavati et al., (2008) used Landsat Thematic Mapper (TM) and Enhanced Thematic
Mapper Plus (ETM+) data in order to monitor geomorphologic changes of Hendijan River Delta, southwestern
Iran. In the Indian Context, various researchers have carried out shoreline change studies using remote sensing
data. Mukhopadhyay et al., (2011) analyzed the coastal erosion and associated shoreline change in relation to sea
surface height anomaly in the Chandipur coast in Balasore district of Orissa using multi temporal satellite imagery
during the period 1990 to 2010. Kumar and Jayappa (2009) studied the accretion and erosion pattern along the
Mangalore coast. Choudhary et al., (2013) conducted shoreline detection from Karwar to Gokarna, western coast of
India. Gupta, (2014) carried out a study on Monitoring Shoreline Changes in the Gulf of Khambhat, India During
1966-2004 Using RESOURCESAT-1 LISS-III.
Further, in order to evaluate and predict the extent of geomorphic and ecological changes taking place, coastal
zones require an enormous amount of site specific research, by virtue of the complexities associated with them
even at the regional level (Manimurali and Dinesh Kumar, 2014). The existing land-use/land-cover (LULC) of a
region is a cumulative output of the interaction between natural as well as anthropogenic variables and processes.
Several studies have demonstrated the applicability of Remote Sensing (RS) and Geographical information system
(GIS) as efficient and cost effective tools for analyzing the spatial and temporal dynamics of LULC (Hathout,
2002; Herold et al., 2003; Lambin et al., 2003; Siddiqui and Maajid, 2004; Muttitanon and Tripathi, 2005; Santhiya
et al., 2010; Misra et al., 2013; Gilani et al., 2014) and other natural processes, wherein the former provides a
valuable source of multi-temporal data, and the latter is useful for mapping and assessing the associated patterns.
Thus, these tools provide a unique opportunity to develop information sources and support decision making
activities in a plethora of coastal zone applications.
With this perspective, the current study aims to fulfill two main objectives, firstly to study the shoreline changes
along the districts of South Gujarat using GIS as a tool to demarcate the eroding zones along this coastline.
Previous literature available (Gupta, 2014; Manek and Balaji, 2014; Mahapatra et al., 2013; INCOIS, 2009) for this
region report that shorelines adjoining these districts are particularly infamous for being erodible in nature, due to

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

both natural as well as anthropogenic reasons. Secondly, the LULC classes in this region for the year 2014 are
quantified by using a combination of pre-classification (band ratio) and post classification (hybrid classification)
techniques. This is mainly done to identify the prevailing LULC classes in this area and to understand the
vulnerability of this region to changes. The overall objective of this study is to produce shoreline change and LULC
maps that can be used efficiently by coastal managers to devise an effective coastal zone management plan for this
region.
2. Study Area
The Gulf of Cambay, also called the gulf Khambat is a trumpet shaped gulf in the Arabian Sea that is situated
between the Saurashtra peninsula and mainland Gujarat. Located approximately between latitude 20 30 and 22 20,
71 45 and 72 53 E, it covers an extent of about 3120km2. The GoC is about 70km wide and 130 Km long (Vora et.
al., 1980) with a mid-depth of 30 m. Many rivers drain into the gulf, including the Sabarmati, Mahi, Narmada,
Tapti and Shetrunji and is characterized by a number of estuaries, islands, mudflats, cliffs and mangroves. The gulf
is dominated by strong tidal currents (semi-diurnal types with high tidal range) which are responsible for most of
the depositional and erosional features of the gulf.

Fig.1. Study Area

The study area (Fig.1) includes the three coastal districts of Surat, Navsari and Valsad, located in Southern
Gujarat along the Gulf of Khambat. The district of Surat with a total population of 6,079,231 (Chandramauli, 2011)
covers an area of 7657 sq. km and is bordered by the districts of Bharuch and Narmada in the North, Navsari and
Dangs in the South and the Gulf of Khambat in the West. The town of Hazira is located in this district, which is an
important transshipment port and is popularly known as the 'industrial hub' of India due to the presence of major
industrial facilities. The district of Navsari covers an area of 2211 sq.km with a population of 1329672 and is
bordered by the districts of Surat in the North and Dangs in the East. The district of Valsad is surrounded by the

383

384

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

district of Navsari in the North, Dang in the East, and Maharashtra in the South. The population of Valsad is
1705678 persons for an area of 3034 sq.km (Chandramauli, 2011).

3. Data and Methodology

3.1 Shoreline Change Analysis
The LandSat archive data for different years was downloaded from the websitehttp://earthexplorer.usgs.gov/.The important factors considered for finalizing the satellite images were cloud cover,
similar tide conditions, similar season data, uniform projection factors, etc. The satellite data further underwent
radiometric correction and data scaling to enable maximum visual interpretation. Landsat TM (19 July 1990, TE3.33m); Landsat ETM (11 January 2001, TE- 2.98m) and Landsat OLI_TRS (12 April 2014, TE- 3.15m) datasets
with UTM (Universal Transverse of Mercator) zone 43 North projection system were used for this study region.
3.1.1 Digitization of Shoreline
The shoreline features were identified using the tonal differences between the land and the sea. A band ratio
technique was applied to differentiate the land and water pixels. Further vectorization technique was applied to get
the shoreline features in Arc-GIS environment. To further refine the results, visual interpretation was carried out
for editing the shoreline features to conform to the High Tide Line (HTL).
3.1.2 Study of Shoreline Change
The Digital Shoreline Analysis System (DSAS) is a freely available software application that works within the
Environmental Systems Research Institute (ESRI) Geographic Information System (ArcGIS) software. The
digitized shoreline for the years 1990, 2001 and 2014 in the vector format (. shp) are used as the input to the Digital
Shoreline Analysis System (DSAS) to calculate the rate of shoreline change. The analysis also requires transect
information and hence, transects (in the form of Shape file) are laid at every 500m interval along the shoreline. The
DSAS tool basically estimates the Net Shoreline Movement (NSM) and End Point Rate (EPR) which are used to
derive the output maps of this study. The NSM calculates the distance between the oldest and the youngest
shoreline for each transect and the EPR is obtained by dividing the NSM, by the number of years elapsed between
the two shoreline positions. The linear extents with negative NSM or EPR values indicate erosion whereas those
with positive values indicate accretion.
3.2 Land-use/land-cover Mapping
In this study, the various LULC classes for each district are extracted from the 2014 Landsat imagery by using
band ratio as the image enhancement technique. For this study, different band ratios were made for the study area
by using ENVI 4.5 software. Band ratio using bands 3,4,5,7 were used to derive layers of ratios 3/4, 4/3 and 5/7
which were subsequently used to derive color composite combinations to discriminate different LULC features.
The final color composite is created using 5/7 in red, 4/3 in green and 3/2 in blue which efficiently distinguishes the
various features of the heterogeneous landscape of the study region. Further, the transformed images are used as
image interpretation aids to classify the images of 2014. For this area, a methodology combining both supervised
and unsupervised classification called hybrid classification is adopted. This approach encompasses three main steps
(1) an initial spectral clustering, (2) assignment of clusters to user-defined classes and (3) maximum likelihood (or
similar decision rule) classification of the entire image (Misra et al. 2013). Here, firstly we have applied
unsupervised classification to the images by using the ISODATA clustering algorithm. In the second step,
signatures of classes which overlap or get mixed with other classes are collected and assigned signature file, a
result of the supervised approach. Finally, all signatures either are appended together to classify (as a supervised
classification step) the complete and full extent of the originals image using maximum likelihood algorithm.

385

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

Accuracy assessment is then performed by a collection of 256 points from Google Earth Pro, Ground truth data and
ancillary data and the overall accuracy as well as the Kappa coefficient is reported in Table 1.

Table 1. Overall classification accuracy and Kappa coefficient for all datasets

Study Region

Year of Dataset

Overall Accuracy of Classification

Kappa Coefficient

Surat

Landsat OLI 2014

92.54%

0.92

Navsari

Landsat OLI 2014

94.56%

0.94

Valsad

Landsat OLI 2014

91.82%

0.91

4.1 Shoreline Change

The shoreline change was estimated for the districts of Surat, Navsari and Valsad using the DSAS tool
embedded in ArcGIS 10.1. The extents of the shoreline accreting or eroding for the periods 1990- 2001, 2001-2014
and 1990-2014 are shown in Fig.2. Of the entire length of 117km, about 65% of the shoreline is observed to be
eroding over the period of 24 years. Onjal, Bhat, Danti and Umbergaon as well as the extents from Umbhrat to
Vansi and Fansa to Maroli shows considerable erosional activity. Strong tidal currents are responsible for the
accretion observed along the North Hazira Port. The shoreline changes from 1991 to 2001 show extreme rate of
erosion especially along Umbhrat which has a negative NSM of 234m.

386

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

Fig.2. Shoreline changes from (a) 1990-2001 (b) 2001- 2014 (c) 1990-2014

From 2001- 2014 accretion with an average EPR of 0.65m/yr is witnessed. Overall, from 1990-2014, the
shoreline analysis of this coastline (117km), reveals an eroding trend with an average EPR of -0.54m/yr and
average NSM of -12.4m. The maximum accretion during this period is seen Umarsadi to Udwada where an erosion
of 256.28m is seem from Umbhrat to Vansi. The cumulative action of waves and tides are responsible for the
erosion/accretion trends observed along this coastline. Moreover, coastal protection is difficult as the coastal belt
along GOC is mainly composed of clay sand or silt. As stated before, anthropogenic impacts are immense in these
regions that greatly impact the coastal processes. A large number of sea walls, groynes, breakwaters and jetties
alter the shoreline by disturbing the sediment dynamics of the region (Gupta 2014; Mahapatra et al. 2014). Thus, in
totality the erosional activity of this region is caused by the combining effect of tides, waves and sediment
transport.
The study highlights the advantage of using GIS tools for shoreline studies. The DSAS methodology used here
provides very good qualitative estimates to understand the dynamics in a given area. The major drawback however
is that the tool considers only the initial and final years for the study. Also, it takes into account a transect based
analysis wherein certain important characteristics of the shoreline could be lost in places where there are no
transects available. Nevertheless, the method is important from the perspective of environmental monitoring and
hence can be used extensively for preliminary studies and understanding.

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

4.2 Mapping of land use land cover for Surat, Valsad and Navsari
The study domain of each district was extracted using a buffer zone of 20, 18, 35 km for Surat, Navsari and
Valsad respectively. The domain sizes were chosen so that they cover the immediate coastal zone and urban
development in the vicinity. The predominant classes for this region are water, intertidal mudflats, salt marshes,
mangroves, sandy features, agricultural land, urban built up, aquaculture, vegetation and barren. Fig.3and Table.2
shows the LULC maps for each district and their corresponding percentage distribution.

Fig.3. 2014 LULC maps for Surat, Navsari and Valsad

387

388

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

Table.2. Percentage (%) coverage of each LULC class

LULC Class

Percentage(%) coverage of each LULC class

2014
Surat

Navsari

Valsad

Water

38.91

50.91

65.81

Intertidal Mudflats/Tidalflats

3.33

3.45

0.91

Salt Marshes/Salt Pans

3.02

8.14

0.86

Marshy vegetation/ Mangroves

0.84

0.89

0.14

Shoreline/ Sandy Features

0.08

0.46

0.17

Agricultural/ Fallow Land

37.87

26.25

23.02

Urban Built Up

7.56

2.23

2.47

Aquaculture

1.42

1.27

0.13

Barren

1.69

6.39

6.17

Vegetation

5.29

0.00

0.34

For Surat, agricultural land covers the maximum spatial extent of almost 74425.59 hectares. A total of 141370.7
hectares is covered by Intertidal mudflats/tidal flats, Salt marshes and mangroves. Aquaculture constitutes of about
2791.53 hectares. Only 148.86 hectares is covered by the sandy-features class. Heavy urbanization can be seen as
the Urban built up class covers almost 14862.51hectares of the region under consideration. In case of Navsari,
910.3hectares is covered by mangroves. A considerable spatial coverage of Salt marshes can be seen of about
8292.1hectares. For the area studied for Navsari most of the region is covered with agricultural land. In case of
Valsad it can be seen approx.23% of the area is covered by agricultural land class. The mangrove spatial extent is
relatively less, about 428.76m. The shoreline as represented by the Sandy features class is well represented for this
district covering and area of about 655.29 hectares. The Urban Built Up class covers an area of 9520.38 hectares.
Thus, it is evident that the three districts encompass some extremely fragile ecosystems like mangroves, mudflats,
salt marshes etc. According, to coastal vulnerability studies reported in literature (Manimurali et al., 2013), such
features are generally considered to highly vulnerable to coastal hazards. Hence, these regions require immense
protection from degradation.

5. Conclusion
This study is an effort to understand the dynamics of shoreline changes occurring along the South Gujarat coast
and evaluating the cause of the same. The End Point Rate of -0.54m/yr highlights the eroding trend of the shoreline
which is of immense concern especially from the sustainability point. For the South Gujarat districts, it can be seen
that the urbanization leads to immense pressure on the nearby coastal resources. Further, through the LULC maps
its can be witnessed that these district cover fragile ecosystems of mudflats, salt marshes etc and hence future
environmental degradation along these coast could disrupt the normal functioning of the environment. Therefore, it
is suggested that improved tools of remote sensing (high resolution data) and GIS accompanied with field survey
and numerical modeling can enable better understanding of these regions so that a sound strategy can be devised
for the conservation and restorations of these natural systems. Finally, the present research pertaining to the study
of the shoreline changes and mapping of LULC will enable the decision makers to identify the susceptible zones
and find better solutions to the existing coast problems in these locations.

A. Misra and R. Balaji / Procedia Engineering 116 (2015) 381 389

6. References
Chandramauli, C., 2011. Census of India 2011: provisional population totals paper 1 of 2011 India Series 1,
Chapter 6. New Delhi, India: Office of the Registrar General & Census Commissioner.
Choudhary, R., Gowthaman, R., & SanilKumar, V., 2013. Shoreline change detection from Karwar to GokarnaSouth West coast of India using remotely Sensed data. International Journal of Earth Sciences. 6(3), 489-494.
Cracknell, A. P., 1999. Remote sensing techniques in estuaries and coastal zones an update. International Journal
of Remote Sensing, 20(3), 485-496.
Engineers, U. A. C. O. (2002). Coastal engineering manual. Engineer Manual, 1110, 2-1100.
Ghanavati, E., Firouzabadi, P. Z., Jangi, A. A., & Khosravi, S.,2008. Monitoring geomorphologic changes using
Landsat TM and ETM+ data in the Hendijan River delta, southwest Iran. International Journal of Remote
Sensing,29(4), 945-959.
Gilani, H., Shrestha, H. L., Murthy, M. S. R., Phuntso, P., Pradhan, S., Bajracharya, B., & Shrestha, B. (2015).
Decadal land cover change dynamics in Bhutan. Journal of environmental management, 148, 91-100.
Gupta, M., 2014.Monitoring Shoreline Changes in the Gulf of Khambhat, India During 1966-2004 Using
RESOURCESAT-1 LISS-III. Open Journal of Remote Sensing and Positioning, 1(1), 27-37.
Hathout, S., 2002. The use of GIS for monitoring and predicting urban growth in East and West St Paul, Winnipeg,
Manitoba, Canada. Journal of Environmental management, 66(3), 229-238.
Herold, M., Goldstein, N. C., & Clarke, K. C.,2003. The spatiotemporal form of urban growth: measurement,
analysis and modeling. Remote sensing of Environment, 86(3), 286-302.
Integrated Coastal and Marine Area Management(ICMAM), Project Directorate, Chennai. (2002). Report on
Critical habitat Information System for the Gulf of Khambat- Gujarat. http://www.icmam.gov.in/GUK.PDF
Kumar, A., & Jayappa, K. S., 2009. Long and short-term shoreline changes along Mangalore coast,
India. International Journal of Environmental Research,3(2), 177-188.
Lambin, E. F., Geist, H. J., & Lepers, E.,2003. Dynamics of land-use and land-cover change in tropical
regions. Annual review of environment and resources, 28(1), 205-241.
Mahapatra, M., Ratheesh, R., & Rajawat, A. S., 2013. Shoreline change monitoring along the south Gujarat coast
using remote sensing and gis techniques. International Journal of Geology, Earth and Environmental science,
3(2), 115-120.
Mahapatra, M., Ratheesh, R., & Rajawat, A. S., 2014. Shoreline Change Analysis along the Coast of South
Gujarat, India, Using Digital Shoreline Analysis System. Journal of the Indian Society of Remote Sensing,
42(4), 869-876.
Manek, N.P., & Balaji, R., 2014.Assessment of shoreline oscillations along south Gujarat coastline, India. Indian
Journal of Marine Sciences, 43(7). (online)
ManiMurali, R., Ankita, M., Amrita, S., & Vethamony, P., 2013. Coastal vulnerability assessment of Puducherry
coast, India, using the analytical hierarchical process. Natural Hazards and Earth System Sciences, 13(12),
3291-3311.
Misra, A., & Vethamony, P., 2013. Assessment of the land use/land cover (LU/LC) and mangrove changes along
the MandoviZuari estuarine complex of Goa, India. Arabian Journal of Geosciences, 8(1), 267-279.
Mukhopadhyay, A., Mukherjee, S., Hazra, S. and Mitra, D., 2011. Sea Level Rise And Shoreline Changes: A
Geoinformatic appraisal Of Chandipur Coast, Orissa. International Journal of Geology, Earth and
Environmental Sciences, Vol. 1 (1), 9-17.
Murali, R. M., & Kumar, P. D. (2015). Implications of sea level rise scenarios on land use/land cover classes of the
coastal zones of Cochin, India. Journal of environmental management, 148, 124-133.
Muttitanon, W., & Tripathi, N. K., 2005. Land use/land cover changes in the coastal zone of Ban Don Bay,
Thailand using Landsat 5 TM data. International Journal of Remote Sensing, 26(11), 2311-2323.
Pandian, P. K., Ramesh, S., Murthy, M. V. R., Ramachandran, S., & Thayumanavan, S., 2004. Shoreline changes
and near shore processes along Ennore coast, East coast of south India. Journal of Coastal Research, 828-84.
Santhiya, G., Lakshumanan, C., & Muthukumar, S.,2010. Mapping of Landuse/Landcover Changes of Chennai
Coast and Issues related to Coastal Environment Using Remote Sensing and GIS. International Journal of
Geomatics and Geosciences, 1(3), 563-576.
Siddiqui, M. N., & Maajid, S., 2004. Monitoring of geomorphological changes for planning reclamation work in
coastal area of Karachi, Pakistan. Advances in Space Research, 33(7), 1200-1205.
Vora, K.H., Gujar, A.R., and Karisiddaiah, S.M., 1980. Sandwaves of the Gulf Of Khambhat, Indian Journal of
Marine Science, 9, 90-93.

389