Coastal wetland and shoreline change mapping of Pichavaram, south east coast of India using Satellite data
Hong Yeon CHO, Lakshumanan.C
Korea Ocean Research and Development Institute
Ansan, P.O Box29, Seoul 425-600. Korea
Centre for Environmental Studies
Chennai- 600 025, India
Most of the human settlements along the Indian coast are located along the estuaries and deltas. In India, mangrove forests are traditionally been used for a variety of purposes like, boat-building, tannin extraction, firewood, stakes for fishing, fodder, fertilizer etc. Anthropogenic activities eliminate the protection provided by mangroves to the coastal and the inland ecosystems, further disturbing wildlife habitat and biodiversity. In south East Asia, there is a severe drive for the conversion of mangrove lands for agricultural and industrial purposes, aquaculture. The factors that severely affect mangrove ecosystems are 1) diminishing fresh water inflow, 2) increasing salinity and, 3) nutrient supply (MOEF, 1987). Prawn culture in the mangroves of Chorao Island (Goa), Chilka lagoon (Orissa) and Pichavaram (Tamil Nadu) is of great concern to different environmental groups in India. In general, the Indian mangroves are considered as degraded (Krishnamoorthy, 1995).
Importace of Wetlands in Ecological Functions
Coastal and inland wetlands play an important role in the ecological functions of wetlands in various ways. They are detailed as follows:
Mangroves distribution in India occupies an area of about 4250 km2 (MOEF, 1994). Due to general intertidal slope and heavy siltation, the delta environment of east coast supports extensive mangroves. The largest stretch of mangroves in the country is located at Sundarbans, which covers an area of about 2119 km2. Next to Sundarbans, the Andaman & Nicobar island mangroves are the second largest, which covers and area of about 966 km2. Small patches of mangroves are found in Gujarat, Maharastra and some other coastal states (MOEF, 1987).
In Tamil Nadu, mangroves are well-developed in Pichavaram and Muthupet. The Pichavaram mangrove is a typical swamp, extending between Vellar and Coleroon estuaries. In Pichavaram mangroves sixteen Angiosperm species were recorded, fourteen of them are exclusively mangrove species (Krishnamurthy et al., 1981)
Six zones have been demarcated in Pichavaram; Aricenniamarina is dominant in Zone-I; Arthrocnemum indium, Exoecaria agallocha, Salicornia brachiata, Sesuvium portulacastrum and Suaeda maritima are sporadically distributed in the sandy region Zone II includes the bank of three creeks lying parallel to the shore the fringe of shoreward belt is occupied by Salicornia brachiata and the inward belt has Avicennia apiculata, Rhizophora apiculata and R.Stylosa. The inner region is found to have mixed community of Excocaria agallocha and Salicornia brachiata and Arthrocnemum indium occurs in this zone. Luxuriant mangrove vegetation exists in zone III with maximum number of species. The channels fringes areas are bordered by Rhizophora apiculata and R. Muconata, Acanthus ilicifolins and Derris Heterophylla found in the zone IV and A continuous stretch of Suaeda maritima is observed in zone V. Salicornia brachiata is dominant exists in zone VI near Coleroon estuary (Krishnamurthy, 1995)
Pichavaram is located 51 km north east of Chidambaram, Cuddalore district, Tamil Nadu, between latitude 11°20' to 11°30' north and longitudes 79°45' to 79°55' east (Fig.1). It is an estuarine type of mangrove situated at the confluence of Uppanar, a tributary of the Coleroon River. Fishing villages, croplands, and Aquaculture ponds surrounds the area. This mangrove environment is attracting large number of tourists.
The Pichavaram mangrove wetland has 51 islets and the total area of the Vellar-Pichavaram-Coleroon estuarine complex is 2335.5 ha of which only 241 ha. is occupied by dense mangrove vegetation. Nearly 593 ha, of this wetland is occupied by helophytic vegetation like Suaeda, 262.5 ha. barren mud flats and 1238.5 ha. Barren high saline soil (Krishnamoorthy et al., 1994) out of the 2335.5 ha of this mangrove wetland only 1100 ha. Comprising the entire mangrove vegetation located in the middle portion of the Vellar-Pichavaram-Coleroon wetland has been declared as a reserved forest. Department of Forest, Government of Tamil Nadu, declared the Pichavaram mangrove as a Reserved Forest.
Fig.1. Study Area Location Map
Two major rivers namely Vellar River and Coleroon River in this area drained into Bay of Bengal. The area between the two rivers is identified as brackish water with mangrove vegetation. The area is covered by alluvium in the western part and fluvial marine and beach sands cover eastern part of the area. Geomorphology of area is major area is covered by flood plain, sedimentary plain and beach sand. Major portion of the area falls under nearly level sloping category. During the nine year period from 1990 to 1999 the maximum rainfall of 1844 mm was recorded in 1990 and minimum rainfall of 875 mm was recorded in 1996. The soil group of the area is Hydrological soil group `C' (USDA) low infiltration and moderate runoff potential found 50% area. The soil group `B' with moderate runoff is covering about 45% areas. The remaining 5% area is occupied by soil group `A' with high infiltration low runoff potential. Hence, this study envisages to understanding the overall changes in the wetland ecosystem of Pichavaram mangroves.
Satellites and Sensors
Earth Resources Technology satellite ERTS-1 launched by NASA later renamed as Landsat-1 for mapping and monitoring of earth resources. Many other countries France launched SPOT; India lunched Indian Remote Sensing Satellite Series (IRS). NASA have been launched five satellites in the series of Landsat, first three and first generation satellites and carried Return Beam Vedicon (RBV) and Multispectral Scanner (MSS) imaging sensors and the second generation satellites carry advanced imaging sensor called Thematic Mapper (TM). Two satellites in the series of SPOT launched by France provide data in MSS and panchromatic spectral band with normal and stereoscopic mode. Indian Remote Sensing satellites series IRS-1A, IRS-1B, IRS-1C and IRS-1D the indigenously developed satellites have been put into orbit 1988, 1991, 1996, 1997 respectively and IRS-1A & 1B carried LISS I and LISS II multispectral sensors, IRS-1C & 1D contains LISS III, with multispectral scanner, panchromatic stereo viewing capability sensor provide good quality data.
Image Interpretation and Analysis
The satellite data products are available to the scientists in form of photographic products such as films, diapositives and paper prints of geo-coded in various scales or digital data stored in the form of Computer Compatible Tape (CCT) cartridges, floppies and C.D ROMs. The desired information can be extracted from the above data products through visual interpretation and digital image processing techniques. Both visual and digital image analysis techniques are complementary to each other. For large areas and spectrally homogenous scenes digital image processing techniques may provide a quick and cost effective means of image analysis. Smaller areas and spectrally heterogeneous scenes visual interpretation method is more suitable. However in order to bring out the subtle variation and quantitative measurements it is essential to perform the digital image processing techniques since classification of digital data can be done through computers with help of image processing software.
Visual Interpretation Techniques
Visual interpretation of remote sensing images for extracting desired information could be achieved in an efficient and effective manner by using several basic interpretation keys (or) elements (Floyd F. Sabins Jr. 1987). The basic interpretation keys are i) Tone ii) texture iii) pattern iv) shape v) size and vi) location or association. All these interpretation elements are qualitative attributes and they are subjective depends on the experience and personal bias of an interpreter.
Digital Image Processing
Remotely sensed data compounds to different earth features collected by the satellite sensors and stored in computer compatible tape, cartridge CD-ROM, floppy in regular line and columns. The pixel represents brightness value having a specific Digital Number (DN) value depends on the reflected energy from the earth surface in a specific wavelength or band or channel. Therefore, each one of the earth features is sensed by the sensors simultaneously and provides a set of DN values. The DN values of each pixel are ranging from zero for black to some higher value (255) for white, based on the radiometric resolution. The availability of remote sensed data in digital form helps in carrying out digital image processing with aid of ERDAS-imagine image processing software. The digital image processing techniques provide flexibility in data handling due to the fact that the digital data can be numerically manipulated by using an equation (or) set of equations to get the desired details in the graphic display (or) pictorial form for further analysis (Lillesand TM and Kiefer RW, 1987)
There are many procedures/methods available for image data manipulation they can be broadly grouped into 3 categories viz. 1) Image rectification and restoration also called preprocessing 2) Image enhancement and 3) Image classification.
Image rectification and restoration
Preprocessing operations are intended to eliminate or correct the distortions or errors caused due to geometric distortions, radiometric distortion presence of noise in the data, etc. This standard products made available to the interpreters are preprocessed therefore generally data can be used directly for image enhancement and classification.
Operations are being implemented to image data to get the enhanced output for subsequent visual interpretations. The enhancement techniques provide better feature exhibition to increase the visual distinction between features contained in a scene. The enhanced output can be seen in the display or can be recorded in the pictorial form as Black and White (or) colour composite images.
Image classification operations are essentially meant to substitute visual analysis of remotely sensed data with quantitative analysis. The classification of the remotely sensed satellite digital data can be carried out either without a prior knowledge about the features present in the scene is called unsupervised classification and with a prior knowledge about the terrain features interactive classification method is called supervised classification i.e. The user can define the class beforehand the supervised classification is the setting of representative pixel values for each class and classifying another pixel values based on this standard set.
In this present study preprocessed satellite (LANDSAT, IRS) were used for Wetland mapping and also supervised classification method were performed using ERDAS Imagine image processing software package.
Wetland map was classified using available satellite data IRS1C LISS –III (23.5 m spatial resolution) in the year 1998 for the Pichavaram Mangrove areas. The area can be divided into three zones Agriculture zone and Wetland zone with vegetation and Non-Vegetated wetland.
Agriculture land is present in the northern part of the Pichavaram area, this delta region formed by rapid deposition of stream borne sediments into a still body of water. The river supplies sand silt and other detrital materials which are deposited. This is very fertile area present in the Cauvery delta region. Flood plains are predominant along the rivers which are linear and parallel to the river course. The standing crop of paddy is almost seen through out the year along the Vellar river course an area of about 116.7 km2.
Non vegetated wetland consists of mudflats beach spit and sandbars along the lagoon estuaries and deltaic regions. They are divided into high tidal, inter tidal and sub tidal regions (Davis, 1972). However, at many places, they could not be further categorized. High tidal flat normally have very fine mud with little or no moisture. Large patches of mudflats occur between Uppanar and Coleroon estuaries. It is not possible to delineate inter tidal zone at all the places for the length of 1,486 km2. The beach is located the eastern part of the area near Vellar mouth, the beach area is narrow and at some places sand is replaced by mud. Beach area under erosion in many places spits are observed near Coleroon and Uppanar mouth area one end of the spit is attached to the mainland and other end terminates into the open sea. Mudflats are also in between mangrove vegetation in the hightidal, intertidal and subtidal zone. Mudflats are associated with less wave energy zones. These mud flats are composed of clay and silt and they are exposed during very low tides. Intertidal mudflats are made up of fine grained soft muds which are deposited under quiet environment. The sub-strata are suitable for the growth of mangroves and marsh vegetation. A variety flora and fauna are present in these areas. Most of sand bars are found near the mouth of rivers and therefore called mouth bars. These mouth bars are small in size in the areas of Vellar, Coleroon and Uppanar estuaries. Sand bars are also present in Uppanar backwater areas and fringed with sparse Mangroves.
The vegetated wetland has been classified in two types such as Mangroves and tidal swamps in Pichavaram area coastal lagoon covers an area of about 8.4 km2. The entire mangroves area consists of 51 islets which are separated by complex network of creeks. A long sand bar separates the whole area from the sea. Extensive mangrove forest, containing narrow Rizophora belt in the frequently inundated areas is followed by wide belt Avicennia on terrains submerged mostly equinoctial tides. The area that is generally very saline and sandy. Which is submerged only during exceptionally high tides contains mostly Suaeda vegetation. The Pichavaram mangroves area (7.29km2) is declared as reserved forest by state government.
The current level of exploitation of Pichavaram mangroves far exceeds sustainable levels, and the habitat is rapidly being degraded. Dense scrub like Avicenia Ilicifolium and Cariops Roxburghianae are mixed with mangroves scrubs are found near Uppanar riverbank and Vellar River, other vegetation such as marsh vegetations, grass, thorny plants and scattered mangroves were classified under this category due to their scattered distribution, it is difficult to separately delineate them, and they are collectively classified as other vegetation with mudflat. Sparse stunted marsh vegetation and degraded mangroves are present on hightidal mudflats near mouth of Vellar river, bank of Uppanar and Near Coleroon riverbank.
Salt deposits on mudflats areas mainly along saltpans in Porto Novo near Vellar river saline areas. The marsh vegetation present on the bar which is separating the lagoon from the sea is dense and water logged areas are also present in between. Flood prone areas were delineated near Coleroon river Vellar and the near banks of Uppanar and Coleroon rivers. Lagoons are observed in this area lying parallel to the coastline and separated from the open sea by barrier of islands. These lagoons are connected to the Bay of Bengal by one or more openings. The sub-rivers like Uppanar, Vellar and Coleroon rivers drain into this lagoon. The water level is fluctuating and suspended sediment concentration is high,
Fig. 2.Wetland Map
mangrove vegetation is present along the verge of lagoon. Classified output checked with sufficient ground truth verification, hence relevant ground data were collected for verification and field checks were made for doubtful areas necessary corrections were made in the classified output.
Coastal Landuse and Shoreline changes
Form 1970 survey of India toposheet and 1987 Landsat TM and 1998 IRS 1C Liss-III satellite data were compared qualitatively and quantitatively bring out the Landuse changes. Various costal Landuse categories were identified viz Mangrove, Aquaculture land, fallow land, degraded mangrove areas, sandy area, forest plantation, mud flat and waterlogged areas. The area under agriculture was around 116.53 km2 in 1970is decreased to 112 km2 in 1987 and 108 km2 in 1998. Mangrove areas in the year 1970 were about 4.9 km2 and reduced to 4.62 km2 in 1987 and to 3.70 km2 in 1998. This indicates various developmental activities. In 1987 3.99 km2 lands was aquaculture activities and in 1998 it increased to 6.99 km2. The fallow land areas found in 1970 was about 8.9 km2 in 1987 was about 7.45 km2 and in 1998 this area was nearly 8.45 km2. Mudflat areas covered about 6.0 km2 in 1970; in 1987 it was nearly 6.45 km2 and in 1998 it was 4.26 km2. Sandy area in 1970 was about 15.67 km2 in 1987 it was around 16.40 km2 and in 1998 it decreased to 13.72 km2. Degraded mangrove area in 1970 was very less, around 0.9 km2. In 1987 the area under degradation was nearly 1.20 km2. Further increase in degraded mangrove to 1.55 km2 was observed in 1998. Forest plantation area was identified 1.2 km2 in 1970 in 1987 it increased to about 1.25 km2 and in 1998 it was around 1.80km2. The waterlogged area in 1998 it was around 4.90 km2 along the fringes of Pichavaram lagoon restoration activities are going on. Pichavaram wetlands are surrounded by green canopy marsh vegetation; scrubs were identified using temporal remotely sensed data. Human dependency on biomass for fuel wood, timber, fodder, fiber and so on has inflicted major damage on plant diversity over the past 20 years. Human activities have resulted in a net loss of approximately 1.6 million km2 in wetlands (IGBP, 1993). Between 1966 and 1990 about 40 percent of mangrove areas decreased (MOEF 1994). Therefore identification of alternative fuel wood source, and generation of new biomass resources such as energy plantation on common lands, becomes very essential.
Shoreline profiles were observed using different year’s satellite data and Survey of India toposheet (58M/15). Depositional and erosional areas were identified from Vellar mouth to Coleroon, from Porto-Novo through Pichavaram, Killai the coastline near MGR Tittu and Chinnavaikkal to Kodigampalayam are erosional in nature. At Mudisodi and near the mouth of Vellar river depositional activity is observed.