International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004
Nayak and Bahuguna, 2001, Blasco and Aizpuru, 2002). Principal
component analysis and band ratio are important techniques for
distinguishing mangroves (Nayak et al. 1985, Nayak, 1994, Green
et al. 1998). It was found that spectral resolution is more crucial
in distinguishing mangroves (Gao, 1999a, 1999b).
Mangroves are characterised by the presence of particular
dominant communities/species, physico-chemical environment,
extent and frequency of inundation by tidal waves, salinity and a
soil types. It is possible to identify major mangrove communities
such as Avicennia spp.; Rhizophora forest, Sonneratia spp.,
Phoenix spp. and mangrove scrub using LISS III and PAN data
on the Indian sub-continent (Fig. 2). The variation in pigment
content, leaf structure and water content in mangroves leaves
along with the associated features and location were exploited to
identify dominant mangrove types. Information regarding
different mangrove community zonation is a vital for bio-
diversity assessment and for preparing management plans for
conservation.
Avicennia alba
Aegialitis
c
Avicennia ep . x : x 3
marina Ug a e cpm c
Marsh
Phoenix
Fig. 2. Mangrove zonation in the part of Sunderbans, India usng
IRS LISS III and PAN merged data
‘3.3 Coral Reefs
The information about extent and condition of coral reefs is
required for planning conservation and preventive measures to
protect this fragile system. Coral reefs cover about 300,000 sq km
areas. IRS LISS III, LANDSAT TM and SPOT multi-spectral
data have shown potential in identifying various coral reef
features. Coral reef features such as type (fringing, atoll, platform,
patch, coral heads, sand cays, etc.), reef-flat, reef vegetation,
degraded reef, lagoons, live corals and coralline shelf have been
mapped using IRS LISS II and III data on 1:50,000 scale for the
Indian reefs (Nayak et al. 1996, Bahuguna and Nayak, 1998).
Uncharted coralline shelf, coral heads, live coral platform and
coral pinnacle were mapped. These maps have been used as a
basic input for identifying the boundaries of protected areas and
biosphere reserves. It was observed that coral reef degradation is
more serious problem than total destruction (Bahuguna and
Nayak, 1998). The felling of mangroves and clearing of forests
have increased sedimentation and affected live coral and species
diversity. It was observed that annual growth of coral reefs
decreases with increase in sediment loading in the Mayaquez bay
in Puerto Rico (Miller and Cruise, 1995).
Coral reefs show distinct morphological and ecological
characters. Water column modifies the apparent reflectance
spectrum of an object on the sea bottom. Water column correction
and contextual editing increased the classification accuracy
(Mumby et al. 1998). IRS-1C LISS III and PAN merged data
have been used for coral reef zonation study (Nayak et al. 1996).
It is possible to identify reef edge, reef crest, algal ridge, moat,
coral knolls, reef platform, reef flat (consisting of sea grass,
algae/seaweed, shallow pools, mud over reef and sand apron).
IRS LISS III green band (520-590 nm) is quite useful. The live
coral area as small as 50-sq. m was possible to identify. The use
of very high-resolution data such as IKONOS has further
improved the accuracy of classification as well as delineation ef
boundary (Fig. 3).
WR
Fig. 3. Coral reef zonation (Gulf of Kachchh, Western India),
using IKONOS data.