Survey and Mapping of Mangrove Cover Using Remote Sensing – A Case...

Survey and Mapping of Mangrove Cover Using Remote Sensing – A Case Study of Sundarbans

SHARE

Dr Alok Saxena
Joint Director
Forest Survey of India
India
Email: [email protected]

Dr J K Rawat, Mr S K Singh

Abstract
Mangrove forests, unique ecosystems existing in the inter-tidal zones of coastal environment are undergoing constant changes because of their dynamic nature and to a greater degree due to various natural and biotic influences. Hence an accurate and upto date information regarding their extent and distribution is an essential pre-requisite for their sustainable management. Remote sensing coupled with geographical information system serves as an ideal tool in discerning these spatial changes over time. Forest Survey of India, an organization under Ministry of Environment & Forests, Government of India, is assessing and mapping mangroves using satellite data on a two-year cycle since 1987. As per its latest assessment (2001), an area of 4482 km2. is under mangrove cover.

In the present study, the spatial changes that had taken place between the period 1998 and 2002 in mangrove cover in Sunderban Biosphere Reserve of West Bengal is studied in details using IRS IC/ID LISS III data. A combination of digital image processing techniques was employed for enhancement and classification process. It is observed from the study that middle infra red band (0.77mm – 0.86mm) is highly suitable for mapping mangroves. Two major classes of mangroves, dense and open mangroves were delineated from the digital data. The study reveals a net increase of 116 km2. in mangrove cover. This includes an increase of 449 km2. in dense mangrove cover and a decrease of 333 km2. in open mangrove cover. The area harbours rich mangrove bio-diversity and certain species exhibit a definite pattern of distribution. The classification techniques carried out could discern this clearly, which can be further utilized for community zonation. Also the utility of mapping mangroves at different scales is compared and was found that a scale of 1:50,000 could depict forest blanks clearly and could aid in accurate area estimation.