Mapping The Central Matang Mangrove Forest Reserve, Perak, Using Remote Sensing And Geographic Information System

Mangroves are characterized by littoral forest formation occurring in all estuaries of the Peninsular Malaysia. It plays an important role to protect the shoreline along the coast. In Malaysia, although mangroves are well managed especially in Perak, Johor and Selangor but the integration of remote sensing with geographic information system (GIS) for mapping and managing mangrove forest is not widely practiced. The purpose of this study is to use remote sensing technique using SPOT and IKONOS data integrated with GIS for mapping the extent of mangrove forest in central part of MMFR and for quantifying temporal changes in stand density and areal extent within the MMFR from year 1989 – 2000.A study in mapping the mangrove forest using remote sensing integration with GIS was carried out in central part of Matang Mangrove Forest Reserve (MMFR) in the Range Kuala Trong, Perak. The study area faces the Straits of Malacca lying between latitudes 438N to 449N and longitudes 10020E to 10036E, where the classification of mangrove forest areas was carried out and recorded. Multispectral SPOT (Systeme Pour‟l Observation de la Terre) images of 1989, 1993, 1997 and 2000 and IKONOS image of 2000 for Kuala Trong areas (based on AOI) were enhanced, classified and vectorized using image processing software for the purpose of mapping the mangrove forest. Spatial data for the mangrove forests such as information of compartments, blocks, names of area digitized by the Forestry Department (Mapping and GIS Section) using ARC/INFO Version 3.4.2 Geographic Information System (GIS) software were used as secondary data in the study. Based on the image analysis of the SPOT images, the mangrove forest reserves were classified as Excellent Forest Reserve, Good Forest Reserve, Poor Forest Reserve, Dryland Forest Reserve and Damaged Forest Reserve. These five classes of mangrove forests, can be further categorised as Productive and Non-productive area. The analysis showed that the average volumes of timber available within the productive areas of the study site were Excellent Forest (362.50m3/ha - 50.82%); Good Forest (256.31m3/ha -5.93%) and Poor Forest (94.54m3/ha - 13.25%) with an overall classification accuracy of more than 70% while the statistics value obtained from Kappa‟s was shown more than 0.6 which is relatively quite good results for image processing. It can be concluded that the satellite remote sensing with the integration of GIS can be successfully used and implemented for mangrove classification and mapping for the advance purposes of providing fast, efficient and accurate information on the mangrove resource.

Preview PDF FH_2006_7a.pdf Download (539kB)

Actions (login required)

View Item