STUDY ON VEGETATION COVER CHANGES IN THE PROVINCE OF SOUTH KALIMANTAN USING RGB-NDVI UNSUPERVISED CLASSIFICATION METHOD
Vegetation is one of the largest supporting structures in biosphere ecological unit. Satellite images are very useful in monitoring vegetation cover changes at regional and global scales. This study employed multitemporal SPOT Vegetation images acquired in 2000, 2002, and 2004 for analyzing recent vegetation cover changes over the Province of South Kalimantan using RGB-NDVI unsupervised classification method. This visual RGB-NDVI method involved creation of color composite images and utilization of additive color theory, where each NDVI from three dates is combined with the red, green, and blue color write functions of the computer monitor. The procedure classified events of change and no-change in the study area in all periods into nine classes of events, with 80% agreement with 810 reference sample
points. Research results showed that locations of vegetation index change in South Kalimantan area in small part located around the boundary area of the province, i.e. in western and northern area, whereas the larger part scattered from southern to eastern areas and in almost all of its coastal zone, and spread
around urban areas. In Laut Island and Sebuku Island, locations of vegetation index changes scattered over the island areas. The mean value of vegetation index change areas tend to decrease with reduction rate of 0.32% of its land region.
Dibuat oleh : Dewi Kania Sari, Edhis Triyono Hermawan, Gun Gun Gunawan Hudman
E-mail: dewiks@itenas.ac.id
Kata kunci : Change Detection, Vegetation Cover, RGB-NDVI Method, SPOT Vegetation, South Kalimantan
Keterangan : Karya Ilmiah ini dipresentasikan pada Konfrensi Internasional Map Asia 2005 Jakarta, 22-25 Agustus 2005 dan dimuat pada Proceeding of International Conference of Map Asia 2005 Jakarta, in CS-ROM