ALGORITMA UNTUK DETEKSI TUMPAHAN MINYAK DI LAUT TIMOR MENGGUNAKAN CITRA MODIS

Muhammad Sudibjo; Vincentius P. Siregar; Jonson Lumban Gaol

doi:10.24319/jtpk.4.41-62

Muhammad Sudibjo Institut Pertanian Bogor
Vincentius P. Siregar Departemen Ilmu dan Teknologi Kelautan Fakultas Perikanan dan Ilmu Kelautan, Institut Pertanan Bogor
Jonson Lumban Gaol Departemen Ilmu dan Teknologi Kelautan Fakultas Perikanan dan Ilmu Kelautan, Institut Pertanan Bogor

DOI: https://doi.org/10.24319/jtpk.4.41-62

Keywords: Oil Spill, MODIS imagery, algorithm

Abstract

The Oil spill in Timor Sea in 2009 has spread the oil to the area of 10.842.81 km2. It was successfully detected by MODIS. The main purposes of this research were to compare the detection result of spill from several algorithms with image by using MODIS image and see the visual difference generated. There were four algorithm used in this research, namely: Oil Spill Index, Fluorescence Index, Principal Component Analysis (PCA), and Normalized Difference Vegetation Index (NDVI). These five algorithms resulted in the following comparisons: the appearance of visualization of oil spill seen on MODIS image with oil spill index and fluorescence index algorithm was brighter than the surrounding water bodies and they had higher pixel value. Mean while, oil visualization using PCA and NDVI algorithm was darker than the surrounding of water bodies and they had lower pixel value. The result of the test accuracy of oil spill index, fluorescence index, PCA, NDVI algorithm were 41%, 46%, 41%, and 60% respectivety.