Author(s)

Syartinilia Wijaya, Satoshi Tsuyuki

Department of Landscape Architecture, Faculty of Agriculture, Bogor Agricultural University (IPB), Bogor, Indonesia.
Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo, Japan.

As landmass of the world is covered by vegetation, taking into account phenology when performing land cover classification may yield more accurate maps. The availability of no-cost Moderate Resolution Imaging Spectrometer (MODIS) NDVI dataset that provides high-quality continuous time series data is representing a potentially significant source of land cover information especially for detection natural forest distribution. This study intends to assess the advantage of MODIS 250 m Normalized Difference Vegetation Index (NDVI) multi-temporal imagery for detection of densely vegetation cover distribution in Java and then for identification of remaining natural forest in Java from densely vegetation cover distribution. Result of this study successfully demonstrated the contribution of MODIS NDVI 250 m for detection the natural forest distribution in Java Island. Therefore, the approach described herein provided classification accuracy comparable to those of maps derived from higher resolution data and will be a viable alternative for regional or national classifications.

Java; MODIS; Multi-Temporal; Natural Forest; NDVI

S. Wijaya and S. Tsuyuki, "Contribution of MODIS NDVI 250 m Multi-Temporal Imagery Dataset for the Detection of Natural Forest Distribution of Java Island, Indonesia," Journal of Geographic Information System, Vol. 4 No. 5, 2012, pp. 462-469. doi: 10.4236/jgis.2012.45050.