Development of Near-Real Time Forest Monitoring (Phase I: Data Preparation)

Deforestation and forest degradation are still major problems in Indonesia, even though the magnitude has been decreasing recently. In order to overcome these problems, an immediate action with accurate and up-to-date information is essensial. Expired information, due to out of date is usually hamper the immediate response. Remote sensing technology seems to be a powerful tool to monitor the change of forest cover. However, an availability of cloud free satellite images in tropical region is another of obstacle. Regarding to the deforestation and forest degradation issues, it needs to develop near-real time forest monitoring, that publicly available to support the transparency in managing the natural resources. As an initial work of the development of early deforestation detection system, we employed the long-term MODIS image datasets, acquired from 2001 to 2013 with 286 time series data totally. Although MODIS data have some advantages in providing basic information related to the real time change detection, these time-series data inevitably contain disturbance caused by atmospheric variability and aerosol scattering. This study is an early phase in developing the near-real time forest monitoring system, which is focused on providing a reliable image datasets to support this system. The results suggested that MODIS data offer great opportunities to support the near-real time forest monitoring system at large scale.