IDENTIFICATION OF MANGROVE CHANGES IN THE MAHAKAM DELTA IN 2007-2017 USING ALOS/PALSAR AND LANDSAT
Keywords:Mangrove, Mahakam Delta, Google Earth Engine (GEE), ALOS/PALSAR, Landsat
The mangrove area in the Mahakam Delta has dynamically changed due to the land-use conversion for various purposes. Various remote sensing data can monitor the changes, for example, ALOS/PALSAR and Landsat imagery. However, there are limited studies that compare the use of both imageries to monitor such changes. This paper aims to compare the ability of two satellite imageries, i.e., ALOS/PALSAR and Landsat, to monitor the dynamic of mangrove areas. Two time-series data of ALOS/PALSAR and Landsat imagery for the acquisition period between 2007 and 2017 were analyzed using the Support Vector Machine (SVM) classification method on the Google Earth Engine (GEE). Landsat analysis results show an increase in the mangrove area of about 17,016 ha and a reduction of about 6,377 ha. ALOS/PALSAR images showed an increase of 15,903 ha and a reduction of 12,713 ha. The change detection results using two different imageries, i.e., Landsat and PALSAR, show slightly different results. Mangrove areas in 2007 and 2017 increased the area as detected from both Landsat and PALSAR. Landsat imaging classification is better at identifying mangroves from non-mangroves, although the 2007 classification results have flaws due to recording errors in striping. Because the quality of PALSAR 2007 and PALSAR 2017 images is not affected, the classification of PALSAR images is deemed more consistent in the area calculation. However, classification results in separating mangrove and non-mangrove near bodies of water are lacking.