DETERMINING BURNED AREAS USING DIFFERENT THRESHOLD VALUES OF NDVI WITH SENTINEL-2 SATELLITE IMAGES ON GEE PLATFORM: A CASE STUDY OF MUĞLA PROVINCE

Authors : Sinan Demir, Ibrahim Dursun

Pages : 117-130

View : 90 | Download : 88

Publication Date : 2023-12-29

Article Type : Research Paper

Abstract :Remote sensing technologies play a critical role in detecting land cover change and burned areas. Therefore, it is aimed to utilize satellite imagery and the Google Earth Engine (GEE) platform for the identification of burned areas and assessment of land cover changes (LULC). This study used the code block in the GEE platform to detect burned areas using high-resolution Sentinel-2 satellite imagery. The study focused on the province of Muğla, and the Normalized Difference Vegetation Index (NDVI) was calculated from Sentinel-2A satellite images for September 2020 and September 2021. Different threshold values (0.2, 0.3, 0.4, 0.5, 0.6) were applied to the different layers created from the dNDVI (Differenced Normalized Difference Vegetation Index) layers, and it was masked in the Global Water Surface dataset, to determine land cover changes and burned areas. The study results were compared with the Moderate Resolution Imaging Spectroradiometer (MODIS) burned area data. As a result, a threshold value of 0.3 was determined for the high-resolution satellite imagery, which can be used to identify land cover changes and burned areas without the interference of mixed pixels. The determined threshold value covered 82.77% (46966.44 hectares) of the area identified by the MODIS satellite. The used code block, utilizing Sentinel-2 satellite imagery within the GEE platform, not only successfully detected changes in burned areas but also holds promise as an effective tool for monitoring and tracking such changes.
Keywords : CBS, orman yangını, uzaktan algılama, mekansal analiz