<![CDATA[Vegetation cover plays an important role in controlling the view, boundaries, air temperature, living place and aesthetics in an area. Vegetation cover changes can be caused by changes in temperature, rainfall and human activities. Google Earth Engine (GEE) provides machine learning algorithms such as NDVI which are very useful in extracting vegetation density levels from imagery. The purpose of this study was to analyze vegetation cover changes by human activities in relation to the geomorphological form of Kendari City. The imagery used in multi-temporal monitoring are Landsat-7 ETM in 2000, Landsat-5 TM in 2005 and 2010 and Landsat 8 OLI in 2015 and 2020. Input machine learning using near infrared (NIR) and red (Red) for the NDVI Algorithm while the geomorphological form uses SRTM imagery. The classification of vegetation cover consists of water bodies, open field, built areas and roads covered with asphalt, paving or soil, plantations/agriculture, bushes, grass, reeds, green open space and forests. Each sub-district experienced a decrease in vegetation cover in the form of plantations/agriculture, bushes, grass, reeds, green open space except for the West Kendari District which tended to be varied. The forest area is getting better every year. The existence of protected forests and geomorphological forms such as lowlands are the driving factors for changes in vegetation cover, while low hills and high hills are flat to steep are contrainst factors. Machine learning in GEE is very helpful in monitoring vegetation cover changes and has an NDVI algorithm that is quite easy to apply.]]>
Copyrights © 2021
