Mapping forest cover change in relation to oil palm commodities development using Google Earth Engine in the Leuser Ecosystem

Abstract

The Leuser ecosystem is a vital biodiversity hotspot providing essential ecosystem services to local communities in Asia. It encompasses diverse landscapes, from rainforests to coastal ecosystems and peatlands. However, human activities, particularly deforestation for palm oil plantations, have severely impacted Leuser ecosystem’s ecological balance. This study analyzes land cover changes in Leuser ecosystem between 2016 and 2022, using satellite imagery from platforms like Sentinel-1 and Landsat 8 on Google Earth Engine. During this period, forest cover decreased by 2.8%, while oil palm plantations expanded by 3.15%, with 23,404 hectares of forested land converted. Unregulated activities by small-scale farmers accounted for 73% of this conversion. The accuracy assessment of the land use classification showed moderate accuracy in 2016, with an overall accuracy of 64.95% and a kappa coefficient of 0.388. However, in 2022, the accuracy improved significantly, with an overall accuracy of 73.20% and a kappa coefficient of 0.520. Though land use classification accuracy improved from 2016 to 2022, more robust monitoring and regulation are necessary to address the threats posed by palm oil development. Striking a balance between economic development and environmental preservation in Leuser ecosystem requires sustainable palm oil practices and conservation initiatives.