Trida Ridho Fariz, Amnan Haris, Nana Kariada Tri Martuti, Norma Eralita, Luthfi Hanum Saputri, Gilang Syahbananto, Cintiya Egi Purwadi, Zahra Rafidah, Sheeny Az-Zahra
This study evaluates land cover mapping using Landsat 8 and Landsat 9 satellite imagery, focusing on the coastal regions of Pekalongan Regency and Kendal Regency in Indonesia, which face significant environmental challenges due to climate change. Utilizing machine learning algorithms from Google Earth Engine (GEE), the research employs a supervised classification technique based on Random Forest (RF) to assess land cover. The analysis reveals overall accuracy values of 0.75 for Landsat 8 and 0.82 for Landsat 9, with Kappa values of 0.68 and 0.78, respectively. The accuracy of individual land cover types varies, with water bodies achieving the highest accuracy and agricultural land the lowest. The study suggests prioritizing Landsat 9 imagery for medium-scale mapping. © Published under licence by IOP Publishing Ltd.
Environmental Science, Universitas Negeri Semarang, Indonesia; Biology, Universitas Negeri Semarang, Indonesia; Science Education, Universitas Negeri Semarang, Indonesia
