Analisis Perubahan Tutupan Mangrove di Segara Anakan Menggunakan Citra Satelit Multitemporal

Abstract

Mangrove merupakan ekosistem penting karena menjadi habitat berbagai spesies, menyerap karbon dan menyaring nitrat dan fosfat dari sungai ke laut. Segara Anakan merupakan habitat mangrove yang mengalami perubahan bentuk lahan akibat proses sedimentasi yang kemudian memicu terjadinya perubahan komposisi mangrove. Penyusutan badan air serta konversi lahan menjadi faktor penyusutan habitat mangrove di Segara Anakan. Penelitian ini bertujuan menganalisis perubahan tutupan mangrove serta menyediakan informasi spasial terbaru yaitu peta tutupan lahan di Segara Anakan. Penelitian ini memanfaatkan Google Earth Engine (GEE) untuk melakukan pemrosesan data citra dan mengaplikasikan indeks mangrove yaitu Mangrove Vegetation Index (MVI), Automatic Mangrove Mapping and Index (AMMI) dan Enhanched Mangrove Index (EMI ) sebagai input dalam klasifikasi terbimbing. Klasifikasi terbimbing dilakukan dengan membandingkan algoritma klasifikasi yaitu Minimum Distance, Random Forest, k-Nearest Neighbhor (KNN), dan Classification and Regression Tree (CART). Hasil uji akurasi menunjukkan bahwa random forest memperoleh nilai akurasi tertinggi yaitu 78%. Berdasarkan hasil klasifikasi dengan random forest, mangrove pohon di Segara Anakan diketahui menyusut hingga 131.4 ha/tahun dalam periode 1990-2020 sejalan dengan badan air laguna yang mengalami penyusutan hingga 111.6 ha/tahun. Sementara luas nipa meningkat hingga 323.9 ha/tahun dalam periode tersebut. Hal ini mengindikasikan bahwa komunitas nipa semakin mendominasi Segara Anakan sementara mangrove pohon terus berkurang.

Mangroves are a crucial ecosystem as they provide habitat for various species, absorb carbon, and filter nitrates and phosphates from rivers to the sea. Segara Anakan is the mangrove habitat who has undergone landform changes due to sedimentation processes, which have subsequently triggered changes in mangrove composition. The reduction of water bodies and land conversion are also factors contributing to the decline of mangrove habitats in Segara Anakan. The purpose of this study is to analyze changes in mangrove cover and provide the latest spatial information, namely a land cover map of Segara Anakan. This study utilizes Google Earth Engine (GEE) for image data processing and applies mangrove indices such as the Mangrove Vegetation Index (MVI), Automatic Mangrove Mapping and Index (AMMI), and Enhanced Mangrove Index (EMI) as inputs in supervised classification. Supervised classification was conducted by comparing classification algorithms, namely Minimum Distance, Random Forest, k-Nearest Neighbor (KNN), and Classification and Regression Tree (CART). Accuracy test results show that Random Forest achieved the highest accuracy, at 78%. Based on the classification results using Random Forest, tree mangroves in Segara Anakan were found to be shrinking by up to 131.4 ha/year during the period 1990-2020, in line with the lagoon water body shrinking by up to 111.6 ha/year during the same period. Meanwhile, nipa area increased by up to 323.9 ha/year during that period. This indicates that the nipa community is increasingly dominating Segara Anakan while tree mangroves continue to decline.