Analisis Sebaran dan Kerapatan Hutan Mangrove dari Pengolahan Citra Sentinel-2B di Daerah Kurau, Bangka Belitung

Abstract

Kawasan hutan mangrove memiliki manfaat dalam aspek sosial, ekonomi, maupun ekologi. Fungsi fisik hutan mangrove sebagai pengendali intrusi air laut terhadap daratan, kawasan penyanga, dan pelindung garis pantai. Algoritma Maximum Likelihood Classification (MLC) digunakan untuk memetakan sebaran mangrove. Analisis Normalized Difference Vegetation Index (NDVI) digunakan untuk mengetahui indeks kerapatan vegetasi suatu daerah. Dari hasil perhitungan Hutan mangrove di Kurau, Bangka Belitung pada 2021 didapatkan luas total 695.94 Ha yang terbagi menjadi tiga kelas kerapatan, yaitu mangrove padat (446.36 Ha), mangrove sedang (110.43 Ha), dan mangrove jarang (80.58 Ha). Uji akurasi dengan confusion matrix mendapatkan akurasi sebaran sebesar 94.02% dengan kappa coefficient 0.88, dan akurasi kerapatan sebesar 90% dengan kappa coefficient sebesar 0.81. Pengujian korelasi antara data survei kerapatan mangrove dengan NDVI didapatkan nilai korelasi (r) sebesar 0.8299 dan nilai determinasi (R2) sebesar 0.6887.

Mangrove areas have benefits in social, economy, and ecology aspects. The physical function of mangrove forests is to control sea water intrusion into land area, buffer areas, and coastlines protector. Maximum Likelihood Classification (MLC) algorithm is used to map mangrove coverage distribution. Normalized Difference Vegetation Index (NDVI) analysis is used to determined the vegetation density index of an area. The results of computations obtained mangrove forest area in Kurau region, Bangka Belitung in 2021 has a total area of 695.45 Ha which is divided into three density classes, namely dense mangrove (446.36 Ha), moderate mangrove (110.43 Ha), and rare mangrove (80.58 Ha). The accuracy test using confusion matrix got a distribution accuracy of 94.02% with a kappa coefficient of 0.88, and a density accuracy of 90% with kappa coefficient of 0.81. The correlation test between mangrove density survey data and NDVI obtained a correlation value (r) of 0.829 and determination value (R2) of 0.6887.