Pemetaan Sebaran dan Kerapatan Mangrove di Pesisir Belitung Timur Menggunakan Citra Pleiades

Abstract

Ekosistem mangrove di pesisir Belitung Timur mengalami penurunan luasan akibat dialihfungsikan menjadi area tambang timah oleh masyarakat. Penelitian ini bertujuan memetakan sebaran dan kerapatan mangrove di pesisir Belitung Timur dengan pendekatan penginderaan jauh dan Sistem Informasi Geografis. Kerapatan mangrove ditinjau dengan menghitung persentase tutupan kanopi mangrove menggunakan metode hemispherical photography dan analisis Normalized Difference Vegetation Index (NDVI). Selain itu, metode pixel-based classification dan Object-Based Image Analysis (OBIA) digunakan untuk mengklasifikasikan citra Pleiades. Hasil menunjukkan dominansi tingkat kerapatan mangrove di lokasi berdasarkan nilai hemispherical photography adalah sedang. Sementara, nilai kerapatan mangrove menggunakan analisis NDVI didominasi oleh kelas rapat. Hasil uji akurasi menggunakan confusion matrix antara kerapatan mangrove hasil NDVI dengan tutupan kanopi hasil data lapang menunjukkan nilai overall accuracy (OA) sebesar 88%. Analisis regresi linear sederhana menunjukkan bahwa hubungan persentase tutupan kanopi mangrove dengan nilai NDVI menghasilkan korelasi yang kuat dengan nilai koefisien korelasi (r) sebesar 0,801 dan nilai koefisien determinasi (R2) sebesar 0,642.

The extent of mangrove ecosystems on the coast of East Belitung has diminished as a consequence of their conversion into tin mining areas by the local community. The objective of this study was to map the distribution and density of mangroves on East Belitung's coast using remote sensing and geographic information system (GIS) approaches. Mangrove density was quantified by calculating the percentage of mangrove canopy cover using hemispherical photography and a Normalized Difference Vegetation Index (NDVI) analysis. Moreover, pixel-based classification and object-based image analysis (OBIA) techniques were utilized to categorize the Pleiades images. The results demonstrated that the predominant level of mangrove density in the area, as determined by hemispherical photography, was classified as medium. In contrast, the NDVI analysis indicated that the dense class constituted the majority of the mangrove density values. The results of the accuracy assessment, conducted using a confusion matrix between mangrove density derived from NDVI and canopy cover data obtained from field observations, demonstrated an overall accuracy (OA) value of 88%. Simple linear regression analysis revealed that there was a strong correlation between the percentage of mangrove canopy cover and NDVI values, with a correlation coefficient (r) of 0.801 and a coefficient of determination (R2) of 0.642.