Pemetaan Kerapatan Mangrove di Cilacap Menggunakan Google Earth Engine (GEE)

Abstract

Eksploitasi mangrove yang berlebihan oleh manusia mengakibatkan luasan mangrove semakin berkurang seiring berjalannya waktu. Monitoring dan penanaman kembali menjadi fokus utama untuk menyelamatkan kelestarian mangrove. Penelitian ini bertujuan memetakan sebaran dan kerapatan mangrove di Kolak Sekancil, Cilacap, menggunakan citra Sentinel-2A. Citra diambil dan diolah menggunakan Google Earth Engine (GEE) dengan algoritma Mangrove Vegetation Index (MVI) dan Normalized Difference Vegetation Index (NDVI). Data lapang mangrove dan bukan mangrove diambil sebanyak 65 titik menggunakan metode hemispherical photography. Algoritma MVI digunakan untuk membedakan wilayah mangrove dan bukan mangrove dengan rentang nilai 2,372 – 18,079 untuk wilayah mangrove. Algoritma NDVI digunakan untuk melihat sebaran dan kerapatan mangrove di wilayah penelitian. Terdapat 3 kelas kerapatan yang dihasilkan, yaitu rapat (11,72 ha), sedang (0,69 ha), dan jarang (1,13 ha). Total luasan mangrove yang didapatkan sebesar 13,54 ha dengan tingkat akurasi mencapai 82,22% dan koefisien kappa 0,89. Korelasi antara persentase kanopi mangrove dengan nilai NDVI ditunjukkan dengan nilai koefisien korelasi (r) sebesar 0,797 dan koefisien determinasi (r2) sebesar 0,636.

The excessive exploitation of mangroves by humans has led to a decline in mangrove coverage over time. Monitoring and replanting have become the primary focus for preserving mangrove sustainability. This study aims to map the distribution and density of mangroves in Kolak Sekancil, Cilacap, using Sentinel-2A imagery. The images were captured and processed using Google Earth Engine (GEE) with the Mangrove Vegetation Index (MVI) and Normalized Difference Vegetation Index (NDVI) algorithms. The field data for mangrove and non-mangrove areas were collected at 65 locations using the hemispherical photography method. The MVI algorithm was used to differentiate mangrove from non-mangrove areas with a value range of 2,372 – 18,079 for mangrove regions. The NDVI algorithm was employed to assess the distribution and density of mangroves in the study area. Three density classes were identified: dense (11.80 ha), moderate (0.40 ha), and sparse (0.28 ha). The total area mapped was 12.48 ha with an accuracy rate of 82.22% and a kappa coefficient of 0.89. The correlation between the mangrove canopy percentage and NDVI value is indicated by a correlation coefficient (r) value of 0.797 and a determination coefficient (r2) of 0.636.