Mapping Shallow Water Benthic Habitats Using High Resolution Multispectral UAV Data: A Case Study of Panggang Island

Devi Aliyah Sari (Department of Marine Technology , Faculty of Fisheries and Marine Sciences, IPB University, 16680, Indonesia)
Vincentius P. Siregar (Department of Marine Science and Technology, Faculty of Fishery and Marine Science, IPB University, Dramaga Campus, Bogor 16680, Indonesia)
Syamsul Bahri Agus (Department of Marine Science and Technology, Faculty of Fishery and Marine Science, IPB University, Dramaga Campus, Bogor 16680, Indonesia)

Jurnal Ilmu dan Teknologi Kelautan Tropis
Vol 18 No 1 (2026): Jurnal Ilmu dan Teknologi Kelautan Tropis

The use of Unmanned Aerial Vehicles (UAVs) has become increasingly important for high-resolution remote sensing applications, particularly for mapping coastal and shallow-water environments. Benthic habitats in shallow marine environments include seagrass meadows, macroalgae, live coral reefs, and degraded coral communities associated with sandy, muddy, and coral rubble substrates. This study mapped benthic habitats on Panggang Island, Indonesia, using multispectral imagery acquired by a DJI Phantom 4 Multispectral UAV with a spatial resolution of 6 cm per pixel. Seven benthic habitat classes were identified; sand , seagrass, live coral, dead coral with algae, coral with algae, rubble, and macroalgae. Habitat classification was performed using a pixel-based Support Vector Machine (SVM) algorithm. Classification accuracy was evaluated using a confusion matrix, yielding an overall accuracy of 87% and a Kappa coefficient of 0.84 The results demonstrate that integrating high-resolution UAV multispectral imagery with pixel-based SVM classification provides an effective approach for detailed benthic habitat mapping in small-island shallow-water environments.