Bathymetry Extraction from Spot 7 Satellite Imagery Using Random Forest Methods

Kuncoro Teguh Setiawan and Nana Suwargana and Devica Natalia Br Ginting and Masita Dwi Mandini Manessa and Nanin Anggraini and Syifa Wismayati Adawiah and Atriyon Julzarika and Surahman and Syamsu Rosid and Agustinus Harsono Supardjo (2019) Bathymetry Extraction from Spot 7 Satellite Imagery Using Random Forest Methods. International Journal of Remote Sensing and Earth Sciences, 16 (1). pp. 23-30. ISSN 0216-6739

[img] Text
Jurnal_Kuncoro TS_Pusfatja_2019.pdf

Download (819kB)


The scope of this research is the application of the random forest method to SPOT 7 data to produce bathymetry information for shallow waters in Indonesia. The study aimed to analyze the effect of base objects in shallow marine habitats on estimating bathymetry from SPOT 7 satellite imagery. SPOT 7 satellite imagery of the shallow sea waters of Gili Matra, West Nusa Tenggara Province was used in this research. The estimation of bathymetry was carried out using two in-situ depth-data modifications, in the form of a random forest algorithm used both without and with benthic habitats (coral reefs, seagrass, macroalgae, and substrates). For bathymetry estimation from SPOT 7 data, the first modification (without benthic habitats) resulted in a 90.2% coefficient of determination (R2) and 1.57 RMSE, while the second modification (with benthic habitats) resulted in an 85.3% coefficient of determination (R2) and 2.48 RMSE. This research showed that the first modification achieved slightly better results than the second modification; thus, the benthic habitat did not significantly influence bathymetry estimation from SPOT 7 imagery

Item Type: Article
Uncontrolled Keywords: bathymetry, random forest, SPOT 7
Subjects: Teknologi Penginderaan Jauh > Pengelolaan dan Pengembangan > Citra Satelit
Divisions: Deputi Penginderaan Jauh > Pusat Pemanfaatan Penginderaan jauh
Depositing User: Dinar Indrasasi
Date Deposited: 27 Jul 2021 23:25
Last Modified: 27 Jul 2021 23:25

Actions (login required)

View Item View Item