Automated Mangrove Detection Method using Combined Machine Learning and Mangrove Index over Indonesia
1Research Center for Geoinformatics, Research Organization for Electronics and Informatics, National Research and Innovation Agency, M.H. Thamrin No. 8, Jakarta Pusat 10340 Indonesia
2Deputy for Research and Innovation Infrastructure, National Research and Innovation Agency, M.H. Thamrin No. 8, Jakarta Pusat 10340 Indonesia
3Center for Data and Information, National Research and Innovation Agency, M.H. Thamrin No. 8, Jakarta Pusat 10340 Indonesia
4Faculty of Social and Political Sciences, Semarang State University, Sekaran, Gunung Pati, Semarang City, Central Java 50229 Indonesia
*Author to whom correspondence should be addressed:
E-mail: kurnia.ulfa.example@university.edu (KU)
E-mail: kurnia.ulfa.example@university.edu (KU)
Received: September 27, 2024 | Revised: February 06, 2025 | Accepted: April 01, 2025 | Published: June 2025
Abstract
The carbon storage potential of mangrove forests is higher than highland tropical forests. It makes mapping mangroves essential to estimate a total carbon storage potential. Integration of satellite data which covers a wide area and machine learning can obtain mangrove detection to be faster and more accurate. The study proposes an automated mangrove detection method using a combined mangrove index and machine learning over Indonesia. Normalized Difference Moisture Index and Mangrove Vegetation Index from Landsat 8/9 images were used to obtain training datasets for the Random Forest process. The results show that the overall accuracy is 0.93, and the kappa accuracy is 0.91. It proves that the proposed method can be used for mapping mangroves and non-mangroves accurately.
Keywords
machine learning ; random forest ; Landsat 8 ; Normalized Difference Moisture Index ; Mangrove Vegetation Index ; Landsat 9 ; mangrove mapping
Available Repositories
Share Article
Article Metrics
--
Views
--
Downloads
--
Citations
Export Citation
Full Text
