Mapping Mangrove Above-Ground Carbon Using Multi-Source Remote Sensing Data and Machine Learning Approach in Loh Buaya, Komodo National Park, Indonesia

A novel hybrid model combining Extreme Gradient Boosting (XGB) and Genetic Algorithm (GA) has demonstrated high accuracy in estimating mangrove above-ground carbon (AGC) in Loh Buaya, Komodo National Park, Indonesia. The model achieved an R² of 0.857 during training and 0.758 during testing, significantly outperforming other machine learning techniques like Random Forest and Support Vector Machine. Using multisource remote sensing data—including optical imagery from Sentinel-2B, Synthetic Aperture Radar (SAR), and a national Digital Elevation Model (DEM)—the study estimated AGC ranging from 2.52 to 123.89 Mg C ha⁻¹, with an average of 57.16 Mg C ha⁻¹. This approach offers a fast, reliable, and scalable method for quantifying mangrove carbon stocks using open-source data, with potential global applications in tropical ecosystems.

Rijal, S. S., Pham, T. D., Noer’Aulia, S., Putera, M. I., & Saintilan, N. (2023). Mapping Mangrove Above-Ground Carbon Using Multi-Source Remote Sensing Data and Machine Learning Approach in Loh Buaya, Komodo National Park, Indonesia. Forests, 14(1), 94. https://doi.org/10.3390/f14010094