Biomass Prediction of Mangrove Forests in Langkat Regency Using Random Forest Regression Method
Rahmat Sukron Ardi Hidayat, Freza Riana, Sahid Agustian Hudjimartsu · e-Jurnal Penyelidikan dan Inovasi · 2025
Mangrove forests are important ecosystems that play a role in maintaining the stability of coastal ecosystems, absorbing carbon dioxide (CO2) more effectively than terrestrial forests, and protecting coastal areas from abrasion and the impact of natural disasters. This research was conducted in Lubuk Kertang Village, Langkat Regency, North Sumatra, with an area of 3026 km². The main objective of this study is to predict the biomass of mangrove trees using the Random Forest Regression (RFR) method by using unmanned aerial vehicles (UAVs) to obtain Biomass on the Ground (AGB). Model validation was carried out using Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE), R-squared (R²). The results of the analysis showed that the RFR model gave excellent results with an R² value of 0.9697, RMSE of 2.55, and MAE of 1.82. The distribution of biomass in the study area showed significant variation, with the average tree biomass of 21,136 Kg/tree, the lowest biomass of 6,588 Kg/tree, and the highest biomass of 46,700 Kg/tree.