| dc.description.abstract | Mangrove forests play a vital role in coastal ecosystems, particularly in carbon sequestration. The calculation of above ground carbon content in mangrove forest ecosystems plays a crucial role in global climate change mitigation efforts, given the significant carbon storage capacity of mangrove forests. Information on carbon stocks can aid in formulating conservation policies and sustainable natural resource management. This study aims to model the above ground carbon content in the mangrove forest of Lubuk Kertang Village, Brandan Barat District, Langkat Regency, using Sentinel-1 satellite imagery. The methodology includes direct biomass measurements in the field using 20 × 20 meter plots and image data analysis employing the Random Forest algorithm and regression. The processed data includes tree species, diameter at breast height (DBH), and backscatter values from Sentinel-1 imagery. The linear regression model using the VV+VH combination, represented by the equation y = 8.6664x + 262, produced the best results with a coefficient of determination (R²) of 86.98% and a Root Mean Square Error (RMSE) of 8.63, indicating a high level of accuracy in estimating above-ground carbon stock. These findings demonstrate that Sentinel-1 imagery is effective in estimating carbon stocks, providing a valuable basis for supporting conservation efforts and sustainable mangrove forest management. | en_US |
