<![CDATA[This study analysed spatiotemporal changes in mangrove canopy density within the coastal areas of Pekalongan Regency and Pekalongan City using multitemporal Sentinel-2 imagery (2019, 2021, and 2024). This study was motivated by the continuous degradation of mangrove ecosystems due to severe tidal flooding, land subsidence, and coastal hydrodynamic disturbances, which necessitate reliable monitoring tools to support mitigation and restoration programs. The objective of this study was to compare the performance of NDVI, NDVI-Red Edge, and mRE-SR vegetation indices in estimating mangrove canopy density and to determine the most accurate index for tidal-affected environments. The methodological framework involved image preprocessing, land cover classification, vegetation index computation, and linear regression modelling validated by in situ canopy measurements obtained through hemispherical photography. The results showed that the mangrove area declined significantly between 2019 and 2021, followed by partial recovery in 2024 in response to rehabilitation efforts. Among the tested indices, NDVI-Red Edge Band 5 yielded the highest accuracy with the lowest RMSE (7.65%), outperforming NDVI and mRE-SR, whereas Bands 6 and 7 showed weak predictive capability. The study concluded that NDVI-RE Band 5 is the most reliable index for mapping mangrove canopy density in dynamic coastal environments affected by tidal inundation. These findings demonstrate the effectiveness of combining Sentinel-2 red-edge information with field-based validation to support mangrove monitoring and coastal management.]]>
