<![CDATA[Mangrove forests are spatially dynamic intertidal ecosystems whose monitoring is constrained by difficult field access, tidal variation, cloud cover, spectral similarity with terrestrial vegetation, and rapid land-use change. Remote sensing has therefore become a central tool for mapping mangrove extent, condition, species composition, structural attributes, and temporal change. This review synthesizes recent developments in optical, synthetic aperture radar (SAR), hyperspectral, unmanned aerial vehicle (UAV), and LiDAR approaches for mangrove mapping, with emphasis on freely available Landsat and Sentinel data and applications relevant to tropical coasts. Studies published mainly during 2016–2026 were examined across four analytical dimensions: sensor characteristics, spectral or structural features, classification algorithms, and accuracy assessment. The evidence shows that Landsat remains valuable for long-term change analysis, whereas Sentinel-2 improves detection of narrow and fragmented stands through 10-m observations. Combining Sentinel-1 SAR and Sentinel-2 optical time series generally improves robustness in persistently cloudy regions. Mangrove-specific indices such as the Mangrove Vegetation Index can simplify extent mapping, while hyperspectral imagery, UAV data, and LiDAR are increasingly important for species discrimination, canopy height, biomass, and restoration assessment. However, no sensor or index removes the need for representative training data and independent validation. Tidal stage, mixed pixels, seasonal compositing, class imbalance, and spatially non-independent validation remain major sources of uncertainty. For Indonesia, an operational strategy should integrate multi-temporal Sentinel-1/2 data, machine learning, field observations, and standardized accuracy reporting to support repeatable mapping for conservation, restoration, blue-carbon accounting, and coastal management.]]>
