Satellite remote sensing of mangrove forests: Recent advances and future opportunities

Abstract

Mangroves are salt tolerant woody plants that form highly productive intertidal ecosystems in tropical and subtropical regions. Despite the established importance of mangroves to the coastal environment, including fisheries, deforestation continues to be a major threat due to pressures for wood and forest products, land conversion to aquaculture, and coastal urban development. Over the past 15 years, remote sensing has played a crucial role in mapping and understanding changes in the areal extent and spatial pattern of mangrove forests related to natural disasters and anthropogenic forces. This paper reviews recent advancements in remote-sensed data and techniques and describes future opportunities for integration or fusion of these data and techniques for large-scale monitoring in mangroves as a consequence of anthropogenic and climatic forces. While traditional pixel-based classification of Landsat, SPOT, and ASTER imagery has been widely applied for mapping mangrove forest, more recent types of imagery such as very high resolution (VHR), Polarmetric Synthetic Aperture Radar (PolSAR), hyperspectral, and LiDAR systems and the development of techniques such as Object Based Image Analysis (OBIA), spatial image analysis (e.g. image texture), Synthetic Aperture Radar Interferometry (InSAR), and machine-learning algorithms have demonstrated the potential for reliable and detailed characterization of mangrove forests including species, leaf area, canopy height, and stand biomass. Future opportunities include the application of existing sensors such as the hyperspectral HYPERION, the application of existing methods from terrestrial forest remote sensing, investigation of new sensors such as ALOS PRISM and PALSAR, and overcoming challenges to the global monitoring of mangrove forests such as wide-scale data availability, robust and consistent methods, and capacity-building with scientists and organizations in developing countries.