Approaches, potential, and challenges in the use of remote sensing to study mangrove and other tropical wetland forests

Abstract

Tropical wetland forests are fragile ecosystems facing critical risks due to global warming and other anthropogenic threats. Hence, gathering accurate and reliable information on them is urgent. Although remote sensing has demonstrated great potential in studying terrestrial ecosystems, remote sensing-based wetland forest research is still in an early stage of development. Mapping wetland forests, particularly mangrove forests, was an initial goal of this approach and is a task that still faces methodological challenges. Initially based on aerial photography only, wetland forest mapping through remote sensing underwent explosive diversification after the launching of artificial satellites in the 1970s. Later, precision in wetland forest mapping increased with the combination of hyperspectral, multispectral, and high and very high resolution imagery. Accurate delimitation of wetland forest extent is also necessary to assess their temporal dynamics (losses, gains, and horizontal displacement). Despite the prevalence of mapping studies, current remote sensing-based research on wetland forests addresses new questions and novel aims, such as describing and predicting wetland forest attributes through mathematical modeling. Although this approach has made substantial progress in recent decades, modeling and predicting wetland forest attributes remain insufficiently explored fields of research. Combining active and passive sensors is a promising alternative to provide a more accurate picture of these communities’ attributes. In particular, LiDAR and radar-based technologies may help overcome difficulties encountered in older studies. In the future, we will witness conceptual and methodological progress that will enable us to surmount the remaining challenges.