Remote Sensing Approaches for Automated Global Mangrove Classification

Authors: Rishi Singh*, Clark University, Ron Eastman, Professor, John Rogan, Professor, Florencia Sangermano, Professor
Topics: Geographic Information Science and Systems, Remote Sensing, Coastal and Marine
Keywords: Mangrove, Landsat, Sentinel, Classification, Remote Sensing, Coastal, Tasseled Cap
Session Type: Poster
Day: 4/12/2018
Start / End Time: 1:20 PM / 3:00 PM
Room: Napoleon Foyer/Common St. Corridor, Sheraton, 3rd Floor
Presentation File: No File Uploaded


Mangrove forests are a critical component of coastal habitats that provide a variety of benefits including coastal protection, water purification, species habitats for nesting and breeding, and carbon sequestration. Despite their unique ecological role, staggering degradation of mangroves has occurred worldwide since the 1980’s, driven by rapid urbanization and popular land use practices like aquaculture and charcoal production. This deforestation has been most pronounced in southeast Asian countries. In response to this biological concern, numerous researchers have utilized remote sensing and GIS technologies to monitor changes in mangrove forests worldwide. While many projects have been successful in detecting mangrove forests for specific case studies, there are growing opportunities to use GIS for automated classification of mangroves at a global scale. To contribute to our understanding of global mangrove classification, this study explores a variety of classification approaches designed to detect mangrove forests for any study area based on multispectral imagery from different sources (e.g. Landsat and Sentinel platforms). Classification techniques explored are based on a variety of approaches using different inputs (e.g. tasseled cap transformation and mangrove indices) and classification methods (e.g. supervised, object-based image analysis, classification tree analysis). Classification methods are compared to one another to assess which method provides the most effective basis for an automated classification approach.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login