Coastal Wetlands are changing at unprecedented rates (Reyes et al., 2000 & 2004) and the ability to monitor their changes remotely over time has become increasingly important (Hilbert, 2006 & Blaschke, 1996). The focus of this project was to complete a historical sub-pixel prediction land use classification map for six land classes: water estuarine emergent wetlands, fresh water wetlands, forests, low intensity developed areas and high intensity developed areas using high spatial resolution imagery and applying the proportions to medium resolution imagery for the past twenty-seven years (1984-2011) for the Western Lower Pascagoula River Basin. The other main objective was to complete an analysis of how the estuarine emergent marsh has changed during this historical time series. Landsat (medium resolution) and Quickbird (high resolution) imagery will be utilized to complete a regression tree based on the digital numbers and predictions will be completed by comparing multiple methods including pruning, bootstrapping and random forest. Then an image differencing will be performed to show areas of marsh change.
","abstract_html":"<p><p>Coastal Wetlands are changing at unprecedented rates (Reyes et al., 2000 & 2004) and the ability to monitor their changes remotely over time has become increasingly important (Hilbert, 2006 & Blaschke, 1996). The focus of this project was to complete a historical sub-pixel prediction land use classification map for six land classes: water estuarine emergent wetlands, fresh water wetlands, forests, low intensity developed areas and high intensity developed areas using high spatial resolution imagery and applying the proportions to medium resolution imagery for the past twenty-seven years (1984-2011) for the Western Lower Pascagoula River Basin. The other main objective was to complete an analysis of how the estuarine emergent marsh has changed during this historical time series. Landsat (medium resolution) and Quickbird (high resolution) imagery will be utilized to complete a regression tree based on the digital numbers and predictions will be completed by comparing multiple methods including pruning, bootstrapping and random forest. Then an image differencing will be performed to show areas of marsh change. </p></p>","abstract_has_math":false,"creators":["Frey, Jennifer"],"institution":null,"degree_name":"Master of Science (MS)","degree_level":"Masters Thesis","degree_discipline":null,"degree_department":null,"school":null,"contributors":["Wei Wu","Greg Carter","Maria Kalcic"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2011,"date_issued":"2011-12-01T08:00:00Z","date_published":"2011-12-01T08:00:00Z","updated_at":"2026-07-24T05:45:40Z","subjects":[],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://aquila.usm.edu/masters_theses/978","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Wei Wu","Greg Carter","Maria Kalcic"]},{"key":"dc:creator","label":"Author","values":["Frey, Jennifer"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2073-07-14T07:00:00Z"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Masters Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MS)"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://aquila.usm.edu/masters_theses/978"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Coastal Wetlands are changing at unprecedented rates (Reyes et al., 2000 & 2004) and the ability to monitor their changes remotely over time has become increasingly important (Hilbert, 2006 & Blaschke, 1996). The focus of this project was to complete a historical sub-pixel prediction land use classification map for six land classes: water estuarine emergent wetlands, fresh water wetlands, forests, low intensity developed areas and high intensity developed areas using high spatial resolution imagery and applying the proportions to medium resolution imagery for the past twenty-seven years (1984-2011) for the Western Lower Pascagoula River Basin. The other main objective was to complete an analysis of how the estuarine emergent marsh has changed during this historical time series. Landsat (medium resolution) and Quickbird (high resolution) imagery will be utilized to complete a regression tree based on the digital numbers and predictions will be completed by comparing multiple methods including pruning, bootstrapping and random forest. Then an image differencing will be performed to show areas of marsh change.
"]},{"key":"dc:title","label":"Title","values":["Sub-Pixel Classification of Historical and Current Marsh Habitat For the Eastern Mississippi Gulf Coast Using Remotely Sensed Images"]}]}],"canonical_facts":{"dc:contributor":["Wei Wu","Greg Carter","Maria Kalcic"],"dc:creator":["Frey, Jennifer"],"dc:date.available":["2073-07-14T07:00:00Z"],"dc:description.abstract":["Coastal Wetlands are changing at unprecedented rates (Reyes et al., 2000 & 2004) and the ability to monitor their changes remotely over time has become increasingly important (Hilbert, 2006 & Blaschke, 1996). The focus of this project was to complete a historical sub-pixel prediction land use classification map for six land classes: water estuarine emergent wetlands, fresh water wetlands, forests, low intensity developed areas and high intensity developed areas using high spatial resolution imagery and applying the proportions to medium resolution imagery for the past twenty-seven years (1984-2011) for the Western Lower Pascagoula River Basin. The other main objective was to complete an analysis of how the estuarine emergent marsh has changed during this historical time series. Landsat (medium resolution) and Quickbird (high resolution) imagery will be utilized to complete a regression tree based on the digital numbers and predictions will be completed by comparing multiple methods including pruning, bootstrapping and random forest. Then an image differencing will be performed to show areas of marsh change.
"],"dc:identifier":["https://aquila.usm.edu/masters_theses/978"],"dc:title":["Sub-Pixel Classification of Historical and Current Marsh Habitat For the Eastern Mississippi Gulf Coast Using Remotely Sensed Images"],"thesis:degree_level":["Masters Thesis"],"thesis:degree_name":["Master of Science (MS)"]},"updated_at":"2026-07-24T05:45:40Z"}