{"id":{"repo_id":"york","oai_identifier":"oai:yorkspace.library.yorku.ca:10315/40620"},"canonical_url":"https://search.dev.ndltd.org/etd/york/oai:yorkspace.library.yorku.ca:10315/40620","repository":{"repo_id":"york","name":"York University","base_url":"https://yorkspace.library.yorku.ca/oai/request"},"display":{"title":"Exploiting Overlapping Landsat Scene Classifications and Focal Context to Identify Boreal Disturbance Mapping Uncertainty","abstract":"The BorealDB dataset is derived from a mosaic of Landsat scenes that were independently classified to identify historic fire and timber harvesting disturbances within Ontario. This thesis identifies and flags areas of classification uncertainty within BorealDB and scrutinizes them to assess classification confidence. The focal context of all orthogonal neighbour states was quantified to feed classification tree (CT) and random forest (RF) classifiers to predict focal disturbance classes. Uncertainty is deemed to exist where BorealDB and predicted CT or RF classes disagree. When RF and CT predictions were compared with the BorealDB classes, RF predicted more uncertainty (58%) than CT predictions (15%). Sampled locations compared with original satellite imagery and visual assessments suggested uncertainty depended on classifier, disturbance type, and spatial neighbours. Timber harvest disturbance classifications had the most uncertainty and CT predictions was the most consistent with neighbouring classifications and visual assessments indicating it is more effective than RF.","abstract_html":"The BorealDB dataset is derived from a mosaic of Landsat scenes that were independently classified to identify historic fire and timber harvesting disturbances within Ontario. This thesis identifies and flags areas of classification uncertainty within BorealDB and scrutinizes them to assess classification confidence. The focal context of all orthogonal neighbour states was quantified to feed classification tree (CT) and random forest (RF) classifiers to predict focal disturbance classes. Uncertainty is deemed to exist where BorealDB and predicted CT or RF classes disagree. When RF and CT predictions were compared with the BorealDB classes, RF predicted more uncertainty (58%) than CT predictions (15%). Sampled locations compared with original satellite imagery and visual assessments suggested uncertainty depended on classifier, disturbance type, and spatial neighbours. Timber harvest disturbance classifications had the most uncertainty and CT predictions was the most consistent with neighbouring classifications and visual assessments indicating it is more effective than RF.","abstract_has_math":false,"creators":["Wu, Wesley Jovia"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Remmel, Tarmo"],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-12-14","date_published":"2022-12-14","updated_at":"2026-07-24T06:34:05Z","subjects":["Geographic information science","Forestry","Geography"],"languages":["en"],"rights":["Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests."],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/10315/40620","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Remmel, Tarmo"]},{"key":"dc:creator","label":"Author","values":["Wu, Wesley Jovia"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2022-12-14T16:21:16Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2022-12-14T16:21:16Z"]},{"key":"dc:date.issued","label":"Date","values":["2022-12-14"]},{"key":"dc:type","label":"Dc Type","values":["Electronic Thesis or Dissertation"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Geographic information science","Forestry","Geography"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/10315/40620"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The BorealDB dataset is derived from a mosaic of Landsat scenes that were independently classified to identify historic fire and timber harvesting disturbances within Ontario. This thesis identifies and flags areas of classification uncertainty within BorealDB and scrutinizes them to assess classification confidence. The focal context of all orthogonal neighbour states was quantified to feed classification tree (CT) and random forest (RF) classifiers to predict focal disturbance classes. Uncertainty is deemed to exist where BorealDB and predicted CT or RF classes disagree. When RF and CT predictions were compared with the BorealDB classes, RF predicted more uncertainty (58%) than CT predictions (15%). Sampled locations compared with original satellite imagery and visual assessments suggested uncertainty depended on classifier, disturbance type, and spatial neighbours. Timber harvest disturbance classifications had the most uncertainty and CT predictions was the most consistent with neighbouring classifications and visual assessments indicating it is more effective than RF."]},{"key":"dc:title","label":"Title","values":["Exploiting Overlapping Landsat Scene Classifications and Focal Context to Identify Boreal Disturbance Mapping Uncertainty"]}]}],"canonical_facts":{"dc:contributor.advisor":["Remmel, Tarmo"],"dc:creator":["Wu, Wesley Jovia"],"dc:date.accessioned":["2022-12-14T16:21:16Z"],"dc:date.available":["2022-12-14T16:21:16Z"],"dc:date.issued":["2022-12-14"],"dc:description.abstract":["The BorealDB dataset is derived from a mosaic of Landsat scenes that were independently classified to identify historic fire and timber harvesting disturbances within Ontario. This thesis identifies and flags areas of classification uncertainty within BorealDB and scrutinizes them to assess classification confidence. The focal context of all orthogonal neighbour states was quantified to feed classification tree (CT) and random forest (RF) classifiers to predict focal disturbance classes. Uncertainty is deemed to exist where BorealDB and predicted CT or RF classes disagree. When RF and CT predictions were compared with the BorealDB classes, RF predicted more uncertainty (58%) than CT predictions (15%). Sampled locations compared with original satellite imagery and visual assessments suggested uncertainty depended on classifier, disturbance type, and spatial neighbours. Timber harvest disturbance classifications had the most uncertainty and CT predictions was the most consistent with neighbouring classifications and visual assessments indicating it is more effective than RF."],"dc:identifier.uri":["http://hdl.handle.net/10315/40620"],"dc:language":["en"],"dc:rights":["Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests."],"dc:subject":["Geographic information science","Forestry","Geography"],"dc:title":["Exploiting Overlapping Landsat Scene Classifications and Focal Context to Identify Boreal Disturbance Mapping Uncertainty"],"dc:type":["Electronic Thesis or Dissertation"]},"updated_at":"2026-07-24T06:34:05Z"}