{"id":{"repo_id":"mtu","oai_identifier":"oai:digitalcommons.mtu.edu:etdr-1852"},"canonical_url":"https://search.ndltd.org/etd/mtu/oai:digitalcommons.mtu.edu:etdr-1852","repository":{"repo_id":"mtu","name":"Michigan Technological University","base_url":"https://digitalcommons.mtu.edu/do/oai/"},"display":{"title":"Application of remote sensing and machine learning modeling to post-wildfire debris flow risks","abstract":"Historically, post-fire debris flows (DFs) have been mostly more deadly than the fires that preceded them. Fires can transform a location that had no history of DFs to one that is primed for it. Studies have found that the higher the severity of the fire, the higher the probability of DF occurrence. Due to high fatalities associated with these events, several statistical models have been developed for use as emergency decision support tools. These previous models used linear modeling approaches that produced subpar results. Our study therefore investigated the application of nonlinear machine learning modeling as an alternative. Existing models identified the burn severity of wildfires as an important input in their development. Currently, the most widespread approach to obtaining this input is the use of the differenced normalized burn ratio (dNBR) index, which is determined using data from optical sensors on satellites. However, progress of this existing protocol is mostly hampered by the presence of cloud coverage during data acquisition since optical sensors cannot penetrate clouds. Radar sensors on the other hand can penetrate clouds and smoke. This study therefore developed a radar based algorithm to be used as an alternative to the dNBR metric. The results showed the SAR metric to perform even better than the dNBR, with an overall accuracy (OA) of ~60% and Kappa of 0.35 in comparison to an OA of ~35% and a kappa of 0.1 from the dNBR approach. Next we developed a nonlinear machine learning model to predict the likelihood of post-wildfire debris flow occurrences. This produced improved results over the linear modeling approach with an average sensitivity of 77%, depicting increased ability to predict ~8 out of 10 DF producing basins. Finally, we performed a case study to validate our DF model that showed the model’s robustness in isolating especially high hazard locations. Having these improved models will furnish emergency responders with an increased ability to better assess the associated risks of potential debris flow producing basins and make informed decisions on mitigation and/ or prevention measures.","abstract_html":"Historically, post-fire debris flows (DFs) have been mostly more deadly than the fires that preceded them. Fires can transform a location that had no history of DFs to one that is primed for it. Studies have found that the higher the severity of the fire, the higher the probability of DF occurrence. Due to high fatalities associated with these events, several statistical models have been developed for use as emergency decision support tools. These previous models used linear modeling approaches that produced subpar results. Our study therefore investigated the application of nonlinear machine learning modeling as an alternative. Existing models identified the burn severity of wildfires as an important input in their development. Currently, the most widespread approach to obtaining this input is the use of the differenced normalized burn ratio (dNBR) index, which is determined using data from optical sensors on satellites. However, progress of this existing protocol is mostly hampered by the presence of cloud coverage during data acquisition since optical sensors cannot penetrate clouds. Radar sensors on the other hand can penetrate clouds and smoke. This study therefore developed a radar based algorithm to be used as an alternative to the dNBR metric. The results showed the SAR metric to perform even better than the dNBR, with an overall accuracy (OA) of ~60% and Kappa of 0.35 in comparison to an OA of ~35% and a kappa of 0.1 from the dNBR approach. Next we developed a nonlinear machine learning model to predict the likelihood of post-wildfire debris flow occurrences. This produced improved results over the linear modeling approach with an average sensitivity of 77%, depicting increased ability to predict ~8 out of 10 DF producing basins. Finally, we performed a case study to validate our DF model that showed the model’s robustness in isolating especially high hazard locations. Having these improved models will furnish emergency responders with an increased ability to better assess the associated risks of potential debris flow producing basins and make informed decisions on mitigation and/ or prevention measures.","abstract_has_math":false,"creators":["Addison, Priscilla"],"institution":null,"degree_name":"Doctor of Philosophy in Geological Engineering (PhD)","degree_level":"Open Access Dissertation","degree_discipline":"Department of Geological and Mining Engineering and Sciences","degree_department":null,"school":null,"contributors":["Thomas Oommen"],"advisors":[],"committee_chairs":[],"committee_members":["Ann Maclean","Stanley Vitton","Qiuying Sha"],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:17:40Z","subjects":["Debris flow prediction","Probability modeling","Decision tree","Hazard assessment","Remote sensing","Synthetic aperture radar","Geological Engineering","Other Earth Sciences","Probability","Statistical Models"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.mtu.edu/etdr/703"],"render_values":[{"text":"https://digitalcommons.mtu.edu/etdr/703","href":"https://digitalcommons.mtu.edu/etdr/703","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.37099/mtu.dc.etdr/703","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/10.37099/mtu.dc.etdr/703","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.mtu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.mtu.edu%3Aetdr-1852","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Thomas Oommen"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Ann Maclean","Stanley Vitton","Qiuying Sha"]},{"key":"dc:creator","label":"Author","values":["Addison, Priscilla"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2019-11-28T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2018-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Department of Geological and Mining Engineering and Sciences"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Open Access Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy in Geological Engineering (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Debris flow prediction","Probability modeling","Decision tree","Hazard assessment","Remote sensing","Synthetic aperture radar","Geological Engineering","Other Earth Sciences","Probability","Statistical Models"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["10.37099/mtu.dc.etdr/703","https://digitalcommons.mtu.edu/etdr/703"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Historically, post-fire debris flows (DFs) have been mostly more deadly than the fires that preceded them. Fires can transform a location that had no history of DFs to one that is primed for it. Studies have found that the higher the severity of the fire, the higher the probability of DF occurrence. Due to high fatalities associated with these events, several statistical models have been developed for use as emergency decision support tools. These previous models used linear modeling approaches that produced subpar results. Our study therefore investigated the application of nonlinear machine learning modeling as an alternative. Existing models identified the burn severity of wildfires as an important input in their development. Currently, the most widespread approach to obtaining this input is the use of the differenced normalized burn ratio (dNBR) index, which is determined using data from optical sensors on satellites. However, progress of this existing protocol is mostly hampered by the presence of cloud coverage during data acquisition since optical sensors cannot penetrate clouds. Radar sensors on the other hand can penetrate clouds and smoke. This study therefore developed a radar based algorithm to be used as an alternative to the dNBR metric. The results showed the SAR metric to perform even better than the dNBR, with an overall accuracy (OA) of ~60% and Kappa of 0.35 in comparison to an OA of ~35% and a kappa of 0.1 from the dNBR approach. Next we developed a nonlinear machine learning model to predict the likelihood of post-wildfire debris flow occurrences. This produced improved results over the linear modeling approach with an average sensitivity of 77%, depicting increased ability to predict ~8 out of 10 DF producing basins. Finally, we performed a case study to validate our DF model that showed the model’s robustness in isolating especially high hazard locations. Having these improved models will furnish emergency responders with an increased ability to better assess the associated risks of potential debris flow producing basins and make informed decisions on mitigation and/ or prevention measures."]},{"key":"dc:description.note","label":"Dc Description Note","values":["<p>Please put an embargo on this submission for a couple months because the accepted manuscript in chapter 3 has not yet been published online by Taylor and Francis. They prefer they publish it first before I make it publicly available through my dissertation. Thank you.</p>"]},{"key":"dc:title","label":"Title","values":["Application of remote sensing and machine learning modeling to post-wildfire debris flow risks"]}]}],"canonical_facts":{"dc:contributor":["Thomas Oommen"],"dc:contributor.committeemember":["Ann Maclean","Stanley Vitton","Qiuying Sha"],"dc:creator":["Addison, Priscilla"],"dc:date.available":["2019-11-28T08:00:00Z"],"dc:date.created":["2018-01-01T08:00:00Z"],"dc:description.abstract":["Historically, post-fire debris flows (DFs) have been mostly more deadly than the fires that preceded them. Fires can transform a location that had no history of DFs to one that is primed for it. Studies have found that the higher the severity of the fire, the higher the probability of DF occurrence. Due to high fatalities associated with these events, several statistical models have been developed for use as emergency decision support tools. These previous models used linear modeling approaches that produced subpar results. Our study therefore investigated the application of nonlinear machine learning modeling as an alternative. Existing models identified the burn severity of wildfires as an important input in their development. Currently, the most widespread approach to obtaining this input is the use of the differenced normalized burn ratio (dNBR) index, which is determined using data from optical sensors on satellites. However, progress of this existing protocol is mostly hampered by the presence of cloud coverage during data acquisition since optical sensors cannot penetrate clouds. Radar sensors on the other hand can penetrate clouds and smoke. This study therefore developed a radar based algorithm to be used as an alternative to the dNBR metric. The results showed the SAR metric to perform even better than the dNBR, with an overall accuracy (OA) of ~60% and Kappa of 0.35 in comparison to an OA of ~35% and a kappa of 0.1 from the dNBR approach. Next we developed a nonlinear machine learning model to predict the likelihood of post-wildfire debris flow occurrences. This produced improved results over the linear modeling approach with an average sensitivity of 77%, depicting increased ability to predict ~8 out of 10 DF producing basins. Finally, we performed a case study to validate our DF model that showed the model’s robustness in isolating especially high hazard locations. Having these improved models will furnish emergency responders with an increased ability to better assess the associated risks of potential debris flow producing basins and make informed decisions on mitigation and/ or prevention measures."],"dc:description.note":["<p>Please put an embargo on this submission for a couple months because the accepted manuscript in chapter 3 has not yet been published online by Taylor and Francis. They prefer they publish it first before I make it publicly available through my dissertation. Thank you.</p>"],"dc:identifier":["10.37099/mtu.dc.etdr/703","https://digitalcommons.mtu.edu/etdr/703"],"dc:subject":["Debris flow prediction","Probability modeling","Decision tree","Hazard assessment","Remote sensing","Synthetic aperture radar","Geological Engineering","Other Earth Sciences","Probability","Statistical Models"],"dc:title":["Application of remote sensing and machine learning modeling to post-wildfire debris flow risks"],"thesis:degree_discipline":["Department of Geological and Mining Engineering and Sciences"],"thesis:degree_level":["Open Access Dissertation"],"thesis:degree_name":["Doctor of Philosophy in Geological Engineering (PhD)"]},"updated_at":"2026-11-19T08:17:40Z"}